Association between HbA1c and peripheral neuropathy in a 10-year follow-up study of people with normal glucose tolerance, impaired glucose tolerance and Type 2 diabetes

Aims To explore the association between HbA(1c) and sural nerve function in a group of people with normal glucose tolerance, impaired glucose tolerance or Type 2 diabetes. Methods We conducted a 10-year follow-up study in 87 out of an original 119 participants. At study commencement (2004), 64 men and 55 women (mean age 61.1 years) with normal glucose tolerance (n=39), impaired glucose tolerance (n=29), or Type 2 diabetes (n=51) were enrolled. At the 2014 follow-up (men, n=46, women, n=41; mean age 71.1 years), 36, nine and 42 participants in the normal glucose tolerance, impaired glucose tolerance and Type 2 diabetes categories, respectively, were re-tested. Biometric data and blood samples were collected, with an electrophysiological examination performed on both occasions. Results At follow-up, we measured the amplitude of the sural nerve in 74 of the 87 participants. The mean amplitude had decreased from 10.9 V (2004) to 7.0 mu V (2014; P<0.001). A 1% increase in HbA(1c) was associated with a similar to 1% average decrease in the amplitude of the sural nerve, irrespective of group classification. Crude and adjusted estimates ranged from -0.84 (95% CI -1.32, -0.37) to -1.25 (95% CI -2.31, -0.18). Although the mean conduction velocity of those measured at both occasions (n=73) decreased from 47.6 m/s to 45.8 m/s (P=0.009), any association with HbA(1c) level was weak. Results were robust with regard to potential confounders and missing data. Conclusions Our data suggest an association between sural nerve amplitude and HbA(1c) at all levels of HbA(1c). Decreased amplitude was more pronounced than was diminished conduction velocity, supporting the notion that axonal degeneration is an earlier and more prominent effect of hyperglycaemia than demyelination.