ORIGINAL ARTICLE
Evidence for Preoperative Aspirin Improving Major Outcomes in Patients With Chronic Kidney Disease Undergoing Cardiac Surgery A Cohort Study Linong Yao, MD, PhD,∗ † Nilas Young, MD,‡ Hong Liu, MD,§ Zhongmin Li, PhD,¶ Will Sun, MS,|| Jordan Goldhammer, MD,∗ Lei Tao, MD,† Jianbin He, MD,† James Diehl, MD,∗∗ and Jianzhong Sun, MD, PhD∗ Background: Effects of aspirin on patients with chronic kidney disease (CKD) remains unclear. This study aimed to examine the effect of preoperative aspirin use on postoperative renal function and 30-day mortality in patients with CKD undergoing cardiac surgery. Methods: A retrospective cohort study was performed on consecutive patients (n = 5175) receiving cardiac surgery in 2 tertiary hospitals. Of all patients, 3585 met the inclusion criteria and underwent the analysis to determine the association of preoperative aspirin with incidence of acute kidney injury (AKI) and death based on estimated glomerular filtration rate (eGFR). Results: Of 3585 patients, 31.5% had CKD (eGFR < 60 mL/min/1.73 m2 ) at baseline and 27.6% had AKI postoperatively. The baseline eGFR had a nonlinear relationship with the incidence and stages of AKI. As eGFR decreased to 15 to 30 from more than or equal to 90 mL/min/1.73 m2 , AKI and 30-day mortality increased to 50.5% from 23.5% and to 11.9% from 2.6%, respectively (P < 0.001). However, preoperative aspirin use was associated with a significant decrease in postoperative AKI and 30-day mortality in patients with CKD undergoing cardiac surgery, in particular, the survival benefit associated with aspirin was greater in patients with CKD (vs normal kidney function): 30-day mortality was reduced by 23.3%, 58.2%, or 70.0% for patients with baseline eGFR more than or equal to 90, 30 to 59, or 15 to 30 mL/min/ 1.73 m2 , respectively (P trend < 0.001). Conclusions: For patients with CKD undergoing cardiac surgery, preoperative aspirin therapy was associated with renal protection and mortality decline. The magnitude of the survival benefit was greater in patients with CKD than normal kidney function. Keywords: acute kidney injury, coronary artery bypass graft, aspirin, cardiac surgery, chronic kidney disease, valve surgery (Ann Surg 2015;261:207–212)
From the ∗ Department of Anesthesiology, Thomas Jefferson University, Philadelphia, PA; †Department of Anesthesiology, Tangdu Hospital, The Fourth Military Medical University, Xian, P. R. China; ‡Division of Cardiothoracic Surgery, University of California Davis Medical Center, Sacramento, CA; §Department of Anesthesiology and Pain Medicine, University of California Davis Medical Center, Sacramento, CA; ¶Department of Internal Medicine, University of California Davis Medical Center, Sacramento, CA; ||Jefferson Medical College, Philadelphia, PA; and ∗∗ Division of Cardiothoracic Surgery, Thomas Jefferson University, Philadelphia, PA. A part of this study was presented in an abstract at Euroanaesthesia 2012, Paris, June 9–12, 2012. Disclosure: The authors have declared that no conflict of interest exists. This study was partially supported by the authors’ institution and department. Dr Yao was also supported by Science and Technology Development Fund (No. 2008K1313), Shaanxi Province, China. Reprints: Jianzhong Sun, MD, PhD, Department of Anesthesiology, Jefferson Medical College, Thomas Jefferson University, Suite G8490, 111 South 11th Street, Philadelphia, PA 19107. E-mail: [email protected]. C 2014 by Lippincott Williams & Wilkins Copyright ISSN: 0003-4932/14/26101-0207 DOI: 10.1097/SLA.0000000000000641
Annals of Surgery r Volume 261, Number 1, January 2015
C
hronic kidney disease (CKD) is common in the general population, particularly in patients with cardiovascular disease (CVD).1–3 In previous studies, more than 20% of patients undergoing cardiac surgery were found to have renal insufficiency or CKD.4,5 In addition, acute kidney injury (AKI) is a common postoperative complication for patients undergoing cardiac surgery, occurring in up to 30% of all cardiac surgery patients.5–7 CKD, AKI, or perioperative kidney function has a significant impact on morbidity and mortality in patients undergoing cardiac surgery.4,6,8–10 However, no therapy has been proven effective in reducing or preventing postoperative AKI or major complications in patients with renal dysfunction undergoing cardiac surgery.11,12 Previous clinical studies have linked aspirin to improvement of renal and cardiovascular outcomes in cardiac surgery patients. In a multicenter and observational study, Mangano13 showed that the early use of postoperative aspirin was associated with a significant reduction in the incidence of postoperative renal failure in patients undergoing coronary artery bypass graft (CABG), suggesting that aspirin may have beneficial effects in patients with renal dysfunction undergoing cardiac surgery. On the basis of previous studies including from ours,13,14 we hypothesized that preoperative aspirin therapy may protect kidneys against ischemia/reperfusion injury-induced by cardiac surgery, and the protection may be correlated with preoperative kidney function, that is, baseline estimated glomerular filtration rate (eGFR). This study was undertaken to test this hypothesis; and the specific goals were to determine (1) whether preoperative aspirin therapy is associated with the reduction of postoperative AKI and death in patients with CKD undergoing cardiac surgery, and (2) if yes, what relationship exists among the death, preoperative use of aspirin, and baseline kidney function or eGFR.
METHODS Study Design This study was a cohort, retrospective study involving consecutive patients (n = 5175) receiving cardiac surgery including CABG, valve surgery, CABG plus valve surgery, or other cardiac surgery at 2 tertiary medical centers (Thomas Jefferson University hospital and UC Davis Medical Center, dated from 2001 to 2010). The study was in compliance with the Declaration of Helsinki, approved by the local institutional review board, and individual consent was waived in compliance with the HIPAA (Health Insurance Portability and Accountability Act) regulations and the waiver criteria. The patients excluded were those with preoperative anticoagulants, adenosine diphosphate receptor inhibitors, glycoprotein IIb/IIIa inhibitors, antiplatelets, unknown aspirin use, or those without records of preoperative or postoperative creatinine. Of all patients, 3585 met the inclusion criteria and were divided into 2 groups: those who use www.annalsofsurgery.com | 207
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Yao et al
(n = 2407) and those who do not use (n = 1178) preoperative aspirin (within 5 days preceding surgery) (Fig. 1). On the basis of preoperative eGFR (mL/min/1.73 m2 ), the patients (n = 3585) were divided into 5 groups at baseline: normal eGFR (>90), mildly decreased eGFR (60–89), moderately decreased eGFR (30–59), severely decreased eGFR (15–29), or on dialysis/kidney failure (eGFR < 15).15,16 eGFR was calculated on the basis of Modification of Diet in Renal Disease equations.17 CKD is defined by eGFR less than 60 mL/min/1.73 m2 .15 Postoperatively, AKI was defined and divided into 3 stages on the basis of Acute Kidney Injury Network criteria: increase in serum creatinine above baseline more than or equal to 1.5- to twofold or creatinine increase of 0.3 mg/dL or more (stage 1); more than 2- to threefold (stage 2); more than threefold or creatinine increase of more than 4.0 mg/dL or acute increase of more than 0.5 mg/dL (stage 3), respectively.18 As described in other reports,5,6 the last serum creatinine before surgery and the highest serum creatinine during the first 7 days after surgery were taken for measuring baseline and postoperative kidney functions, respectively.
Data Collection The data were prospectively compiled into a database by independent investigators. The patient data collected included demographics, patient history, medical record information, preoperative risk factors, preoperative medications, intraoperative data, postoperative cardiocerebral events, 30-day all cause mortality, and pre- and postoperative creatinine. Preoperative use of aspirin indicates use of aspirin in the patient within 5 days preceding surgery. The outcomes of this study were postoperative AKI and 30-day all cause mortality.
Statistical Analysis
Continuous and categorical variables were reported as mean ± SD or percentages, and compared with a 2-sample t test, an analysis of variance or a χ 2 test (2 tailed) as appropriate, respectively. Univariate and multivariate logistic regression were performed to assess associations of demographic, therapeutic, and clinical outcome
variables. Missing values were imputed to the median for continuous variables (after stratifying on relevant variables to enhance prediction of the missing value), and the most frequent value for binary and polytomous variables.19 As detailed previously,14,20 a propensity score-adjusted analysis21,22 was used to reduce the potential bias resulting from nonrandom assignments to the 2 groups. In brief, a propensity score was derived, reflecting the probability that a patient would receive preoperative aspirin. This was accomplished by performing a multivariate logistic regression analysis using preoperative aspirin as the dependent variable and entering all baseline variables and the type of surgery that clinically would likely affect the probability of using preoperative aspirin. And then, with all relevant variables including the propensity score, the multivariate logistic regression analysis was performed to identify independent predictors or risk factors for postoperative AKI and mortality. To achieve model parsimony and stability, the backward stepwise selection procedure was applied with the dropout criterion P > 0.2. On the basis of variables contained in the database, clinical relevance and previous studies, the following variables were included in the analysis: (1) baseline factors including age, sex, and body mass index (BMI); (2) patient history including diabetes, hypertension, smoker, peripheral vascular disease, cerebrovascular disease, chronic lung disease, and family History of coronary artery disease (CAD); (3) preoperative risk factors including angina, congestive heart failure, previous myocardial infarction (MI), and preoperative medications such as β-blockers, digitalis, diuretics and rennin-angiotensin system inhibitors [RAS inhibitors including angiotensin-converting enzyme (ACE) Inhibitors or angiotensin-II receptor blockers (ARB)], lipid lowering drugs in addition to aspirin; and (4) intraoperative factors including perfusion time, cross-clamp time, CABG, valve, CABG plus valve or other cardiac surgery. The Hosmer-Lemeshow goodness-of-fit statistic was applied for the models fit and the C statistic as a measure of predictive power. Percentages, odds ratios (ORs), 95% confidence intervals (CI), and P (2-sided) < 0.05 were given in
Eligible subjects identified (n=5175) (1) Age 18+ years (2) CABG, Valve and other cardiac surgery
Subjects excluded (n=1590) (1) Preoperative anticoagulants(n=649) (2) Preoperative ADP inhibitors(n=114) (3) Preoperative Gp IIbIIIa inhibitors(n=32) (4) Preoperative antiplatelets(n=277) (5) Without Pre- or Postoperative creatinine(460) (6) Unknown aspirin use(58) Subjects included (n=3585)
With preoperative aspirin (n=2407) (1) CABG (n=1178) (2) Valve (n=594) (3) Valve plus CABG (n=365) (4) Others (n=270)
Without preoperative aspirin (n=1178) (1) CABG (n=548) (2) Valve (n=309) (3) Valve plus CABG (n=175) (4) Others (n=146)
FIGURE 1. Selection of study sample. ADP indicates adenosine diphosphate. 208 | www.annalsofsurgery.com
C 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
the results. Death rates were also analyzed for changes over degrees of renal dysfunction using the Cochran-Armitage test for trend. SPSS 17.0 software for Windows (SPSS Inc, Chicago, IL) was used for the statistical analysis and SAS version 9.3 (SAS Institute, Inc, Cary, NC) for the Cochran-Armitage test.
RESULTS Baseline and Intraoperative Parameters Of 5175 eligible patients in the database, 3585 patients met the inclusion criteria and were divided into 2 groups: those who use (n = 2407) and those who do not use (n = 1178) preoperative aspirin (Fig. 1). Demographic and clinical data of the patients are presented in Table 1. No significant differences were found between the 2 groups in age, sex, BMI, smoking, family history of CAD, history of chronic lung disease, cerebrovascular disease, congestive heart failure, or preoperative use of ACE inhibitor, ARB, or lipid lowering drugs. However, the patients with preoperative aspirin were more likely to have diabetes, hypertension, peripheral vascular disease, angina, previous MI, and be using preoperative beta-blockers. Surgical factors such as perfusion time, cross-clamp time, and type of surgery including CABG, valve, CABG + valve, or other cardiac surgery were similar between 2 groups (Table 1).
Association of CKD, AKI, and Mortality Of 3585 patients at baseline, 822, 1634, 916, and 101 showed no, mild, moderate, and severe renal disease, respectively, and 112 patients were on dialysis indicating that CKD was common and presented in 31.5% of this cohort of patients undergoing cardiac surgery. Postoperatively 27.6% of patients (989 of 3585) had various stages of AKI (Table 2). Overall, the incidence and stage of AKI were closely correlated with the baseline GFR or renal function, that is, as renal dysfunction advanced or eGFR decreased at baseline, postoperative AKI occurred more frequently (ranged from 23.5% if eGFR ≥90 mL/min/1.73 m2 to 50.5% if eGFR at 15– 30 mL/min/1.73 m2 ) and more severe (stage 3 of AKI was 2.2% if eGFR ≥90 mL/min/1.73 m2 but 18.8%% if eGFR at 15–30 mL/ min/1.73 m2 ) (Table 2 and Fig. 2). As illustrated in Figure 2, there was a nonlinear relationship between baseline eGFR and postoperative probability of AKI: while eGFR was more than 90 mL/min/1.73 m2 , the probability of AKI declined sharply; while eGFR decreased from 90 mL/min/1.73 m2 to 30 mL/min/1.73 m2 , the probability of AKI rose slowly; and while eGFR declined to less than 30 mL/min/1.73 m2 , the probability of AKI rose sharply. In addition, as baseline eGFR decreased from 90 or more (normal) to 15 to 30 (severe renal disease) or to less than 15 mL/min/ 1.73 m2 (on dialysis), 30-day mortality increased from 2.6% to 11.9% (by 4.6-fold) or to 13.4% (by 5.2-fold), respectively (Table 2).
Aspirin and Chronic Kidney Disease in Cardiac Surgery
Independent Risk Factors for AKI With multivariate logistic regression analysis on independent risk factors for postoperative AKI, preoperative use of aspirin was associated with a significant protective effect against AKI (OR = 0.533, 95% CI: 0.446–0.636, P < 0.001). Other independent risk factors of postoperative AKI were male sex, preoperative renal dysfunction, family history of CAD, diabetes, hypertension, angina, previous MI, congestive heart failure, and perfusion time (Table 3).
Association Between Preoperative Aspirin and Postoperative Outcomes Overall, preoperative aspirin is associated with significant beneficial effects on major postoperative outcomes: AKI and 30-day mortality (Figs. 3, 4). The unadjusted univariate analysis showed that AKI was 25.8% (620 of 2407) versus 31.3% (369 of 1178) for patients
TABLE 1. Demographic and Clinical Characteristics Aspirin Yes (N = 2407) No (N = 1178)
Characteristics Age, yrs Male sex, % BMI, kg/m2 Past medical history Diabetes Hypertension Smoker Cerebrovascular disease Peripheral vascular disease Chronic lung disease Family history CAD Clinical pattern Angina Congestive heart failure Previous MI Medical therapy Beta blockers ACE inhibitors or ARB Lipid lowering Perfusion time, min Cross-clamp time, min CABG Valve CABG + Valve Others
P
62.2 ± 13.4 1641 (68.2) 29.0 ± 6.6
62.1 ± 13.8 798 (67.7) 29.2 ± 6.2
821 (34.1) 1665 (69.2) 641 (26.6) 378 (15.7) 321 (13.3) 516 (21.4) 847 (35.2)
279 (23.7) 707 (60.0) 303 (25.7) 159 (13.5) 113 (9.6) 231 (19.6) 410 (34.8)
Evidence for Preoperative Aspirin Improving Major Outcomes in Patients With Chronic Kidney Disease Undergoing Cardiac Surgery A Cohort Study Linong Yao, MD, PhD,∗ † Nilas Young, MD,‡ Hong Liu, MD,§ Zhongmin Li, PhD,¶ Will Sun, MS,|| Jordan Goldhammer, MD,∗ Lei Tao, MD,† Jianbin He, MD,† James Diehl, MD,∗∗ and Jianzhong Sun, MD, PhD∗ Background: Effects of aspirin on patients with chronic kidney disease (CKD) remains unclear. This study aimed to examine the effect of preoperative aspirin use on postoperative renal function and 30-day mortality in patients with CKD undergoing cardiac surgery. Methods: A retrospective cohort study was performed on consecutive patients (n = 5175) receiving cardiac surgery in 2 tertiary hospitals. Of all patients, 3585 met the inclusion criteria and underwent the analysis to determine the association of preoperative aspirin with incidence of acute kidney injury (AKI) and death based on estimated glomerular filtration rate (eGFR). Results: Of 3585 patients, 31.5% had CKD (eGFR < 60 mL/min/1.73 m2 ) at baseline and 27.6% had AKI postoperatively. The baseline eGFR had a nonlinear relationship with the incidence and stages of AKI. As eGFR decreased to 15 to 30 from more than or equal to 90 mL/min/1.73 m2 , AKI and 30-day mortality increased to 50.5% from 23.5% and to 11.9% from 2.6%, respectively (P < 0.001). However, preoperative aspirin use was associated with a significant decrease in postoperative AKI and 30-day mortality in patients with CKD undergoing cardiac surgery, in particular, the survival benefit associated with aspirin was greater in patients with CKD (vs normal kidney function): 30-day mortality was reduced by 23.3%, 58.2%, or 70.0% for patients with baseline eGFR more than or equal to 90, 30 to 59, or 15 to 30 mL/min/ 1.73 m2 , respectively (P trend < 0.001). Conclusions: For patients with CKD undergoing cardiac surgery, preoperative aspirin therapy was associated with renal protection and mortality decline. The magnitude of the survival benefit was greater in patients with CKD than normal kidney function. Keywords: acute kidney injury, coronary artery bypass graft, aspirin, cardiac surgery, chronic kidney disease, valve surgery (Ann Surg 2015;261:207–212)
From the ∗ Department of Anesthesiology, Thomas Jefferson University, Philadelphia, PA; †Department of Anesthesiology, Tangdu Hospital, The Fourth Military Medical University, Xian, P. R. China; ‡Division of Cardiothoracic Surgery, University of California Davis Medical Center, Sacramento, CA; §Department of Anesthesiology and Pain Medicine, University of California Davis Medical Center, Sacramento, CA; ¶Department of Internal Medicine, University of California Davis Medical Center, Sacramento, CA; ||Jefferson Medical College, Philadelphia, PA; and ∗∗ Division of Cardiothoracic Surgery, Thomas Jefferson University, Philadelphia, PA. A part of this study was presented in an abstract at Euroanaesthesia 2012, Paris, June 9–12, 2012. Disclosure: The authors have declared that no conflict of interest exists. This study was partially supported by the authors’ institution and department. Dr Yao was also supported by Science and Technology Development Fund (No. 2008K1313), Shaanxi Province, China. Reprints: Jianzhong Sun, MD, PhD, Department of Anesthesiology, Jefferson Medical College, Thomas Jefferson University, Suite G8490, 111 South 11th Street, Philadelphia, PA 19107. E-mail: [email protected]. C 2014 by Lippincott Williams & Wilkins Copyright ISSN: 0003-4932/14/26101-0207 DOI: 10.1097/SLA.0000000000000641
Annals of Surgery r Volume 261, Number 1, January 2015
C
hronic kidney disease (CKD) is common in the general population, particularly in patients with cardiovascular disease (CVD).1–3 In previous studies, more than 20% of patients undergoing cardiac surgery were found to have renal insufficiency or CKD.4,5 In addition, acute kidney injury (AKI) is a common postoperative complication for patients undergoing cardiac surgery, occurring in up to 30% of all cardiac surgery patients.5–7 CKD, AKI, or perioperative kidney function has a significant impact on morbidity and mortality in patients undergoing cardiac surgery.4,6,8–10 However, no therapy has been proven effective in reducing or preventing postoperative AKI or major complications in patients with renal dysfunction undergoing cardiac surgery.11,12 Previous clinical studies have linked aspirin to improvement of renal and cardiovascular outcomes in cardiac surgery patients. In a multicenter and observational study, Mangano13 showed that the early use of postoperative aspirin was associated with a significant reduction in the incidence of postoperative renal failure in patients undergoing coronary artery bypass graft (CABG), suggesting that aspirin may have beneficial effects in patients with renal dysfunction undergoing cardiac surgery. On the basis of previous studies including from ours,13,14 we hypothesized that preoperative aspirin therapy may protect kidneys against ischemia/reperfusion injury-induced by cardiac surgery, and the protection may be correlated with preoperative kidney function, that is, baseline estimated glomerular filtration rate (eGFR). This study was undertaken to test this hypothesis; and the specific goals were to determine (1) whether preoperative aspirin therapy is associated with the reduction of postoperative AKI and death in patients with CKD undergoing cardiac surgery, and (2) if yes, what relationship exists among the death, preoperative use of aspirin, and baseline kidney function or eGFR.
METHODS Study Design This study was a cohort, retrospective study involving consecutive patients (n = 5175) receiving cardiac surgery including CABG, valve surgery, CABG plus valve surgery, or other cardiac surgery at 2 tertiary medical centers (Thomas Jefferson University hospital and UC Davis Medical Center, dated from 2001 to 2010). The study was in compliance with the Declaration of Helsinki, approved by the local institutional review board, and individual consent was waived in compliance with the HIPAA (Health Insurance Portability and Accountability Act) regulations and the waiver criteria. The patients excluded were those with preoperative anticoagulants, adenosine diphosphate receptor inhibitors, glycoprotein IIb/IIIa inhibitors, antiplatelets, unknown aspirin use, or those without records of preoperative or postoperative creatinine. Of all patients, 3585 met the inclusion criteria and were divided into 2 groups: those who use www.annalsofsurgery.com | 207
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Yao et al
(n = 2407) and those who do not use (n = 1178) preoperative aspirin (within 5 days preceding surgery) (Fig. 1). On the basis of preoperative eGFR (mL/min/1.73 m2 ), the patients (n = 3585) were divided into 5 groups at baseline: normal eGFR (>90), mildly decreased eGFR (60–89), moderately decreased eGFR (30–59), severely decreased eGFR (15–29), or on dialysis/kidney failure (eGFR < 15).15,16 eGFR was calculated on the basis of Modification of Diet in Renal Disease equations.17 CKD is defined by eGFR less than 60 mL/min/1.73 m2 .15 Postoperatively, AKI was defined and divided into 3 stages on the basis of Acute Kidney Injury Network criteria: increase in serum creatinine above baseline more than or equal to 1.5- to twofold or creatinine increase of 0.3 mg/dL or more (stage 1); more than 2- to threefold (stage 2); more than threefold or creatinine increase of more than 4.0 mg/dL or acute increase of more than 0.5 mg/dL (stage 3), respectively.18 As described in other reports,5,6 the last serum creatinine before surgery and the highest serum creatinine during the first 7 days after surgery were taken for measuring baseline and postoperative kidney functions, respectively.
Data Collection The data were prospectively compiled into a database by independent investigators. The patient data collected included demographics, patient history, medical record information, preoperative risk factors, preoperative medications, intraoperative data, postoperative cardiocerebral events, 30-day all cause mortality, and pre- and postoperative creatinine. Preoperative use of aspirin indicates use of aspirin in the patient within 5 days preceding surgery. The outcomes of this study were postoperative AKI and 30-day all cause mortality.
Statistical Analysis
Continuous and categorical variables were reported as mean ± SD or percentages, and compared with a 2-sample t test, an analysis of variance or a χ 2 test (2 tailed) as appropriate, respectively. Univariate and multivariate logistic regression were performed to assess associations of demographic, therapeutic, and clinical outcome
variables. Missing values were imputed to the median for continuous variables (after stratifying on relevant variables to enhance prediction of the missing value), and the most frequent value for binary and polytomous variables.19 As detailed previously,14,20 a propensity score-adjusted analysis21,22 was used to reduce the potential bias resulting from nonrandom assignments to the 2 groups. In brief, a propensity score was derived, reflecting the probability that a patient would receive preoperative aspirin. This was accomplished by performing a multivariate logistic regression analysis using preoperative aspirin as the dependent variable and entering all baseline variables and the type of surgery that clinically would likely affect the probability of using preoperative aspirin. And then, with all relevant variables including the propensity score, the multivariate logistic regression analysis was performed to identify independent predictors or risk factors for postoperative AKI and mortality. To achieve model parsimony and stability, the backward stepwise selection procedure was applied with the dropout criterion P > 0.2. On the basis of variables contained in the database, clinical relevance and previous studies, the following variables were included in the analysis: (1) baseline factors including age, sex, and body mass index (BMI); (2) patient history including diabetes, hypertension, smoker, peripheral vascular disease, cerebrovascular disease, chronic lung disease, and family History of coronary artery disease (CAD); (3) preoperative risk factors including angina, congestive heart failure, previous myocardial infarction (MI), and preoperative medications such as β-blockers, digitalis, diuretics and rennin-angiotensin system inhibitors [RAS inhibitors including angiotensin-converting enzyme (ACE) Inhibitors or angiotensin-II receptor blockers (ARB)], lipid lowering drugs in addition to aspirin; and (4) intraoperative factors including perfusion time, cross-clamp time, CABG, valve, CABG plus valve or other cardiac surgery. The Hosmer-Lemeshow goodness-of-fit statistic was applied for the models fit and the C statistic as a measure of predictive power. Percentages, odds ratios (ORs), 95% confidence intervals (CI), and P (2-sided) < 0.05 were given in
Eligible subjects identified (n=5175) (1) Age 18+ years (2) CABG, Valve and other cardiac surgery
Subjects excluded (n=1590) (1) Preoperative anticoagulants(n=649) (2) Preoperative ADP inhibitors(n=114) (3) Preoperative Gp IIbIIIa inhibitors(n=32) (4) Preoperative antiplatelets(n=277) (5) Without Pre- or Postoperative creatinine(460) (6) Unknown aspirin use(58) Subjects included (n=3585)
With preoperative aspirin (n=2407) (1) CABG (n=1178) (2) Valve (n=594) (3) Valve plus CABG (n=365) (4) Others (n=270)
Without preoperative aspirin (n=1178) (1) CABG (n=548) (2) Valve (n=309) (3) Valve plus CABG (n=175) (4) Others (n=146)
FIGURE 1. Selection of study sample. ADP indicates adenosine diphosphate. 208 | www.annalsofsurgery.com
C 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
the results. Death rates were also analyzed for changes over degrees of renal dysfunction using the Cochran-Armitage test for trend. SPSS 17.0 software for Windows (SPSS Inc, Chicago, IL) was used for the statistical analysis and SAS version 9.3 (SAS Institute, Inc, Cary, NC) for the Cochran-Armitage test.
RESULTS Baseline and Intraoperative Parameters Of 5175 eligible patients in the database, 3585 patients met the inclusion criteria and were divided into 2 groups: those who use (n = 2407) and those who do not use (n = 1178) preoperative aspirin (Fig. 1). Demographic and clinical data of the patients are presented in Table 1. No significant differences were found between the 2 groups in age, sex, BMI, smoking, family history of CAD, history of chronic lung disease, cerebrovascular disease, congestive heart failure, or preoperative use of ACE inhibitor, ARB, or lipid lowering drugs. However, the patients with preoperative aspirin were more likely to have diabetes, hypertension, peripheral vascular disease, angina, previous MI, and be using preoperative beta-blockers. Surgical factors such as perfusion time, cross-clamp time, and type of surgery including CABG, valve, CABG + valve, or other cardiac surgery were similar between 2 groups (Table 1).
Association of CKD, AKI, and Mortality Of 3585 patients at baseline, 822, 1634, 916, and 101 showed no, mild, moderate, and severe renal disease, respectively, and 112 patients were on dialysis indicating that CKD was common and presented in 31.5% of this cohort of patients undergoing cardiac surgery. Postoperatively 27.6% of patients (989 of 3585) had various stages of AKI (Table 2). Overall, the incidence and stage of AKI were closely correlated with the baseline GFR or renal function, that is, as renal dysfunction advanced or eGFR decreased at baseline, postoperative AKI occurred more frequently (ranged from 23.5% if eGFR ≥90 mL/min/1.73 m2 to 50.5% if eGFR at 15– 30 mL/min/1.73 m2 ) and more severe (stage 3 of AKI was 2.2% if eGFR ≥90 mL/min/1.73 m2 but 18.8%% if eGFR at 15–30 mL/ min/1.73 m2 ) (Table 2 and Fig. 2). As illustrated in Figure 2, there was a nonlinear relationship between baseline eGFR and postoperative probability of AKI: while eGFR was more than 90 mL/min/1.73 m2 , the probability of AKI declined sharply; while eGFR decreased from 90 mL/min/1.73 m2 to 30 mL/min/1.73 m2 , the probability of AKI rose slowly; and while eGFR declined to less than 30 mL/min/1.73 m2 , the probability of AKI rose sharply. In addition, as baseline eGFR decreased from 90 or more (normal) to 15 to 30 (severe renal disease) or to less than 15 mL/min/ 1.73 m2 (on dialysis), 30-day mortality increased from 2.6% to 11.9% (by 4.6-fold) or to 13.4% (by 5.2-fold), respectively (Table 2).
Aspirin and Chronic Kidney Disease in Cardiac Surgery
Independent Risk Factors for AKI With multivariate logistic regression analysis on independent risk factors for postoperative AKI, preoperative use of aspirin was associated with a significant protective effect against AKI (OR = 0.533, 95% CI: 0.446–0.636, P < 0.001). Other independent risk factors of postoperative AKI were male sex, preoperative renal dysfunction, family history of CAD, diabetes, hypertension, angina, previous MI, congestive heart failure, and perfusion time (Table 3).
Association Between Preoperative Aspirin and Postoperative Outcomes Overall, preoperative aspirin is associated with significant beneficial effects on major postoperative outcomes: AKI and 30-day mortality (Figs. 3, 4). The unadjusted univariate analysis showed that AKI was 25.8% (620 of 2407) versus 31.3% (369 of 1178) for patients
TABLE 1. Demographic and Clinical Characteristics Aspirin Yes (N = 2407) No (N = 1178)
Characteristics Age, yrs Male sex, % BMI, kg/m2 Past medical history Diabetes Hypertension Smoker Cerebrovascular disease Peripheral vascular disease Chronic lung disease Family history CAD Clinical pattern Angina Congestive heart failure Previous MI Medical therapy Beta blockers ACE inhibitors or ARB Lipid lowering Perfusion time, min Cross-clamp time, min CABG Valve CABG + Valve Others
P
62.2 ± 13.4 1641 (68.2) 29.0 ± 6.6
62.1 ± 13.8 798 (67.7) 29.2 ± 6.2
821 (34.1) 1665 (69.2) 641 (26.6) 378 (15.7) 321 (13.3) 516 (21.4) 847 (35.2)
279 (23.7) 707 (60.0) 303 (25.7) 159 (13.5) 113 (9.6) 231 (19.6) 410 (34.8)
