- Email: [email protected]
Differences and similarities in risk factors for postoperative acute kidney injury between younger and older adults undergoing cardiac surgery
Saydy et al
Perioperative Management
Differences and similarities in risk factors for postoperative acute kidney injury between younger and older adults undergoing cardiac surgery Nadim Saydy,a Amine Mazine, MD, MSc,a Louis-Mathieu Stevens, MD, PhD,b Hughes Jeamart, MD, MSc,a,c Philippe Demers, MD, MSc,a,c Pierre Page, MD,a,c Yoan Lamarche, MD, MSc,a,c and Ismail El-Hamamsy, MD, PhDa,c ABSTRACT Objectives: Acute kidney injury is a frequent complication after cardiac surgery. The purpose of this study was to assess the risk factors for acute kidney injury in patients 60 years of age undergoing cardiac surgery and to compare these risk factors with those identified in patients 65 years of age.
Results: In both age groups, the following variables were associated independently with greater degrees of postoperative increase in serum creatinine on multivariable analysis: greater body mass index, peripheral vascular disease, preoperative use of diuretics, lower preoperative hemoglobin, preoperative intra-aortic balloon pump, urgent or emergent status, long cardiopulmonary bypass time, and hemofiltration. In younger patients, a greater increase in serum creatinine was associated with diabetes, and previous cardiac surgery, whereas female sex was associated with a lower degree of increase in serum creatinine. In older patients, a greater increase in serum creatinine was associated with age, hypertension, smoking, and lower left ventricular left ejection fraction. Operation type and coronary artery disease had a different impact on postoperative creatinine increase between younger and older patients. Conclusions: This study identified both common and distinct risk factors associated with postoperative increase in serum creatinine between patients 60 years and those 65 years undergoing cardiac surgery. Importantly, all potentially modifiable risk factors were present in both groups. (J Thorac Cardiovasc Surg 2017;-:1-10)
From the aDivision of Cardiac Surgery, Montreal Heart Institute, bDivision of Cardiac Surgery, Centre Hospitalier Universitaire de Montreal, and cDivision of Cardiac Surgery, Sacre-Cœur Hospital, University of Montreal School of Medicine, Montreal, Quebec, Canada. N.S. and A.M. contributed equally to this work and share co–first authorship. Received for publication Oct 4, 2016; revisions received Aug 1, 2017; accepted for publication Aug 18, 2017. Address for reprints: Ismail El-Hamamsy, MD, PhD, Department of Surgery, University of Montreal, Division of Cardiac Surgery, Montreal Heart Institute, 5000 Belanger St, Montreal, Quebec H1T 1C8, Canada (E-mail: i.elhamamsy@ icm-mhi.org). 0022-5223/$36.00 Copyright Ó 2017 by The American Association for Thoracic Surgery http://dx.doi.org/10.1016/j.jtcvs.2017.08.039
Factors associated with postoperative increase in creatinine on multivariable analysis. Central Message This study identified both common and distinct risk factors associated with postoperative increase in serum creatinine between patients 60 years and those 65 years undergoing cardiac surgery.
PM
Methods: From 2010 to 2012, 1253 patients 60 years (mean age 52 9 years) and 2488 patients 65 years (mean age 74 6 years) underwent cardiac surgery. Linear regression models using least absolute shrinkage and selection operator methods and mixed effects linear regression models were used to assess factors associated with maximum postoperative increase in serum creatinine in these two cohorts.
Perspective Postoperative acute kidney injury (AKI) is an important predictor of morbidity and mortality after cardiac surgery. Several risk factors have been identified; however, few studies have focused on young and middle-aged adults. We found that AKI is associated with negative outcomes, even in these younger patients. We identified both common and distinct risk factors for AKI between patients 60 years and those 65 years.
See Editorial Commentary page XXX.
Acute kidney injury (AKI) after cardiac surgery is a frequent complication that occurs in up to 40% of patients,1 with dialysis required in 1% to 2% of all patients.2 Importantly, AKI has been associated with increased in-hospital mortality and reduced long-term survival.3,4 In view of the lack of active Scanning this QR code will take you to the article title page.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
1
Perioperative Management
Abbreviations and Acronyms AKI ¼ acute kidney injury BMI ¼ body mass index CABG ¼ coronary artery bypass graft CPB ¼ cardiopulmonary bypass IABP ¼ intra-aortic balloon pump LASSO ¼ least absolute shrinkage and selection operator RRT ¼ renal-replacement therapy
PM
treatment for AKI, prevention and management of risk factors remain critical. Many studies have sought to identify preoperative and perioperative risk factors that predispose to AKI after cardiac surgery; however, few have focused specifically on young and middle-aged adults. The general aim of this study was to assess the risk factors for AKI in patients 60 years of age and younger undergoing cardiac surgery and to compare these risk factors with those identified in patients 65 years of age and older. The specific goals were to (1) compare the baseline and operative characteristics, as well as the postoperative clinical outcomes, between patients who did and did not develop postoperative AKI in each age group independently and (2) to model the maximum increase in postoperative serum creatinine by the use of linear regression techniques to identify specific and common risk variables in the younger and older patients. Our hypothesis was that traditional risk factors, such as atherosclerosis related factors, are less likely to be associated with AKI in younger patients than in those older than 65 years. This cohort did not include patients with complex adult congenital heart disease. METHODS Study Population The study was approved by the institutional review board, and a waiver of individual patient consent was obtained. This is a retrospective analysis of data collected between October 2010 and December 2012 in 2 centers affiliated with the University of Montreal: Montreal Heart Institute and Sacre-Cœur Hospital. All consecutive patients older than 18 years of age undergoing cardiac surgery were screened for eligibility. Exclusion criteria were age 61 to 64 years (n ¼ 588), off-pump coronary artery bypass grafting (CABG) surgery (n ¼ 293), and renal-replacement therapy (RRT) before surgery (n ¼ 32). After the exclusion criteria were applied, the final cohort consisted of 3741 consecutive patients. Of these, 1253 were 60 years of age or younger (study group), and 2488 were 65 years of age or older (control group). Their procedures were performed by 1 of 12 surgeons involved in the study. Preoperative medication was reviewed for the 7 days before surgery. Angiotensin-converting enzyme inhibitors and/or diuretic use were not discontinued before surgery in all patients. During the study period, tranexamic acid was the only antithrombolytic agent used. Data collected included baseline preoperative characteristics, operative details, and postoperative outcomes.
Definitions Patients who required postoperative dialysis or who presented a greater than 50% increase in serum creatinine in the first 7 postoperative days
2
Saydy et al
compared with the last baseline value before surgery were deemed to have AKI. Operative mortality was defined as death occurring within 30 days of surgery or during the index hospitalization. Surgery was considered urgent in patients who had not been electively admitted for operation but who underwent surgery during the same admission for medical reasons. Emergent operation was defined as operation within 6 hours of admission. Weaning from cardiopulmonary bypass (CPB) was considered difficult if any of the following was required: 3 vasoactive agents at low dose, 2 vasoactive agents at high dose, return on CPB, or mechanical support.
Statistical Analysis The general statistical plan for this study was to first compare the baseline and operative characteristics, as well as the postoperative clinical outcomes, between patients who did and did not develop postoperative AKI in the younger (60 years) and older (65 years) age groups independently and to then model the maximum increase in postoperative serum creatinine by the use of linear regression techniques to identify specific and common risk variables in each group. The specific statistical methods used to achieve these goals are detailed below. Baseline characteristics, intraoperative data, and early outcomes were compared within each group—ie, patients 60 years and 65 years— between patients who developed postoperative AKI and those who did not. Continuous variables are expressed as mean standard deviation (when normally distributed) or median and interquartile range (when not normally distributed), and were compared by means of Student t test or the Mann-Whitney U test, as appropriate. Categorical variables are expressed as frequency (%) and were compared by means of Pearson c2 test or Fisher exact test, as appropriate. Factors associated with maximum postoperative increase in serum creatinine were identified via a combination of 2 methods. First, linear regression models using least absolute shrinkage and selection operator (LASSO) selection methods were used to identify candidate variables.5 Variables assessed as potential risk factors were the preoperative baseline characteristics described in Table 1, as well as the operative data presented in Table 2. For variables that were not distributed according to the normal distribution (eg, serum creatinine levels, CPB duration), a logarithmic transformation was used. The LASSO is a regression analysis method that applies a penalty to large coefficients, thus resulting in more interpretable regression models. Because of its ability to filter through a large number of variables and shrink effect coefficients, the LASSO method is especially useful when a large number of potential variables are available for model fitting. In addition, the LASSO mitigates many of the issues associated with the stepwise variable selection method used in traditional regression models, such as overfitting resulting from multicollinearity between covariates. To account for the effect of the individual surgeon and institution, multivariable analyses using mixed effects linear regression models were then conducted with the variables identified through LASSO (the MIXED procedure, SAS software, version 9.4; SAS Institute, Cary, NC). Every mixedeffect model included the last baseline preoperative serum creatinine value as a covariate. Only factors that were significant under both the LASSOselected and mixed effects models were considered to be validated for the final model. This analysis sequence—ie, variable preselection through LASSO modeling followed by validation through mixed effect linear regression modeling—was first applied separately to patients 60 years and those 65 years, resulting in 2 distinct models. In the final step of model building, all variables identified through the separate models were forced into a common model that included all patients, and all 2-way interaction effects between these variables and age group were examined. Of note, the multivariable model was a linear regression model aimed at modeling maximum postoperative increase in serum creatinine, not a logistic regression model aimed at modeling the incidence of AKI. All analyses were carried out using the SAS software, version 9.4 (SAS Institute). Statistical significance was set at a ¼ 0.05.
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Perioperative Management
TABLE 1. Baseline characteristics
Variables Demographic data Age, y, mean SD Female sex Body mass index, kg/m2, mean SD
Patients 60 years old AKI n ¼ 192 P value
Std Diff
No AKI n ¼ 1851
Patients 65 years old AKI n ¼ 637 P value
Std Diff
52 9 241 (23%) 28 5
52 8 62 (32%) 30 7
.5 <.01 <.001
0.06 0.21 0.30
74 6 556 (30%) 28 5
74 6 280 (44%) 29 6
.001 <.001 <.001
Comorbidities Hypertension Dyslipidemia Diabetes Coronary artery disease Peripheral vascular disease Smoking Aorta pathology
584 (55%) 560 (53%) 196 (18%) 557 (52%) 37 (3%) 578 (55%) 94 (9%)
120 (63%) 111 (59%) 59 (31%) 106 (55%) 11 (6%) 100 (52%) 20 (11%)
.04 .1 <.001 .5 .1 .8 .5
0.17 0.11 0.29 0.05 0.11 0.4 0.07
1499 (81%) 1379 (75%) 592 (32%) 1340 (72%) 119 (6%) 512 (28%) 94 (9%)
551 (86%) 478 (75%) 233 (37%) 450 (71%) 54 (8%) 196 (31%) 20 (11%)
<.01 .7 .03 .4 .08 .2 .5
0.15 0.02 0.10 0.04 0.08 0.07 0.08
Preoperative medication* Diuretics ACE inhibitors Statins ASA Clopidogrel Vasopressors Inotropes
205 (19%) 435 (41%) 615 (58%) 636 (60%) 260 (25%) 12 (1%) 6 (1%)
60 (32%) 84 (45%) 108 (57%) 124 (65%) 46 (24%) 11 (6%) 7 (4%)
<.001 .4 .9 .2 .9 <.001 <.001
0.28 0.07 0.02 0.10 0.006 0.28 0.22
666 (36%) 956 (52%) 1429 (77%) 1346 (73%) 442 (24%) 21 (1%) 13 (1%)
310 (49%) 327 (51%) 473 (74%) 454 (71%) 166 (26%) 7 (1%) 9 (1%)
<.001 .9 .1 .5 .3 .9 .1
0.26 0.008 0.07 0.03 0.05 0.003 0.07
141 16 215 60 84 [73, 96]
134 20 219 67 79 [65, 96]
<.001 .4 <.01
0.37 0.05 0.09
131 18 205 61 91[77, 108]
126 18 215 73 88 [72, 112]
<.001 <.01 .05
0.32 0.15 0.07
88 [75, 98]
90 [74, 102]
.3
0.09
67 [54, 80]
67 [50, 81]
.5
0.08
479 (45%)
95 (49%)
140 (8%)
71 (11%)
495 (47%) 81 (8%) 6 (1%)
68 (35%) 24 (13%) 5 (3%)
1052 (57%) 60 (33%) 52 (3%)
305 (48%) 233 (37%) 28 (4%)
57 11
54 12
.02
57 11
18 (2%) 12 (1%) 57 (5%) 82 (8%)
10 (5%) 6 (3%) 18 (9%) 27 (14%)
<.001 .03 .03 <.01
0.21 0.14 0.15 0.20
Predicted EuroSCORE II, median [IQR]
1.1 [0.8, 1.8]
1.5 [1.1, 3.2]
<.001
Predicted STS mortality, median [IQR]
0.6 [0.4, 1.2]
0.9 [0.5, 2.4]
Predicted STS renal failure, median [IQR]
1.3 [0.9, 2.1]
2.0 [1.0, 3.8]
Preoperative laboratory valuesy Hemoglobin, g/dL, mean SD Platelets, 3103/mL, mean SD Creatinine, mmol/L, median [IQR] eGFR, mL/min per 1.73 m2, median [IQR] Preoperative CKD No CKD or stage 1, ClCr > 90 mL/min CKD stage 2, ClCr 60-89 mL/min CKD stage 3, ClCr 30-59 mL/min CKD stage 4-5, ClCr < 30 mL/min Preoperative cardiopulmonary status Left ventricular ejection fraction (%, mean SD) Cardiogenic shock Preoperative endotracheal intubation Previous cardiac surgery Intra-aortic balloon pump
<.001
0.15 0.29 0.26
<.001
0.20
55 12
<.001
0.19
27 (1%) 15 (1%) 123 (7%) 132 (7%)
13 (2%) 13 (2%) 49 (8%) 81 (13%)
.3 .01 .4 <.001
0.04 0.10 0.04 0.18
0.52
2.4 [1.5, 4.5]
3.5 [1.9, 6.4]
<.001
0.36
<.001
0.57
2.4 [1.4, 4.5]
3.4 [1.8, 6.0]
<.001
0.40
<.001
0.47
3.9 [2.3, 6.8]
4.9 [2.9, 9.1]
<.001
0.29
0.29
0.19
AKI, Acute kidney injury; Std Diff, standard difference; SD, standard deviation; ACE, angiotensin-converting enzyme; ASA, aspirin; IQR, interquartile range; eGFR, estimated glomerular filtration rate; CKD, chronic kidney disease; ClCr, creatinine clearance; STS, Society of Thoracic Surgeons. *In the 7 days before surgery. yLast preoperative value available.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
3
PM
No AKI n ¼ 1061
Perioperative Management
Saydy et al
TABLE 2. Operative data
Variables
No AKI n ¼ 1061
Patients 60 years old AKI n ¼ 192 P value
PM
Status Elective Urgent (same hospital stay) Extreme urgency (<6 h)
692 (65%) 351 (33%) 18 (2%)
122 (64%) 63 (33%) 7 (4%)
Type of surgery Isolated CABG surgery Isolated valvular surgery Combined CABG þ valvular surgery Other (no valvular or CABG intervention)
495 (47%) 417 (39%) 57 (5%) 92 (9%)
77 (40%) 74 (39%) 28 (15%) 13 (7%)
Approach Sternotomy Minimally invasive
945 (89%) 116 (11%)
183 (95%) 9 (5%)
82 [60,112]
100 [71,152]
261 (25%) 361 (34%) 218 (21%) 104 (10%) 117 (11%) 58 [39,85] 43 (4%) 187 (18%)
30 (16%) 55 (29%) 32 (17%) 26 (14%) 49 (26%) 67 [46,114] 11 (6%) 57 (30%)
<.001 .3 <.001
28 4 29 4 1587 737
27 4 27 4 1671 788
50 (5%) 41 (4%)
26 (14%) 6 (3%)
Intraoperative strategy CPB time, min, median [IQR] CPB time (grouped) CPB <60 min CPB 60-89 min CPB 90-119 min CPB 120-149 min CPB 150 min Crossclamp time, min, median [IQR] Circulatory arrest Hemofiltration Perioperative hemodilution Hematocrit during CPB, g/dL, mean SD Hematocrit post CPB, g/dL, mean SD Volume of hemodilution, L, mean SD Intraoperative complications Difficult CPB weaning Intraoperative dialysis
Std Diff
.2
No AKI n ¼ 1851
Patients 65 years old AKI n ¼ 637 P value
0.15
.3
0.07
<.001
0.22
.3
0.05
<.001 <.001
0.28 0.34
131 (21%) 190 (30%) 150 (24%) 84 (13%) 82 (13%) 62 [41,90] 26 (4%) 181 (28%)
<.001 .03 <.001
0.18 0.10 0.19
26 4 27 4 1528 675
25 5 26 3 1691 797
<.001 <.001 <.001
0.21 0.30 0.21
105 (6%) 36 (2%)
73 (11%) 18 (3%)
<.001 .2
0.20 0.06
1132 (61%) 695 (38%) 24 (1%)
395 (62%) 229 (36%) 13 (2%)
965 (52%) 498 (27%) 339 (18%) 49 (3%)
271 (43%) 176 (28%) 167 (26%) 23 (4%)
1756 (95%) 95 (5%)
611 (96%) 26 (4%)
79 [59,103]
89 [64,121]
0.19 0.09 0.29
464 (25%) 698 (38%) 422 (23%) 182 (10%) 85 (5%) 55 [38,78] 44 (2%) 372 (20%)
<.001 <.001 .2
0.32 0.37 0.10
<.001 .6
0.32 0.04
<.001
0.32
<.01
0.24
<.001 <.001
0.45 0.44
Std Diff
AKI, Acute kidney injury; Std Diff, standard difference; CABG, coronary artery bypass graft; IQR, interquartile range; CPB, cardiopulmonary bypass; SD, standard deviation.
RESULTS Baseline Characteristics A total of 192 patients (18%) developed AKI in the younger group versus 637 (34%) in the older group. Preoperative patient characteristics are presented in Table 1. The mean age was 52 9 years in the younger group and 74 6 years in the older age group. In the younger group, 76% of patients were male, compared with 66% in the older group. In both groups, there was a greater proportion of female patients who developed AKI compared with those who did not. Patients who developed AKI also had a greater mean body mass index (BMI) in both groups. Patients in the younger group who developed AKI had greater rates of hypertension (P ¼ .04) and diabetes (P <.001) compared with those who did not. They were also more likely to be taking diuretics, vasopressors, and inotropes in the week preceding surgery (P <.001). Their last preoperative hemoglobin mean value also was lower (P < .001). Finally, young patients with postoperative 4
AKI had a lower preoperative left ventricular ejection fraction (P ¼ .02) and greater rates of preoperative cardiogenic shock (P ¼ .002), endotracheal intubation (P ¼ .03), previous cardiac surgery (P ¼ .03), and intra-aortic balloon pump (IABP) (P ¼ .004) compared with those who did not develop AKI. These differences between patients who developed postoperative AKI and those who did not were all present in the older group, with the exception of vasopressor (P ¼ .94) and inotrope (P ¼ .10) use, as well as preoperative cardiogenic shock (P ¼ .32) and previous cardiac surgery (P ¼ .37). Patients in both age groups who developed AKI had greater predicted Society of Thoracic Surgeons renal failure scores (P < .001), greater predicted Society of Thoracic Surgeons mortality scores (P <.001), and greater EuroSCORE II scores (P <.001) (Table 1). Operative Details Operative details are shown in Table 2. Most patients underwent elective surgery (63%). The vast majority of
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Perioperative Management
TABLE 3. Early postoperative outcomes
Mortality (30 d) RRT Stroke
Patients 60 years old AKI n ¼ 192
P value
No AKI n ¼ 1851
Patients 65 years old AKI n ¼ 637
P value
3 (0%)
15 (8%)
<.001
21 (1%)
68 (11%)
<.001
29 (15%)
<.001
82 (13%)
<.001
5 (0%)
3 (2%)
.08
17 (1%)
12 (2%)
.05
Prolonged ventilation
50 (5%)
21 (12%)
<.001
135 (7%)
124 (21%)
<.001
Any reoperation
59 (6%)
30 (16%)
<.001
123 (7%)
98 (15%)
<.001
Prolonged 14-d length of stay
36 (3%)
31 (16%)
<.001
123 (7%)
166 (26%)
<.001
Perioperative myocardial infarction
13 (1%)
7 (4%)
.01
16 (1%)
14 (2%)
Low output syndrome Atrial fibrillation Pacemaker
.008
7 (1%)
11 (6%)
<.001
28 (2%)
46 (7%)
<.001
166 (16%)
53 (28%)
<.001
606 (33%)
279 (44%)
<.001
35 (3%)
15 (8%)
71 (4%)
49 (8%)
<.001
.003
Hospital stay, d, median [IQR]
5 [4,7]
7 [5,10]
<.001
5 [4,7]
7 [5,10]
<.001
ICU stay, d, median [IQR]
1 [1,2]
2 [1,5]
<.001
1 [1,2]
2 [1,5]
<.001
Mediastinal drainage, mL, median [IQR]
638 [415, 1040]
845 [570, 1505]
<.001
638 [415, 1040]
845 [570, 1505]
<.001
421 (40%) 343 (32%) 248 (23%) 151 (14%)
130 (68%) 125 (65%) 69 (36%) 57 (30%)
<.001 <.001 <.001 <.001
1150 (62%) 343 (32%) 248 (23%) 151 (14%)
515 (81%) 125 (65%) 69 (36%) 57 (30%)
<.001 <.001 <.001 <.001
Any postoperative transfusion Packed red blood cells Platelets Fresh-frozen plasma
PM
Variables
No AKI n ¼ 1061
AKI, Acute kidney injury; RRT, renal-replacement therapy; ICU, intensive care unit; IQR, interquartile range.
procedures were performed by sternotomy (93%). Operations performed consisted in isolated CABG in 48% of patients, isolated valvular surgery in 31%, combined CABG and valvular surgery in 16%, and other procedures in 5%. These other procedures consisted mainly in ascending aorta and aortic arch operations, as well as septal myectomies. This cohort did not include any patient with complex adult congenital heart disease (eg, single-ventricle physiology, Fontan circulation, etc). In both age groups, patients who developed AKI were more likely to have had a combined CABG and valvular intervention compared with patients with no AKI. In both age groups, patients with AKI had longer CPB and aortic crossclamp times, a greater incidence of difficult CPB weaning, as well as lower hematocrit values both during and after CPB. In patients aged 65 and older—but not in those aged 60 and younger—larger hemodilution volumes and more frequent use of circulatory arrest were seen in patients who developed AKI compared with those who did not. Clinical Impact of AKI Early postoperative outcomes are detailed in Table 3. Among the 192 patients who developed postoperative AKI in the younger group, RRT was required in 15% (n ¼ 29). Among the 637 patients who developed postoperative AKI in the older group, RRT was required in 13%
(n ¼ 82). In both groups, patients who developed AKI had greater 30-day mortality compared with those with no AKI (8% vs 0% in the younger group; 11% vs 1% in the older group). In both groups, patients with AKI also had greater rates of prolonged ventilation, reoperation, perioperative myocardial infarction, low output syndrome, atrial fibrillation, permanent pacemaker requirement, postoperative blood product transfusion, as well as longer hospital and intensive care unit stays. Multivariable Linear Regression Analysis of Variables Associated With Maximum Postoperative Serum Creatinine Increase Variables associated with maximal postoperative serum creatinine increase in the final multivariable models are presented in Tables 4 and 5, as well as in Figure 1. In both younger and older patients, the following variables were associated independently with postoperative increase in creatinine: greater BMI, peripheral vascular disease, use of diuretics, lower preoperative hemoglobin, preoperative IABP, urgent or emergent status, long CPB time, and hemofiltration. In addition, certain variables associated with creatinine increase differed between the 2 groups. Although coronary artery disease was protective in young patients, it was harmful in older patients. In younger patients, a larger increase in
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
5
Perioperative Management
Saydy et al
TABLE 4. Multivariable analysis of factors associated with maximum postoperative serum creatinine increase in patients 60 years old Estimate
Standard error
P value
Demographic data Body mass index Female sex
0.04467 0.06777
0.00118 0.00741
<.001 <.001
Comorbidities Diabetes Coronary artery disease Peripheral vascular disease
0.04953 0.1048 0.1135
0.000129 0.0167 0.0215
<.001 <.001 <.001
Risk factors
Preoperative medication* Diuretics
0.05709
0.01848
.002
0.7288 0.02404
0.04294 0.00343
<.001 <.001
Preoperative cardiopulmonary status Previous cardiac surgery Intra-aortic balloon pump
0.1159 0.07113
0.00271 0.00568
<.001 <.001
Status Urgent or emergent status
0.03824
0.00245
<.001
Intraoperative strategy CPB time Hemofiltration
0.1262 0.05732
0.00919 0.0101
<.001 <.001
0.0283 0.00089 0.0300
.021 <.001 .037
Preoperative laboratory valuesy Last preoperative creatinine Last preoperative hemoglobin
PM
Type of surgery (ref: CABG) Isolated valvular surgery Combined CABG þ valvular surgery Other surgery (no CABG or valvular surgery)
0.1268 0.0989 0.1074
Variables in bold represent significant risk factors in both age groups. CPB, Cardiopulmonary bypass; CABG, coronary artery bypass graft. *In the 7 days before surgery. yLast preoperative value available.
serum creatinine was associated with diabetes (P <.001), whereas in older patients, it was associated with hypertension (P ¼ .003) and smoking (P ¼ .005). Female sex (P<.001) was associated with a lesser degree serum creatinine increase in young patients but not in the older group. Previous cardiac surgery was associated with larger increases in serum creatinine in the younger group (P<.001) but not in the older group. Conversely, lower preoperative left ventricular ejection fraction was associated with larger increases in serum creatinine in the older age group (P <.001) but not in the younger group. In younger patients, isolated valvular and other (ie, neither CABG nor valvular) operations were associated with a lower degree of postoperative serum creatinine increase (P ¼ .021 and P ¼ .037, respectively), whereas combined CABG and valvular surgery was associated with a greater degree of creatinine increase (P <.001). In contrast, in older patients, isolated valvular and other operations were associated with a greater degree of postoperative serum creatinine increase (P <.001 and P ¼ .013, respectively). Interaction analyses did not significantly alter either multivariable model. DISCUSSION Young adults represent a specific population in the realm of cardiac disease. In contrast to older patients, the impact 6
of atherosclerotic risk factors on their vasculature and overall physiology has been less sustained and therefore may be less clinically relevant at the time of surgery. In this study, we sought to specifically determine the risk factors of AKI in young adults undergoing cardiac surgery and compare these with risk factors observed in an older patient population with more established atherosclerotic disease. Interestingly, the main findings of this study suggest that although a number of risk factors are present across both age groups, others appear to be specific to either younger or older patients. The present study identified associations between postoperative creatinine increase and the following preoperative factors in all age groups: high BMI, coronary artery disease, peripheral vascular disease, use of diuretics within 7 days before surgery, high preoperative creatinine, low preoperative hemoglobin, and presence of an IABP. The link between obesity and postoperative outcomes after cardiac surgery is elusive. In the last decade, several investigators have observed a trend referred to as the ‘‘obesity paradox.’’ Despite being at an increased cardiovascular risk and having more comorbidities, patients with a greater BMI do not have less favorable outcomes after cardiac surgery.6-10 Whether this ‘‘paradox’’ applies to postoperative AKI is a matter of debate. Although some studies have found no association between BMI and AKI,3,11 others observed a
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Perioperative Management
TABLE 5. Multivariable analysis of factors associated with maximum postoperative serum creatinine increase in patients 65 years old Estimate
Standard error
P value
Demographic data Age Body mass index
0.00487 0.0484
0.000967 0.003282
<.001 <.001
Comorbidities Hypertension Coronary artery disease Peripheral vascular disease Smoking
0.0388 0.0380 0.0364 0.0248
0.01293 0.007696 0.005607 0.008833
.003 <.001 <.001 .005
Preoperative medication* Diuretics
0.0327
0.01291
.011
Preoperative laboratory valuesy Last preoperative creatinine Last preoperative hemoglobin
0.7977 0.0238
0.03979 0.003047
<.001 <.001
Preoperative cardiopulmonary status Intra-aortic balloon pump Left ventricular ejection fraction
0.106 0.00144
0.01094 0.000384
<.001 <.001
Status Urgent or emergent status
0.0109
0.000072
<.001
Intraoperative strategy CPB time Hemofiltration
0.1302 0.0523
0.02056 0.01307
<.001 <.001
Type of surgery (ref: CABG) Isolated valvular surgery Combined CABG þ valvular surgery Other (no CABG or valvular)
0.0287 0.0107 0.0441
0.001912 0.01734 0.008211
<.001 .58 .013
PM
Risk factors
Variables in bold represent significant risk factors in both age groups. CPB, Cardiopulmonary bypass; CABG, coronary artery bypass graft. *In the 7 days before surgery. yLast preoperative value available.
decreased rate of AKI and mortality in obese patients.8,10 In contrast, our study demonstrated a clear association between increased BMI and postoperative serum creatinine increase, further contributing to the equipoise surrounding this question. Anemia, preoperative use of diuretics, preexisting kidney disease, and preoperative IABP insertion have been identified previously as potentially modifiable risk factors of AKI after cardiac surgery.12-16 Of these, anemia is the most readily modifiable risk factor. Anemia contributes to the pathogenesis of AKI by reducing tissue oxygen delivery, since hemoglobin is the main factor that determines arterial oxygen concentration.17 One might have expected to see a lesser effect of preoperative anemia in young adults because of the shorter exposure to atherosclerotic risk factors in these patients, and a presumably greater ischemic reserve. Nonetheless, in the present study, multivariable analyses showed that low preoperative hemoglobin levels were associated with an increase in postoperative serum creatinine levels in both age groups. These findings suggest that just like their older counterpart, younger patients are vulnerable to hypoxemic renal injury secondary to preoperative anemia and that measures aimed at increasing preoperative hemoglobin levels should be sought in these young
adults undergoing elective cardiac procedures, similar to what is performed in older patients. The nephrotoxic effect of diuretics in the context of cardiac surgery can be explained by the association of volume depletion and renal hypoperfusion,18 resulting in renal hypoxia, one of the main mechanisms of AKI. The association between diuretic use and postoperative creatinine increase observed in this study highlights the fact that diuretics should only be administered when specifically indicated, both preoperatively and in the postoperative setting.15,18,19 High preoperative creatinine—in the setting of chronic kidney disease or surgery after cardiogenic shock with end-organ damage—is also known to be associated with postoperative AKI.3,20 This study also identified several risk factors for postoperative serum creatinine increase that differed between younger and older patients. Diabetes mellitus—previously identified as a risk factor for postoperative AKI13—was associated independently with creatinine increase in patients 60 years of age and younger but not in those 65 years of age and older. Conversely, atherosclerotic risk factors such as smoking and hypertension were identified as independent risk factors in the older patient group but not in patients 60 years of age and younger. This is likely due to the
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
7
Perioperative Management
Saydy et al
PM FIGURE 1. Factors associated with postoperative increase in serum creatinine on multivariable linear regression analysis. Variables associated with postoperative increase in serum creatinine in patients 60 years (green) only are presented in the top panel. Variables associated with postoperative increase in serum creatinine in patients 65 years (red) only are presented in the bottom panel. Variables associated with postoperative increase in serum creatinine in patients in both age groups are presented in the middle panel. Estimates are presented with 95% confidence interval bars. CABG, Coronary artery bypass graft; BMI, body mass index; PVD, peripheral vascular disease; Hb, hemoglobin; IABP, intra-aortic balloon pump; CPB, cardiopulmonary bypass; LVEF, left ventricular ejection fraction.
8
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Perioperative Management
(I2 ¼ 96%).23 Our finding of a greater degree of creatinine increase in young and middle-aged adults undergoing CABG compared with valvular surgery may be due to the fact that patients who undergo CABG before the age of 60 typically have very aggressive atherosclerotic disease, which may predispose them to renal injury. The main limitation of this study is its retrospective, observational nature. Even though all patients aged 60 years and less were included in this study, most patients were closer to 60 years of age (mean age: 52 years). Furthermore, outcomes beyond the immediate perioperative period were unfortunately not readily available for analysis, such that no comment can be made on the differential long-term impact of postoperative AKI between younger and older patients. Finally, comparisons between patients in the younger and older groups should be interpreted cautiously as these 2 groups had different baseline and operative characteristics, which were not accounted for in the analysis. Despite these limitations, this study is, to our knowledge, the first to specifically examine the risk factors for postoperative creatinine increase in young and middle-aged adults undergoing cardiac surgery. CONCLUSIONS This study identified both common and distinct factors associated with postoperative increase in serum creatinine between patients aged 60 years and younger and those 65 years and older undergoing cardiac surgery. Of note, all modifiable (ie, preoperative hemoglobin, preoperative diuretic use) or potentially modifiable (ie, preoperative IABP, CPB duration) risk factors were present in both groups, highlighting the importance of preoperative and intraoperative strategies aimed at mitigating these various factors in all patients, regardless of age. Conflict of Interest Statement Authors have nothing to disclose with regard to commercial support. References 1. Mariscalco G, Lorusso R, Dominici C, Renzulli A, Sala A. Acute kidney injury: a relevant complication after cardiac surgery. Ann Thorac Surg. 2011;92:1539-47. 2. Rosner MH, Okusa MD. Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol. 2006;1:19-32. 3. Loef BG, Epema AH, Smilde TD, Henning RH, Ebels T, Navis G, et al. Immediate post-operative renal function deterioration in cardiac surgical patients predicts in-hospital mortality and long-term survival. J Am Soc Nephrol. 2005;16: 195-200. 4. Chertow GM, Levy EM, Hammermeister KE, Grover F, Daley J. Independent association between acute renal failure and mortality following cardiac surgery. Am J Med. 1998;104:343-8. 5. Tibshirani R. Regression shrinkage and selection via the lasso. J R Stat Soc Ser B. 1996;58:267-88. 6. Stamou SC, Nussbaum M, Stiegel RM, Reames RK, Skipper ER, Robicsek F, et al. Effect of body mass index on outcomes after cardiac surgery: is there an obesity paradox? Ann Thorac Surg. 2001;91:42-7.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
9
PM
accumulation of repeated insults to the renal and coronary microvasculature in these older patients, which ultimately leads to decreased autoregulation of renal perfusion.21,22 This hypothesis is further supported by the observation that age—a known risk factor for impaired renal autoregulation—was associated significantly with postoperative creatinine increase in the older patient group but not in patients 60 years and younger. The present study also identified several operative variables associated with postoperative increase in serum creatinine for both younger and older patients: long duration of CPB, hemofiltration, and urgent operative status. This is in line with findings from a recently published metaanalysis that found CPB duration and emergent operative status to be associated with AKI.23 CPB duration is a well-established risk factor for AKI in the general cardiac surgery patient population.24-28 Other variables, such as lower CPB temperatures, low CPB flow and a large difference between preoperative mean arterial pressure and mean arterial pressure during CPB, also have been shown to be associated with AKI.29-33 We hypothesized that CPB duration may not have such a direct and marked impact on postoperative renal function in younger patients. However, findings from the present study support the notion that young patients are also vulnerable to the cytotoxic effects of prolonged CPB. A large number of young adults undergoing cardiac surgical procedures often require long operative times because of the complex nature of their lesions, requiring concomitant procedures or the need for hypothermic circulatory arrest in aortic pathologies. In these patients, efforts should be made during CPB to ensure adequate perfusion flow and pressure, optimize hematocrit levels, and limit the use of vasopressor agents during CPB. Another interesting finding of this study is the differential impact of the type of surgery on postoperative creatinine increase between younger and older patients. Specifically, in patients 65 years or older, isolated CABG was associated with a lower postoperative increase in creatinine compared with isolated valvular or ‘‘other’’ operations. In contrast, in patients 60 years or younger, isolated CABG was associated with a greater postoperative increase in creatinine compared with isolated valvular or ‘‘other’’ operations but with a lower increase in creatinine compared with combined CABG and valvular surgery. The association between combined operations (ie, CABG and valvular) and greater rates of AKI is not surprising and has been previously demonstrated.20 However, in the case of isolated procedures, the differential association of CABG and valvular operations with AKI in either age group is more controversial. This uncertainty was highlighted in a recently published metaanalysis that showed no significant association between AKI and type of surgery (CABG vs valvular) but identified very high heterogeneity among published studies
Perioperative Management
PM
7. Le-Bert G, Santana O, Pineda AM, Zamora C, Lamas GA, Lamelas J. The obesity paradox in elderly obese patients undergoing coronary artery bypass surgery. Interact Cardiovasc Thorac Surg. 2011;13:124-7. 8. Engel AM, McDonough S, Smith JM. Does an obese body mass index affect hospital outcomes after coronary artery bypass graft surgery? Ann Thorac Surg. 2009;88:1793-800. 9. van Straten AH, Bramer S, Soliman Hamad MA, van Zundert AA, Martens EJ, Schonberger JP, et al. Effect of body mass index on early and late mortality after coronary artery bypass grafting. Ann Thorac Surg. 2010;89:30-7. 10. Thourani VH, Keeling WB, Kilgo PD, Puskas JD, Lattouf OM, Chen EP, et al. The impact of body mass index on morbidity and short- and long-term mortality in cardiac valvular surgery. J Thorac Cardiovasc Surg. 2011;142:1052-61. 11. Engleberger L, Suri RM, Connolly HM, Li Z, Abel MD, Greason KL, et al. Increased risk of acute kidney injury in patients undergoing tricuspid valve surgery. Eur J Cardiothorac Surg. 2013;43:993-9. 12. Karkouti K, Wijeysundera DN, Beattie WS. Risk associated with pre-operative anemia in cardiac surgery: a multicenter cohort study. Circulation. 2008;117:478-84. 13. Karkouti K, Wijeysundera DN, Yau TM, Callum JL, Cheng DC, Crowther M, et al. Acute kidney injury after cardiac surgery: focus on modifiable risk factors. Circulation. 2009;119:495-502. 14. Chertow GM, Lazarus JM, Christiansen CL, Cook F, Hammermeister KE, Grover F, et al. Pre-operative renal stratification. Circulation. 1997;95:878-84. 15. Lassnigg A, Donner E, Grubhofer G, Presterl E, Druml W, Huesmayr M. Lack of renoprotective effects of dopamine and furosemide during cardiac surgery. J Am Soc Nephrol. 2000;11:97-104. 16. Lombardi R, Ferreiro A, Servetto C. Renal function after cardiac surgery: adverse effect of furosemide. Renal Fail. 2003;25:775-86. 17. Nangaku M. Chronic hypoxia and tubulointerstitial injury: a final common pathway to end-stage renal failure. J Am Soc Nephrol. 2006;17:17-25. 18. Lombardi R, Ferreiro A. Risk factors profile for acute kidney injury after cardiac surgery is different according to the level of baseline renal function. Ren Fail. 2008;30:155-60. 19. Thiele RH, Isbell JM, Rosner MH. Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol. 2015;10:500-14. 20. Thakar CV, Arrigain S, Worley S, Yared JP, Paganini EP. A clinical score to predict acute renal failure after cardiac surgery. J Am Soc Nephrol. 2005;16: 162-8. 21. Abuelo JG. Normotensive acute renal failure. N Engl J Med. 2007;357:797-805. 22. Bidani AK, Polichnowski AJ, Loutzenhiser R, Griffith KA. Renal microvascular dysfunction, hypertension and CKD progression. Curr Opin Nephrol Hypertens. 2013;22:1-9.
10
Saydy et al
23. Yi Q, Li K, Jian Z, Xiao YB, Chen L, Zhang Y, et al. Risk factors for acute kidney injury after cardiovascular surgery: evidence from 2157 cases and 49 777 controls—a meta-analysis. Cardiorenal Med. 2016;6:237-50. 24. Mehta RH, Castelvecchio S, Ballotta A, Frigiola A, Bossone E, Ranucci M. Association of gender and lowest hematocrit on cardiopulmonary bypass with acute kidney injury and operative mortality in patients undergoing cardiac surgery. Ann Thorac Surg. 2013;96:133-40. 25. Karkouti K, Beattie WS, Wijeysundera DN, Rao V, Chan C, Dattilo KM, et al. Hemodilution during cardiopulmonary bypass is an independent risk factor for acute renal failure in adult cardiac surgery. J Thorac Cardiovasc Surg. 2005;129:391-400. 26. Bove T, Calabro MG, Landoni G, Aletti G, Marino G, Crescenzi G, et al. The incidence and risk of acute renal failure after cardiac surgery. J Cardiothorac Vasc Anesth. 2004;18:442-5. 27. Suen WS, Mok CK, Chiu SW, Cheung KL, Lee WT, Cheung D, et al. Risk factors for development of acute renal failure (ARF) requiring dialysis in patients undergoing cardiac surgery. Angiology. 1998;49:789-800. 28. Sirvinskas E, Andrejaitiene J, Raliene L, Nasvytis L, Karbonskiene A, Pilvinis V, et al. Cardiopulmonary bypass management and acute renal failure: risk factors and prognosis. Perfusion. 2008;23:323-7. 29. Kourliouros A, Valencia O, Philips SD, Collinson PO, van Besouw JP, Jahangiri M. Low cardiopulmonary bypass perfusion temperatures are associated with acute kidney injury following coronary artery bypass surgery. Eur J Cardiothorac Surg. 2010;37:704-9. 30. Kanji HD, Schulze CJ, Hervas-Malo M, Wang P, Ross DB, Zibdawi M, et al. Difference between pre-operative and cardiopulmonary bypass mean arterial pressure is independently associated with early cardiac surgery-associated acute kidney injury. J Cardiothorac Surg. 2010;5:71. 31. Ranucci M, Pavesi M, Mazza E, Bertucci C, Frigiola A, Menicanti L, et al. Risk factors for renal dysfunction after coronary surgery: the role of cardiopulmonary bypass technique. Perfusion. 1994;9:319-26. 32. Fang WC, Helm RE, Krieger KH, Rosengart TK, DuBois WJ, Sason C, et al. Impact of minimum hematocrit during cardiopulmonary bypass on mortality in patients undergoing coronary artery surgery. Circulation. 1997;96(suppl II):194-9. 33. Swaminathan M, Phillips-Bute BG, Conlon PJ, Smith PK, Newman MF, Stafford-Smith M. The association of lowest hematocrit during cardiopulmonary bypass with acute renal injury after coronary artery bypass surgery. Ann Thorac Surg. 2003;76:84-92.
Key Words: acute kidney injury, cardiac surgery, young patients
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Differences and similarities in risk factors for postoperative acute kidney injury between younger and older adults undergoing cardiac surgery Nadim Saydy, Amine Mazine, MD, MSc, Louis-Mathieu Stevens, MD, PhD, Hughes Jeamart, MD, MSc, Philippe Demers, MD, MSc, Pierre Page, MD, Yoan Lamarche, MD, MSc, and Ismail ElHamamsy, MD, PhD, Montreal, Quebec, Canada This study identified both common and distinct risk factors associated with postoperative increase in serum creatinine between patients 60 years and those 65 years undergoing cardiac surgery.
PM
000
Perioperative Management
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
Perioperative Management
Differences and similarities in risk factors for postoperative acute kidney injury between younger and older adults undergoing cardiac surgery Nadim Saydy,a Amine Mazine, MD, MSc,a Louis-Mathieu Stevens, MD, PhD,b Hughes Jeamart, MD, MSc,a,c Philippe Demers, MD, MSc,a,c Pierre Page, MD,a,c Yoan Lamarche, MD, MSc,a,c and Ismail El-Hamamsy, MD, PhDa,c ABSTRACT Objectives: Acute kidney injury is a frequent complication after cardiac surgery. The purpose of this study was to assess the risk factors for acute kidney injury in patients 60 years of age undergoing cardiac surgery and to compare these risk factors with those identified in patients 65 years of age.
Results: In both age groups, the following variables were associated independently with greater degrees of postoperative increase in serum creatinine on multivariable analysis: greater body mass index, peripheral vascular disease, preoperative use of diuretics, lower preoperative hemoglobin, preoperative intra-aortic balloon pump, urgent or emergent status, long cardiopulmonary bypass time, and hemofiltration. In younger patients, a greater increase in serum creatinine was associated with diabetes, and previous cardiac surgery, whereas female sex was associated with a lower degree of increase in serum creatinine. In older patients, a greater increase in serum creatinine was associated with age, hypertension, smoking, and lower left ventricular left ejection fraction. Operation type and coronary artery disease had a different impact on postoperative creatinine increase between younger and older patients. Conclusions: This study identified both common and distinct risk factors associated with postoperative increase in serum creatinine between patients 60 years and those 65 years undergoing cardiac surgery. Importantly, all potentially modifiable risk factors were present in both groups. (J Thorac Cardiovasc Surg 2017;-:1-10)
From the aDivision of Cardiac Surgery, Montreal Heart Institute, bDivision of Cardiac Surgery, Centre Hospitalier Universitaire de Montreal, and cDivision of Cardiac Surgery, Sacre-Cœur Hospital, University of Montreal School of Medicine, Montreal, Quebec, Canada. N.S. and A.M. contributed equally to this work and share co–first authorship. Received for publication Oct 4, 2016; revisions received Aug 1, 2017; accepted for publication Aug 18, 2017. Address for reprints: Ismail El-Hamamsy, MD, PhD, Department of Surgery, University of Montreal, Division of Cardiac Surgery, Montreal Heart Institute, 5000 Belanger St, Montreal, Quebec H1T 1C8, Canada (E-mail: i.elhamamsy@ icm-mhi.org). 0022-5223/$36.00 Copyright Ó 2017 by The American Association for Thoracic Surgery http://dx.doi.org/10.1016/j.jtcvs.2017.08.039
Factors associated with postoperative increase in creatinine on multivariable analysis. Central Message This study identified both common and distinct risk factors associated with postoperative increase in serum creatinine between patients 60 years and those 65 years undergoing cardiac surgery.
PM
Methods: From 2010 to 2012, 1253 patients 60 years (mean age 52 9 years) and 2488 patients 65 years (mean age 74 6 years) underwent cardiac surgery. Linear regression models using least absolute shrinkage and selection operator methods and mixed effects linear regression models were used to assess factors associated with maximum postoperative increase in serum creatinine in these two cohorts.
Perspective Postoperative acute kidney injury (AKI) is an important predictor of morbidity and mortality after cardiac surgery. Several risk factors have been identified; however, few studies have focused on young and middle-aged adults. We found that AKI is associated with negative outcomes, even in these younger patients. We identified both common and distinct risk factors for AKI between patients 60 years and those 65 years.
See Editorial Commentary page XXX.
Acute kidney injury (AKI) after cardiac surgery is a frequent complication that occurs in up to 40% of patients,1 with dialysis required in 1% to 2% of all patients.2 Importantly, AKI has been associated with increased in-hospital mortality and reduced long-term survival.3,4 In view of the lack of active Scanning this QR code will take you to the article title page.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
1
Perioperative Management
Abbreviations and Acronyms AKI ¼ acute kidney injury BMI ¼ body mass index CABG ¼ coronary artery bypass graft CPB ¼ cardiopulmonary bypass IABP ¼ intra-aortic balloon pump LASSO ¼ least absolute shrinkage and selection operator RRT ¼ renal-replacement therapy
PM
treatment for AKI, prevention and management of risk factors remain critical. Many studies have sought to identify preoperative and perioperative risk factors that predispose to AKI after cardiac surgery; however, few have focused specifically on young and middle-aged adults. The general aim of this study was to assess the risk factors for AKI in patients 60 years of age and younger undergoing cardiac surgery and to compare these risk factors with those identified in patients 65 years of age and older. The specific goals were to (1) compare the baseline and operative characteristics, as well as the postoperative clinical outcomes, between patients who did and did not develop postoperative AKI in each age group independently and (2) to model the maximum increase in postoperative serum creatinine by the use of linear regression techniques to identify specific and common risk variables in the younger and older patients. Our hypothesis was that traditional risk factors, such as atherosclerosis related factors, are less likely to be associated with AKI in younger patients than in those older than 65 years. This cohort did not include patients with complex adult congenital heart disease. METHODS Study Population The study was approved by the institutional review board, and a waiver of individual patient consent was obtained. This is a retrospective analysis of data collected between October 2010 and December 2012 in 2 centers affiliated with the University of Montreal: Montreal Heart Institute and Sacre-Cœur Hospital. All consecutive patients older than 18 years of age undergoing cardiac surgery were screened for eligibility. Exclusion criteria were age 61 to 64 years (n ¼ 588), off-pump coronary artery bypass grafting (CABG) surgery (n ¼ 293), and renal-replacement therapy (RRT) before surgery (n ¼ 32). After the exclusion criteria were applied, the final cohort consisted of 3741 consecutive patients. Of these, 1253 were 60 years of age or younger (study group), and 2488 were 65 years of age or older (control group). Their procedures were performed by 1 of 12 surgeons involved in the study. Preoperative medication was reviewed for the 7 days before surgery. Angiotensin-converting enzyme inhibitors and/or diuretic use were not discontinued before surgery in all patients. During the study period, tranexamic acid was the only antithrombolytic agent used. Data collected included baseline preoperative characteristics, operative details, and postoperative outcomes.
Definitions Patients who required postoperative dialysis or who presented a greater than 50% increase in serum creatinine in the first 7 postoperative days
2
Saydy et al
compared with the last baseline value before surgery were deemed to have AKI. Operative mortality was defined as death occurring within 30 days of surgery or during the index hospitalization. Surgery was considered urgent in patients who had not been electively admitted for operation but who underwent surgery during the same admission for medical reasons. Emergent operation was defined as operation within 6 hours of admission. Weaning from cardiopulmonary bypass (CPB) was considered difficult if any of the following was required: 3 vasoactive agents at low dose, 2 vasoactive agents at high dose, return on CPB, or mechanical support.
Statistical Analysis The general statistical plan for this study was to first compare the baseline and operative characteristics, as well as the postoperative clinical outcomes, between patients who did and did not develop postoperative AKI in the younger (60 years) and older (65 years) age groups independently and to then model the maximum increase in postoperative serum creatinine by the use of linear regression techniques to identify specific and common risk variables in each group. The specific statistical methods used to achieve these goals are detailed below. Baseline characteristics, intraoperative data, and early outcomes were compared within each group—ie, patients 60 years and 65 years— between patients who developed postoperative AKI and those who did not. Continuous variables are expressed as mean standard deviation (when normally distributed) or median and interquartile range (when not normally distributed), and were compared by means of Student t test or the Mann-Whitney U test, as appropriate. Categorical variables are expressed as frequency (%) and were compared by means of Pearson c2 test or Fisher exact test, as appropriate. Factors associated with maximum postoperative increase in serum creatinine were identified via a combination of 2 methods. First, linear regression models using least absolute shrinkage and selection operator (LASSO) selection methods were used to identify candidate variables.5 Variables assessed as potential risk factors were the preoperative baseline characteristics described in Table 1, as well as the operative data presented in Table 2. For variables that were not distributed according to the normal distribution (eg, serum creatinine levels, CPB duration), a logarithmic transformation was used. The LASSO is a regression analysis method that applies a penalty to large coefficients, thus resulting in more interpretable regression models. Because of its ability to filter through a large number of variables and shrink effect coefficients, the LASSO method is especially useful when a large number of potential variables are available for model fitting. In addition, the LASSO mitigates many of the issues associated with the stepwise variable selection method used in traditional regression models, such as overfitting resulting from multicollinearity between covariates. To account for the effect of the individual surgeon and institution, multivariable analyses using mixed effects linear regression models were then conducted with the variables identified through LASSO (the MIXED procedure, SAS software, version 9.4; SAS Institute, Cary, NC). Every mixedeffect model included the last baseline preoperative serum creatinine value as a covariate. Only factors that were significant under both the LASSOselected and mixed effects models were considered to be validated for the final model. This analysis sequence—ie, variable preselection through LASSO modeling followed by validation through mixed effect linear regression modeling—was first applied separately to patients 60 years and those 65 years, resulting in 2 distinct models. In the final step of model building, all variables identified through the separate models were forced into a common model that included all patients, and all 2-way interaction effects between these variables and age group were examined. Of note, the multivariable model was a linear regression model aimed at modeling maximum postoperative increase in serum creatinine, not a logistic regression model aimed at modeling the incidence of AKI. All analyses were carried out using the SAS software, version 9.4 (SAS Institute). Statistical significance was set at a ¼ 0.05.
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Perioperative Management
TABLE 1. Baseline characteristics
Variables Demographic data Age, y, mean SD Female sex Body mass index, kg/m2, mean SD
Patients 60 years old AKI n ¼ 192 P value
Std Diff
No AKI n ¼ 1851
Patients 65 years old AKI n ¼ 637 P value
Std Diff
52 9 241 (23%) 28 5
52 8 62 (32%) 30 7
.5 <.01 <.001
0.06 0.21 0.30
74 6 556 (30%) 28 5
74 6 280 (44%) 29 6
.001 <.001 <.001
Comorbidities Hypertension Dyslipidemia Diabetes Coronary artery disease Peripheral vascular disease Smoking Aorta pathology
584 (55%) 560 (53%) 196 (18%) 557 (52%) 37 (3%) 578 (55%) 94 (9%)
120 (63%) 111 (59%) 59 (31%) 106 (55%) 11 (6%) 100 (52%) 20 (11%)
.04 .1 <.001 .5 .1 .8 .5
0.17 0.11 0.29 0.05 0.11 0.4 0.07
1499 (81%) 1379 (75%) 592 (32%) 1340 (72%) 119 (6%) 512 (28%) 94 (9%)
551 (86%) 478 (75%) 233 (37%) 450 (71%) 54 (8%) 196 (31%) 20 (11%)
<.01 .7 .03 .4 .08 .2 .5
0.15 0.02 0.10 0.04 0.08 0.07 0.08
Preoperative medication* Diuretics ACE inhibitors Statins ASA Clopidogrel Vasopressors Inotropes
205 (19%) 435 (41%) 615 (58%) 636 (60%) 260 (25%) 12 (1%) 6 (1%)
60 (32%) 84 (45%) 108 (57%) 124 (65%) 46 (24%) 11 (6%) 7 (4%)
<.001 .4 .9 .2 .9 <.001 <.001
0.28 0.07 0.02 0.10 0.006 0.28 0.22
666 (36%) 956 (52%) 1429 (77%) 1346 (73%) 442 (24%) 21 (1%) 13 (1%)
310 (49%) 327 (51%) 473 (74%) 454 (71%) 166 (26%) 7 (1%) 9 (1%)
<.001 .9 .1 .5 .3 .9 .1
0.26 0.008 0.07 0.03 0.05 0.003 0.07
141 16 215 60 84 [73, 96]
134 20 219 67 79 [65, 96]
<.001 .4 <.01
0.37 0.05 0.09
131 18 205 61 91[77, 108]
126 18 215 73 88 [72, 112]
<.001 <.01 .05
0.32 0.15 0.07
88 [75, 98]
90 [74, 102]
.3
0.09
67 [54, 80]
67 [50, 81]
.5
0.08
479 (45%)
95 (49%)
140 (8%)
71 (11%)
495 (47%) 81 (8%) 6 (1%)
68 (35%) 24 (13%) 5 (3%)
1052 (57%) 60 (33%) 52 (3%)
305 (48%) 233 (37%) 28 (4%)
57 11
54 12
.02
57 11
18 (2%) 12 (1%) 57 (5%) 82 (8%)
10 (5%) 6 (3%) 18 (9%) 27 (14%)
<.001 .03 .03 <.01
0.21 0.14 0.15 0.20
Predicted EuroSCORE II, median [IQR]
1.1 [0.8, 1.8]
1.5 [1.1, 3.2]
<.001
Predicted STS mortality, median [IQR]
0.6 [0.4, 1.2]
0.9 [0.5, 2.4]
Predicted STS renal failure, median [IQR]
1.3 [0.9, 2.1]
2.0 [1.0, 3.8]
Preoperative laboratory valuesy Hemoglobin, g/dL, mean SD Platelets, 3103/mL, mean SD Creatinine, mmol/L, median [IQR] eGFR, mL/min per 1.73 m2, median [IQR] Preoperative CKD No CKD or stage 1, ClCr > 90 mL/min CKD stage 2, ClCr 60-89 mL/min CKD stage 3, ClCr 30-59 mL/min CKD stage 4-5, ClCr < 30 mL/min Preoperative cardiopulmonary status Left ventricular ejection fraction (%, mean SD) Cardiogenic shock Preoperative endotracheal intubation Previous cardiac surgery Intra-aortic balloon pump
<.001
0.15 0.29 0.26
<.001
0.20
55 12
<.001
0.19
27 (1%) 15 (1%) 123 (7%) 132 (7%)
13 (2%) 13 (2%) 49 (8%) 81 (13%)
.3 .01 .4 <.001
0.04 0.10 0.04 0.18
0.52
2.4 [1.5, 4.5]
3.5 [1.9, 6.4]
<.001
0.36
<.001
0.57
2.4 [1.4, 4.5]
3.4 [1.8, 6.0]
<.001
0.40
<.001
0.47
3.9 [2.3, 6.8]
4.9 [2.9, 9.1]
<.001
0.29
0.29
0.19
AKI, Acute kidney injury; Std Diff, standard difference; SD, standard deviation; ACE, angiotensin-converting enzyme; ASA, aspirin; IQR, interquartile range; eGFR, estimated glomerular filtration rate; CKD, chronic kidney disease; ClCr, creatinine clearance; STS, Society of Thoracic Surgeons. *In the 7 days before surgery. yLast preoperative value available.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
3
PM
No AKI n ¼ 1061
Perioperative Management
Saydy et al
TABLE 2. Operative data
Variables
No AKI n ¼ 1061
Patients 60 years old AKI n ¼ 192 P value
PM
Status Elective Urgent (same hospital stay) Extreme urgency (<6 h)
692 (65%) 351 (33%) 18 (2%)
122 (64%) 63 (33%) 7 (4%)
Type of surgery Isolated CABG surgery Isolated valvular surgery Combined CABG þ valvular surgery Other (no valvular or CABG intervention)
495 (47%) 417 (39%) 57 (5%) 92 (9%)
77 (40%) 74 (39%) 28 (15%) 13 (7%)
Approach Sternotomy Minimally invasive
945 (89%) 116 (11%)
183 (95%) 9 (5%)
82 [60,112]
100 [71,152]
261 (25%) 361 (34%) 218 (21%) 104 (10%) 117 (11%) 58 [39,85] 43 (4%) 187 (18%)
30 (16%) 55 (29%) 32 (17%) 26 (14%) 49 (26%) 67 [46,114] 11 (6%) 57 (30%)
<.001 .3 <.001
28 4 29 4 1587 737
27 4 27 4 1671 788
50 (5%) 41 (4%)
26 (14%) 6 (3%)
Intraoperative strategy CPB time, min, median [IQR] CPB time (grouped) CPB <60 min CPB 60-89 min CPB 90-119 min CPB 120-149 min CPB 150 min Crossclamp time, min, median [IQR] Circulatory arrest Hemofiltration Perioperative hemodilution Hematocrit during CPB, g/dL, mean SD Hematocrit post CPB, g/dL, mean SD Volume of hemodilution, L, mean SD Intraoperative complications Difficult CPB weaning Intraoperative dialysis
Std Diff
.2
No AKI n ¼ 1851
Patients 65 years old AKI n ¼ 637 P value
0.15
.3
0.07
<.001
0.22
.3
0.05
<.001 <.001
0.28 0.34
131 (21%) 190 (30%) 150 (24%) 84 (13%) 82 (13%) 62 [41,90] 26 (4%) 181 (28%)
<.001 .03 <.001
0.18 0.10 0.19
26 4 27 4 1528 675
25 5 26 3 1691 797
<.001 <.001 <.001
0.21 0.30 0.21
105 (6%) 36 (2%)
73 (11%) 18 (3%)
<.001 .2
0.20 0.06
1132 (61%) 695 (38%) 24 (1%)
395 (62%) 229 (36%) 13 (2%)
965 (52%) 498 (27%) 339 (18%) 49 (3%)
271 (43%) 176 (28%) 167 (26%) 23 (4%)
1756 (95%) 95 (5%)
611 (96%) 26 (4%)
79 [59,103]
89 [64,121]
0.19 0.09 0.29
464 (25%) 698 (38%) 422 (23%) 182 (10%) 85 (5%) 55 [38,78] 44 (2%) 372 (20%)
<.001 <.001 .2
0.32 0.37 0.10
<.001 .6
0.32 0.04
<.001
0.32
<.01
0.24
<.001 <.001
0.45 0.44
Std Diff
AKI, Acute kidney injury; Std Diff, standard difference; CABG, coronary artery bypass graft; IQR, interquartile range; CPB, cardiopulmonary bypass; SD, standard deviation.
RESULTS Baseline Characteristics A total of 192 patients (18%) developed AKI in the younger group versus 637 (34%) in the older group. Preoperative patient characteristics are presented in Table 1. The mean age was 52 9 years in the younger group and 74 6 years in the older age group. In the younger group, 76% of patients were male, compared with 66% in the older group. In both groups, there was a greater proportion of female patients who developed AKI compared with those who did not. Patients who developed AKI also had a greater mean body mass index (BMI) in both groups. Patients in the younger group who developed AKI had greater rates of hypertension (P ¼ .04) and diabetes (P <.001) compared with those who did not. They were also more likely to be taking diuretics, vasopressors, and inotropes in the week preceding surgery (P <.001). Their last preoperative hemoglobin mean value also was lower (P < .001). Finally, young patients with postoperative 4
AKI had a lower preoperative left ventricular ejection fraction (P ¼ .02) and greater rates of preoperative cardiogenic shock (P ¼ .002), endotracheal intubation (P ¼ .03), previous cardiac surgery (P ¼ .03), and intra-aortic balloon pump (IABP) (P ¼ .004) compared with those who did not develop AKI. These differences between patients who developed postoperative AKI and those who did not were all present in the older group, with the exception of vasopressor (P ¼ .94) and inotrope (P ¼ .10) use, as well as preoperative cardiogenic shock (P ¼ .32) and previous cardiac surgery (P ¼ .37). Patients in both age groups who developed AKI had greater predicted Society of Thoracic Surgeons renal failure scores (P < .001), greater predicted Society of Thoracic Surgeons mortality scores (P <.001), and greater EuroSCORE II scores (P <.001) (Table 1). Operative Details Operative details are shown in Table 2. Most patients underwent elective surgery (63%). The vast majority of
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Perioperative Management
TABLE 3. Early postoperative outcomes
Mortality (30 d) RRT Stroke
Patients 60 years old AKI n ¼ 192
P value
No AKI n ¼ 1851
Patients 65 years old AKI n ¼ 637
P value
3 (0%)
15 (8%)
<.001
21 (1%)
68 (11%)
<.001
29 (15%)
<.001
82 (13%)
<.001
5 (0%)
3 (2%)
.08
17 (1%)
12 (2%)
.05
Prolonged ventilation
50 (5%)
21 (12%)
<.001
135 (7%)
124 (21%)
<.001
Any reoperation
59 (6%)
30 (16%)
<.001
123 (7%)
98 (15%)
<.001
Prolonged 14-d length of stay
36 (3%)
31 (16%)
<.001
123 (7%)
166 (26%)
<.001
Perioperative myocardial infarction
13 (1%)
7 (4%)
.01
16 (1%)
14 (2%)
Low output syndrome Atrial fibrillation Pacemaker
.008
7 (1%)
11 (6%)
<.001
28 (2%)
46 (7%)
<.001
166 (16%)
53 (28%)
<.001
606 (33%)
279 (44%)
<.001
35 (3%)
15 (8%)
71 (4%)
49 (8%)
<.001
.003
Hospital stay, d, median [IQR]
5 [4,7]
7 [5,10]
<.001
5 [4,7]
7 [5,10]
<.001
ICU stay, d, median [IQR]
1 [1,2]
2 [1,5]
<.001
1 [1,2]
2 [1,5]
<.001
Mediastinal drainage, mL, median [IQR]
638 [415, 1040]
845 [570, 1505]
<.001
638 [415, 1040]
845 [570, 1505]
<.001
421 (40%) 343 (32%) 248 (23%) 151 (14%)
130 (68%) 125 (65%) 69 (36%) 57 (30%)
<.001 <.001 <.001 <.001
1150 (62%) 343 (32%) 248 (23%) 151 (14%)
515 (81%) 125 (65%) 69 (36%) 57 (30%)
<.001 <.001 <.001 <.001
Any postoperative transfusion Packed red blood cells Platelets Fresh-frozen plasma
PM
Variables
No AKI n ¼ 1061
AKI, Acute kidney injury; RRT, renal-replacement therapy; ICU, intensive care unit; IQR, interquartile range.
procedures were performed by sternotomy (93%). Operations performed consisted in isolated CABG in 48% of patients, isolated valvular surgery in 31%, combined CABG and valvular surgery in 16%, and other procedures in 5%. These other procedures consisted mainly in ascending aorta and aortic arch operations, as well as septal myectomies. This cohort did not include any patient with complex adult congenital heart disease (eg, single-ventricle physiology, Fontan circulation, etc). In both age groups, patients who developed AKI were more likely to have had a combined CABG and valvular intervention compared with patients with no AKI. In both age groups, patients with AKI had longer CPB and aortic crossclamp times, a greater incidence of difficult CPB weaning, as well as lower hematocrit values both during and after CPB. In patients aged 65 and older—but not in those aged 60 and younger—larger hemodilution volumes and more frequent use of circulatory arrest were seen in patients who developed AKI compared with those who did not. Clinical Impact of AKI Early postoperative outcomes are detailed in Table 3. Among the 192 patients who developed postoperative AKI in the younger group, RRT was required in 15% (n ¼ 29). Among the 637 patients who developed postoperative AKI in the older group, RRT was required in 13%
(n ¼ 82). In both groups, patients who developed AKI had greater 30-day mortality compared with those with no AKI (8% vs 0% in the younger group; 11% vs 1% in the older group). In both groups, patients with AKI also had greater rates of prolonged ventilation, reoperation, perioperative myocardial infarction, low output syndrome, atrial fibrillation, permanent pacemaker requirement, postoperative blood product transfusion, as well as longer hospital and intensive care unit stays. Multivariable Linear Regression Analysis of Variables Associated With Maximum Postoperative Serum Creatinine Increase Variables associated with maximal postoperative serum creatinine increase in the final multivariable models are presented in Tables 4 and 5, as well as in Figure 1. In both younger and older patients, the following variables were associated independently with postoperative increase in creatinine: greater BMI, peripheral vascular disease, use of diuretics, lower preoperative hemoglobin, preoperative IABP, urgent or emergent status, long CPB time, and hemofiltration. In addition, certain variables associated with creatinine increase differed between the 2 groups. Although coronary artery disease was protective in young patients, it was harmful in older patients. In younger patients, a larger increase in
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
5
Perioperative Management
Saydy et al
TABLE 4. Multivariable analysis of factors associated with maximum postoperative serum creatinine increase in patients 60 years old Estimate
Standard error
P value
Demographic data Body mass index Female sex
0.04467 0.06777
0.00118 0.00741
<.001 <.001
Comorbidities Diabetes Coronary artery disease Peripheral vascular disease
0.04953 0.1048 0.1135
0.000129 0.0167 0.0215
<.001 <.001 <.001
Risk factors
Preoperative medication* Diuretics
0.05709
0.01848
.002
0.7288 0.02404
0.04294 0.00343
<.001 <.001
Preoperative cardiopulmonary status Previous cardiac surgery Intra-aortic balloon pump
0.1159 0.07113
0.00271 0.00568
<.001 <.001
Status Urgent or emergent status
0.03824
0.00245
<.001
Intraoperative strategy CPB time Hemofiltration
0.1262 0.05732
0.00919 0.0101
<.001 <.001
0.0283 0.00089 0.0300
.021 <.001 .037
Preoperative laboratory valuesy Last preoperative creatinine Last preoperative hemoglobin
PM
Type of surgery (ref: CABG) Isolated valvular surgery Combined CABG þ valvular surgery Other surgery (no CABG or valvular surgery)
0.1268 0.0989 0.1074
Variables in bold represent significant risk factors in both age groups. CPB, Cardiopulmonary bypass; CABG, coronary artery bypass graft. *In the 7 days before surgery. yLast preoperative value available.
serum creatinine was associated with diabetes (P <.001), whereas in older patients, it was associated with hypertension (P ¼ .003) and smoking (P ¼ .005). Female sex (P<.001) was associated with a lesser degree serum creatinine increase in young patients but not in the older group. Previous cardiac surgery was associated with larger increases in serum creatinine in the younger group (P<.001) but not in the older group. Conversely, lower preoperative left ventricular ejection fraction was associated with larger increases in serum creatinine in the older age group (P <.001) but not in the younger group. In younger patients, isolated valvular and other (ie, neither CABG nor valvular) operations were associated with a lower degree of postoperative serum creatinine increase (P ¼ .021 and P ¼ .037, respectively), whereas combined CABG and valvular surgery was associated with a greater degree of creatinine increase (P <.001). In contrast, in older patients, isolated valvular and other operations were associated with a greater degree of postoperative serum creatinine increase (P <.001 and P ¼ .013, respectively). Interaction analyses did not significantly alter either multivariable model. DISCUSSION Young adults represent a specific population in the realm of cardiac disease. In contrast to older patients, the impact 6
of atherosclerotic risk factors on their vasculature and overall physiology has been less sustained and therefore may be less clinically relevant at the time of surgery. In this study, we sought to specifically determine the risk factors of AKI in young adults undergoing cardiac surgery and compare these with risk factors observed in an older patient population with more established atherosclerotic disease. Interestingly, the main findings of this study suggest that although a number of risk factors are present across both age groups, others appear to be specific to either younger or older patients. The present study identified associations between postoperative creatinine increase and the following preoperative factors in all age groups: high BMI, coronary artery disease, peripheral vascular disease, use of diuretics within 7 days before surgery, high preoperative creatinine, low preoperative hemoglobin, and presence of an IABP. The link between obesity and postoperative outcomes after cardiac surgery is elusive. In the last decade, several investigators have observed a trend referred to as the ‘‘obesity paradox.’’ Despite being at an increased cardiovascular risk and having more comorbidities, patients with a greater BMI do not have less favorable outcomes after cardiac surgery.6-10 Whether this ‘‘paradox’’ applies to postoperative AKI is a matter of debate. Although some studies have found no association between BMI and AKI,3,11 others observed a
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Perioperative Management
TABLE 5. Multivariable analysis of factors associated with maximum postoperative serum creatinine increase in patients 65 years old Estimate
Standard error
P value
Demographic data Age Body mass index
0.00487 0.0484
0.000967 0.003282
<.001 <.001
Comorbidities Hypertension Coronary artery disease Peripheral vascular disease Smoking
0.0388 0.0380 0.0364 0.0248
0.01293 0.007696 0.005607 0.008833
.003 <.001 <.001 .005
Preoperative medication* Diuretics
0.0327
0.01291
.011
Preoperative laboratory valuesy Last preoperative creatinine Last preoperative hemoglobin
0.7977 0.0238
0.03979 0.003047
<.001 <.001
Preoperative cardiopulmonary status Intra-aortic balloon pump Left ventricular ejection fraction
0.106 0.00144
0.01094 0.000384
<.001 <.001
Status Urgent or emergent status
0.0109
0.000072
<.001
Intraoperative strategy CPB time Hemofiltration
0.1302 0.0523
0.02056 0.01307
<.001 <.001
Type of surgery (ref: CABG) Isolated valvular surgery Combined CABG þ valvular surgery Other (no CABG or valvular)
0.0287 0.0107 0.0441
0.001912 0.01734 0.008211
<.001 .58 .013
PM
Risk factors
Variables in bold represent significant risk factors in both age groups. CPB, Cardiopulmonary bypass; CABG, coronary artery bypass graft. *In the 7 days before surgery. yLast preoperative value available.
decreased rate of AKI and mortality in obese patients.8,10 In contrast, our study demonstrated a clear association between increased BMI and postoperative serum creatinine increase, further contributing to the equipoise surrounding this question. Anemia, preoperative use of diuretics, preexisting kidney disease, and preoperative IABP insertion have been identified previously as potentially modifiable risk factors of AKI after cardiac surgery.12-16 Of these, anemia is the most readily modifiable risk factor. Anemia contributes to the pathogenesis of AKI by reducing tissue oxygen delivery, since hemoglobin is the main factor that determines arterial oxygen concentration.17 One might have expected to see a lesser effect of preoperative anemia in young adults because of the shorter exposure to atherosclerotic risk factors in these patients, and a presumably greater ischemic reserve. Nonetheless, in the present study, multivariable analyses showed that low preoperative hemoglobin levels were associated with an increase in postoperative serum creatinine levels in both age groups. These findings suggest that just like their older counterpart, younger patients are vulnerable to hypoxemic renal injury secondary to preoperative anemia and that measures aimed at increasing preoperative hemoglobin levels should be sought in these young
adults undergoing elective cardiac procedures, similar to what is performed in older patients. The nephrotoxic effect of diuretics in the context of cardiac surgery can be explained by the association of volume depletion and renal hypoperfusion,18 resulting in renal hypoxia, one of the main mechanisms of AKI. The association between diuretic use and postoperative creatinine increase observed in this study highlights the fact that diuretics should only be administered when specifically indicated, both preoperatively and in the postoperative setting.15,18,19 High preoperative creatinine—in the setting of chronic kidney disease or surgery after cardiogenic shock with end-organ damage—is also known to be associated with postoperative AKI.3,20 This study also identified several risk factors for postoperative serum creatinine increase that differed between younger and older patients. Diabetes mellitus—previously identified as a risk factor for postoperative AKI13—was associated independently with creatinine increase in patients 60 years of age and younger but not in those 65 years of age and older. Conversely, atherosclerotic risk factors such as smoking and hypertension were identified as independent risk factors in the older patient group but not in patients 60 years of age and younger. This is likely due to the
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
7
Perioperative Management
Saydy et al
PM FIGURE 1. Factors associated with postoperative increase in serum creatinine on multivariable linear regression analysis. Variables associated with postoperative increase in serum creatinine in patients 60 years (green) only are presented in the top panel. Variables associated with postoperative increase in serum creatinine in patients 65 years (red) only are presented in the bottom panel. Variables associated with postoperative increase in serum creatinine in patients in both age groups are presented in the middle panel. Estimates are presented with 95% confidence interval bars. CABG, Coronary artery bypass graft; BMI, body mass index; PVD, peripheral vascular disease; Hb, hemoglobin; IABP, intra-aortic balloon pump; CPB, cardiopulmonary bypass; LVEF, left ventricular ejection fraction.
8
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Perioperative Management
(I2 ¼ 96%).23 Our finding of a greater degree of creatinine increase in young and middle-aged adults undergoing CABG compared with valvular surgery may be due to the fact that patients who undergo CABG before the age of 60 typically have very aggressive atherosclerotic disease, which may predispose them to renal injury. The main limitation of this study is its retrospective, observational nature. Even though all patients aged 60 years and less were included in this study, most patients were closer to 60 years of age (mean age: 52 years). Furthermore, outcomes beyond the immediate perioperative period were unfortunately not readily available for analysis, such that no comment can be made on the differential long-term impact of postoperative AKI between younger and older patients. Finally, comparisons between patients in the younger and older groups should be interpreted cautiously as these 2 groups had different baseline and operative characteristics, which were not accounted for in the analysis. Despite these limitations, this study is, to our knowledge, the first to specifically examine the risk factors for postoperative creatinine increase in young and middle-aged adults undergoing cardiac surgery. CONCLUSIONS This study identified both common and distinct factors associated with postoperative increase in serum creatinine between patients aged 60 years and younger and those 65 years and older undergoing cardiac surgery. Of note, all modifiable (ie, preoperative hemoglobin, preoperative diuretic use) or potentially modifiable (ie, preoperative IABP, CPB duration) risk factors were present in both groups, highlighting the importance of preoperative and intraoperative strategies aimed at mitigating these various factors in all patients, regardless of age. Conflict of Interest Statement Authors have nothing to disclose with regard to commercial support. References 1. Mariscalco G, Lorusso R, Dominici C, Renzulli A, Sala A. Acute kidney injury: a relevant complication after cardiac surgery. Ann Thorac Surg. 2011;92:1539-47. 2. Rosner MH, Okusa MD. Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol. 2006;1:19-32. 3. Loef BG, Epema AH, Smilde TD, Henning RH, Ebels T, Navis G, et al. Immediate post-operative renal function deterioration in cardiac surgical patients predicts in-hospital mortality and long-term survival. J Am Soc Nephrol. 2005;16: 195-200. 4. Chertow GM, Levy EM, Hammermeister KE, Grover F, Daley J. Independent association between acute renal failure and mortality following cardiac surgery. Am J Med. 1998;104:343-8. 5. Tibshirani R. Regression shrinkage and selection via the lasso. J R Stat Soc Ser B. 1996;58:267-88. 6. Stamou SC, Nussbaum M, Stiegel RM, Reames RK, Skipper ER, Robicsek F, et al. Effect of body mass index on outcomes after cardiac surgery: is there an obesity paradox? Ann Thorac Surg. 2001;91:42-7.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
9
PM
accumulation of repeated insults to the renal and coronary microvasculature in these older patients, which ultimately leads to decreased autoregulation of renal perfusion.21,22 This hypothesis is further supported by the observation that age—a known risk factor for impaired renal autoregulation—was associated significantly with postoperative creatinine increase in the older patient group but not in patients 60 years and younger. The present study also identified several operative variables associated with postoperative increase in serum creatinine for both younger and older patients: long duration of CPB, hemofiltration, and urgent operative status. This is in line with findings from a recently published metaanalysis that found CPB duration and emergent operative status to be associated with AKI.23 CPB duration is a well-established risk factor for AKI in the general cardiac surgery patient population.24-28 Other variables, such as lower CPB temperatures, low CPB flow and a large difference between preoperative mean arterial pressure and mean arterial pressure during CPB, also have been shown to be associated with AKI.29-33 We hypothesized that CPB duration may not have such a direct and marked impact on postoperative renal function in younger patients. However, findings from the present study support the notion that young patients are also vulnerable to the cytotoxic effects of prolonged CPB. A large number of young adults undergoing cardiac surgical procedures often require long operative times because of the complex nature of their lesions, requiring concomitant procedures or the need for hypothermic circulatory arrest in aortic pathologies. In these patients, efforts should be made during CPB to ensure adequate perfusion flow and pressure, optimize hematocrit levels, and limit the use of vasopressor agents during CPB. Another interesting finding of this study is the differential impact of the type of surgery on postoperative creatinine increase between younger and older patients. Specifically, in patients 65 years or older, isolated CABG was associated with a lower postoperative increase in creatinine compared with isolated valvular or ‘‘other’’ operations. In contrast, in patients 60 years or younger, isolated CABG was associated with a greater postoperative increase in creatinine compared with isolated valvular or ‘‘other’’ operations but with a lower increase in creatinine compared with combined CABG and valvular surgery. The association between combined operations (ie, CABG and valvular) and greater rates of AKI is not surprising and has been previously demonstrated.20 However, in the case of isolated procedures, the differential association of CABG and valvular operations with AKI in either age group is more controversial. This uncertainty was highlighted in a recently published metaanalysis that showed no significant association between AKI and type of surgery (CABG vs valvular) but identified very high heterogeneity among published studies
Perioperative Management
PM
7. Le-Bert G, Santana O, Pineda AM, Zamora C, Lamas GA, Lamelas J. The obesity paradox in elderly obese patients undergoing coronary artery bypass surgery. Interact Cardiovasc Thorac Surg. 2011;13:124-7. 8. Engel AM, McDonough S, Smith JM. Does an obese body mass index affect hospital outcomes after coronary artery bypass graft surgery? Ann Thorac Surg. 2009;88:1793-800. 9. van Straten AH, Bramer S, Soliman Hamad MA, van Zundert AA, Martens EJ, Schonberger JP, et al. Effect of body mass index on early and late mortality after coronary artery bypass grafting. Ann Thorac Surg. 2010;89:30-7. 10. Thourani VH, Keeling WB, Kilgo PD, Puskas JD, Lattouf OM, Chen EP, et al. The impact of body mass index on morbidity and short- and long-term mortality in cardiac valvular surgery. J Thorac Cardiovasc Surg. 2011;142:1052-61. 11. Engleberger L, Suri RM, Connolly HM, Li Z, Abel MD, Greason KL, et al. Increased risk of acute kidney injury in patients undergoing tricuspid valve surgery. Eur J Cardiothorac Surg. 2013;43:993-9. 12. Karkouti K, Wijeysundera DN, Beattie WS. Risk associated with pre-operative anemia in cardiac surgery: a multicenter cohort study. Circulation. 2008;117:478-84. 13. Karkouti K, Wijeysundera DN, Yau TM, Callum JL, Cheng DC, Crowther M, et al. Acute kidney injury after cardiac surgery: focus on modifiable risk factors. Circulation. 2009;119:495-502. 14. Chertow GM, Lazarus JM, Christiansen CL, Cook F, Hammermeister KE, Grover F, et al. Pre-operative renal stratification. Circulation. 1997;95:878-84. 15. Lassnigg A, Donner E, Grubhofer G, Presterl E, Druml W, Huesmayr M. Lack of renoprotective effects of dopamine and furosemide during cardiac surgery. J Am Soc Nephrol. 2000;11:97-104. 16. Lombardi R, Ferreiro A, Servetto C. Renal function after cardiac surgery: adverse effect of furosemide. Renal Fail. 2003;25:775-86. 17. Nangaku M. Chronic hypoxia and tubulointerstitial injury: a final common pathway to end-stage renal failure. J Am Soc Nephrol. 2006;17:17-25. 18. Lombardi R, Ferreiro A. Risk factors profile for acute kidney injury after cardiac surgery is different according to the level of baseline renal function. Ren Fail. 2008;30:155-60. 19. Thiele RH, Isbell JM, Rosner MH. Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol. 2015;10:500-14. 20. Thakar CV, Arrigain S, Worley S, Yared JP, Paganini EP. A clinical score to predict acute renal failure after cardiac surgery. J Am Soc Nephrol. 2005;16: 162-8. 21. Abuelo JG. Normotensive acute renal failure. N Engl J Med. 2007;357:797-805. 22. Bidani AK, Polichnowski AJ, Loutzenhiser R, Griffith KA. Renal microvascular dysfunction, hypertension and CKD progression. Curr Opin Nephrol Hypertens. 2013;22:1-9.
10
Saydy et al
23. Yi Q, Li K, Jian Z, Xiao YB, Chen L, Zhang Y, et al. Risk factors for acute kidney injury after cardiovascular surgery: evidence from 2157 cases and 49 777 controls—a meta-analysis. Cardiorenal Med. 2016;6:237-50. 24. Mehta RH, Castelvecchio S, Ballotta A, Frigiola A, Bossone E, Ranucci M. Association of gender and lowest hematocrit on cardiopulmonary bypass with acute kidney injury and operative mortality in patients undergoing cardiac surgery. Ann Thorac Surg. 2013;96:133-40. 25. Karkouti K, Beattie WS, Wijeysundera DN, Rao V, Chan C, Dattilo KM, et al. Hemodilution during cardiopulmonary bypass is an independent risk factor for acute renal failure in adult cardiac surgery. J Thorac Cardiovasc Surg. 2005;129:391-400. 26. Bove T, Calabro MG, Landoni G, Aletti G, Marino G, Crescenzi G, et al. The incidence and risk of acute renal failure after cardiac surgery. J Cardiothorac Vasc Anesth. 2004;18:442-5. 27. Suen WS, Mok CK, Chiu SW, Cheung KL, Lee WT, Cheung D, et al. Risk factors for development of acute renal failure (ARF) requiring dialysis in patients undergoing cardiac surgery. Angiology. 1998;49:789-800. 28. Sirvinskas E, Andrejaitiene J, Raliene L, Nasvytis L, Karbonskiene A, Pilvinis V, et al. Cardiopulmonary bypass management and acute renal failure: risk factors and prognosis. Perfusion. 2008;23:323-7. 29. Kourliouros A, Valencia O, Philips SD, Collinson PO, van Besouw JP, Jahangiri M. Low cardiopulmonary bypass perfusion temperatures are associated with acute kidney injury following coronary artery bypass surgery. Eur J Cardiothorac Surg. 2010;37:704-9. 30. Kanji HD, Schulze CJ, Hervas-Malo M, Wang P, Ross DB, Zibdawi M, et al. Difference between pre-operative and cardiopulmonary bypass mean arterial pressure is independently associated with early cardiac surgery-associated acute kidney injury. J Cardiothorac Surg. 2010;5:71. 31. Ranucci M, Pavesi M, Mazza E, Bertucci C, Frigiola A, Menicanti L, et al. Risk factors for renal dysfunction after coronary surgery: the role of cardiopulmonary bypass technique. Perfusion. 1994;9:319-26. 32. Fang WC, Helm RE, Krieger KH, Rosengart TK, DuBois WJ, Sason C, et al. Impact of minimum hematocrit during cardiopulmonary bypass on mortality in patients undergoing coronary artery surgery. Circulation. 1997;96(suppl II):194-9. 33. Swaminathan M, Phillips-Bute BG, Conlon PJ, Smith PK, Newman MF, Stafford-Smith M. The association of lowest hematocrit during cardiopulmonary bypass with acute renal injury after coronary artery bypass surgery. Ann Thorac Surg. 2003;76:84-92.
Key Words: acute kidney injury, cardiac surgery, young patients
The Journal of Thoracic and Cardiovascular Surgery c - 2017
Saydy et al
Differences and similarities in risk factors for postoperative acute kidney injury between younger and older adults undergoing cardiac surgery Nadim Saydy, Amine Mazine, MD, MSc, Louis-Mathieu Stevens, MD, PhD, Hughes Jeamart, MD, MSc, Philippe Demers, MD, MSc, Pierre Page, MD, Yoan Lamarche, MD, MSc, and Ismail ElHamamsy, MD, PhD, Montreal, Quebec, Canada This study identified both common and distinct risk factors associated with postoperative increase in serum creatinine between patients 60 years and those 65 years undergoing cardiac surgery.
PM
000
Perioperative Management
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -