[THE IMPACT OF ADMISSION TO THE MEDICAL WARD ON HEMOGLOBIN A1C LEVELS AMONG PATIENTS WITH TYPE-2 DIABETES].

Introduction: Blood glucose monitoring is a routine medical practice in the medical ward regardless of the cause of hospitalization of patients with diabetes. In this study we prospectively evaluated the impact of hospitalization on hemoglobin A1c (HbA1c) among patients with type-2 diabetes admitted to the medical ward, for reasons unrelated directly to their diabetes.

Methods: Patients with type-2 diabetes who were admitted to the medical ward for reasons not directly related to diabetes were asked to participate in our study. After consent , demographic, clinical and laboratory parameters were documented on admission, including age, sex, social status, years of study, duration of diabetes, type of anti-diabetic treatment, background medical problems and treatment, diagnosis on admission, temperature, blood glucose levels on admission to the medical ward, complete blood count and serum creatinine. In addition, HbA1c levels on admission, first 24-hour blood glucose levels at the medical ward, change of therapeutic treatment of their diabetes, diagnosis at discharge and recommended medication at discharge, were also documented. Three months following admission, the HbA1c study was repeated again. Paired t-test and Wilcoxon's rank sign test were used to compare between HbA1c levels at baseline and 3 months later among all the patients and among the subgroups. Multivariate regression analysis was used also to predict change in HbA1c levels among these patients. Physicians taking care of the patients on the medical ward were not aware of the study.

Results: Forty-four patients were recruited and 40 patients completed the study. There were 24 men (60%), with a mean age of 69.625±11.53 years for all the patients. The most common diagnosis on admission was infection followed by congestive heart failure. Mean 24-hour blood glucose level following admission was 209 mg% ±95. Among 12 patients there was an augmentation in diabetes treatment and in 3 there was dis-augmentation/discontinuation. Mean hemoglobin A1c levels on admission and 3 months later was 7.29% and 6.80% respectively (p=0.095). However, sub-analysis of patients with HbA1c levels >7.5% on admission, showed a significant decrease in the repeated HbA1c levels (p=0.003).

Conclusions: Medical ward admission was associated with reduced HbA1C levels, yet not significant, among patients with type-2 diabetes who were admitted for reasons not directly related to diabetes.