Kidney function in patients undergoing coronary revascularization

Chronic kidney disease (CKD) is recognized as a major global public health problem. CKD affects 10 to 16% of the adult population in Asia, Australia, Europe and USA, and increases the risk of all-cause mortality and cardiovascular disease across broad range of populations, that is, general or high-risk populations.¹ Patients who underwent coronary revascularization are not an exception. In 2002, Best et al.² reported that kidney dysfunction significantly increases the risk of death during and in the first few years after percutaneous coronary intervention (PCI) in a dose-response manner. Importantly, increased mortality after PCI has been observed even in patients with mildly or moderately reduced kidney function. A few other studies have confirmed the clinical significance of CKD on prognosis after PCI.^{3, 4}

In this issue of Hypertension Research, Kasai et al.⁵ have extended our comprehension about the impact of CKD on patients who underwent coronary revascularization in several aspects. Although reduced kidney function currently attracts attention of interventional cardiologists mainly because of its strong impact on the risk of contrast-induced nephropathy, Kasai et al. found that reduced glomerular filtration rate (GFR) contributes to increased risk of mortality for over 10 years after coronary revascularization. Thus, the clinical usefulness of kidney function evaluation before PCI extends beyond short-term risk assessment for contrast-induced nephropathy.

In the study of Kasai et al., patients with CKD stage 3 (estimating GFR (eGFR) 30–59 ml min⁻¹ per 1.73 m²) had 2- to 3-fold increased mortality after coronary revascularization as compared with those who had eGFR ⩾90 ml min⁻¹ per 1.73 m², even after adjusting for potential confounders such as diabetes mellitus. Of note, relative risks of CKD stage 3 for all-cause mortality and cardiac mortality (hazard ratios of 2–3) were greater than those of other predictors such as diabetes (hazard ratios ∼1.5) or previous history of stroke (hazard ratios ∼2), suggesting that patients with CKD undergoing PCI need special attention.

It is worth emphasizing that, in study by Kasai et al., the mean serum creatinine level among patients with CKD stage 3 (eGFR 30–59 ml min⁻¹ per 1.73 m²) was 1.1 mg dl⁻¹, a level lower than the threshold, indicating clinical actions for prevention of contrast-induced nephropathy. This emphasizes the importance of eGFR from serum creatinine rather than interpreting a serum creatinine level itself for clinical management. One strategy to encourage healthcare providers to use eGFR is automated laboratory reporting system of eGFR. Indeed, about 70% of laboratories in the USA are currently reporting eGFR along with serum creatinine results.⁶ Such an automated reporting system would facilitate general physicians, interventional and general cardiologists, and cardiac surgeons to evaluate patients’ kidney function and future risk more precisely and efficiently.

Best et al.² demonstrated that patients with kidney dysfunction tended not to receive complete revascularization, and this may be one of mechanisms by which reduced kidney function is associated with adverse outcomes in patients with indication of coronary revascularization. However, Kasai et al. observed that the impact of CKD on mortality is significant even when patients received complete revascularization by PCI with adjunction of bypass graft surgery in some patients. This may be due to the fact that CKD has been associated with broad range of cardiovascular outcomes, for example, heart failure, stroke or peripheral artery disease,⁷ and also non-cardiovascular diseases, for example, pneumonia or cancer.^{8, 9} Therefore, it is still a challenge to optimize care for CKD patients undergoing coronary revascularization.

Several issues remain to be elucidated. First of all, it is not clear how we can improve the prognosis of CKD patients who undergo coronary revascularization. Renin–angiotensin system inhibitors are currently recommended for patients with coronary artery disease, particularly when they have reduced cardiac function, hypertension, diabetes or albuminuria.¹⁰ Statin are also good candidates to improve outcomes of CKD patients. In general, CKD patients have been understudied in clinical randomized trials, leaving uncertainty of effectiveness of these drugs for CKD patients with coronary artery disease. Preliminary report of the results of the SHARP Study this year suggest that ezetemibe plus simvastatin reduced the risk of major atherosclerotic events in CKD patients by 17% (P=0.0022), although patients with coronary artery disease at baseline were not included in the SHARP Study.¹¹ In the study of Kasai et al., coronary revascularizations were conducted in late 80 s or early 90 s, and thus, use of rennin–angiotensin system inhibitors and statins were quite low (<5 and <20%, respectively), providing less information on the effects of these drugs in this population.

Second, the impact of albuminuria on prognosis among patients who are undergoing coronary revascularization needs to be investigated further. A growing body of evidence suggests that albuminuria is a strong predictor of mortality and cardiovascular disease independently of reduced GFR.¹ Importantly, albuminuria has also been shown to be a predictor of acute kidney injury with a clinical implication for risk evaluation or prevention of contrast-induced nephropathy.¹² Recently, several investigators, including an international conference have proposed revising the CKD staging system to give more equal emphasis to GFR and albuminuria levels.¹² Therefore, it would be important to evaluate independent and combined impact of GFR and albuminuria on short- and long-term prognosis among patients who are planning coronary revascularization.

Third, the impact of CKD on prognosis after PCI with drug-eluting stents needs yet to be elucidated. Drug-eluting stents have substantially decreased stent restenosis, although there is a concern about the potentially increased risk of late-stent thrombosis.¹³ End-stage renal disease has been reported as a potent risk factor of late-stent thrombosis.¹³ Therefore, it is important to evaluate the impact of mildly or moderately reduced GFR on long-term prognosis after drug-eluting stent implantation. Given the period when patients received PCI in study carried out by Kasai et al. (late 80’s and early 90’s), none of study participants received stent implantation.

Finally, it is still controversial as to which equation is the best for eGFR in international settings. Kasai et al.⁵ have used a recently proposed equation for Japanese and showed that GFR estimated by this new equation is a good predictor of mortality after coronary revascularization. In Western populations, the Modification of Diet in Renal Disease Study (MDRD) equation has been most frequently used for eGFR in clinical practice or research.^{6, 12} In 2009, The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) published a new equation using the identical variables as the MDRD equation, namely, age, gender, race and serum creatinine.⁶ GFR estimated by this new CKD-EPI equation has been shown to categorize individuals’ clinical risk better than the MDRD equation.⁶ However, this CKD-EPI equation has not been validated in Asian populations. Several equations have been proposed from Eastern Asian countries. Therefore, studies are needed to compare those equations with respect to risk prediction among Asian populations. Several methodological challenges to compare GFR estimated by different equations with respect to risk prediction have been reported.⁶ In general, we anticipate that all equations will do better than serum creatinine alone, but risk related to absolute cutoffs, which simplify clinical care (for example, <60 ml min⁻¹ per 1.73 m²), will be quite sensitive to the equation used and calibration of the serum creatinine assay.

In summary, kidney dysfunction is a predictor not only of periprocedure acute kidney injury and mortality but also of long-term prognosis in patients undergoing coronary revascularization, even when they receive complete revascularization. For better evaluation of kidney function, eGFR, rather than a level of creatinine itself, should be used. CKD affects broad range of clinical aspects during and after coronary revascularization (Figure 1). Thus, physicians who are involved in either revascularization procedures or long-term care of CKD patients with coronary artery disease should be aware of short- and long-term impact of CKD and provide comprehensive management to these patients.

Figure 1

Issues related to CKD among patients undergoing coronary revascularization.