By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Due to essential maintenance the subscribe/renew pages will be unavailable on Wednesday 26 October between 02:00- 08:00 BST/ 09:00 – 15:00 SGT/ 21:00- 03:00 EDT. Apologies for the inconvenience.
Death with function causes half of late kidney transplant failures, and cardiovascular disease (CVD) is the most common cause of death in these patients. We examined the use of potentially cardioprotective medications in a prospective observational study at seven transplant centers in the United States and Canada. Among 935 patients, 87% received antihypertensive medications at both 1 and 6 months after transplantation. Similar antihypertensive regimens were used for patients with and without diabetes and CVD, but with wide variability among centers. In contrast, while 44% of patients were on angiotensin converting enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARB) at the time of transplantation, the proportion taking these agents dropped to 12% at month 1, then increased to 24% at 6 months. Fewer than 30% with CVD or diabetes received ACEI/ARB therapy 6 months posttransplant. Aspirin use was uncommon (<40% of patients). Even among those with diabetes and/or CVD, fewer than 60% received aspirin and only half received a statin at 1 and 6 months. This study demonstrates marked variability in the use of cardioprotective medications in kidney transplant recipients, a finding that may reflect, among several possible explanations, clinical uncertainty due the lack of randomized trials for these medications in this population.
Given the dramatic progress in short-term graft survival after kidney transplantation, attention has increasingly turned to improving long-term outcomes. While over 90% of recipients now keep their kidneys at least 1 year, the annual rate of allograft failure beyond the first year has remained largely unchanged (1). Most intermediate and long-term allograft losses are attributed to either ‘chronic allograft nephropathy’ or death with a functioning allograft (2). The former remains an ill-defined entity of uncertain pathogenesis, while the latter is seemingly more important as transplant recipients become older and remain longer on dialysis awaiting transplantation.
Death with function accounts for approximately half of all allograft losses and often cuts short the projected longevity of a transplanted kidney (3,4). Although transplant patients are also at increased risk of dying from infection and malignancy, the most common cause of death is cardiovascular disease (CVD) (2). Studies in the general population have shown that aspirin, beta-blockers, angiotensin converting enzyme inhibitors (ACEI), angiotensin receptor blockers (ARB) and statins can each reduce the risk of major CVD events by as much as 25%, with additional risk reduction from combination therapies (5). However, there are few randomized controlled trials examining the safety and efficacy of strategies to reduce CVD in kidney transplant recipients, and it is unclear whether clinicians are prepared to apply guidelines from the general population to kidney transplant recipients.
The Long-Term Deterioration of Kidney Allograft Function (DeKAF) study is an ongoing, prospective, multicenter observational project designed to better characterize the causes of late kidney allograft failure. The protocol calls for the tabulation of medications utilized after kidney transplantation, including those that might impact CVD. As accumulating data indicated wide variability among participating centers in the use of medications shown to be cardioprotective in other populations, and aware of the potential clinical impact of out findings, we examined usage of these agents in detail.
The DeKAF study consists of two distinct cohorts of kidney transplant recipients enrolled at seven transplant centers in the United States and Canada: (1) Prospective Cohort: all consecutive patients transplanted after October 1, 2005 and (2) Cross-Sectional Cohort: previously transplanted patients with new-onset graft dysfunction. Over 80% of those eligible to participate are enrolled in the trial, with a goal of enrolling 7000 patients (5000 in the Prospective Cohort and 2000 in the Cross-Sectional Cohort). A detailed description of the study can be found at http://www.clinicaltrials.gov (NCT00270712). Institutional Review Board approval was obtained at all participating sites.
In this study, we examined medication use at 1 and 6 months after transplantation in the Prospective Cohort. Medication history collected for all patients at the time of transplant, and 1 and 6 months after transplant included: antihypertensive, anticoagulant, antiplatelet, lipid-lowering and hypoglycemic agents. Measured serum lipid levels were not necessarily obtained as a fasting specimen. Logistic regression was used to assess whether preexisting CVD or diabetes was associated with a higher likelihood of prescribing antihypertensive agents, specifically beta-blockers and ACEI/ARB, as well as aspirin and statins. Each model was stratified by clinical center. Odds ratios (OR) are given with 95% confidence intervals (CI). Significance was defined a priori as p < 0.05. All p-values are two-sided and are not adjusted for multiple comparisons. All analyses were conducted using SAS software (version 9, SAS Institute Inc., Cary, NC).
At the time of this report, 1890 patients had been enrolled in the Prospective Cohort, and follow-up data were available for 935 at months 1 and 6. Of these 935 patients, 476 (51%) had no history of CVD or diabetes, 375 (40%) had preexisting diabetes mellitus (22% of whom also had received pancreas transplants) and 281 (30%) had documented preexisting CVD, defined as either coronary artery disease, myocardial infarction, congestive heart failure, stroke or peripheral vascular disease (Table 1). Because their exclusion did not alter any examined metric, diabetic patients with pancreas transplants were included in all analyses. Thus, patients who could be considered at high risk for CVD, that is, those with either known CVD or diabetes, made up almost half (49.1%) of the study population (Table 1).
Table 1. Study population characteristics
Diabetes and CVD
No diabetes or CVD
1The p-values comparing the four groups are from a chi-square test (comparing proportions) and a Kruskal–Wallis test (comparing medians).
2Values are median (interquartile range).
To convert: creatinine mg/dL to μmol/L multiply by 88.4; cholesterol and LDL to mmol/L multiply by 0.02586; triglycerides to mmol/L multiply by 0.01129.
Overall, 86% of all subjects received antihypertensive therapy at month 1, almost identical to the proportion on these agents at 6 months (85%) (Table 2). The most commonly used antihypertensive medications were beta-blockers (67 and 63%) and calcium blockers (49 and 43%) at 1 and 6 months, respectively. Those with a history of CVD or diabetes were significantly more likely to be taking antihypertensive agents at the time of transplant; after transplantation, preexisting CVD but not diabetes was associated with increased likelihood of antihypertensive therapy. (Table 2) Similar trends were detected for use of beta-blockers.
Table 2. Medication use
Entire cohort (%) (N = 935)
CVD (%) (N = 281)
Diabetes (%) (N = 375)
No diabetes or CVD (%) (N = 476)
CVD OR (95% CI)1
Diabetes OR (95% CI)1
1Logistic regression models are stratified by clinical center; CVD is compared to no CVD; diabetes is compared to no diabetes. Models include both CVD and diabetes status and their interaction if significant.
There was a significant interaction between diabetes and CVD (p = 0.02).
OR (95% CI) for each group are as follows:
Diabetes/no CVD: 5.2 (3.6, 7.5)
CVD/no diab: 3.7 (2.4, 6.3)
Diabetes and CVD: 3.0 (2.1, 4.4)
2.5 (1.8, 3.4)
1.6 (1.2, 2.2)
There was a significant interaction between diabetes and CVD (p = 0.03).
OR (95% CI) for each group are as follows:
Diabetes/no CVD: 2.0 (1.4, 2.9)
CVD/no diab: 4.1 (2.5, 6.8)
Diabetes and CVD: 3.8 (2.6, 5.5)
There was a significant interaction between diabetes and CVD (p = 0.001).
OR (95% CI) for each group are as follows:
Diabetes/no CVD: 12.9 (8.6, 19.4)
CVD/no diab: 8.6 (5.1, 14.6)
Diabetes and CVD: 4.6 (3.1, 6.9)
2.9 (2.1, 4.1)
3.5 (2.5, 5.0)
2.8 (2.0, 4.0)
3.7 (2.7, 5.3)
In contrast to the liberal use of beta-blockers at both months 1 and 6, the overall proportion of patients on an ACEI or ARB after transplant was small. Though 45% of recipients were on these drugs at the time of transplant, only 12% were taking ACEI or ARB at 1 month, and only 24% at 6 months, a percentage that did not appear to vary by preexisting comorbidity. (Table 2).
Fewer than half of all patients were treated with aspirin, and only 60% with diabetes and/or CVD received aspirin. Preexisting CVD and diabetes were associated with use of aspirin (Table 2). Among the patients with diabetes and/or CVD, approximately half were prescribed a statin at month 1 and 6. Preexisting CVD and diabetes were associated with a higher likelihood of statin use (Table 2).
For all these agents there was wide variability in the prescription practice between centers (Figure 1). At some centers, a majority of patients received a combination regimen including aspirin, beta-blockers and statins. At others, some agents were used widely, while others were not. In at least one center, few patients received any of these agents.
In an attempt to discern the basis for use of lipid-lowering agents in DeKAF, statin usage in patients with diabetes or CVD was compared to the use among patients without these risk factors, stratified by serum levels of low-density lipoprotein (LDL) cholesterol. Figure 2 illustrates the use of statins by risk group and LDL level. Among the lower-risk patients, those with documented hypercholesterolemia (LDL >130 mg/dL) were more likely than those with more acceptable LDL cholesterol to receive statins (p = 0.03).
Death with function accounts for half of late allograft failure, with CVD its most frequent cause. Much effort is now focused on defining CVD risk after kidney transplantation, and pharmaceutical companies have spent millions marketing relatively minor advantages in CVD risk associated with one immunosuppressive medication versus another. There now appears to be an emerging consensus in favor of managing CVD risk factors including hypertension, diabetes mellitus and hyperlipidemia, after kidney transplantation (6–8). Indeed, in this study, the increased usage of statins in patients with preexisting CVD or diabetes independent of hyperlipidemia may indicate such a predisposition. In the general population, there have been a large number of randomized controlled trials supporting use of medications to reduce CVD and mortality. However, there have been few such trials in kidney transplant recipients, and it is unclear whether guidelines for reducing the incidence of CVD in the general population are applicable to kidney transplant recipients. Our findings indicate marked heterogeneity in clinical practice in this regard, perhaps reflecting significant uncertainty among transplant physicians given the paucity of randomized trials in the population we treat.
The current data can be interpreted as a glass either ‘half full ’ or ‘half empty’. Indeed, approximately 75% of the patients with diabetes and/or CVD are receiving beta-blockers from the time of transplantation onward, and approximately two-thirds are on statins by 6 months after transplantation. Nonetheless, practice patterns are highly variable among, and even within, transplant centers. How is this variability to be explained? Some patients receive the bulk of posttransplant care from community physicians, many of whom may be reluctant to prescribe new agents. Use of anticoagulants in some patients may limit prescription of aspirin. There is also a hierarchy of needs that occupies the time spent in patient encounters at transplant centers; a transplant physician or surgeon struggling to eliminate BK nephropathy or donor-specific antibody may not be left with sufficient time to address cardioprotection. However, the importance of these data is reflected in the response of at least one DeKAF center to their availability: systematic identification of all patients not receiving aspirin prophylaxis after transplant, with a dramatic increase in the percentage of patients now taking low-dose aspirin (personal communication). It is this change in practice patterns at our centers that leads us to report these data in their current form at this stage of the project.
Beyond mere practice patterns, there are several important medical justifications that may limit use of potentially cardioprotective agents after transplantation. The drug–drug interaction between statins and calcineurin inhibitors is very real, especially among cyclosporine-treated patients, with some wariness on the part of patient and physician to use the combination (7). More recipients are now taking tacrolimus, which may have less potential to exacerbate statin-related adverse effects like rhabdomyolysis and myalgias, and experience has shown patients generally able to tolerate low-dose statins without difficulty (9,10). The current data indicate that many patients taking ACEI/ARB therapy at the time of transplantation have those medications discontinued, a not-unwarranted response to fear of suboptimal allograft function postoperatively. In addition, use of ACEI/ARB therapy may contribute to significant anemia after transplantation, again justifying some caution. Finally, the cost of these agents may be daunting for some patients; at the current time, multiple preparations within each category of cardioprotective therapy are available in the United States in a generic formulation.
The only randomized controlled trial assessing the effect of a medication on reducing CVD events is the assessment of Lescol in renal transplantation (ALERT) study, and even this trial did not show a statistically significant effect of the statin on the primary endpoint (11,12). It is unclear whether the results of trials to reduce CVD in the general population are applicable to kidney transplant recipients, and this ambiguity could explain the lack of more consistent use of possible cardioprotective agents in this population. The heterogeneity in the use of these agents also suggests that there is an opportunity to conduct appropriate placebo-controlled trials, given that there remains equipoise between placebo and the use of most, if not all, of these medications. However, it is far from clear whether such trials will be conducted, nor what their results might be (13,14).
In summary, preliminary data from the DeKAF study indicate heterogeneity in the use of potentially cardioprotective agents in kidney transplant recipients, even among patients considered to be at high risk due to preexisting diabetes or CVD. These results indicate that there is an opportunity to conduct placebo-controlled trials for CVD in kidney transplant patients, and if well-designed clinical trials demonstrate the efficacy of cardioprotective agents in kidney transplant recipients, there is great potential for reducing CVD events in this population.