Optic Nerve and Retinal Imaging Methods

Number: 0344

Table Of Contents

Policy

Scope of Policy

This Clinical Policy Bulletin addresses optic nerve and retinal imaging methods.

1. Medical Necessity

   Aetna considers optic nerve and retinal imaging methods medically necessary for documenting the appearance of the optic nerve head and retina in the following diagnoses/individuals:

   • Age-related macular degeneration
   • Cystoid macular edema following cataract surgery
   • Diabetic retinopathy
   • Dysthyroid optic neuropathy
   • Epiretinal membrane
   • Ethambutol-induced optic neuropathy
   • Glaucoma
   • Glaucoma suspects
   • Macular edema
   • Macular hole
   • Non-arteritic anterior ischemic optic neuropathy
   • Posterior vitreous detachment
   • Pseudotumor cerebri
   • Screening and monitoring for chloroquine (Aralen), ethambutol (Myambutol), ezogabine (Potiga), hydroxychloroquine (Plaquenil), ponatinib (Iclusig), siponimod (Mayzent), and vigabatrin (Sabril) toxicityFootnote1*
   • Sudden onset vitreous hemorrhage
   • Vitreomacular traction and vitreomacular adhesion
   • Vogt-Koyanagi-Haradas (to quantify subretinal fluid and to follow individuals during treatment)
   • Other diseases where the optic nerve head and retina have been affected.

   Optic nerve imaging for glaucoma more frequently than once per year is considered not medically necessary.

   Note: Accepted optic nerve and retinal imaging methods include the following:

   • Confocal laser scanning ophthalmoscopy
   • Nerve fiber layer testing or analysis (confocal laser scanning tomography with polarimetry)
   • Optical coherence tomography (OCT)
   • Stereophotogrammetry.

   Aetna considers optical coherence tomography angiography (OCTA) a medically necessary alternative to fluorescein angiography (FA) for the following:

   1. Diagnosis of chorioretinal vascular abnormalities related to age-related macular degeneration, choroidal neovascularization, and non-infectious vasculitis;
   2. Evaluation of acute exudative inflammation (e.g. toxoplasmosis and optic disc edema), diabetic retinopathy, and intra-ocular tumors; and
   3. Monitoring treatment of conditions amenable to laser photocoagulation.

   Aetna considers performance of both OCTA and FA for the same indication or performance of both OCT and OCT-A for the same indication as not medically necessary.

   Footnote1* In addition to annual screening that should begin after 5 years of use (or sooner it there are unusual risk factors), a  baseline study of optic nerve and retinal imaging is considered medically necessary before initiation of chloroquine, hydroxychloroquine, or vigabatrin therapy.

2. Experimental, Investigational, or Unproven

   1. Aetna considers optic nerve and retinal imaging methods experimental, investigational, or unproven for the following (not an all-inclusive list):

      • As biomarker for peripheral artery disease
      • Cataracts
      • Dry eye diseases
      • Evaluation of the neurodegeneration pattern in individuals with intra-cranial tumors
      • Evaluation of Parinaud oculoglandular syndrome (cat scratch disease)
      • Evaluation of schizophrenia spectrum disorders
      • Evaluation of visual snow syndrome
      • Imaging following intra-ocular lens (IOL) exchange following IOL dislocation
      • Imaging of the retina as a biomarker for neurodegeneration in frontotemporal degeneration, multiple sclerosis and optic neuritis
      • Ocular histoplasmosis
      • Posterior capsule opacification
      • Routine screening of asymptomatic persons for glaucoma and other retinal diseases
      • Screening/monitoring of persons on fingolimod (Gilenya).
   2. Aetna considers the use of patient-initiated image capture and transmission to a remote surveillance center via the optical coherence tomography (OCT) device experimental, investigational, or unproven because the effectiveness of this approach has not been established.
   3. Aetna considers OCTA experimental, investigational, or unproven for diagnosing and monitoring of glaucoma because the effectiveness of this approach has not been established.
   4. Aetna considers ultrasonographic (US) assessment of optic nerve sheath diameter (ONSD) experimental, investigational, or unproven as a screening tool for intracranial hypertension in traumatic brain injury because the effectiveness of this approach has not been established.
   5. Aetna considers ultrasonographic (US) assessment of optic nerve sheath diameter (ONSD) experimental, investigational, or unproven for assessment of intracranial pressure in aneurysmal subarachnoid hemorrhage because the effectiveness of this approach has not been established.

3. Related Policies

   • CPB 0563 - Retinopathy Telescreening Systems

Background

The appearance of the optic nerve head (the disc) and the nerve fiber layer is evaluated in the diagnosis and follow-up of glaucoma.  The standard methods of detecting glaucoma include ophthalmoscopy, tonometry, perimetry, and gonioscopy.  These procedures are considered part of the comprehensive ophthalmologic examination.  Recently, other methods of measuring the optic disc and the nerve fiber layer have been developed in an attempt to create more accurate and reproducible methods of screening, detecting, and following structural parameters related to glaucoma.  These methods include the following:

Confocal Laser Scanning Ophthalmoscopy

Confocal laser scanning ophthalmoscopy, also known as scanning laser ophthalmoscopy (SLO), is a method of examining the eye using confocal laser scanning microscopy (stereoscopic videographic digitized imaging) to make quantitative topographic measurements of the optic nerve head and surrounding retina. This may be done with either reflection or fluorescence. Targeted tissues can be viewed in 3-dimensional (3D) high-resolution planes running parallel to the line of sight. 

The confocal laser scanning tomographic ophthalmoscope scans layers of the retina to make quantitative measurements of the surface features of the optic nerve head and fundus.  It has been used as an alternative to standard ophthalmologic methods of evaluating the optic nerve head and fundus in patients with glaucoma, papilledema, and other disorders affecting the retina.  Other terms for confocal laser scanning tomography include: laser scanning topography, confocal scanning laser topography, confocal laser scanning tomography, scanning laser ophthalmoscopy (SLO), and electro-fundus imaging. Types of confocal laser scanning ophthalmoscopes include:

• Heidelberg Laser Tomographic Scanner or Heidelberg Retina Tomograph (HRT) (Heidelberg Engineering, Dossenheim, Germany),
• TopSS Topographic Scanning System (Laser Diagnostic Technologies, San Diego, CA); and
• Zeiss Confocal Laser Scanning Ophthalmoscope. (Zeiss Humphrey Systems, Dublin, CA)

Nerve Fiber Layer Testing or Analysis (Laser Scanning Polarimetry)

Thinning of the nerve fiber layer is associated with glaucomatous damage and has been shown to be correlated with visual field loss.  Scanning laser polarimetry, also called confocal scanning laser polarimetry, measures change in the linear polarization (retardation) of light. It uses both a scanning laser ophthalmoscope and a polarimeter (an optical device to measure linear polarization change) to measure the thickness of the nerve fiber layer of the retina. The confocal scanning laser polarimeter is essentially a confocal scanning laser ophthalmoscope with an additional polarization modulator, a cornea polarization compensator and a polarization detection unit. 

The GDx Nerve Fiber Analysis System (Laser Diagnostic Technologies, Inc., San Diego, CA) is a confocal laser scanning ophthalmoscope with an integrated polarimeter.  Instead of measuring topography, or height of the retina, like other confocal laser scanners, GDx measures the thickness of the retinal nerve fiber layer and then analyzes the results and compares them to a database of normative values.

The Retinal Thickness Analyzer (RTA) Digital Fundus Imaging (Talia Technology, Inc., Tampa, FL) uses a scanning laser biomicroscope that uses a laser to create a series of slit images of the retina that are digitalized and converted into a topographic map that quantifies retinal thickness.

Optical Coherence Tomography

Optical coherence tomography (OCT) is a noninvasive, transpupillary, retinal imaging technology, which uses near-infrared light to produce high-resolution cross-sectional images. It is suggested for diagnostic use as an alternative to standard excisional biopsy and to guide interventional procedures.

OCT (e.g., Humphrey OCT Scanner (Zeiss Humphrey, Dublin, CA)) has also been used for screening, diagnosis, and management of glaucoma and other retinal diseases.  In OCT, low coherence near-infrared light is split into a probe and a reference beam.  The probe beam is directed at the retina while the reference beam is sent to a moving reference mirror (AHFMR, 2003).  The probe light beam is reflected from tissues according to their distance, thickness, and refractive index, and is then combined with the beam reflected from the moving reference mirror.  When the path lengths of the two light beams coincide (known as constructive interference) this provides a measure of the depth and reflectivity of the tissue that is analogous to an ultrasound A scan at a single point.  A computer then corrects for axial eye movement artifacts and constructs a two dimensional B mode image from successive longitudinal scans in the transverse direction.  A map of the tissue is then generated based on the different reflective properties of its components, resulting in a real-time cross-sectional histological view of the tissue.

Barella et al (2013) examined the diagnostic accuracy of machine learning classifiers (MLCs) using retinal nerve fiber layer (RNFL) and optic nerve (ON) parameters obtained with spectral-domain optical coherence tomography (SD-OCT).  A total of 57 patients with early-to-moderate POAG and 46 healthy patients were recruited.  All 103 patients underwent a complete ophthalmological examination, achromatic standard automated perimetry, and imaging with SD-OCT.  Receiver operating characteristic curves were built for RNFL and ON parameters.  Ten MLCs were tested.  Areas under ROC curves (aROCs) obtained for each SD-OCT parameter and MLC were compared.  The mean age was 56.5 ± 8.9 years for healthy individuals and 59.9 ± 9.0 years for glaucoma patients (p = 0.054).  Mean deviation values were -1.4 dB for healthy individuals and -4.0 dB for glaucoma patients (p < 0.001).  Spectral domain-OCT parameters with the greatest aROCs were cup/disc area ratio (0.846) and average cup/disc (0.843).  Areas under ROC curves obtained with classifiers varied from 0.687 (CTREE) to 0.877 (RAN).  The aROC obtained with RAN (0.877) was not significantly different from the aROC obtained with the best single SD-OCT parameter (0.846) (p = 0.542).  The authors concluded that MLCs showed good accuracy, but did not improve the sensitivity and specificity of SD-OCT for the diagnosis of glaucoma.

Bidot et al (2013) noted that OCT is used primarily in neuro-ophthalmology to measure thinning of the RNFL in optic neuropathies and to rule out a subtle maculopathy in patients complaining of blurred vision with a "normal" fundoscopic appearance.  Only a few studies address the role of OCT in papilledema secondary to intra-cranial hypertension.  Optical coherence tomography has been proposed as a diagnostic tool for mild papilledema, assisting the clinician in differentiating papilledema from optic nerve head drusen (ONHD), and for following the RNFL thickening from papilledema.  However, the contribution of OCT in intra-cranial hypertension management is still unclear with the exception of its role in detecting associated maculopathy.  Currently, OCT does not replace visual field testing and fundus examination.

In a comparative case-series study, Kulkarni et al (2014) evaluated the clinical utility of SD-OCT in differentiating mild papilledema from buried ONHD.  A total of 16 eyes of 9 patients with ultrasound-proven buried ONHD, 12 eyes of 6 patients with less than or equal to Frisen grade 2 papilledema owing to idiopathic intra-cranial hypertension were included in this study.  Two normal fellow eyes of patients with buried ONHD were included.  A raster scan of the ON and analysis of the RNFL thickness was performed on each eye using SD-OCT.  Eight eyes underwent enhanced depth imaging SD-OCT.  Images were assessed qualitatively and quantitatively to identify differentiating features between buried ONHD and papilledema.  Five clinicians trained with a tutorial and masked to the underlying diagnosis independently reviewed the SD-OCT images of each eye to determine the diagnosis.  Main outcome measures were differences in RNFL thickness in each quadrant between the 2 groups and diagnostic accuracy of 5 independent clinicians based on the SD-OCT images alone.  These investigators found no difference in RNFL thickness between buried ONHD and papilledema in any of the 4 quadrants.  Diagnostic accuracy among the readers was low and ranged from 50 % to 64 %.  The kappa coefficient of agreement among the readers was 0.35 (95 % confidence interval [CI]: 0.19 to 0.54).  The authors concluded that SD-OCT is not clinically reliable in differentiating buried ONHD and mild papilledema.

Stereo-Photogrammetry

Stereo-photogrammetry, (Glaucoma-Scope (OIS, Sacramento, CA)) measures the dimensions of the optic disc in three-dimensional space using stereo-photography.  Stereo-photographs are taken from two camera positions with parallel optical axes.  Stereo-analysis of these photographs are used to determine the three-dimensional characteristics of the optic nerve head, and for following glaucomatous change of the optic nerve head over time.  Stereo-plotters and digital computer processing of scanned images have been used in an attempt to provide more quantitative, objective, and reproducible methods of measuring optic nerve disc changes.

Each of these methods has been used to image the optic nerve head in glaucoma patients. According to available guidelines, these methods may be used for documenting the appearance of the optic nerve head and retina in persons with glaucoma and other retinal diseases.  But these devices have not been proven to be of value for screening of asymptomatic persons. 

Available methods of optic nerve imaging (e.g., Heidelberg Retinal Tomograph, GDx confocal laser scanning polarimeter, Humphrey OCT Scanner, Glaucoma-Scope) were cleared by the U.S. Food and Drug Administration (FDA) based on a 510(k) premarket notification due to their “substantial equivalence” to other devices on the market.  Thus, the manufacturers were not required to submit to the FDA the evidence that would be required to support a premarket approval application (PMA).

An AAO Preferred Practice Pattern on Primary Open-Angle Glaucoma (2005), which focuses on management of patients with evidence of glaucomatous damage as manifested by acquired optic nerve or nerve fiber layer abnormalities or typical visual field loss, states that optic nerve head and retinal nerve fiber layer analysis should be performed to document optic nerve head morphology.

Finnish evidence-based guidelines on open-angle glaucoma (Tuulonen et al, 2003) state: “High technology instruments (the Heidelberg retina tomograph, retinal nerve fiber analyzer and optical coherence tomography), developed for nerve fiber layer and optic disc imaging and measurements, are not yet ready for routine glaucoma diagnostics. Due to their sensitivity and specificity, some instruments may be used for the follow-up of glaucoma.”

Aetna’s position on optic nerve and retinal imaging devices is based on the use of these devices as standard of care for documenting the appearance of the optic nerve head and retina, in place of retinal drawings or fundus photographs.  In addition, optic nerve head imaging methods have been used as a noninvasive alternative to fluorescein angiography and slit-lamp biomicroscopy in assessing the retinal nerve fiber layer, although fluorescein angiography is also a sensitive test to detect leakage of incompetent retinal vessels.  Because of the slow rate of progression of glaucoma, repeated optic nerve head imaging is not necessary more frequently than once every year.

Although optic nerve and retinal imaging devices have been used to document the appearance of the optic nerve head and retina, there is a lack of evidence from prospective clinical studies demonstrating that clinical outcomes are improved by incorporating this technology into glaucoma screening.  A number of structured evidence reviews have concurred that there is limited evidence of the clinical utility of optic nerve head imaging methods in these situations (AHP, 1996; AHFMR, 1996; Lee, et al., 1996; TEC, 2001; AHFMR, 2003: TEC, 2003; IECS, 2003; AHFMR, 2006).  A BlueCross BlueShield Association Technology Evaluation Center (2003) assessment of optic nerve imaging devices (termed RNFL analysis (RNFLA) in the report) in the diagnosis and management of glaucoma and concluded that they do not meet TEC criteria.  Using data from the Ocular Hypertension Treatment Study, the assessment found that RNFLA would not be useful in deciding whether to initiate early treatment of glaucoma or change treatment regimens, as the vast majority of patients with abnormal RNFLA test results would not be expected to go on to develop glaucoma.  The assessment concluded: "The scientific evidence is insufficient to permit conclusions concerning the effects of RNFLA for the diagnosis or management of POAG [primary open angle glaucoma]; therefore, it is not possible to determine whether the procedure improves net health outcome." 

The TEC assessment (BCBSA, 2003) focused on the best available evidence for retinal nerve fiber layer analysis.  Although there are many published studies of RNFLA, almost all of them are cross-sectional studies that evaluate the sensitivity and specificity of RNFA by comparing RNFA measurements of normal persons to persons with glaucoma or ocular hypertension. The assessment explains, however, that cross-sectional studies do not follow persons over time and are not designed to assess the relationship between a test result and subsequent development of disease.  Studies that follow persons over time (longitudinal studies) are necessary to evaluate the ability of a test to predict which persons will develop disease and need treatment from those that will not develop disease.

Another limitation of published cross-sectional studies of RNFA is that they only compare normal persons to persons with glaucoma or ocular hypertension.  Because these studies do not include persons with other ocular conditions, they do not provide information on the ability of RNFLA to distinguish patients with glaucoma or ocular hypertension from other ocular conditions.  The subjects of these cross-sectional studies do not accurately reflect the spectrum of conditions that one would expect to see in the usual clinical practice setting (BCBSA, 2001).  The external validity of these studies could be strengthened by selecting study subjects from a representative sample of the population of patients suspected of having the disease.

Cross-sectional studies cannot determine whether abnormalities that are detected by RNFA but not by any other standard method are early disease that might benefit from treatment (BCBSA, 2003).  The most reliable evidence for assessing the clinical impact of RNFLA would be direct evidence from randomized trials comparing the impact on visual field defects of treatment initiated by different thresholds of RNFLA test results.  As no such studies are available, technology assessments of RNFLA have had to rely on indirect evidence. 

The most reliable indirect evidence to estimate diagnostic performance of RNFLA comes from longitudinal studies with a clinical population of patients suspected of having glaucoma, using follow-up for visual loss as a reference standard (BCBSA, 2003).  The key issue in the early detection of glaucoma is how well test results predict the future development of visual loss.  Thus, it is critical that the diagnostic performance of an early test be measured against follow-up for visual changes as the reference standard, in a longitudinal study.

The TEC assessment noted that no longitudinal study has yet appeared that selected subjects for whom RNFLA results are most likely to influence management decisions, that is, persons who have normal intraocular pressure and who do not meet conventional diagnostic criteria for glaucoma (BCBSA, 2003).  The report identified only two published longitudinal studies that present data showing whether RNFLA changes precede development of visual field defects, on an individual patient basis.  The most useful longitudinal evidence for the indication concerning detection of glaucoma comes from a subset of the Ocular Hypertension Treatment Study (Kamal et al, 2000).  In this study, 21 patients progressed from ocular hypertension to glaucoma (converters) and 164 patients did not progress (nonconverters).  Of the 21 converters, 13 had abnormal RNFLA results and in 11 of these the tests were positive prior to development of visual field defects (average lead time was 5.4 months).  Of the 164 nonconverters, 47 had abnormal RNFLA results.  Using this study’s results to estimate diagnostic performance, the positive predictive value of RNFLA, which is the most relevant index for early detection of glaucoma, is 22 % (13/60), so that 78 % of persons with signs of progression of glaucoma on RNFLA would not go on to develop visual field changes within the 6-year follow-up period of this study.  The TEC assessment concluded that "a positive predictive value at this level does not appear to be sufficiently high for use in deciding to initiate early treatment or to change treatment regimens" (BCBSA, 2003).  In addition, this study only included persons with ocular hypertension, and may not accurately reflect the performance of RNFLA in a cohort of glaucoma suspects without ocular hypertension.  Given that the predictive values are dependent on the prevalence of the target condition in the population, the predictive value RNFLA in a population that includes individuals without ocular hypertension is likely to be even lower than the estimate from this study of ocular hypertensives, given that persons with ocular hypertension, an established glaucoma risk factor, are more likely to develop glaucoma than persons without ocular hypertension.

Chauhan et al (2001) reported on the results of a longitudinal study of RNFLA and visual field testing (perimetry) in patients with glaucoma who were followed for a median of 5.5 years.  In 29 percent of cases, progression was detected by both perimetry and RNFLA; when progression was detected by both tests, it was just as common for perimetry to detect it first as it is for RNFLA to detect it first.  Although progression was observed by RNFLA alone in 40 % of patients, it is unclear from this study how often such patients experience visual progression (true positives versus false positives).  The TEC assessment (2003) emphasizes that the key issue in the early detection of glaucoma is how well test results predict the future development of visual loss, as loss of vision is an endpoint that patients experience in terms of quality of life and ability to function.

A report on glaucoma screening prepared for the UK National Screening Committee (Spry and Sparrow, 2003) stated that methods of assessing optic nerve head appearance using images acquired by digital scanning laser instrumentation are quantitative and rapid to perform.  The report concluded, however, that “[t]o date, however, scant longitudinal information is available on individuals who exhibit apparent structural abnormalities with these techniques but no glaucomatous loss of visual function.”  The ability to utilize optic nerve head appearance in assessing glaucoma risk is limited by the fact that “considerable overlap exists between the distribution of relevant parameters found in patients with glaucoma and normal individuals.”

The AAO and the American Glaucoma Society prepared a work group report to provide a “rationale for [insurance] coverage” of optic nerve head imaging (Remey 2002; AAO, 2003; AAO/AGS, 2003).  The AAO/AGS Work Group statement (2003) on the clinical utility of optic nerve scanning devices in screening focused exclusively on comparisons with automated perimetry or photography used alone.  However, the standard methods of detecting and monitoring glaucoma include ophthalmoscopy (to inspect the optic disk and nerve fiber layer), drawings of the optic nerve head and stereoscopic disc photographs (to document the status of the optic nerve head), tonometry (to measure intraocular pressure), perimetry (to measure visual fields), and gonioscopy (to measure the angle of the anterior chamber). 

None of the studies cited in the AAO/AGS Work Group statement (2003) represented prospective clinical outcome studies.  The need for prospective clinical outcome studies comparing optic nerve scanning devices to standard methods of evaluation is especially critical given that there is no established gold standard comparison test for predicting the risk of glaucoma development prior to the onset of visual field defects.  Although RNFLA devices have been in commercial use for more than a decade, the quality of evidence supporting their use remains limited, and the best available evidence indicates that the ability of RNFLA to predict progression to glaucoma is limited. 

The monograph from the AAO/AGS Work Group (2003) commented on the limited quality of evidence supporting the use of optic nerve scanning devices.  The monograph stated: "In clinical practice, many patients now are tested with more advanced visual field techniques designed to detect glaucoma earlier.  Yet these variations on visual field testing have not required rigorous TEC assessment to determine if they are useful.  Clinical experience, cross-sectional studies, and a few longitudinal cohort studies have shown that they are useful improvements in our ability to detect glaucoma and/or progression.  The case is similar to RNFLA, where clinicians and researchers have determined that this newer technology surpasses or equals the current clinical assessment of the optic nerve and nerve fiber layer."

Assessment of the evidence supporting the use of optic nerve scanning devices is required because these techniques have been presented as a new technology rather than as an incremental advance over an existing technology such as ophthalmoscopy or other established methods of evaluating the retina and optic nerve head.

The Center for Medicare and Medicaid Services (CMS) has not established a national coverage position on optic nerve imaging devices; CMS has left coverage of optic nerve imaging devices to the discretion of local Medicare carriers. 

Optical coherence tomography (OCT) (e.g., Humphrey OCT Scanner (Zeiss Humphrey Systems, Dublin, CA) has also been used for screening, diagnosis, and management of glaucoma and other retinal diseases. Optical coherence tomography (OCT) is a relatively new non-invasive imaging modality that uses reflected light in a manner analogous to the use of sound waves in ultrasonography to create high-resolution (10 micron) cross-sectional images of the vitreoretinal interface, retina and sub-retinal space, analogous to histological sections seen through a light microscope.  OCT also gives quantitative information about the peripapillary retinal nerve fiber layer thickness. 

The Alberta Heritage Foundation for Medical Research (2003) assessed the value of optical coherence tomography (OCT) in the diagnosis of retinal diseases.  It stated that “OCT appears promising for diagnosing patients with cystoid macular edema and moderate glaucoma.”

OCT can determine the presence of cystoid macular edema (CME) by visualizing the fluid-filled spaces in the retina.  The amount of CME can be monitored over time by quantifying the area of cystoid spaces on a cross-sectional image through the macula.  Studies have reported OCT to be comparable to fluorescein angiography in the evaluation of CME.  However, fluorescein angiography may be a more sensitive study for leakage of incompetent retinal vessels (Roth, 2001).

OCT, scanning laser ophthalmoscope, and confocal laser tomography may also be useful in macular holes, in establishing the status of the vitreomacular interface and distinguishing full-thickness holes from lamellar holes and macular cystic lesions (Valero and Atebara, 2001).  According to the American Academy of Ophthalmology (2003), “[i]n most cases the diagnosis [of macular hole] is made by clinical evaluation. Optical coherence tomography provides information on the anatomy of the macular hole and may aid in the diagnosis and staging.”

OCT has also been used in a variety of other retinal diseases.  In diabetic retinopathy, OCT has been used to evaluate retinal swelling and serous retinal detachment.  OCT has been able to demonstrate a moderate correlation between retinal thickness and best-corrected visual acuity, and it has been able to demonstrate three basic structural changes of the retina from diabetic retinal edema, i.e., retinal swelling, cystoid edema, and serous retinal detachment (Khan and Lam, 2004).  There is, however, a lack of prospective clinical studies demonstrating that clinical outcomes are improved by incorporating OCT into screening of persons with diabetes for retinopathy.  Current guidelines from the American Diabetes Association do not incorporate OCT into diabetic retinopathy screening algorithms.  Optical coherence tomography can be useful for quantifying retinal thickness, monitoring partial resolution of macular edema, and identifying vitreomacular traction in selected patients with diabetic macular edema caused by a taut posterior hyaloid face (AAO, 2003).  The AAO (2003) states that this test might be considered in diabetic retinopathy patients unresponsive to laser treatment for macular edema for whom the ophthalmologist is considering vitrectomy with removal of the posterior hyaloid face.

OCT has also been used to determine the presence of sub-retinal fluid and in documenting the degree of retinal thickening in age-related macular degeneration (AMD).  This study has shown decreased reflectance at the level of the rod-cone layer indicating that atrophy is present in this layer (Maturi, 2005).  According to guidelines from the AAO (2003), “the value of this test in evaluating and treating AMD remains unknown.”

Thiele and colleagues (2020) noted that relative ellipsoid zone reflectivity (rEZR) represents a potential biomarker of photoreceptor health on spectral-domain OCT (SD-OCT).  Because manual quantification of rEZR is laborious and lacks spatial resolution, automated quantification of the rEZR would be beneficial.  The se researchers examined the reliability and reproducibility of an automated rEZR quantification method.  The rEZR was acquired using a manual and an automated approach in eyes with age-related macular degeneration (ARMD) and healthy controls.  The rEZR obtained from both methods was compared and the agreement between the methods and their reproducibility assessed.  A total of 40 eyes of 40 subjects aged 65.2 ± 7.8 years (mean ± standard deviation) were included in this study.  Both the manual and automated method showed that control eyes exhibited a greater rEZR than ARMD eyes (p < 0.001).  Overall, the limits of agreement between the manual and automated method were -7.5 to 7.3 arbitrary units (AU) and 95 % of the data points had a difference in the rEZR between the methods of ±8.2 %.  An expected perfect reproducibility was observed for the automated method, whereas the manual method had a coefficient of repeatability of 6.3 AU.  The authors concluded that the automated quantification of rEZR method was reliable and reproducible.  Moreover, these researchers stated that further studies of the rEZR as a novel biomarker for ARMD severity and progression are needed.

The authors stated that this study had several drawbacks.  First, the sample size of the study was relatively small with 40 subjects; however, there were 7 discrete data points collected from each study eye, generating a total of 280 data points, which were sufficient for examining the agreement and reproducibility of the 2 methods.  Second, the study only examined the rEZR on single line scan.  This work was an initial study to test the authors’ automated algorithm and it was not feasible to carry out manual quantification of the rEZR in such a large volume scan.  Because the data from this study showed a high agreement between the automated and manual methods, the next step will be to expand this research to examine the rEZR of the entire volume scan.  These investigators believed that the adaptation of the automated algorithm on volumetric OCT data is easy to implement.  Because the process of analyzing volumetric data is exactly the same as that used in a single OCT line scan, except that it is repeated on multiple B-scans, they did not expect any difference in the performance of the automatic method when applied to volumetric OCT data.  They stated that further studies are needed to assess the performance of their proposed automated method in determining the rEZR on OCT images acquired from other systems and on other retinal conditions.  Third, the coordinates of the segmentation lines were generated semi-automatically in this study; thus, the entire process of obtaining the rEZR was not fully automated because these researchers still needed to carry out manual review and correction of OCT segmentation.  However, the actual detection of the peaks and measurement of the reflectivity was an automated process.  This step was their first in the development toward a fully automated approach.  They were working toward implementing artificial intelligence in the process to perform the segmentation automatically in the future.  They believed that the implantation of artificial intelligence not only improves the efficiency of the rEZR determination process, but also allows them to obtain the rEZR data from 3D OCT volume scans and in context of large and longitudinal AMD studies.  Finally, in this study, the RPE was flattened; the current automated method required a relatively straight horizontal reference line for an accurate detection of the reflectivity peaks.  In the authors’ experience, this reference line does not have to be absolutely flat for the algorithm to detect the peaks, as demonstrated by a high peak detection rate in this study.  The process of flattening the RPE layer inevitably also altered the contour of the retinal layers above it.  Although changes in the retinal contour did not appear to affect the rEZR quantification, to overcome this limitation and further improve the accuracy of the peak detection, these researchers are working on a new algorithm to detect the coordinates of the EZ and ELM without the need to flatten the RPE layer.

OCT is also being investigated in evaluating choroidal neovascularization (CNV).  Well-defined and diffuse CNV have characteristic appearances on OCT, as do sub-retinal hemorrhages and retinal detachments.  Despite the many advantages of OCT, fluorescein angiography remains the imaging modality of choice in the management of CNV.  Currently, OCT cannot replace fluorescein angiography in the management of CNV (Wu, 2005).  

Alberta Heritage Foundation for Medical Research (2003) stated that “while OCT appears promising for diagnosing patients with cystoid macular edema and moderate glaucoma, it still has a number of practical and theoretical limitations.  Its ability to detect any other of the myriad retinal diseases is unknown.  It is clear that OCT in its current state of development is ineffective as a stand alone diagnostic test, but a study has not yet been conducted to assess its value as part of a serial testing strategy.  The position of OCT in the scheme of testing needs to be established so that an optimal testing strategy can be identified that is both highly accurate and clinically practical.  Randomized controlled trials are also needed to establish the clinical impact of OCT diagnostic imaging on the management, treatment options, and outcomes of patients”.  The AHFMR reasoned that “[w]hile OCT appears promising as a tool for diagnosing retinal disease, there are many questions relating to its clinical utility that are not likely to be answered by a cross-sectional study.  Longitudinal studies are needed to determine the temporal relationship between OCT RNFL thickness measurements and visual field defects, and to identify any changes in RNFL thickness that could predict future visual deterioration.”

A more recent assessment by the Alberta Heritage Foundation for Medical Research (AHFMR, 2006) summarized the current status of ophthalmic scanning devices in glaucoma.  The assessment concluded: "Based on results from three systematic reviews, the value of CSLO [confocal scanning laser ophthalmoscopy] and SLP [scanning laser polarimetry] as diagnostic tools for the detection of early glaucoma remains unclear, although the HRT and GDx methods hold considerable promise for the detection of glaucoma-associated structural change.  The available evidence showed that HRT and GDx are able to differentiate between normal individuals and those with glaucoma.  However, whether these devices have the sensitivity and specificity to detect the early onset of glaucoma, before the onset of visual field loss, remains to be determined.  The available CSLO and SLP devices still await prospective validation against accepted measures of structural and functional change in terms of whether the use of a test results improves patient outcomes and is helpful in patient management and obviates unnecessary treatment."

The United Kingdom National Health Service National Coordinating Centre for Healthcare Technology Assessment (NCCHTA) conducted primary research and a comprehensive systemic review of the ophthalmic scanning devices in glaucoma screening (Kwartz et al, 2005).  The assessment concluded "[t]he findings of the glaucoma imaging study suggest that, although optic nerve head tomography and scanning laser polarimetry provide good-quality digital images, their data may contribute little to a patient's clinical diagnosis but would add significantly to the cost of their assessment."

Hickman (2007) reviewed the last 10 years of progress in the imaging of the optic nerve with a particular focus on applications to multiple sclerosis (MS).  Development of magnetic resonance imaging (MRI) of the optic nerve has lagged behind imaging of other parts of the CNS.  These limitations are due to technical challenges related to the small size and mobility of the optic nerves and artefacts caused by surrounding cerebrospinal fluid, orbital fat, and air-bone interfaces.  Nonetheless, the last 10 years has seen significant progress with regard to detecting optic nerve atrophy following optic neuritis, the use of fat- and cerebrospinal fluid (CSF)-suppressed high resolution imaging, the ability to measure magnetization transfer ratio and diffusivity in the optic nerve, and the emergence of SPIR-FLAIR for increasing sensitivity to inflammatory demyelination.  Remaining challenges include further reduction of movement artifacts, testing ultra-high field MRI systems and dedicated surface coils, and developing automated segmentation techniques to improve the reproducibility of quantitative measurements.  Finally the role of OCT as a marker of retinal damage needs to be clarified further through correlations with MRI, clinical, and electrophysiologic data.

In a report on optic nerve head and retinal nerve fiber layer analysis by the American Academy of Ophthalmology, Lin and associates (2007) evaluated the current published literature on the use of optic nerve head (ONH) and retinal nerve fiber layer (RNFL) measurement devices in diagnosing open-angle glaucoma and detecting progression.  The authors concluded that ONH and RNFL imaging devices provide quantitative information for the clinician.  Based on studies that have compared the various available technologies directly, there is no single imaging device that outperforms the others in distinguishing patients with glaucoma from controls.

In a report on laser scanning imaging for macular disease by the American Academy of Ophthalmology, McDonald and colleagues (2007) examined if  laser scanning imaging is a sensitive and specific tool for detecting macular disease when compared with the current standard technique of slit-lamp biomicroscopy or stereoscopic fundus photography.  Literature searches conducted in December 2004 and in August 2006 retrieved 370 citations.  The Retina Panel members selected 65 articles for the panel methodologist to review and rate according to the strength of the evidence.  Of the 65 articles reviewed, 6 provided level I evidence, 9 provided level II evidence, and 50 provided level III evidence.  A level I rating was assigned to studies that reported an independent masked comparison of an appropriate spectrum of consecutive patients, all of whom had undergone both the diagnostic test and the reference standard.  A level II rating was assigned to an independent masked or objective comparison; a study performed in a set of non-consecutive patients or confined to a narrow spectrum of study individuals (or both), all of whom had undergone both the diagnostic test and the reference standard; or an independent masked comparison of an appropriate spectrum, but the reference standard had not been applied to all study patients.  A level III rating was assigned when the reference standard was un-objective, unmasked, or not independent; positive and negative tests were verified using separate reference standards; or the study was performed in an inappropriate spectrum of patients.  There are high-level studies of the use of laser scanning imaging to quantify macular thickness and, thereby, macular edema in patients with diabetic retinopathy and to examine patients with a macular hole.  There is lower-quality evidence on the use of laser scanning imaging for other diseases of the macula.  There is insufficient evidence to compare the different instruments.  The authors concluded that there is level I evidence that laser scanning imaging can accurately and reliably quantify macular thickness in patients with diabetic retinopathy.  There is level I evidence that OCT provides additional information to clinical examination when used in patients with a macular hole.  Laser scanning imaging provides important information that is helpful in patient management by allowing objective serial quantitative measurements.  Although further studies are needed to develop an optimal testing strategy using these imaging modalities, laser scanning imaging is a sensitive, specific, reproducible tool for diagnosing macular edema and, therefore, is likely to be useful for managing diseases that result in macular edema.

Thenappan and colleagues (2021) stated that OCT summary measures have been suggested as a way to detect progression in eyes with advanced glaucoma.  These researchers showed that these measures have serious flaws largely due to segmentation errors; however, inspection of the images and thickness maps can be clinically useful.  These investigators tested the hypothesis that recently suggested global OCT measures for detecting progression in eyes with advanced progression are seriously affected by segmentation mistakes and other errors that limit their clinical utility.  A total of 45 eyes of 38 patients with a 24-2 mean deviation worse than -12 dB had at least 2 spectral domain OCT sessions (0.8 to 4.4 years apart) with 3.5-mm circle scans of the disc and cube scans centered on the fovea.  Average (global) circum-papillary retinal nerve fiber layer thickness (GcRNFL), and ganglion cell plus inner plexiform layer thickness (GGCLP) were obtained from the circle and cube scan, respectively.  To evaluate progression, ΔGcRNFL was calculated for each eye as the GcRNFL value at time 2 minus the value at time 1, and ΔGGCLP was calculated in a similar manner.  The b-scans of the 6 eyes with the highest and lowest ΔGcRNFL and ΔGGCLP values were examined for progression as well as segmentation, alignment, and centering errors.  Progression was a major factor in only 7 of the 12 eyes with the most negative values of either ΔGcRNFL or ΔGGCLP, whereas segmentation played a role in 8 eyes and was the major factor in all 12 eyes with the largest positive values.  Furthermore, alignment (1 eye) and other (3 eyes) errors played a secondary role in 4 of the 6 eyes with the most negative ΔGcRNFL values.  The authors concluded that for detecting the progression of advanced glaucoma, common summary metrics have serious flaws largely due to segmentation errors, which limited their utility in clinical and research settings.

In a prospective, controlled, single-center study, Ibrahim and colleagues (2010) examined the applicability of tear meniscus height (TMH) measurement using Visante OCT in the diagnosis of dry eye disease.  A total of 24 right eyes of 24 patients (6 males, 18 females; mean age of 63.14 +/- 13.4 years) with definite dry eye according to the Japanese dry eye diagnostic criteria and 27 right eyes of 27 control subjects (12 males, 15 females; mean age of 56.04 +/- 14.22 years) were recruited.  All subjects underwent slit-lamp TMH measurement, OCT upper and lower TMH measurements, tear film breakup time (BUT) measurements, vital staining, and Schirmer test.  The results were compared between the 2 groups by Mann-Whitney test.  Main outcome measures were the correlation between the clinical findings of slit-lamp TMH, strip meniscometry examination, tear functions, vital staining scores, and the OCT upper and lower TMH parameters were tested by Spearman's correlation test.  Receiver operating characteristic (ROC) curve technique was used to evaluate the sensitivity, specificity and cut-off values of OCT TMH examination in the diagnosis of dry eye.  The OCT upper and lower TMH values, slit-lamp TMH, strip meniscometry, tear film BUT, and vital staining scores were significantly lower in the dry eye patients compared with controls (p < 0.001).  A significant correlation between the OCT upper and lower TMH measurements as well as slit-lamp TMH, strip meniscometry, tear functions, vital staining scores, and the Schirmer test was found.  The ROC curve technique analysis of the OCT lower TMH showed that, when the cut-off value was set at less than 0.30 mm, the sensitivity and specificity of the testing were 67 % and 81 %, respectively.  The authors concluded that Visante OCT is a quick, non-invasive method for assessing the TMH, with acceptable sensitivity, specificity, and repeatability, and may have potential applications for the diagnosis and evaluation of dry eye disease.  They also stated that further studies on OCT TMH should determine age- and gender-specific cut-off values, sensitivity, specificity of the test in the diagnosis of dry eye disease when performed alone or in conjunction with other dry eye tests.

In a prospective, cross-sectional study, Jeoung et al (2010) evaluated quantitatively the degree of diffuse retinal nerve fiber layer (RNFL) atrophy using Stratus optical OCT.  A total of 102 eyes of 102 patients with diffuse RNFL atrophy and 102 healthy eyes of 102 age-matched subjects were enrolled in the Diffuse Atrophy Imaging Study.  Two experienced observers graded RNFL photographs of diffuse RNFL atrophy eyes using a previously reported standardized protocol with a 4-level grading system.  Readings were taken from the superior and inferior RNFL areas.  The OCT-measured RNFL thickness parameters were compared among normal eyes and diffuse atrophy subgroups.  Area under the ROCs (AROCs) was calculated for various OCT RNFL parameters.  Main outcome measures were average and segmental (4 quadrants and 12 clock-hours) OCT-measured RNFL thicknesses and AROCs for various OCT parameters.  For superior and inferior RNFL areas, diffuse atrophy grading by 2 observers agreed in 82.5 % and 83.3 % of cases, respectively, with a substantial agreement (kappa value = 0.760 [p < 0.001] and 0.777 [p < 0.001]).  Significant differences were observed in RNFL thickness among normal and all diffuse atrophy subgroups, especially in the 7 and 11 o'clock sectors (p < 0.0001).  The OCT RNFL thickness measurements decreased with increasing severity of RNFL damage.  The 7 and 11 o'clock sectors showed the highest AROCs for discrimination of mild RNFL atrophy from normal eyes (0.972 and 0.979, respectively).  The authors concluded that the OCT RNFL thickness parameters showed excellent quantitative correlation with the degree of diffuse RNFL atrophy.  These findings suggested that Stratus OCT may serve as a useful adjunct in accurately and objectively assessing the degree of diffuse RNFL atrophy.  Moreover, the authors noted that further studies are needed to assess the diagnostic ability of the Stratus OCT with its internal normative database to detect diffuse RNFL atrophy.

Cettomai and associates (2010) performed clinical and OCT examinations on 240 patients attending a neurology clinic.  Using OCT 5th percentile to define abnormal RNFL thickness, these investigators compared eyes classified by neurologists as having optic atrophy to RNFL thickness, and afferent pupillary defect (APD) to RNFL thickness ratios of eye pairs.  Mean RNFL thickness was less in eyes classified by neurologists as having optic atrophy (79.4 +/- 21 μm; n = 63) versus those without (97.0 +/- 15 μm; n = 417; p < 0.001, t-test) and in eyes with an APD (84.1 +/- 16 μm; n = 44) than without an APD (95.8 +/- 17 μm; n = 436; p < 0.001).  Physicians' diagnostic accuracy for detecting pallor in eyes with an abnormal RNFL thickness was 79 % (sensitivity = 0.56; specificity = 0.82).  Accuracy for detecting a retinal APD in patients with mean RNFL ratio (affected eye to unaffected eye) less than 0.90 was 73 % (sensitivity = 0.30; specificity = 0.86).  Ability to detect visual pathway injury via assessment of atrophy and APD differed between neurologists.  The authors concluded that OCT reveals RNFL abnormality in many patients in whom eyes are not classified by neurologic examiners as having optic atrophy.  They stated that further study is needed to define the role of OCT measures in the context of examinations for optic atrophy and APD by neuro-ophthalmologists.

Marmor et al (2011) stated that the AAO recommendations for screening of chloroquine (CQ) and hydroxychloroquine (HCQ) retinopathy were published in 2002, but improved screening tools and new knowledge about the prevalence of toxicity have appeared in the ensuing years.  No treatment exists as yet for this disorder, so it is imperative that patients and their physicians be aware of the best practices for minimizing toxic damage.  New data have shown that the risk of toxicity increases sharply toward 1 % after 5 to 7 years of use, or a cumulative dose of 1000 g of HCQ.  The risk increases further with continued use of the drug.  The prior recommendation emphasized dosing by weight.  However, most patients are routinely given 400 mg of HCQ daily (or 250 mg CQ).  This dose is now considered acceptable, except for individuals of short stature, for whom the dose should be determined on the basis of ideal body weight to avoid over-dosage.  A baseline examination is advised for patients starting these drugs to serve as a reference point and to rule out maculopathy, which might be a contraindication to their use.  Annual screening should begin after 5 years (or sooner if there are unusual risk factors).  Newer objective tests, such as multi-focal electroretinogram (mfERG), spectral domain-OCT(SD-OCT), and fundus auto-fluorescence (FAF), can be more sensitive than visual fields.  It is now recommended that along with 10-2 automated fields, at least one of these procedures be used for routine screening where available.  When fields are performed independently, even the most subtle 10-2 field changes should be taken seriously and are an indication for evaluation by objective testing.  Because mfERG testing is an objective test that evaluates function, it may be used in place of visual fields.  Amsler grid testing is no longer recommended.  Fundus examinations are advised for documentation, but visible bull's-eye maculopathy is a late change, and the goal of screening is to recognize toxicity at an earlier stage.  Patients should be aware of the risk of toxicity and the rationale for screening (to detect early changes and minimize visual loss, not necessarily to prevent it).  The drugs should be stopped if possible when toxicity is recognized or strongly suspected, but this is a decision to be made in conjunction with patients and their medical physicians.

Scanning Computerized Ophthalmic Diagnostic Imaging (OCT) for Patients with Multiple Sclerosis

Saidha et al (2015) examined if atrophy of specific retinal layers and brain substructures are associated over time, in order to further validate the utility of OCT as an indicator of neuronal tissue damage in patients with MS.  Cirrus high-definition OCT (including automated macular segmentation) was performed in 107 MS patients biannually (median follow-up of 46 months).  Three-Tesla magnetic resonance imaging brain scans (including brain-substructure volumetrics) were performed annually.  Individual-specific rates of change in retinal and brain measures (estimated with linear regression) were correlated, adjusting for age, sex, disease duration, and optic neuritis (ON) history.  Rates of ganglion cell + inner plexiform layer (GCIP) and whole-brain (r = 0.45; p < 0.001), gray matter (GM; r = 0.37; p < 0.001), white matter (WM; r = 0.28; p = 0.007), and thalamic (r = 0.38; p < 0.001) atrophy were associated.  GCIP and whole-brain (as well as GM and WM) atrophy rates were more strongly associated in progressive MS (r = 0.67; p < 0.001) than relapsing-remitting MS (RRMS; r = 0.33; p = 0.007).  However, correlation between rates of GCIP and whole-brain (and additionally GM and WM) atrophy in RRMS increased incrementally with step-wise refinement to exclude ON effects; excluding eyes and then patients (to account for a phenotype effect), the correlation increased to 0.45 and 0.60, respectively, consistent with effect modification.  In RRMS, lesion accumulation rate was associated with GCIP (r = −0.30; p = 0.02) and inner nuclear layer (r = −0.25; p = 0.04) atrophy rates.  The authors concluded that over time GCIP atrophy appeared to mirror whole-brain, and particularly GM, atrophy, especially in progressive MS, thereby reflecting underlying disease progression.  They stated that these findings supported OCT for clinical monitoring and as an outcome in investigative trials.

The authors stated that the results of this study indicated that GCIP and brain atrophy in MS closely parallel each other over time, suggesting a role for OCT as a valuable biomarker not only for the purpose of tracking patients clinically, but also in clinical trials for objective investigation of putative neuro-protective and/or neuro-restorative therapies.  Although GCIP and brain atrophy are associated in RRMS (especially after refinement for ON history; a factor that should be borne in mind in the interpretation of GCIP measures longitudinally), the associations between GCIP and brain atrophy in progressive MS appeared to be exceptional.  These researchers noted that although their findings require independent verification, and should be replicated across larger MS cohorts.

Behbehani and colleagues (2017) stated that OCT with retinal segmentation analysis is used in assessing axonal loss and neuro-degeneration in MS by in-vivo imaging, delineation and quantification of retinal layers.  There is evidence of deep retinal involvement in MS beyond the inner retinal layers.  The ultra-structural retinal changes in MS in different MS phenotypes can reflect differences in the pathophysiologic mechanisms.  There is limited data on the pattern of deeper retinal layer involvement in progressive MS (PMS) versus relapsing remitting MS (RRMS).  In a cross-sectional study, these researchers compared the OCT segmentation analysis in patients with RRMS and PMS.  A total of 113 MS patients (226 eyes) (29 PMS, 84 RRMS) and 38 healthy controls (72 eyes) were included in this trial; SD-OCT using the macular cube acquisition protocol and segmentation of the retinal layers for quantifying the thicknesses of the retinal layers were carried out.  Segmentation of the retinal layers was performed utilizing Orion software for quantifying the thicknesses of individual retinal layers.  The retinal nerve finer layer (RNFL) (p = 0.023), the ganglion-cell/inner plexiform layer (GCIPL) (p = 0.006) and the outer plexiform layer (OPL) (p = 0.033) were significantly thinner in PMS compared to RRMS.  There was significant negative correlation between the outer nuclear layer (ONL) and EDSS (r = -0.554, p = 0.02) in PMS patients.  In RRMS patients with prior optic neuritis, the GCIPL correlated negatively (r = -0.317; p = 0.046), while the photoreceptor layer (PR) correlated positively with EDSS (r = 0.478; p = 0.003).  The authors concluded that patients with PMS exhibited more atrophy of both the inner and outer retinal layers than RRMS.  The ONL in PMS and the GCIPL and PR in RRMS can serve as potential surrogate of disease burden and progression (EDSS).  The specific retinal layer predilection and its correlation with disability may reflect different pathophysiologic mechanisms and various stages of progression in MS.  Moreover, they stated that longitudinal studies using OCT segmentation analysis can better define the significance and the dynamics of the changes in retinal layers in different MS phenotypes and how they relate to disease progression.

The authors noted that this study was limited by its cross-sectional design and its relatively small sample size.  Most of the PMS cohort were composed of secondary progressive MS (SPMS) with under-representation of primary progressive MS (PPMS) due to rarity of the latter phenotype.  In addition, these investigators combined the primary and secondary progressive cohort as a single group, which may have influenced their findings.  However, there is increasing evidence of the phenotypic similarities between PPMS and SPMS and, common genetic susceptibility to MS are similar between and measures of global brain tissue damage and magnetization transfer imaging.  Despite that, the OCT findings in this study in the progressive cohort did not strictly apply to PPMS, which had its unique aspects of indolent course with less frequent visual pathway involvement and thus relative preservation of the inner retinal layers.

Scanning Computerized Ophthalmic Diagnostic Imaging (OCT) for Patients with Vogt-Koyanagi-Haradas

Sakata et al (2014) noted that Vogt-Koyanagi-Harada (VKH) disease is a systemic autoimmune disorder that affects pigmented tissues of the body, with its most dire manifestations affecting the eyes.  This review focused on the diagnostic criteria of VKH disease, including some information on history, epidemiology, appropriate clinical and classification criteria, etiopathogenesis, treatment and outcomes.  Expert review of most relevant literature from the disease's first description to 2013 and correlation with the experience in the care of VKH disease patients at a tertiary Uveitis Service in Brazil gathered over the past 40 years.  The clinical manifestations and ancillary assessment of VKH disease have been summarized in the Revised Diagnostic Criteria proposed in 2001 in a manner that allows systematic diagnosis of both acute and chronic patients.  It includes the early acute uveitic manifestations (bilateral diffuse choroiditis with bullous serous retinal detachment and optic disk hyperemia), the late ocular manifestations (diffuse fundus depigmentation, nummular depigmented scars, retinal pigment epithelium clumping and/or migration, recurrent or chronic anterior uveitis), besides the extra-ocular manifestations (neurological/auditory and integumentary).  There are 2 exclusion criteria, i.e., absence of previous ocular penetrating trauma or surgery and any other ocular disease that could be confounded with VKH disease.  HLA-DRB1*0405 plays an important role in pathogenesis, rendering carriers more susceptible to disease.  The primary ocular pathological feature is a diffuse thickening of the uveal tract in the acute phase.  Later on, there may be a compromise of choriocapillaris, retinal pigment epithelium and outer retina, mostly due to an "upstream" effect, with clinical correlates as fundus derangements.  Functional tests (ERG and visual field testing) as well as imaging modalities (retinography, fluorescein/indocyanine green angiography (FA/ICGA), OCT and ultrasound) play an important role in diagnosis, severity grading as well as disease monitoring.  Though high-dose systemic corticosteroids remain gold-standard therapy, refractory cases may need other agents (cyclosporine A, anti-metabolites and biological agents).  In spite of good visual outcomes in the majority of patients, knowledge about disease progression even after the acute phase and its impact on visual function warrant further investigation.

Komuku et al (2015) stated that VKH disease and central serous chorioretinopathy (CSCR) develop serous retinal detachment; however, the treatment of each disease is totally different.  Steroids treat VKH but worsen CSC; therefore, it is important to distinguish these diseases.  These investigators reported a case with CSCR, which was diagnosed by en face OCT imaging during the course of VKH disease.  A 50-year old man was referred with blurring of vision in his right eye.  Fundus examination showed bilateral optic disc swelling and macular fluid in the right eye; OCT showed thick choroid, and en face OCT images depicted blurry choroid without clear delineation of choroidal vessels.  Combined with angiography findings, this patient was diagnosed with VKH disease and treated with steroids.  Promptly, fundus abnormalities resolved with the reduction of the choroidal thickness and the choroidal vessels became visible on the en face images.  During the tapering of the steroid, serous macular detachment in the right eye recurred several times.  Steroid treatment was effective at first; however, at the 4th appearance of sub-macular fluid, the patient did not respond.  At that time, the choroidal vessels on the en face OCT images were clear, which significantly differed from the images at the time of recurrence of VKH.  Angiography also suggested CSCR-like leakage.  The tapering of the steroids was effective in resolving the fluid.  Secondary CSCR may develop in the eye with VKH after steroid treatment.  The authors concluded that en face OCT observation of the choroid may be helpful to distinguish each condition.

Tsuboi et al (2015) characterized patients with (VKH disease with choroidal folds (CFs) and determine how the foveal choroidal thickness changes after initial treatment using high-penetration OCT (HP-OCT).  In this retrospective observational study, these researchers analyzed 42 eyes of 21 patients with new-onset VKH disease to determine the demographic and clinical differences between patients with and without CFs; 24 eyes (57.1 %) of 13 patients with VKH disease had CFs.  The mean age (p = 0.0009) of patients with CFs was significantly higher than that of those without CFs (49.1 versus 39.4 years, respectively).  The frequency of disc swelling (p = 0.0001) was significantly higher in eyes with CFs than in those without CFs (95.8 % versus 38.9 %).  The choroidal thickness at the first visit (p = 0.0011) was significantly greater in eyes with CFs than in those without CFs (794 ± 144 μm versus 649 ± 113 μm).  The choroid 6 months after the initial treatment (p = 0.0118) was significantly thinner in eyes with CFs than in those without CFs (270 ± 92 μm versus 340 ± 80 μm).  The frequency of sunset glow fundus at 6 months (p = 0.0334) in eyes with CFs was significantly higher than in those without CFs (62.5 % versus 27.8 %).  The authors concluded that the development of CFs in patients with VKH disease was significantly correlated with age, disc swelling, and choroidal thickness.  The eyes with CFs frequently developed a sunset glow fundus.  They stated that these findings suggested that patients with CFs might have severe and longstanding inflammation of the choroidal tissues.

Hashizume et al (2016) determined the clinical significance of retinal pigment epithelium (RPE) undulations in the acute stage of VKH disease.  Retinal pigment epithelium undulations were detected and classified into 3 grades: Grade 1, slight; Grade 2, moderate; and Grade 3, severe undulations, in the EDI-OCT images.  The relationship between the clinical characteristics and the presence of RPE undulations was investigated.  Among the 61 eyes of 31 patients with VKH disease, 40 eyes had some degree of RPE undulations (Grade 1 = 12, Grade 2 = 15, and Grade 3 = 13).  The patients with RPE undulations in both eyes were significantly older at the onset (p = 0.0002).  The eyes with RPE undulations were more likely to develop posterior recurrences (p = 0.032) and have worse vision at 12 months (p = 0.043).  Multiple regression analysis revealed that RPE undulations were an independent predictor of posterior recurrences (p = 0.009) and poor visual outcomes (p = 0.035).  The authors concluded that retinal pigment epithelium undulations detected by EDI-OCT were relatively frequent occurrences at the acute stage of VKH, and their presence is a predictor of posterior recurrences and poor visual outcomes after high-dose steroid therapy.

Bae et al (2017) examined if the inflammatory composition of sub-retinal fluid in VKH serous retinal detachments is predictive of photoreceptor injury, and quantified photoreceptor recovery, following resolution of these detachments.  Optical density (OD) measurements of spectral-domain OCT (SD-OCT) scans were used to derive the fibrinous index, a measure of the inflammatory composition of sub-retinal fluid. In order to assess photoreceptor status, photoreceptor outer segment (PROS) volume was measured from SD-OCT scans.  The fibrinous index of sub-retinal fluid in VKH uveitis was strongly correlated with the PROS volume following resolution of sub-retinal fluid (r = -0.70, p = 0.006).  Following fluid resolution, both PROS volume (p < 0.0001) and visual acuity (p = 0.0015) improved.  The authors concluded that the fibrinous index of sub-retinal fluid during the acute stage of VKH can predict photoreceptor status following resolution of sub-retinal fluid; PROS volume is a useful measure of photoreceptor recovery in VKH.

Aggarwal et al (2018) reported the imaging characteristics of acute VKH disease using OCT angiography (OCTA).  In this prospective study, patients with acute VKH (n = 10; mean age of 30.5 ± 13.43 years) underwent multi-modal imaging (baseline and follow-up) using fundus photography, FA, ICGA, OCT, and OCTA.  The OCTA images were analyzed to assess the retino-choroidal vasculature and compared with other imaging techniques.  During the active stage, all eyes showed multiple foci of choriocapillaris flow void that correlated with ICGA.  These foci decreased in number and size after initiation of therapy.  In 1 patient, flow void areas re-appeared after cessation of therapy without any detectable change on ICGA.  This patient soon developed clinical recurrence requiring re-initiation of immunosuppression.  The authors concluded that OCTA allowed high-resolution imaging of inflammatory foci suggestive of choriocapillaris hypo-perfusion in acute VKH disease non-invasively.  They stated that OCTA may be very helpful in the follow-up of such patients.

Liu et al (2016) examined the diagnostic value of OCT for the detection of acute VKH disease.  Clinical charts and OCT images were retrospectively reviewed for patients consecutively diagnosed with acute VKH, sub-acute VKH, multi-focal CSCR, and posterior scleritis.  All patients underwent OCT, fundus photography, and FA before treatment.  The characteristics of OCT and FA were analyzed and recorded.  The study included 80 eyes with acute VKH, 32 eyes with sub-acute VKH, 33 eyes with CSCR, and 13 eyes with posterior scleritis.  The most common OCT features of VKH disease were hyper-reflective dots (70/80; 88 %), sub-retinal membranous structures (64/80; 80 %), retinal detachment higher than 450 μm (63/80; 79 %), and retinal pigment epithelium (RPE) folds (44/80; 55 %).  For the detection of VKH disease, sensitivity and specificity were for sub-retinal membranous structures 80 % and 95.6 %, respectively, for high retinal detachment 78.8 % and 76.1 %, respectively, for sub-retinal hyper-reflective dots, 87.5 and 60.9 %, respectively, and for RPE folds 55 % and 80.4 % respectively.  Sub-retinal membranous structures showed the highest positive predictive value (97.3 %) and negative predictive value (65.7 %) of all OCT assessed features.  The authors concluded that OCT-related morphological signs had a relatively high predictive value for the diagnosis of acute VKH.

Chee et al (2017) compared EDI-OCT and swept source OCT (SS-OCT) in assessment of VKH disease.  All consecutive VKH patients seen at Singapore National Eye Centre during 2012 to 2013 were imaged using both modalities.  Sub-foveal choroidal thickness (SFCT) was measured by one masked trained observer.  A total of 137 pairs of scans were obtained from 48 patients. SFCT was more likely to be measurable on SS-OCT than EDI-OCT (112, 81.8 %; 84, 61.3 %; p < 0.001 Fisher's Exact test).  There was good inter-OCT correlation of SFCT when both scans were measurable (mean of the difference in SFCT ± 2 standard deviations (SD) of -14.5 ± 21.0 μm).  The authors concluded that SS-OCT images were superior to EDI-OCT; but the SFCT measurements are comparable when both are readable.

Furthermore, the American Academy of Ophthalmology (2016) states that “The diagnosis of VKH syndrome is essentially clinical; exudative retinal detachment during the acute disease and sunset glow fundus during the chronic phase are highly specific to this entity.  In patients presenting without extra-ocular changes, FA, ICG angiography, OCT, FAF imaging, lumbar puncture, and ultrasonography may be useful confirmatory tests.  During the acute uveitic stage, FA typically reveals numerous punctate hyper-fluorescent foci at the level of the RPE in the early stage of the study followed by pooling of dye in the sub-retinal space in areas of neurosensory detachment.  The vast majority of patients show disc leakage, but CME and retinal vascular leakage are uncommon.  In the convalescent and chronic recurrent stages, focal RPE loss and atrophy produce multiple hyper-fluorescent window defects without progressive staining … OCT may be useful in the diagnosis and monitoring of serous macular detachments, CME, and choroidal neovascular membranes.  More recently, the combined use of FAF imaging and SD-OCT offers a non-invasive assessment of RPE and outer retina changes in patients with chronic VKH syndrome that may not be apparent on clinical examination”. 

Frontotemporal Degeneration

Kim and colleagues (2017) noted that whereas Alzheimer disease (AD) is associated with inner retina thinning visualized by SD-OCT, these researchers sought to determine if the retina has a distinguishing biomarker for frontotemporal degeneration (FTD).  Using a cross-sectional design, these investigators examined retinal structure in 38 consecutively enrolled patients with FTD and 44 controls using a standard SD-OCT protocol.  Retinal layers were segmented with the Iowa Reference Algorithm.  Subgroups of highly predictive molecular pathology (tauopathy, TAR DNA-binding protein 43, unknown) were determined by clinical criteria, genetic markers, and a CSF biomarker (total tau: β-amyloid) to exclude presumed AD.  These researchers excluded eyes with poor image quality or confounding diseases; SD-OCT measures of patients (n = 46 eyes) and controls (n = 69 eyes) were compared using a generalized linear model accounting for inter-eye correlation, and correlations between retinal layer thicknesses and Mini-Mental State Examination (MMSE) were evaluated.  Adjusting for age, sex, and race, patients with FTD had a thinner outer retina than controls (132 versus 142 μm, p = 0.004).  Patients with FTD also had a thinner outer nuclear layer (ONL) (88.5 versus 97.9 μm, p = 0.003) and ellipsoid zone (EZ) (14.5 versus 15.1 μm, p = 0.009) than controls, but had similar thicknesses for inner retinal layers.  The outer retina thickness of patients correlated with MMSE (Spearman r = 0.44, p = 0.03).  The highly predictive tauopathy subgroup (n = 31 eyes) also had a thinner ONL (88.7 versus 97.4 μm, p = 0.01) and EZ (14.4 versus 15.1 μm, p = 0.01) than controls.  The authors concluded that FTD was associated with outer retina thinning, and this thinning correlated with disease severity.

The authors stated that one drawback of this study was the different demographics of controls and patients.  While all patients were recruited consecutively, differences reflected the different populations of FTD versus controls recruited during a routine eye examination.  Another drawback of these findings was the limited number of patients in the non-tauopathy subgroups; this must be considered before generalizing the results to all patients with FTD.  These investigators stated that the findings of this study suggested that measurements of retinal thickness have the potential to serve as biomarkers for FTD and may relate to disease severity; future work should focus on direct comparison of AD patients with FTD patients and comparison of the different subgroups of FTD using similar methods and longitudinal studies with autopsy confirmation.

Diagnosis of Optic Neuritis

In a retrospective, observational study, Xu and colleagues (2019) examined the sensitivity of OCT in detecting prior unilateral optic neuritis.  Patients who presented from January 1, 2014, to January 6, 2017, with unilateral optic neuritis and OCT available at least 3 months after the attack were enrolled in this trial.  These investigators compared OCT RNFL and GCIPL thicknesses between affected and unaffected contralateral eyes.  They excluded patients with concomitant glaucoma or other optic neuropathies.  Based on analysis of normal controls, thinning was considered significant if RNFL was at least 9 µm or GCIPL was at least 6 µm less in the affected eye compared to the unaffected eye.  A total of 51 patients (18 male and 33 female) were included in the study; RNFL and GCIPL thicknesses were significantly lower in eyes with optic neuritis compared to unaffected eyes (p < 0.001); RNFL was thinner by greater than or equal to 9 µm in 73 % of optic neuritis eyes compared to the unaffected eye; GCIPL was thinner by greater than or equal to 6 µm in 96 % of optic neuritis eyes, which was more sensitive than using RNFL (p < 0.001).  When using a threshold less than or equal to 1st percentile of age-matched controls, sensitivities were 37 % for RNFL and 76 % for GCIPL, each of which was lower than those calculated using the inter-eye difference as the threshold (p < 0.01).  The authors concluded that these findings supported the use of OCT in the diagnosis of prior optic neuritis, especially in those with unilateral presentation.  There were no patients who had optic neuritis with complaints of vision loss who did not have thinning of the GCIPL on OCT.  These researchers stated that although larger prospective studies are needed to confirm the optimal criteria for identifying pathologic thinning of the inner retina by OCT, it is a highly sensitive method of detecting a history of unilateral optic neuritis.  This study provided Class III evidence that OCT accurately identified patients with prior unilateral optic neuritis.

The authors stated that this study had several drawbacks.  Because this patient cohort consisted of unilateral optic neuritis, these findings were not directly applicable to patients with bilateral optic neuritis or patients with prior episodes of optic neuritis in the concomitant eye.  This study excluded any patients with optic nerve pathology in the fellow eye.  In patients with bilateral optic neuritis or any other optic neuropathy in the fellow eye, the 1st percentile threshold may be a more sensitive method than the inter-eye difference threshold for detecting optic neuritis, given that inter-eye difference decreases with any bilateral process.  Furthermore, depending on the provider's preference or patient's schedule, not all unilateral chronic optic neuritis patients seen during the time period of the study had OCT data (3/54 patients or 5.6 % were excluded due to this).  Thus, this could have introduced sampling bias in this retrospective, observational study.  The follow-up period for OCT was variable and was as short as 3 months.  This could have contributed to the decreased sensitivity in RNFL because continued thinning is expected for at least 6 months after an initial optic neuritis.  Nonetheless, these investigators found similar sensitivities when they examined a subset of patients who had follow-up OCT images beyond 6 months and, subsequently, the sensitivities were valid in this cohort.  Another drawback of the study was that patients with optic neuritis had a variety of causes, including MS, neuromyelitis optica spectrum disorder (NMOSD), and idiopathic. The cause of optic neuritis can influence the expected degree of RNFL and GCIPL thinning.  However, even after excluding patients with myelin oligodendrocyte glycoprotein (MOG)-IgG and aquaporin-4 (AQP4)-IgG-NMOSD associated optic neuritis, the sensitivity using the proposed 99th percentile cut-offs for inter-eye variability was still 70 % and 96 % for RNFL and GCIPL, respectively; therefore, OCT remained sensitive for detecting prior optic neuritis for typical demyelinating optic neuritis patients.  Lastly, the control group had a higher percentage of male participants than this cohort of optic neuritis.  However, male and female participants had the same inter-eye RNFL or GCIPL difference in both the control group and the cohort of patients with optic neuritis (unpublished data).  Although traumatic brain injury (TBI) could be a potential confounder in OCT measurements, the control cohort excluded any veterans with TBI.  Also, no difference between the 1st-season OCT measurements of the football players and track team players was found in the healthy controls.

In an editorial that accompanied the afore-mentioned study, Saidha and Naismith (2019) states that “The current study is an excellent start, but more work is required before fully recommending OCT as a routine tool for diagnosing ON and subclinical optic neuropathy.  Larger studies should be performed within specific disease states such as MS, with subset analyses to evaluate patients at older age or many years from their suspected demyelinating event.  Longitudinal studies can help clarify whether inter-eye differences change during the course of disease.  The identification of inter-eye asymmetry in an individual patient may be an uncertain basis upon which to conclude that there is definitive evidence of prior ON, especially if asymmetry in both measures are incongruent (e.g., fulfilled by RNFL but not GCIPL).  Despite the limitations noted, and the need for further, larger, longitudinal studies, the results of this study are promising.  Potentially, OCT could fulfill multiple roles towards diagnosis, prognosis, and treatment monitoring in ON, MS, and related disorders”.

Evaluation and Screening for Ethambutol Toxicity

The American Academy of Ophthalmology’s guideline on “Drug-related adverse effects of clinical importance to the ophthalmologist” (Fraunfelder, 2014) stated that “Ethambutol optic neuropathy is usually retrobulbar and bilateral, though sometimes asymmetric.  Ethambutol toxicity may affect only the small caliber papillo-macular bundle axons, which are hard to visualize, and optic atrophy will not develop for months after the fibers are lost.  This means objective findings on the fundus exam are frequently unrecognized.  Optic neuropathy may occur, on average, at 2 to 5 months after starting therapy.  The earliest ophthalmologic findings in toxic optic neuropathy from ethambutol may be loss of visual acuity, color vision loss or central scotomas.  Ethambutol also has an affinity for the optic chiasm with bi-temporal visual field defects manifesting with toxicity … Consider optical coherence tomography or contrast sensitivity testing as these tests could pick up early ethambutol toxicity not detected with the baseline examination. Optical coherence tomography (OCT) may be the future for following toxic optic neuropathies as subtle retinal nerve fiber layer (NFL) swellings can be visualized with the acute insult and NFL thinning can be visualized from chronic toxicity”.

Furthermore, the Royal College of Ophthalmologists’ RCOphth statement on ethambutol toxicity (2017) stated that “Ethambutol is an effective antibiotic used to treat tuberculosis but optic neuropathy is a potentially serious side effect of the drug, thought to be due to zinc chelation causing mitochondrial dysfunction.  Ethambutol toxicity in adults is rare, occurring in less than 2 % of patients on the standard dosage of 15 mg/kg/day, but impaired renal function and smoking may increase the risk.  Onset of optic neuropathy is typically 2to 5 months after starting therapy, but may occur within days.  Symptoms can be highly variable and may initially be unilateral.  Loss of visual acuity, color vision impairment and central/paracentral scotomata may occur; bi-temporal field defects have also been reported due to an affinity of ethambutol for the chiasm.  Although optic atrophy will subsequently develop, signs may be absent in early stages of toxicity, but visual evoked potentials and optical coherence tomography show promise in detecting subclinical optic neuropathy”.

In a review on “Ethambutol optic neuropathy”, Chamberlaina and colleagues (2017) provided a summary of the epidemiology, clinical findings, management and outcomes of ethambutol-induced optic neuropathy (EON).  Ethambutol-induced optic neuropathy is a well-known, potentially irreversible, blinding but largely preventable disease.  Clinicians should be aware of the importance of patient and physician education as well as timely and appropriate screening.  Two of the largest epidemiologic studies investigating EON to-date showed the prevalence of EON in all patients taking ethambutol to be between 0.7 and 1.29 %, a value consistent with previous reports of patients taking the doses recommended by the World Health Organization (WHO).  Several studies evaluated the utility of OCT in screening for EON.  These showed decreased RNFL thickness in patients with clinically significant EON, but mixed results in their ability to detect such changes in patients taking ethambutol without visual symptoms.  The authors concluded that ethambutol-induced optic neuropathy is a well-known and devastating complication of ethambutol therapy.  It may occur in approximately 1 % of patients taking ethambutol at the WHO recommended doses, although the risk increases substantially with increased dose.  All patients on ethambutol should receive regular screening by an ophthalmologist including formal visual field testing.  Visual evoked potentials and OCT may be helpful for EON screening, but more research is needed to clarify their clinical usefulness.  Patients who develop signs or symptoms of EON should be referred to the ethambutol-prescribing physician immediately for discontinuation or a reduction in ethambutol dosing.

Evaluation of the Neurodegeneration Pattern in Individuals with Intra-Cranial Tumors

Banc and colleagues (2018) noted that OCT is a non-invasive, high-resolution imaging technique that was suggested to be a powerful biomarker of neurodegeneration.  These researchers examined the pattern of retinal OCT changes in patients with visual pathway tumors.  A prospective clinical study was conducted and patients with single cerebral tumors with potential of compression on the visual pathway were included.  Patients with multiple and/or metastatic tumors were excluded.  Each patient underwent a neurosurgical and ophthalmologic evaluation, cranial-cerebral MRI, and ocular OCT in both eyes.  The OCT parameters included circum-papillary RNFL thickness (average and sector thickness) and retinal thickness in the macular area (average and sector thickness).  A total of 50 patients were examined clinically and by MRI, and 18 patients were excluded; 32 patients were eligible for the study and completed the retinal OCT; 18 patients had tumors with compressive potential on the optic chiasm, 11 patients had tumors close to the optic radiations, and 3 patients had tumors in the occipital lobe.  A specific pattern of OCT changes was found for each site.  Regional parameters of both optic nerve and macula were altered.  The authors concluded that retinal OCT is a promising tool for the in-vivo assessment of the neurodegeneration pattern in patients with intra-cranial tumors.  They stated that the evaluation of single intra-cranial tumors with compressive potential on the visual pathway is a good candidate for the study of neurodegeneration.

Evaluation of Parinaud Oculo-Glandular Syndrome (Cat Scratch Disease)

Perez and colleagues (2010) noted that cat scratch disease (CSD) is the main clinical presentation of Bartonella henselae infection.  However, ocular manifestations of bartonellosis occur in about 5 to 10 % of the patients, mainly presenting as neuro-retinitis, choroiditis or oculo-glandular syndrome of Parinaud.  The authors described 2 patients with documented B. henselae infection and typical ocular compromise.  Both patients were treated and had a favorable visual outcome.

An UpToDate review on “Microbiology, epidemiology, clinical manifestations, and diagnosis of cat scratch disease” (Spach and Kaplan, 2019) states that “Parinaud oculo-glandular syndrome is an atypical form of CSD, which is reported in 2 to 8 % of patients with CSD.  Parinaud oculo-glandular syndrome is characterized by tender regional lymphadenopathy of the preauricular, submandibular, or cervical lymph nodes associated with infection of the conjunctiva, eyelid, or adjacent skin surface.  Usual complaints include unilateral red eye, foreign body sensation, and excessive watering of the eyes.  Discharge may be serous or purulent and copious in some patients.  The inoculation of the organism occurs via a cat bite or lick near (or in) the eye, as well as by self-inoculation from another site … Some patients develop a stellate macular exudate (known as a "macular star").  Macular stars are due to vascular leakage from the optic nerve head, and can be seen on fluorescein angiography or optical coherence tomography angiography.  Patients with B. henselae-induced neuro-retinitis may not develop a macular star until 1 to 4 weeks after initial presentation, and the exudate may persist for months, despite resolution of the neuro-retinitis”.  However, there is no mentioning of OCT in the “Summary and Recommendations” of this review.

Monitoring of Plaquenil (Hydroxychloroquine) Toxicity

In a retrospective, observational cohort study, Browning (2013) determined the impact of the revised American Academy of Ophthalmology (AAO) guidelines on screening for hydroxychloroquine retinopathy.  The setting was a private practice of 29 doctors; study population entailed a total of 183 patients for follow-up and 36 patients for baseline screening.  Review of charts, 10-2 visual fields (VFs), multi-focal electroretinograms (mfERG), and spectral-domain optical coherence tomography (SD-OCT) images before and after the revised guidelines.  Main outcome measure was rates of use of ancillary tests and clinical intervention, costs of screening, follow-up schedules, and comparative sensitivity of tests.  New hydroxychloroquine toxicity was found in 2 of 183 returning patients (1.1 %).  Dosing above 6.5 mg/kg/day was found in 28 of 219 patients (12.8 %), an under-estimate because patient height, weight, and daily dose were not determined in 77 (35.1 %), 84 (38.4 %), and 59 (26.9 %), respectively.  In 10 of the 28 (35.7 %), the dose was reduced, in 2 (7.1 %) hydroxychloroquine was stopped, but in 16 (57.1 %) no action was taken.  The cost of screening rose 40 %/patient after the revised guidelines.  Fundus auto-fluorescence (FAF) imaging was not used.  No toxicity was detected by adding mfERG or SD-OCT.  In no case was a 5-year period free of follow-up recommended after baseline screening in a low-risk patient.  The author concluded that detection of toxic daily dosing was a cost-effective way to reduce hydroxychloroquine toxicity, but height, weight, and daily dose were commonly not checked.  The revised guidelines, emphasizing mfERG, SD-OCT, or FAF, raised screening cost without improving case detection.  The recommended 5-year screening-free interval for low-risk patients after baseline examination was ignored.

In a retrospective, observational, case-series study, Leung et al (2015) reported rapid onset of retinal toxicity in a series of patients followed on high-dose (1,000 mg daily) hydroxychloroquine during an oncologic clinical trial studying hydroxychloroquine with erlotinib for non-small cell lung cancer (NSCLC).  Ophthalmic surveillance was performed on patients in a multi-center clinical trial testing high-dose (1,000 mg daily) hydroxychloroquine for advanced NSCLC.  The Food & Drug Administration (FDA)-recommended screening protocol included only visual acuity testing, dilated fundus examination, Amsler grid testing, and color vision testing.  In patients seen at Stanford, additional sensitive screening procedures were added at the discretion of the retinal physician: high-resolution SD-OCT, FAF imaging, Humphrey visual field (HVF) testing, and mfERG.  Out of the 7 patients having exposure of at least 6 months, 2 developed retinal toxicity (at 11 and 17 months of exposure).  Damage was identified by OCT imaging, mfERG testing, and, in 1 case, visual field testing.  Fundus auto-fluorescence imaging remained normal.  Neither patient had symptomatic visual acuity loss.  The authors concluded that these cases showed that high doses of hydroxychloroquine could initiate the development of retinal toxicity within 1 to 2 years.  Although synergy with erlotinib is theoretically possible, there are no prior reports of erlotinib-associated retinal toxicity despite over a decade of use in oncology.  These results also suggested that sensitive retinal screening tests should be added to ongoing and future clinical trials involving high-dose hydroxychloroquine to improve safety monitoring and preservation of vision.

In a case-series study, Latasiewicz et al (2017) raised awareness of the emerging issue of serious retinal damage caused by the prolonged use of hydroxychloroquine (HCQ) and the importance of adequate and appropriate monitoring of visual function during treatment.  This was a small retrospective case series of 3 patients on long-term HCQ who developed serious symptomatic retinal toxicity confirmed on imaging and functional testing.  All 3 patients were treated with HCQ for over 15 years; 2 for rheumatoid arthritis (RA), and the 3rd for systemic lupus erythematosus (SLE).  All 3 patients had macular involvement varying in severity confirmed with characteristic features on imaging and functional testing (OCT, AF and Humphrey 10-2 visual fields).  The authors concluded that HCQ is widely used to treat autoimmune conditions with a proven survival benefit in patients with SLE.  However, long-term use can be associated with irreversible retinal toxicity.  These cases highlighted that HCQ, like chloroquine, could also cause visual loss in susceptible individuals.  These researchers stated that early detection of pre-symptomatic retinal changes by the introduction of appropriate screening and monitoring is mandatory to limit the extent of irreversible visual loss due to HCQ retinal toxicity.

Kowalski et al (2018) reported the findings of 2 patients with dermatological conditions who developed retinal toxicity after treatment with HCQ that exceeded dosing recommendations.  There was no treatment for HCQ retinal toxicity and associated visual loss, so appropriate monitoring is imperative.  All members of a patient's multi-disciplinary team should be aware of the ocular risks of HCQ, the importance of dosing within recommended guidelines and appropriate monitoring in reducing the risk of visual loss.

Furthermore, an UpToDate review on “Antimalarial drugs in the treatment of rheumatic disease” (Wallace, 2020) states that “We advise assessment of ocular health within 1 year of starting long-term antimalarial drug therapy.  The baseline examination should include a fundus examination of the macula to rule out any underlying disease that may interfere with the interpretation of screening tests.  The frequency of subsequent screening during the first 5 years of treatment may be individualized based upon assessment of risk.  We prefer annual screening exams for all patients, but the AAO has suggested that for patients with a normal baseline exam who do not have major risk factors for toxic retinopathy, follow-up examinations may be deferred until there have been 5 years of exposure .  Major risk factors for toxic retinopathy include a daily dose of HCQ greater than 5 mg/kg real body weight or a daily dose of chloroquine greater than 2.3 mg/kg real body weight, antimalarial use for greater than 5 years, the presence of renal disease, concomitant tamoxifen use, and/or the presence of macular disease.  Patients should be alert for any change in visual acuity and should seek medical attention promptly if any visual loss is noted.  Anti-malarials should be discontinued immediately if there is any suspicion of retinopathy”.

Screening and Monitoring of Ethambutol (Myambutol) Toxicity

Menon et al (2009) evaluated various visual parameters for early detection of ethambutol toxicity.  This was a prospective study of 104 eyes of 52 patients being treated with ethambutol in the Directly Observed Treatment Strategy Centre (Dr R P Centre for Ophthalmic Sciences, New Delhi, India).  Visual acuity (VA), visual fields, visual evoked responses (VER), stereo-acuity and retinal nerve fiber layer (RNFL) thickness on optical coherence tomography (OCT) were assessed.  Examinations were done before the start of therapy, after 1 and 2 months of treatment, and 1 month after stopping ethambutol.  No visual functional defect was noted at baseline.  On follow-up, VA, color vision, contrast sensitivity, fundus and stereo-acuity were not affected in any patient.  Visual field defects developed in 7.69 % (8/104) of the eyes.  Pattern-VER showed an increased mean latency of the P(100) wave after 1 and 2 months of therapy (p < 0.001 for both) with 14.42 % (15/104) of eyes showing more than 10 ms increase in latency.  On OCT, significant loss of mean temporal RNFL thickness was detected in 2.88 % (3/104) of eyes individually.  Overall, 19.23 % (20/104) of the studied eyes showed sub-clinical toxicity.  Reversal of this observed toxicity on pattern-VER and visual fields was observed in 80 % of eyes after 1 month of stoppage of ethambutol; however, mean VER latency remained delayed (p = 0.002).  The authors concluded that pattern-VER and visual field examinations were sensitive tests to detect early toxicity.  Together with OCT, they may help to identify patients who are likely to develop clinical toxicity.

Gumus and Oner (2015) examined the effect of anti-tubercular treatment on RNFL thickness and the efficiency of OCT on early diagnosis of optic neuropathy.  A total of 20 patients diagnosed with either pulmonary or extra-pulmonary tuberculosis that were treated with anti-tubercular treatment (isoniazid (INH), rifampicin, ethambutol (ETM), and pyrazinamide) were enrolled in the study.  RNFL thicknesses of the patients were measured via OCT, at baseline (before starting anti-tubercular treatment) and after the 2-month treatment period.  Standard ophthalmologic examinations were also performed.  Compared to baseline values, after the 2-month treatment period, thinning was detected in the right eye's average and superior quadrant RNFLs (p = 0.024 and p = 0.006 respectively) and in the left eye's average, superior quadrant, and inferior quadrant RNFLs (p = 0.001, p = 0.008, p < 0.001, respectively).  The authors reported that patients receiving INH and ETM, which were the basic medicines of anti-tubercular treatment, experienced thinning in RNFL after the 2-month treatment period.  These researchers stated that patients receiving these drugs can be followed via OCT in terms of reduction in RNFL thicknesses for early diagnose of INH and ETM toxicity.

Kim and Park (2016) noted that tuberculosis in developed countries is on the rise, and the main treatment ethambutol is known to induce ocular toxicity.  However, to-date, there are unknown tests or protocols for detecting sub-clinical ethambutol-induced ocular toxicity, which is important as early detection is related to symptom reversibility.  These researchers defined ethambutol-induced ocular toxicity as statistically significant change of visual function that was induced by ethambutol.  They identified a visual function test for the early detection of sub-clinical ethambutol-induced ocular toxicity.  Furthermore, these investigators examined the continuity or reversibility of early sub-clinical changes that were observed during the visual function tests after stopping ethambutol treatment.  The age range of 31 patients was from 13 to 72 years.  The range of dosage was 15 to 19 mg/kg/day.  The average period of dosage was 5 months.  These researchers performed a VA test, visual field test, color vision test, contrast sensitivity test, fundus examination, RNFL OCT per month and pattern visual evoked potential test (pattern VEP) every 2 months before and during ethambutol treatment in 62 eyes of 31 patients.  Among these patients, selected 21 patients were re-examined by these tests at the 3, 6 and 12 months after stopping ethambutol treatment.  These investigators compared the test results from the last follow-up during ethambutol treatment and after ethambutol stoppage with those obtained before ethambutol treatment (baseline).  RNFL OCT showed that average RNFL thickness increased 5 months after ethambutol treatment (p = 0.032), and pattern VEP showed that P100 latency was delayed in 2 and 4 months after ethambutol treatment (p = 0.001; p < 0.001, respectively).  These early changes observed on RNFL OCT and pattern VEP progressed 6 months after ethambutol stoppage in 21 patients.  Twelve months after ethambutol stoppage, these early changes returned to baseline levels.  During the study, no changes in VA, color vision, fundus, contrast sensitivity or visual field were observed.  The authors concluded that pattern VEP and RNFL OCT were suitable tests for the early detection of sub-clinical ethambutol-induced ocular toxicity.  These tests should be performed until 12 months after ethambutol stoppage.

Pavan Taffner et al (2018) evaluated, through OCT, alterations in retinal thickness, secondary to use of ethambutol in the treatment of patients with tuberculosis, in addition to studying the use of simpler semiological tools, such as Amsler and Ishihara, in the screening of these cases.  A total of 30 patients with ethambutol were recruited from the reference service of tuberculosis treatment at the Federal University of Espírito Santo from May 2015 to July 2016.  After clinical history, the following parameters were analyzed; best corrected visual acuity (BCVA), biomicroscopy, tonometry, photo-motor reflex testing, Ishihara test, Amsler's grid test, color digital retinography and OCT with CIRRUS HD-OCT (Humphrey-Zeiss) every 2 months during treatment with ethambutol.  They were divided into 2 groups according to the treatment: standard group, 2 months of ethambutol; extended group, 9 to 12 months of ethambutol.  There was a significant reduction in OCT thickness between the pre- and post-treatment times in 10 eyes of the extended group, mean reduction of 7.8 microns and in 7 eyes of the standard group, with an average of 5.57 microns.  During the study, a significant reduction of retinal thickness was observed in both groups at 2 months of treatment, and the delta percentage was higher in those patients who presented reduction of VA and / or change in the Ishihara test.  The authors concluded that there was a significant reduction in the thickness of the nerve fiber layer by OCT in the patients studied, being more pronounced in those submitted to the extended treatment regimen.  This reduction was observed 2 months after the start of therapy, and was more significant in the cases that presented changes in the Ishihara test.  Moreover, these researchers stated that further studies are needed to elucidate the risk factors and intervals required between OCT screening tests for early signs of ethambutol optic neuropathy.

Jin et al (2019) longitudinally evaluated the visual function and structure of patients taking ethambutol by various modalities and identified useful tests for detection of sub-clinical ethambutol-induced optic toxicity.  This retrospective study enrolled 84 patients with newly diagnosed tuberculosis treated with ethambutol; BCVA, color vision, contrast sensitivity, fundus and RNFL photography, automated visual field (VF) test, and OCT were performed: prior to starting; every month during administration, and 1 month after stoppage.  These researchers longitudinally compared visual function and structure with the baseline and identified the occurrence of sub-clinical toxicity.  BCVA, color vision, and contrast sensitivity showed no change from the baseline.  Mean temporal RNFL thickness was significantly increased at 6 months (p = 0.014).  Sub-clinical toxicity was found in 22 eyes of 14 patients (i.e., 13 % of 168 eyes), in the forms of VFI decrease (VF index, 9 eyes of 6 patients), quadrant RNFL thickness increase (5 eyes of 4 patients), and VF pattern defect (12 eyes of 6 patients); 73 % of the patients showed recovery to the baseline at 1 month post-stoppage.  The risk factors for occurrence of sub-clinical toxicity were age, cumulative dose, and medication duration.  The authors concluded that mean temporal RNFL thickness increased after administration.  The VFI, quadrant RNFL thickness, and VF pattern defect could prove useful in assessment of sub-clinical toxicity; medication duration was shown to be a strong risk factor for occurrence of sub-clinical toxicity.

These investigators noted that in this study, the incidence of clinical toxicity could not be rated, because no patient complained of clinical symptoms.  However, they could assume that the incidence would be lower than 1.2 % (1/84), which is compatible with the results of previous studies.  The authors stated that this study had several drawbacks.  First, none of the participants experienced clinical symptoms, and therefore, incidence of clinical optic neuropathy after sub-clinical change could not be ruled out.  They did not stop administration of ethambutol in the sub-clinical cases, and none of these patients developed clinical ethambutol-induced optic neuropathy until 1 month after administration.  Thus, the implications of sub-clinical ethambutol-induced toxicity for actual occurrence of clinical toxicity remain to be elucidated in another long-term, prospective studies.  Second, these researchers could not evaluate the patients for a sufficient span of time after stoppage of drug administration.  Given the retrospective study design, they were unable to control the follow-up visitation, and so 50 % of subjects with sub-clinical changes failed to visit after stoppage of administration (7 of 14 patients), and these investigators were also were unable to collect data beyond 1 month after stoppage.  In reversible cases, the resolution of ethambutol-induced optic toxicity typically occurred 3 months after cessation.  With a longer follow-up period, sub-clinical changes in VF pattern and RNFL thickness, which remained at 1 month after stoppage in this study, might have been shown to have recovered to the baseline.  Furthermore, although GCC analysis was possible with up-graded Cirrus HD-OCT software, the up-graded software was not available at the authors’ institute at the time of the study.  Changes in GCC thickness might be more dramatic than changes in RNFL thickness, but they might also be less specific, as they involved 3 different innermost retinal layers instead of just one.  These researchers stated that future study including foveal GCC or GC-IPL should be conducted with more advanced modalities.  Finally, the concurrent effect of isoniazid could not be ruled out.

Furthermore, an UpToDate review on “Ethambutol: An overview” (Drew, 2020) states that “Monitoring -- It is generally recommended that patients receiving ethambutol as part of combination therapy for treatment of a mycobacterial infection undergo baseline Snellen visual acuity and red-green color perception testing.  All patients should be advised of the side effects associated with ethambutol, most notably those associated with the development of optic neuritis.  The need for routine periodic visual acuity testing during therapy is controversial, especially if a dose of 15 mg/kg is chosen, but patients noting changes in their vision should be referred to an ophthalmologist for careful monitoring.  In all patients receiving combination therapy for tuberculosis or MAC infections, baseline laboratory studies should be obtained and repeated in the event of suspected drug-related toxicity.  Although serum concentration monitoring of ethambutol is not routinely performed, it may be useful in cases of severe renal insufficiency or suspected malabsorption (as demonstrated in some HIV-infected patients receiving anti-tuberculous therapy).  If serum drug concentration monitoring is performed, the proposed therapeutic range 2 hours post-dose is 2 to 6 mcg/mL”.

Screening and Monitoring of Ponatinib (Iclusig) Toxicity

An UpToDate review on “Ocular side effects of systemically administered chemotherapy” (Liu et al, 2020) states that “Fibroblast growth factor receptor (FGFR) inhibitors -- Several inhibitors of FGFR (including ponatinib, dovitinib, and erdafitinib) are in clinical trials for a variety of malignancies.  Erdafitinib has now been approved for the treatment of advanced urothelial cancers that harbor certain FGFR mutations.  All of these drugs appear to be associated with a similar type of serous retinopathy (foci of sub-retinal fluid) to that seen with the MEK inhibitors, possibly because the FGFR pathway intersects with the MEK pathway.  In the phase II BLC2001 trial, which included 87 patients with locally advanced or metastatic urothelial cancer that had susceptible FGFR2 or FGFR3 mutations, ocular toxicity resulting in a visual field defect was reported in 25 %, with a median time to first onset of 50 days.  Grade 3 symptoms, defined as involving the central field of vision causing vision worse than 20/40 or >3 lines of worsening from baseline, were reported in 3 % of patients.  Dry eye symptoms occurred in 28 % of patients during treatment and were grade 3 in 6 %.  Ocular symptoms resolved in 13 % and were ongoing at the study cutoff in 13 % … The United States prescribing information for erdafitinib recommends that all patients receive dry eye prophylaxis with ocular lubricants as needed.  Monthly ophthalmologic examinations (including an assessment of visual acuity, slit lamp examination, fundus examination, and optical coherence tomography) are recommended during the first 4 months of treatment and every 3 months thereafter, with urgent reevaluation at any time for visual symptoms.  It is recommended that the drug be withheld when serous retinal toxicity occurs, regardless of vision, and permanently discontinued if it does not resolve in 4 weeks or if it is grade 4 in severity (i.e., visual acuity 20/200 or worse in the affected eye).  However, there were no data provided on the percentage of patients whose symptoms resolved within 4 weeks.  There are also recommended dose modification guidelines for patients who develop ocular adverse reactions”.

Evaluation of Visual Snow Syndrome

According to NIH’s Genetic and Rare Diseases Information Center (GARD) webpage, visual snow syndrome is diagnosed based on the symptoms and a specific set of criteria.  In order for a person to be diagnosed with visual snow syndrome, other potential causes of the symptoms must be ruled out.  Most people with visual snow syndrome have normal vision tests and normal brain structure on imaging studies.  Symptoms of visual snow syndrome may include:

• Tiny, snow-like dots across the visual field
• Sensitivity to light (photophobia)
• Continuing to see an image after it is no longer in the field of vision (palinopsia)
• Difficulty seeing at night (nyctalopia)
• Seeing images from within the eye itself (entoptic phenomena).

Less common symptoms may include migraines, tinnitus, and fatigue.  In general, the symptoms of visual snow syndrome don't change with time.  Some people with visual snow syndrome have depression or anxiety related to their symptoms.  The symptoms of visual snow syndrome can start at any age, but usually occur in early adulthood.  The underlying cause of visual snow syndrome is unknown.  It is thought to be due to a problem with how the brain processes visual images.  There is no mentioning on “fundus photography” or “scanning computerized ophthalmic diagnostic imaging, posterior segment”.  Visual Snow Syndrome

Puledda et al (2020) validated the current criteria of visual snow and described its common phenotype using a substantial clinical data-base.  These investigators carried out a web-based survey of patients with self-assessed visual snow (n = 1,104), with either the complete visual snow syndrome ( VSS; n = 1,061) or visual snow without the syndrome (n = 43).  They also described a population of patients (n = 70) with possible hallucinogen persisting perception disorder who presented clinically with visual snow syndrome.  The visual snow population had an average age of 29 years and had no sex prevalence.  The disorder usually started in early life, and approximately 40 % of patients had symptoms for as long as they could remember.  The most commonly experienced static was black and white.  Floaters, after-images, and photophobia were the most reported additional visual symptoms.  A latent class analysis showed that visual snow does not present with specific clinical endophenotypes.  Severity can be classified by the amount of visual symptoms experienced.  Migraine and tinnitus had a very high prevalence and were independently associated with a more severe presentation of the syndrome.  The authors concluded that clinical characteristics of visual snow did not differ from the previous cohort in the literature, supporting validity of the current criteria.  Visual snow likely represents a clinical continuum, with different degrees of severity.  On the severe end of the spectrum, it is more likely to present with its common co-morbid conditions, migraine and tinnitus.  Visual snow does not depend on the effect of psychotropic substances on the brain.  This study does not mention fundus photography or scanning computerized ophthalmic diagnostic imaging as diagnostic/management tools.

Traber et al (2020) noted that visual snow is considered a disorder of central visual processing resulting in a perturbed perception of constant bilateral whole-visual field flickering or pixelation.  When associated with additional visual symptoms, it is referred to as VSS.  Its pathophysiology remains elusive.  These researchers highlighted the visual snow literature focusing on recent clinical studies that add to the understanding of its clinical picture, pathophysiology, and treatment.  Clinical characterization of VSS is evolving, including a suggested modification of diagnostic criteria.  Regarding pathophysiology, 2 recent studies tested the hypothesis of dysfunctional visual processing and occipital cortex hyper-excitability using electrophysiology.  Likewise, advanced functional imaging shows promise to allow further insights into disease mechanisms.  A retrospective study provided Class IV evidence for a possible benefit of lamotrigine in a minority of patients.  The authors concluded that scientific understanding of VSS is growing.  Major challenges remain the subjective nature of the disease, its overlap with migraine, and the lack of quantifiable outcome measures, which are necessary for clinical trials.  In that context, refined perceptual assessment, objective electrophysiological parameters, as well as advanced functional brain imaging studies, are promising tools in the pipeline.

Eren and Schankin (2020) stated that VSS is a debilitating disorder characterized by tiny flickering dots (like TV static) in the entire visual field and a set of accompanying visual (palinopsia, enhanced entoptic phenomena, photophobia, nyctalopia), non-visual (e.g., tinnitus) and non-perceptional (e.g., concentration problems, irritability) symptoms.  Its pathophysiology is enigmatic and therapy is often frustrating.  These researchers summarized the current understanding of pathophysiology and treatment of VSS.  They carried out a systematic search of PubMed data-base using the key word "visual snow" and pre-defined inclusion and exclusion criteria.  The results were stratified into "treatment" and "pathophysiology".  In additional, these investigators performed a search with the key words "persistent migraine aura" and "persistent visual aura" and screened for mis-diagnosed patients actually fulfilling the criteria for visual snow syndrome.  The reference lists of most publications and any other relevant articles known to the authors were also reviewed and added if applicable.  From the 50 original papers found by searching for "visual snow, 21 were included according to the inclusion and exclusion criteria.  An additional 4 publications came searching for "persistent migraine aura" or "persistent visual aura".  Further publications derived from literature references resulting in a total of 20 articles for pathophysiology and 15 for treatment with some overlaps.  Regarding pathophysiology, hyper-excitability of the visual cortex and a processing problem of higher order visual function were assumed; however, the location is still being debated.  In particular, it is unclear if the primary visual cortex, the visual association cortex or the thalamo-cortical pathway is involved.  Regarding treatment, data are available on a total of 153 VSS patients with medication mentioned for 54 resulting in a total of 136 trials.  From the 44 different medications tried, only 8 were effective at least once.  The best data are available for lamotrigine being effective in 8/36 (22.2 %, including 1 total response and no worsening), followed by topiramate being effective in 2/13 (15.4 %, no total response and 1 worsening).  The only other medication resulting in worsening of VSS was amitriptyline according to the literature review.  The others reported to be effective at least once were valproate, propranolol, verapamil, baclofen, naproxen and sertraline.  The non-pharmacological approach using color filters of the yellow-blue color spectrum might also be helpful in some patients.  The authors concluded that VSS is still far from being fully understood.  In respect of pathophysiology, a disorder of visual processing is likely.  The best pharmacological evidence exists for lamotrigine, which can be discussed off-label.  As non-pharmacological option, patients might benefit from tinted glasses for everyday use.

Yoo et al (2020) noted that the findings of ophthalmic examinations have not been systematically investigated in VSS.  These researchers examined the abnormal neuro-ophthalmologic findings in a patient cohort with symptoms of VSS.  They retrospectively reviewed 28 patients who were referred for symptoms of visual snow to a tertiary referral hospital from November 2016 to October 2019.  These investigators defined the findings of best corrected visual acuity (BCVA), visual field testing, pupillary light reflex, contrast sensitivity, full-field and multi-focal electroretinography (mf-ERG), and optical coherence tomography (OCT).  A total of 20 patients (71 %) were finally diagnosed as VSS.  Their additional visual symptoms included illusionary palinopsia (61 %), enhanced entoptic phenomenon (65 %), disturbance of night vision (44 %), and photophobia (65 %).  A history of migraine was identified in 10 patients (50 %).  The mean BCVA was less than 0.1 logarithm of the minimum angle of resolution, and electrophysiology showed normal retinal function in all patients.  Contrast sensitivity was decreased in 2 of the 7 patients tested.  Medical treatment was administered to 5 patients which all turned out to be ineffective.  Among the 8 patients who were excluded, 1 was diagnosed with rod-cone dystrophy and another with idiopathic intra-cranial hypertension.  The authors concluded that neuro-ophthalmologic findings were mostly normal in patients with VSS; retinal or neurological diseases must be excluded as possible causes of visual snow.

Patient-Initiated Optical Coherence Tomography (OCT) with Mobile Devices

Mehta et al (2017) noted that OCT is widely used in ophthalmology clinics and has potential for more general medical settings and remote diagnostics.  In anticipation of remote applications, these researchers developed wireless interactive control of an OCT system using mobile devices.  A web-based user interface (WebUI) was developed to interact with a hand-held OCT system.  The WebUI consisted of key OCT displays and controls ported to a webpage using HTML and JavaScript.  Client–server relationships were created between the WebUI and the OCT system computer.  The WebUI was accessed on a cellular phone mounted to the hand-held OCT probe to wirelessly control the OCT system.  A total of 20 subjects were imaged using the WebUI to assess the system.  System latency was measured using different connection types (wireless 802.11n only, wireless to remote virtual private network [VPN], and cellular).  Using a cellular phone, the WebUI was successfully used to capture posterior eye OCT images in all subjects.  Simultaneous interactivity by a remote user on a laptop was also demonstrated.  On average, use of the WebUI added only 58, 95, and 170 ms to the system latency using wireless only, wireless to VPN, and cellular connections, respectively.  Qualitatively, operator usage was not affected.  The authors concluded that using a WebUI, they demonstrated wireless and remote control of an OCT system with mobile devices.  These researchers stated that by enabling wireless viewing and control of an OCT imaging session by multiple users, the WebUI provides a promising platform to develop remote OCT applications.  This is particularly important for eye care delivery in acute or general care settings, which may have limited access to specialty ophthalmic care.  The WebUI has the potential to enable primary, non-ophthalmologist providers to receive real-time feedback and input from remotely located specialists simultaneously during OCT imaging sessions (or asynchronously for stored OCT data); and ultimately improve eye care for those patients with ocular pathology who present to non-specialty acute care settings.  They noted that although there have been prior works in store-and-forward tele-ophthalmology using OCT, the presented work is, to the best of the authors’ knowledge, the 1st demonstration of live, remote interactivity and control of ocular imaging with an OCT system.

Malone et al (2019) stated that OCT is the gold standard for quantitative ophthalmic imaging.  The majority of commercial and research systems require patients to fixate and be imaged in a seated upright position, which limits the ability to perform ophthalmic imaging in bedridden or pediatric patients.  Hand-held OCT devices overcome this limitation; however, image quality often suffers due to a lack of real-time aiming and patient eye and photographer motion.  These researchers described a hand-held spectrally encoded coherence tomography and reflectometry (SECTR) system, which enabled simultaneous en face reflectance and cross-sectional OCT imaging.  The hand-held probe employs a custom double-pass scan lens for fully tele-centric OCT scanning with a compact optomechanical design and a rapid-prototyped enclosure to reduce the overall system size and weight.  These researchers also introduced a variable velocity scan waveform that allowed for simultaneous acquisition of densely sampled OCT angiography (OCTA) volumes and wide-field reflectance images, which enabled high-resolution vascular imaging with precision motion-tracking for volumetric motion correction and multi-volumetric mosaicking.  Finally, these investigators demonstrated in-vivo human retinal OCT and OCTA imaging using hand-held SECTR on a healthy volunteer.  Clinical translation of hand-held SECTR will allow for high-speed, motion-corrected wide-field OCT and OCTA imaging in bedridden and pediatric patients who may benefit ophthalmic disease diagnosis and monitoring.  The authors concluded that while further work is needed to overcome limitations associated with this new technology, they have demonstrated the feasibility of wide-field ophthalmic OCTA using a hand-held probe.

Chopra et al (2021) stated that OCT is a paragon of success in the translation of biophotonics science to clinical practice.  OCT systems have become ubiquitous in eye clinics; however, access beyond this is limited by their cost, size and the skill needed to operate the devices.  Remarkable progress has been made in the development of OCT technology to improve the speed of acquisition, the quality of images and into functional extensions of OCT such as OCT angiography.  However, more works to be carried out to radically improve the access to OCT by addressing its limitations and enable penetration outside of typical clinical settings and into under-served populations.  Beyond high-income countries, there are 6.5 billion people with similar eye-care needs, which could not be met by the current generation of bulky, expensive and complex OCT systems.  These investigators noted that recent regulatory approvals and feasibility studies highlighted the emerging spread of miniaturized, portable and hand-held OCT systems, lending their use as a point-of-care diagnostic tool in non-traditional settings such as intensive care, as well as in the home environment for remote monitoring of chronic conditions such as AMD.  The latter in particular could be analogous to continuous monitoring; thus, providing opportunities for personalized treatment plans for conditions such as wet AMD that benefit from close monitoring and often require indefinite follow-up. 

Hillmann (2021) noted that OCT has become one of the most important techniques in ophthalmic diagnostics, as it is the only way to provide a 3D visualization of morphological changes in the layered structure of the retina at a high resolution.  Furthermore, OCT is applied for countless medical and technical purposes.  Recent developments pave the way for small-footprint OCT systems at significantly reduced costs, thereby extending possible use cases.  The author stated that it appears increasingly likely that, in the near future, OCT will find its way into many more industrial and medical applications, including disease monitoring at home. 

Well-designed studies are needed to determine whether patient-initiated OCT would improve health outcomes of patients with ophthalmic diseases.

Optic Nerve Sheath Diameter Measurement for Detection of Increased Intra-Cranial Pressure

Guidelines from the American College of Emergency Physicians (ACEP, 2016) state that "The use of emergency US in the eye has described for the detection of posterior chamber and orbital pathology. Specifically, US has been described to detect retinal detachment, vitreous hemorrhage, and dislocations or disruptions of structures. In addition the structures posterior to the globe such as the optic nerve sheath diameter may be a reflection of other disease in the central nervous system." Among the references that these guidelines include as support for this statement a study by Tayal et al (2007) describing emergency department use of ultrasound measurement of optic nerve sheath diameter to detect findings of increased intracranial pressure in adult head injury patients.

Raffiz and Abdullah (2017) stated that bedside ultrasound measurement of optic nerve sheath diameter (ONSD) is emerging as a non-invasive technique to examine and predict raised intra-cranial pressure (ICP).  It has been shown in previous studies that ONSD measurement has good correlation with surrogate findings of raised ICP such as clinical and radiological findings suggestive of raised ICP.  In a prospective, observational study, these researchers examined the correlation between sonographic measurements of ONSD value with ICP value measured via the gold standard invasive intra-cranial ICP catheter; and determined the cut-off value of ONSD measurement in predicting raised ICP, along with its sensitivity and specificity value.  This trial was carried out using convenience sample of 41 adult neurosurgical patients treated in neurosurgical intensive care unit (ICU) with invasive ICP monitoring placed in-situ as part of their clinical care.  Portable SonoSite ultrasound (US) machine with 7-MHz linear probe were employed to measure ONSD using the standard technique.  Simultaneous ICP readings were obtained directly from the invasive monitoring.  A total of 75 measurements were carried out on 41 patients.  The non-parametric Spearman correlation test revealed a significant correlation at the 0.01 level between the ICP and ONSD value, with correlation coefficient of 0.820.  The ROC curve generated an area under the curve (AUC) with the value of 0.964, and with standard error (SE) of 0.22.  From the ROC curve, these investigators found that the ONSD value of 5.205 mm is 95.8 % sensitive and 80.4 % specific in detecting raised ICP.  The authors concluded that ONSD value of 5.205 was sensitive and specific in detecting raised ICP.  Bedside US measurement of ONSD was readily learned; and was reproducible and reliable in predicting raised ICP.  These researchers stated that this non-invasive technique could be a useful adjunct to the current invasive intracranial catheter monitoring, and has wide potential clinical applications in district hospitals, emergency rooms and ICUs.

Robba and associates (2018) noted that although invasive intracranial devices (IIDs) are the gold standard for the measurement of ICP, US of the ONSD has been suggested as a potential non-invasive ICP estimator.  In a meta-analysis, these researchers examined the diagnostic accuracy of sonographic ONSD measurement for the evaluation of intracranial hypertension (IH) in adult patients.  They searched on electronic databases (Medline/PubMed, Scopus, Web of Science, ScienceDirect, Cochrane Library) until May 31, 2018 for comparative studies that examined the effectiveness of sonographic ONSD versus ICP measurement with IID.  Data were extracted independently by 2 authors; they used the QUADAS-2 tool for assessing the risk of bias (RB) of each study.  A diagnostic meta-analysis following the bi-variate approach and random-effects model was carried out.  A total of 7 prospective studies (320 patients) were evaluated for IH detection (assumed with ICP greater than 20 mmHg or greater than 25 cmH2O).  The accuracy of included studies ranged from 0.811 (95 % CI: 0.678 to 0.847) to 0.954 (95 % CI: 0.853 to 0.983); 3 studies were at high RB.  No significant heterogeneity was found for the DOR, positive likelihood ratio (PLR) and negative likelihood ratio (NLR), with I2 < 50 % for each parameter.  The pooled DOR, PLR and NLR were 67.5 (95 % CI: 29 to 135), 5.35 (95 % CI: 3.76 to 7.53) and 0.088 (95 % CI: 0.046 to 0.152), respectively.  The area under the hierarchical summary ROC curve (AUHSROC) was 0.938.  In the subset of 5 studies (275 patients) with IH defined for ICP greater than 20 mmHg, the pooled DOR, PLR and NLR were 68.10 (95 % CI: 26.8 to 144), 5.18 (95 % CI: 3.59 to 7.37) and 0.087 (95 % CI: 0.041 to 0.158), respectively, while the AUHSROC was 0.932.  The authors concluded that although the wide 95 % CI in the pooled DOR suggested caution, US ONSD may be a potentially useful approach for assessing IH when IIDs are not indicated or available.

In a systematic review and meta-analysis, Aletreby and colleagues (2022) examined the diagnostic accuracy of ONSD when compared to the standard invasive ICP measurement.  These researchers carried out a systematic search of PubMed and Embase for studies including adult patients with suspected elevated ICP and comparing sonographic ONSD measurement to a standard invasive method.  Quality of studies was examined using the QUADAS-2 tool by 2 independent authors.  They used a bi-variate model of random effects to summarize pooled sensitivity, specificity, and diagnostic odds ratio (DOR).  Heterogeneity was investigated by meta-regression and sub-group analyses.  These investigators included 18 prospective studies (16 studies including 619 patients for primary outcome).  Only 1 study was of low quality, and there was no apparent publication bias.  Pooled sensitivity was 0.9 [95 % CI: 0.85 to 0.94], specificity was 0.85 (95 % CI: 0.8 to 0.89), and DOR was 46.7 (95 % CI: 26.2 to 83.2) with partial evidence of heterogeneity.  The AUC of the summary ROC was 0.93 (95 % CI: 0.91 to 0.95, p < 0.05).  No co-variates were significant in the meta-regression.  Subgroup analysis of severe TBI and parenchymal ICP found no heterogeneity.  ICP and ONSD had a correlation coefficient of 0.7 (95 % CI: 0.63 to 0.76, p < 0.05).  The authors concluded that ONSD is a useful adjunct in ICP evaluation; however, it is currently not a replacement for invasive methods where they are feasible.

Dixon et al (2020) commented that "Sonographic measurement of the optic nerve sheath diameter showed initial promise in detecting elevated intracranial pressure (ICP). In the setting of trauma, elevated ICP is likely to suggest hematoma or significant edema. However, Oberfoell et al. found significant interrater variability in optic nerve sheath diameter measurements, even among emergency physicians with fellowship training in ultrasound. At this time, there are no formal recommendations for ultrasound as the sole imaging modality for diagnostic purposes in head trauma."

An UpToDate review on “Evaluation and management of elevated intracranial pressure in adults” (Smith and Amin-Hanjani, 2022) states that “Noninvasive systems -- A number of devices designed to record ICP noninvasively have been studied, but most have not demonstrated reproducible clinical success or have not been studied in large clinical trials.  We do not use these in clinical practice … Ocular sonography can provide a noninvasive measure of optic nerve sheath diameter, which has been found to correlate with ICP.  A number of studies have found that diameters of 5 to 6 mm have the ability to discriminate between normal and elevated ICP in patients with intracranial hemorrhage and traumatic brain injury”.  

Optic Nerve Sheath Diameter Measurement for Prediction of Neurologic Outcomes in Patients with Post-Cardiac Arrest Return of Spontaneous Circulation

In a review of neurologic prognostication after resuscitation in cardiac arrest, Lupton, et al. (2020) explained: "More recently, ultrasound has been studied as an adjunct to the clinic examination through measurement of the optic nerve sheath diameter as a correlate for neurologic outcome. A larger optic nerve diameter may be indicative of increased intracranial pressure from cerebral edema and thus a worse neurologic outcome. A meta‐analysis of 3 studies (102 patients) evaluating optic nerve sheath diameter measured by ultrasound as a predictor of poor neurologic outcomes after out‐of‐hospital cardiac arrest reported a pooled sensitivity and specificity of 77% and 98%, respectively [citing Lee, et al., 2019].  However, this study did not include ultrasound studies conducted within a 6‐hour window from return of spontaneous circulation or in the ED. In a subsequent study evaluating optic nerve sheath diameter in 36 patients by 1 operator immediately post‐return of spontaneous circulation and at 24‐, 48‐, and 72‐hour intervals, there was no correlation with measurements obtained immediately post‐return of spontaneous circulation and patient outcome [citing Park, et al., 2019]. At 24 hours, the authors found that an optic nerve sheath diameter over 4.9 mm had a sensitivity of 83.3% and a specificity of 94.4% and a PPV of 93.7% for predicting a poor neurologic outcome."

In a systematic review and meta-analysis, Lee and Yun (2019) examined the diagnostic performance of ONSD for the prediction of neurologic outcome in post-cardiac arrest (CA) patients and relative prediction performance according to ONSD measurement modality.  These investigators searched PubMed and Embase databases for diagnostic accuracy studies that used ocular US or brain computed tomography (CT) for prediction of neurologic outcome.  Bi-variate modelling and HSROC modelling were carried out to examine diagnostic performance.  A pooled DOR with a 95 % CI not including 1 was considered informative.  Subgroup analysis was carried out according to the modality (ocular US versus brain CT).  Methodologic quality was evaluated using the Quality Assessment of Diagnostic Accuracy Studies-2 tool.  These investigators conducted meta-regression analyses for heterogeneity exploration.  A total of 8 studies including 766 patients were included.  For prediction of poor neurologic outcome, ONSD showed pooled sensitivity 0.41, pooled specificity 0.99, and area under the ROC curve 0.86.  According to the pooled DORs, ONSD was informative for prediction of neurologic outcome.  In subgroup analysis, ONSD on ocular US showed significantly higher sensitivity and similar specificity than that on brain CT.  On meta-regression analysis, locale, time to examination after return of spontaneous circulation (ROSC), cause of CA, and reference standard were sources of heterogeneity.  The authors concluded that ONSD may be useful for predicting neurologic outcomes in post-CA patients.

Lupton and associates (2020) noted that out-of-hospital CA (OHCA) remains a leading cause of mortality in the U.S., and the majority of patients who die after achieving ROSC die from withdrawal of care due to a perceived poor neurologic prognosis.  Unfortunately, withdrawal of care often occurs during the 1st day of admission and research suggested this early withdrawal of care may be premature and result in unnecessary deaths for patients who would have made a full neurologic recovery.  These investigators examined the evidence for neurologic prognostication in the emergency department (ED) for patients who achieve ROSC after an OHCA.  The authors concluded that for the OHCA patient with ROSC in the ED there are no clinical examination findings, imaging results, laboratory studies, monitoring tests, or scoring systems that predict poor neurologic outcomes with enough statistical certainty to inform withdrawal of care discussions or preclude the need for targeted temperature management and ICU admission.  Based on the available data, physicians should avoid attempts at early withdrawal of care in the ED.

Lee and colleagues (2021) stated that ONSD can aid in predicting the neurologic outcomes of patients with post-CA ROSC.  In a retrospective, single-center study, these investigators examined the effect of ONSD changes before and after CA on neurologic outcomes in patients with ROSC after CA using brain CT.  This trial included patients hospitalized after CA, who had undergone pre- and post-CA brain CT between January 2001 and September 2020.  Subjects were divided into good and poor neurologic outcome (GNO and PNO, respectively) groups based on their neurologic outcome at hospital discharge.  These researchers carried out between-group comparisons of the amount and rate of ONSD changes in brain CT and calculated the AUC to determine their predictive value for neurologic outcomes.  Among the 96 enrolled patients, 25 had GNO.  Compared with the GNO group, the PNO group showed a significantly higher amount (0.30 versus 0.63 mm; p = 0.030) and rate (5.26 % versus 12.29 %; p = 0.041) of change.  The AUC for predicting PNO was 0.64 (95 % CI: 0.53 to 0.73; p = 0.04), and patients with a rate of ONSD change greater than 27.2 % had PNO with 100 % specificity and positive predictive value (PPV).  The authors concluded that ONSD changes may be useful in predicting neurologic outcomes in patients with post-CA ROSC.  Moreover, these researchers stated that there is a need for large, prospective studies to confirm these findings.

The authors stated that this study had several drawbacks.  First, this was a single-center study with a limited sample size that led to insufficient statistical power; however, they carried out a power analysis to calculate the sample size, which was relatively large compared with those of other studies.  Second, this retrospective study included patients who underwent both pre- and post-CA brain CT, which could lead to selection bias affecting the results.  Third, although these researchers attempted to extensively collect variables based on the Utstein Resuscitation Registry Templates, there may still be hidden confounders.  Fourth, there could have been minor measurement errors given the very small size of the ONSD in brain CT; however, to minimize these errors, 2 blinded emergency physicians carried out measurements using a standardized method showing consensus.  Fifth, current guidelines recommend a neurologic outcome assessment at 3 months after discharge; however, these investigators measured the neurologic outcomes at discharge and did not determine the long-term outcomes.  Sixth, OHCA and in-hospital cardiac arrest (IHCA) were both included and analyzed in this study, despite the differences in the characteristics and proportion of GNO and PNO.  Follow-up studies that include only OHCA or IHCA patients may be needed.  This was a retrospective study, and the clinical utility of the predictive value for prognosis remains unclear.

AngioPlex Optical Coherence Tomography Angiography (OCTA) for Evaluation of Diabetic Non-Proliferative Retinopathy and Hypertensive Retinopathy

Johannesen et al (2019) employed OCT angiography (OCTA) to examine foveal microvascular changes in diabetes by comparing the area of foveal avascular zone (FAZ) in healthy controls and patients with diabetes with no diabetic retinopathy (NDR) as well as different stages of diabetic retinopathy (DR).  These researchers carried out a systematic literature search based on the population, intervention, comparison and outcome (PICO) strategy by 2 independent reviewers.  The search was performed in PubMed, Embase and Cochrane Library, including keywords “diabetes mellitus”, “DR” and “OCTA”.  Of 358 studies initially identified, 215 studies were screened after duplicate removal.  Of these, these investigators included 12 (9 cross-sectional and 3 retrospective) studies in this review.  With the data at hand, it was not possible to perform a meta-analysis.  The selected studies included patients with NDR (n = 8), non-proliferative diabetic retinopathy (NPDR, n = 8) and proliferative diabetic retinopathy (PDR, n = 6).  Several of the studies provided information for more than 1 diabetic group.  In general, there was a trend towards a larger area of FAZ in patients with diabetes.  As compared to healthy controls, this was reported in patients with NDR (5 of 8 studies), NPDR (7 of 8 studies) and PDR (6 of 6 studies).  The authors concluded that OCTA was non-invasively able to identify foveal capillary non-perfusion as an early event in DR.  In some studies, this has even been identified in patients without clinically identifiable microvascular lesions.  Moreover, these researchers stated that longitudinal studies are needed to examine if OCTA-findings are able to predict long-term structural and functional outcome.

Akil et al (2019) stated that OCTA is a non-invasive method that enables visualization of blood flow within retinal vessels down to the size of capillaries by detecting motion contrast from moving blood cells.  OCTA provides a safe procedure to assess retinal microvasculature with higher contrast and resolution than conventional fluorescence angiography (FA).  The different capillary plexuses are displayed separately; and their perfusion density can be quantified.  Imaging capabilities such as these have led to an emerging field of clinical application for OCTA in vascular diseases such as DR.  Evaluation of parameters such as para-foveal capillary perfusion density could be a biomarker for disease diagnosis and progression.  Typical microvascular changes in DR such as capillary non-perfusion, micro-aneurysms, intra-retinal microvascular abnormalities, and neovascularization can be reliably detected in OCT angiograms, characterized in detail and attributed to the different capillary plexuses.  Monitoring of these lesions in-vivo gives potential novel insight into the pathophysiology in DR.  The authors summarized the potential applications/utility of OCTA in DR reported in the literature.

Vaz-Pereira et al (2020) noted that DR is a leading cause of blindness due to diabetic macular edema (DME) or complications of proliferative DR (PDR); OCT is a non-invasive imaging technique well-established for DME but less used to assess neovascularization in PDR.  Developments in OCT imaging and the introduction of OCTA have shown significant potential in PDR.

In an observational, cohort, cross-sectional study, Yang et al (2020) examined microvascular abnormalities in diabetic patients without conventional clinical signs of DR.  The study group included randomly chosen participants of a community-based cohort with diabetes type 2 without DR, and the control group consisted of non-diabetic individuals from a population-based study.  All participants underwent OCTA.  Upon OCTA, 118 (40.4 %) eyes of the study group (n = 292 eyes) showed microvascular abnormalities including FAZ erosion (95 (32.5 %) eyes), non-perfusion areas in the superficial and deep retinal layers (39 (13.4 %) eyes and 19 (6.5 %) eyes, respectively), and micro-aneurysms in the superficial and deep retinal layers (22 (7.5 %) eyes and 31 (10.6 %) eyes, respectively).  None of these abnormalities was detected in the control group (n = 80).  The study group showed a lower vessel density in the superficial retinal vascular layer in all regions except for the foveal region (p < 0.001), and higher vessel density in the para-foveal region in the deep retinal vascular layer (p = 0.01).  Higher diabetes prevalence was associated with lower superficial retinal vascular density (p = 0.005) in multi-variable analysis.  A lower radial peripapillary capillary flow density was correlated (regression coefficient r, 0.62) with higher fasting blood concentration of glucose (p < 0.001) in multi-variable analysis.  The authors concluded that OCTA revealed microvascular abnormalities in 40 % of eyes of diabetic patients without ophthalmoscopically detectable diabetic fundus changes in a community-based population.  These researchers stated that the early stage of DR may be re-defined upon OCTA.

Li et al (2021) noted that DR is the most common microvascular complication of diabetes; however, early changes in retinal micro-vessels are difficult to detect clinically, and a patient's vision may have begun to deteriorate by the time a problem is identified; OCTA is an innovative tool for observing capillaries in-vivo.  In a retrospective, observational, cross-sectional study, these researchers analyzed retinal vessel density and thickness changes in patients with diabetes.   Between August 2018 and February 2019, these investigators collected OCTA data from healthy subjects and diabetics from the First Affiliated Hospital of Harbin Medical University.  They analyzed the retinal vessel density and thickness changes of the participants.  A total of 97 diabetic patients with diabetes at different severity stages of DR and 85 controls were involved in this study.  Diabetic patients exhibited significantly lower retinal VD (particularly in the deep vascular complexes), thickening of the neurosensory retina, and thinning of the retinal pigment epithelium compared with controls.  In the control group, non-DR group and mild DR group, superficial VD was significantly correlated with retinal thickness (r = 0.3886, p < 0.0001; r = 0.3276, p = 0.0019; r = 0.4614, p = 0.0024, respectively).  The authors concluded that patients with diabetes exhibit ischemia of the retinal capillaries and morphologic changes in-vivo before vision loss; thus, OCTA may be useful as a quantitative method for the early detection of DR.

Multiple Sclerosis and Optic Neuritis

Kenney et al (2022) noted that recent studies have suggested that inter-eye differences (IEDs) in peripapillary RNFL (pRNFL) or ganglion cell + inner plexiform (GCIPL) thickness by SD-OCT may identify individuals with a history of unilateral ON; however, this requires further validation.  Machine learning classification may be useful for validating thresholds for OCT IEDs and for examining added utility for visual function tests, such as low-contrast letter acuity (LCLA), in the diagnosis of people with MS (PwMS) and for unilateral ON history.  In a cross-sectional study, participants were from 11 sites within the International Multiple Sclerosis Visual System consortium.  pRNFL and GCIPL thicknesses were measured using SD-OCT.  A composite score combining OCT and visual measures was compared individual measurements to determine the best model to distinguish PwMS from controls.  These methods were also used to distinguish those with a history of ON among PwMS.  ROC curve analysis was performed on a training data set (2/3 of cohort) and then applied to a testing data set (1/3 of cohort).  Support vector machine (SVM) analysis was used to assess whether machine learning models improved diagnostic capability of OCT.  Among 1,568 PwMS and 552 controls, variable selection models identified GCIPL IED, average GCIPL thickness (both eyes), and binocular 2.5 % LCLA as most important for classifying PwMS versus controls.  This composite score performed best, with AUC = 0.89 (95 % CI: 0.85 to 0.93), sensitivity = 81 %, and specificity = 80 %.  The composite score ROC curve performed better than any of the individual measures from the model (p < 0.0001).  GCIPL IED remained the best single discriminator of unilateral ON history among PwMS (AUC = 0.77, 95 % CI: 0.71 to 0.83, sensitivity = 68 %, specificity = 77 %).  SVM analysis performed comparably with standard logistic regression models.  The authors concluded that a composite score combining visual structure and function improved the capacity of SD-OCT to distinguish PwMS from controls.  GCIPL IED best distinguished those with a history of unilateral ON.  SVM performed as well as standard statistical models for these classifications.  Classification of Evidence = III.

The authors stated that the limitations of this study included the cross-sectional design.  Because longitudinal data for these investigations are not yet available, predictions of change over time, such as the capacity to predict future ON or future conversion to MS based on current diagnostic criteria, are yet to be determined.  Another limitation was the fact that GCIPL thickness and high-contrast visual acuity (HCVA) were not measured for all sites.  Furthermore, 74 subjects did not have information on ON history.  These investigators would not expect bias to be introduced from the small amount of missing data.  Subjects with no GCIPL, HCVA, or ON history data were similar in all baseline characteristics compared with those with these measures included.  Participants who did not have LCLA had slightly longer disease durations than did those with LCLA measurements (10.2 versus 6.7 years); however, distributions of EDSS scores, disease subtypes, pRNFL and GCIPL values were similar.  It was unlikely that this subgroup had more severe optic nerve disease; thus, these investigators would not expect the results to be biased by these characteristics.  The healthy control cohort included 55 % female patients while the PwMS cohort was 70 % female.  In a previous study using these data, sensitivity analyses were performed to compare the cohorts; it was determined that no significant bias would come from these discrepancies in sex distribution.  Although there could be site-related effects, site was not included as a co-variate in the statistical models.  Site would not be interpretable in the composite score or machine learning models.  To overcome potential site differences, the conversion equation described was used to normalize data.  Furthermore, other factors such as age, sex, and race/ethnicity were included in the models.  Given these adjustments, these researchers did not expect any bias was introduced from site-related effects.

Noll et al (2022) OCTA is a novel technique allowing non-invasive assessment of the retinal vasculature.  During RRMS, retinal vessel loss occurs in eyes suffering from acute ON and recent data suggested that retinal vessel loss might also be evident in non-affected eyes.  These investigators examined if alterations of the retinal vasculature are linked to the intra-thecal immunity and whether they allow prognostication of the future disease course.  This study included 2 different patient cohorts recruited at a tertiary German academic MS center between 2018 and 2020 and a cohort of 40 healthy controls.  A total of 90 patients with RRMS undergoing lumbar puncture and OCTA analysis were enrolled into a cross-sectional cohort study to examine associations between the retinal vasculature and the intra-thecal immune compartment.  They recruited another 86 RRMS patients into a prospective, observational, cohort study who underwent clinical examination, OCTA and cerebral MRI at baseline and during annual follow-up visits to clarify whether alterations of the retinal vessels are linked to RRMS disease activity.  Eyes with a history of ON were excluded from the analysis.  Rarefication of the superficial vascular complex occurred during RRMS and was linked to higher frequencies of activated B cells and higher levels of the pro-inflammatory cytokines -- interferon gamma (IFN-γ), tumor necrosis factor alpha (TNF-α) and interleukin (IL)-17 -- in the CSF.  During a median follow-up of 23 (inter-quartile range [IQR] of 14 to 25) months, vessel loss within the superficial (hazard ratio [HR] 1.6 for a 1 %-point decrease in vessel density, p = 0.01) and deep vascular complex (HR 1.6 for a 1 %-point decrease, p = 0.05) was associated with future disability worsening.  The authors concluded that ON independent rarefication of the retinal vasculature might be linked to neuroinflammatory processes during RRMS and might predict a worse disease course; therefore, OCTA might be a novel biomarker to monitor disease activity and predict future disability.  These researchers stated that If confirmed by others and validated within bigger trials with longer clinical follow-up durations, evaluation of the retinal vasculature could allow for estimating disease activity and prognosis.

The authors stated that this study had several drawbacks.  First, the longitudinal follow-up period of this trial was short and only a small portion of patients faced disability worsening.  These circumstances might result in false positive results; thus, these findings need to be discussed with caution.  However, OCTA is a novel technique and longer follow-up durations are thereby limited.  Second, the results of the CSF analysis were limited by the fact that some cytokines, especially IL-17, were only traceable in a small portion of patients.  Third, OCTA is technically challenging.  Especially for eyes with visual impairment, OCTA becomes extremely susceptible to imaging artifacts that affect vessel density measurements.  All examinations were carried out by experienced technicians, and these researchers employed a rigorous approach for OCTA quality control to ensure reliable OCTA results.  Widely accepted OCTA quality criteria, however, are missing to-date.  Fourth, OCTA measurements were device-specific.  These investigators could not exclude a device-specific effect on these findings and conclusions, and studies using and comparing different OCTA machines are needed.  Fifth, there were methodological issues that restrict the interpretability of these findings.  OCTA provided information regarding retinal perfusion patterns, but not on vessel morphology and vessel integrity.  An automatic and robust differentiation of retinal vessel structures into veins and arteries is not possible and the currently used technique does not allow to distinguish whether a decrease in retinal vessel density reflects true vessel loss, constriction, or shrinkage of vessels.  Therefore, advances in both hardware and software solutions are urgently needed.  Sixth, the physiological influence of age and sex and other environmental factors on OCTA measures are focus of current research and thus not entirely clear.  It is probable that age especially might influence retinal vessel densities; thus, these findings might not be transferable to younger or older patient cohorts.  However, to address this issue, these researchers corrected all their statistical models for the covariate age to reduce an influence of purely age-related changes on their findings.

Annual OCT for the Follow-Up of Thyroid Ophthalmopathy After Orbital Decompression

In an observational, cross-sectional, controlled study, Forte et al (2010) examined retinal nerve fiber layer (RNFL) thickness in eyes with Graves' orbitopathy (GO), in eyes with ocular hypertension (OHT) and in a control group of healthy eyes.  These investigators evaluated all patients with primary open-angle glaucoma (POAG) and all patients with GO and intra-ocular pressure (IOP) of greater than 23 mm Hg in primary position examined from March 2006 to June 2007.  A total of 40 apparently healthy patients (80 eyes) were enrolled as a control group.  Complete ophthalmic evaluation, visual field (VF) examination with the Humphrey Visual Field Analyzer and RNFL thickness measurement with optic nerve tracking OCT (ONT-OCT) were performed.  Among 116 eyes with POAG [58 patients, 32 men, 26 women, mean age of 62 (46 to 71) years], RNFL was reduced in 87 eyes (75 %, p = 0.05) when compared with the control group, and a good correlation was found between RNFL thickness and VF abnormalities (Spearman's rho 0.822; p = 0.001).  Among 60 eyes [30 patients, 12 men, 18 women, mean age of 56 (50 to 69) years] with GO and OHT, non-glaucomatous diffuse abnormalities of the VF were detected in 44 eyes (73.3 %, p = 0.03), while RNFL thinning was present in 14 eyes (9 patients, 23.3 %, p = 0.03).  No correlation was found between RNFL thickness and VF abnormalities (Spearman's rho 0.365; p = 0.02).  No significant differences in RNFL pattern were present between the group with GO, OHT and RNFL thinning and the group with POAG.  The authors concluded that in patients with GO and OHT, evaluation of RNFL thickness with ONT-OCT may represent an objective diagnostic technique for detecting optic neuropathy.

Park et al (2016) examined the influence of optic nerve compression (ONC) on the peripapillary RNFL thickness in eyes with acute and chronic dysthyroid optic neuropathy (DON).  Patients with DON and healthy control subjects underwent peripapillary OCT scanning with the Cirrus HD-OCT.  Patients were classified as acute (within 6 months from the onset of DON) versus chronic (6 months or more from the onset of DON) DON.  The thickness of peripapillary RNFL was compared between eyes with acute and chronic DON and control eyes.  Baseline factors associated with visual acuity (VA) at the last visit were also analyzed.  The mean temporal peripapillary RNFL thickness was thinnest in chronic DON at 66 ± 12 μm compared to 76 ± 8 μm in eyes with acute DON and 73 ± 12 μm in control eyes (p = 0.014).  In a multi-variable analysis, patients with greater inferior peripapillary RNFL thickness and younger age tended to have better VA at the last visit (p = 0.034, odds ratio [OR] = 1.038 and p = 0.007, OR = 0.912, respectively).  The authors concluded that these findings revealed a notable difference in temporal peripapillary RNFL thickness in eyes with chronic DON compared to eyes with acute DON and control eyes.  They also found a significant association between inferior peripapillary RNFL thickness and VA at the last visit.  Thicker inferior peripapillary RNFL thickness was associated with better visual outcome.  These researchers stated that further studies with large sample sizes using a prospective design should more clearly reveal the time aspect of the association between the onset of DON and the changes in peripapillary RNFL, and their clinical significance.

In an observational, cross-sectional study, De-Pablo-Gomez-de-Liano et al (2019) evaluated the correlation between OCT and MRI measurements of extra-ocular rectus muscle thickness in patients with Graves' ophthalmopathy.  This trial was carried out in 62 eyes of 31 patients with Graves' ophthalmopathy.  The disease phase was inactive in 20 patients and active in the remaining 11.  The OCT measurements obtained included medial rectus thickness at 7.2 and 9.2 mm from the limbus and lateral rectus thickness at 8.5 mm from the limbus.  MRI measurements were maximum transversal diameter (T-MRI), cranio-caudal diameter (CC-MRI), and muscle area (A-MRI).  For the whole patient cohort, correlation emerged between the OCT-MR and T-MRI measurements (r = 0.428 to 0.576; p ≤ 0.002), A-MRI (r = 0.562 to 0.674; p < 0.001), and CC-MRI (r = 0.286 to 0.293; p ≤ 0.046).  In patients with clinically active Graves' ophthalmopathy, correlations with T-MRI (r = 0.576 to 0.604; p ≤ 0.010) and A-MRI (r = 0.678 to 0.706; p < 0.001) were higher.  No correlations were observed between OCT and MRI measurements of lateral rectus thickness (p ≥ 0.177), regardless of disease phase.  The authors concluded that the correlations observed suggested that OCT could be a complementary assessment or screening method to detect thickening of the anterior portion of the medial rectus muscle in patients with Graves' ophthalmopathy, which may be especially useful when MRI is not available.

In a prospective, single-center, case-series study, Lewis et al (2019) examined the use of OCTA in the evaluation of GO and response to orbital decompression in patients with and without DON.  This trial included 12 patients (24 orbits) with GO and ± DON, (6 orbits) who underwent bilateral orbital decompression.  All patients underwent pre- and post-operative OCTA of the peripapillary area.  Vessel density indices were calculated in a 4.5 mm × 4.5 mm ellipsoid centered on the optic disk using split-spectrum amplitude decorrelation angiography algorithm, producing the vessel density measurements.  Mean change in vessel density indices was compared between pre- and post-operative sessions and between patients with and without DON.  Patient 1, a 34-year-old man with GO and unilateral DON OD, showed a significant reduction in blood vessel density indices oculus dexter (OD) (DON eye) after decompression while a more modest reduction was found oculus sinister (OS) with the greatest change noted intra-papillary.  Patient 2, a 50-year-old man with DON OU, showed worsening neuropathy following decompression OD that was confirmed by angiographic density indices.  Patient 3, a 55-year-old woman with DON, showed a reduction in blood vessel density OD and increased density OS.  Patients without DON showed overall less impressive changes in indices as compared to those with DON.  Using OCTA, response to surgical treatment in GO orbits, more so in orbits with DON, could be demonstrated and quantified using vessel density indices with reproducibility.

Furthermore, an UpToDate review on “Clinical features and diagnosis of Graves' orbitopathy (ophthalmopathy)” (Davies and Burch, 2023) does not mention OCT as a management option.

OCT for Evaluation of Non-Arteritic Anterior Ischemic Optic Neuropathy

An UpToDate chapter on clinical features and diagnosis of non-arteritic anterior ischemic optic neuropathy (NAION) (Tamhankar and Volpe, 2023) stated that, in acute stages of NAION, optical coherence tomography (OCT) will show thickness of the inner and outer layers of the peripapillary retina, while in chronic NAION cases, atrophy of the ganglion cell layer and its axons is seen as inner retinal thinning. 

In a systematic review and meta-analysis, Chou et al (2022) examined microvascular alterations with OCTA in eyes with non-arteritic anterior ischemic optic neuropathy (NAION) and the unaffected fellow eyes.  These investigators carried out a comprehensive literature search in the PubMed and Embase databases through September 6, 2020, and identified the studies on NAION and the unaffected fellow eyes using OCTA.  Eligible studies and data of interest were extracted and analyzed by RevMan Software v. 5.4 and Stata Software v.14.0.  The weighted mean differences (WMD) and 95 % CIs were used to evaluate the strength of the association.  A total of 17 observational comparative studies, including 379 eyes with NAION, 175 unaffected contralateral eyes and 470 eyes of healthy controls, were identified.  Compared to those of the healthy controls, the perfusion density (PD) of radial peripapillary capillary (RPC) and peripapillary superficial capillary plexus (ppSCP) of NAION were significantly lower.  Moreover, the PD of the macular SCP (mSCP) in NAION was significantly reduced in the whole image, superior quadrant and temporal quadrant, while the macular deep capillary plexus (mDCP) showed a decreasing PD only within the whole image.  Between unaffected fellow eyes and healthy eyes, significant differences of PD were demonstrated in the whole image and some peripapillary regions of the RPC and ppSCP.  The authors concluded that these findings suggested that compared to those of healthy controls, the eyes affected by NAION and unaffected fellow eyes demonstrated significant microvascular impairments in different regions.  Between acute and non-acute NAION, macular OCTA parameters showed different characteristic patterns.

These researchers stated that this study showed that there are several potential knowledge gaps that may guide the direction of future studies.  First, more studies of macular OCTA parameters of NAION with different disease durations are needed to examine the microcirculation change within the foveal and peri-foveal areas.  Second, the changes between unaffected fellow eyes and the healthy controls or acute versus non-acute NAION detected by OCTA require more direct comparisons within different areas and at different disease phases to examine if the difference in OCTA parameters between them is significant.  Third, follow-up studies may be needed to examine the association between the change in OCTA in unaffected fellow eyes and the subsequent incidence.  Admittedly, this meta-analysis had several drawbacks: First, although these investigators accumulated all available studies and data, the sample size of some OCTA measurements was relatively small.  Second, since the acute and chronic periods of NAION were not standardized, these researchers could not perform a subgroup analysis based on specific duration.  Limited by sufficient studies using different OCTA devices, the authors could not carry out subgroup analysis based on that too.  Third, the heterogeneities of some outcomes could not be eliminated after sensitivity analysis and subgroup analysis.  Fourth, since only 4 studies reported systemic co-morbidities (such as diabetes, obstructive sleep apnea, and systemic hypertension) that might affect retinal microcirculation, these researchers could not analyze those potential confounding factors in the present study.

OCT for Evaluation of Schizophrenia Spectrum Disorders

Komatsu et al (2022) stated that the retina shares structural and functional similarities with the brain.  In addition , structural changes in the retina have been observed in patients with schizophrenia spectrum disorders (SSDs). In a systematic review and meta-analysis, these investigators examined retinal abnormalities and their association with clinical factors for SSD.  Studies related to retinal layers in SSD patients were retrieved from PubMed, Scopus, Web of Science, Cochrane Controlled Register of Trials, International Clinical Trials Registry Platform, and PSYNDEX databases from inception to March 31, 2021.  These researchers screened and evaluated the eligibility of the identified studies.  EZR ver.1.54 and the metafor package in R were used for the meta-analysis and a random-effects or fixed-effects model was used to report standardized mean differences (SMDs).  A total of 23 studies (2,079 eyes of patients and 1,571 eyes of controls) were included in the systematic review and meta-analysis.  The average pRNFL thickness, average macular thickness (MT), and macular ganglion cell layer-inner plexiform layer (GCL-IPL) thickness were significantly lower in patients than in controls (n = 14, 6, and 3, respectively; SMD = -0.33, -0.49, and -0.43, respectively).  Patients also had significantly reduced macular volume (MV) compared to controls (n = 7; SMD = -0.53).  The optic cup volume (OCV) was significantly larger in patients than in controls (n = 3; SMD = 0.28).  The meta-regression analysis indicated an association between several clinical factors, such as duration of illness and the effect size of the pRNFL, macular GCL-IPL, MT, and MV.  The authors concluded that thinning of the pRNFL, macular GCL-IPL, MT, and MV and enlargement of the OCV in SSD were observed.  Retinal abnormalities may be applicable as state/trait markers in SSDs.  Moreover, these researchers stated the accumulated evidence was mainly cross-sectional and needs verification by longitudinal studies to characterize the relationship between OCT findings and clinical factors.

OCT for Differentiation of Ischemic Optic Neuropathy from Glaucoma

Smeets et al (2024) stated that AION can mimic glaucoma and consequently cause difficulties in differential diagnosis.  These investigators examined differences in diagnostic tests that could aid in conducting a correct diagnosis.  The search strategy was carried out according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2009 guidelines; and 4 databases were used: Medline, Embase, Web of Science and Cochrane.  A total of 772 references were eligible; 39 were included after screening with respect to inclusion criteria that included English language and published in the 20 years before search date.  A total of 90 % (n = 35) of included studies used OCT.  Glaucomatous eyes had a significantly greater cup area, volume and depth, cup-to-disc ratio, and a lower rim volume and area, and a thinner Bruch's Membrane opening-minimum rim width.  RNFL thinning in glaucomatous eyes occurred primarily at the supero-temporal, infero-temporal and infero-nasal sectors, while AION eyes showed mostly supero-nasal thinning.  Glaucoma eyes showed greater macular ganglion cell layer thickness, except at the infero-temporal sector.  OCT-angiography measurements showed a significant decrease in superficial and deep macular vessel density (VD) in glaucoma compared to AION with similar degree of VF damage; and the para-papillary choroidal VD was spared in AION eyes compared to glaucomatous eyes.  The authors concluded that by means of OCT imaging, ONH parameters appeared most informative to distinguish between glaucoma and AION.  These investigators stated that although both diseases affect the RNFL thickness, it appeared to do so in different sectors.  In addition, differences in structure and vascularity of the macula could aid in making the differential diagnosis.

Ocular Imaging (e.g., OCT, and OCT Angiography) as a Biomarker for Dysthyroid Optic Neuropathy

In a systematic review and meta-analysis, Yang et al (2023) examined the available evidence regarding the role of OCT and OCTA in DON.  Studies in the literature that focused on OCT, OCTA and DON were retrieved by searching PubMed, Embase, Cochrane databases and Clinical Trial before June 20, 2023.  The methodological quality was assessed using the Newcastle-Ottawa scale.  The quantitative calculation was performed using Review Manager V.5.3.  A total of 12 studies met the eligibility criteria and were included.  DON group presented lower macular ganglion cell complex in the overall, superior, and inferior hemifields compared with the non-DON group.  In addition, the ganglion cell layer and inner plexiform layer in DON group was thinner in contrast to the non-DON group.  The ONH vessel density was lower in the DON group than that in the non-DON group.  A reduction of radial peri-papillary capillary vessel density could be observed in the DON group than the non-DON group in overall, inside disc, peri-papillary, superior-hemifield, temporal and nasal.  Furthermore, the macular superficial retinal capillary layer of non-DON and DON was lower than the healthy control group.  The authors concluded that this study supported the potential value of OCT and OCTA metrics as novel biomarkers of DON.  Moreover, these researchers stated that future, longitudinal studies with larger sample sizes and more potential confounders controlled are needed to confirm these findings.

The authors stated that this meta-analysis had several drawbacks.  First, heterogeneity in this meta-analysis may limit the generalizability of the pooled result; and the source of heterogeneity could not be discerned by a subgroup analysis.  Second, the number of studies in this meta-analysis was relatively small.  Third, no single protocol completely characterized DON at present, and diagnostic criteria were inconsistent across studies.  Fourth, all samples included in this analysis were from Asia, and lack of coherence from other continents.

The American Academy of Ophthalmology EyeWiki on dysthyroid optic neuropathy (Pakdel et al, 2024) stated that "Thinning of the retinal nerve fiber layer and macular ganglion cell complex could be seen by means of OCT and this finding correlates with patients' visual function. This finding is an objective way to measure atrophy or thinning prior to any fundus abnormality."

Ocular Imaging (e.g., OCT, and OCT Angiography) as a Biomarker for Peripheral Artery Disease

Prem Senthil et al (2024) examined the available evidence exploring the use of non-invasive ocular imaging for the diagnosis of peripheral artery disease (PAD).  The search was carried out in early April 2022 and included the databases Medline, Scopus, Embase, Cochrane, and others.  A total of 5 studies were included in the final review.  Of the 5 studies that used ocular imaging in PAD, 2 used retinal color fundus photography, 1 used OCT, and 2 used OCTA to examine the ocular changes in PAD.  PAD was associated with both structural and functional changes in the retina.  Structural alterations around the optic disc and temporal retinal vascular arcades were observed in color fundus photography of patients with PAD compared to healthy individuals.  The presence of retinal hemorrhages, exudates, and micro-aneurysms in color fundus photography was associated with an increased future risk of PAD, especially the severe form of the disease.  The RNFL was significantly thinner in the nasal quadrant in patients with PAD compared to age-matched healthy controls in OCT.  Similarly, the choroidal thickness in the sub-foveal region was significantly thinner in patients with PAD compared to healthy controls.  Patients with PAD also had a significant reduction in the retinal and choroidal circulation in OCTA compared to healthy controls.  As PAD causes thinning and ischemic changes in retinal vessels, examination of the retinal vessels using retinal imaging techniques could provide useful information regarding early micro-vascular damage in PAD.  The authors concluded that only a few studies had examined the association between retinal micro-vascular changes (using OCT or OCTA imaging) and PAD.  These investigators stated that there is a strong clinical need for more specific biomarkers for PAD because clinical assessment for PAD has a relatively poor predictive value.  Using a non-invasive ophthalmic imaging technique in PAD would significantly improve the rate of diagnosis without putting the patients at risk of complications.  These researchers stated that ocular imaging could potentially serve as a biomarker for PAD; and early detection of PAD and subsequent treatment may have the potential to enhance quality of life (QOL) and improve clinical outcomes for patients with PAD.

The authors stated that the study had several drawbacks.  First, this review only included 5 studies that examined the use of retinal imaging in patients with PAD; however, there is no minimum number of studies for inclusion in a systematic review and the number of studies that were included largely depends on the research topic as well as the amount of supportive evidence available.  A smaller number of studies indicated a lack of research and knowledge gaps in this field.  All the studies that were included in this review were assessed for risk of bias using the risk of bias assessment tool.  Second, only 2 studies used OCTA as an imaging modality to determine retinal changes in PAD.  This could be because OCTA is a relatively new technique and its use in a systemic disease such as PAD has not been well-established.  Third, none of the studies examined the association between retinal changes and the severity of the disease.  This might be because of the small sample size in these studies and future research examining the use of retinal imaging in PAD should also examine the relationship between retinal changes and the severity of the condition.

Optic Nerve Sheath Diameter Sonography for the Diagnosis of Intracranial Hypertension in Traumatic Brain Injury

Berhanu et al (2023) examined the optimal diagnostic cut-off of ultra-sonographic optic nerve sheath diameter (ONSD) in the diagnosis of ICP.  These investigators carried out a systematic search of available studies evaluating the use of ONSD US in patients with suspected ICP.  Meta-analysis of diagnostic accuracy of ultra-sonographic ONSD was carried out using a bi-variate model of random effects to summarize pooled sensitivity and specificity; and a summary ROC (SROC) curve was plotted.  Accuracy measures associated with ONSD cut-off and pre-defined co-variates were examined with meta-regression.  These investigators included 38 studies, comprising a total of 2,824 patients.  A total of 21 studies used invasive techniques as a reference standard estimation of IC;  and meta-analysis showed a pooled sensitivity of 0.90 (95 % CI: 0.85 to 0.93) and specificity of 0.87 (95 % CI: 0.80 to 0.91).  Optimal ONSD cut-off values ranged between 4.1 mm and 7.2 mm.  Meta-regression analysis showed that ONSD cut-off values of 5.6 to 6.3 mm were associated with higher pooled specificity compared to cut-off values of 4.9 to 5.5 mm (0.93, 95 % CI: 0.85 to 0.97 versus 0.78, 95 % CI: 0.65 to 0.87; p = 0.036).  The authors concluded that US of ONSD demonstrated a high diagnostic accuracy for ICP, with high pooled sensitivity and specificity.  Furthermore, larger cut-off values appeared to significantly increase specificity without compromising sensitivity, which supported their use as optimal ONSD cut-off.  These investigators stated that ONSD US is a fast and accessible technique that may be useful in the screening of patients with suspected ICP, however the significant heterogeneity between included studies suggested the need for further standardized studies with more homogenous criteria to accurately determine the optimal ONSD cut-off and its diagnostic accuracy.

Bakola et al (2024) noted that trans-orbital sonography (TOS) provides a non-invasive tool to detect ICP by assessing ONSD and optic disc elevation (ODE).  In a case-control, single-center study, these researchers compared TOS-acquired ONSD and ODE among intracranial hypertension (IIH)-cases versus patients with other neurological diseases (controls).  In addition, a systematic review and meta-analysis was carried out to present pooled MDs and diagnostic measures of ONSD and ODE between IIH-cases and controls.  This study consisted of 31 IIH-cases, and 34 sex- and age-matched controls; ONSD values were higher among IIH-cases than controls (p < 0.001), while ODE was more prevalent in cases (65 % versus 15 %; p < 0.001).  The ROC-curve analysis revealed that the optimal cut-off value of ONSD for predicting IIH was 5.15 mm, with an AUC of 0.914 (95 % CI: 0.861 to 0.967) and sensitivity and specificity values of 85 % and 90 %, respectively.  In a meta-analysis of 14 included studies with 415 IIH-cases, ONSD and ODE values were higher in IIH-cases than controls (MD in ONSD 1.20 mm; 95 % CI: 0.96 to 1.44 mm and in ODE 0.3 mm; 95 % CI: 0.33 to 0.67 mm).  With regard to ONSD, pooled sensitivity, specificity, and DOR were calculated at 85.5 % (95% CI: 77.9 % to 90.8 %), 90.7 % (95 % CI: 84.6 % to 94.5 %), and 57.394 (95 % CI: 24.597 to 133.924), respectively.  The AUC in summary ROC-curve analysis was 0.878 (95 % CI: 0.858 to 0.899) with an optimal cut-off point of 5.0 mm.  The authors concluded that TOS exhibited a high diagnostic utility for the non-invasive diagnosis of IIH; it represents a very promising bedside diagnostic tool that may deserve wider implementation in everyday clinical practice.

The authors stated that this case-control study and subsequent meta-analysis had several drawbacks.  First, in this trial, patients were evaluated by TOS at a single time-point and before IIH diagnosis, while subsequent measurements to estimate the potential treatment response using TOS were not part of the study protocol.  Second, regarding the meta-analysis, 6 of the included studies compared ONSD or ODE among IIH cases versus patients with other neurological diseases and not with healthy individuals.  This was also the case with their single-center study; however, when subgroup analysis stratifying by controls was performed, there was no significant difference with regard to ONSD mean difference.  On the contrary, the ODE mean difference was smaller when IIH cases were compared to patients with other neurological diseases versus healthy individuals.  Yet, the subgroup assessment for this outcome could be biased due to the limited number of studies included.  Third, other sources of bias were also disclosed among included studies in this systematic review, including selecting, performance, and publication bias.  Fourth, significant heterogeneity emerged in the assessment of most outcomes, probably owing to the different cut-off values used and the limited sample size of each study.

Xu et al (2024) stated that timely diagnosis and management of elevated ICP in patients with TBI could significantly reduce mortality rates; and US examination of the ONSD is considered a potential, non-invasive, and effective method for assessing ICP.  In a systematic review and meta-analysis of ONSD US detection and invasive ICP monitoring methods, these investigators examined the diagnostic accuracy of ONSD US detection methods for IH in patients with TBI.  They searched the Web of Science, PubMed, and Embase databases to examine the diagnostic accuracy of ONSD US for predicting increased ICP; 2 authors independently extracted the collected data.  Simultaneously, the QUADAS-2 tool was used to examine the bias risk of each study and conducted random-effects meta-analyses for the accuracy and specificity of diagnosis, and calculated pooled estimates.  A total of 10 studies with 512 patients were included.  The diagnostic accuracy of ONSD US for IH was revealed as a pooled sensitivity of 0.85 (95 % CI: 0.79 to 0.89) and specificity of 0.88 (95 % CI: 0.80 to 0.93), compared with the invasive ICP monitoring standard for patients with TBI.  The authors concluded that ONSD US may be a useful method for predicting increased ICP in adult patients with TBI.  Moreover, these researchers stated that further clinical studies are needed to confirm the diagnostic value of ONSD US.

An American Academy of Ophthalmology Preferred Practice Pattern on Idiopathic Epiretinal Membrane and Vitreomacular Traction (AAO, 2019) states that spectral domain optical coherence tomography "is a highly sensitive and routine methodology" to diagnose and characterize epiretinal membrane, vitreomacular traction, and associated retinal changes.

Optical Coherence Tomography Angiography

The American Academy of Ophthalmology EyeWiki on “Optical coherence tomography angiography” (Sousa, 2024) noted that OCTA has emerged as a non-invasive technique for imaging the microvasculature of the retina and the choroid.  The key advantages of OCTA  are the shorter acquisition time and that it is a non-invasive process.  Fluorescein and indocyanine-green angiography require an injectable dye (which takes time to reach retinal vessels, and may be associated with systemic adverse effects and even anaphylactic reactions).  OCTA has been reported to be useful in the diagnosis and understanding of many retinal conditions, including CSCR, DR, dry/wet AMD, and macular telangiectasia.  OCTA is also gaining increasing popularity for optic nerve disorders assessment (e.g., glaucoma).

Diabetic Retinopathy

Loria et al (2021) examined quantitative alterations of the choriocapillaris in swept-source optical coherence tomography angiography (SS-OCTA) in diabetic patients.  This trial included normal patients and diabetic patients with and without diabetic retinopathy (DR), excluding patients with macular edema.  Angiograms in 3×  mm were acquired with Plexelite 9000 SS-OCTA.  Choroidal flow voids were analyzed following removal of projection artifacts.  The principal evaluation was the correlation between choroidal flow voids area (FVA-CC) and DR stage.  A total of 120 eyes of 72 patients were analyzed.  There were 17 eyes from healthy subjects, 30 eyes without DR, 22 eyes with minimal non-proliferative DR, 30 eyes with moderate non-proliferative DR, 16 eyes with severe non-proliferative DR, and 5 eyes with proliferative DR (PDR).  The percentage of FVA-CC for each group was, respectively, 10.9 ± 3.4 %, 14.6 ± 4.8 %, 17.6 ± 3.5 %, 20.7 ± 5.9 %, 19.9 ± 2.9 %, and 26.6 ± 4.4 %.  FVA-CC and DR stage significantly correlated (p < 0.0001).  FVA-CC was significantly increased in diabetic patients without DR compared with healthy subjects (p = 0.008).  The authors concluded that diabetes was associated with quantifiable choriocapillaris alterations in SS-OCTA; these alterations preceded clinical signs of DR and were correlated with DR stage.

Nouri et al (2024) stated that DR is characterized by retinal vasculopathy and is a leading cause of visual impairment; OCTA is an innovative imaging technology that can detect various pathologies and quantifiable changes in retinal microvasculature.  These investigators described its functional principles and advantages over FA and carried out a comprehensive review on its clinical uses in the screening or management of individuals with pre-diabetes, diabetes without clinical retinopathy (NDR), non-PDR (NPDR), PDR, and DME.  OCTA demonstrated early microvascular alterations in pre-diabetic and NDR eyes, which may co-exist with sub-clinical neuro-retinal dysfunction.  Its uses in NPDR include measuring ischemia, detecting retinal neovascularization, and timing of early treatment via predicting the risk of retinopathy worsening or development of DME.  In PDR, OCTA aided in characterizing the flow within neovascular complexes and assessing their progression or regression in response to treatment.  In eyes with DME, OCTA perfusion parameters may be of predictive value regarding the visual and anatomical gains associated with treatment.

In an observational, cross-sectional study, Yuan et al (2024) examined the prevalence of foveal neovascularization (FNV) and its associated clinical features in PDR eyes.  Subjects underwent ultra-widefield photography, OCT, and SS-OCTA.  FNV was defined as a hyper-reflective lesion breaching the internal limiting membrane and displaying flow signal on OCTA, within 1-mm of foveal avascular zone.  Vascular metrics were obtained from the ARI Network portal.  Ischemic index (ISI) and inner choroid flow deficit percentage were calculated using FIJI from 12×12 and 6×6-mm scans, respectively.  Logistic regression models were used to compare eyes with and without FNV.  This trial included 249 eyes of 164 patients (age of 58 years [50 to 65]).  FNV was identified in 20 eyes (8 %).  Univariate logistic regression showed significant associations between FNV and younger age (p = 0.03), higher maximal HbA1c (p = 0.04), worse VA (p = 0.01), presence of disorganization of retinal inner layers (DRIL) (p = 0.01), no macular posterior vitreous detachment (PVD) (p = 0.03), neovascularization elsewhere (NVE) and at the disc (NVD) (p = 0.01 and p = 0.001), and greater ISI (p = 0.04).  In multi-variable analysis, a significant association remained between FNV and worse VA (p = 0.04), NVD (p < 0.001), DRIL (p < 0.001), and absence of macular PVD (p = 0.01).  No associations were observed with SS-OCTA vascular metrics.  The authors concluded that this study provided a comprehensive characterization of FNV in PDR.  FNV was identified in 8 % of this cohort, being more prevalent in younger patients with severe PDR, as evidenced by NVD and DRIL presence.  The absence of macular PVD may explain its association with younger age.

Furthermore, an UpToDate review on “Diabetic retinopathy: Classification and clinical features” (D’Amico and Shah, 2024) states that “Although РDR can be diagnosed by fundus examination, fluorescein angiography (a photographic study in which the transit of intravenously injected flսοrеѕceiո dye is recorded by photography with a special camera) is useful to document capillary nonperfusion and leakage from new blood vessels.  An alternative, rapid, and dye-free noninvasive technique called optical coherence tomographic angiography (ΟСΤ-A) can also be used to document retinal circulatory abnormalities, but it is less widely used and less sensitive than flսοrеѕсеiո angiography”.

Diagnosing and Monitoring of Glaucoma

In a report by the AAO, WuDunn et al (2021) examined the available evidence on the use of OCTA to aid in detecting changes associated with the diagnosis of primary open-angle glaucoma (POAG).  These investigators carried out searches of the peer-reviewed literature in March 2018, June 2018, April 2019, December 2019, and June 2020 in the PubMed and Cochrane Library data-bases.  Abstracts of 459 studies were examined to exclude reviews and non-English studies.  After inclusion and exclusion criteria were applied, 75 studies were selected, and the panel methodologist rated them for strength of evidence.  A total of 3 studies were rated level I, and 57 articles were rated level II, and the 15 level III articles were excluded.  OCTA can detect decreased capillary vessel density within the peri-papillary nerve fiber layer (level II) and macula (level I and II) in patients with suspected glaucoma, pre-perimetric glaucoma, and perimetric glaucoma.  The degree of vessel density loss correlated significantly with glaucoma severity both overall and topographically (level II) as well as longitudinally (level I).  For differentiating glaucomatous from healthy eyes, some studies found that peripapillary and macular vessel density measurements by OCTA showed a diagnostic ability (AUC) that was comparable with structural OCT retinal nerve fiber and ganglion cell thickness measurements, whereas other studies found that structural OCT measurements performed better.  Choroidal or deep-layer micro-vasculature drop-out as measured by OCTA was also associated with glaucoma damage (level I and II).  Lower peri-papillary and macular vessel density and choroidal micro-vasculature drop-out were associated with faster rates of disease progression (level I and II).  The authors concluded that vessel density loss associated with glaucoma could be detected by OCTA.  OCTA changes in glaucomatous eyes were consistent with glaucomatous pathophysiologic features and suggested that OCTA of the peri-papillary and macular microvasculature could be useful in the diagnosis of POAG.  Moreover, these investigators noted that OCTA provided complementary information to the traditional  structural and functional parameters.  It may be especially helpful in the evaluation of patients who are glaucoma suspects and in the monitoring of advanced disease when structural OCT parameters have reached the floor of their dynamic range.

The authors stated that most published studies on the use of OCTA in glaucoma diagnosis and management have been cross-sectional.  Further investigations should provide more information on key parameters of diagnostic discrimination, such as sensitivity, specificity, as well as PPV and negative predictive value (NPV).  Longitudinal studies should examine OCTA parameters for their capability of predicting glaucoma onset and progression and for their usefulness in monitoring glaucoma progression.  An important question that may be addressed is whether vascular changes would lead to structural and functional loss, or whether vascular changes result from functional loss resulting from decreased metabolic demand.  Another important question is whether OCTA findings of vessel density loss are irreversible or whether vessel density increases with lowering of IOP.  These researchers stated that as hardware and software technology advances, improvements could be expected in overcoming current limitations, such as projection artifacts, motion blur, difficulty in imaging deep optic nerve vasculature, and limited ability to quantitate blood flow through individual vessels.  These investigators stated that OCTA may improve the understanding of the mechanisms of glaucoma; therefore, resulting in improved care for patients with glaucoma.

Haja and Mahadevappa (2023) stated that as a leading cause of irreversible vision loss, glaucoma needs early detection for effective management; and IOP is a significant risk factor for glaucoma.  Convolutional Neural Networks (CNN) showed exceptional capabilities in analyzing retinal fundus images, a non-invasive and cost-effective imaging technique widely used in the diagnosis of glaucoma.  By learning from large data-sets of annotated images, CNN can identify subtle changes in the optic nerve head and retinal structures indicative of glaucoma.  This enables early and precise glaucoma diagnosis, allowing clinicians to implement timely interventions.  CNNs excel in analyzing complex medical images, detecting subtle changes indicative of glaucoma with high precision.  Another valuable diagnostic tool for glaucoma evaluation, OCT, provides high-resolution cross-sectional images of the retina.  CNN can effectively analyze OCT scans and extract meaningful features, facilitating the identification of structural abnormalities associated with glaucoma.  Visual field testing, conducted by means of devices like the Humphrey Field Analyzer, is important in evaluating functional vision loss in glaucoma.  The integration of CNN with retinal fundus images, OCT scans, visual field testing, as well as IOP measurements represents a transformative approach to glaucoma detection.  These advanced technologies have the potential to revolutionize ophthalmology by enabling early detection, personalized management, and improved patient outcomes.  CNNs facilitate remote expert opinions and enhance treatment monitoring.  Overcoming challenges such as data scarcity and interpretability can optimize CNN use in glaucoma diagnosis.  Measuring RNFL thickness as a diagnostic marker proves valuable.  CNN implementation reduces healthcare costs and improves access to quality eye care.  The authors concluded that future research should focus on optimizing architectures and incorporating novel biomarkers.  CNN integration in glaucoma detection revolutionizes ophthalmology, improving patient outcomes and access to care.  Moreover, this review did not mention OCTA as a diagnostic option..

Tsai et al (2024) noted that glaucoma, a leading cause of irreversible blindness, poses a significant health burden world-wide.  Early detection is essential for effective management and prevention of vision loss.  These investigators presented a collection of novel structural biomarkers in glaucoma diagnosis.  By means of advanced imaging techniques and data analysis algorithms, researchers can recognize indicators of glaucomatous progression.  Many research studies have demonstrated a correlation between the structural changes in the eye or brain, especially in the optic nerve head and RNFL, and the progression of glaucoma.  These biomarkers showed value in distinguishing glaucomatous eyes from healthy ones, even in the early stages of the disease.  By facilitating timely detection and monitoring, they hold the potential to mitigate vision impairment and improve patient outcomes.  The authors stated that this study marked an advancement in the field of glaucoma, offering a promising avenue for enhancing the diagnosis and possible management.

These investigators stated that OCTA is a new method that allows for non-invasive and detailed imaging of the small blood vessels in the retina, choroid, as well as optic disc area.  It also offers measurements of the amount of blood vessels in each layer, known as vascular density.  One thing that should be brought to everyone’s attention is the fact that every OCT device that is now on the market detects, represents, and analyzes OCTA signals and microvascular perfusion using a distinct algorithm; hence, the results from various OCTA devices might not be directly comparable.  Moreover, these researchers stated that relying merely on a single biomarker may not adequately evaluate glaucomatous changes; however, combining multiple biomarkers and functional tests increases the sensitivity and specificity of early glaucoma diagnosis.  They stated that further investigations are needed to examine the diagnostic and prognostic significance of these biomarkers combined with others, including perimetry and electrophysiological tests.

In a prospective study, Demir et al (2024) examined if changes in OCTA values can predict progression in the VF and RNFL in patients with glaucoma progression.  All patients in the glaucoma out-patient clinic of SBU Gulhane Medicine Faculty between 2021 and 2023 underwent RNFL, 24:2 VF, and simultaneous peri-papillary OCTA.  A total of 130 eyes that progressed were included in the study.  Thinning of more than 5 μ in any quadrant in the RNFL or a decrease of more than 1 dB in the MD value in the VF within 6 months was accepted as a progression criterion.  The ability of changes in OCTA thickness values and OCTA radial peri-papillary capillary plexus (RPCP) analysis to predict progression in RNFL and VF was prospectively examined.  The mean age of the patients was 66.9 ± 11.8 years.  There was progression in VF in 70 eyes, and RNFL in 89 eyes between baseline and 6-month controls.  The change in OCTA thickness values had insufficient ability to predict the progression of RNFL in the superior, inferior, nasal, temporal and total quadrants (p = 0.55, 0.40, 0.84, 0.91, 0.39, respectively).  The changes in OCTA thickness values failed to predict VF MD progression in the superior, inferior, nasal, temporal, and total quadrants (p = 0.40, 0.11, 0.24, 0.44, 0.10, respectively).  The changes in OCTA RPCP values did not show superiority in the ability to predict RNFL progression in superior, inferior, nasal, temporal, and total quadrants (p = 0.21, 0.53, 0.39, 0.39, 0.29, respectively).  The changes in OCTA RPCP values did not show superiority in the ability to predict VF progression in superior, inferior, nasal, temporal, and total quadrants (p = 0.96, 0.29, 0.77, 0.42, 0.21, respectively).  The authors concluded that although OCTA is a non-invasive imaging test whose use and popularity have been increasing in recent years, this study could not show superiority compared to RNFL and VF tests in demonstrating glaucoma progression.

Belbase et al (2024) compared OCTA retinal vasculature measurements in healthy, glaucoma suspect (GS), and glaucoma patients.  A total of 114 eyes with good quality OCTA pictures were included from 38 healthy, 38 GS, and 38 POAG patients.  The information on retinal vasculature was summarized as a vessel density map and as vessel density (%), which is the fraction of the flowing vessel area over the total area examined.  The superior, inferior, nasal, and temporal quadrants, as well as whole vessel density (wVD) and peri-papillary vessel density (ppVD) data taken from the RNFL, were studied.  Global indices of the visual field were correlated with vessel density measurement.  Mean vessel density was significantly lower in POAG eyes compared with GSs and healthy eyes (wVD) 45.34 % ± 6.64 %, 50.06 % ± 1.97 %, and 53.06 % ± 2.12 %, respectively (p < 0.001), and ppVD 47.42 % ± 7.73 %, 47.42 % ± 7.73 %, and 56.074 % ± 2.71 %, respectively (p < 0.001).  A linear relationship between vessel density (wVD and ppVD) and global indices of the VF (mean deviation [MD] and pattern standard deviation [PSD]) showed a significant (p < 0.001) relation.  R2 was 0.35, 0.36, 0.36, and 0.33 for MD and wVD, PSD and wVD, PSD and ppVD, and MD and ppVD, respectively.  The authors concluded that for distinguishing between healthy and glaucoma eyes, OCTA vessel density showed near similar diagnostic accuracy to VF tests.  These researchers stated that these findings implied that OCTA measurements reflected damage to tissues important in the pathogenesis of POAG.

The authors stated that this trial had several drawbacks.  First, it was unclear whether the decreased vessel density is the result of glaucomatous optic nerve head (ONH) damage or the cause of it.  According to Hollo (2016) measures of peri-papillary angio-flow density could detect decreased peri-papillary perfusion in early glaucoma before the development of severe RNFL destruction and VF degeneration.  During the OCTA scanning, no blood pressure (BP) measures were taken in this study. A  link between BP and ONH blood flow is attainable; however, previous research found no significant link between BP and ONH vessel density.  Second, these investigators  were unable to analyze intra-observer variability since the OCT images were obtained by a single operator but only 1 measurement was taken.  Third, in the glaucoma group, the majority of the patients were using several ocular anti-hypertensive eye-drops.  With the small sample size (n = 38 in the POAG group), these researchers could not assess their individual effects on disc perfusion, and they could not completely rule out the idea that the glaucoma drops were to blame for the lower disc perfusion.  Fourth, the RNFL’s retinal ganglion cells are primarily concentrated in the macula, and several studies have established a strong association between ganglion cell-inner plexiform layer thickness and glaucoma.  However, this trial did not examine the vessel density in the macula.

In a prospective,  cross-sectional study, Angmo et al (2024) compared the diagnostic ability of macular ganglion cell inner plexiform layer (mGCIPL) and peri-papillary RNFL (pRNFL) thickness on OCT and macular and peri-papillary perfusion changes using OCTA in GS and early POAG.  A total of 90 patients (30 normal, 30 GS, and 30 early POAG) were recruited in this trial.  The average thickness of mGCIPL and pRNFL on spectral domain-OCT and macular vessel density, ONH perfusion, and ONH flux index (FI) on OCTA were evaluated for early diagnosis of glaucoma.  Macular vessel density, ONH perfusion, and ONH FI were significantly reduced in early POAG eyes compared to normal.  The best correlation was observed between ONH FI and the average RNFL in both glaucoma suspects (r = 0.47, p < 0.01) and early POAG patients (r = 0.53, p < 0.01).  Out of all the measured OCTA parameters, only ONH perfusion was significantly lower in glaucomatous eyes compared to glaucoma suspects (p < 0.001).  Average GCIPL (0.82) and macular vessel density (0.76) had the highest AUC curve value among all the OCT and OCTA parameters, respectively, for differentiating glaucoma suspects from controls.  Rim area (0.92) and ONH FI (0.81) had the highest AUC curve value among all the OCT and OCTA parameters for differentiating early POAG patients from controls.  The authors concluded that OCTA vascular parameters had a good correlation with structural damage both at the disc and the macula.  However, OCT parameters were superior to OCTA parameters for diagnosis of glaucoma, although OCTA parameters are deranged very early in the disease.

Wu et al (2024) examined longitudinal OCTA changes in macula and ONH in healthy, GS and POAG eyes.  Healthy, GS and POAG eyes from Diagnostic Innovations in Glaucoma Study with 2 years or more follow-up and 4 visits of macular/ONH OCTA imaging were included.  Rates of macular wiVD (whole-image vessel density) and ONH wiCD (whole-image capillary density) changes were calculated for each diagnosis group using join mixed-effect modelling.  Differences in wiVD/wiCD change rates across diagnoses were examined via pair-wise comparison.  Relationships of baseline 24-2 VF mean deviation (MD) with wiVD/wiCD change rates were evaluated for POAG eyes.  There were 36, 57 and 79 eyes (25, 38 and 50 subjects) in healthy, GS and POAG groups, respectively.  Rates of wiVD (range of -0.72 to -0.92 %/year) and wiCD (range of -0.28 to -0.66 %/year) loss were different from zero in all groups (p < 0.05).  The rates of wiCD loss differed across all diagnosis groups (p < 0.001), while wiVD change rates did not increase when comparing healthy to GS eyes (p = 0.167).  Baseline VF MD showed a significant but modest correlation with the rates of both wiVD and wiCD loss (p < 0.05), and the correlation with wiCD change rate was slightly stronger (R2 = 0.27 versus 0.16).  The authors concluded that in glaucoma, there was earlier micro-vasculature loss in the ONH than in the macula.  Moreover, ONH vessel density loss revealed a slightly stronger association with baseline VF than macular VD.  These researchers stated that observing ONH vessel density loss with OCTA may aid in monitoring early glaucoma, which should be confirmed by future larger studies.

Furthermore, UpToDate reviews on “Open-angle glaucoma: Epidemiology, clinical presentation, and diagnosis” (Jacobs, 2024), “Angle-closure glaucoma” (Weizer, 2024), and “Overview of glaucoma in infants and children” (Reynolds and Reynolds, 2024)) do not mention optical coherence tomography angiography as a diagnostic/monitoring tool.

Age-Related Macular Degeneration

In a prospective, comparative study, Ma et al (2024) examined the differences of CNV measurements between SS-OCTA and spectral-domain OCTA (SD-OCTA) in neovascular AMD (nAMD) and the imaging reliability of the 2 devices.  SS-OCTA and SD-OCTA were used to scan the same eye with the modes of 3×3 and 6×6 mm2 centered on the neovascularization.  Only qualified images were chosen and the border of CNV was manually delineated by 2 graders independently.  The area of CNV (ACNV), vascular perfusion density (PD), and vessel length density (VLD) within the delineation were calculated using Image J.  The differences of CNV measurements between the 2 OCTA devices were compared using Bland-Altman analysis.  The agreement between the 2 graders on the measurements of each device was compared using the intra-class correlation coefficient (ICC).  A total of 18 patients (22 eyes) with nAMD were included.  The measurements of ACNV, PD, and VLD were 7.247 ± 4.586 and 4.901 ± 3.741 mm2, 43.202 ± 9.636 and 34.904 ± 10.489, 6.339 ± 1.228 and 5.908 ± 1.741 mm-1 for SS-OCTA and SD-OCTA, respectively.  The differences between the 2 devices were 2.346 ± 3.030 mm2 (Z = -3.782, p < 0.0001), 8.298 ± 14.160 (Z = -2.419, p = 0.016), and 0.431 ± 2.114 mm-1 (Z = -0.828, p = 0.408) for ACNV, PD and VLD, respectively.  The ICC between 2 graders were 0.893 (p < 0.001), 0.902 (p < 0.001), 0.885 (p < 0.001) for ACNV, PD, VLD in SS-OCTA, and 0.971 (p < 0.001), 0.976 (p < 0.001), 0.973 (p < 0.001) in SD-OCTA, respectively.  The authors concluded that both OCTA devices exhibited high imaging reliability.  Compared with SD-OCTA, SS-OCTA had a larger ACNV measurements, but did not demonstrate better resolution of internal vessels of CNV and well signal strength.

Furthermore, an UpToDate review on “Age-related macular degeneration” (Vavvas, 2024) states that “Additional specialized imaging of the retina includes color photograph of the retina, auto-fluorescence imaging of the retina, and an ΟCT.  Depending on the suspected stage (intermediate to advanced wet ΑMD), a fluorescein angiogram or an indocyanine green angiogram, as well as ՕCT angiography may also be obtained to evaluate for the presence of wet AΜD”.

Ultrasonographic Assessment of Optic Nerve Sheath Diameter as a Screening Tool for Intracranial Hypertension in Traumatic Brain Injury

Lioi et al (2024) noted that severe TBI is a condition burdened by high morbidity and mortality.  Prevention of secondary insults is one of the main objectives of management and intra-cranial pressure (ICP) monitoring is a cornerstone in management of TBI.  The relationship between ICP and optic nerve sheath is known from the literature.  Optic nerve sheath US could represent a method added to the armamentarium for monitoring ICP.  These investigators examined how US-measured optic nerve sheath diameter (ONSD) varies as a function of ICP in a cohort of patients with severe blunt head injury in whom an intra-parenchymal sensor was placed (20 adult TBI patients: 14 males and 6 females, with a mean age of 43.1 ± 15.8 years; range of 18 to 76).  They evaluated the accuracy of ONSD in distinguishing dichotomized ICP cut-offs and analyzed the learning curve and its potential as screening tool to select TBI patients most in need of invasive ICP monitoring.  ONSD and ICP have a linear relationship; however, there are limits of evaluating the one-to-one correspondence between those 2 variables.  These researchers selected a cut-off of sonographic ONSD above which there is a concerned elevation of ICP (ICP greater than 15 mmHg) worthy of invasive 2nd-line monitoring.  Therefore, it is possible to use ONSD as a 1st-line non-invasive tool to intercept patients at risk of developing frank intra-cranial hypertension (ICH).  The authors proposed the use of ONSD US as a screening investigation for post-traumatic ICH in the context of an emergency department, especially in contexts where there is limited availability of ICP monitors.

The authors stated that this study had 2 main drawbacks.  First, the sample examined was small (n = 20), because this was a preliminary study for the validation of a diagnostic practice recently introduced in the authors’ center.  Second, these researchers had few ICP measurements above 20 mmHg; this was mainly due to the fact that patients in whom it would have been possible to measure that value presumably directly underwent urgent surgery due to a clinical-radiological certainty of the necessity of it.  Nevertheless, these investigators believed it is important to emphasize the possibility of performing this diagnostic maneuver having high clinical value (i.e., measuring ONSD) especially for centers where there is no immediate availability of intra-parenchymal ICP sensors.

Ferreira et al (2025) stated that ICP can worsen the clinical condition of patients with TBI.  One non-invasive and easily bedside-performed technique to estimate ICP is US measurement of ONSD.  These investigators examined ONSD and correlated it with ICP values obtained by intra-parenchymal monitoring to establish the ONSD threshold value for elevated ICP and reference range of ONSD in severe TBI patients.  A total of 40 severe TBI patients (Glasgow Coma Scale [GCS] score of 8 or less) were included.  US measurement of ONSD was carried out and compared with intra-parenchymal ICP monitoring to examine their association and determine the ONSD threshold value.  Exclusion criteria included individuals under 18 years of age, penetrating TBI, or direct ocular trauma.  A total of 53 ONSD measurements were carried out in all patients.  The mean ONSD value in the group with ICP of less than 20 mmHg was 5.4 mm ± 1.0, while in the group with ICP of 20 mmHg or higher, it was 6.4 mm ± 0.7 (p = 0.0026).  A positive and statistically significant correlation, albeit weak (r = 0.33), was observed between US measurement of ONSD and intra-parenchymal ICP monitoring.  The statistical analysis of the ROC curve identified the best cut-off as 6.18 mm, with 77.8 % sensitivity and 81.8 % specificity.  The authors concluded that the findings of this study showed a positive, albeit weak, correlation between US measurement of ONSD and intra-parenchymal ICP monitoring, with an ONSD threshold value of 6.18 mm.  Achieving only 77.8 % sensitivity and considering the substantial variability between ONSD measurements (SD at 1.0) might limit the reliability of ICP assessment based solely on ONSD measurements.

Automated Measurements of Optic Nerve Sheath Diameter for Assessment of Intracranial Pressure in Aneurysmal Subarachnoid Hemorrhage

Netteland et al (2025) stated that ONSD is a promising non-invasive parameter for ICP assessment; however, in the setting of aneurysmal subarachnoid hemorrhage (aSAH), several previous studies have reported no association between US-measured ONSD and ICP.  In a prospective, single-center study, these researchers assessed ONSD in patients with aSAH using a novel method of automated real-time US measurements and examined if factors such as having undergone surgery would affect its association to ICP.  This trial included adult patients with aSAH undergoing invasive ICP monitoring; ONSD was obtained using a prototype US machine with software for real-time automated measurements at the bedside.  Correlation between ONSD and ICP was examined, and the ability of ONSD to discriminate dichotomized ICP was evaluated.  Afore-mentioned analyses were conducted for the whole cohort and repeated for subgroups by whether the basal cisterns had been surgically entered before US examination.  A total of 26 US examinations were carried out in 20 patients.  There was a positive correlation between ONSD and ICP (R = 0.43; p = 0.03).  In the subgroup where the basal cisterns had not been surgically entered before US examination, there was a stronger correlation (R = 0.55; p = 0.01), whereas no correlation was observed in the subgroup where the basal cisterns had been surgically entered (R = - 0.16; p = 0.70).  ONSD displayed an ability to discriminate ICP dichotomized at 15 mm Hg or higher (AUC = 0.84, 95 % CI: 0.65 to 0.96).  Subgroup analysis showed a perfect discriminatory ability (AUC = 1, 95 % CI: 0.81 to 1) where the basal cisterns had not been surgically entered and no discriminatory ability (AUC = 0.47, 95 % CI: 0.16 to 0.84) where the basal cisterns had been surgically entered before US examination.  The authors concluded that automatically measured ONSD correlated well with ICP and demonstrated a perfect discriminatory ability in patients with aSAH in whom the basal cisterns had not been entered surgically before US examination, and may be a clinically valuable non-invasive marker of ICP in these patients.  Moreover, these researchers stated that caution should be exercised in using ONSD in patients in whom the basal cisterns have been entered surgically before ONSD measurements, as no association was observed in this subgroup.

The authors stated that this trial had several drawbacks.  First, it was a single-center study; thus, findings require verification.  Second, the sample size was small (n = 20 patients), which had implications for power in the analyses, especially so in the subgroup analyses.  Third, the number of observations in the high ICP range was limited.  This was related to the patient group studied, as a main objective of neuro-intensive care of patients with aSAH is to manage high ICP.  Although the typical treatment threshold in aSAH is 20 mm Hg, what constitutes “normal ICP” is by most considered lower, and limits of 10 mm Hg and 15 mm Hg have been suggested.  Moreover, the threshold associated with harm on a group level is not known with certainty, and data have suggested that sustained ICP of greater than 10 mm Hg may be associated with worse outcomes in ICP-monitored neuro-critically ill patients.  To examine ONSD’s ability to discriminate ICP, these investigators chose to dichotomize ICP at 15 mm Hg or higher to maintain a relative group size balance.  Although this dichotomization does not reflect the typical treatment threshold used, these researchers believed that using this cut-off still provided valid information on the discriminatory ability of ONSD.  Fourth, the authors observed that the ICP values did not distribute along approximate normality, which implied that the p-values of the Pearson correlations in which outliers were not excluded should be interpreted with some caution.  The distribution of the observed ICP values may be specific to the current data set but may also be seen in context with the non-linear pressure-volume curve relating to the Monro-Kellie doctrine.  Statistically, these researchers explored both natural log transformation of ICP and exclusion of the 2 ICP outliers.  Both approaches yielded approximated normal distributions of ICP, and both yielded very similar end results in terms of strength of correlation between ONSD and ICP.  Because the 2 outliers differed qualitatively from the rest of the observations in that active management had been withdrawn and cerebral perfusion pressure was near zero, indicating tamponade, these investigators chose exclusion of the outliers, as this approach was considered better rooted in physiological factors and not just statistical ones.  However, natural log transformation of ICP did yield a more linear relationship between ONSD and ICP for all ICP values.  Whether this is a trait specific to the current data set or holds true also in external data sets warrants further consideration.

References