Article Summary Management of Diabetes and Hyperglycemia in Hospitals. Clement SC, Braithwaite SS, Magee MF, Ahmann A, Smith EP, Schafer RG, Hirsh IB on behalf of the Diabetes in Hospitals Writing Committee. Diabetes Care . 2004;27:553-591. Objective: To evaluate the literature associated with the management of diabetes and hyperglycemia in the hospital setting, with a focus on their effects on outcomes. Methods: To provide data regarding treatment of various acute conditions, a number of questions were examined and recommendations provided: 1) What is the link between high blood glucose (BG) and poor outcomes? 2) What are the target BG levels for the hospitalized patient? 3) How are target BG levels best achieved in the hospital? 4) How can system design and implementation improve diabetes care in the hospital? 5) What is the role of the Diabetes Self-Management Education for the hospitalized patient? 6) What is the role of the medical nutrition therapy in the hospitalized patient with diabetes? 7) What is the role of bedside glucose monitoring in the hospitalized patient? 8) Is improved diabetes care in hospitals cost effective? In this review article, studies that address the association of BG levels with outcomes are grouped into the medical and surgical areas of general medicine and surgery, cardiovascular disease, critical care, and neurologic disorders. Results: Studies of patients admitted for general medicine and surgery showed mortality in normoglycemic patients at 1.7% compared with 3% for patients with known diabetes and 16% in patients with new hyperglycemia (n=1886). BG over 220 mg/dL was a sensitive predictor (85%) resulting in a 5.9-fold increase for serious infections, eg, sepsis, pneumonia, and wound (n=97). A meta-analysis of 14 review articles regarding cardiovascular disease and critical care for acute myocardial infection (AMI) showed a 3.9-fold increase in relative risk for mortality in subjects without diabetes and BG at or above 109.8 mg/dL. The risk for coronary heart failure and cardiogenic shock also increased. Intensive insulin therapy followed by multi-injection treatment in patients with AMI resulted in 29% reduction in mortality at 1 year, with benefits extending to 3.4 years (n=620). Observational studies of patients hospitalized for cardiac surgery found that initiating insulin infusion reduced mortality and incidence of deep-sternum wound infections nondiabetic levels. In prospective, randomized, controlled trials of patients admitted to surgical critical care units and on mechanical ventilation, insulin infusion therapy to target BG levels of 80 mg/dL to 110 mg/dL reduced mortality by 40%, compared with target levels of 180 mg/dL to 200 mg/dL in conventional therapy (n=1548). In a review of several other prospective studies, similar findings were noted of improved outcomes associated with reduction in BG in patients admitted with acute stroke. Conclusion: This Technical Review cites numerous examples of improved outcomes when hyperglycemia is measured and treated, regardless of the presenting condition. Clinical trials are strongly urged to definitively answer the questions posed in this position statement.