Prevalence of Neuropathy in Prediabetes Estimated by Quantitative Assessment of Sudomotor Function
Co-authors
R. Gabriel1,2. JH. Calvet3, A. Vilier3, N. Boukichou2, J. Tuomilehto 2,4 on behalf of ePREDICE Study Group.
1Instituto de Salud Carlos III, Escuela Nacional de Sanidad, Madrid, Spain.
2. Evidem Consultores. ePREDICE Coordinating Centre, Madrid. Spain
3. IMPETO MEDICAL. Paris, France
4. Dasman Diabetes Institute. Kuwait City, Kuwait

Background
Sudomotor function is used to evaluate the peripheral autonomic system. Several studies have shown sudomotor dysfunction in diabetes and prediabetes. Measurement of Electrochemical Skin Conductance (ESC) based on the electrochemical reaction between the chloride ions in sweat and stainless steel plate electrodes is a simple, non-invasive, and quick method allowing assessment of sweat gland function. Previous studies have reported that ESC measurement correlates well with other validated sudomotor standards. However, sweat function is influenced by factors as age, sex, race, BMI, and glycaemic status. Therefore a need remains to describe the performance of this technology in the prediabetic stage.
Aim
1) To determine ESC values and distribution measured by SUDOSCAN in adults with prediabetes, 
2) To estimate the prevalence of neuropathy and cardiorrespiratory autonomic neuropathy (CAN) among adults with prediabetes 
3) To explore factors potentially influencing ESC-derived indicators of neuropathy, including age, sex, BMI, and glycemic parameters.
Method
Study population. 447 prediabetic individuals (59% females), 45-74 years old, with IFG or IGT or combined  IFG+IGT, recruited in the ongoing multicentre European Randomized Clinical Trial ePREDICE (Early Prevention of Diabetes Complications in People with Hyperglycaemia in Europe).
Measurements: SUDOSCAN test, HbA1c, fasting and 2h plasma glucose, BMI and waist circumference were measured at baseline. Neuropathy was defined according to SUDOSCAN results: a) Normal: hands ESC ≥ 60 µS and feet ESC ≥ 70 µS; b) Neuropathy: hands ESC between 40 – 60 µS or feet ESC between 50-70 µS; and c) Severe neuropathy: hand ESC < 40 µS or feet ESC < 50 µS. CAN was defined according to CAN risk score : <30 no risk of CAN, ≥ 30 at risk of CAN. The percentage of Neuropathy and CAN in prediabetes by HbA1c, fasting and 2h- glucose levels, in neuropathy and CAN groups were estimated.
Results
Table 1. Percentage of neuropathy defined by SUDOSCAN in the population (No neuropathy n=310 (69%), Neuropathy n=117 (26%), and severe neuropathy n=20 (5%)).
Hands ESC (µS), n (%)	
<40	                                           18 (4.0)
40-60	                                 92 (20.6)
≥60	                                       337 (75.4)
Feet ESC (µS), n (%)	
<50	                                              4 (0.9)
50-70	                                 50 (11.2)
≥70	                                       393 (87.9)
CAN risk score, n (%)	
<30	                                       200 (44.7)
≥30	                                       203 (55.3)

Table 2. Mean values of hands and feet ESC in groups of neuropathy.
		                   No neuropathy  	 Neuropathy	 Severe neuropathy 	 p value
Hands ESC (µS)		75.7 ± 7.7	         56.0 ± 9.5           37.1 ± 10.9     	<0.0001
Feet ESC (µS)	        83.0 ± 5.1	         73.3 ± 10.3         65.6 ± 16.6	        <0.0001

Percentage of neuropathy in groups of 2h OGTT (<7.8 mmol/l: no neuropathy 69%, neuropathy 26%, severe neuropathy 5%; 7.8-11 mmol/l : no neuropathy 72%, neuropathy 24%, severe neuropathy 4%; ≥11 mol/l: no neuropathy 35%, neuropathy 59%, severe neuropathy 6%).

Table 3 Mean values of hands and feet ESC in groups of OGTT-2h.
		                                     Hands ESC (µS)    Feet ESC (µS)       CAN risk score (%)
                                 	N	        Mean ± SD	           Mean ± SD	               Mean ± SD
OGTT120 (mmol/l)				
<7.8	                      184        68.8 ± 14.3	            80.2 ± 8.8	                31.9 ± 11.4
7.8-11	                      246        69.5 ± 13.1	            79.7 ± 9.1	                34.0 ± 11.2
≥11	                                17	        59.4 ± 16.4	            74.2 ± 13.0	    37.5 ± 10.1
P-value		                                   0.0133	               0.0361	                   0.0424

Table 4 Percentage of neuropathy in normal, pre-diabetes and diabetes groups defined by HbA1c, fasting glucose and 2h-OGTT groups.
                                       	N	  No neuropathy   Neuropathy   Severe neuropathy  p value
Hba1c (%)				                                                                                	       0.0771
<5.7	                     154             103 (66.9)	       48 (31.2)              3 (1.9)	 
5.7-6.5	                     265             187 (70.6)	       61 (23.0)            17 (6.4)	 
≥6.5	                       28                20 (71.4)	         8 (28.6)                 0 (0)	 
Fasting Glucose (mmol/l)					                                                               0.3056
<5.5	                      28	           15 (53.6)	       12 (42.9)              1 (3.6)	 
5.5-7	                    410              288 (70.2)	     103 (25.1)            19 (4.6)	 
≥7	                                 9                  7 (77.8)	         2 (22.2)                 0 (0)	 
2h-OGTT (mmol/l)			                                                                       		       0.0302
<7.8	                    184              127 (69.0)	       48 (26.1)              9 (4.9)	 
7.8-11	                    246              177 (72.0)	       59 (24.0)            10 (4.1)	 
≥11	                              17                   6 (35.3)	       10 (58.8)              1 (5.9)	 

Discussion
Overall prevalence of peripheral neuropathy is very high (31%) in the prediabetic adult European population.  ESC means, the frequency of Neuropathy and of CAN risk score all increase and are significantly associated with 2h-postprandial blood glucose levels but NOT with fasting glucose or HbA1c values. 
SUDOSCAN is a simple, rapid, and highly reproducible test for use as a screening tool, and as an endpoint in clinical trials on neuropathy in prediabetes.

Prevalence of neuropathy in prediabetics estimated by quantitative assessment of sudomotor function