Presence of Acanthosis Nigricans Indicates Nephropathy In People With Type 2 Diabetes

Introduction

Type 2 diabetes (T2D) is a major public health concern worldwide, and in South Asia.^1,2 The magnitude of diabetes-related complications, therefore, is substantial and increasing. Among microvascular and macrovascular complications, cardiovascular disease and nephropathy are more prevalent in South Asians than in Caucasians.³

Among the commonly assessed parameters that can help predict the risk of T2D-related complications without the need for laboratory testing are age, degree and pattern of obesity, family history of diabetes, blood-pressure status, physical inactivity, dietary habits, and tobacco or alcohol use. However, clinical phenotypic markers like acanthosis nigricans (AN) have not been adequately researched. Previously, we found that those with AN had significantly higher rates of hepatic fat accumulation and liver fibrosis, suggesting it could serve as an easily identifiable clinical marker for liver complications in people with diabetes.⁴

AN is a skin condition that causes melanocytosis, basal hyperpigmentation, papillomatosis, hyperkeratosis, and epidermal and dermal hyperplasia. It is marked by dark, thickened, velvety patches that are usually seen in body folds such as the neck, axillae, and groin. It is seen as a visible early indicator of metabolic syndrome and T2D.⁵ Although it is a reactive cutaneous change that is directly linked to obesity, insulin resistance, endocrinopathies, or malignancies, it is frequently associated with T2D.⁶ Ethnicity appears to be a separate influence, with AN being common among African-Americans, Hispanics, and American Indians, less frequent in whites, and showing varying prevalence, particularly high among Asians and those of mixed ancestry.^7,8 The frequency of AN among individuals with T2D varied greatly across multiple studies, ranging from 5.5–62.6%, depending on demographic and geographic characteristics. Research about AN is sparse in India. Prevalence has been reported to be higher in North India (62.6%)⁹ and 14.5% in the general population of adolescents in South India.¹⁰

AN has recently garnered attention as a potential marker for underlying vascular dysfunction in diabetes.¹¹ However, clinical trials directly addressing this association are lacking. Notably, there is no robust evidence evaluating whether the presence or severity of AN can reliably reflect the risk or burden of chronic microvascular and macrovascular complications in individuals with established T2D, highlighting a critical and unmet need in current literature.

Addressing this knowledge gap, we hypothesized that moderate to severe AN in people with T2D is associated with a greater magnitude of both microvascular and macrovascular complications compared to those with T2D but without non-AN (NAN). Given its high visibility and ease of clinical recognition, AN may serve as a simple, non-invasive marker for early identification of vascular risk in T2D.

Materials and Methods

We conducted a prospective, observational, cross-sectional comparative study between January 2024 and September 2025 from Fortis C-DOC Hospital for Diabetes and Allied Sciences, New Delhi, India. The participant consent and institutional ethics committee approval (F.No.1.1/FCDOC/EC/HAO/2024-25) was obtained. The study was registered with the Clinical Trial registry India (CTRI/2024/05/067815) and Clinicaltrials.gov (USA) (NCT06386939). The inclusion criteria for the AN group were as following: T2D, diabetes duration up to 30 years, age 20–70 years, both genders and presence of moderate or severe AN, based on physical examination by the clinician.

People having T2D with AN (n, 150) and NAN (n, 150) were recruited. Patients with congestive heart disease, severe end-organ damage, or chronic diseases: renal/hepatic failure, any malignancy, major systemic illness, etc., were excluded.

The following assessments were carried out among the study participants. General physical examination included the patient’s blood pressure, height, weight, body mass index (BMI), waist circumference (WC), hip circumference (HC), and handgrip strength using a Jamar dynamometer (as previously described), were recorded.¹² Waist-to-hip-circumference ratio (W-HR) was calculated. The external phenotypic markers such as, buffalo hump, xanthelasma, double chin, and arcus were observed. The biochemical investigations included fasting blood glucose (FBG), post-prandial blood glucose (PPBG), glycated hemoglobin (HbA1c), total cholesterol (TC), serum triglycerides (TG), high density lipoprotein-cholesterol (HDL-C), low density lipoprotein-cholesterol (LDL-C), serum glutamic oxaloacetic transaminase (SGOT), serum glutamic pyruvic transaminase (SGPT) and gamma-glutamyl transferase (GGT) levels, serum creatinine (Cr), microalbumin urine test (MAU), albumin-to-creatinine ratio (ACR), hemoglobin (Hb) and platelets.¹³ eGFR was estimated using the Chronic Kidney Disease-Epidemiology Collaboration equation (CKD-EPI) and CKD was staged according to KDIGO 2012 Clinical Practice Guideline.

Evaluation of AN

We did not include people having mild AN in this study. For the evaluation of moderate and severe AN, neck texture and severity were assessed using a graded system. Moderate AN was characterized by visually observable coarseness of the neck skin (texture grade 2) and scaly, velvety pigmented patches extending to the posterior border of the sternocleidomastoid (most frequently 3-6 inches), which are not visible from the front (severity grade 3). Severe AN was defined by extremely coarse skin with visibly raised “hills and valleys” (texture grade 3) and scaly pigmented patches extending anteriorly (∼ >6 inches), clearly visible when the patient is viewed from the front (severity grade 4). Figure 1 shows the pictures of severe AN.

Close

Figure 1: Figure shows pictures of severe acanthosis at the neck region from a 69-year-old male with type 2 diabetes, duration of 10 years, having body mass index 38.06 kg/m², waist circumference 132 cm, Waist-to-hip-circumference ratio 1.07. His glycemic parameters were - fasting blood glucose:190 mg/dL, 2-hour postprandial glucose: 258 mg/dL, and HbA1c:10.2%. The panel shows (a) left lateral neck view, (b) posterior neck view, (c) anterior neck view, (d) right lateral neck view. AN was also present over the axilla and face in some patients (not shown in the figure).

Export to PPT

Results

Of 300 participants with T2D, 150 had AN and 150 NAN. Groups were age-matched (mean age 52.1 vs. 53.8 years, p = 0.120), though the AN cohort included proportionally more men (76.7% vs. 64.7%, p = 0.031) and had slightly shorter diabetes duration (10.4 vs. 11.9 years, p = 0.044) [Table 1].

Mean BMI was significantly elevated in the AN group (p <0.001), as were WC (p <0.001) and W-HR (p = 0.015). Physical features, including double chin (p <0.001), buffalo hump (p <0.001), and arcus (p = 0.001), were significantly increased in the AN group [Table 1].

Lifestyle factors differed substantially between groups. Regular exercise was less common among those with AN (p = 0.004), while smoking or tobacco use was threefold higher (p <0.001) and alcohol consumption more frequent (p = 0.034) [Table 1].

The majority of participants in both groups were receiving metformin (90.7 vs. 94.0%) and SGLT2 inhibitors (76.0 vs. 78.7%). Use of sulfonylureas, including gliclazide and glimepiride, as well as DPP-4 inhibitors, repaglinide, α-glucosidase inhibitors, and insulin therapy was similar across groups. Use of semaglutide was substantially higher among patients with AN compared with those without AN (22.7 vs. 6.7%). Use of other GLP-1 based drugs (non-semaglutide) was slightly higher in the AN group (4.7 vs. 2.0%). Statin therapy and antihypertensive treatment were equally prevalent in both groups (75.3 vs. 72.0% and 48.7 vs. 48.0%, respectively), [Supplementary Table S2].

Despite comparable glycemic control, with similar FBG (p = 0.904) and PPBG (p = 0.532), the AN group demonstrated significant dyslipidemia. Serum TG was elevated (p = 0.007), while HDL cholesterol was reduced (p = 0.002). Mean TC and LDL cholesterol showed no significant intergroup differences. Liver function tests (SGOT, SGPT, and GGT) and hematological parameters were comparable between groups [Table 2].

Microvascular complications

Mean eGFR was comparable between the AN and NAN groups (93.8 ± 22.0 vs 92.2 ± 18.7 mL/min/1.73m²; p = 0.523). CKD stage distribution also did not differ significantly between the groups (p = 0.304) [Table 2]. MAU was significantly elevated (p=0.004) in AN group compared to NAN group, despite similar serum creatinine levels (p=0.537). Specifically, nephropathy was significantly more prevalent in the AN group compared with the NAN group (50.0 vs. 17.3%, absolute difference 32.7%, 95% CI 22.1-43.3). Unadjusted analysis revealed a nearly fivefold increased odds (OR 4.77, 95% CI 2.77-8.21, p <0.001). After adjustment for age, sex, diabetes duration, BMI, exercise, smoking, and alcohol use, the association remained significant (adjusted OR 4.21, 95% CI 2.35-7.53, p <0.001), indicating that AN was associated with more than fourfold higher odds of nephropathy after multivariable adjustments. In the secondary analysis, the prevalence of AN was 75.0% among those with nephropathy versus 37.5% among those without nephropathy (p <0.001), yielding an odds ratio for AN in the nephropathy group of 4.93 (95% CI 2.78–8.74), consistent with our primary finding. In contrast to neuropathy (p = 0.94) and retinopathy (p = 0.73) showed no association with AN presence [Table 3, Figure 2].

Table 3: Prevalence and regression analysis of diabetes-related complications in patients with AN vs. NAN

Complication	AN (n = 150)	NAN (n = 150)	Absolute difference (%)	Unadjusted OR (95% CI)	p value	Adjusted OR* (95% CI)	p value
Microvascular complications, n (%)
Nephropathy	75 (50.0)	26 (17.3)	+32.7	4.8 (2.8–8.2)	<0.001	4.2 (2.4–7.5)	<0.001
Retinopathy	33 (22.0)	34 (22.7)	-0.7	1.0 (0.6–1.7)	0.89	1.0 (0.6–1.7)	0.94
Neuropathy	83 (55.3)	87 (58.0)	-2.7	0.9 (0.6–1.4)	0.66	0.9 (0.6–1.5)	0.73
Macrovascular complications, n (%)
Macrovascular disease	14 (9.3)	15 (10.0)	-0.7	0.9 (0.42–2.1)	0.86	1.0 (0.4–2.4)	0.96
Abnormal carotid doppler	72 (48.0)	90 (60.0)	-12.0	0.6(0.4–1.0)	0.045	0.7 (0.4–1.1)	0.11
Peripheral vascular disease	19 (12.7)	23 (15.3)	-2.7	0.8 (0.4–1.6)	0.55	0.8 (0.4–1.7)	0.60

Data are presented as n (%) for prevalence. OR: Odds ratio, CI: Confidence interval, AN: Acanthosis nigricans, NAN: Non-Acanthosis nigricans. *Adjusted for age, sex, diabetes duration, body mass index, regular exercise, smoking/tobacco use, and alcohol consumption.

Close

Figure 2: Forest plot showing odds ratios (OR) with 95% confidence intervals (CI) for diabetes-related conditions in individuals with Acanthosis nigricans (AN) vs. without (NAN). Red circles represent unadjusted ORs; blue circles represent adjusted ORs controlling for age, sex, diabetes duration, body mass index, exercise, smoking, and alcohol use. The vertical dashed line at OR = 1.0 indicates no association. An odds ratio (OR >1) indicates increased odds of the complication. Values right of the line indicate increased odds in the AN group; values left indicate reduced odds. Confidence intervals crossing the line indicate non-significant associations. OR: Odds ratio, CI: Confidence interval.

Export to PPT

Discussion

Our findings demonstrated that nephropathy was significantly more frequent in individuals with T2D and AN, while associations with retinopathy, neuropathy, and macrovascular disease were weaker and inconsistent. To our knowledge, this is the first study to report a statistically significant association between AN and diabetic nephropathy in T2D. This research is a continuation of our prior work, where we showed that AN independently predicts hepatic fibrosis and steatosis in T2D.⁴

The link between AN and systemic hyperinsulinemia is well established. Fasting insulin and Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) are significantly higher in those with AN.¹⁵ Guevara-Gutierrez et al. also showed that the prevalence of abnormal intima-media thickness was higher in patients with AN vs. healthy participants (62.2% vs. 35.5%, p = 0.02).¹⁶ Another study showed the relationship between atherosclerosis and cardiovascular disease in people with AN.¹¹

IR, a hallmark of AN, adversely has the potential to adversely affect renal health. IR is highly prevalent in CKD and contributes to the progression of renal disease by worsening renal hemodynamics through various mechanisms, including activation of the sympathetic nervous system, sodium retention, and downregulation of the natriuretic peptide system.¹⁷ Impaired insulin signaling damages podocytes and tubular function, whereas insulin-driven sodium retention worsens hypertension.¹⁸ Insulin resistance independently predicts CKD even without diabetes, as reported in the Chinese population, reinforcing the biological relevance of our findings.¹⁹

Evidence for other microvascular diseases in people with AN has been variable. Gad et al.²⁰ showed adolescents with obesity and AN with small nerve fiber loss, indicative of a sub-clinical neuropathy.

CKD is an escalating burden in India and South Asia, where its onset in diabetes is often earlier and progression faster than other complications.^21,22 Despite this, CKD remains under-recognized until late irreversible stages. Identifying AN as a clinical predictor may facilitate earlier suspicion and screening, particularly in resource-limited settings lacking routine albuminuria or eGFR monitoring.

The clinical implications are considerable. AN is a cost-free, easily identifiable marker observable in primary care. Our results indicate that moderate-to-severe AN in T2D should prompt heightened vigilance for nephropathy, with appropriate renal evaluation if not done already. While longitudinal data are required to establish causality, AN should not be viewed as a benign cutaneous finding but as a visible indicator of nephropathy risk. If validated across larger cohorts, AN may be incorporated into risk-prediction models for diabetes-related complications and influence routine clinical practice.

This study has a few limitations. Being a single-center study conducted in an urban tertiary care setting, the generalizability of these findings to rural populations and other ethnic groups may be limited. 24-h urinary protein estimation and insulin resistance parameters such as HOMA-IR were not assessed due to logistical constraints, cost considerations, and challenges with patient compliance in an outpatient setting. Kidney biopsy, although a definitive diagnostic tool, was not feasible given the observational design and challenges in obtaining patient consent. Additionally, dyslipidemia parameters (serum TG and HDL-C) were not included as covariates in the multivariable model due to the influence of lipid-lowering therapies, including statins. The relatively modest sample size may further limit generalizability.

In conclusion, AN independently predicts diabetic nephropathy with ∼4-fold increased risk, representing a previously underrecognized association and offering a low-cost clinical indicator for early identification of renal risk. This novel finding suggests that the presence of AN in people with T2D needs intensified nephropathy surveillance.