Coronary artery disease (CAD) in end‐stage renal disease (ESRD)
The poor prognosis of patients with ESRD is largely due to the high incidence of cardiovascular disease, which accounts for almost 50% of deaths . Twenty per cent of these deaths are due to acute myocardial infarction. The excessive risk of CAD  is highest in elderly patients and in diabetics. In addition to a high prevalence of traditional risk factors, a number of uraemia‐specific factors contribute to coronary atherosclerosis and myocardial ischaemia [3,4]. As there is hardly any evidence based on prospective trials in this group of patients, there is still uncertainty about the best strategy to diagnose and treat CAD in ESRD.
Diagnostic approach to CAD in ESRD
Symptoms of angina pectoris are often misleading in patients with ESRD. There is a high prevalence of silent myocardial ischaemia. Significant CAD is found in 30–50% of asymptomatic or mildly symptomatic diabetics screened before renal transplantation [5,6]. On the other hand, about 25% of ESRD patients with angina pectoris have no significant narrowing of epicardial coronary arteries due to small‐vessel disease and microcirculatory dysfunction. The diagnostic information of baseline electrocardiogram is often obscured by non‐specific changes due to left ventricular (LV) hypertrophy, electrolyte disturbances, or uraemic pericarditis. Resting echocardiography is a valuable tool to detect regional wall motion abnormalities and to diagnose global LV dysfunction, which is significantly associated with severe CAD.
For non‐invasive stress testing, several methods can be used. The value of electrocardiographic (ECG) ergometry in ESRD is limited because of non‐specific basal ECG abnormalities, excessive hypertension during exercise, and poor exercise tolerance, less than half of dialysis patient reaching target heart rates. Exercise thallium scintigraphy also requires that the patient is able to exercise submaximally. It has a sensitivity of 90% for detection of CAD, but a specifity of only 68%, most probably due to the high prevalence of LV hypertrophy [5,6]. Pharmacological stress testing using dipyridamole scintigraphy may be an alternative, but in patients with ESRD its value as a screening tool is limited. Widely varying sensitivities and specificities have been reported , which may be explained by the fact that dipyridamole induces myocardial blood flow heterogeneity rather than true myocardial ischaemia. Recently the combination of dipyridamole and exercise thallium imaging has been proposed as a more accurate method . With a negative predictive value of 98% this test may be especially useful to rule out significant CAD in asymptomatic patients with ESRD.
Dobutamine stress echocardiography is independent of exercise capacity and may therefore be particularly valuable in patients with ESRD. In such patients it detects CAD with a high sensitivity and specificity and identifies a cohort of patients at low risk of further cardiac complications . In a recent study in renal transplant candidates, the sensitivity and specificity were 75 and 71% to detect CAD as defined by quantitative coronary angiography, with a positive and negative predictive value of 45 and 90% respectively . Therefore dobutamine stress echocardiography appears to be a useful but imperfect screening method.
The gold standard for the diagnosis of CAD remains coronary angiography. In renal failure, however, the periprocedural risk is increased. Patients with preterminal renal insufficiency have a high risk of contrast‐induced renal failure, which may be prevented by adequate hydration [11,12]. In uraemic patients, especially in diabetics with LV failure, pulmonary oedema may occur as a result of contrast medium associated volume load. Also bleeding complications occur more frequently.
Even without considering cost, it is evident that coronary angiography cannot be used as a screening method, and that catheterization should be reserved for those patients with a high risk of CAD and those who would benefit from revascularization. A useful diagnostic management strategy for screening patients planned for renal transplantation has been proposed that stratifies patients into groups with low, intermediate, and high risk [6,13]. Younger patients (<50 years) without diabetes, without symptoms suggestive of CAD or heart failure and a normal ECG may be considered to carry a low cardiovascular risk and do not require invasive cardiac evaluation. Patients aged >50 years or patients with diabetes, but without cardiac symptoms, are considered ‘intermediate risk’. These should be evaluated by non‐invasive tests like exercise or dipyridamole scintigraphy or dobutamine echocardiography, and should undergo cardiac catheterization in the event of a positive result. Patients with cardiac symptoms or evidence of previous myocardial infarction are high‐risk patients and should all undergo coronary angiography. There is a debate as to whether all patients with diabetes of longer duration should be considered ‘high risk’ and evaluated accordingly.
Treatment of CAD in ESRD
Of paramount importance is aggressive risk‐factor modification (diabetes, hyperlipidaemia, smoking, etc.) by life‐style changes, dietary intervention, and pharmacological measures. Normalization of blood pressure should be aimed at. Further medical therapy includes, as in non‐uraemic patients, administration of acetyl salicylic acid in all patients (alternatively ticlopidine or clopidogrel) and, if possible β‐blockers. In addition, nitrates and calcium‐channel antagonists can be used for the treatment of angina pectoris. For most of these drugs, dose adjustments are not required. However, for some substances such as atenolol or acebutolol, dose reduction is required in uraemia. HMG‐CoA reductase inhibitors (‘statins’) are the agents of choice for primary or secondary prevention in hypercholesterolaemia. In renal transplant recipients receiving cyclosporin they have to be used cautiously and in reduced dosage because of an increased risk of rhabdomyolysis.
The effectiveness of medical therapy for the treatment of CAD has not been studied specifically in patients with ESRD. There is some evidence for the benefit of β‐blocker therapy. In a retrospective analysis, the use of a β‐blocker after myocardial infarction was associated with a 40% reduction in 2‐year mortality. Subgroup analyses showed that patients with diabetes and renal insufficiency start with a much higher risk, but relative risk reductions by a β‐receptor antagonist were similar (−36 and −35%) . Unfortunately, it appears that not all ESRD patients in need of this intervention receive this potential benefit .
Some aspects of therapy are specific for uraemic patients: hypervolaemia, which increases myocardial O2 demand, should be avoided using dietary sodium restriction, possibly diuretics, haemodialysis, and optimization of dry weight. Modalities and techniques of dialysis have to be adjusted to avoid hypotension and tachycardia. In some patients, peritoneal dialysis may be preferable to haemodialysis. Anaemia should be corrected to increase O2 delivery using erythropoietin. Whether the haematocrit to be aimed should be 30–35% or higher is still a matter of debate . Hyperphosphataemia should be treated and an increased calcium–phosphorus product avoided to prevent myocardial and coronary calcification.
The key question is when and how to perform a revascularization procedure. Most studies investigating the benefit of revascularization over medical therapy have excluded uraemic patients. There is only one very small, prospective study in patients with ESRD . Twenty‐six asymptomatic or mildly symptomatic diabetic patients with significant CAD documented by pretransplant coronary angiography were randomly assigned to medical treatment (aspirin plus a calcium antagonist) or revascularization (angioplasty or bypass surgery). Ten of 13 medically managed patients reached a cardiovascular end‐point and five died within a median of 8 months as compared to two of 13 (1 death) in the revascularization group. Although the study group was too small to allow general conclusions and the form of medical treatment would not be regarded adequate today, these results underline the poor prognosis of diabetic patients with ESRD and silent CAD, and do suggest that revascularization procedures may decrease morbidity and mortality in this population. But what is the best revascularization method?
Dialysis patients undergoing bypass surgery have a high perioperative mortality of about 9%  which is at least three times higher than in non‐uraemic patients . Identifiable risk factors are dyspnoea at rest, dialysis >5 years, NYHA class IV, emergency operation, older age, and LV dysfunction . LV hypertrophy, LV dysfunction, and myocardial and coronary calcification are associated with a high incidence of perioperative infarction, arrhythmia and low‐output syndrome. In addition, uraemic patients have more bleeding complications and infections.
Long‐term outcome of dialysis patients after bypass surgery is poor. Dialysis patients from the US Renal Data System database operated between 1978 and 1995 had a 5‐year survival rate of only 26.5% , in contrast to a 5‐year survival rate of about 90% in non‐uraemic patients . In a retrospective analysis of 51 dialysis patients undergoing bypass surgery between 1988 and 1997, however, 5‐year survival was 65% . Although this was still worse than in non‐dialysis patients and the study was small, these more optimistic results may reflect advances in surgical technique, especially regarding the use of arterial grafts, and may be more relevant to current practice.
Balloon angioplasty (PTCA)
Regarding PTCA in dialysis patients, low acute success rates and high complication rates [22–24] as well as poor outcomes have been reported in earlier small studies. After 6 months, recurrent ischaemia was observed in 63% of patients, myocardial infarction in 23%, and death in 13% . Several factors may contribute to the unfavourable acute and long‐term results of PTCA in these patients: more complex lesions with diffuse disease and extensive calcification, smaller vessel diameters, a large proportion of diabetics and patients with multivessel disease, as well as prothrombotic factors. Restenosis rate following angioplasty was found to be increased to 60% in ESRD patients, compared with 35% in controls . Factors associated with restenosis were significantly smaller diameters of the coronary reference segment, pointing to a more diffuse disease, as well as raised plasma fibrinogen levels, indicating increased prothrombotic risk.
Bypass vs PTCA
In ESRD patients the different revascularization strategies have not been compared in prospective randomized trials. A retrospective analysis evaluated 84 dialysis patients revascularized by PTCA or bypass surgery . Although the bypass group had more severe CAD, the outcome was better regarding freedom from angina, myocardial infarction, and cardiovascular death, although there was no difference in overall mortality. However, patient groups may not have been comparable. The PTCA group included more diabetics and patients who were ‘not surgical candidates’. The largest retrospective analysis is based on over 14000 dialysis patients from the US Renal Data System database revascularized between 1978 and 1995 . Despite a higher in‐hospital mortality of surgically treated patients (12.5 vs 5.4% in the PTCA group), their overall long‐term survival was better. However, the 5‐year mortality rate was still as high as 73.5% (vs 76.8% in the PTCA group). Also the incidence of myocardial infarction was lower in the bypass group. The most powerful predictors of poor outcome in both groups were advanced age (>75) and diabetes. A higher proportion of such high risk patients were in the PTCA group. This explains at least in part, the difference in outcome. Randomized studies in non‐dialysis patients have shown that diabetics with multivessel disease are the main group of patients in whom the prognosis after bypass surgery is better than after PTCA .
Survival after myocardial infarction was recently examined in 640 dialysis patients in relation to treatment modality . Patients treated medically had a very low survival rate of only 45% at 1 year. Survival was better in patients treated by PTCA (54%) and was best in patients undergoing bypass surgery (69%). However, after adjustment for covariates such as history of stroke, heart failure, ejection fraction, age, and body mass index, these differences were no longer significant. Although this study shows that dialysis patients who are revascularized do better after myocardial infarction, especially if they undergo bypass surgery, these results probably reflect the selection of younger patients with lower risk and fewer comorbid conditions.
New interventional technologies
All of the above‐quoted studies on interventional therapy examined plain balloon angioplasty. What is the place of new interventional techniques such as stenting? Coronary stenting has been shown to reduce acute complications of coronary intervention as well as restenosis rate [29,30]. These results have revolutionized interventional cardiology, with the effect that most interventional procedures nowadays include the placement of a stent. Furthermore, new and effective antiplatelet therapy like the thienopyridines and glycoprotein IIb/IIIa‐receptor antagonists, given in addition to aspirin, have reduced complications and improved outcome after coronary interventions [31,32]. Coronary stenting together with glycoprotein IIb/IIIa‐receptor blockade is associated with better 1‐year survival than stenting alone, and may be the best currently available interventional strategy . It is evident that dialysis patients with CAD should benefit from these new therapies as well. Preliminary data indicate that the procedural success of coronary stenting in such patients is not different from non‐dialysis patients, and cardiac event rate as well as target lesion revascularization rate during follow up is only slightly, but not significantly, higher .
In a preliminary report of a retrospective analysis of dialysis patients undergoing coronary revascularization in the US between 1995 and 1996, 1‐year survival rate after coronary stenting was 80%, which was significantly higher than in patients undergoing balloon angioplasty alone . Interestingly, this was also higher than in patients undergoing bypass surgery. While these data indicate a better outcome in dialysis patients with coronary stenting than with balloon angioplasty alone, there is no data on the role of other modern interventional techniques. As coronary lesions in dialysis patients are often heavily calcified , rotational atherectomy to remove calcified plaque mechanically, followed by stenting, may be a particularly valuable treatment strategy . Brachytherapy, i.e. exposure of stenosed and dilated or stented segments to β‐ or γ‐ radiation, has been shown to reduce the rate of restenosis after coronary intervention [38,39] and might be especially useful in dialysis patients. The value of other techniques like the cutting balloon technique, directional atherectomy, laser angioplasty, and extraction atherectomy is not known for dialysis patients and will probably be restricted to selected cases.
Dialysis patients with CAD have a high cardiovascular risk and a poor prognosis. Current clinical practice regarding diagnosis and management of CAD in these patients is not based on firm evidence, due to the lack of prospective studies. It has to rely on retrospective analyses, extrapolations from studies in non‐uraemic patients, and on clinical judgement. Although prospective randomized trials would be desirable to guide treatment strategies, they may not be feasible because of established practice and the small number of patients with ESRD and CAD relative to the large number of patients with ischaemic heart disease.
A thorough diagnostic evaluation, often including coronary angiography, is indicated especially in patients scheduled for renal transplantation, even if the patients are not or are only mildly symptomatic. Aggressive medical treatment aimed at primary and secondary prevention remains the basis of CAD therapy in ESRD. Studies indicate that these patients benefit from coronary revascularization and do worse with medical therapy alone. In patients who are not candidates for renal transplantation because of age or comorbidity, treatment strategies should probably be more conservative and guided mainly by symptoms, as these patients may not be expected to benefit from revascularization.
Most data suggest that bypass surgery should be the revascularization strategy of choice in dialysis patients, especially in multivessel disease and in diabetics, because of better long‐term outcome as compared to balloon angioplasty. However, the price of a high perioperative mortality has to be taken into account. Recent preliminary data are optimistic regarding new interventional techniques, especially stenting, which is superior to balloon angioplasty alone and might even offer an outcome similar to bypass surgery. There may also be a place for other new interventional techniques and new adjunctive pharmacological therapy in this subset of patients.
A recent editorial was entitled: ‘Poor long‐term survival of dialysis patients after acute myocardial infarction: Bad treatment or bad disease?’ . It is certainly a bad disease with a poor outcome, but a disease that deserves the best possible treatment. Recent developments are encouraging, suggesting that an improvement of outcome may be possible, which warrants further common effort of nephrologists, cardiologists, and cardiovascular surgeons.
Correspondence and offprint requests to: Prof. Dr Dietmar Elsner, Klinik und Poliklinik für Innere Medizin II, Klinikum der Universität, P‐93042 Regensburg, Germany.