Improvement of Electrophysiological Neuropathy After Islet Transplantation for Type 1 Diabetes: A 5-Year Prospective Study

The consequences of islet transplantation (IT) on neuropathy remain unclear, especially since neurotoxic effects of tacrolimus have been reported (1). The restoration of a euglycemic state in the Diabetes Control and Complications Trial and Epidemiology of Diabetes Interventions and Complications study and through alone or combined whole pancreas transplantation has been shown to halt the progression of diabetic neuropathy (2). A significant improvement of a nerve conduction velocity (NCV) score compared with baseline was observed after IT-after-kidney (IAK), without any difference from a control group of nine kidney-only transplanted patients (3). No significant improvement of NCV was noticed 3 years after IT-alone (ITA) in a crossover study compared with intensive medical therapy (4). There has been no study of cardiac autonomic neuropathy (CAN) after IT and the results after kidney-pancreas transplantation are discordant. The aim of our study is to describe the evolution of neuropathy after IT. This single-arm, 5-year prospective, longitudinal study included 21 type 1 diabetic patients who underwent IT (13 ITA, 8 IAK) in a single center. Immunosuppression was induced by interleukin-2 receptor antibodies and maintained with tacrolimus and sirolimus as described (5). The main study end points were continuous 24-h mean blood pressure and 72-h continuous glucose monitoring (CGM), as well as electrophysiological and cardiovascular autonomic testing, pre-IT and yearly until 5 years post-IT. Ten of the 21 patients (48%) were insulin-independent 5 years post-IT. Four patients lost their islet graft during follow-up, but were analyzed on an intention-to-treat basis.

The metabolic and sensory parameters improved between 0 and 5 years: (median [interquartile range], respectively 0 vs. 5; A_1c: 8.3 [7.3–8.9] vs. 6.9 [6.0–7.9]% or 67 [56–74] vs. 52 [42–63] mmol/mol, P < 0.01; fasting blood glucose: 237 [148–289] vs. 108 [98–139] mg/dL, P < 0.001; fasting/postprandial C-peptide: 0 [0–0] vs. 0.9 [0.5–2.0]/2.7 [0.6–3] ng/mL, P < 0.0001; CGM mean glucose and SD: 147 [129–198] vs. 126 [109–159] mg/dL, P = 0.02 and 58 [43–76] vs. 33 [15–55] mg/dL, P < 0.01; sensory NCV: 47.5 [40.4–53.3] vs. 41.2 [33.5–43.9] m/s, P < 0.01; and sensory action potentials: 5.5 [3.7–7.3] vs. 11.8 [5.8–14.5] mV, P < 0.01) (Fig. 1B, C, E, F). Moreover, the sensory parameters significantly improved over the 5-year posttransplantation period (Fig. 1A−D). The kidney (creatinine, Modification of Diet in Renal Disease, microalbuminuria) and lipid parameters, mean 24-h blood pressure, and motor electrophysiological and CAN parameters did not vary significantly between pre- and 5 years post-IT. Enzyme conversion inhibitors and statins were added in one-third of patients between 0 and 5 years. Both sensory (r = −0.28, P < 0.01) and motor (r = −0.27, P < 0.01) action potentials and sensory and motor NCV (r = −0.24, P = 0.01 and r = −0.34, P < 0.001, respectively) were negatively correlated with CGM mean glucose. Only sensory NCV were negatively correlated with CGM SD (r = −0.19, P < 0.01). Motor action potentials and NCV were negatively correlated with tacrolimus (r = −0.23, P = 0.02 and r = −0.25, P = 0.01, respectively) and triglycerides (r = −0.35, P < 0.001 and r = −0.21, P = 0.02, respectively). Motor action potentials were negatively correlated with mean systolic blood pressure (r = −0.41, P < 0.0001).

Half of the patients were insulin-independent at the 5-year assessment, a figure close to the results of pancreas transplantation. Sensory neuropathy improved 5 years after IT. The favorable effect of IT on peripheral neuropathy in association with CGM mean glucose was burdened with the deleterious consequences of tacrolimus, mainly on the motor component. Low-dose tacrolimus and tight glucose, triglycerides, and blood pressure control are recommended.