Screening for Coronary Artery Disease Before Renal Transplantation—Rational or Rationing?

Premature cardiovascular disease (CVD) is the leading cause of death in patients with end-stage renal disease (ESRD), receiving renal replacement therapy and the major cause of early deaths after renal transplantation. As a result of these observations, it has become commonplace to screen potential renal transplant recipients for coronary artery disease (CAD), a strategy that is supported by many published guidelines. This strategy is reliant on the assumptions that (a) CAD is the cause of premature CVD and (b) identification and correction of CAD will improve the outcomes of patients awaiting transplantation regardless of whether they ultimately receive a transplant. It seems like a good idea, but is it? Until recently, we have had little information on the impact of cardiovascular (CV) screening: specifically, the pickup rate of significant CAD and the intervention rate. The article by De Lima et al. published in this issue of Transplantation describes a registry analysis (with the inherent difficulties in interpretation) of CV screening in San Paulo, Brazil—one of the largest and most active transplant centers in the world. One thousand twenty-five consecutive patients underwent CV screening before acceptance on the transplant waiting list (1). Using myocardial scintigraphy combined with clinical findings to identify patients likely to have CAD, 519 (50.6%) patients underwent coronary angiography. Although 230 (22.4%) had significant coronary stenoses, a majority were managed medically with only 30 (2.9%) patients undergoing coronary revascularization. In this small group of patients, no survival benefit was seen with coronary intervention compared with medical therapy. A group of 16 patients who were offered intervention but refused did exceptionally badly, but the reasons for refusal—that are likely to include comorbidity—are not described. What do these results tell us? First, although the selection criteria for screening are not clearly described, CV screening will identify a large proportion of patients with abnormal noninvasive tests and subsequent coronary angiography will identify a high prevalence of structural CAD. However, intervention is seldom performed, and the survival of patients with CAD, with and without intervention, is similar. The low intervention rate confirms the findings of others—Kasiske et al. (2) and our own group among them (3)—and probably reflects the reluctance of cardiologists and cardiac surgeons to intervene in asymptomatic patients in the light of the COURAGE trial (4), the failure of CV screening to improve the outcomes of high-risk patients undergoing major noncardiac surgery (5), and the high rate of complications in patients with ESRD after interventions. Second, these data show that the CV investigations described help identify prognostic groups. Thus, patients with negative myocardial scintigraphy fare better than those with positive tests; patients with structural CAD fare worse than those without CAD—although this is independent of conventional CV risk factors, it is not investigated by multivariate analysis. De Lima et al. imply that the identification of structural CAD may allow maximization of medical therapy and that this may improve survival. However, there is scant evidence to support the use of any cardioprotective therapy in patients on hemodialysis therapy and considerable information to suggest that it is not CAD, but uremic cardiomyopathy that contributes to the excess CV mortality in ESRD (6). Thus, one should not read too much into these observational registry data. Of more concern is the impact of CV screening on the delay between consideration of potential transplant recipients and their ultimate acceptance onto the waiting list, a delay that must ultimately delay transplantation. The overall mortality of patients awaiting transplantation is, at best, three times that of patients with functioning transplants and the absolute mortality rate of patients who undergo cardiac screening is likely to be high. If we accept that the relative survival benefits of transplantation are likely to be the same across all groups of potential transplant recipients, delay in wait listing will necessarily be associated with a higher mortality rate for all patients during the screening period and selection of fitter patients who are more likely to survive the screening period. If the screening period is short and the benefits of intervention are great, then this may be acceptable. However, if screening results in a long delay and the intervention is uncommon or relatively ineffective, then the strategy may be unacceptable. Although De Lima et al. do not discuss the duration of screening, in the United Kingdom this can lead to protracted delays in transplant listing, and with low intervention rates and comparable outcomes in those with and without intervention, it is difficult to see how a strategy of screening for CAD will improve overall survival, although survival bias may result in apparent benefit. When the cost of CV screening is added, and the cost of rescreening for those who remain on the waiting list for a prolonged period, we are faced with an expensive practice with little evidence of benefit to individuals or to the ESRD population as a whole. It is difficult to imagine that we would pursue such a strategy if the supply of organs for transplantation was unlimited. With the information we currently have, the benefits (or otherwise) of screening needs to be addressed in a clinical trial that randomizes patients without symptoms of ischemic heart disease to CV screening, or not, and reports survival and subsequent CV event rates. Such a trial would also contribute to a broader understanding of the pathophysiology of CVD in the peritransplant period and the extent to which CAD and other factors, such as uremic cardiomyopathy contributes to the poor CV outcomes that are dominated by sudden death and death due to heart failure rather than myocardial infarction. At least one trial is planned in this area, led by Kasiske et al. in the United States. In the meantime, the study by De Lima et al. adds to accumulating evidence that negative noninvasive CV investigations identify a low-risk population of potential transplant recipients, but that the intervention rate after pretransplant CV screening is low, and any benefits are hard to find.