Aim. This study aimed to investigate whether in vivo corneal confocal microscopy (CCM) can detect the improvement of corneal nerve parameters following glycemic control in patients with type 2 diabetes in natural history. Methods. Thirty-two patients with diabetes complicated by DPN and 12 age-matched control subjects underwent detailed clinical examination and were assessed per the Toronto Clinical Scoring Scale for DPN, nerve conduction studies, and IVCCM at baseline and after approximately one year from the first visit. Results. At follow-up, 16 diabetic patients had improved glycemic control (group A, HbA1c < 7.0%, 7.78 ± 1.62% versus 6.52 ± 0.59%, P=0.005), while the remainder continued to have elevated HbA1c levels (group B, HbA1c ≥ 7.0%, 8.55 ± 1.57% versus 8.79 ± 1.05%, P=0.527). For patients in group A, corneal nerve fiber density (CNFD) (18.55 ± 5.25 n/mm2 versus 21.78 ± 6.13 n/mm2, P=0.005) and corneal nerve fiber length (CNFL) (11.62 ± 2.89 mm/mm2 versus 13.04 ± 2.44 mm/mm2, P=0.029) increased significantly compared to baseline. For patients in group B, sural sensory nerve conduction velocity (47.93 ± 7.20 m/s versus 44.67 ± 6.43 m/s, P=0.024), CNFD (17.19 ± 5.31 n/mm2 versus 15.67 ± 4.16 n/mm2, P=0.001), corneal nerve branch density (19.33 ± 12.82 n/mm2 versus 14.23 ± 6.56 n/mm2, P=0.033), and CNFL (11.16 ± 2.57 mm/mm2 versus 9.90 ± 1.75 mm/mm2, P=0.011) decreased significantly. Conclusions. The results of this study suggest that morphological repair of corneal nerve fibers can be detected when glycemic control improves. In vivo CCM could be a sensitive method that can be applied in future longitudinal or interventional studies on DPN.
Early Detection of Nerve Fiber Loss by Corneal Confocal Microscopy and Skin Biopsy in Recently Diagnosed Type 2 Diabetes