Impact of hyperglycemia on morbidity and mortality, length of hospitalization and rates of re-hospitalization in a general hospital setting in Brazil

Abstract

Background: Hyperglycemia in hospitalized patients is known to be related to a higher incidence of clinical and surgical complications and poorer outcomes. Adequate glycemic control and earlier diagnosis of type 2 diabetes during hospitalization are cost-effective measures.

Methods: This prospective cohort study was designed to determine the impact of hyperglycemia on morbidity and mortality in a general hospital setting during a 3-month period by reviewing patients' records. The primary purposes of this trial were to verify that hyperglycemia was diagnosed properly and sufficiently early and that it was managed during the hospital stay; we also aimed to evaluate the relationship between in-hospital hyperglycemia control and outcomes such as complications during the hospital stay, extent of hospitalization, frequency of re-hospitalization, death rates and number of days in the ICU (Intensive Care Unit) after admission. Statistical analyses utilized the Kruskall-Wallis complemented by the "a posteriori" d.m.s. test, Spearman correlation and Chi-squared test, with a level of significance of 5% (p < 0.05).

Results: We reviewed 779 patient records that fulfilled inclusion criteria. The patients were divided into 5 groups: group (1) diabetic with normal glycemic levels according to American Diabetes Association criteria for in-hospital patients (n = 123); group (2) diabetics with hyperglycemia (n = 76); group (3) non-diabetics with hyperglycemia (n = 225); group (4)diabetics and non-diabetics with persistent hyperglycemia during 3 consecutive days (n = 57) and group (5) those with normal glucose control (n = 298). Compared to patients in groups 1 and 5, patients in groups 2, 3 and 4 had significantly higher mortality rates (17.7% vs. 2.8%) and Intensive Care Unit admissions with complications (23.3% vs. 4.5%). Patients in group 4 had the longest hospitalizations (mean 15.5 days), and group 5 had the lowest re-hospitalization rate (mean of 1.28 hospitalizations). Only 184 (51.4%) hyperglycemic patients had received treatment. An insulin "sliding-scale" alone was the most frequent treatment used, and there was a wide variation in glucose target medical prescriptions. Intra Venous insulin infusion was used in 3.8% of patients in the ICU. Glycohemoglobin(A1C) was measured in 11 patients(2.2%).

Conclusions: Hospital hyperglycemia was correlated with, among other parameters, morbidity/mortality, length of hospitalization and number of re-hospitalizations. Most patients did not have their glycemic levels measured at the hospital; despite the high number of hyperglycemic patients not diagnosed as diabetics, A1C was not frequently measured. Even when patients are assessed for hyperglycemia, they were not treated properly.

Figure 1

Total number of patients in-hospital during 3 months and inclusion algorithm in the study (Glucose is expressed in mg/dL).