Organised By
Organised By
Hyperglycaemia is commonly observed during acute and critical illness. Recent studies have investigated the risk of developing diabetes after acute and critical illnesses, but the relationship between degrees of in-hospital hyperglycaemia and new onset diabetes has not been investigated. This study examines the evidence for the relationship between in-hospital hyperglycaemia and prevalence of new onset diabetes following acute and critical illness. A literature search was performed of MEDLINE, EMBASE, and Scopus databases for relevant studies published from January 1, 2000, through August 4, 2016. Patients with no history of diabetes prior to hospital discharge were included in the systematic review. In-hospital glucose concentration was classified as normoglycaemia, mild hyperglycaemia, or severe hyperglycaemia for the meta-analysis. Twenty-three studies were included in the systematic review, and 18 of these (111,078 patients) met eligibility criteria for meta-analysis. The prevalence of new onset diabetes was significantly related to in-hospital glucose concentration and was 4% (95% CI 2% to 7%), 12% (95% CI 9% to 15%), and 28% (95% CI 18% to 39%) for normoglycaemia, mild hyperglycaemia, and severe hyperglycaemia patients, respectively. The prevalence of new onset diabetes was not influenced by acute disease setting, follow-up duration, or study design. In summary, this study found a stepwise growth in the prevalence of new onset diabetes with increasing in-hospital glucose concentration. Patients with severe hyperglycaemia are at highest risk, with 28% developing diabetes after hospital discharge.