Predictive role of weight-adjusted waist index in diabetic peripheral neuropathy among patients with type 2 diabetes mellitus

ABSTRACT:

Background

Obesity is a known risk factor for diabetic peripheral neuropathy (DPN), but the relationship between body mass index (BMI) and DPN remains controversial.

Aims

This study explores the relationship between weight-adjusted waist index (WWI) and DPN in Chinese patients with Type 2 Diabetes Mellitus (T2DM), with the goal of establishing WWI as an effective tool for the early detection and risk stratification of DPN.

Methods

A total of 1790 participants were included. The relationships between WWI, BMI, WC, and DPN were analyzed using logistic regression, restricted cubic spline, and subgroup analysis, with ROC-AUC used to assess the diagnostic performance of the obesity indices.

Results

During this cross-sectional study, a total of 895 cases of DPN were recorded. Each 1-SD increase in WWI was associated with an increased risk of DPN in Model 1 (OR: 1.72, 95% CI: 1.37–2.16), Model 2 (OR: 1.41, 95% CI: 1.10–1.80), and Model 3 (OR: 1.37, 95% CI: 1.06–1.77). A linear relationship was observed between WWI and DPN when WWI was greater than 10.757 cm/√kg (OR: 1.375, 95% CI: 1.115–1.7). BMI was nonlinearly associated with DPN (p-nonlinear < 0.05), and age significantly modified the relationship between BMI and DPN. ROC analysis indicated that WWI had superior diagnostic performance compared to BMI and WC.

Conclusion

WWI is linearly and independently associated with DPN and demonstrates greater stability and predictive value than BMI and WC. Therefore, WWI may serve as a more reliable indicator for identifying individuals at risk for DPN.

CORE TIP

Weight-adjusted waist index (WWI) is a reliable and comprehensive indicator of obesity. The aim of our present study was to assess the degree of association between WWI, WC and BMI and the risk of diabetic peripheral neuropathy (DPN). We found that WWI independently predicted DPN, with a subsequent increase in the risk of DPN in diabetic patients when WWI exceeded 10.757 cm/√kg.

1. Introduction

Diabetes has emerged as a prevalent public health concern, impacting a large number of individuals worldwide as an ongoing medical condition [1]. As the prevalence of diabetes increases, so does the incidence of diabetes-related comorbidities, including chronic microvascular complications such as diabetic peripheral neuropathy (DPN). According to the International Diabetes Federation, the global prevalence of diabetes is projected to rise significantly, from 536.6 million cases in 2021 to 783.2 million by 2045 [2]. Chronic microvascular complications of diabetes, such as DPN, are quite common. It is estimated that about 50% of diabetic patients can eventually develop DPN, of which 30%–40% experience painful symptoms [3]. DPN is the main driver of nontraumatic amputation in diabetic patients [4]. Chronic hyperglycemia, insulin resistance, prolonged duration of diabetes, and dyslipidemia are acknowledged risk factors for DPN [5–8]. Meanwhile, studies have proven that obesity is also considered a risk factor for DPN [9,10].

Obesity can be classified into generalized obesity and abdominal obesity. Body Mass Index (BMI) and waist circumference, which indicate general and abdominal obesity, respectively, are linked to a heightened likelihood of neuropathy [11,12]. However, there has been controversy regarding the relationship between BMI and DPN, with studies reporting positive, negative, and no associations, as well as a “U-shaped” relationship [13–15]. This is partly because BMI does not differentiate between the body’s fat and muscle content, thus leading to the seemingly contradictory results between obesity and neuropathy [16]. Weight-adjusted waist index (WWI), a novel anthropometric index, is strongly correlated with high fat and low muscle content, making it a useful measure of central obesity [17]. Compared to BMI, WWI is a better predictor of hypertension and diabetic nephropathy [18–20]. Furthermore, WWI predicts myasthenia gravis better than WC in patients with Type 2 Diabetes (T2D) [21]. The above studies show that WWI is a reliable and comprehensive indicator of obesity. A cohort study in rural China identified WWI as an independent predictor of T2D [22]. While previous studies have established WWI as a reliable and comprehensive indicator of obesity, there remains a gap in the research regarding its association with DPN, particularly in the Han Chinese population. In China, where diabetes and its complications are highly prevalent, understanding the role of WWI in predicting DPN is crucial. The aim of this study was to assess the association between WWI, WC, and BMI in relation to DPN risk in Han Chinese patients with T2D, and to explore the potential of WWI as a tool for early detection and risk stratification of DPN in this high-risk population.

2. Methods

2.1. Study design and participants

Participants in this study included 1790 patients with type 2 diabetes who had complete data on all variables at baseline. This retrospective study was conducted at the Department of Endocrinology, Tongji Hospital, Huazhong University of Science and Technology, from March 2023 to December 2023. We employed a hospital-based convenience sampling method. All participants underwent blood and urine sample collection, anthropometric measurements, and screening for diabetic complications during the initial visit. Diabetes diagnosis followed the criteria set by the American Diabetes Association in 2022 [23]. Exclusion criteria included: (1) age under 18 or over 80 years; (2) presence of neuropathies unrelated to diabetes; (3) acute diabetic complications such as diabetic ketoacidosis and hyperosmolar hyperglycemic syndrome; (4) chronic diseases including chronic kidney disease and cardiovascular conditions; (5) missing key clinical or anthropometric data; and (6) malignancies. The study protocol was approved by the Human Research Committee of Tongji Hospital and conducted in accordance with the ethical principles set forth in the Declaration of Helsinki (TJ-IRB20230881). As this was a retrospective study, the Institutional Review Board of the Department of Endocrinology, Tongji Hospital, Huazhong University of Science and Technology waived the requirement for informed consent, and all methods followed relevant guidelines and regulations.

2.2. Study variables

Patient data, including demographic, anthropometric, and laboratory variables, were collected. Physical measurements such as heart rate, blood pressure, WC, BMI, and WWI were obtained under resting conditions. Laboratory variables included liver function (ALT, AST), kidney function (creatinine and eGFR), blood glucose levels, and lipid profiles (total cholesterol, triglycerides, HDL, and LDL). WWI was calculated as WC divided by the square root of body weight (cm/√kg) [24]. After an 8-h fast, blood and first morning urine samples were collected and analyzed immediately. All measurements followed standardized protocols. For example, WC was measured at the level of the umbilicus while the patient was standing, using a flexible tape measure. BMI was calculated using the formula: Weight (kg)/[Height (m)]². Nerve Conduction Velocity (NCV) was measured using a Kipoint-4 electromyography system (Medtronic, Denmark). All equipment was regularly calibrated to ensure measurement accuracy. The studies were performed on the median nerve at the wrist and the tibial nerve at the ankle. All personnel involved in the measurements underwent specialized training. All equipment was calibrated before use and on a regular basis to ensure measurement accuracy and reliability.

2.3. Assessment of DPN

DPN is diagnosed based on a combination of symptoms and signs. Symptoms of neuropathy include numbness, reduced sensation, burning pain, or tingling, primarily in the feet, with symptoms spreading upward from the lower limbs. Signs of neuropathy include abnormalities in pinprick sensation, vibration sense, pressure sense, temperature sense, and ankle reflexes [25]. These five signs are assessed using a calibrated 128 Hz tuning fork for vibration testing, a Tiptherm rod for temperature testing, and 10-g monofilament and pinprick tests. These assessments are conducted according to the methods described in the report by the American Diabetes Association’s Foot Care Interest Group Workgroup [26]. After excluding neuropathies unrelated to diabetes, a clinical diagnosis of DPN was established if the patient exhibited typical neuropathic symptoms and at last one positive signs. In cases where symptoms and signs are atypical, nerve function tests and sensory threshold tests are used as adjuncts for diagnosis, with abnormalities in amplitude, nerve conduction velocity, or latency considered indicative of abnormal nerve conduction. The consensus criteria for diagnosing diabetic neuropathy include abnormalities in nerve conduction properties in at least one nerve among two different nerves (≥ 99th or ≤ 1st percentile), with the sural nerve being mandatory [27]. Although electrodiagnostic testing is a routine assessment for large fiber dysfunction, electrodiagnostic abnormalities are not required for clinical neuropathy diagnosis according to the ADA definition [4].

2.4. Statistical analysis

We described continuous variables using means or medians (interquartile ranges) based on their distribution. The significance of differences among continuous variables was assessed using ANOVA or the Kruskal–Wallis test, while the chi-square test was applied to categorical variables. Participants were divided into three groups based on WWI tertiles. We assessed the relationship between WWI and DPN using multivariate logistic regression, where DPN was treated as the dependent variable. Three models were developed to adjust for potential confounders: Model I did not adjust for any factors; model II adjusted for age and sex; and Model III adjusted for age, sex, diabetes duration, BMI, HbA1c, SBP, DBP, HDL, and LDL. Covariate selection was guided by previous research and carefully considered to prevent multicollinearity. HbA1c, BMI, diabetes duration, sex, and age are commonly used in combination to predict the occurrence of DPN [28,29]. SBP, DBP, HDL, and LDL have been demonstrated to be closely associated with the progression of DPN [30]. Restricted cubic splines were used to model the relationship between BMI, WC and WWI with DPN using logistic regression models. This approach was chosen to assess potential nonlinear associations between these variables and DPN, as previous studies have indicated that the relationship between obesity metrics and diabetes complications may not be strictly linear. Four knots were selected at the 5th, 35th, 65th, and 95th percentiles of the distribution to allow for flexibility in modeling the data while minimizing overfitting. Nonlinearity was tested by comparing the model with only linear terms to the model with both linear and cubic spline terms using a likelihood ratio test. Subgroup analyses were performed based on factors including gender (male or female), age (<65 or ≥65 years), BMI (<28 or ≥28 kg/m²), smoking status (yes or no), and diabetes duration (<5 years or ≥5 years). All analyses were performed using R software (version 4.3.1; the R Foundation for Statistical Computing, Vienna, Austria). Restricted cubic spline analysis was conducted using the rms package, logistic regression analysis was performed with the stats package, collinearity detection was carried out using the car package, and ROC analysis was performed using the pROC package. Statistical significance was defined as p < 0.05.

3. Results

The baseline characteristics of all participants are shown in Table 1. The average age was 55.4 years, and 694 participants were women. A total of 1790 participants were included in the study, of whom 895 were diagnosed with DPN, and 895 were diagnosed with diabetes alone (NDPN). Participants were categorized into three tertiles based on WWI values. Significant variations (p < 0.001) in sociodemographic factors, duration of diabetes, scr, and eGFR were observed among the WWI tertile groups. As WWI increased, several trends emerged. The highest WWI tertile was associated with a higher prevalence of DPN, indicating a potential link between central obesity and DPN. Higher WWI tertiles were more common among females, suggesting a potential role of gender in obesity distribution and its impact on neuropathy risk. Additionally, WC increased with WWI, reflecting central obesity and its association with higher abdominal fat, which may elevate the risk of diabetes-related complications, including DPN. Older participants were more likely to have higher WWI values, implying that age may contribute to longer durations of central obesity or adiposity. Table S1 presents the baseline characteristics of all participants, stratified by the presence or absence of diabetic neuropathy. Significant differences were observed between the two groups in terms of age, BMI, diabetes duration, and WWI. In Figure 1, we used restricted cubic splines to flexibly model and visualize the relation of predicted BMI, WC and WWI with DPN in the total study population.

Table 1.

Baseline characteristics of study participants by weight-adjusted waist index.

Variables	Q1 (<10.7912)	Q2 (10.7912-11.3846)	Q3 (>11.3846)
	N = 600	N = 593	N = 597	P value	ALL
Disease:				<0.001
NDPN	333 (55.5%)	311 (52.4%)	251 (42.0%)		850(50%)
DPN	267 (44.5%)	282 (47.6%)	346 (58.0%)		850(50%)
Gender:				<0.001
Female	142 (23.7%)	212 (35.8%)	340 (57.0%)		694
Male	458 (76.3%)	381 (64.2%)	257 (43.0%)		1096
Age, y	52.1 (9.97)	54.7 (9.89)	59.5 (9.77)	<0.001	55.4
BMI, kg/m2	24.5 (2.62)	24.6 (2.56)	25.1 (2.82)	<0.001	24.76
WC, cm	86.8 (6.62)	91.4 (6.92)	96.7 (7.43)	<0.001	91.6
Smoking	179 (29.8%)	171 (28.8%)	171 (28.6%)	0.888	521(29.8%)
Pulse, times/min	85.3 (12.6)	86.5 (12.1)	86.9 (12.9)	0.090	86.2
SBP, mmHg	134 (20.5)	132 (20.3)	131 (18.5)	0.049	132.1
DBP, mmHg	80.8 (12.3)	81.8 (11.5)	82.5 (12.2)	0.054	81.72
Duration of diabetes, y	10.0 [5.00;15.0]	9.00 [4.00;13.0]	6.00 [3.00;11.0]	<0.001	9.3
Glycemia, mmol/L	11.9 [8.35;16.2]	11.7 [8.04;15.7]	12.8 [8.40;17.3]	0.032	13
HbA1C, %	8.30 [7.20;10.1]	8.30 [7.10;10.2]	8.80 [7.30;10.4]	0.060	8.8
ALT, U/L	18.0 [13.0;27.0]	18.0 [13.0;26.0]	20.0 [14.0;29.0]	0.004	23.5
AST, U/L	19.0 [15.0;24.0]	18.0 [15.0;22.0]	18.0 [15.0;23.0]	0.108	20.9
TC, mmol/L	4.27 [3.58;5.11]	4.34 [3.58;5.10]	4.41 [3.78;5.09]	0.082	4.5
TG, mmol/L	1.95 [1.30;3.15]	1.95 [1.29;3.07]	2.12 [1.32;3.49]	0.127	2.7
HDL-C, mmol/L	1.02 [0.85;1.25]	1.05 [0.88;1.25]	1.03 [0.85;1.20]	0.089	1.07
LDL-C, mmol/L	2.56 (0.92)	2.64 (0.94)	2.71 (0.87)	0.029	2.6
Scr, umol/L	83.0 [68.0;98.0]	71.0 [61.0;83.0]	64.0 [55.0;74.0]	<0.001	74.4
eGFR, ml/min/1.73m2	81.4 [64.9;93.6]	94.4 [80.3;103]	104 [95.8;112]	<0.001	91.1

Note: Data are expressed as mean, medians (interquartile range), or number (%).

Abbreviations: DPN, diabetic peripheral neuropathy; NDPN, non-diabetic peripheral neuropathy; BMI, body mass index; WC, waist circumference; WWI, weight-adjusted-waist index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HbA1c, hemoglobin A1c; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; scr, serum creatinine; eGFR, estimated glomerular filtration rate.

Figure 1.

Restricted cubic spline analysis of BMI, WC, and WWI for the estimation of the risk of DPN after adjusting for multivariate covariates in total study population. Odds ratios are indicated by solid lines and 95% CIs by shaded areas. Reference point is lowest value for each of BMI and WWI (27.38 kg/m² for BMI and 10.757 cm/√kg for WWI), with knots placed at the 5th, 35th, 65th, and 95th percentiles of BMI distribution. A: BMI; B: WC; C: WWI. A and B: adjusted for age, gender, systolic blood pressure, diastolic blood pressure, duration of diabetes, hemoglobin A1c, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol. C: adjusted for age, body mass index, gender, systolic blood pressure, diastolic blood pressure, duration of diabetes, hemoglobin A1c, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol. (A) shows the non-linear association between BMI and DPN risk, with a threshold at 7.38 kg/m². (B) shows the non-linear association between WC and DPN risk, with no significant threshold identified. (C) shows the non-linear association between WWI and DPN risk, with a threshold at 10.757 cm/√kg.

3.1. Demographic characteristics of the participants

Table 2 illustrates the connection between WWI and the risk of DPN. WWI was divided into three categories: low (Q1), medium (Q2) and high (Q3) based on values. Across all models, WWI and DPN remained independently associated. In the crude model, participants in the highest WWI tertile (Q3) showed a markably increased risk of DPN (OR: 1.72, 95% CI: 1.37–2.16). Model 2 showed that individuals in Q3 had a 41% increased risk of DPN (OR: 1.41, 95% CI: 1.10–1.80). After adjusting for age, gender, BMI, diabetes duration, HbA1c, HDL-C, LDL-C, SBP, and DBP, individuals in the highest tertile (≥11.3846 cm/√kg) of the WWI had a 37% higher probability of developing DPN than those in the lowest tertile (≤10.7912 cm/√kg; OR: 1.37, 95% CI: 1.06–1.77). The results suggest that individuals in the highest WWI tertile have an elevated risk of DPN, underscoring the predictive value of WWI in identifying high-risk patients (Table 2). Collinearity analysis indicated that the variance inflation factor (VIF) for all variables was below 5. Restricted cubic spline analysis revealed no significant non-linear association between WWI and DPN in the fully adjusted model accounting for demographics and clinical factors. Above a WWI of 10.757 cm/√kg, the odds ratio per standard deviation increase in predicted WWI was 1.375 (95% CI: 1.115–1.7; Figure 1C).

Table 2.

Association between WWI and DPN by WWI tertiles.

WWI (cm/√kg)	OR (95% CI), P-value
	Model 1	Model 2	Model 3
Q1 (<10.7912)	Reference	Reference	Reference
Q2 (10.7912–11.3846)	1.13 (0.9, 1.42) 0.29	1.05 (0.83, 1.33) 0.672	1.03 (0.82, 1.31) 0.778
Q3 (>11.3846)	1.72 (1.37, 2.16) <0.001	1.41 (1.10, 1.80) 0.007	1.37 (1.06, 1.77) 0.015
P for trend	<0.001	0.014	0.027

Note: Model 1: crude model; model 2: adjusted for age and gender; model 3: adjusted for age, gender, BMI, systolic blood pressure, diastolic blood pressure, duration of diabetes, hemoglobin A1c, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol.

Abbreviations: WWI, weight-adjusted-waist index; DPN, diabetic peripheral neuropathy; or, odds ratio; CI, confidence interval.

3.2. Association analysis of BMI and WC with DPN

Estimated associations between BMI and DPN are shown using non-linear spline models (Figure 1A) and piecewise linear models (Table 3) in total study population. After adjusting for duration, HbA1c, SBP, DBP, HDL, and LDL, the inflection point for DPN was identified at a BMI of 27.38 kg/m² (P values for log-likelihood ratio < 0.05). For BMI values above this BMI, the risk of DPN increased (OR: 1.145, 95% CI: 1.023–1.287), while below this BMI, an inverse association was observed (OR: 0.901, 95% CI: 0.857–0.947). However, the associations between WC and DPN was insignificant (Figure 1B).

Table 3.

Threshold effect analysis of BMI on DPN in all participants.

	The effect size, 95%CI	P-value
Model 1 Fitting model by standard linear regression	0.955 (0.92–0.99)	0.013
Model 2 Fitting model by two-piecewise linear regression
Inflection point	27.38
<27.38	0.901 (0.857–0.947)	<0.05
>27.38	1.145 (1.023–1.287)	0.02
P for likelihood ratio test		<0.001

Note: Adjusted for age, gender, systolic blood pressure, diastolic blood pressure, duration of diabetes, hemoglobin A1c, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol.

Moreover, we categorized the subjects into two groups based on the value of BMI: those below 27.38 kg/m² and those above 27.38 kg/m². We did stratified analyses to explore whether the association of BMI with DPN varied across sex and age. We observed a significant interaction between the BMI and age (p < 0.05). Age-stratified analyses showed that the association between BMI and DPN risk in age ≥ 60 in the multivariable-adjusted reference model had a significant inverse association (OR: 0.59, 95% CI: 0.37–0.94; Table 4). Considering the interaction between age and BMI, the subjects were divided into two groups with the cut-off age of 60, and the relationship between BMI and DPN was analyzed separately for each group (Figure 2). It was found that in patients age < 60 years old, BMI was negatively correlated with DPN for BMIs < 24.8 kg/m² and positively correlated for higher BMI. There was no significant interaction observed between sex and BMI.

Table 4.

ORs (95% CIs) Of DPN according to the body mass index.

OR (95%CI) P-value
BMI	<27.38	≥27.38	P interaction
Overall	1	0.91 (0.7,1.19) 0.494
Gender			0.626
Male	1	0.97 (0.7,1.35) 0.864
Female	1	0.82 (0.53,1.28) 0.387
Age, years			0.033
<60	1	1.12 (0.81,1.54) 0.5
≥60	1	0.59 (0.37,0.93) 0.024

Note: Stratified analysis of the relationship between BMI and DPN in patients. Age, gender, systolic blood pressure, diastolic blood pressure, duration of diabetes, hemoglobin A1c, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol were adjusted. In the stratified analyses, the models were not tuned for the stratification variables themselves.

Figure 2.

Association between BMI and DPN among diabetes, by age. (A) < 60 years old; (B) ≥ 60 years old. The reference point is 24.8 kg/m² for BMI, with knots placed at the 5th, 35th, 65th, and 95th percentiles of BMI distribution. The relationship between BMI and DPN was analyzed separately for two age groups, with a cut-off age of 60. In patients aged < 60 years, BMI was negatively correlated with DPN for BMI < 24.8 kg/m² and positively correlated for higher BMI.

3.3. Subgroup analysis

To assess the stability of the relationship between WWI and DPN across different population settings, subgroup analyses and interaction tests were conducted. Figure 3 shows the results of these subgroup analyses and interaction tests. The effects of WWI on DPN were evaluated using subgroups based on gender, age, BMI, current smoking status, and diabetes duration. These analyses suggested a stronger positive association between WWI and DPN in younger (≤65 years), female, and non-obese diabetic patients (BMI < 28 cm/√kg). However, no significant interactions were observed between these subgroups and WWI in relation to DPN (all P for interaction >0.05). Therefore, further studies are needed to analyze these trends.

Figure 3.

Subgroup analysis of the relationship between WWI and DPN in patients. Age, gender, BMI, systolic blood pressure, diastolic blood pressure, duration of diabetes, hemoglobin A1c, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol were adjusted. In the subgroup analyses, the models were not turned for the stratification variables themselves. sSubgroup analyses showed a stronger positive association between WWI and DPN in younger (≤ 65 years), female, and non-obese (BMI < 28 kg/m²) diabetic patients. No significant interactions were found (all P for interaction > 0.05), indicating the need for further research.

3.4. ROC analysis

Analysis revealed that the AUC for BMI, WC, and WWI in predicting diabetic neuropathy were 0.53, 0.50, and 0.56, respectively (Table S2). While the differences are modest, WWI demonstrated slightly better predictive performance, suggesting potential clinical advantages in identifying individuals at risk of DPN.

4. Discussion

This study addresses the gap in research regarding the predictive utility of the WWI for DPN. Our findings demonstrate a significant association between increased WWI and a higher DPN risk, which remained significant after adjusting for potential confounders such as gender, age, BMI, and diabetes duration. Notably, WWI emerged as an independent predictor of DPN, with a cutoff point of 10.757 cm/√kg, above which the risk of DPN increased. WWI exhibited the highest AUC value compared to the other two indices. While the advantage of WWI in terms of AUC is not substantial, it still demonstrates greater utility in identifying metabolic abnormalities among individuals with a normal BMI and providing a more comprehensive assessment of central obesity. Given the limitations of BMI in distinguishing central obesity and the inconsistent findings regarding the relationship between WC and DPN, WWI may serve as a more reliable predictor, particularly when combined with other obesity-related metrics, warranting further investigation.

With the increasing obesity rate in China, the risk of diabetes and its complications has also risen. According to data from the China Chronic Disease and Risk Factor Surveillance (CCDRFS) program, obesity prevalence rose from 3.1% in 2004 to 8.1% in 2008 [31]. A population-based study of obesity suggested that 14.1% of the study population was obese, and 34.8%were overweight in China [32]. Approximately two-thirds of Chinese adults will be affected by obesity, and it is expected that by 2030 [33]. Obesity is intrinsically related to numerous diseases, including cardiovascular ailments, all-cause mortality, and diabetes and its complications [34–37]. Recent studies have found that central obesity measurements are key independent risk factors for cardiovascular and microvascular diseases in individuals with prediabetes and diabetes, even in those with a normal BMI [38]. Gao et al. discovered that the fat mass index (FMI) and WC are superior to BMI in identifying the risk of obesity-related diabetic neuropathy, thereby supporting the association between obesity and the progression of diabetic neuropathy [39]. Studies have shown that central obesity, particularly abdominal fat accumulation, is a critical predictor of diabetic microvascular complications. This underscores the importance of finding convenient and effective indices to assess obesity.

Existing studies on the association between BMI, WC, and the risk of DPN have reported conflicting results. Some studies have found a positive correlation between BMI and DPN risk, while others report no significant association or even a negative correlation. Similarly, the relationship between WC and DPN risk has shown variability across different populations. In our study, we used multiple anthropometric measures, including BMI, WC, and the WWI, to assess DPN risk. Our findings indicate that WWI, which accounts for both fat and muscle content, is a more reliable predictor of DPN risk than BMI or WC alone. This comprehensive approach helps clarify the conflicting evidence and provides a more accurate assessment of obesity-related risk factors for DPN.

The prevalence of obesity has risen markedly, underscoring the importance of exploring effective indices for obesity assessment. BMI and WC are the two most widely used obesity measurements, which are selected to assess overall adiposity and abdominal obesity, respectively. Previous studies have emphasized that BMI and WC were independently correlated with several somatic diseases, including hypertension, non-alcoholic fatty liver disease (NAFLD), diabetes and its complications [40–42]. Cumulative evidence has indicated that reduced muscle mass and increased visceral fat content can increase the risk of physical disability, metabolic disorders, and cardiovascular disease [43–45]. However, traditional anthropometric measurements cannot distinguish muscle and fat content. In our study, WC was not associated with DPN, whereas BMI was nonlinearly associated with DPN. For BMI, a reduction of the risk was seen within the lower range until 27.38 kg/m², with an OR of 0.901 (0.857–0.947) per standard deviation, which increased thereafter (P for non-linearity = 0.003). In a prospective cohort study in the United States, BMI was J-associated with all-cause mortality in men. The J-shaped correlation between BMI and all-cause mortality changed to a linear correlation after excluding those with low lean body mass. Differences in body fat and muscle content between patients will affect the relationship between BMI and disease [46]. Given the aforementioned, WWI was proposed as an innovative adiposity index in 2018. Unlike traditional anthropometric measurements, WWI offers a more nuanced insight into body composition, correlating positively with abdominal fat and negatively with muscle mass, making it an exemplary metabolic index for disease prediction. Previous research has demonstrated that WWI was, respectively, positively and negatively associated with abdominal fat measures and abdominal muscle mass measures [47]. This may explain why WWI is an excellent metabolic index to BMI for predicting several diseases.

Accumulating data indicates that obesity is linked to a higher risk of peripheral neuropathy [48]. The underlying mechanism may be that the peripheral nervous system (PNS) lacks protection from the blood-brain (nerve) barrier. This leads to the loss of peripheral sensory neurons and tiny nerve fibers within the epidermis, which induces the corresponding symptoms. Obesity affects lipid metabolic processes by over-activating the autonomic nervous system, thereby affecting lipid metabolism. Furthermore, obesity-mediated increases in lipid metabolites, such as free fatty acids and long-chain fatty acids, alter normal lipid metabolism in the PNS and induce endoplasmic reticulum dysfunction, thereby exacerbating PNS damage [49]. Nerve damage is an irreversible process. Obesity, as a variable risk factor, deserves to bet more attention. Our findings demonstrate that WWI is a reliable anthropometric indicator for predicting DPN, particularly in non-obese people with a BMI of less than 27.38 kg/m². Given this, we should pay more attention to WWI, and individuals need to be encouraged to maintain WWI < 10.757 cm/√kg to prevent DPN.

To our collective knowledge, this is the inaugural cross-sectional research to identify the correlation between WWI and DPN. The meticulous study design and data collection are two other advantages of our study. All of our subjects passed a rigorous neuropathic deterioration, which enabled the relationship between WWI and DPN more accurate and realistic.

This study does have some disadvantages. First, its retrospective design limits the ability to establish causality between WWI and DPN, the lack of longitudinal data prevents us from evaluating the temporal relationship between WWI and DPN or assessing the long-term effects of WWI on DPN risk. Second, although we adjusted for covariates, there were still potential variables that could be related to DPN that were not included in our study, such as glucose-lowering medication use, the duration of DPN, and other chronic conditions. Third, our study was conducted at a single center, which may limit the generalizability of our findings, making them only applicable to the Chinese population. Fourth, the lack of comprehensive data on lifestyle factors (e.g. physical activity, dietary habits) and medication use, which may serve as potential confounders, limits the understanding of the relationship between WWI and DPN. Finally, potential measurement biases, such as variability in nerve conduction tests, may affect the results of our study.

In conclusion, we discovered that elevated WWI may be associated with a higher risk of DPN. Compared to BMI, WWI is linearly associated with DPN and is more stable, less affected by other variables. In the future, it could be incorporated into risk calculators or routine anthropometric assessments, with the WWI threshold (10.757 cm/√kg) helping clinicians more effectively identify high-risk populations. However, further research is still needed to clarify our results, after which it may be widely applied in clinical practice.

Funding Statement

This study was supported by grants from the National Natural Science Foundation of China (grant nos.82100922H.R, 82270855 G.Y) and the Ministry of Science and Technology of Hubei (2022BCA014).

Disclosure statement

No potential conflict of interest was reported by the authors.

Institutional review board statement

This investigation was approved by the Human Research Committee of Wuhan Tongji Hospital.

Informed consent statement

The need for patient consent was waived due to the retrospective nature of the study (TJ-IRB20230881).

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to their containing information that could compromise the privacy of research participants.