Sudomotor dysfunction in people with type 1 diabetes with and without clinical peripheral neuropathy: A study from Pune, India Authors: Rohan Shah, Rishikesh Behere, Aboli Bhalerao, Sanat Phatak, Alma Baptist, Chinmay S Marathe, Chittaranjan S Yajnik Published: 2026 ABSTRACT ABSTRACTBackground and AimsIt is appreciated that current screening tools for diabetic peripheral neuropathy (DPN) often fail to detect early disease. Sudomotor (sweat gland) dysfunction, among the earliest manifestations of DPN, can be measured objectively by the SUDOSCAN® device. We compared SUDOSCAN® with conventional validated screening (Michigan Neuropathy Screening Instrument or MNSI) in type 1 diabetes (T1D).MethodsT1D participants (n = 264, 46.2% male, mean age 26.3 years, mean diabetes duration 13.2 years) enrolled in the prospective PICTURE study (Diabetes Unit, KEM Hospital, Pune, India) underwent simultaneous assessments for sudomotor dysfunction (electrochemical skin conductance ESC <60 μS indicating dysfunction) and DPN (MNSI) and were classified as: (a) MNSI abnormal (‘clinical DPN’), (b) MNSI normal, but ESC <60 (‘subclinical DPN) and (c) Both normal (‘no neuropathy’).ResultsSudomotor dysfunction was more common than clinical DPN (46.2 vs 22.3%) with a sensitivity of 72.8% and specificity of 61.4% for detecting clinical DPN. Those with sudomotor dysfunction (ESC <60 μS) were 4.3 times more likely to have clinical DPN compared with participants without sudomotor dysfunction (ESC ≥60 μS). Among participants with normal MNSI (i.e., without clinical DPN), subclinical DPN was detected in 38.5% and was associated with younger age, shorter duration of diabetes, anemia, vitamin D deficiency, and microvascular complications.ConclusionSudomotor testing is an objective and sensitive method for DPN screening in T1D, aiding early detection.Keywords: Sudomotor dysfunction, SUDOSCAN, Type 1 diabetes Abbreviations area under the curve Body Mass Index Coronary Artery Disease cardiovascular reflex tests Complete Blood Count chemiluminescent microparticle immunoassay diabetic peripheral neuropathy distal symmetrical polyneuropathy electrocardiographic electrochemical skin conductance Early Treatment Diabetic Retinopathy Study Fibrocalculous Pancreatic Diabetes glycated hemoglobin high‐density lipoprotein holotranscobalamin High‐Performance Liquid Chromatography intraepidermal nerve fiber density King Edward Memorial low‐density lipoprotein Michigan Neuropathy Screening Instrument nerve conduction studies Neuropathy Impairment Score‐Lower Limbs Prevalence and Incidence of Complications of Type one diabetes in pUne at a diabetes centRE quantitative autonomic function testing quantitative sensory testing Receiver Operating Characteristics Statistical Package for the Social Sciences Semmes–Weinstein monofilament type 1 diabetes type 2 diabetes urine albumin–creatinine ratio microsiemens INTRODUCTION Diabetic peripheral neuropathy (DPN) refers to a ‘heterogeneous group of neurological disorders clinically evident or subclinical that occur in the setting of diabetes mellitus and when other etiologies of peripheral neuropathy have been excluded’ 1 . The most common presentation of DPN is distal symmetrical polyneuropathy (DSPN), and prevalence rates as high as 50% have been reported, especially in type 2 diabetes (T2D) with long duration of disease 2 . It is widely acknowledged that DPN is associated with increased morbidity, mortality, healthcare costs, and a reduced quality of life 3 . Up to 50% of those affected may be asymptomatic, and the diagnosis, not surprisingly, is often made later in the clinical course, when irreversible damage has occurred 4 . While DPN is believed to be rare in early type 1 diabetes and prevalence said to increase with a longer duration of diabetes and poor glycemic control 5 , 6 , 7 , 8 , the diagnosis has conventionally relied on routine bedside clinical tools, such as sensitivity to a 10 g Semmes–Weinstein monofilament (SWMF), vibration sensation using tuning forks, or ankle reflexes, often combined with assessment of neurological symptoms 9 which are useful in identifying the ‘foot at‐risk for ulceration or amputation’. These tests often fail to detect early stages of neuropathy 10 , especially small‐fiber neuropathy, and thus the true prevalence of DPN in type 1 diabetes may have been underestimated. Sudomotor nerve dysfunction is one of the earliest detectable small nerve fiber abnormalities 4 . It can now be easily evaluated using devices such as SUDOSCAN®, a noninvasive quantitative and objective method for the early detection of sudomotor dysfunction 11 , 12 , which assesses autonomic control of sweat gland activity. We and others have reported a good correlation of SUDOSCAN® with both DSPN as well as autonomic neuropathy, particularly cardiovascular autonomic neuropathy 3 , 13 , 14 , 15 . In people with type 2 diabetes, we found that SUDOSCAN® measurements are in good agreement with conventional testing (questionnaire‐based methods), with a sensitivity and specificity of 73% and 62%, respectively 16 . However, only a few studies have reported the use of SUDOSCAN® in type 1 diabetes 3 , 4 , 17 , 18 and these studies typically have a small sample size (between 50 and 100). The major focus of these initial studies was the ‘agreement’ of SUDOSCAN® with conventional neuropathy testing. Subclinical DPN is defined as the ‘electrophysiological abnormality of nerve function in the absence of clinical symptoms or signs of peripheral nerve disease’ 17 , 19 . The prevalence of subclinical DPN in adult type 1 diabetes has not been well studied but is thought to be common and under‐recognized. We investigated the utility of SUDOSCAN® to diagnose peripheral neuropathy in an Indian cohort with type 1 diabetes as part of the PICTURE trial, a prospective study of diabetes‐related complications in people with type 1 diabetes being conducted at the Diabetes Unit, KEM Hospital, Pune and compared it with a standard clinical assessment (The Michigan Neuropathy Screening Instrument or MNSI), thus allowing to define its role in identifying subclinical DPN. We also evaluated its associations with risk factors for DPN. METHODS Study participants The Diabetes Unit at King Edward Memorial (KEM) Hospital, Pune, India, provides specialized care for the treatment of diabetes and over 1,000 patients with type 1 diabetes are registered in the clinic. The diagnosis of type 1 diabetes was based on pragmatic clinical criteria, supported by circulating C‐peptide concentrations and autoimmune markers, such as GAD antibodies. Participants from this registry were followed up as part of the ‘Prevalence and Incidence of Complications of Type one diabetes in pUne at a diabetes centRE’ (PICTURE) study. The primary objective of the PICTURE study was to determine the prevalence of long‐term complications among type 1 diabetes patients. Type 1 diabetes patients with a duration of diabetes of more than 1 year were included in this study. Those diagnosed with Fibrocalculous Pancreatic Diabetes (FCPD), neonatal diabetes, syndromic diabetes, and those suspected to have monogenic diabetes were excluded. Pregnant or lactating women with type 1 diabetes were also excluded. Study procedure The study was conducted between June 2016 and June 2019. type 1 diabetes patients attending the outpatient clinic were invited to participate in the study if they fulfilled the inclusion and exclusion criteria. Clinical information, including age at diagnosis and duration of diabetes, was obtained from medical records and a clinical interview. Measurements included blood pressure and anthropometric data (height and weight). Retinal images were obtained using a fundus camera. Nonfasting venous blood and urine samples were collected for hematologic and biochemical measurements, including lipid profile, creatinine, and albuminuria as well as ferritin and vitamins B12 and D. This is a post hoc exploratory analysis employing data from the PICTURE study. The original protocol aimed to assess the prevalence of diabetes complications in type 1 diabetes, and while the SUDOSCAN® and MNSI data were collected prospectively, the specific hypothesis regarding its utility in identifying subclinical DPN was developed post‐data acquisition. Physical measurements Blood pressure was measured in the supine position on the right arm, using a digital blood pressure monitor (UA 767PC, A&D Instruments Ltd, Oxford, UK) with an appropriately sized cuff. Two readings were taken 5 min apart, and the average of these readings was used for analysis. Weight was measured to the nearest 0.01 kg using an electronic weighing balance (CONWEIGH Electronic Inst P. LTD). Height was measured using a wall‐mounted stadiometer to the nearest 0.1 cm (CMS Instruments Ltd., London, UK). Body Mass Index (BMI) kg/m2 was calculated as body weight in kilograms divided by the square of standing height in meters. Regular calibration of all instruments was done. Assessment of neuropathy Michigan neuropathy screening instrument (MNSI) The MNSI is a standard method for the assessment of neuropathy and includes two parts. Part A consists of questions that assess neuropathic symptoms, and Part B involves a clinical examination (Appendix S1). The clinical assessment includes foot inspection for deformities, skin changes, and infection, ankle reflex testing, vibration sensation using a 128 Hz tuning fork, and touch perception using 10 g monofilament. A composite score is calculated separately for symptoms (i.e., MNSI‐A) and clinical examination (i.e., MNSI‐B). MNSI‐A (score >7.5) and MNSI‐B (score >2) suggest neuropathy 20 . Sudomotor nerve function assessment Sudomotor nerve function was measured using SUDOSCAN® (Impeto Medical, Paris). The device consists of two sets of electrodes for hands and feet connected to a computer to record the output. The subject places the palms of the hands and the soles of the feet on the electrodes 11 , 12 . A low voltage current (<4 V) applied through the stainless steel electrodes leads to the attraction of chloride ions present in the heavily concentrated sweat glands of the palms and soles. The machine measures electrochemical sweat conductance (ESC), defined as the ratio of the current measured to the constant power applied, expressed in microsiemens (μS), and measured separately for the hands and feet (right and left sides). The ESC values of the left and right sides were averaged to obtain mean ESC scores for the hands and the feet. Unlike nerve conduction studies (NCS), electrochemical skin conductance (ESC) measurement using SUDOSCAN® does not require active nerve stimulation or measurement of conduction velocity and is therefore less sensitive to minor variations in skin temperature 21 . Nevertheless, to minimize the influence of environmental and physiological factors, all measurements were performed under standardized conditions. Testing was conducted in a temperature‐controlled room, with participants seated comfortably and allowed to rest before measurement. All participants were instructed to wash their hands and feet with soap and water to remove dirt or surface contaminants, and to thoroughly dry their skin with a cloth and air‐dry before electrode contact. Measurements were performed in accordance with the manufacturer's recommended protocol. These procedures were applied uniformly to all participants. Sudomotor dysfunction was defined as ESC <60 μS and is reported as abnormal ESC at any (hand or feet) 4 , 14 , 22 . The ESC threshold of <60 μS used to define abnormal sudomotor function was chosen a priori based on prior validation and normative studies. Previous work has commonly used ESC values around 60 μS as a conservative cutoff to flag possible sudomotor impairment, and normative investigations have shown that mean hand and foot ESC values and appropriate cutoffs vary by population and ethnicity. We therefore selected ESC <60 μS as our primary threshold for comparability with published diagnostic and screening studies 4 , 13 , 14 , 22 . Based on the MNSI and sudomotor nerve function assessment, participants were divided into three groups: (a) MNSI abnormal (i.e., ‘clinical DPN’) (b) MNSI normal, but ESC <60 (i.e., ‘subclinical DPN’) and (c) MNSI normal with ESC >60 (i.e., ‘no neuropathy’). The rationale for this classification was based on the complementary nature of the two assessment tools. The MNSI is a validated clinical instrument that primarily reflects large‐fiber dysfunction and is widely used to define clinically overt DPN 20 . In contrast, sudomotor dysfunction assessed by SUDOSCAN® reflects small fiber involvement, which is known to occur early in the course of diabetic neuropathy 11 , 12 . Accordingly, participants with abnormal MNSI were classified as having clinical DPN irrespective of ESC values, while those with normal MNSI but abnormal ESC (<60 μS) were classified as having subclinical DPN, indicating early small fiber dysfunction in the absence of clinically detectable neuropathy. Evaluation of vascular complications Retinopathy Retinal images were obtained after pupillary dilation using a 3nethra Classic fundus camera (Forus Health Pvt Ltd, India). To categorize retinopathy, a senior ophthalmologist graded retinal photographs for each eye using the Early Treatment Diabetic Retinopathy Study (ETDRS) grading scale 23 . Nephropathy Participants with eGFR <90 mL/min/1.73 m2 and/or urine albumin–creatinine ratio (UACR) ≥30 mg/g were categorized to have nephropathy 24 . Coronary artery disease (CAD) CAD was diagnosed based on documented myocardial infarction or a history of angioplasty or bypass surgery, coronary angiography findings, or suggestive electrocardiographic (ECG) changes (Minnesota code 2009) 25 . Laboratory measurements A nonfasting venous blood sample and urine were collected. The following were measured: Complete Blood Count (CBC) using a Beckman Coulter AC.T diff TM Analyzer (Miami, Florida, USA). HbA1c was measured on a BIO‐RAD D‐10 (BIO‐RAD Laboratories India Private Limited) using the high‐performance liquid chromatography (HPLC) method. UACR was measured by the immunoturbidimetric method. Total and high‐density lipoprotein (HDL) cholesterol, triglycerides, and creatinine were measured using an automated biochemistry analyzer (Hitachi 902, Roche Diagnostics, Germany) using standard reagents. Low‐density lipoprotein (LDL) cholesterol was calculated using the Friedewald‐Fredrickson formula. Serum 25‐hydroxy vitamin D3 was measured using the IDS Enzyme‐Linked Immunosorbent Assay (ELISA) kit. Plasma vitamin B12 and holotranscobalamin (Holo‐TC) (‘Active’ B12) were measured by chemiluminescent microparticle immunoassay (CMIA) using the Architect system (Abbott GmbH & Co. KG, Germany). Ferritin was measured using the DiaMetra, Italy, ELISA kit. Statistical analysis Data on sociodemographic, clinical, biochemical parameters, and SUDOSCAN® measurements are represented as median (25th–75th centile) for continuous variables and number (percentage) for categorical variables. Variables not normally distributed were natural log‐transformed for further analysis. We examined the association of clinical, biochemical, and nutritional parameters with neuropathy as classified into ‘no neuropathy’, ‘subclinical DPN’, and ‘clinical DPN’, as defined above. Significance was checked using Trend and ancova (adjusted for covariates: age and sex), with a Bonferroni correction for multiple comparisons. The area under the curve (AUC) of the receiver operating characteristics (ROC) curve was calculated to measure the efficiency of SUDOSCAN® in the diagnosis of neuropathy against MNSI. Logistic regression analysis was performed to estimate ODDs of ESC abnormal and for MNSI normal and ESC < 60. The independent associations of these ‘risk factors’ were further assessed using multivariable logistic regression. To assess whether the association between sudomotor dysfunction and clinical DPN was independent of known risk factors, multivariable logistic regression models were constructed with clinical DPN (MNSI abnormal) as the dependent variable. Sudomotor dysfunction (ESC <60 μS) was entered as the primary independent variable, with adjustment for sex, duration of diabetes, BMI, HbA1c, anemia, and vitamin B12 status. Given the wide age range of participants, age was included as a covariate in all comparative and multivariable analyses. Age‐stratified analyses were considered; however, these were not performed due to the limited number of pediatric participants with neuropathy, particularly those with clinically overt DPN. To examine the potential influence of body size on sudomotor function, correlations between ESC values and body weight and BMI were assessed using Spearman's correlation. In addition, BMI was included as a covariate in multivariable models assessing the association between sudomotor dysfunction and clinical DPN. Data were analyzed using the Statistical Package for the Social Sciences (SPSS) [IBM Corp Version 26.0. Armonk, NY: IBM Corp]. P‐value <0.05 is considered statistically significant. Ethical considerations The study was approved by the Institutional Ethics Committee of the King Edward Memorial Hospital Research Centre (KEMHRC), Pune (Reference: KEMHRC ID: Ph.D. 19) and was prospectively registered on clinicaltrials.gov (Study Identification Number: NCT03069196), with primary outcomes focused on micro‐ and macrovascular, pulmonary complications, and growth in people with type 1 diabetes. The use of SUDOSCAN® to assess sudomotor function was not a predefined outcome in the registry and the hypothesis regarding its diagnostic utility was developed post hoc. All participants provided signed informed consent; those under 18 years old provided assent, and their parents or guardians signed consent. RESULTS Participant characteristics Out of a total of 327 type 1 diabetes patients enrolled in the PICTURE study, ESC values were available for 309 participants, and MNSI values were available for 264 participants. The 18 participants who did not have ESC values and the 63 who did not have MNSI values were not significantly different from the remainder of the cohort in terms of age, duration of diabetes, hemoglobin, and HbA1c levels. We report results for 264 participants for whom both MNSI and SUDOSCAN® measurements were available. The characteristics of participants are described in Table 1. The median age of participants was 26.3 years (min 8.8, max 65.1), 122 (46.2%) were male, the median duration of diabetes was 13.2 years (min 1.1, max 53.9), and HbA1c was 8.9% (min 5.2, max 16.3) OR 74 mmol/mol (min 33, max 151). Thirty‐nine percent of the participants were anemic (age and sex‐specific World Health Organization (WHO) criteria) 26 , and 17.8% had low vitamin B12 concentrations (total B12 < 150 pmol/L). Retinopathy was present in 11.9% (ETDRS grading), and albuminuria in 23% (defined as UACR >30 mg/g OR 3.5). Table 1Characteristics of participants by neuropathy statusClinical DPNSubclinical DPN, MNSI normal, Any ESC < 60 (n = 79) – BNo Neuropathy MNSI normal, Both ESC > 60 (n = 126) – C P‐value (anova) P‐value (anova) P‐value (anova) P‐value (anova)MNSI Abnormal (n = 59) – AA vs. B vs. CA vs. BA vs. CB vs. CDemographic characteristicsAge (years)36.3 (31.1, 43.7)26.0 (18.6, 35.5)22.0 (15.6, 29.2)<0.001<0.001<0.0010.067Male35 (59.3%)29 (36.7%)58 (46.0%)0.0310.0080.0920.189Height (cm)160.7 (150.5, 168.5)157 (151.4, 164.5)157.9 (152.7, 165.7)<0.001<0.001<0.0010.117Weight (kg)59.5 (52.0, 66.4)52.9 (44.2, 62.5)52.7 (43.9, 63.7)<0.0010.046<0.0010.055BMI22.9 (19.9, 24.9)21.1 (18.9, 24.2)20.7 (17.7, 24.6)0.1040.5810.0430.179Diabetes‐related variablesHbA1c (%)9.3 (8.0, 10.6)8.9 (7.7, 10)8.6 (7.6, 9.8)<0.0010.0100.0030.079Duration of diabetes (years)23.3 (15.4, 30)12.9 (7.6, 21.7)10.3 (6.6, 16.1)0.007<0.001<0.0010.280Hematologic and biochemical variablesHemoglobin (g/dL)12.1 (10.8, 13.4)11.9 (10.7, 13)12.6 (11.4, 14.1)<0.0010.361<0.0010.004Anemia (WHO 2011)25 (42.4%)40 (51.9%)37 (30.8%)0.0120.2680.1270.003Ferritin (ng/mL)24.0 (9.0, 46.0)29 (9.5, 56.1)32 (13.5, 59.5)0.0350.1170.0060.293Vitamin B12 (pmol/L)243.0 (162.0, 428.0)270 (193.5, 437.5)209 (157, 350.5)0.0920.1000.7980.033B12‐Deficient (Total B12 < 150 pmol/L)22.0 (37.3%)21.0 (26.6%)58.0 (46.0%)0.0200.1790.2630.005Holo‐TC (pmol/L)44.5 (24.9, 69.3)44.9 (29.2, 100.6)37.4 (24.2, 64.9)0.0670.0150.8140.05625‐Hydroxy Vitamin D‐IDS (nmol/L)41.1 (29.4, 63.4)39.6 (25.5, 60.5)32.8 (23.8, 51.6)0.1650.1060.7090.052Cardiometabolic variablesCholesterol (mg/dL)160.0 (140.5, 184)162.5 (136.5, 177.5)156.0 (135.5, 182)0.7170.3140.4080.520Triglyceride (mg/dL)76.0 (63.0, 117.5)72.5 (52.8, 105.2)84.0 (62.0, 103.8)0.3160.1100.2910.270HDL (mg/dL)44.0 (39.3, 50.8)47.0 (40.0, 56.8)45.0 (38.0, 52.0)0.4250.1970.7820.340Creatinine (mg/dL)0.9 (0.7, 1.0)0.7 (0.6, 0.8)0.7 (0.6, 0.9)0.3850.9480.6800.179Systolic BP (mmHg)116 (107, 130)106 (101, 114)111 (103, 118)0.0510.1090.7790.020Diastolic BP (mmHg)70 (65, 79)68 (63, 73)67 (62, 74)0.5820.1320.4770.821Vascular complications and ESC valuesAt least one Microvascular complication29 (49.2%)22 (27.8%)22 (17.6%)<0.0010.010<0.0010.083Feet ESC (μS)53.5 (41.0, 70.0)53.5 (41.0, 62.5)79 (71.5, 84.0)<0.0010.981<0.001<0.001Hand ESC (μS)58.0 (38.0, 71.5)52.5 (46.0, 64.0)75.3 (69.0, 82.5)<0.0010.783<0.001<0.001Data represented as median (25th, 75th percentile) or n (%), P‐value <0.05 considered significant and was calculated by anova adjusted for age and gender or Chi‐square test. Clinical DPN was defined based on abnormal MNSI irrespective of ESC values; subclinical DPN was defined as normal MNSI with ESC < 60 μS. ESC cutoffs: ESC < 60 μS used to define abnormal sudomotor function. Cutoff chosen based on prior validation and commonly used thresholds in the literature; population differences (including ethnic variation) have been reported and may influence optimal cutoffs 35 , 36 . Characteristics of participants by neuropathy status Data represented as median (25th, 75th percentile) or n (%), P‐value <0.05 considered significant and was calculated by anova adjusted for age and gender or Chi‐square test. Clinical DPN was defined based on abnormal MNSI irrespective of ESC values; subclinical DPN was defined as normal MNSI with ESC < 60 μS. ESC cutoffs: ESC < 60 μS used to define abnormal sudomotor function. Cutoff chosen based on prior validation and commonly used thresholds in the literature; population differences (including ethnic variation) have been reported and may influence optimal cutoffs 35 , 36 . Prevalence of clinical and subclinical DPN Fifty‐nine participants (22.3%) had clinical DPN as per MNSI criteria. Among participants with clinical DPN (MNSI abnormal), 43 (72.9%) had abnormal sudomotor function (ESC <60 μS), while the remaining 16 (27.1%) had ESC values ≥60 μS. This discordance highlights that MNSI and sudomotor testing capture partially overlapping but distinct aspects of neuropathy. Of the 205 participants with a normal MNSI, 79 (38.5%) were found to have ESC <60 μS (either hand or feet) (i.e., subclinical DPN), while 126 participants had ESC ≥60 μS, (no neuropathy). Within the clinical DPN group (MNSI abnormal, n = 59), the vast majority of participants (n = 57) had abnormalities detected on the clinical examination component (MNSI‐B), while only two participants had both symptom‐based (MNSI‐A) and examination‐based (MNSI‐B) abnormalities. No participants had isolated MNSI‐A abnormalities. Given this distribution, meaningful subgroup comparisons between MNSI‐A and MNSI‐B categories were not feasible. Overall, sudomotor dysfunction in the whole study cohort (n = 264) (ESC <60 μS in either hand or foot) was evident in 122 participants (46.2%). Patients with type 1 diabetes who had sudomotor dysfunction (ESC <60 μS) had a higher HbA1c concentration, lower hemoglobin levels, and a higher prevalence of other microvascular complications compared with those without (P < 0.05, all). Vitamin B12 and Holo‐TC concentrations were marginally higher. The association of clinical and subclinical DPN with biochemical characteristics Participants with ‘clinical DPN’ were the eldest and with the longest duration of diabetes, highest HbA1c, and highest prevalence of microvascular complications (see Table 1). They also had a higher prevalence of anemia and lower ferritin concentrations but higher vitamin B12, HOLO‐TC, and vitamin D concentrations than those without neuropathy. Those with subclinical DPN were also older, had a longer duration of diabetes, higher HbA1c, higher prevalence of other microvascular complications, and higher levels of vitamin B12 compared with those without neuropathy. However, compared with those with ‘clinical DPN’, those with subclinical DPN were younger, had a lower duration of diabetes, lower HbA1c, lower prevalence of other microvascular complications, and comparable Holo‐TC (see Table 1). Lipid concentrations were similar across the groups. Among the 79 participants with subclinical DPN (normal MNSI and ESC <60 μS), 50 (63.3%) were female. Anemia was observed in 14.3% of males and 73.5% of females in this subgroup, suggesting a possible sex‐specific association between anemia and low ESC. The number of participants with neuropathy in the <18 years was small. Among those with clinical DPN (n = 59), only four participants (5.8%) were aged <18 years, while the remaining 55 were adults. Similarly, among participants with subclinical DPN, 19 (27.5%) were aged <18 years and 60 were ≥18 years. Due to the small number of those in the <18 years cohort, particularly in the clinical DPN group, age‐stratified analyses were not undertaken. Electrochemical skin conductance showed no significant association with body size measures. Hand ESC demonstrated a weak, nonsignificant inverse correlation with body weight (Spearman r = −0.082, P = 0.186) and BMI (r = −0.082, P = 0.184). Similarly, feet ESC showed a weak, nonsignificant correlation with body weight (r = −0.078, P = 0.208) and BMI (r = −0.100, P = 0.106). Sensitivity and specificity of SUDOSCAN® to predict clinical DPN When neuropathy prevalence by MNSI and sudomotor dysfunction (either hands or feet) were compared, the latter had a specificity of 61.4% and a sensitivity of 72.8% for the diagnosis of neuropathy. The negative predictive value was 88.7%, and the positive predictive value was 35.2% and the AUC was 0.672. DISCUSSION In this study, we investigated sudomotor (small nerve fiber) function using the SUDOSCAN® device in people with type 1 diabetes to screen for peripheral neuropathy in an outpatient setting. To the best of our knowledge, this is the largest study of sudomotor testing employing SUDOSCAN® in people with type 1 diabetes worldwide. The key findings are: (i) In those without evidence of overt DPN (i.e., normal MNSI), sudomotor dysfunction (ESC < 60) was identified in 79/205 (38.5%) participants suggesting higher sensitivity to diagnose nerve involvement in type 1 diabetes and at a younger age and a lower duration of diabetes. (ii) Compared with those without any neuropathy, those with subclinical DPN had a longer duration of diabetes, higher HbA1c, and lower hemoglobin concentrations, providing an insight into its etiology. (iii) Our results show a graded distribution of risk factors for neuropathy across the three severity groups. (iv) Overall, the SUDOSCAN® showed acceptable sensitivity and negative predictive value for early diagnosis of neuropathy in people with type 1 diabetes. The ease of performance makes it a suitable recommendation for use as a screening test in busy outpatient clinics. Diagnosis of DPN has conventionally relied on bedside clinical tests, which primarily assess so‐called ‘large’ nerve fibers, comprising myelinated Aα and Aβ fibers (mediating touch, proprioception, position sense, vibration, and muscle control). It is now appreciated that the large nerve fibers are affected relatively late in the natural history of diabetic neuropathy 26 . The ‘small’ nerve fibers, comprising thinly myelinated Aδ or nonmyelinated C fibers, are involved early in the course of diabetic neuropathy. These constitute the majority of peripheral nerve fibers (as high as 90%) and mediate sweating, pain perception, temperature, tissue blood flow, and regulate the autonomic system 27 , 28 . Thus, the detection of small‐fiber neuropathy will facilitate early detection and timely intervention to limit or reverse the progression of neuropathy and help prevent serious foot complications 29 . While NCS are regarded as the ‘gold standard’ for diagnosing peripheral neuropathy 30 , a significant limitation is that these tests assess large, myelinated nerves and cannot detect changes in the smaller fibers. Thus, NCS are not suitable for small fiber assessment and are used instead to exclude other etiologies of large‐fiber neuropathy and atypical neurological presentations such as amyotrophy 2 . Sweat glands are innervated by small, unmyelinated sympathetic C nerve fibers responsible for the sweat response 31 , and sudomotor nerve dysfunction is recognized as one of the earliest detectable small nerve fiber abnormalities. The SUDOSCAN® assesses sweat gland function by measuring ESC based on chloride ion concentrations in sweat 3 , 32 and has been introduced as a screening tool to assess DPN. Small‐fiber neuropathy is most commonly characterized by a reduction in intraepidermal nerve fiber density (IENFD) on skin biopsy, which is considered a reference standard for the structural assessment of small nerve fibers 33 , 34 . Several studies have demonstrated that sudomotor dysfunction assessed by SUDOSCAN® correlates with established markers of small‐fiber neuropathy, including reduced IENFD as well as abnormalities on quantitative sensory and autonomic testing 4 , 14 , 31 . Because sweat glands are innervated by unmyelinated sympathetic C fibers, impaired ESC reflects functional compromise of the same fiber population that contributes to reduced IENFD 5 , 28 . While SUDOSCAN® does not directly measure nerve fiber density, it provides a noninvasive functional surrogate of small‐fiber integrity that is well suited for screening purposes in clinical and epidemiological settings 4 , 14 , 15 , 28 , 31 , 35 . Aging is an established determinant of small‐fiber integrity, with several studies demonstrating an age‐related decline in IENFD even in individuals without diabetes 33 , 34 . Age‐related alterations in small fiber and autonomic function have also been described, and it is therefore plausible that sudomotor function, as assessed by ESC, may decline with advancing age 31 , 35 . In the present study, age was included as an adjustment variable in all relevant analyses, and the association between sudomotor dysfunction and clinical DPN persisted after age adjustment, suggesting that the observed findings are not solely attributable to aging. We found that the sensitivity and specificity of SUDOSCAN® in identifying DPN were 72.8 and 61.4%, respectively. A comparison with other studies in type 1 diabetes is not straightforward due to considerable differences in study methodologies including differences in the neuropathy screening tools, characteristics of study populations, and variable ESC cutoffs. For example, Casellini et al. compared the SUDOSCAN® with DPN assessment using Neuropathy Impairment Score‐Lower Legs (NIS‐LL), quantitative autonomic function testing (QAFT), and quantitative sensory testing (QST) in 21 patients with type 1 diabetes (duration of diabetes not stated). They reported that feet ESC (values <60 indicating neuropathy) had a sensitivity of 78% and specificity of 92% in detecting DPN 4 . Selvarajah et al. compared the SUDOSCAN® with clinical examinations, neurophysiological assessments, and five standard cardiovascular reflex tests (CARTs) as well as NCS using the American Academy of Neurology criteria in 45 patients with type 1 diabetes (duration of diabetes 29.7 years for those with DPN). They reported a sensitivity of 65% and specificity of 80% to the cardiac autonomic neuropathy (CAN) score 4 . In general, our observations may be considered comparable to sensitivity values for diagnosis of DPN by SUDOSCAN® reported in previous studies in both type 1 diabetes and type 2 diabetes (range from 78.0 to 87.5) 3 , 4 , 11 , 33 , 34 . Taken together, these findings confirm that SUDOSCAN® is a reasonable screening device for early DPN in both type 1 diabetes and type 2 diabetes. Our results strongly support its use to screen type 1 diabetes, without clinically overt DPN. Our findings align with those of Veloso et al. 17 , who reported sudomotor dysfunction in 26% of type 1 diabetes patients without clinical neuropathy and found associations with longer diabetes duration, higher HbA1c, and retinopathy, similar to our results. Feldman et al. 20 have proposed a structured two‐step approach for diagnosing and staging diabetic neuropathy using clinical and electrophysiological criteria, albeit focusing on large fiber damage. Our study adds to this by identifying additional metabolic risk factors, such as anemia and low vitamin D, which may contribute to small fiber dysfunction detectable via SUDOSCAN®. Our finding of higher circulating ‘active’ vitamin B12 levels in those with sudomotor dysfunction could be due to reverse causality, that is, family physicians are more likely to prescribe vitamin B12 in those with early symptoms. On the other hand, the association with lower hemoglobin (anemia) may represent a causal role through reduced tissue oxygen supply. We found sudomotor dysfunction in 38.5% of people with type 1 diabetes with normal MNSI, at a younger age and shorter duration of diabetes, compared with those with clinical DPN. These findings challenge the conventional wisdom that DPN is uncommon in early, uncomplicated type 1 diabetes. This so‐called ‘subclinical DPN’ has been poorly studied in type 1 diabetes cohorts and warrants longitudinal exploration. Veloso et al. studied people with type 1 diabetes in Brazil who did not have clinical evidence of DPN (i.e., MNSI normal) and found 16 (26.2%) to have sudomotor dysfunction based on SUDOSCAN® readings, supporting our findings and highlighting the need for further well‐designed studies 17 , 20 . Body weight and adiposity have previously been suggested as potential determinants of ESC, possibly reflecting differences in sweat gland density, skin properties, or other physiological factors influencing chloride ion flux 15 , 35 , 36 . In the present study, however, ESC values showed only weak and nonsignificant correlations with body weight and BMI for both hands and feet, suggesting that body size did not materially influence sudomotor measurements in this cohort. These findings are consistent with prior validation studies reporting that while anthropometric factors may contribute modestly to inter‐individual variability in ESC, they do not fully account for the association between sudomotor dysfunction and diabetic neuropathy 4 , 14 . Importantly, the persistence of the association between ESC abnormalities and clinically defined neuropathy after multivariable adjustment further supports the utility of SUDOSCAN® as a screening tool that is not primarily driven by body size. A further consideration is the use of a single ESC cutoff value across a wide age range. While ESC thresholds of <60 μS have been widely applied in adult populations, age‐specific normative values for children and adolescents are not well established. In our cohort, formal age‐stratified analyses were not feasible due to the small number of pediatric participants with neuropathy, particularly those with clinically overt DPN. Although all analyses were adjusted for age, the application of a uniform cutoff may still influence diagnostic performance metrics, such as AUC, PPV, and NPV, especially in younger participants. Future studies are needed to establish age‐specific reference ranges and optimal cutoffs for sudomotor function in pediatric populations. Finally, we acknowledge that normative ESC values and optimal diagnostic thresholds vary between populations and ethnic groups. Normative studies have demonstrated differences in mean ESC values across ethnic groups and have highlighted the importance of population‐specific reference ranges when defining screening cutoffs 35 , 36 . Lower mean ESC values have been reported in several non‐Caucasian populations, including Asian cohorts, supporting the need for locally derived normative data 13 . In the absence of large, locally derived reference datasets, we used the widely referenced ESC <60 μS threshold to facilitate comparison with prior validation and screening studies 4 , 35 . A major strength of our study is the large number of people with type 1 diabetes. We also present detailed clinical and biochemical measurements to investigate possible etiological factors of early neuropathy. The inclusion of people with a short disease duration allowed us to study the early stages of neuropathy, that is, ‘subclinical DPN’. The following limitations should also be considered: (i) our assessment was cross‐sectional and therefore the observed associations do not necessarily imply causality. A longitudinal follow‐up would provide information on the long‐term predictive utility of SUDOSCAN®. (ii) We did not compare SUDOSCAN® results with NCS, which are considered the gold standard and require strict control of skin temperature. Sudomotor assessment using SUDOSCAN® is relatively robust to minor environmental variations, making it particularly suitable for use in busy outpatient and epidemiological settings when standardized testing conditions are maintained. However, as described earlier, NCS are unsuitable for small fiber assessment. Instead, we compared SUDOSCAN® measurement with the MNSI, which is validated and the most widely used clinical and research tool for assessment of diabetic neuropathy, (iii) we have not ruled out other rare causes of neuropathy in our participants (amyloidosis, autoimmune conditions) or sweating‐related abnormalities such as anhidrosis which could potentially affect SUDOSCAN® measurements. (iv) Although the study included a wide age range, the number of participants with neuropathy in the pediatric age group was limited, particularly for clinically overt DPN. While all analyses were adjusted for age and sex, this restricted our ability to perform meaningful age‐stratified analyses. Future studies with larger cohorts comprising participants under the age of 18 years are needed to better characterize age‐specific patterns of sudomotor dysfunction and neuropathy in type 1 diabetes. (v) Although the duration of diabetes in the subclinical DPN group was not short in absolute terms, sudomotor dysfunction was detected at an earlier stage relative to the development of clinically overt neuropathy. (vi) Although SUDOSCAN® assessments were performed as part of the broader PICTURE cohort study, sudomotor analysis was not prespecified in the trial registration. Consequently, this work should be interpreted as an exploratory, post hoc analysis. CONCLUSIONS We found the SUDOSCAN® to be an acceptable screening device for DPN in type 1 diabetes. Compared with the standard clinical method (MNSI), this method identified early neuropathy in a substantial number of people with type 1 diabetes. Our results highlight the presence of nerve damage even in individuals with type 1 diabetes without clinically overt neuropathy when assessed using conventional screening methods and have the potential in clinical practice to help motivate patients to be extra vigilant about metabolic control to prevent more serious complications. Future directions Prospective follow‐up studies are warranted to establish the natural history of early neuropathy in type 1 diabetes, the validity of sudomotor testing in people with relatively shorter duration of diabetes as well as to establish a normative reference range for sudomotor impairment (measured as ESC) in people with type 1 diabetes under the age of 18 years. DISCLOSURE The authors declare no conflict of interest. Approval of the research protocol: Approved by the Institutional Ethics Committee of King Edward Memorial Hospital Research Centre, Pune (Reference: KEMHRC ID: Ph.D. 19). Informed consent: Obtained from all participants; those <18 years provided assent and parental consent. Approval date of registry and the registration no. of the study/trial: Prospectively registered at ClinicalTrials.gov (NCT03069196: 27/02/2017). Animal studies: N/A. Supporting information Appendix S1. Michigan Neuropathy Screening Instrument (MNSI) assessment tool. Acknowledgments The authors would like to acknowledge the T1D participants and thank them for their time. We also thank the Diabetes Unit research and clinical outpatient staff, including diabetes consultants for their support. Many participants in this study were supported by philanthropic grants from the Hinduja Foundation, Mumbai Mukul Madhav Foundation (MMF), Pune and Nityaasha Foundation, Pune. We are extremely grateful to Impeto Medical, Paris, France for providing the SUDOSCAN® device free of cost, enabling research and clinical care. This study was investigator‐initiated and led (C.S.Y) and the funders or device providers had no input in the study design, data collection, data analysis, or manuscript preparation. CSM is supported by the Central Adelaide Local Health Network Florey Fellowship. Open access publishing facilitated by Adelaide University, as part of the Wiley ‐ Adelaide University agreement via the Council of Australasian University Librarians Contributor Information Chinmay S. Marathe, Email: chinmay.marathe@adelaide.edu.au. Chittaranjan S. Yajnik, Email: csyajnik@gmail.com. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available from the corresponding author upon reasonable request. References Associated Data Supplementary Materials Appendix S1. Michigan Neuropathy Screening Instrument (MNSI) assessment tool. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.