Random or fasting glucose measurements can be used to screen for diabetes in patients admitted for acute myocardial infarction, a study found.
The prospective study included 1,574 patients at 24 hospitals who were not taking any glucose-lowering medications when they were admitted for acute myocardial infarction (MI). All of the patients had a random plasma glucose (RPG) taken at admission and at least 2 fasting plasma glucose (FPG) levels measured during hospitalization. All also received an HbA1c test (with a result ≥6.5% considered a diabetes diagnosis) to evaluate the accuracy of RPG and FPG as screening tools. Results were published online Nov. 4 by BMJ Open Diabetes Research & Care.
The study found that a RPG >140 mg/dL (7.8 mmol/L) or an FPG ≥126 mg/dL (7.0 mmol/L) was highly sensitive for diabetes diagnosis. Using those glucose levels as cutoffs for HbA1c testing would have identified 86% of the patients with diabetes and necessitated only half of the overall MI group getting their HbA1c tested. With those cutoffs, the number needed to screen (NNS) would be 3.3 to identify 1 case of diabetes versus 5.6 with universal HbA1c testing of MI patients. Raising the cutoff to RPG >180 identified 82% of diabetics, required 40% of MI patients to get an HbA1c, and had an NNS of 2.7.
These selective screening methods could identify the vast majority of incident diabetes in MI patients with less-than-universal HbA1c testing, the study authors concluded. Targeting screening would be more cost-effective in practice, and these cutoffs could also be useful for research in which HbA1c results are unavailable, they noted. The patients whose diabetes would be missed by this screening method were the ones with relatively mild elevations in HbA1c, perhaps lessening the consequences of the missed diagnoses, the researchers noted, although they also acknowledged that this screening system would not identify patients with pre-diabetes for intervention.
Screening for diabetes with RPG or FPG could be particularly useful in resource-limited settings, to accomplish the important goal of identifying diabetes during hospitalization for acute MI, and targeting diabetic patients for educational, lifestyle and medical interventions, the study authors concluded.