Renal function at the time of a myocardial infarction maintains prognostic value for more than 10 years

BACKGROUND: Renal function is an important predictor of mortality in patients with myocardial infarction (MI), but changes in the impact over time have not been well described. We examined the importance of renal function by estimated GFR (eGFR) and se-creatinine as an independent long-term prognostic factor. METHODS: Prospective follow-up of 6653 consecutive MI patients screened for entry in the Trandolapril Cardiac Evaluation (TRACE) study. The patients were analysed by Kaplan-Meier survival analysis, landmark analysis and Cox proportional hazard models. Outcome measure was all-cause mortality. RESULTS: An eGFR below 60 ml per minute per 1.73 m(2), consistent with chronic renal disease, was present in 42% of the patients. We divided the patients into 4 groups according to eGFR. Overall, Cox proportional-hazards models showed that eGFR was a significant prognostic factor in the two groups with the lowest eGFR, hazard ratio 1,72 (confidence interval (CI) 1,56-1,91) in the group with the lowest eGFR. Using the eGFR group with normal renal function as reference, we observed an incremental rise in hazard ratio. We divided the follow-up period in 2-year intervals. Landmark analysis showed that eGFR at the time of screening continued to show prognostic effect until 16 years of follow-up. By multivariable Cox regression analysis, the prognostic effect of eGFR persisted for 12 years and of se-creatinine for 10 years. When comparing the lowest group of eGFR with the group with normal eGFR, prognostic significance was present in the entire period of follow-up with a hazard ratio between 1,97 (CI 1,65-2,35) and 1,35 (CI 0,99-1,84) in the 2-year periods. CONCLUSIONS: One estimate of renal function is a strong and independent long-term prognostic factor for 10-12 years following a MI.