Diabetic neuropathy and axon reflex-mediated neurogenic vasodilatation in type 1 diabetes

Abstract

Objective: Axon reflex-mediated neurogenic vasodilatation in response to cutaneous heating may reflect early, pre-clinical small fibre dysfunction. We aimed to evaluate the distribution of the vascular flare area measured by laser doppler imaging ("LDI(FLARE) area") in type 1 diabetes and in healthy volunteers.

Research and methods: Concurrent with clinical and electrophysiological examination to classify diabetic sensorimotor polyneuropathy (DSP), LDI(FLARE) area (cm(2)) was determined in 89 type 1 diabetes subjects matched to 64 healthy volunteers. We examined the association and diagnostic performance of LDI with clinical and subclinical measures of DSP and its severity.

Results: Compared to the 64 healthy volunteers, the 56 diabetes controls without DSP had significantly lower LDI(FLARE) area (p = 0.006). The 33 diabetes cases with DSP had substantially lower LDI(FLARE) area as compared to controls without DSP (p = 0.002). There was considerable overlap in LDI(FLARE) area between all groups such that the ROC curve had an AUC of 0.72 and optimal sensitivity of 70% for the detection of clinical DSP. Use of a subclinical definition for DSP, according to subclinical sural nerve impairment, was associated with improved AUC of 0.75 and sensitivity of 79%. In multivariate analysis higher HbA1c and body mass index had independent associations with smaller LDI(FLARE) area.

Conclusions: Axon reflex-mediated neurogenic vasodilatation in response to cutaneous heating is a biomarker of early nerve dysfunction in DSP. Its independent association with glycemic exposure in diabetes subjects and both glycemic exposure and BMI in healthy volunteers highlights the existence of small-fibre dysfunction in the natural history of DSP.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Box-And-Whisker Plots Demonstrating The Distribution Of LDIFLARE Area In 64 Healthy Volunteers And 89 Type 1 Diabetes Subjects According To Neuropathy Status.

Compared to the healthy volunteer group LDIFLARE area was significantly smaller in subjects with type 1 diabetes without DSP (p = 0.006). Compared to subjects with type 1 diabetes controls without DSP, LDIFLARE area was smaller in cases with DSP (p = 0.0.0002). As indicated in the figure, among controls the LDIFLARE area was not different according to presence or absence of subclinical sural nerve impairment. Similarly, the LDIFLARE area was similar among cases with DSP regardless of severity. The criteria for mild neuropathy were two or more abnormal nerve conduction parameters in the lower limb (sural and peroneal nerve distributions), and moderate and severe neuropathy were defined by four and five abnormal parameters, respectively. NCS, nerve conduction study. LDIFLARE, laser doppler imaging flare. DSP, diabetic sensorimotor polyneuropathy.

Figure 2. Receiver Operating Characteristic (ROC) Curves For The Identification Of Diabetic Sensorimotor Polyneuropathy (Panel A) And Subclinical Sural Nerve Impairment (Panel B) By LDIFLARE Area In The 89 Subjects With Type 1 Diabetes.

The ROC curve in Panel A employed as the outcome the “England criteria” for DSP. Area under the curve was 0.72 and the threshold on the curve with optimal operating characteristics (≤1.90 cm2 for ruling in cases, indicated by the asterisk) had a sensitivity of 66% and a specificity of 70%. The ROC curve in Panel B employed as the outcome the “sural nerve criteria” to define an earlier stage of nerve impairment that did the “England Criteria” definition for DSP. The area under the curve was 0.75and the single point on the curve with optimal operating characteristics (also ≤1.90 cm2 for ruling in cases, indicated by the asterisk) had a sensitivity of 79% and a specificity of 60%.