CLINICAL INVESTIGATION
Renal Dysfunction and hsCRP Predict Long-term Outcomes of Percutaneous Coronary Intervention in Acute Myocardial Infarction Chong-hui Wang, MD, Shu-yang Zhang, MD, Quan Fang, MD, Zhu-jun Shen, MD, Zhong-jie Fan, MD, Xiao-feng Jin, MD, Yong Zeng, MD, Zhen-yu Liu, MD and Hong-zhi Xie, MD
Abstract: Background: This study assessed the combined utility of estimated glomerular filtration rate (eGFR) and serum high-sensitivity C-reactive protein (hsCRP) levels to predict long-term mortality and cardiovascular outcomes of patients with acute ST-segment elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (PCI). Elevated CRP levels and renal dysfunction have both been shown to independently and jointly predict mortality and cardiovascular outcomes after PCI in the short term. However, long-term results in patients with acute STEMI undergoing PCI have not been reported. Methods: A total of 262 patients with acute STEMI undergoing primary PCI were classified at admission into quartiles according to eGFR (,60, 60–70, 70–80 and $80 mL$min21$1.73 m22) and hsCRP (,3 and $3 mg/L). Mortality, nonfatal myocardial infarction (MI) and major adverse cardiac events (MACEs) were compared among the groups. Results: During a median follow-up of 48.3 months, the composite of all-cause mortality and nonfatal MI (mortality + MI) was significantly higher (35.09%) in the group with the lowest eGFR compared with that of the other 3 eGFR groups (14.29%, 3.77% and 9.43%, respectively, P , 0.0001) and the group with elevated hsCRP (34.29%) versus that with hsCRP ,3 mg/L (4.41%, P , 0.0001). A combined analysis showed an exaggerated hazard in patients with the lowest eGFR and highest hsCRP (hazard ratio: 44.658; 95% confidence interval: 5.955–111.890). Conclusions: Renal dysfunction and elevated hsCRP predict a high long-term incidence of MACE in patients with acute STEMI undergoing primary PCI, with the combination being of prognostic significance for long-term mortality and MI in these patients. Key Indexing Terms: High-sensitivity C-reactive protein; Estimated glomerular filtration rate; Mortality; Major adverse cardiac events. [Am J Med Sci 2015;349(5):413–420.]
R
enal dysfunction and vascular inflammation are major risk factors for cardiovascular (CV) diseases. Estimated glomerular filtration rate (eGFR) and high-sensitivity C-reactive protein (hsCRP) serve as easily measured indicators of renal dysfunction and vascular inflammation, respectively, and are closely related to the prognosis of CV diseases in both the general population1–3 and in patients with preexisting CV disease.4–11 In patients with acute coronary syndromes (ACS), hsCRP can predict mortality and a range of CV outcomes.12,13 Renal dysfunction is associated with an increased risk of death and CV events in patients with a variety of CV diseases, including ACS,4,6,12,14 acute myocardial infarction (AMI),5,7,15–18 From the Department of Cardiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Submitted August 5, 2014; accepted in revised form December 31, 2014. The authors have no financial or other conflicts of interest to disclose. Correspondence: Chong-hui Wang, MD, Department of Cardiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China (E-mail: [email protected]).
The American Journal of the Medical Sciences
coronary artery disease with subsequent percutaneous coronary intervention (PCI)8,19–21 and heart failure.9,10,22 Understanding the prognostic significance of renal dysfunction and elevated CRP in patients with CV diseases may be important for guiding appropriate treatment. Although renal dysfunction is common in patients with heart failure, these patients may not receive the medications they need.9 Its presence may also hamper a doctor’s ability to decide whether to perform primary PCI in patients with AMI who may benefit from this procedure. Moderate or severe renal insufficiency was found to be associated with unfavorable long-term outcomes in patients undergoing successful PCI with drug-eluting stents (DESs).21 Decreased GFR has also been shown to be an independent predictor of in-hospital mortality in patients with ST-segment elevation myocardial infarction (STEMI) undergoing primary PCI.15 Despite these findings, not much is known regarding the long-term prognostic significance of renal dysfunction and elevated hsCRP in patients with acute STEMI undergoing PCI. Acute STEMI has a long-term mortality of almost 30%.23 When available, PCI is widely recognized as the most effective treatment for acute STEMI.24,25 For patients undergoing PCI, both CRP levels and indicators of kidney function have been shown to provide independent markers of risk of long-term mortality, with high levels of ultrasensitive CRP and low eGFR found to significantly negatively affect survival after PCI.26 Although a number of other studies have shown levels of hsCRP and measures of renal dysfunction to be independent predictors of mortality and CV outcomes, only a few studies have shown the combination to be of prognostic significance in patients undergoing PCI.12,19,26 These studies, however, had a short follow-up and were performed in a heterogeneous patient population. The authors thus sought to extend these findings by evaluating whether renal dysfunction and elevated hsCRP could jointly predict long-term (.3 years) mortality, nonfatal myocardial infarction (MI) and major adverse cardiac events (MACEs) in patients exclusively with acute STEMI undergoing primary PCI.
METHODS Study Population This was a retrospective study. From January 2005 to December 2007, 312 patients with acute STEMI underwent primary PCI in Peking Union Medical College Hospital. Individuals eligible for primary PCI fulfilled all of the following inclusion criteria: chest pain for .30 minutes, ST-segment elevation of $1 mm in at least 2 contiguous electrocardiograph leads or with new left bundle-branch block and hospital admission within 12 hours from symptom onset or between 12 and 24 hours from symptom onset with worsened ischemic chest pain. All patients fulfilled current diagnostic criteria for AMI, although there were 2 cases with new left bundle-branch block
Volume 349, Number 5, May 2015
413
Wang et al
that had undergone PCI. However, these patients had elevated biomarkers of AMI, including cardiac troponin I (CTnI) and creatine kinase MB isoenzyme (CK-MB). For patients undergoing more than 1 procedure, only the 1st was used for analysis. Patients lost to follow-up or having incomplete information on hsCRP or serum creatinine levels were excluded from the study. Clinical follow-up data were obtained by a dedicated registry data manager for outpatient records or telephone interviews. The study was approved by the institutional review board of our hospital. Written informed consent was obtained from every patient. eGFR and hsCRP Patients’ blood was collected for serum creatinine analysis at admission, within 60 to 90 minutes before PCI. Serum creatinine was measured using the picric acid method with an Olympus 5400 analyzer (Tokyo, Japan). eGFR was used as an indicator of renal dysfunction and calculated using the Modification of Diet in Renal Disease equation.27 eGFR baselines were available in 251 of the 262 patients, and patients were grouped into quartiles according to their baseline eGFR (,60, 60–70, 70–80, $80 mL$min21$1.73 m22). Patients’ blood was collected for serum hsCRP analysis within 24 hours after PCI. Serum hsCRP levels were measured using a turbidimetric immunoassay with an Olympus 5400 analyzer. Patients were subdivided on the basis of cutoffs of hsCRP known to predict CV risk in the general population (,3 and $3 mg/L).28 Baseline levels of hsCRP were available in 241 of the 262 patients. Definitions and Follow-up The primary outcome was the composite of cumulative all-cause mortality and nonfatal MI. Secondary outcomes were the occurrence of MACE, defined as the composite of all-cause death, nonfatal MI and target vessel revascularization. MI was defined as ischemic chest pain with either a new ST segment and T wave changes or recurrent elevation of creatine kinase (CK) more than 2 times the upper limit of normal range, with an increase in CK-MB. Target vessel revascularization was defined as any reintervention (PCI or coronary artery bypass grafting) performed on the treated vessel, driven by ischemia and with documentation of lesions with .50% diameter stenosis. Definite stent thrombosis was defined as angiographically defined thrombosis with thrombolysis in myocardial infarction flow grade 0 or 1 or the presence of flow-limiting thrombus of the previously stented infarction-related artery. Clinical follow-up was performed at our outpatient clinic or by telephone interview every 6 months. Statistical Analysis All values are expressed as mean 6 standard deviation or as numbers (percentages). Baseline values between groups were compared with analysis of variance for continuous variables and the x2 test for categorical data. Event-free survival curves were obtained using the Kaplan-Meier method, and outcomes were compared using the log-rank test. Baseline factors with a P value ,0.05 were then entered into a forward stepwise multivariate Cox proportional hazards model. Independent predictors of all-cause mortality and MI were identified using a Cox proportional hazards regression model. All statistical analyses were performed using SAS software (version 8.2; Cary, NC); P value ,0.05 was considered statistically significant.
414
RESULTS eGFR as an Independent Predictor of Long-term Outcomes After PCI Of 262 patients enrolled for analysis, 145 received DES and 117 received bare-metal stents. Patients were divided into quartiles according to baseline eGFR (mL$min21$1.73 m22) (Q1: ,60 [n 5 57], Q2: 60–70 [n 5 35], Q3: 70–80 [n 5 53], Q4: $80 [n 5 106]). Table 1 shows patients’ demographic and baseline characteristics. There were no significant differences among all groups in terms of ACS risk factors (smoking, obesity, diabetes, hypertension and hyperlipidemia), serum levels of cardiac infarction biomarkers, angiographic characteristics, PCI procedure (DES rate) and the timing of PCI reflected in symptomonset-to balloon-time (P . 0.05). However, patients in eGFR Q1 were significantly older and had more severe cardiac dysfunction reflected by poorer left ventricular ejection fraction (LVEF) compared with patients in the other 3 eGFR quartiles (Table 1). None of the patients in the study had received dialysis before PCI. Within 2 weeks after PCI, no patients needed dialysis. Over the next several years, 3 patients needed dialysis because of diabetic or hypertensive nephropathy, which did not have any relationship with the PCI. During the median follow-up of 48.3 months, patients in eGFR Q1 showed a significantly higher incidence of all-cause mortality and nonfatal MI (35.09%) than that of patients in eGFR Q2 (14.29%), Q3 (3.77%) or Q4 (9.43%) (P , 0.0001). The occurrence of MACE was also significantly higher in patients in eGFR Q1 (42.11%) than in eGFR Q2 (17.14%), Q3 (11.32%) or Q4 (16.98%) (P 5 0.0002) (Table 2). The mortality rate of patients in eGFR Q1 was significantly higher than that of patients in the other eGFR quartiles during the follow-up period (P , 0.0001). Survival analysis with the log-rank test indicated that the mortality and MI rates of patients in eGFR Q1 were significantly higher than those of patients in eGFR Q4 during the follow-up period (P , 0.0001) (Figure 1A). hsCRP as an Independent Predictor of Long-term Outcomes After PCI The authors divided patients into 2 groups according to hsCRP (mg/L) (low: ,3 [n 5 136] and high: $3 [n 5 105]; Table 3). There were no significant differences among groups in terms of ACS risk factors, angiographic characteristics and timing of PCI (symptom-onset-to balloon-time) (P . 0.05). In contrast to the eGFR categorization analysis, there were no sex differences according to levels of hsCRP, although older patients had a higher level of hsCRP (P , 0.05) than did younger patients. Interestingly, high levels of hsCRP were correlated with low eGFR (P , 0.05) (Table 3). Long-term outcomes on the basis of hsCRP are shown in Table 4. During a median follow-up of 48.3 months, the rate of cumulative all-cause mortality and MI was significantly higher in patients with hsCRP $3 mg/L compared with patients in the low hsCRP group (34.29% versus 4.41%, respectively, P , 0.0001). The occurrence of MACE was also significantly higher in the group of patients with hsCRP $3 mg/L than in the group with low hsCRP (47.62% versus 6.62%, respectively, P , 0.0001). Mortality of patients with high hsCRP was significantly higher than that of patients with low hsCRP during the follow-up period (P , 0.0001). Survival analysis with the log-rank test indicated that the mortality and MI rates of patients with high hsCRP were significantly higher than those of patients with low hsCRP during the follow-up period (P , 0.0001) (Figure 1B). Volume 349, Number 5, May 2015
eGFR/hsCRP Jointly Predict PCI Outcomes
TABLE 1. Baseline patient characteristics stratified by eGFR eGFR (mL$min21$1.73 m22)
Renal Dysfunction and hsCRP Predict Long-term Outcomes of Percutaneous Coronary Intervention in Acute Myocardial Infarction Chong-hui Wang, MD, Shu-yang Zhang, MD, Quan Fang, MD, Zhu-jun Shen, MD, Zhong-jie Fan, MD, Xiao-feng Jin, MD, Yong Zeng, MD, Zhen-yu Liu, MD and Hong-zhi Xie, MD
Abstract: Background: This study assessed the combined utility of estimated glomerular filtration rate (eGFR) and serum high-sensitivity C-reactive protein (hsCRP) levels to predict long-term mortality and cardiovascular outcomes of patients with acute ST-segment elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (PCI). Elevated CRP levels and renal dysfunction have both been shown to independently and jointly predict mortality and cardiovascular outcomes after PCI in the short term. However, long-term results in patients with acute STEMI undergoing PCI have not been reported. Methods: A total of 262 patients with acute STEMI undergoing primary PCI were classified at admission into quartiles according to eGFR (,60, 60–70, 70–80 and $80 mL$min21$1.73 m22) and hsCRP (,3 and $3 mg/L). Mortality, nonfatal myocardial infarction (MI) and major adverse cardiac events (MACEs) were compared among the groups. Results: During a median follow-up of 48.3 months, the composite of all-cause mortality and nonfatal MI (mortality + MI) was significantly higher (35.09%) in the group with the lowest eGFR compared with that of the other 3 eGFR groups (14.29%, 3.77% and 9.43%, respectively, P , 0.0001) and the group with elevated hsCRP (34.29%) versus that with hsCRP ,3 mg/L (4.41%, P , 0.0001). A combined analysis showed an exaggerated hazard in patients with the lowest eGFR and highest hsCRP (hazard ratio: 44.658; 95% confidence interval: 5.955–111.890). Conclusions: Renal dysfunction and elevated hsCRP predict a high long-term incidence of MACE in patients with acute STEMI undergoing primary PCI, with the combination being of prognostic significance for long-term mortality and MI in these patients. Key Indexing Terms: High-sensitivity C-reactive protein; Estimated glomerular filtration rate; Mortality; Major adverse cardiac events. [Am J Med Sci 2015;349(5):413–420.]
R
enal dysfunction and vascular inflammation are major risk factors for cardiovascular (CV) diseases. Estimated glomerular filtration rate (eGFR) and high-sensitivity C-reactive protein (hsCRP) serve as easily measured indicators of renal dysfunction and vascular inflammation, respectively, and are closely related to the prognosis of CV diseases in both the general population1–3 and in patients with preexisting CV disease.4–11 In patients with acute coronary syndromes (ACS), hsCRP can predict mortality and a range of CV outcomes.12,13 Renal dysfunction is associated with an increased risk of death and CV events in patients with a variety of CV diseases, including ACS,4,6,12,14 acute myocardial infarction (AMI),5,7,15–18 From the Department of Cardiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Submitted August 5, 2014; accepted in revised form December 31, 2014. The authors have no financial or other conflicts of interest to disclose. Correspondence: Chong-hui Wang, MD, Department of Cardiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China (E-mail: [email protected]).
The American Journal of the Medical Sciences
coronary artery disease with subsequent percutaneous coronary intervention (PCI)8,19–21 and heart failure.9,10,22 Understanding the prognostic significance of renal dysfunction and elevated CRP in patients with CV diseases may be important for guiding appropriate treatment. Although renal dysfunction is common in patients with heart failure, these patients may not receive the medications they need.9 Its presence may also hamper a doctor’s ability to decide whether to perform primary PCI in patients with AMI who may benefit from this procedure. Moderate or severe renal insufficiency was found to be associated with unfavorable long-term outcomes in patients undergoing successful PCI with drug-eluting stents (DESs).21 Decreased GFR has also been shown to be an independent predictor of in-hospital mortality in patients with ST-segment elevation myocardial infarction (STEMI) undergoing primary PCI.15 Despite these findings, not much is known regarding the long-term prognostic significance of renal dysfunction and elevated hsCRP in patients with acute STEMI undergoing PCI. Acute STEMI has a long-term mortality of almost 30%.23 When available, PCI is widely recognized as the most effective treatment for acute STEMI.24,25 For patients undergoing PCI, both CRP levels and indicators of kidney function have been shown to provide independent markers of risk of long-term mortality, with high levels of ultrasensitive CRP and low eGFR found to significantly negatively affect survival after PCI.26 Although a number of other studies have shown levels of hsCRP and measures of renal dysfunction to be independent predictors of mortality and CV outcomes, only a few studies have shown the combination to be of prognostic significance in patients undergoing PCI.12,19,26 These studies, however, had a short follow-up and were performed in a heterogeneous patient population. The authors thus sought to extend these findings by evaluating whether renal dysfunction and elevated hsCRP could jointly predict long-term (.3 years) mortality, nonfatal myocardial infarction (MI) and major adverse cardiac events (MACEs) in patients exclusively with acute STEMI undergoing primary PCI.
METHODS Study Population This was a retrospective study. From January 2005 to December 2007, 312 patients with acute STEMI underwent primary PCI in Peking Union Medical College Hospital. Individuals eligible for primary PCI fulfilled all of the following inclusion criteria: chest pain for .30 minutes, ST-segment elevation of $1 mm in at least 2 contiguous electrocardiograph leads or with new left bundle-branch block and hospital admission within 12 hours from symptom onset or between 12 and 24 hours from symptom onset with worsened ischemic chest pain. All patients fulfilled current diagnostic criteria for AMI, although there were 2 cases with new left bundle-branch block
Volume 349, Number 5, May 2015
413
Wang et al
that had undergone PCI. However, these patients had elevated biomarkers of AMI, including cardiac troponin I (CTnI) and creatine kinase MB isoenzyme (CK-MB). For patients undergoing more than 1 procedure, only the 1st was used for analysis. Patients lost to follow-up or having incomplete information on hsCRP or serum creatinine levels were excluded from the study. Clinical follow-up data were obtained by a dedicated registry data manager for outpatient records or telephone interviews. The study was approved by the institutional review board of our hospital. Written informed consent was obtained from every patient. eGFR and hsCRP Patients’ blood was collected for serum creatinine analysis at admission, within 60 to 90 minutes before PCI. Serum creatinine was measured using the picric acid method with an Olympus 5400 analyzer (Tokyo, Japan). eGFR was used as an indicator of renal dysfunction and calculated using the Modification of Diet in Renal Disease equation.27 eGFR baselines were available in 251 of the 262 patients, and patients were grouped into quartiles according to their baseline eGFR (,60, 60–70, 70–80, $80 mL$min21$1.73 m22). Patients’ blood was collected for serum hsCRP analysis within 24 hours after PCI. Serum hsCRP levels were measured using a turbidimetric immunoassay with an Olympus 5400 analyzer. Patients were subdivided on the basis of cutoffs of hsCRP known to predict CV risk in the general population (,3 and $3 mg/L).28 Baseline levels of hsCRP were available in 241 of the 262 patients. Definitions and Follow-up The primary outcome was the composite of cumulative all-cause mortality and nonfatal MI. Secondary outcomes were the occurrence of MACE, defined as the composite of all-cause death, nonfatal MI and target vessel revascularization. MI was defined as ischemic chest pain with either a new ST segment and T wave changes or recurrent elevation of creatine kinase (CK) more than 2 times the upper limit of normal range, with an increase in CK-MB. Target vessel revascularization was defined as any reintervention (PCI or coronary artery bypass grafting) performed on the treated vessel, driven by ischemia and with documentation of lesions with .50% diameter stenosis. Definite stent thrombosis was defined as angiographically defined thrombosis with thrombolysis in myocardial infarction flow grade 0 or 1 or the presence of flow-limiting thrombus of the previously stented infarction-related artery. Clinical follow-up was performed at our outpatient clinic or by telephone interview every 6 months. Statistical Analysis All values are expressed as mean 6 standard deviation or as numbers (percentages). Baseline values between groups were compared with analysis of variance for continuous variables and the x2 test for categorical data. Event-free survival curves were obtained using the Kaplan-Meier method, and outcomes were compared using the log-rank test. Baseline factors with a P value ,0.05 were then entered into a forward stepwise multivariate Cox proportional hazards model. Independent predictors of all-cause mortality and MI were identified using a Cox proportional hazards regression model. All statistical analyses were performed using SAS software (version 8.2; Cary, NC); P value ,0.05 was considered statistically significant.
414
RESULTS eGFR as an Independent Predictor of Long-term Outcomes After PCI Of 262 patients enrolled for analysis, 145 received DES and 117 received bare-metal stents. Patients were divided into quartiles according to baseline eGFR (mL$min21$1.73 m22) (Q1: ,60 [n 5 57], Q2: 60–70 [n 5 35], Q3: 70–80 [n 5 53], Q4: $80 [n 5 106]). Table 1 shows patients’ demographic and baseline characteristics. There were no significant differences among all groups in terms of ACS risk factors (smoking, obesity, diabetes, hypertension and hyperlipidemia), serum levels of cardiac infarction biomarkers, angiographic characteristics, PCI procedure (DES rate) and the timing of PCI reflected in symptomonset-to balloon-time (P . 0.05). However, patients in eGFR Q1 were significantly older and had more severe cardiac dysfunction reflected by poorer left ventricular ejection fraction (LVEF) compared with patients in the other 3 eGFR quartiles (Table 1). None of the patients in the study had received dialysis before PCI. Within 2 weeks after PCI, no patients needed dialysis. Over the next several years, 3 patients needed dialysis because of diabetic or hypertensive nephropathy, which did not have any relationship with the PCI. During the median follow-up of 48.3 months, patients in eGFR Q1 showed a significantly higher incidence of all-cause mortality and nonfatal MI (35.09%) than that of patients in eGFR Q2 (14.29%), Q3 (3.77%) or Q4 (9.43%) (P , 0.0001). The occurrence of MACE was also significantly higher in patients in eGFR Q1 (42.11%) than in eGFR Q2 (17.14%), Q3 (11.32%) or Q4 (16.98%) (P 5 0.0002) (Table 2). The mortality rate of patients in eGFR Q1 was significantly higher than that of patients in the other eGFR quartiles during the follow-up period (P , 0.0001). Survival analysis with the log-rank test indicated that the mortality and MI rates of patients in eGFR Q1 were significantly higher than those of patients in eGFR Q4 during the follow-up period (P , 0.0001) (Figure 1A). hsCRP as an Independent Predictor of Long-term Outcomes After PCI The authors divided patients into 2 groups according to hsCRP (mg/L) (low: ,3 [n 5 136] and high: $3 [n 5 105]; Table 3). There were no significant differences among groups in terms of ACS risk factors, angiographic characteristics and timing of PCI (symptom-onset-to balloon-time) (P . 0.05). In contrast to the eGFR categorization analysis, there were no sex differences according to levels of hsCRP, although older patients had a higher level of hsCRP (P , 0.05) than did younger patients. Interestingly, high levels of hsCRP were correlated with low eGFR (P , 0.05) (Table 3). Long-term outcomes on the basis of hsCRP are shown in Table 4. During a median follow-up of 48.3 months, the rate of cumulative all-cause mortality and MI was significantly higher in patients with hsCRP $3 mg/L compared with patients in the low hsCRP group (34.29% versus 4.41%, respectively, P , 0.0001). The occurrence of MACE was also significantly higher in the group of patients with hsCRP $3 mg/L than in the group with low hsCRP (47.62% versus 6.62%, respectively, P , 0.0001). Mortality of patients with high hsCRP was significantly higher than that of patients with low hsCRP during the follow-up period (P , 0.0001). Survival analysis with the log-rank test indicated that the mortality and MI rates of patients with high hsCRP were significantly higher than those of patients with low hsCRP during the follow-up period (P , 0.0001) (Figure 1B). Volume 349, Number 5, May 2015
eGFR/hsCRP Jointly Predict PCI Outcomes
TABLE 1. Baseline patient characteristics stratified by eGFR eGFR (mL$min21$1.73 m22)
