Corneal Confocal Microscopy as a Measure of Diabetic Neuropathy

Distal symmetric polyneuropathy is the most common pattern of nerve injury in patients with diabetes (1,2). In fact, up to 50% of patients with diabetes will develop distal symmetric polyneuropathy at some point during their illness (3). Currently, the only effective treatment to prevent this prevalent condition is glucose control (4). Unfortunately, demonstrating an improvement in neuropathy over time has been much more difficult to achieve than preventing progression (5). Possible explanations for the lack of improvement in outcome measures include a lack of an effective treatment and/ or a lack of a sensitive test for nerve fiber repair. No consensus exists as to which neuropathy outcomes should be obtained after a therapeutic intervention, and all currently available options have advantages and disadvantages. Clinical assessments and scales have the advantage of taking into account patients’ symptoms and neurologic examination, but recent work has indicated that this approach may have poor reproducibility (6). Nerve conduction studies have long been used in studies of glucose control in type 1 and type 2 diabetes, and they have the advantage of being quantitative. However, improvement in nerve conduction studies may take years to manifest, and these studies do not assess small fiber nerve function, which are often the first nerves to be injured and perhaps the first to be repaired. Intraepidermal nerve fiber density (IENFD), measured by counting nerves that cross the dermal-epidermal junction, is a newer technique that is a measure of small fiber nerves and may be more sensitive to change (7) (Fig. 1). However, this technique is invasive, requiring a skin biopsy at one or two places each time it is measured. Quantitative sudomotor axon reflex testing is another measure of small fiber nerves, but it has the advantage of being noninvasive. However, the reproducibility of this study has been called into question (8). Quantitative sensory testing is another noninvasive, quantitative neuropathy outcome measure that has the potential to measure both large and small fiber function. This approach has been shown to be reproducible; however, it requires an alert, attentive, and motivated subject and there is no way to distinguish those with real from those with feigned sensory loss (9). Given the advantages and disadvantages of commonly used outcome measures, a need for a noninvasive, sensitive measure of neuropathy exists. Several groups have reported the use of corneal confocal microscopy (CCM) evaluation of corneal nerve structures as a reliable measure of diabetic neuropathy (10–12) (Fig. 1). CCM has been shown to be effective as a rapid, noninvasive, repeatable evaluation that allows detection of neuropathy in patients with diabetes (10). Tavakoli et al. (13) have previously shown CCM evidence of early regeneration of corneal nerves in 20 patients with type 1 diabetes following simultaneous pancreas-kidney transplantation (SPK). In this issue of Diabetes, Tavakoli et al. (14) present longer-term data on 15 patients with significant neuropathy who underwent successful SPK. In this well-designed and carefully executed study, patients were evaluated at baseline and at 6 and 12 months after surgery with a detailed assessment of their neurologic status, including nerve conduction studies, quantitative sensory testing, and skin biopsies as well as CCM and corneal sensitivity testing (14). Evaluation of the recovery of diabetic neuropathy following SPK has been difficult to assess, but in this study the authors show that corneal nerve fiber density, corneal nerve branch density, and corneal nerve length as measured by CCM all improve significantly by 1 year after SPK. A major strength of this study is the simultaneous evaluation of peripheral nerve function using traditional examination methods, which showed no improvement at either the 6-month or 12-month postoperative visit. The authors conclude that CCM provides a more sensitive measure of assessing nerve repair than other currently used methods of evaluating neuropathy. Despite the small number of subjects, the lack of a randomized control group or blinding, and the relatively large number of comparisons performed, this study provides compelling evidence of the utility of CCM as a measure of early nerve regeneration. It should be noted that the current study did not evaluate IENFD in thigh skin biopsy specimens (7), upper extremity nerve conduction studies (15), or sudomotor testing (16)— measures that have previously been shown to improve in studies of the recovery of neuropathy. Further, since no other peripheral nerve measures studied showed improvement over the time course of this study, it is unclear how the early nerve regeneration seen in the cornea relates to functional improvements of peripheral neuropathy. Overall, the results of this study provide further evidence for the role of CCM as a biomarker to evaluate potential therapies in future clinical trials of human diabetic neuropathy. Importantly, CCM is a noninvasive technique that is fast, reproducible, quantitative, and measures small fiber nerves. However, many important questions need to be addressed. First, future studies need to determine if improvements in CCM parameters predict improvements in traditional neuropathy outcome measures such as clinical assessments and nerve conduction studies, as well as patient-oriented outcomes such as pain, disability, and quality of life. The possibility remains that corneal nerves and sensory/motor nerves in the feet are unrelated. From the Department of Ophthalmology, University of Michigan, Ann Arbor, Michigan; and the Department of Neurology, University of Michigan, Ann Arbor, Michigan. Corresponding author: Brian C. Callaghan, [email protected]. DOI: 10.2337/db12-1114 2013 by the American Diabetes Association. Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/licenses/by -nc-nd/3.0/ for details. See accompanying original article, p. 254.