The Relationship Between Smoking Status and C-Peptide Levels in Patients with Type 2 Diabetes Mellitus

Background/Objectives: This study aimed to investigate the relationship between smoking status and C-Peptide levels in patients with Type 2 Diabetes Mellitus (T2DM). Methods: The study included patients over the age of 18 with type 2 diabetes mellitus, without any malignancy, who presented to our hospital’s Internal Medicine Outpatient Clinic between January 2022 and April 2023. A total of 204 patients were recruited and examined in three groups: nonsmokers (n = 142), current smokers (n = 37), and former smokers (n = 25). Results: In the nonsmoker group, fasting c-peptide was significantly correlated with neck circumference (r = 0.202), waist circumference (r = 0.311), hip circumference (r = 0.270), insulin (r = −0.358), HbA1c (r = −0.235) and High Dense lipoprotein (r = −0.192). In the smoker group, fasting c-peptide was significantly correlated with diabetes age (r = −0.466), and sodium glucose cotransporter drugs (SGLT-2 inhibitor) usage (r = −0.335). In former smokers, fasting c-peptide was significantly correlated with SBP (r = −0.441), insulin (r = −0.578) and fasting blood sugar (r = 0.520; p < 0.01). In former smokers, fasting c-peptide (r = 0.522) and postprandial c-peptide (r = 0.505; p < 0.01) were significantly and positively correlated with smoking years. In former smokers, fasting c-peptide (r = −0.411) and postprandial c-peptide (r = −0.552) were significantly and negatively correlated with years since quitting. Multivariate analysis results showed that associations of fasting C-peptide levels with tease dummy (B = −1.176; p < 0.05) and insulin dummy (B = 0.052; p < 0.05) were statistically significant. Postprandial C-peptide was significantly associated with age (B = 0.077; p < 0.01), DM age (B = −0.074; p < 0.05), BMI (B = 0.177; p < 0.01) and insulin dummy (B = −2.910; p < 0.01). Conclusions: In former smokers, smoking years were positively correlated with both fasting and postprandial C-peptide levels in univariate analysis; however, this relationship was attenuated and became non-significant when evaluated within the multivariable regression model.