Detection of Atherosclerotic Cardiovascular Disease in Patients with Advanced Chronic Kidney Disease in the Cardiology and Nephrology Communities

Cardiovascular disease (CVD) is the leading cause of death among patients with advanced chronic kidney disease (CKD), i.e., with an estimated glomerular filtration rate (eGFR) < 30 mL/min/1.73 m², especially end-stage renal disease (ESRD), with approximately 41% mortality in ESRD attributed to CVD [1, 2]. Nevertheless, patients with advanced CKD and ESRD are underrepresented in clinical trials of CVD [3]. The Kidney Disease Outcome and Quality Initiative 2005 Cardiovascular Disease in Dialysis Patients guidelines recommend the assessment and treatment of risk factors and end organs for CVD in dialysis patients, but they lack detailed specifics on screening modalities and risk stratification beyond transthoracic echocardiography for valvular disease in dialysis patients [4]. The most extensive set of guidelines has been outlined for patients with CKD undergoing transplant evaluation [5]. The majority of recommendations were class IIa or lower due to a lack of strong randomized controlled trial data addressing the topics outlined. No similar set of guidelines is available for advanced CKD or ESRD populations who are not being evaluated for renal transplantation.

Understanding variation in clinical practice in assessing and managing CVD risk factors in such patients can guide future clinical trial designs in addressing knowledge gaps, thereby leading to higher-class recommendations. We evaluated clinical practice patterns among nephrologists and cardiologists regarding the identification and management of atherosclerotic CVD (ASCVD), the most common cause of CVD in the population with advanced CKD – i.e., with an eGFR < 30 mL/min/1.73 m² – and ESRD.

A 22-question survey (5 general questions and 17 clinical scenarios), designed by the authors (A.A., M.S.S., R.I.C., R.L., R.O.M., and S.B.), that assessed provider perceptions and practice patterns concerning the epidemiology, diagnosis, and management of ASCVD in patients with advanced CKD stages 4–5 (eGFR ≤30 mL/min/1.73 m²) and ESRD (eGFR < 15 mL/min/1.73 m² and on dialysis – hemodialysis or peritoneal dialysis, or with a kidney transplant) was developed by a group of 3 nephrologists and 3 cardiologists based on clinical guidelines and the published literature.

This de novo survey (see online suppl. Appendix for a full survey; for all online suppl. material, see www.karger.com/doi/10.1159/000490768) consisted of 3 broad domains: (1) epidemiological perspectives – the prevalence of coronary artery disease (CAD) in advanced CKD and ESRD; (2) screening – the ideal screening modality for CAD in advanced CKD and ESRD, as well as appropriate patient selection for screening; and (3) therapeutics – the best treatment for CAD in advanced CKD and ESRD, as well as appropriate targets for CAD risk reduction among patients with advanced CKD and ESRD. It was approved by the Albany Medical College Institutional Review Board and distributed to nephrology (n = 130) and cardiology (n = 20) providers (physicians, physician assistants, and nurse practitioners) who were contacts of the authors, and nephrologists with Fresenius Medical Care (n = 300), an international provider of dialysis services (only the US providers were contacted). All invited participants received a link to the survey via email (hereinafter, participants will be identified as respondents). The survey was administered between November 2016 and June 2017 using Qualtrics® (https://www.qualtrics.com). Data summaries were derived using the R statistical software package (https://cran.r-project.org/) and Qualtrics®.

The overall response rate was 25% (30% among cardiology providers and 17% among nephrology providers); the complete response rate was 18%. Of the respondents with complete results, 96.3% were physicians, of which 77% were nephrologists and 5% cardiologists. The general demographic features of the respondents with complete information are presented in Table 1.

The respondents generally agreed (96%) that CAD was more common among patients with CKD, irrespective of the CKD stage, than among those without CKD; 91% indicated that an atypical presentation of CAD was more common in CKD. In addition, 94% of the respondents felt that for an individual with advanced CKD, death due to cardiovascular causes was the more likely outcome than was ESRD requiring renal replacement therapy.

For patients with advanced CKD not yet on renal replacement therapy, the majority of the respondents (55%) preferred myocardial perfusion scintigraphy (MPS), followed by coronary angiography (16%) and exercise treadmill testing (11%) (Fig. 1a). For patients with ESRD on dialysis, coronary angiography was the preferred diagnostic modality for CAD (50%) (Fig. 1b).

We ascertained which clinical scenarios would prompt screening for CAD in patients with advanced CKD or ESRD on hemodialysis (Fig. 2). Ninety-five percent of the respondents would not screen an asymptomatic patient with ESRD without known traditional CAD risk factors. In patients symptomatic for possible cardiac disease (e.g., dyspnea on exertion or persistent hypotension on dialysis), or with objective evidence of reduced systolic function as reflected in the left ventricular ejection fraction, at least 85% of the respondents preferred evaluation for CAD as a possible cause of the findings in each scenario. The preference for preoperative cardiac testing depended on the type of surgery. In contrast to the pre-renal transplant workup in essentially asymptomatic patients (in whom > 95% of the respondents would evaluate for CAD), prior to laparoscopic bariatric surgery on an advanced CKD patient with traditional risk factors for CAD (hypertension, diabetes, and smoking), only 73% indicated they would formally evaluate for CAD.

Four case scenarios were presented to the respondents of patients with advanced CKD or ESRD (renal transplant) with symptoms or stress test findings suggestive of CAD with a goal to assess management strategies (Fig. 3). In a patient found to have a positive stress test as part of the kidney transplantation evaluation, 73% (73% of the nephrologists, 80% of the cardiologists) preferred some form of coronary revascularization plus medical therapy over medical therapy alone. Prior to nontransplant surgery (arteriovenous fistula creation), 61% (60% of the nephrologists, 90% of the cardiologists) preferred a revascularization approach over medical therapy alone. In each of these nontransplant case scenarios, the patient had CKD stage 5 (eGFR < 15 mL/min/1.73 m²), and a need for hemodialysis or peritoneal dialysis was not explicitly stated.

In two separate patient scenarios with stable renal transplant function (Fig. 3), the respondents felt that a coronary revascularization approach was most appropriate (74%) when the patient was presenting with the classic symptom of chest pain at rest. Only 59% favored this approach for the patient with diabetes and atypical symptoms (dyspnea on exertion).

The majority of the respondents were split regarding the systolic blood pressure (SBP) target in CKD patients with minimal albuminuria (< 30 mg/g): 140 mm Hg (45%) or 130 mm Hg (40%) (Fig. 4). However, 57% favored the lower SBP target of 130 mm Hg for CKD patients with marked albuminuria (> 300 mg/g). More liberal SBP targets of 140 mm Hg (59%) and 150 mm Hg (25%) were favored for patients with ESRD on maintenance hemodialysis therapy.

Sixty-seven percent of the respondents indicated that they would prescribe high-intensity statin to a dialysis patient previously on statin therapy with multivessel CAD requiring coronary artery bypass grafting (Fig. 5). The likelihood of prescribing high-intensity statin was somewhat lower (55%) for a dialysis patient previously on moderate-intensity statin with nonobstructive 2-vessel CAD on coronary angiography.

The results of this survey study shed light on the current management of CAD among patients with advanced CKD or ESRD, and we feel that they afford important insights into this area. There is general agreement among providers that ASCVD is common, and CVD outcomes are more likely to occur than developing ESRD in patients with advanced CKD; this is well supported by a large mass of epidemiological data [6]. Whereas advanced CKD was widely recognized as an independent risk factor for ASCVD by the survey respondents, there was variation in their responses regarding the diagnosis and management of ASCVD in such patients.

The choice of the screening modality depended on the need for renal replacement therapy (hemodialysis or peritoneal dialysis). The respondents felt that noninvasive modalities, primarily MPS (55%), should be utilized for case detection rather than invasive methods such as coronary angiography (3%) for nondialysis patients with advanced CKD. In patients already on hemodialysis, the preference was for coronary angiography, but only about half of the respondents chose this modality. The reasons for this discrepancy in the use of coronary angiography were not elicited; however, given the difference based on the need for hemodialysis, the concern about avoiding iodinated contrast medium in nondialyzed patients may have been the primary motivating factor. An analysis from the NCDR CathPCI registry showed that the risk of acute kidney injury following percutaneous coronary intervention (PCI) in patients with advanced CKD is higher than in those without CKD, but the risk may be independent of the dye load [7]. An important caveat from this analysis is that all patients were admitted for an acute coronary syndrome; the risk may be attenuated if the PCI is conducted under elective conditions.

Currently there are no randomized controlled trials to guide the choice of a noninvasive screening modality for CAD in patients with advanced CKD or ESRD. The best available data are from a meta-analysis by Wang et al. [8]. The authors identified that MPS (sensitivity 0.69, specificity 0.77) and dobutamine stress echocardiography (sensitivity 0.80, specificity 0.89) had the best test performance amongst the modalities studied, including stress electrocardiography, digital subtraction fluorography, and exercise ventriculography; the difference between MPS and dobutamine stress echocardiography did not meet statistical significance (p = 0.07).

In deciding which patient to send for CAD evaluation, the presence of CKD stage 5 alone, without symptoms suggesting cardiac disease or risk factors for cardiac disease, was not deemed as a reason for testing: 85% of the respondents decided against screening. This is an interesting finding, in that the majority of the respondents felt that cardiovascular morbidity and mortality was the most likely outcome of advanced CKD or ESRD. In other words, the implicit acknowledgement that advanced CKD and ESRD are possibly a CAD equivalent, as is suggested by some [9], has not translated into the evaluation and management of patients with advanced CKD or ESRD as such. On the other hand, when presented with a patient with ESRD and multiple traditional risk factors for CAD, most respondents (73%) felt that screening was appropriate. The KDOQI 2005 Cardiovascular Disease in Dialysis Patients guidelines recommend that “all patients – regardless of symptoms – require assessment for cardiovascular disease […] as well as screening for both traditional and nontraditional cardiovascular risk factors” [4]. The authors noted that the guidelines were vague due to the sparse nature of the available evidence and lack of randomized controlled trials regarding cardiac evaluation of patients with ESRD, and guidelines are lacking for screening of ASCVD in CKD patients not on dialysis, or being worked up for renal transplantation.

Evaluation for kidney transplantation has routinely involved screening for CAD. This includes assessing the presence of traditional risk factors (diabetes) as well as of nontraditional risk factors such as being on dialysis for > 1 year, as occurred in our case scenarios. The survey respondents seemed to identify with the most recent guidelines regarding CAD evaluation prior to kidney transplantation: > 95% would screen for CAD prior to kidney transplant evaluation and 73% favored attempting revascularization if a positive stress test had been obtained prior to kidney transplant evaluation [5]. The ACC/AHA guidelines from 2015 suggest that screening for CAD prior to kidney transplantation should occur if ≥3 risk factors are present; these guidelines have been endorsed by the American Society of Transplantation, the American Society of Transplant Surgeons, and the National Kidney Foundation [5].

The KDOQI guidelines on cardiovascular evaluation provide additional guidance for non-transplant candidates on maintenance hemodialysis. Any patient on hemodialysis with recurrent intradialytic hypotension, inability to achieve a target weight due to hypotension, or a documented drop in ejection fraction (to < 40%) should be evaluated for CAD. Intradialytic hypotension has been associated with a modest 14% increase in the relative risk for developing myocardial infarction [10].

It should be noted that guidelines do not always ensure adequate adherence to evidence-based therapies among providers. The CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation) study sought to understand current acute coronary syndrome guideline implementation across the USA [11, 12]. The investigators identified that the mere presence of guidelines alone does not guarantee adherence by providers. Further measures that improve adherence to guidelines include collaboration between health care providers and administrators, demonstration of improved outcomes in patient care environments, and demonstrating institutional benefits from guideline implementation [12]. To engage administrators will require not only randomized controlled trial data but also cost-benefit analyses.

Of the > 90% of the respondents advocating an evaluation for CAD prior to kidney transplantation workup, the majority (73%) would recommend coronary revascularization. The number was lower for nontransplant surgery: 61% advocated for revascularization prior to arteriovenous fistula creation. In both scenarios, the patients had positive stress test results. This likely reflects the uncertainty based on a lack of clinical trial data regarding the benefit of revascularization for stable ischemic heart disease in patients with advanced CKD. In one of the very few randomized controlled trials on such patients, published in 1992 by Manske et al. [13], coronary revascularization (2 of 13), as compared to medical therapy alone (10 of 13), was associated with a lower incidence of cardiovascular events before or after kidney transplantation in type 1 diabetic patients with advanced CKD (p = 0.002), resulting in an early termination of the study. At the time of that study, optimal medical therapy consisted of aspirin and a beta-blocker or non-dihydropyridine calcium channel blocker and long-acting nitrates. Since then, no randomized trial of coronary revascularization versus medical therapy has focused on patients with advanced CKD or ESRD. The ongoing International Study of Comparative Health Effectiveness of Medical and Invasive Approaches-Chronic Kidney Disease (ISCHEMIA-CKD) (NCT01985360) and the Coronary Artery Disease Screening in Kidney Transplant Candidates (CAD Screening) (NCT02082483) study will help elucidate the best treatment strategy in this population. Furthermore, Bangalore et al. [14] assessed propensity-matched scores in 5,920 patients with CKD who underwent PCI (with everolimus-eluting stents) versus coronary artery bypass grafting for multivessel CAD. PCI was associated with lower short-term (within 30 days) risk of death (HR = 0.55; 95% CI: 0.35–0.87), stroke (HR = 0.22; 95% CI: 0.12–0.42), and early repeat revascularization; however, there was no difference in mortality rate in the long term (mean 2.9 years) with an increased risk of repeat revascularization in the PCI group (HR = 2.42; 95% CI: 2.05–2.85).

Several studies have evaluated the effect of blood pressure reduction in patients with CKD [15-17]. The peculiarity in blood pressure management among patients with CKD is the importance of albuminuria in selecting targets for blood pressure reduction. The Kidney Disease Improving Global Outcomes (KDIGO) Clinical Practice Guidelines for Management of Blood Pressure in CKD recommend lower blood pressure goals for the CKD population with micro- or macroalbuminuria than for those without micro- or macroalbuminuria (> 30 or > 300 mg/24 h, respectively) [18]. The respondents demonstrated adherence to these recommendations in their survey responses. The SBP targets were more liberal for the ESRD patient, with most respondents (64%) suggesting 140 or 150 mm Hg as the target. This finding likely reflects the numerous uncertainties about blood pressure management in patients with ESRD, including appropriate blood pressure monitoring and which outcomes derive a benefit from blood pressure reductions [19]. This is an area in great need of randomized controlled trial data to better guide clinical practice and inform guidelines in the future.

Preliminary evidence suggests that the 10-year Pooled Cohort risk assessment for CAD has validity in CKD patients not on hemodialysis [20]. However, a 10-year risk > 7.5% is very common in this population; hence, the KDIGO Lipid Guidelines recommend blanket statin use for CKD patients > 50 years of age and not on dialysis [21]. This recommendation is in contrast to that for patients on dialysis, for whom neither Die Deutsche Diabetes Dialyse Studie nor the Rosuvastatin and Cardiovascular Events in Patients Undergoing Hemodialysis (AURORA) trials demonstrated any benefit in reducing primary cardiovascular end points with statin therapy [22, 23]. LDL and total cholesterol levels alone do not adequately risk-stratify dialysis patients, in whom low levels of both lipids are common due to malnutrition and chronic disease, despite their high risk for cardiovascular mortality [23]. The Study of Heart and Renal Protection (SHARP) randomized 446 patients on dialysis to statin plus ezetimibe or placebo, but it was not powered to detect the outcome of atherosclerotic events within the dialysis population independent of the larger CKD cohort [24]. Additionally, limited data are currently available regarding the safety profile of higher-dose statins in the CKD population. Therefore, the KDIGO guidelines recommend statin dosing based on the limited randomized trials done on the CKD population, and do not recommend initiation of statin therapy for patients on dialysis [21].

The limitations of the study include its small sample size and low response rate. Additionally, there was a disproportionally higher number of nephrology providers who responded compared to cardiology respondents, translating into underrepresentation of the latter group.

The uncertainties regarding the most appropriate diagnostic strategy, risk factor reduction, and treatment of CAD in patients with advanced CKD or ESRD can only be rectified by greater inclusion of such patients in randomized trials. Steps to rectify this deficiency have been initiated by the SHARP and the ongoing ISCHEMIA-CKD, as well as CAD screening trials. Higher-quality evidence from such trials, as well as a cost-benefit analysis of the utilization of resources, is needed to better define optimal diagnostic and therapeutic strategies for this high-risk population.

The authors have no ethical conflicts to disclose.

S. Bangalore received grant support from the National Heart, Lung, and Blood Institute (NHLBI) for the ISCHEMIA-CKD. The content of this manuscript is solely the responsibility of the authors and does not necessarily reflect the views of the NHLBI, National Institutes of Health, the US Department of Health and Human Services, or the US Department of Veterans Affairs.

No funding was received for this study.