Aims: To determine the diagnostic utility of corneal confocal microscopy (CCM) compared to intra- epidermal nerve fiber density (IENFD), and warm/cold detection thresholds (W/CDT) in patients with short duration type 2 diabetes.
Methods: In this cross-sectional study, 277 adults with short duration type 2 diabetes and 97 age-matched controls were studied. Diabetic peripheral neuropathy (DPN) was defined according to the Toronto consensus as: 1) probable neuropathy, based on positive symptoms and signs of DPN; and 2) definite neuropathy, based on symptoms and signs with abnormal IENFD and/or NCS. DPN was additionally classified into: 1) small fiber neuropathy (SFN); 2) large fiber neuropathy (LFN) and 3) mixed fiber neuropathy (MFN). The Toronto clinical neuropathy score (TCNS) was used to grade DPN severity. CCM images were quantified for corneal nerve fiber density (CNFD), length (CNFL), and branch density (CNBD).
Results: There were 63 participants ‘without DPN’ (median TCNS=1, HbA1c=6.6% and diabetes duration=5.5 years) and 214 participants ‘with DPN’ (n=88 probable / n=126 confirmed DPN; SFN=7% / LFN=9.3%; / MFN=83.2%; with median TCNS=7, HbA1c=6.7% and diabetes duration=6 years). In patients ‘without DPN’ compared to controls, there was no significant difference in CNFD, CNBD, CNFL, IENFD, and W/CDT z-scores. However, in patients ‘with DPN’ compared to those ‘without DPN’, CNFD, CNBD, CNFL, IENFD and W/CDT were significantly lower. There were significant correlations between TCNS and CNFD (r=-0.23), CNFL (-0.24), CNBD (r=-0.17), W/CDT (r=-0.45/-0.34), and IENFD (r=-0.58). Diagnostic performance analysis showed that IENFD had the highest sensitivity/specificity/AUC (0.51/0.9/0.71), followed by W/CDT (0.31/0.85/0.58), CNFL (0.14/0.96/0.55), CNBD (0.08/0.96/0.52) and CNFD (0.08/0.96/0.52). Diagnostic performance was comparable when limited to patients with confirmed DPN.
Conclusions: CCM showed limited sensitivity to diagnose DPN in short duration type 2 diabetes compared to other small fiber measures.
Comments. In DPN, SFN occurs early and its identification with timely risk factor reduction may be critical for successful disease modification. Skin biopsy with IENFD assessment is advocated as the gold standard method to identify small fiber damage. Multiple studies have shown that quantification of the corneal sub-basal plexus using CCM has diagnostic and prognostic value and may be a useful end point to assess early nerve regeneration in clinical trials. The present study is the first to challenge this view, especially in relation to the diagnostic utility of CCM for early neuropathy. However, overall small fiber testing showed poor diagnostic performance even with gold standard IENFD, but more so for thermal thresholds and CCM. Whilst none of the patients ‘without DPN’ had an abnormal IENFD or NCS, a proportion already had abnormal thermal thresholds (14.3%) and CCM (3.1-4.8%) suggesting presence of subclinical DPN. Conversely, a far greater percentage of patients ‘with DPN’ had abnormal IENFD (51.0%) and NCS (37.1%), whilst abnormal thermal thresholds were present in 30.5% and abnormal CCM in 7.5-13.1%. A diagnostic definition incorporating an independent small fiber measure is needed to optimally assess the diagnostic and prognostic performance of both IENFD and CCM.
Ioannis N. Petropoulos
Reference. Gylfadottir SS, Itani M, Kristensen AG, Nyengaard JR, Sindrup SH, Jensen TS, Finnerup NB, Karlsson P. Assessing Corneal Confocal Microscopy and Other Small Fiber Measures in Diabetic Polyneuropathy. Neurology. 2023 Apr 18;100(16):e1680-e1690. doi: 10.1212/WNL.0000000000206902. Epub 2023 Feb 7. PMID: 36750383; PMCID: PMC10115507.
