Diabetes mellitus is a global health concern, and diabetic nephropathy (DN) is a serious complication associated with increased morbidity and mortality. Early identification of at-risk individuals may enable timely intervention. This study aimed to evaluate whether mean platelet volume (MPV) serves as a marker for DN and to identify associated risk factors. In this retrospective cohort study using routinely collected clinical data, patients with type 2 diabetes aged ≥ 20 years who attended the Primary Care Unit of Songklanagarind Hospital between 2013 and 2017 were followed until 2023. Medical records were reviewed for anthropometric, clinical, and laboratory data, including MPV and urine albumin-to-creatinine ratio (UACR). DN was defined as UACR > 30 mg/g on ≥ 2 of 3 measurements within 6 months or estimated glomerular filtration rate 3 months. Of 681 records screened, 450 patients were included after excluding those with prior DN, incomplete key data, or insufficient follow-up assessments, which may introduce selection bias. Cox regression analysis was used to identify factors associated with DN. Among 450 participants (mean age 61.3 ± 10.4 years; 58.2% female), unadjusted baseline comparisons showed that those who developed DN had higher baseline weight and serum creatinine. Over 2,373 person-years (median follow-up 5.21 years), 36 incident cases occurred (cumulative incidence 8%; incidence rate 15.1 per 1,000 person-years). In the multivariate time-dependent Cox model, male sex was significantly associated with an increased risk of DN (HR, 2.39; 95% CI, 1.21–4.72). MPV showed a strong association with DN (HR, 25.81; 95% CI, 1.54–433.79), with evidence of attenuation over time (interaction HR, 0.46; 95% CI, 0.21–1.01). Age and hypertension were not significantly associated with the outcome. Male sex and higher MPV were independently associated with incident DN in this cohort, with MPV demonstrating a time-dependent effect characterized by attenuation over time. These findings suggest that higher MPV may be associated with early risk; however, its predictive performance was modest, and its clinical utility for risk stratification remains uncertain.
Pipatsrisawat et al. (2026) studied this question.