The present study was approved by the Medical Ethics Committee of Nantong First People's Hospital, as indicated by reference number 2023KT228. Initiated in 2023, the study involved the recruitment of eligible patients with type 2 diabetes (T2D) from the Department of Endocrinology. The inclusion criteria were as follows: (1) a diagnosis of T2D in accordance with the 2020 Diabetes Management Guidelines issued by the American Diabetes Association [9]; (2) an age range from 20 to 75 years; and (3) full understanding of the study procedures with informed consent provided for participation. The exclusion criteria were as follows: (1) presence of autoantibodies associated with diabetes; (2) diagnosis of malignant tumors; (3) chronic viral hepatitis or liver cirrhosis; (4) cardiovascular conditions, including stroke, myocardial infarction, cardiovascular revascularization, and peripheral artery occlusion; (5) chronic kidney disease, characterized by an estimated glomerular filtration rate (eGFR) of less than 60 ml/min/1.73 m2; (6) use of glucocorticoids or sex hormone therapy; (7) administration of glucagon-like peptide-1 receptor agonists; (8) presence of anemia or deficiencies in folic acid or vitamin B12; (9) cervical and lumbar disorders; and (10) connective tissue diseases. Finally, the study enrolled a total of 654 eligible patients with T2D who had complete data. This cohort included 174 individuals with both T2D and obesity (T2D/OB, defined as a BMI ≥ 28 kg/m2) and 480 individuals with T2D and nonobesity (T2D/non-OB, defined as a BMI < 28 kg/m2), utilizing the Chinese BMI cutoff point for obesity classification [10].

A comprehensive collection of clinical data was obtained, encompassing anthropometric and demographic parameters such as age, sex, height, weight, BMI, and systolic/diastolic blood pressure (SBP/DBP). Additionally, data on the duration of diabetes, prescribed medications (including statins and hypoglycemic agents), and biochemical markers were included.

Following an 8-h fasting period, peripheral venous blood samples were obtained to quantify the concentrations of alanine aminotransferase (ALT), albumin, triglyceride (TG), total cholesterol (TC), uric acid (UA), cystatin C (CysC), hemoglobin, glycated hemoglobin (HbA1c), glycated albumin, and fasting C-peptide. Concurrently, morning urine samples were collected to assess albumin and creatinine levels, enabling the calculation of the urinary albumin-to-creatinine ratio (ACR). Additionally, the estimated glomerular filtration rate (eGFR) was calculated via the Modification of Diet in Renal Disease (MDRD) equation [11].

A 75-g oral glucose tolerance test (OGTT) was employed to evaluate the function of pancreatic α-cells and was administered to all participants in the morning after an overnight fast. Venous blood samples were obtained at baseline (fasting) and at 30, 60, 120, and 180 min following glucose administration to concurrently quantify plasma glucagon concentrations (GLA0min, GLA30min, GLA60min, GLA120min, and GLA180min). The overall glucagon response during the OGTT was evaluated by the area under the curve for glucagon (AUCgla). Plasma glucagon was measured using a chemiluminescence immunoassay (Glucagon Kit, JINDE BIOTECH, Guangzhou, China) with the HomoG100 analyser. The kit employs a double monoclonal antibody sandwich method, achieving a detection precision bias within ± 10% and both intra-assay and interassay CVs under 10%.

Nerve conduction studies conducted through electromyography are recognized as the most sensitive, objective, and reliable techniques for evaluating DPN and quantifying nerve function, especially in asymptomatic patients [12, 13].

We used an electromyography (MEB-9200 K, Nihon Kohden, Japan) to assess peripheral nerve function in all patients. Nerve latency, amplitude, and conduction velocity (NCV) were measured in the median (MN), ulnar (UN), common peroneal (CPN), posterior tibial (PTN), superficial peroneal (SPN), and sural (SN) nerves.

After standardizing the functional data using Z-scores, overall composite Z-scores for latency, amplitude, and nerve conduction velocity (NCV) were calculated by averaging the respective parameters across all peripheral nerves. Specifically, the composite Z-score for latency was derived by calculating the mean of the latency Z-scores across all peripheral nerves, a methodology that has been previously documented in the literature [13, 14].

In addition, we computed the composite Z-scores for motor nerves (MN, UN, CPN, and PTN) as well as the composite Z-scores for sensory nerves (MN, UN, SPN, and SN).

Statistical analyses were conducted using IBM SPSS Statistics, Version 25.0. A p value of less than 0.05 was deemed indicative of statistical significance.

Initially, descriptive statistical analyses were undertaken. Continuous data following a normal distribution are reported as the means and standard deviations, whereas nonnormally distributed continuous data are presented as medians and interquartile ranges. Categorical data are expressed as frequencies and percentages. AUCgla and other nonnormally distributed data were transformed using the natural logarithm for further analytical procedures, such as lnAUCgla.

Second, we utilized the Student's t test, the Mann–Whitney U test, and the chi-square test to assess differences in normally distributed, skewed, and categorical data, respectively, between the T2D/OB and T2D/non-OB groups.

Third, Pearson's correlation analysis was conducted to evaluate the relationships between glucagon levels (AUCgla) and peripheral nerve functional indices within the T2D/OB and T2D/non-OB groups.

Finally, multivariate linear regression analyses were employed to adjust for additional clinical variables, thereby determining whether glucagon levels (AUCgla) were independently associated with peripheral nerve function in patients with T2D.