Obesity-Related Phenotype Linked to Diabetic Kidney Disease Risk
Key Takeaways
- Among people with type 2 diabetes at a Beijing center, the obesity-related metabolic risk (OMR) class, which made up 39.8% of the cohort, was significantly associated with incident kidney disease and neuropathy compared with the low-risk (LR) class (45.9% of the cohort).
- The behavioral-visceral adiposity (BVA) class (14.3% of the cohort) was significantly associated with neuropathy, but its kidney disease association was not statistically significant.
- Higher SES was associated with greater BVA membership and lower OMR membership relative to LR.
- Exploratory within-SES associations differed in where significance was observed, without establishing an interaction.
- Follow-up was short (about two years) with few events, and the BVA class was almost entirely men who smoked and drank.
In a cohort study, investigators retrospectively analyzed patients at the National Metabolic Management Center, Fangshan Hospital, Beijing University of Chinese Medicine, China, enrolled January 2021–December 2025. Latent class analysis (LCA) included 706 of 3,257 patients screened, after exclusions for missing follow-up, missing data, and comorbidities. Both complications were assessed in outcome-specific risk sets built by excluding people who already had that condition, leaving 537 at risk for incident diabetic kidney disease (DKD) and 559 for diabetic peripheral neuropathy (DPN). Behavioral inputs covered smoking, alcohol, activity, sodium, and produce intake; metabolic inputs covered blood pressure, body mass index, visceral fat, C-peptide-based insulin resistance, glycated hemoglobin, and the triglyceride-to-high-density-lipoprotein cholesterol ratio. The three-class solution had the lowest Bayesian information criterion (BIC), although the Akaike information criterion and adjusted BIC continued to improve with four classes; entropy, classification certainty, and interpretability also informed the choice.
Kidney disease required persistent new abnormalities (urine albumin-to-creatinine ratio of at least 30 mg/g and/or eGFR below 60 for at least three months); neuropathy required abnormal nerve conduction on a point-of-care device with symptoms or signs after other causes were excluded. Cox models adjusted for age, sex, diabetes duration, fasting glucose, and baseline sodium–glucose cotransporter 2 inhibitor, glucagon-like peptide 1 receptor agonist, and insulin use. Socioeconomic status (SES) combined education, occupation, and household income; analyses included exploratory SES stratification and multiple imputation.
Median follow-up was 23.0 months for DKD and 21.9 months for DPN. The analyses included 110 incident DKD events (39 in the LR class, 13 in BVA, 58 in OMR) and 70 incident DPN events (22, 14, and 34, respectively).
The OMR class combined obesity, inactivity, insulin resistance, and unfavorable blood pressure; the LR class had a more favorable profile and served as the reference. OMR was associated with incident DKD versus LR (hazard ratio 2.006, 95% confidence interval [CI] 1.324–3.040). The BVA class was not significantly associated with DKD.
OMR was also associated with incident DPN versus LR (HR 2.056, 95% CI 1.182–3.576). BVA combined relatively low body mass index with greater visceral adiposity, unfavorable behaviors, dyslipidemia, and moderate insulin resistance. About 90% of this class were men, all were current or occasional smokers, and about 95% drank alcohol. It was associated with incident DPN versus LR (HR 2.467, 95% CI 1.189–5.121); because of its composition, the neuropathy association could partly reflect alcohol or smoking effects or sex differences. Higher SES was associated with more BVA and less OMR membership relative to LR. In exploratory SES strata, OMR associations with both complications were significant in the lower-SES group, whereas the BVA–DPN association was significant in the higher-SES group.
Additional adjustment for baseline glycated hemoglobin and blood pressure did not materially change the primary associations. Substituting a triglyceride-glucose measure produced less consistent class assignments (56.7% agreement with the original classification) when the original lipid-ratio input remained than when it was removed (96.6% agreement).
The retrospective, single-center Beijing design limits geographic generalization and leaves room for residual confounding; baseline kidney function and albuminuria were not included in the Cox models. Only about 22% of screened patients entered the analysis, and the cohort was relatively affluent. Limited outcome events, especially within SES strata, categorized continuous inputs, self-reported behaviors, and short follow-up further constrain interpretation. Significance in one SES stratum but not another does not establish effect modification; no formal interaction result was reported.
These classes are not validated clinical tests, the study did not test whether they improve prediction beyond simpler risk factors, and the associations do not show that changing a profile changes complication risk.
The authors concluded that multidimensional behavioral and metabolic profiles were linked to differing observed complication-association patterns, with SES differences requiring confirmation.
Clinician Questions
How was incident diabetic kidney disease defined among people with type 2 diabetes in the Beijing phenotype cohort?
Among participants with type 2 diabetes who had no DKD at baseline, newly diagnosed DKD required a urine albumin-to-creatinine ratio ≥30 mg/g and/or estimated glomerular filtration rate <60 mL/(min·1.73 m²), persisting for at least 3 months.
Why was diabetic retinopathy not assessed alongside kidney disease and neuropathy in the Beijing type 2 diabetes cohort?
Diabetic retinopathy could not be reliably evaluated because standardized fundus photography was not systematically available at Fangshan Hospital. The DKD and DPN associations do not establish retinopathy risk.
Recommended Reading
- For more on kidney and nerve complications in type 2 diabetes: Higher AIP Linked to Kidney and Nerve Complications in Type 2 Diabetes
This content is intended for healthcare professionals for educational purposes. The article and its image were developed with AI assistance and reviewed by members of the ReachMD team prior to publication. Please refer to the original source linked for the complete and most up-to-date information.