The Implications of Renal Impairment Among Patients Undergoing Percutaneous Coronary Intervention

ABSTRACT: The clinical implications of even mild impairment of renal function in percutaneous coronary interventions (PCI) have, to date, been poorly appreciated. The progressive loss of renal function is marked by the onset of a thrombo-inflammatory state not unlike what occurs in progressive coronary artery disease (CAD). In the presence of renal insufficiency, procedural success is reduced, peri-procedural myocardial infarction (MI) is more frequent and increased late mortality, myocardial ischemia and target vessel revascularization are more common. Vessel stenting reduces the rate of target vessel revascularization in renal failure patients, but fails to impact the frequency of peri-procedural ischemic events. Consequently, a rational, evidence-based approach to adjunctive anticoagulation in renal failure patients undergoing PCI is warranted if peri-procedural ischemic events are to be better managed. Unfortunately, current recommendations are hampered by a paucity of clinical trial evidence specifically examining the efficacy of available therapies. In an attempt to go beyond this lack of clinical evidence, some therapeutic insight may be drawn from the known hemostatic derangements associated with renal impairment. These changes include a well-described decrease in platelet aggregatory activity and an increase in thrombin generation. While more definitive clinical trial evidence of adjunctive anticoagulation efficacy among patients with renal impairment is in process, this article will briefly review the pathophysiological changes associated with renal impairment, as well as the evidence supporting each of the current therapeutic options available to the interventional cardiologist. Key words: bivalirudin, GP IIb/IIIa receptor inhibitors, low molecular weight heparin, percutaneous coronary interventions The substantial impact of renal impairment on both short- and long-term outcomes following percutaneous coronary intervention (PCI) has long been recognized. Reduced renal function is associated with an increased rate of peri-procedural bleeding, ischemic complications and target vessel revascularization.^1,2 How best to identify renal failure remains problematic, since the serum creatinine level is poorly representative of the true glomerular filtration rate (GFR). For example, substantial renal failure can exist in patients with a serum creatinine value in the 1.2–1.6 mg/dl range and otherwise go undetected unless the serum creatinine value is indexed by muscle mass. Although PCI trials have actively excluded renal failure patients, they have typically done so utilizing an elevated serum creatinine — that is, a serum creatinine value above the normal population reference range for normal values as reported on clinical chemistry panels — as the penultimate protocol exclusionary criterion. The limitations of this approach are quite obvious. Several reports suggest that up to a quarter of patients enrolled in PCI trials have significant impairment in renal function despite the relatively stringent exclusion criteria based on serum creatinine values. This proportion is even greater in registries that are more reflective of contemporary clinical experience, where patients are frequently older and have a greater prevalence of diabetes and hypertension and thereby are at increased risk for renal insufficiency.³ Additional factors contributing to excess bleeding and ischemic risk in this population include a greater atherosclerotic disease burden, complex lesion morphology, poor vascular integrity and the presence of a prothrombotic-proinflammatory state.⁴ These considerations, coupled with the well-described derangements in hemostatic function associated with renal impairment, call for rational, evidence-based approaches to antithrombotic therapy in patients undergoing PCI. Unfortunately, despite the numerous PCI clinical trials of novel antithrombotic therapies, few studies have specifically examined the efficacy of these therapies in patients with renal impairment. This article will review the current literature addressing outcomes in this high-risk patient population undergoing PCI, explore the pathophysiological derangements possibly underlying these adverse outcomes, and examine the current evidence supporting the use of the various antithrombotic therapies among patients undergoing percutaneous revascularization. Clinical outcomes following PCI among patients with renal impairment Following PCI, patients with renal dysfunction experience an increased rate of in-hospital mortality, myocardial infarction and/or target vessel revascularization. Similarly, long-term repeat revascularization rates in the range of 41–80% following balloon angioplasty have been reported.⁵ In addition, an increase in bleeding events and vascular complications correlates with the presence of reduced renal function.⁶ Among 23 chronic dialysis patients, Marso et al. observed a 17% incidence of abrupt vessel closure requiring emergent intervention; in this same series, event-free survival from recurrent ischemia was only 20%.⁷ This report confirms the adverse prognosis associated with severe renal impairment among patients undergoing PCI. More recently, in a cohort of 362 renal failure patients with pre-procedural creatinine levels of > 1.5 mg/dL (dialysis and non-dialysis patients), coronary stenting has been associated with improvements in long-term event-free survival, although this benefit of coronary stenting may come at the cost of increased in-hospital cardiac events when compared to angioplasty alone.⁸ Despite the confounding effect of reduced renal clearance on creatinine kinase (CK) levels among patients with renal dysfunction, an elevated CK following PCI was associated with a doubling in late mortality (CK elevation, 34.5% versus no CK elevation, 16.9%; p = 0.034) in a study of 190 patients with creatinine values > 2.0 mg/dL.9 Importantly, an elevated post-procedural CK was observed in 17.6% of patients. Hence, considered in tandem, these studies demonstrate that even with stenting, acute ischemic complications continue to be problematic among patients with renal impairment and these events correlate with long-term mortality. Even more compelling is the observational data of 5,031 patients undergoing PCI at the Mayo Clinic from 1994–1999. Highlighting the inadequacy of serum creatinine in detecting substantial reductions in renal function, a creatinine clearance of less than 70 ml/minute, calculated by the Cockroft-Gault formula, was observed in 49% of patients. Compared to patients with normal renal function (creatinine clearance > 70 ml/minute) patients with creatinine clearance 50–69 ml/minute, 30–49 ml/minute and 3 Consequently, the optimal use of adjunctive pharmacotherapies remains an important consideration during PCI if these outcomes are to be improved upon among patients with impaired renal function. Pathophysiological derangements.Renal dysfunction affects each of the essential components of the hemostatic process, namely platelets, the coagulation system and the endothelium (Table 1). These changes culminate in a combined risk of increased bleeding as well as the development of a prothrombotic state.^10,11 The level of hemostatic competence is also influenced by the degree of renal impairment, the underlying renal pathology and therein the level of proteinuria and to some extent, the mode of replacement treatment (peritoneal or hemodialysis). These considerations should be weighed in the selection of an antithrombotic regimen in a renal failure patient for whom a coronary intervention is contemplated. Thrombin and fibrinogen. Sensitive assays of thrombin generation document activation of the coagulation system in the patient with end-stage renal disease (ESRD).¹² Chronic uremia is associated with elevated plasma levels of prothrombin fragments F1+2, thrombin-antithrombin complex and fibrinopeptide A, and increased levels of D-dimer. Similarly, factor VII coagulant activity is also increased correlating with the increased levels of F1+2, as well as total and VLDL cholesterol and triglycerides.¹¹ Current evidence indicates that these increases are the result of greater rates of thrombin generation rather than reduced renal clearance. A reduced ability to inhibit endogenous thrombin formation is also observed in patients with ESRD. Moreover, antithrombin, an inhibitor of serine protease coagulation factors, is reduced in up to 60% of patients with renal disease, a finding that is most often seen in association with proteinuria. The competency of antithrombin is further compromised by the uremic environment, which reduces its activity. Of note, proteinuria itself has also been strongly associated with an increased cardiovascular event rate, particularly among diabetic patients.^13,14 Overall, the available evidence supports the observation that the prothrombotic state of renal disease is one characterized by excess thrombin generation, above and beyond the other hematologic derangements common to renal dysfunction. Platelet defects. Reduced renal function is associated with defects in platelet activation, degranulation, and ultimately aggregation.^{10,11,15–17} Although usually within the normal range, platelet counts are often lower among renal patients than in healthy controls suggesting excess consumption and/or underproduction of platelets,^18,19 and reduced platelet survival times have been demonstrated in dialysis-dependent patients. Platelet responses to agonists [such as adenosine diphosphate (ADP) and epinephrine] are also affected, with studies showing both a lower threshold for platelet activation of platelets and/or a reduced response.^18–21 However, uremia also leads to an acquired storage pool deficiency in platelets, with a reduction in thromboxane A2, ADP and serotonin reserves.²² Hence, the impaired platelet function of renal insufficiency is a condition wherein platelets are partially “spent” or “exhausted”. Table 1 lists the abnormalities in platelet function observed in patients with renal dysfunction. Other factors such as anemia, which is prevalent among renal failure patients, may also contribute to this characteristically poor platelet function. In this regard, Escolar et al. observed a partial return of platelet function when the hematocrit was increased in uremic patients.²³ The direct mechanisms leading to reduced platelet function among patients with renal impairment are probably related to retention of urea, creatinine, guanidosuccinic acid, phenolic acid ester and other similar middle molecular weight uremic toxins. This is supported by the observation that either peritoneal dialysis or hemodialysis can at least partially correct the platelet dysfunction and reverse the bleeding tendency, presumably by the dialytic clearance of these compounds.^10,11,17,24 Recently, other potential products directly affecting platelet function have been proposed. Similar to the small molecular weight peptides developed as potent inhibitors of the glycoprotein (GP) IIb/IIIa receptor, generation and accumulation of endogenous Arg-Gly-Asp (RGD) peptide-binding sequence-containing molecules has been suggested. These inhibitory peptides may account for many of the effects of uremic plasma on normal platelets.^17,25–27 Reduced shear-induced platelet aggregation mediated by decreased availability of GP IIb/IIIa receptors and changes in glycoprotein composition of the GP IIb/IIIa receptor may further affect platelet adhesion and aggregation among patients with renal dysfunction.²⁸Inflammation and atherogenesis. The role of inflammation in atherogenesis and/or plaque instability has been more carefully characterized over the last several years.²⁹ This accentuated inflammatory response, as well as persistent endothelial injury, may contribute to the thrombotic risk associated with uremia. Potentially, the central process resulting in increased thrombin and plasmin generation, as well as platelet dysfunction, are a platelet and endothelial cell response to chronic uremic injury. This hypothesis is supported by the observation that von Willebrand factor, which is secreted from injured endothelial cells and platelets, is elevated in conditions associated with inflammation and thrombosis, including renal dysfunction.³⁰ Possible mechanisms linking chronic injury of the endothelium in renal failure and defects in platelet function included increased production of prostacyclin and nitric oxide (NO). Furthermore, a heightened monocyte response has been seen among these patients. ESRD is associated with elevated levels of interleukin-1 (IL-1), tumor necrosis factor-a (TNFa), fibrinogen, monocyte surface expression tissue factor, and other adhesion molecules.³¹ These changes appear to correlate with elevated prothrombin fragments, F1+2 and lipoprotein (a) levels in these patients. Adding to these thrombo-inflammatory derangements, uremic patients also appear to have differences in coronary atherosclerotic plaque morphology. In particular, coronary arteries among these patients have a greater degree of intimal and medial thickening, a higher degree of calcification and overall lower luminal area.⁴ These morphological factors contribute to the more complex nature of these lesions and a considerably increased peri-procedural risk. These changes appear to link plaque progression, thrombosis and inflammation and may contribute to the excess risk seen among renal impairment patients undergoing PCI. Rational approach to adjunctive pharmacotherapy Clinical outcome data in renal failure patients undergoing PCI, which address the risk-benefit ratio of many of the currently used antithrombotic therapies, has been slow to come forward. In this regard, several clinical trials, particularly those employing the small molecular weight GP IIb/IIIa inhibitors, have intentionally excluded patients with substantial elevations in serum creatinine, since these agents are eliminated largely by renal clearance. Until the necessary supporting clinical trial experience becomes available, the known biologic perturbations in hemostasis associated with renal failure/ESRD can provide insight into what may be preferred approaches to peri-procedural anticoagulation in these high-risk patients. Antiplatelet therapy. To date, clinical trials specifically addressing the efficacy of aspirin among renal impaired patients undergoing PCI have not been performed. Aspirin is both hepatically metabolized and renally excreted as the intact molecule, with the latter representing as much as 35% of its systemic clearance. Despite the fact that aspirin undergoes some degree of renal elimination, its potential for accumulation in the renal failure patient is minimized by the alternative hepatic elimination pathway. In recognition of the efficacy of aspirin in preventing acute ischemic events and the relative lack of associated bleeding complications, the routine use of aspirin without dose adjustment can be advocated in this population (Table 2). Aspirin is significantly dialyzed; therefore, post-dialysis dosing is advisable in the patient requiring a consistent aspirin effect in the interdialytic period. Similarly, the thienopyridines, clopidogrel and ticlopidine, undergo hepatic metabolism and therefore dose adjustment is not required in patients with renal impairment (Table 2). Among patients in the PCI-CURE (Clopidogrel in Unstable Angina to prevent Recurrent ischemic Events) trial, pretreatment with clopidogrel for a median of 6 days with variable clopidogrel dosing post-procedure was associated with a 31% reduction in cardiovascular death or myocardial infarction without any significant increase in bleeding risk.³² Further conceptual support for peri-interventional clopidogrel use comes from the observation that it exhibits some anti-inflammatory activity.^33,34 Noteworthy in the PCI-CURE trial was the fact that renal impairment was not a specific exclusion criterion. Although these data were not specifically analyzed by level of renal function, certain inferences can be made directed toward the renal failure population. First, considering the observed benefit of clopidogrel in the overall PCI-CURE study population and the fact that mild-to-moderate renal failure is common in such a cohort (cohort average age, 61.6 ± 11.2 years; 19% diabetic; 27.3% with a prior myocardial infarction), it is likely that renal failure patients saw similar benefits as the overall study population. Second, the fact that clopidogrel is hepatically and not renally eliminated simplifies dosing consideration in the renal failure population; accordingly, clopidogrel therapy should be strongly considered in a renal failure patient undergoing PCI. Substantial clinical evidence supports the use of the GP IIb/IIIa inhibitors among patients undergoing PCI. Several trials have now demonstrated a 35–50% reduction in 30-day death, myocardial infarction or need for urgent revascularization among patients treated with these agents when compared with heparin alone.³⁵ Yet, despite this wealth of data, relatively little is known about the relative safety and efficacy of these agents in PCI patients with impairment of renal function. Nevertheless, current dosing guidelines have been modified in light of the known pharmacological characteristics of these compounds pending outcome data, which might offer insight into either dose or concentration-response relationships in the renal failure population. Specifically, the currently available small molecule GP IIb/IIIa antagonists, eptifibatide and tirofiban, are largely eliminated via renal excretion; therefore, dose adjustment among patients with severe renal insufficiency (creatinine clearance of less than 30 ml/minute) is recommended (Table 2). The dosing recommendations for these compounds can be overly broad. For example, the recommended serum creatinine values at which eptifibatide dose is reduced by 50% are between 2–4 mg/dl with no additional information available in more severe renal impairment. A serum creatinine range such as this, however, includes significant numbers of severe renal failure patients, particularly in those at the upper limits of this serum creatinine range. Factors such as these make both safe and effective dosing of glycoprotein IIb/IIIa inhibitors quite problematic in the renal failure patient. Finally, it is important to note that trials with these agents in PCI have excluded patients with an elevated serum creatinine value at baseline; consequently, safety and efficacy data with these agents among patients with substantial renal impairment are presently not available. Similarly, the Enhanced Suppression of the Platelet Glycoprotein IIb/IIIa Receptor Using Integrilin Therapy trial (ESPIRIT), Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy (PURSUIT) trial and the Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis (RESTORE) trial examined the relative benefit and risk of these agents over heparin alone and have yet to publish subset analyses in those patients with reduced renal function in these studies. Interestingly, among patients with acute coronary syndromes, evidence with the small molecule oral glycoprotein IIb/IIIa inhibitors suggests that renal impairment increases the risk of bleeding, while very high levels of these agents may be associated with toxicity.^36,37 Nevertheless, attempts to tailor the dose of these agents according to the level of renal function creatinine clearance have failed to improve safety or efficacy with lamifiban or the oral agents, sibrafiban and lotrafiban. Alternatively, abciximab is widely used among patients with renal impairment in whom GP IIb/IIIa inhibition is considered warranted. Being a monoclonal antibody fragment, this agent undergoes almost no renal excretion and is eliminated through platelet degradation by the reticuloendothelial system. Within trials of abciximab in PCI, such as the Evaluation of c7E3 for the Prevention of Ischemic Complications (EPIC) trial; Evaluation in Percutaneous Transluminal Coronary Angioplasty to Improve Long-term Outcome With Abciximab Glycoprotein IIb/IIIa Blockade (EPILOG) trial; and the Evaluation of IIb/IIIa Platelet Inhibitor for Stenting (EPISTENT) trial, renal function was not a reason for exclusion. Specific efficacy and safety data among patients with marked reductions in creatinine clearance are not available from these trials. However, an analysis of the Mayo Clinic registry of 1,146 patients demonstrated no “interaction” between abciximab, ischemic or bleeding events and creatinine clearance on a multiplicative scale.³⁹ While models of interaction are relatively insensitive to differences in risk and benefit across strata, these data would seem to suggest similar efficacy with abciximab in patients with reduced renal function. Nevertheless, considering these data in conjunction with the potential for GP IIb/IIIa inhibitory molecules associated with substantial renal impairment, cautious and individualized use of this agent is recommended in the renal failure patient. When these agents are used, it is prudent that platelet aggregation be measured using point-of-care testing. Thrombin inhibition Considering the pathological derangements associated with renal impairment, including evidence of ongoing thrombin generation, effective thrombin inhibition is a pertinent consideration in renal failure patients undergoing PCI. Heparin therapy, dating to the late 1970s, has been the traditional antithrombin agent for PCI. Despite its long history, robust evidence supporting the dosing of this agent in PCI is lacking and evidence of its safety and effectiveness among patients with renal impairment is similarly limited. The half-life of the anticoagulant effect of heparin is reported to be approximately 1.5 hours and it does not increase with dose when assessed with a wide-range of anticoagulant tests. Alternatively, the plasma half-life of bioassayed heparin increases with increasing relative dose (dose per unit body weight) with a half-life of approximately 60 minutes with the doses used during PCI.⁴⁰ Nevertheless, despite clear evidence of increased bleeding complications with heparin among patients with impairment of renal function, no dose adjustment has been suggested. Importantly, heparin therapy is dependent upon antithrombin levels for its anticoagulant effect. Consequently, attenuation in efficacy may be expected among patients with renal impairment where levels of this factor are diminished in association with proteinuria. Evidence of the association between renal impairment, adverse events, and less than optimum outcomes can be found in a randomized trial of 4,312 patients comparing bivalirudin with heparin in angioplasty patients. Despite high doses of heparin therapy (175 units/kg bolus targeting activated clotting time levels of > 350 seconds), a progressive increase in ischemic complications was observed with declining renal function. Thus, considering these relatively high doses compared to currents standards, it would appear that heparin therapy is unable to adequately ameliorate the thrombo-inflammatory milieu associated with renal impairment. Consequently, more effective antithrombin agents are needed. Recently, low molecular weight heparins have emerged as useful therapies among patients with acute coronary syndromes and are now the focus of trials within the catheterization laboratory. However, among patients with renal dysfunction, these agents share some of the limitations associated with unfractionated heparin. Specifically, the low molecular weight heparins undergo substantial renal clearance with the apparent clearance of enoxaparin, for example, reduced by 30% in patients with severe renal impairment. Consequently, dose adjustment is required (Table 2). In the absence of safety and efficacy data with these agents among patients with renal dysfunction, particular caution should be exercised in the elderly and patient with low body weight (less than 45 kg), where moderate impairment in renal function may be poorly reflected in the serum creatinine. As a small molecule direct thrombin inhibitor, bivalirudin undergoes substantial renal excretion with dose adjustment being required in patients with moderate to severe renal impairment (creatinine clearance 43 As mentioned earlier, stratification of these patients by renal function was associated with a progressive increase in ischemic events in the heparin-treated group coupled with a striking increase in bleeding. In contrast, bivalirudin was associated with a lesser gradation of risk in bleeding or ischemia across the strata of creatinine clearance. Therefore, while dedicated prospective confirmation of this finding is warranted, these findings provide evidence that bivalirudin maintains its efficacy and safety profile among patients with renal impairment (Figure 1) and its use should be given strong consideration in patients with renal impairment who are undergoing PCI. Other direct thrombin inhibitors that are currently available include lepirudin (a recombinant form of hirudin) and argatroban. Clinical outcome data supporting the use of lepirudin among patients with renal impairment have not been published; however, the use of lepirudin is problematic due to its long half-life and substantial renal excretion, potentially compounding the bleeding risk associated with this agent. Consequently, prolonged anticoagulation with this agent can be expected, suggesting that lepirudin is inappropriate in this setting. Alternatively, argatroban is metabolized within the liver and undergoes almost no renal excretion.⁴⁴ Argatroban anticoagulation, compared with historical control subjects, improves clinical outcomes in patients who have heparin-induced thrombocytopenia without increasing bleeding risk.⁴⁵ There are no outcome data with this agent for renal dysfunction patients undergoing routine PCI. Consequently, based on the weight of evidence for argatroban, there are few data to support the use of this agent in patients with renal impairment undergoing PCI. Conclusions The pharmacological treatment of patients with substantial renal impairment presenting for coronary intervention remains unsettled. The excess bleeding and ischemic risk observed in this patient population complicates the choice of an anticoagulant regimen. Even as a regimen is established, in many cases the altered pharmacokinetics and/or pharmacodynamics of the agents used makes safe dosing a challenge in renal failure. In the case of the easily recognized renal failure patient — such as one undergoing maintenance hemodialysis — the need to concentrate on appropriately adjusting the therapeutic regimen is obvious. Such patients, however, represent but a small fraction of the overall renal failure population, many of whom go unrecognized when renal function is estimated based on conventional measures of renal function, such as a serum creatinine value. An additional important consideration is that the use of stenting, which is a common procedure in renal failure patients, does not lessen the need for optimizing pharmacotherapy. While clinical trial and registry data supporting the use of most therapies in this patient group are generally lacking, in consideration of the multiple derangements in platelet and thrombin biology present, a greater emphasis on thrombin inhibition seems prudent in the renal failure patient. Furthermore, despite the ongoing nature of the data acquisition with the direct thrombin inhibitor bivalirudin, it appears to offer certain advantages when used in the renal failure patient undergoing a PCI. Therefore, given the risk profile of these patients and the limited supporting data with most available therapies, individualization of treatment choices is warranted, with the balance between bleeding risk and suppressing post-procedure ischemic events to be carefully weighed with each agent.