Low Serum C-Peptide and Albuminuria Severity in Type 2 Diabetes Mellitus: A Cross-Sectional Study

Background/Objectives: Lower serum C-peptide, reflecting reduced beta-cell reserve, has been linked to diabetic complications, but whether these associations are independent of diabetes duration and glycaemic control is unclear. The objective of this study was to determine which microvascular and macrovascular complications of type 2 diabetes mellitus (T2DM) remain associated with fasting C-peptide after adjustment for these factors. Methods: In this single-centre, cross-sectional study with consecutive prospective enrolment, non-pregnant adults with an established diagnosis of T2DM attending routine outpatient follow-up were eligible; those with pancreatic malignancy or exocrine pancreatic insufficiency, monogenic or autoimmune diabetes, or acute hyperglycaemic crises were excluded. A total of 590 patients were stratified by fasting C-peptide into insufficient (<1.0 ng/mL; n = 35), borderline (1.0-1.8 ng/mL; n = 212) and normal/high (≥1.8 ng/mL; n = 343) groups. Logistic regression adjusted for age, sex, diabetes duration, HbA1c and body mass index assessed each complication. Receiver operating characteristic analysis using DeLong's test assessed incremental discrimination. Results: In the unadjusted analyses, every complication was more frequent at lower C-peptide. After adjustment, C-peptide was not associated with the presence of albuminuria (insufficient versus normal/high, odds ratio 1.60, 95% CI 0.16-15.86) but was strongly associated with its severity: for severely increased or nephrotic-range albuminuria, the adjusted odds ratios were 13.47 (5.33-34.07) and 6.90 (3.83-12.44) in the insufficient and borderline groups (both p < 0.001). Adding C-peptide to a model of diabetes duration, HbA1c and fasting glucose was associated with a statistically significant, but modest, improvement in discrimination for severe albuminuria (area under the curve of 0.748 to 0.783; ΔAUC = 0.035, p = 0.027). Associations with macrovascular complications were present in the unadjusted analyses but were substantially attenuated after adjustment and no longer statistically significant, except for peripheral arterial disease and foot ulcer, which remained associated with insufficient C-peptide in models based on small numbers of events and should be regarded as exploratory. Conclusions: Low C-peptide was independently associated with the severity of albuminuria rather than its presence and was associated with a modest incremental improvement in discrimination. These exploratory findings suggest that C-peptide, an inexpensive and widely available measurement, may provide information associated with advanced diabetic kidney disease; prospective, externally validated studies are needed before clinical application.