Study on the mechanism causing elevation of serum CA19-9 levels in diabetic patients.

Serum CA19-9 levels are often elevated in diabetic patients. To elucidate this mechanism, we investigated the metabolism of CA19-9 in diabetic patients without obvious cancer. Study 1 included 119 patients in whom HbA1c, glycated albumin (GA) and CA19-9 were measured at the time of hospital admission. Study 2 examined 6 patients with markedly elevated CA19-9 levels (≥100 U/mL). Their half-lives for HbA1c, GA, and serum CA19-9 were calculated using the data before and after diabetes treatment. Three diabetic patients with pancreatic cancer were also examined as controls. In Study 1, serum CA19-9 (logarithmically transformed value) was significantly correlated with fasting plasma glucose (FPG), HbA1c and GA. On multivariate analysis, GA and FPG, but not HbA1c, were significant explanatory variables for serum CA19-9. In Study 2, serum CA19-9 decreased together with HbA1c and GA after diabetes treatment. The calculated half-lives for HbA1c, GA, and serum CA19-9 were 33.8 days, 16.1 days, and 10.9 days, respectively. The half-life of serum CA19-9 was longer in the study patients than that reported in patients with malignant tumors. By contrast, in the diabetic patients with pancreatic cancer serum CA19-9 showed a marginal decrease after diabetes treatment. Taken all together, prolonged half-life of serum CA19-9 may contribute to the increase in serum CA19-9 levels in diabetic patients without obvious cancer.

Usefulness of glycated albumin for early detection of deterioration of glycemic control state after discharge from educational admission.

Glycated albumin (GA) reflects shorter term glycemic control state than HbA1c. This study evaluated whether GA is useful for early detection of deterioration of glycemic control state after discharge from educational admission. Among the patients with educational admission, this study included 21 diabetic patients who were followed for at least 10 weeks after discharge from educational admission. Deterioration was defined that GA after discharge increased by ≥0.6% compared to the previous GA. Thirteen patients without deterioration up to 10 weeks after discharge were used as controls. In 8 patients with deterioration within 10 weeks after discharge, their HbA1c and GA at the time of admission and the decrease in HbA1c and GA during hospitalization were not significantly different from the control patients. At the time of deterioration, GA in the patients with deterioration increased significantly from 18.7 ± 2.7% to 21.0 ± 2.5%, whereas HbA1c decreased significantly from 9.1 ± 0.7% to 8.3 ± 0.6%. Subsequently, HbA1c increased significantly to 9.0 ± 0.8% together with GA. Thus, GA is useful for early detection of deterioration of glycemic control state after discharge from educational admission.