Novel Eye Measure Finds Neuropathy in Prediabetes

Miriam E. Tucker

July 09, 2014

People with impaired glucose tolerance (IGT) have evidence of neuropathy that can be detected via the eye using corneal confocal microscopy, a new study finds.

The results were published online June 26 in Diabetes Care by Omar Asghar, of the University of Manchester, United Kingdom, and colleagues.

Used initially in ophthalmology by the Manchester group and others, corneal confocal microscopy (CCM) is now increasingly being adopted as a surrogate measure of small-fiber neuropathy in a variety of conditions including diabetes, principal investigator Rayaz A. Malik, professor of medicine and consultant physician at Weill Cornell Medical School, Qatar, and Central Manchester University Teaching Hospitals, United Kingdom, told Medscape Medical News.

"The use of corneal confocal microscopy to assess neuropathy has steadily grown, and it now has more publications than skin biopsy, which is the so-called gold standard for assessing small-fiber neuropathy," Dr. Malik said.

The group's new study included 37 patients aged 30 to 75 years with 2-hour oral glucose tolerance test values of 7.8 to 11.1 mmol/L (about 140 to 200 mg/dL) and 20 age-matched controls. Use of corneal confocal microscopy revealed that 15 of the IGT group (40.5%) had corneal nerve fiber densities that were 2 standard deviations below those of the control group's average of 24 n/mm2.

This suggests that "IGT is no longer a benign condition, as it is associated with significant nerve damage in about 40% of subjects. Corneal confocal microscopy uses a simple, noninvasive ophthalmic instrument that allows you to detect this early nerve damage and risk-stratify your subjects with IGT," Dr. Malik told Medscape Medical News.

But Peter J. Dyck, MD, professor of neurology and director of the Peripheral Nerve Research Laboratory at the Mayo Clinic, Rochester, Minnesota, disagrees. In 2012, he led a study that found no increased prevalence of either typical or atypical diabetic polyneuropathy among people with impaired glycemia.

"I find the evidence less compelling that IGT on its own causes [typical] …or atypical diabetic polyneuropathy. Physicians who evaluate patients with impaired glycemia and polyneuropathy should encourage weight loss and increased exercise with the goal of preventing type 2 diabetes but should also search for other causes of diabetic polyneuropathy, especially treatable ones," Dr. Dyck told Medscape Medical News.

Dr. Dyck does not advocate using estimates of corneal nerve fibers as a measure of diabetic polyneuropathy in diabetes patients, either. "Even in the most severe patients with diabetic polyneuropathy, decrease of corneal nerve fibers is asymptomatic. To use estimates of corneal nerve fibers — assuming they are decreased — as a surrogate measure of diabetic polyneuropathy would require much more research than reported here or in previous articles," he asserts.

Significant Differences in IGT vs Controls

In the study, the 37 IGT subjects and 20 controls underwent a comprehensive evaluation of neuropathy by assessing symptoms, neurological deficits, nerve-conduction studies, quantitative sensory testing, heart-rate-variability deep breathing, skin biopsy, and corneal confocal microscopy

Those with IGT had significantly higher values on the neuropathy symptom profile (4.1 vs 0.5; P < .001), the McGill pain index (2.8 vs 0.2; P < .001), the neuropathy disability score (2.9 vs 0.6; P = .001), vibration perception threshold (15.9 vs 6.5; P = .002), and warm threshold (40.6 vs 37.6; P = .006).

The IGT group also had significantly lower values on intraepidermal nerve fiber density as assessed by skin biopsy (P = .03) and of corneal nerve fiber density (27.6 vs 37.4; P < .001), corneal nerve branch density (P = .02), and corneal nerve fiber length (P = 0.05), all 3 of which were assessed by corneal confocal microscopy.

However, there were no significant differences in neurophysiology, including sensory and motor nerve amplitude and conduction velocity, or in heart-rate-variability deep breathing.

Compared with the IGT patients without neuropathy, those with neuropathy had significantly greater self-reported pain intensity on the McGill pain index ( P = .04) and significantly lower corneal nerve fiber density (P = .02) and corneal nerve fiber length (P < .001), the authors report.

"This study shows evidence of neuropathy in subjects with IGT, as evidenced by abnormal symptoms, signs, quantitative sensory testing, skin biopsy, and corneal confocal microscopy, but not neurophysiology," the authors conclude.

Debating the Findings

Dr. Malik told Medscape Medical News that corneal confocal microscopy detects small-fiber structure while nerve-conduction tests detect function of large myelinated fibers, which constitute only about 10% of peripheral nerves.

Small fibers mediate pain and blood flow and therefore underlie the 2 main outcomes of diabetic neuropathy: painful neuropathy and foot ulceration. Detection of small-fiber structure, constituting 90% of peripheral nerve fibers, allows for quantification of both degeneration and regeneration, he said.

Previous work by the Manchester group has shown that corneal confocal microscopy detects the earliest evidence of nerve damage in patients who have normal results on nerve-conduction studies and other tests of neuropathy (Diabetes Care. 2013;36: 3646-3651), Dr. Malik said.

He added that this new study calls into question the diagnostic cutoff for diabetes, which was based on epidemiological studies using retinopathy as a basis to define all complications.

"Indeed, these findings challenge this concept, as we are showing the presence of nerve damage at lower levels of abnormal glucose and indeed before the occurrence of retinopathy [Diabet Med. 2014;31: 431-438]."

But Dr. Dyck said that the new study suffers from several methodological problems, including lack of statistical power due to its small size and lack of a definition for either typical or atypical diabetic polyneuropathy.

"What they show is a different distribution of test values between the 2 comparative groups, inferring that it is from a difference of IGT. In fact, the difference might be explained by improper selection of patients and controls, inadequate correction of test results for differences in anthropomorphic variables, or nonrecognition of other neuropathic diseases," he told Medscape Medical News.

And, he noted, "A major concern is that much larger studies, using highly standardized and referenced approaches and masked assessment of patients, did not show the differences in measured attributes of nerve conduction, vibratory detection thresholds, and other end points they show with IGT."

How Should Neuropathy Be Measured?

According to Dr. Dyck, "We do not need additional time-consuming, technical, and expensive surrogate markers of diabetic polyneuropathy. We already have reliable measures of neuropathic signs, neurophysiologic tests, and quality-of-health scores by which to measure not only the occurrence but also the severity of both typical and atypical small-fiber diabetic polyneuropathy."

But Dr. Malik argues that conventional testing does not detect small-fiber neuropathy in patients with IGT. He told Medscape Medical News, "I would not recommend either nerve conduction or autonomic testing, as these were normal. If I were going to be selective as to which IGT subject should have corneal confocal microscopy, then perhaps those with neuropathic symptoms."

Dr. Dyck notes, however, that "irrespective of which [testing] approach is used, an independent assessment of the cause of diabetic polyneuropathy must be made considering more than the IGT status."

This research was funded by awards from the National Institutes of Health and JDRF International. Dr. Malik and coauthors have reported no relevant financial relationships. Dr. Dyck has received financial support within the last year from ISIS Pharmaceuticals and Alnylam to teach investigators for trials of transthyretin amyloid polyneuropathy. His laboratory receives financial support from these 2 pharmaceutical organizations for quality assurance of trial data.

Diabetes Care. Published online June 26, 2014. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.