Aim To investigate whether admission and fasting glucose predict abnormal glucose tolerance in acute myocardial infarction (MI) patients and the role of oral glucose tolerance testing (OGTT)in predicting mortality.
Methods We performed an OGTT on day 3 of admission, on patients presenting with ST or non-ST elevation MI to our coronary care unit over a period of 3 years. Patients with diabetes were excluded. Data were also collected on admission glucose and mortality at 3 years.
Results OGTT was performed on 478 patients. Admission glucose was available for 362 patients. 50 patients (10.9%) died during the study period and the average time to death was 259.6 days (range 2–899 days). The mean (± SD) age was 66 years (±12.6) and 72% were men. Average follow up was 1.7 years (±0.8). There was no significant difference in the diagnosis of impaired glucose tolerance (IGT) or diabetes mellitus (DM) between men and women (37.1 vs 34.6%, p = 0.61 and 20% vs 18%, p = 0.63, respectively). 62 (63.3%) patients diagnosed with DM had a normal fasting glucose (±6.0 mmol/l). 50 (10.9%) patients died during the course of the study; the average time to death was 259.6 days (range 2–899). Of those who died, significantly more had IGT or diabetes (78% vs 53.3%, p = 0.004); however, there was no difference in the fasting glucose (5.3 ± 0.9 mmol/l vs 5.2 ± 0.8 mmol/l, p = 0.16) but the 2-h glucose was significantly higher (10.1 ± 3.2 mmol/l vs 8.6 ± 3.2 mmol/l, p = 0.0009). The admission glucose was not predictive of IGT/DM; of those with an admission glucose ⩾6.1 mmol/l, 21.8% had DM, 37.8% IGT and 40.4% normal glucose tolerance. 61.5% of patients with an admission glucose greater than 11.1 mmol/l did not have DM on OGTT. A raised admission glucose (⩾6.1 mmol/l) did not predict death when compared with normal admission glucose (8.5% vs 11.1%, p = 0.51) in contrast to the OGTT, which identified that patients with IGT or DM were at an increased risk of death compared witho patients with normal glucose tolerance (14.2% vs 5.2%, p = 0.001).
Conclusion There was a high incidence of insulin resistance in acute MI patients. Our data suggest that patients diagnosed with diabetes on OGTT had a significantly higher admission glucose. However, many hyperglycaemic MI patients did not have diabetes, and other factors must therefore contribute to the raised glucose levels on admission. Fasting glucose is a poor predictor of diabetes and outcome. All patients should undergo OGTT for an accurate assessment of IGT and risk stratification.