Impact of pre-diabetes, well-controlled diabetes, and poorly controlled diabetes on anastomotic leakage after esophagectomy for esophageal cancer: a two-center retrospective cohort study of 1901 patients.

Abstract

Diabetes is known to be associated with anastomotic leakage (AL) after esophagectomy. However, it is unknown whether well-controlled diabetes is also associated with AL. We conducted a two-center retrospective cohort database study of patients who underwent oncological esophagectomy (2011-2019). Patients were divided into four groups: normoglycemia, pre-diabetes, well-controlled diabetes (hemoglobin A1c [HbA1c] < 7.0%), and poorly controlled diabetes (HbA1c ≥ 7.0%). The occurrence of AL and length of stay were compared between groups using multivariable analyses. The relationship between categorical HbA1c levels and AL was also investigated in patients stratified by diabetes medication before admission. Among 1901 patients, 1114 (58.6%) had normoglycemia, 480 (25.2%) had pre-diabetes, 180 (9.5%) had well-controlled diabetes, and 127 (6.7%) had poorly controlled diabetes. AL occurred in 279 (14.7%) patients. Compared with normoglycemia, AL was significantly associated with both well-controlled diabetes (odds ratio 1.83, 95% confidence interval [CI] 1.22-2.74) and poorly controlled diabetes (odds ratio 1.95, 95% CI 1.23-3.09), but not with pre-diabetes. Preoperative HbA1c levels showed a J-shaped association with AL in patients without diabetes medication, but no association in patients with diabetes medication. Compared with normoglycemia, only poorly controlled diabetes was significantly associated with longer hospital stay after surgery, especially in patients with operative morbidity (unstandardized coefficient 14.9days, 95% CI 5.6-24.1). Diabetes was associated with AL after esophagectomy even in well-controlled patients, but pre-diabetes was not associated with AL. Operative morbidity, including AL, in poorly controlled diabetes resulted in prolonged hospital stays compared with normoglycemia.