Among studies evaluating associations between cardiac stress testing results and angiographically defined obstructive CAD in cohorts with chronic kidney disease (CKD) stage 5 (glomerular filtration rate (GFR) <15 mL/min/1.73 m² or dialysis dependence), DSE and MPS have sensitivities varying from 0.44 to 0.89 and 0.29 to 0.92 and specificities ranging from 0.71 to 0.94 and 0.67 to 0.89, respectively, for identifying one or more coronary stenoses ≥70 % [23–31]. In a recent meta-analysis assessing the pooled sensitivity and specificity of DSE and MPS in kidney transplant candidates, both DSE and MPS had moderate sensitivity and specificity in detecting significant coronary artery stenosis on invasive coronary angiography in patients who are kidney transplant candidates (DSE (13 studies), pooled sensitivity 0.79 (95 % CI 0.67 to 0.88) and pooled specificity 0.89 (95 % CI 0.81 to 0.94); MPS (9 studies), pooled sensitivity 0.74 (95 % CI 0.54 to 0.87) and pooled specificity 0.70 (95 % CI 0.51 to 0.84)) [32]. When limited to studies that defined CAD as ≥70 % stenosis on coronary angiography, there was little change in these pooled estimates of accuracy (DSE (nine studies), pooled sensitivity 0.76 (95 % CI 0.60 to 0.87) and specificity 0.88 (95 % CI 0.78 to 0.94); MPS (seven studies), pooled sensitivity 0.67 (95 % CI 0.48 to 0.82) and pooled specificity 0.77 (95 % CI 0.61 to 0.88)) [32]. There was evidence that DSE had improved accuracy over MPS (p = 0.02) when all studies were included in the analysis, but this was not significant when the authors excluded studies which did not verify all results with coronary angiography or use a reference standard threshold of ≥70 % stenosis (p = 0.09) [32].

The operational characteristics of the type of stress imaging may differ in ESRD compared with non-ESRD patients. In one study of coronary flow reserve in 64 patients with normal epicardial arteries, 57 % (12/21) of those with diabetic nephropathy had high resting coronary basal flow with no incremental response to adenosine compared with 18 % (2/11) of diabetic patients without renal failure and 9 % (3/32) of nondiabetic patients [33], suggesting impaired vasodilator reserve in diabetic ESRD patients. Overall, the accuracy of inotropic stress echocardiography for the purpose of screening to identify high-risk coronary anatomy may be somewhat superior to that of vasodilator stress nuclear perfusion imaging; however, patient-specific variables, such as echocardiography imaging quality or prior testing results, may guide providers in the selection of stress imaging modality. Comparative studies using contemporary imaging technology (e.g., attenuation correction in MPS) are needed.

Despite the imperfect accuracy of stress testing for the detection of coronary stenoses, abnormal MPS and DSE test results have shown prognostic value for cardiac events and mortality in the ESRD population [21, 27, 30, 31, 34–39]. In a meta-analysis of 12 studies employing either thallium-201 MPS or DSE, ESRD patients with inducible ischemia had approximately six times the risk of myocardial infarction and four times the risk of cardiac death as patients without inducible defects, and patients with fixed defects had nearly five times the risk of cardiac death [40]. The percentage of ischemic segments by DSE was an independent predictor of mortality among 485 patients with advanced renal failure and offered prognostic information beyond clinical characteristics alone [41]. In a study of 126 ESRD patients who underwent 99 m-technetium MPS as part of their pretransplant assessment, the presence of a reversible defect was associated with three times the risk of posttransplant cardiac events and nearly twice the risk of death compared with normal test results [42].

Angiographic studies from the 1970s have documented a high prevalence of coronary stenoses among patients on long-term dialysis [43–47]. More recently, angiographically significant CAD was found in 53 % of a sample of 30 incident ESRD patients without known cardiac history who consented to screening angiography, including 10 (83 %) of the 12 participants with diabetes, although notably angiographic significance was liberally defined as lesions >50 % [48]. Recent reports of angiography in patients undergoing transplant evaluation identified angiographically significant CAD in 42–81 % of samples, with a higher prevalence in subgroups defined as “high risk” by clinical criteria and with use of more liberal angiographic definitions of CAD [22, 29, 30, 49–53].

Despite the high prevalence of CAD in ESRD patients, reported associations of angiographic coronary stenoses with subsequent clinical events in ESRD patients have varied. In one prospective study of 126 renal transplant candidates clinically classified as moderate (age ≥50 years) or high (diabetes, extracardiac vascular disease, or known CAD) coronary risk, MPS, DSE, and coronary angiography were performed in all participants [29]. Clinical risk stratification and coronary angiography predicted MACE after median follow-up of 46 months, but results of MPS and DSE were not significantly correlated with MACE. Several observational studies reported increased unadjusted risk of all-cause mortality and MACE in patients with angiographic CAD [30, 50], while other studies found correlations of angiography and outcomes only in certain patient subgroups, such as those with proximal CAD [49] or with nondiabetic renal failure [51]. Several recent studies have found no associations of CAD with survival of ESRD patients and kidney transplant candidates [22, 52, 53].

There are significant gaps in the literature regarding the outcomes of prophylactic coronary revascularization before kidney transplantation. Only one small randomized trial of revascularization has been performed in this patient population. In 1992, Manske et al. randomly assigned 31 insulin-dependent diabetic transplant candidates with CAD (>75 % stenosis) to revascularization or medical therapy with a calcium channel blocker and aspirin [54]. Ultimately, 10 of 13 medically managed and 2 of 13 revascularized patients reached the primary MACE endpoint including unstable angina, myocardial infarction, or cardiac death. The contemporary relevance of these findings is limited by the small sample size, high event rate among the medically managed group, and subsequent advances that have occurred in medical management of CAD.

While there are no contemporary randomized trials of prophylactic coronary revascularization in kidney transplant candidates, several observational studies have described outcomes after coronary revascularization in selected cohorts of potential kidney transplant candidates. In an observational study of 300 patients who underwent multimodality testing as part of the candidate evaluation at one center, crude survival was not different in patients who underwent revascularization compared with those who underwent angiography without revascularization or no angiography [22]. Among 3,698 patients evaluated for kidney transplant at a single center in 2001–2004, MPS was performed in 60 %, and 7 % of the sample subsequently underwent coronary angiography. The presence and severity of CAD on angiography were not predictive of survival, and coronary revascularization was associated with survival only in patients with 3-vessel CAD [52].

A recent study described one center’s experience (2006–2009) under a protocol in which potential kidney transplant candidates were evaluated by angiography based on the presence of any of the following criteria: age >50 years, diabetes mellitus, any cardiac symptoms, or electrocardiogram (ECG) evidence of ischemia or prior infarction [55]. Among the 657 patients who underwent angiography, significant CAD (defined as >75 % stenosis of one or more coronary arteries, >50 % left main stem stenosis, or an equivocal lesion with flow limitation) was found in 28 %, of whom 55 % were asymptomatic and without prior CAD history. Patients with significant CAD who underwent revascularization followed by transplant (n = 51, 1-year survival 100 %, 3-year survival 97 %) or by continued waiting (n = 177, 1-year survival 95 %, 3-year survival 81 %) had superior survival to the 16 patients who declined revascularization (1-year survival 75 %, 3-year survival 37 %). While this study demonstrates excellent survival in transplant recipients who received preemptive revascularization, the lack of a comparator group of similar patients who did not undergo angiography before transplant prevents inferences regarding the impact of the management approach on outcomes.

In stable general population samples, recent randomized trials have failed to support benefit of revascularization over contemporary medical management, including the CARP and DECREASE-V trials performed among patients awaiting major vascular surgery [56, 57]. Randomized comparison of PCI plus optimal medical therapy among 2,238 patients with stable CAD (≥70 % stenosis in at least one proximal epicardial coronary artery and objective evidence of myocardial ischemia) also found no benefit on the risk of MACE over 4.6 years compared with optimal medical therapy alone [57]. Secondary analysis of 320 participants with CKD defined as estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m² found that, while CKD was associated with increased risk of death or nonfatal MI, the incidence of death or MI in CKD patients did not differ by treatment arm [58].

Weighing these data, the 2012 AHA/ACCF writing group acknowledged that there are “no definitive data at this time for or against screening for myocardial ischemia among kidney transplant candidates who are free of active cardiac conditions” [14]. However, until more data are available, the authors recommended consideration of CAD risk factors to target screening of patients with the highest pretest likelihood of prognostically significant CAD. Risk factors for CAD deemed relevant to transplant candidates in a 2007 report from an international collaboration of the NKF and the Transplantation Society called the Lisbon Report include diabetes, prior cardiovascular disease, >1 year on dialysis, left ventricular hypertrophy, age >60, smoking, hypertension, and dyslipidemia [59]. When compared with the ACC/AHA “Guidelines on Perioperative Cardiovascular Evaluation and Care for Noncardiac Surgery,” screening transplant candidates with multiple risk factors defined by the Lisbon report appeared to improve sensitivity and specificity for the identification of CAD (sensitivity 94 % vs. 77 %; specificity 33 % vs. 24 %) and reduce the overall frequency of testing in a retrospective analysis of data from one center [18]. The Class IIb/LOE C recommendation of the 2012 AHA/ACCF Scientific Statement on consideration of noninvasive stress testing in kidney transplant candidates with no active cardiac conditions based on the presence of multiple CAD risk factors regardless of functional status is shown in Table 10.1. The Scientific Statement also recommends a randomized trial of the impact of screening for CAD on pre- and posttransplant MACE among potential transplant candidates, such as a randomized comparison of four approaches to the preoperative cardiovascular evaluation: (1) cardiac catheterization for all candidates, including fractional flow reserve measurement in indeterminate lesions, (2) noninvasive imaging stress testing (DSE or MPS) for all patients, (3) noninvasive stress testing based on presence of ≥3 cardiac risk factors, and (4) ACC/AHA cardiac evaluation and care algorithm for noncardiac surgery [14].

While there are limited data on the association of preoperative ECG with posttransplant outcomes, data are available on the natural history and prognostic value of baseline ECG abnormalities in dialysis patients. In one study of 12-lead ECG findings among 221 chronic hemodialysis patients compared with patients non-dialysis requiring CKD and patients without kidney disease, the prevalence of ECG abnormalities was 65 %, 41 %, and 5 %, respectively [65]. Significant differences between dialysis patients and normal controls included left ventricular hypertrophy (19 % vs. 0.7 %), evidence of ischemia (7.2 % vs. 0.6 %), premature ventricular contractions (6.8 % vs. 0.3 %), nonspecific ST-T changes (6 % vs. 0.3 %), atrial fibrillation (5.4 % vs. 0 %), left atrial enlargement (2.7 % vs. 0 %), and old MI (1.4 % vs. 0 %). Eighty-seven hemodialysis patients were followed by serial ECG for an average of 7.5 years, during which 39 % maintained normal ECGs, 31 % had abnormal but stable ECGs, 25 % had worsening of their ECG findings, and 5 % had apparent improvement. In another study of 1,738 asymptomatic diabetic patients who were free of known CAD, the presence of pathologic Q waves on ECG was the strongest independent correlate of abnormal “high-risk” MPS results (OR 3.92) [66]. Sixty-one percent of patients with high-risk MPS results were found to have angiographic evidence of left main CAD, 3-vessel CAD, or 1- or 2-vessel with proximal left anterior descending CAD.

Given the low cost and the prognostic implications of ECG abnormalities, the 2012 AHA/ACCF Scientific Statement on pretransplant cardiac evaluation recommends a preoperative resting ECG among all potential transplant candidates but with a stronger evidence class among those with symptoms or known atherosclerotic vascular disease (Table 10.1) [14]. Serial ECG changes in dialysis patients can be expected and thus periodic monitoring of ECGs (e.g., annually) while on the waitlist may be appropriate, although there are no data to support an optimal time interval between repeated ECG tests.

Exercise as the form of cardiovascular stress including exercise treadmill stress testing (ETT) alone or in combination with stress echocardiography or MPS, is the most commonly used method for the performance of stress testing, in ambulatory general population patients in whom there is no evidence of left bundle branch block or significant resting ST segment deviation. Exercise duration and the occurrence and severity of exercise-induced chest discomfort and ECG changes provide important diagnostic and prognostic information. However, there is limited data regarding the diagnostic and prognostic value of regular ETT in pretransplant populations. Based on documented improvements in test accuracy using stress imaging modalities as compared to ETT and the high rate of baseline ST segment abnormalities and left ventricular hypertrophy in pretransplant patients that may limit accuracy of ETT, the use of ETT has not been widely studied in this population [30]. In addition, the majority of studies evaluating stress echocardiography and MPS have used inotropic or vasodilator stress imaging likely as a way to standardize stress imaging protocols among the study populations [32]. Given the increased cost (and risks in the case of dobutamine administration) associated with pharmacologic stress imaging, studies evaluating the accuracy and prognostic value of exercise stress imaging protocols among appropriately selected patients requiring kidney transplantation are needed.

Multiple studies have consistently demonstrated the diagnostic and prognostic value of resting echocardiography among dialysis patients [30, 67, 68]. The “NKF/KDOQI Clinical Practice Guidelines for Cardiovascular Disease in Dialysis Patients” recommend that a resting echocardiogram be performed in all patients at the initiation of dialysis, once the patient has achieved dry weight (ideally within 1–3 months of dialysis initiation) [16]. Two reports from one large center using stress MPS in potential candidates meeting AST criteria for pretransplant ischemia evaluation found left ventricular systolic dysfunction, defined as LVEF less than 40–45 %, in 16–18 % of patients [69, 70]. The majority (61–63 %) of these patients did not have evidence of ischemia by perfusion imaging. Median survival in patients with LVEF <40 % was 49 months compared with 72 months in patients with higher LVEF; after adjustment for ischemia and other risk factors, the relative risk of mortality increased by 2.5 % for each percent decline in LVEF [70]. Further, the 2007 ACC/AHA guidelines on preoperative evaluation in patients undergoing noncardiac surgery recommend the performance of testing to assess left ventricular function among patients with unexplained dyspnea or with known heart failure and a change in functional status [17]. As described in the 2012 AHA/ACCF Scientific Statement, it is reasonable to perform preoperative echocardiography in potential kidney transplantation candidates (Table 10.1) [14].

In addition to MPS and DSE, noninvasive image-based testing for CAD is available using cardiac computed tomography (CT) (Table 10.3). Noncontrast CT for detection and quantification of coronary artery calcification (CAC scoring) has been shown to significantly improve cardiovascular risk prediction as compared to the Framingham score and highly sensitive C-reactive protein in asymptomatic patients without kidney disease [71–73]. Elevated CAC scores (median Agatston score 595, interquartile range 76–1,600) were identified in a sample of 205 adult maintenance hemodialysis patients [74], and similar studies have reported markedly greater coronary calcification in ESRD compared with non-ESRD patients, with greatest disparities according to ESRD status in young cohorts [75–77]. Although one study reported Agatston CAC scores to be an independent predictor of death in chronic hemodialysis patients [78], the role of calcium scoring as a prognostic marker in the ESRD population is uncertain [79]. Other studies report poor correlation between CAC scores and the likelihood of angiographic CAD in patients with advanced kidney disease [80–82], a finding that may reflect the high burden of medial vascular calcification in ESRD compared to the intimal atherosclerotic plaque calcification seen in the non-ESRD population [83]. Modern (64–320-slice and dual-source) coronary computed tomography angiography (CTA) is a highly accurate and prognostically informative noninvasive tool for the diagnosis of CAD when used to evaluate symptomatic patients with low-to-intermediate pretest probability of obstructive coronary stenosis [84–86]. However, coronary CTA has not been adequately studied in patients with significant kidney disease, and its accuracy may be limited in this population due to a high burden of calcified coronary atherosclerosis. Further, the safety of CTA in patients with advanced kidney disease is a concern related to the attendant exposure to iodinated contrast. The “ACCF/SCCT/ACR/AHA/ASE/ASNC/NASCI/SCAI/SCMR 2010 Appropriate Use Criteria for Cardiac Computed Tomography” consider coronary CTA as an option among patients undergoing cardiac surgery for noncoronary indications (e.g., valve replacement) when the pretest CAD risk is either intermediate (appropriate) or low (uncertain) but deem that there are currently no appropriate indications for coronary CTA as part of the preoperative evaluation for noncardiac surgery [87]. Currently, the usefulness of noncontrast CT for CAC scoring and coronary CT angiography is uncertain for the assessment of pretransplant cardiovascular risk (Table 10.1) [14].