USING THE FRAMINGHAM DATABASE TO PREDICT CORONARY HEART DISEASE (CHD) EVENTS IN RENAL TRANSPLANT PATIENTS

Abstract

853 Cardiovascular disease is the single most significant cause of death in end-stage renal disease (ESRD). The utilization of cardiac risk factors in the general population as predictors for coronary heart disease (CHD) events in a patient undergoing renal transplantation has not been described. Andersen et al. developed a Framingham heart study equation to predict the probability of CHD events using the traditional cardiac risk factors (age, gender, systolic BP (SBP), total cholesterol (Chol), HDL, tobacco use, diabetes [DM], and LVH). Our goal was to examine the applicability of the Framingham CHD risk profile in ESRD patients at the time of renal transplantation. 249 patients (pts) from a single-center renal transplant (Tx) program were analyzed. All pts underwent renal Tx in the calendar year 1995. 214 (86%) underwent 1° renal Tx (49% cadaveric, 26% living-donor, and 22% kidney-pancreas). Average f/u was 3.4 years. The population, with a mean age of 40.7±13 yrs, was 60% male. Average SBP was 139±18.3 mmHg, total Chol 229±55.3 mg/dl, and HDL was 58.4±22.5 mg/dl. 23% of the pts were smokers, 21% had LVH on baseline electrocardiogram and 41% of pts had DM. 10% of the cohort had pre-Tx CHD. 29 pts (11.6%) experienced a post-Tx CHD event defined as angina, CHF, acute MI, cardiac arrest, or sudden death. This was significantly greater than the expected CHD event rate computed for the general population without preexisting heart disease (7.02/249;2.7%), reflecting a 3.8-fold increase in CHD risk in our Tx pt population (p<0.001). The risk of a CHD event was highest in the early post-Tx period. Univariate analyses showed that age (p=0.01), creatinine at discharge (p=0.04), and pretransplant CHD (p=0.004) were associated with post-Tx CHD events. Multivariate analysis demonstrated that pre-Tx CHD (p=0.04) was the only variable predictive of a post-Tx CHD event. Other variables including gender, SBP, diabetes, smoking, LVH, Chol, or HDL were not independently predictive of post-Tx CHD events. Despite underestimating the incidence, the Framingham equation predicted 4-yr risk was highly associated with the hazard of CHD in Tx pts (p=0.0028, Cox prop. hazards). This analysis confirms the increased incidence of CHD events in renal Tx patients. The Framingham risk model appears to be a useful tool in identifying Tx pts at risk for CHD events. For assessing short-term (<5 years) risk however, the calculated result should be multiplied by a factor of 3.8 to obtain a more accurate prediction of a post-Tx CHD event.