Evaluation of a simple management protocol for hyperglycaemic crises using intramuscular insulin in a resource-limited setting

Background: Management of hyperglycaemic crises requires expensive and labour-intensive procedures that are not achievable in all clinical settings. Intramuscular (IM) insulin therapy is a more feasible alternative, but remains insufficiently evaluated. We report here on an audit of clinical outcomes of a simple management protocol that involves IM insulin therapy, careful rehydration and inexpensive monitoring in a resource-limited setting. Methods: In June 2006, we began the routine use of a protocol based on IM insulin administration, careful rehydration and affordable monitoring for the management of hyperglycaemic crises in Yaoundé Central Hospital. Clinical records of patients admitted for hyperglycaemic crises 6 months before and 6 months after introduction of the protocol were independently coded and compared for clinical outcomes, including the 48-hour death rate as a primary endpoint. Secondary endpoints were blood glucose (BG) normalization and duration of hospital stay. Results: A total of 112 patients' files fulfilled the inclusion criteria, including 57 before and 55 after the introduction of the IM protocol (intervention). Patients of the pre-intervention group were aged 56.4 ± 2.1 years versus 53.9 ± 2.3 years in the intervention group (p = 0.41), with 23% versus 40%, respectively, with newly diagnosed diabetes (p = 0.05), and 45% versus 41%, respectively, with significant ketosis on admission (p = 0.84). As for the primary endpoint, 15.8% of the pre-intervention group died within 48 hours of admission versus 3.6% in the intervention group (p = 0.03). BG was normalized within 24 hours of admission in 28.1% patients of the pre-intervention group versus 90.9% of the intervention group (p < 0.001). However, the overall duration of hospitalization was similar in both groups. Septic shock, ketosis and high serum creatinine on admission were associated with poor outcomes in both groups. Conclusion: The proposed protocol using IM insulin can be safely used to treat hyperglycaemic crises, with mortality rates comparable to those in specialized centres in developed countries.