Systolic Dysfunction Portends Increased Mortality among Those Waiting for Renal Transplant

Individuals waiting for a renal transplant experience excessive cardiovascular mortality, which is not fully explained by the prevalence of ischemic heart disease in this population. Overt heart failure is known to increase the mortality of patients with ESRD, but the impact of lesser degrees of ventricular systolic dysfunction is unknown. For examination of the association between left ventricular ejection fraction (LVEF) and mortality of renal transplant candidates, the records of 2718 patients evaluated for transplantation at one institution were reviewed. During 6355 patient-years (median 27 mo) of follow-up, 681 deaths occurred. Patients with systolic dysfunction (LVEF ≤0.40) had significantly lower survival than those with higher systolic function (median 49 ± 3.1 versus 72 ± 4.0 mo; P 60% (n = 656). The same pattern was seen with cardiac mortality: Crude HR 6.4 (4.15 to 10.01) for LVEF ≤30%, 3.5 (2.26 to 5.33) for LVEF 31 to 40%, 1.8 (1.16 to 2.67) for LVEF 41 to 50%, and 0.8 (0.51 to 1.37) for LVEF >60%. After accounting for the presence of ischemia and all other risk factors, the adjusted HR for SD was 1.7 (95% CI 1.43 to 2.07), an independent association from diabetes, LVH, and abnormal perfusion (Table 2). A stepwise relationship is suggested by the presence of a graded-reverse association between LVEF and mortality (Figure 3). This finding was further confirmed by entering LVEF into the models as a continuous variable (adjusted HR 0.975; 95% CI 0.968 to 0.981; P 140 mmHg, a diastolic BP >90 mmHg, or the use of antihypertensive drugs. LVH was diagnosed by 12-lead electrocardiogram24 or 2DE.25 BMI was calculated by weight (kg)/square of height (m). Duration of dialysis, age, gender, race, diabetes, serum albumin and creatinine levels, and tobacco smoking history were extracted from the database. The study protocol was reviewed and approved by the institutional review board for human use at UAB. Statistical Analyses Distribution of patients' characteristics at evaluation was tested by χ2 analysis or t test as appropriate. Results were described as means ± SD. Event-free survival curves were constructed using the product-limit method (Kaplan-Meier), and differences among survival curves were estimated by the log-rank test. Cox regression modeling was used to estimate crude (univariate) and adjusted (multivariate) risks. Variables entered in the models were the ones described as being associated with cardiovascular mortality: Age, gender, SES, obesity, race, presence of diabetes, hypertension, LVH, preexisting ischemia, anemia, low albumin level, tobacco smoking, and duration of dialysis. Direct visualization of Log −log (survival time) versus log (survival time) plots and examination of Martingale residuals26 were used to validate the proportionality of hazard increments assumption. LVEF by SPECT, age, BMI, and duration of dialysis were tested in the models both as a continuous and as categorical variables (LVEF ≤30, 31 to 40, 41 to 50, 51 to 60, or >60%; age 35 kg/m2; dialysis >2, 1 to 2, or 40%, no previous CVE, and normal perfusion pattern). We then performed multivariable analysis in which the attributed risk for SD was adjusted by the presence of ischemia and other cardiovascular risk factors. In addition, models including the interaction term “ischemia × LVEF” were built and their HR compared with main-effect models. Subsequent analyses were conducted using cardiac death as the outcome of interest. Internal consistency was assessed by building several models across strata of risk factor categories: Dialysis and nondialysis, male and female, older and younger than 50 yr, black and nonblack, with and without ischemia. Estimated risks were reported as HR with corresponding 95% CI. All P values reported were two-sided. SPSS 11.5 for Windows (SPSS, Chicago, IL) software was used in the analyses. DISCLOSURES None.Figure 1: (A) Overall survival after evaluation according to categories of LVEF. (B) Overall survival according to presence of ischemia or systolic dysfunction (Systolic DF: left ventricular ejection fraction ≤40%) at the time of transplant evaluation.Figure 2: Cardiovascular death survival after evaluation, according to categories of LVEF. All other causes of death were considered censoring HR for all-cause mortality, according to categories of LVEF. were for age, gender, SES, obesity, race, presence of diabetes, hypertension, LVH, preexisting ischemia, anemia, low albumin level, tobacco smoking, and duration of dialysis. < 0.001; = = 1: Demographic data according to left ventricular systolic 2: Mortality after the work of the UAB Renal Transplant Database and transplant