Aetna considers magnetic resonance spectroscopy (MRS) (also known as NMR spectroscopy) experimental and investigational because there is a lack of evidence of its efficacy in the medical literature.

Magnetic resonance spectroscopy (MRS), also known as nuclear magnetic resonance (NMR) spectroscopy, is a non-invasive analytical technique that has been used to study metabolic changes in brain tumors, strokes, seizure disorders, Alzheimer's disease, depression and other diseases affecting the brain. It has also been used to study the metabolism of other organs.

MRS can be done as part of a routine MRI on commercially available MRI instruments. The probe accessory necessary to perform MRS was granted 510(k) clearance from the FDA. MRS and MRI use different software to acquire and mathematically manipulate the signal. Whereas MRI creates an image, MRS creates a graph or 'spectrum' arraying the types and quantity of chemicals in the brain or other organs.

The role of MRS in diagnosis and therapeutic planning has not been established by adequate clinical studies. Specifically, there have been no clinical trials demonstrating improved outcomes in patients evaluated with MRS compared to patients evaluated with conventional imaging modalities.

The consensus in the literature is that further studies are necessary to determine MRS' role in the diagnosing and planning treatment in neurological diseases.

An assessment of MRS prepared by the Tuft's-New England Medical Center Evidence-Based Practice Center for the Agency for Healthcare Research and Quality (AHRQ) (Jordan et al, 2003) reached the following conclusions: “Human studies conducted on the use of MRS for brain tumors demonstrate that this non-invasive method is technically feasible and suggest potential benefits for some of the proposed indications. However, there is a paucity of high quality direct evidence demonstrating the impact on diagnostic thinking and therapeutic decision-making. In addition, the techniques of acquiring the MRS spectra and interpreting the results are not well standardized. In summary, while there are a large number of studies that confirm MRS' technical feasibility, there are very few published studies to evaluate its diagnostic accuracy and whether it can positively affect diagnostic thinking and therapeutic choice. Those studies that do address these areas often have significant design flaws including inadequate sample size, retrospective design and other limitations that could bias the results.”

A structured evidence review of MRS for evaluation of suspected brain tumor conducted by the BlueCross BlueShield Association Technology Evaluation Center (2003) concluded that “[t]he evidence is insufficient to permit conclusions concerning the effect of magnetic resonance spectroscopy on health outcomes.”

The Center for Medicare and Medicaid Services (2004) has determined that there is insufficient evidence to deem MRS “reasonable and necessary” for brain tumor diagnosis. Due to “methodological shortcomings” in the 11 studies reviewed on the use of MRS for brain lesion detection and a lack of a controlled comparison of MRS and traditional diagnostic strategies, CMS has announced that it will continue its current national non-coverage determination.

Gluch (2005) stated that ex vivo and in vivo applications of MRS have been developed, which aid in distinguishing malignant from normal tissues. Studies of breast, colon, cervix, esophageal, and prostate cancer reveal both the successes and failings of present technology. The author noted that verification that these non-invasive tests might supplant conventional histology in obtaining spatial diagnostic and chemical prognostic information remains for the time being illusive.

Willmann et al (2006) evaluated the additional pre-operative value of (1)H MRS in identifying the epileptogenic zone (EZ) for epilepsy surgery by performing a meta-analysis. The authors concluded that MRS still remains a research tool with clinical potential. Their findings indicated the connection of ipsilateral MRS abnormality to good outcome. The ability for prediction of post-operative outcome may depend on the assessed population. They noted that prospective studies limited to non-localized ictal scalp electroencephalography or MRI-negative patients are needed for validation of these data. Furthermore, Hollingworth et al (2006) stated that (1)H MRS is a potentially useful adjunct to anatomical MRI in the characterization of brain tumors. These investigators performed an updated systematic review of the evidence. They concluded that the current evidence on the accuracy of (1)H MRS in the characterization of brain tumors is promising. However, additional high-quality studies are needed to convince policy makers.

Wetter et al (2006) evaluated a routine protocol for combined MR and spectroscopic imaging of the prostate for staging accuracy. A total of 50 patients with biopsy-proven prostate carcinoma were examined with a sequence protocol, which consisted of T2-weighted fast spin-echo sequences and a pelvic T1-weighted spin-echo sequence. For spectroscopy, a 3D chemical shift imaging (CSI) spin-echo sequence was used. Image interpretation was performed by two radiologists. The total number of tumor voxels and tumor voxels per slice were counted to estimate the tumor volume in every patient. The potential of MRS to differentiate between T2 and T3 tumors, based on the estimated tumor volumes, was compared with the staging performance of MRI. The MR measurement time was 19.01 minutes, and the total procedure time averaged 35 minutes. Seventy-six percent of the spectroscopic examinations were successful. Statistically significant differences in the number of tumor voxels per slice and tumor volumes were found between T2 and T3 tumors. The descriptive parameters of MRI and MRS did not differ significantly; sensitivity and specificity were 75 % and 87 %, respectively, for MRI; and 88 % and 70 %, respectively, for MRS. The combination of both methods resulted in only a slight improvement in staging performance and was not statistically significant. The authors concluded that combined MRI and MRS of the prostate has no diagnostic advantage in staging performance over MRI alone. The mean tumor volumes, estimated by MRS, differ statistically significantly between T2 and T3 tumors.

Rajesh et al (2007) noted that three-dimensional MRS is emerging as a new and sensitive tool in the metabolic evaluation of prostate cancer. Zapotoczna et al (2007) stated that the increasing sensitivity and specificity of MRS to the prostate is causing new interest in its potential role in the definition of target subvolumes at higher risk of failure following radical radiotherapy. Prostate MRS might also play a role as a non-invasive predictive factor for tumor response and treatment outcome. However, guidelnes on the pre-treatment staging of prostate cancer by the American College of Radiology (ACR)'s expert panel on urologic imaging and radiation oncology (Israel et al, 2007) stated that one group of investigators have demonstrated that prostate cancers have a characteristic loss of the citrate peak and gain in the choline/creatine peak on MRS imaging. Moreover, the ratio of choline to citrate is related to the Gleason score, suggesting that MRS imaging may provide information about tumor aggressiveness. Improvements in diagnostic accuracy and staging have been reported. However, MRS imaging is technically demanding and time consuming. It has not been proven in multi-institutional trials, although a clinical trial under the auspices of the American College of Radiology Imaging Network (ACRIN) is currently underway. Thus, MRS imaging can not yet be considered a routine diagnostic tool.

Vedolin and colleagues (2007) examined the influence of aging on conventional MRI and MRS findings of patients with mucopolysaccharidosis (MPS), and tested the correlation of enzyme levels, urinary glycosaminoglycans (GAG), and neuroimaging findings. A total of 60 patients with MPS types I (n = 8), II (n = 31), IV-A (n = 4), and VI (n = 17) underwent T2, fluid-attenuated inversion recovery (FLAIR), and MRS of the brain. For analysis of MRI variables, the researchers measured the normalized cerebral volume (NCV), CSF volume (NCSFV), ventricular volume (NVV), and lesion load (NLL) on FLAIR using semi-automated and automated segmentation techniques. For MRS, a point-resolved spectroscopy technique was used. Voxels were positioned at the white and gray matter. Statistical analysis involved Pearson or Spearman tests for correlation between neuroimaging, age, enzyme levels, and urinary GAG. The median age at onset of the disease was 20 months. Patients with longer disease duration had more NLL in the white matter (r = 0.28, p = 0.03), and this difference was more pronounced in MPS II patients (r = 0.44, p = 0.02). Metabolites ratios in MRS, NCV, NCSFV, and NVV did not correlate with disease duration or age of the patients (p > 0.05). MRI and MRS variables in either the white or the gray matter did not correlate with enzymatic activity or GAG levels. Patients with MPS II had a lower mean NCV (p < 0.001). The authors concluded that these findings showed that white matter lesion is more extensive as disease duration increases, especially in mucopolysaccharidosis type II patients. MRI and MRS findings did not correlate with either enzymatic or glycosaminoglycan levels.

Boesch et al (2007) evalauted and compared biochemical and volumetric features of the cerebellum in patients with spino-cerebellar ataxia type 2 (SCA2) and patients with the cerebellar variant of multiple system atrophy (MSA-C). Nine genetically assigned SCA2 patients and 6 MSA-C patients who met the clinical criteria of MSA-C underwent a clinical and neuro-radiological workup with respect to cerebellar features. The MR protocol consisted of a sagittal T1-weighted 3-dimensional fast low-angle shot (3D FLASH) sequence and a transversal T2- and spin-density-weighted turbo spin-echo sequence. The proton magnetic resonance spectroscopic imaging ((1)H-MRSI) protocol consisted of two chemical shift imaging (CSI) sequences (echo time (TE) = 20 and 135 msec). Both short- and long-TE MR spectroscopy (MRS) images showed significant decreases in values for N-acetylaspartate to creatine (NAA/Cr), and choline to creatine (Cho/Cr) ratios in MSA-C and SCA2 compared to normal controls, though there was no difference between the two patient groups. In contrast, distinct cerebellar lactate (Lac) peaks were detected in seven SCA2 patients, and small peaks were detected in two. However, these investigators did not detect any definite Lac peak in MSA-C or control subjects. The authors concluded that MRSI revealed Lac pathology in SCA2 but not in MSA-C. Whether this indicates distinct pathogenetic mechanisms of cerebellar degeneration remains to be established.

Dyke et al (2007) explored (1)H MRSI as a means to assess peri-tumoral tissue response post-resection and Gliadel((R)) implantation in patients with high-grade gliomas. Pilot (1)H MRSI data are presented that demonstrate non-invasive, serial monitoring of metabolic changes at the tumor site following Gliadel implantation. Three patients with newly diagnosed glioblastoma multiforme (GBM) underwent MRI and (1)H MRSI at 3.0 Tesla prior to resection and at 3-5 and > or =12 weeks post-operatively. Baseline MRS spectra of tumor tissue from all patients were characterized by marked increases of choline (CHO) and lactate (LAC), and a decrease of N-acetylaspartate (NAA), typical of GBM compared with normal contra-lateral brain tissue. Post-operatively, spectra were analyzed from the resection cavity and peri-tumoral regions and compared with normal tissue from the contra-lateral brain at baseline. In 2 of 3 patients, peri-tumoral NAA/CRE increased and CHO/NAA decreased compared to contra-lateral brain at 3-5 weeks compared with baseline following Gliadel therapy and surgery but prior to radiotherapy. This study indicated that (1)H MRSI has the ability to localize regions of heterogeneous response following Gliadel treatment. Although data are limited, these results suggested that metabolic indicators of outcome can be successfully monitored pre- and post-surgical resection and Gliadel implantation with (1)H MRSI. Additional study of patients receiving Gliadel Wafers using (1)H MRSI may serve to aid clinicians in assessing tumor regression and gauging efficacy of this chemotherapy treatment.

De Stefano et al (2007) reviewed current MRS clinical applications in multiple scloersis (MS), and discussed the potential and limitations of the technique, and suggested recommendations for the application of MRS to clinical trials. The authors concluded that despite some important limitations, proton MRS has the potential to be implemented in large, multi-centere clinical trials of MS. The usefulness of MRS-derived outcome measures in MS clinical trial has yet to be proven....Future studies and the few clinical trials that are currently incorporating MRS into their imaging protocols will reveal if MRS has a role in quantifying the impact of therapeutic intervention on tissue damage in MS and will help to determine if MRS can become a standard and accepted part of the assessment of MS treatment in the near future.  European Federation of Neurological Societies guidelines on the use of neuroimaging in the management of MS (Filippi et al, 2006) noted that the performance and contribution of diffusion tensor MRI (DT-MRI) and MRS) in multi-center studies still have to be evaluated.

Biomarkers of disc degeneration have been previously described using NMR spectroscopy, but the link between discogenic back pain and biomarkers has not been completely understood. Keshari et al (2008) used quantitative ex vivo proton high resolution magic angle spinning (HR-MAS) NMR spectroscopy to identify biochemical markers associated with discogenic back pain. HR-MAS NMR spectroscopy was performed on snap frozen samples taken from 9 patients who underwent discectomies for painful disc degeneration. The resulting proton NMR spectrums were compared with those from discs harvested from a reference population consisting of 9 scoliosis patients. Spectral analyses demonstrated significantly lower proteoglycan (PG)/collagen (0.31 +/- 0.22 versus 0.77 +/- 0.48) and PG/lactate (0.46 +/- 0.24 versus 2.24 +/- 1.11) ratios, and a higher lactate/collagen (0.77 +/- 0.49 versus 0.40 +/- 0.21) ratio in specimens obtained from discogenic pain patients when compared with scoliosis patients. The authors concluded that these findings suggested that spectroscopic markers of proteoglycan, collagen, and lactate may serve as metabolic markers of discogenic back pain. These results provided a further basis of the potential to develop in vivo MR spectroscopy for the investigation of discogenic back pain.

Guidelines on bone tumors by ACR's expert panel on musculoskeletal imaging (Morrison et al, 2005) noted that MRS has potential to differentiate benign from malignant lesions, however, more research is needed.