Mini - Cec

individual non-member subscription to the journal.

Miniaturized Cardiopulmonary Bypass and Acute

Kidney Injury in Coronary Artery Bypass Graft

Surgery

Umberto Benedetto,

MD,

Remo Luciani,

MD,

Massimo Goracci,

MD,

Fabio Capuano,

MD,

Simone Refice,

MD,

Emiliano Angeloni,

MD,

Antonino Roscitano,

MD,

and Riccardo Sinatra,

MD

Departments of Cardiac Surgery and Nephrology, II School of Medicine, University of Rome La Sapienza, Policlinico S. Andrea, Rome, Italy

Background. Acute kidney injury (AKI) is one of the most important complications after on-pump coronary artery bypass graft surgery (CABG). Miniaturized cardio-pulmonary bypass (mini-CPB) systems have been devel-oped to allow the ease of on-pump surgery but tempering the disadvantages. Whether mini-CPB reduces the inci-dence of AKI remains to be determined.

Methods. Using a propensity score matched analysis, we investigated the occurrence of AKI among patients under-going CABG on mini-CPB (nⴝ 104) versus conventional CPB (nⴝ 601). Acute kidney injury was defined according to the recent Acute Kidney Injury Network classification.

Results. Overall, acute kidney injury developed in 274 of 705 patients (38.8%). A total of 27 of 705 patients (3.8%) required renal replacement therapy. The median postop-erative length of intensive care unit stay in survivors with AKI was 5.4 (3.9 to 6.8) days compared with 2.0 (1.0 to 3.0) days for patients without AKI (p ⴝ 0.0002). The overall incidence of AKI for patients undergoing

mini-CPB was 30 of 104 (28.8%) compared with 244 of 601 (40.5%) for patients undergoing conventional CPB (pⴝ 0.03). In the propensity score matched-pair statistical analysis, mini-CPB was independently associated with a decreased incidence of AKI (adjusted odds ratio [OR] 0.61; 95% confidence interval [CI]: 0.38 to 0.97). Other variables independently associated with AKI were pre-operative glomerular filtration rate (OR 0.988 for 1 mL · minⴚ1· 1.73 mⴚ2 increase; 95% CI: 0.98 to 0.99), postoperative red blood cell transfusion (OR 1.58; 95% CI: 1.12 to 2.23); CPB time (OR 1.005 for 1-minute in-crease; 95% CI: 1.0 to 1.009), and postoperative low output syndrome (OR 1.72; 95% CI: 1.23 to 2.41).

Conclusions. The present study showed that mini-CPB is associated with a lower incidence of AKI when com-pared with conventional CPB among patients undergo-ing CABG.

(Ann Thorac Surg 2009;88:529 –36) © 2009 by The Society of Thoracic Surgeons

A

effect of extracorporeal circulation. However, the beat-ing-heart technique may be challenging, cumbersome, and may have inferior results with respect to long-term graft patency [3]. Therefore, CABG with CPB and car-dioplegic arrest is still the standard method of choice in many institutes. Miniaturized cardiopulmonary bypass (mini-CPB) systems have been developed to allow the ease of on-pump surgery while tempering the disadvan-tages[4]. Major differences to conventional CPB are use of heparinized tubing and oxygenators, minimizing prime volume, use of a centrifugal pump, and elimina-tion of cardiotomy sucelimina-tion and a venous reservoir. Mini-CPB reduces the hemodilution and the foreign surface area blood contact, thus attenuating the systemic inflam-matory response syndrome.

Whether this better biological profile results in re-duced AKI remains to be determined, however, with some discordant results reported[5, 6]. Such discordance

Accepted for publication March 26, 2009.

Address correspondence to Dr Benedetto, Cardiac Surgery Department, II School of Medicine, University of Rome “La Sapienza,” Via di Grotta-rossa 1039, Rome, 00181, Italy; e-mail:u2benedetto@libero.it.

© 2009 by The Society of Thoracic Surgeons 0003-4975/09/$36.00

Published by Elsevier Inc doi:10.1016/j.athoracsur.2009.03.072

ADULT

may be partially explained by the lack of a consensus definition of AKI. There is increasing agreement that a universal kidney injury definition is needed to facilitate progress in the field of acute renal failure. Recently, the Acute Kidney Injury Network has reached a consensus on the definition of AKI [7], according to new data showing that small changes in serum creatinine might be associated with adverse outcomes. Therefore, we aimed to evaluate whether mini-CPB decreases the occurrence of AKI after CABG when compared with conventional CPB, adopting the recently proposed AKI definition.

Patients and Methods

This study was reviewed and approved by the Institu-tional Review Board of the University of Rome, and a waiver of consent was granted. The authors have no conflict of interest to disclose.

Patient Population

In the present study, we analyzed consecutive patients undergoing isolated CABG at our institution between May 2004 and January 2009 who met the following criteria: (1) procedures on CPB; (2) no preoperative renal failure requiring dialysis; and (3) no prior cardiac surgery.

The study population consisted of 705 patients (median age, 68 years; range, 59 to 73); 123 of the 705 were female. The mini-CPB system (Resting Heart System [RHS]; Medtronic, Minneapolis, MN) was used in 104 patients according to surgeon preference. In other cases (n⫽ 601), a conventional extracorporeal circulation system was instituted.

Miniaturized CPB

The RHS is an integrated, low prime, semi– closed-loop CPB system, offering minimal air and blood interface and elimination of antifoam agents. The priming volume of this circuit is 600 mL, and the membrane surface area for gas exchange is 2.5 m2

. Primary blood contact surfaces of this RHS are coated with heparin (BioActive Surface; Carmeda, Stockholm, Sweden) throughout to provide thromboresistance and biocompatibility by mimicking critical characteristics of vascular endothelium. The RHS can offer blood flow from 1.0 to 6.0 L/min. A vent circuit is available in the RHS, and it was used in this study. The

blood from this vent is reinjected into the pump inflow. The absence of cardiotomy reservoirs limits the artificial surface-blood contact that occurs secondary to aspiration of blood. Accordingly, an erythrocyte-scavenging device is necessary when using the RHS. In addition, this system has the technology to detect and remove small air bub-bles in the circuit. If air is entrained from the right atrium, visual and audible alarms alert the surgical team to the condition so that it can be quickly remedied. Two pairs of ultrasonic fluid sensors in the venous air removal device detect air at the inlet of the device. When air enters the device through the venous return line, air bubbles are detected, and the device exerts evident visual and audible indications while removing the venous air. The air is automatically removed from the venous air removal device until its sensors detect no remaining air/blood mixture in the upper area of the device, and then it returns to normal setting.

Conventional CPB

The priming volume amounted to 1,200 mL. The line was coated with heparin (Medtronic). An oxygenator (Affinity NT Oxygenator; Medtronic) and a standard roller pump were included in the set. An arterial filter was used. The intrapericardial suction device was used as usual.

Definition of AKI and Data Collection

Acute kidney injury was defined as an increase of serum creatinine of 0.3 mg/dL or more (26.4␮mol/L) postoper-atively according to the consensus definition proposed by the Acute Kidney Injury Network[7]and stratified into three classes: stage 1, increase of serum creatinine 150% to 200% from baseline; stage 2, increase of serum creati-nine of 200% to 300% from baseline; and stage 3, increase of serum creatinine to more than 300% from baseline or serum creatinine⫽ 4.0 mg/dL (354 ␮mol/L) after a rise of at least 44␮mol/L or treatment with renal replacement therapy. The four-variable Modified Diet and Renal Dis-ease equation was used to estimate baseline glomerular filtration rate (GFR)[8]. We did not use urine output in defining AKI.

Data were prospectively collected and recorded in an electronic database by physicians. Baseline serum creat-inine value was defined as the value recorded just before surgery. Serum creatinine values were taken after patient admission to the intensive care unit and repeated at least once every 24 hours. Based on these values, the difference between the highest serum creatinine value and the baseline value was calculated for each patient. Operative mortality was defined as all deaths occurring within 30 days from surgery.

The European System for Cardiac Operative Risk Eval-uation (EuroSCORE)[9]was derived to assess differences in patient risk profile between the two study groups.

Statistical Analysis

Continuous variables are shown as median with interquar-tile range. All categorical data were displayed as percent-ages. Comparisons were made with Mann-Whitney rank-sum tests and␹2

tests as appropriate. A post hoc␹2 test Abbreviations and Acronyms

AKI ⫽ acute kidney injury

CABG ⫽ coronary artery bypass graft surgery

CI ⫽ confidence interval

CPB ⫽ cardiopulmonary bypass

Mini-CPB⫽ miniaturized cardiopulmonary bypass

NYHA ⫽ New York Heart Association

OR ⫽ odds ratio

RHS ⫽ resting heart system

ADULT

analysis was performed to achieve the study power to detect significant differences for the outcome of interest.

To investigate the effect of mini-CPB on the incidence of AKI, treatment selection bias was controlled for by constructing the propensity score to be operated on mini-CPB. Potential confounding factors considered in the analysis were selected on the basis of a literature review and clinical plausibility. These variables included (1) demographic characteristics such as age, sex, and body mass index (BMI); (2) clinical risk factors including preoperative GFR, diabetes mellitus, chronic obstructive pulmonary disease requiring treatment, hypertension, peripheral vascular disease, preoperative New York Heart Association (NYHA) class III/IV, left ventricular ejection fraction, prior percutaneous coronary interven-tion, myocardial infarction within 30 days, emergency surgery, preoperative medications including␤-blockers,

angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, aspirin, and clopidogrel within 5 days from surgery; (3) procedural characteristics such as pri-mary surgeon, number of grafts per patient, and CPB time; and (4) postoperative variables including red blood cell transfusion and the incidence of postoperative low cardiac output syndrome, defined as the need for postop-erative intra-aortic balloon pump or inotropic support for longer than 30 minutes in the intensive care unit to main-tain the systolic blood pressure at more than 90 mm Hg.

The propensity score was calculated for all 705 patients by means of a multivariate logistic regression analysis, including all significant variables listed in Table 1. To minimize the selection bias of the groups, three nonsig-nificant variables—age, primary surgeon, and the num-ber of grafts per patient—were also used to construct a propensity score.

Table 1. Patient Demographics, Clinical Data, and Procedural Characteristics

Variables

Mini-CPB

Yes (n⫽ 104) No (n⫽ 601) p Value

Demographics

Age median (years), median [IQR] 68 [60.0–74.0] 67 [59.0–73.0] 0.3

Female, % 22.1 16.6 0.2

Body mass index (kg/m), median [IQR] 26.6 [24.4–29.4] 26.6 [24.7–29.0] 0.9

Clinical risk factors

Preoperative GFR (mL · min⫺1· 1.73 m⫺2), median [IQR] 64 [49–78] 66 [54–80] 0.16

Diabetes mellitus, % 35.6 39.8 0.4

Chronic obstructive pulmonary disease, % 7.6 15.6 0.04

Hypertension, % 95.2 92.5 0.4

Peripheral vascular disease, % 10.6 14.6 0.3

Preoperative NHYA functional class III–IV, % 8.6 17.2 0.04

Left ventricular ejection fraction, median [IQR] 55 [47.0–60.0] 55 [45.0–60.0] 0.6

Prior PCI, % 20.2 16 0.3

Myocardial infarction within 30 days, % 45.2 38.1 0.2

Emergency surgery, % 7.6 16.6 0.02

EuroSCORE, median [IQR] 5 [3.0–8.0] 6 [3.0–9.0] 0.3

Preoperative medications

Aspirin within 3 days, % 52.9 43.9 0.1

Clopidogrel within 3 days, % 32.7 19.1 0.002

ACEI/ARB, % 63.5 62.4 0.9 ␤-blockers, % 64.4 56.4 0.1 Procedural characteristics Primary surgeon 0.9 1, % (n) 27.9 (29) 25.2 (151) 2, % (n) 21.1 (22) 21.2 (127) 3, % (n) 18.3 (19) 20.6 (124) 4, % (n) 20.2 (21) 19.5 (117) 5, % (n) 12.5 (13) 13.5 (81)

Number of grafts/patient, median [IQR] 3 [2.0–3.0] 3 [2.0–3.0] 0.1

CPB time (minutes), median [IQR] 104 [84.5–124.5] 103 [85.0–122.0] 0.7

Postoperative variables

Low cardiac output syndrome, % 8.2 12.5 0.2

Red blood cell transfusion rate, % 48.1 65.1 0.001

ACEI/ARB⫽ angiotensin-converting enzyme inhibitors/angiotensin receptor blockers; CPB⫽ cardiopulmonary bypass; GFR⫽ glomerular filtration rate; IQR⫽ interquartile range; NYHA⫽ New York Heart Association; PCI⫽ percutaneous coronary intervention.

ADULT

Each mini-CPB patient was then closely matched with a conventional CPB patient having the same or nearest pro-pensity score. In the matched cohort, a multivariate logistic stepwise regression was performed to identify factors inde-pendently associated with AKI, including all variables sig-nificantly associated to AKI at univariate analysis.

Models fit analysis was evaluated with the Hosmer-Lemeshow goodness-of-fit statistic. The C-statistic was used to assess the predictive power of propensity score. Odds ratios (OR) and their associated 95% confidence intervals (CI) were estimated. In all tests, values of p less than 0.05 were considered significant. The Statistical Package for the Social Sciences, version 11 (SPSS, Chi-cago, IL), and G*Power, version 3.0.5 for Windows (Erd-felder, Faul, & Buchner, 1996, Mannheim, Germany) were used for statistical analysis.

Results

Demographic, Clinical Data, and Perioperative

Characteristics

Demographic, clinical data, and perioperative character-istics are presented in Table 1 according to the use of mini-CPB. Age, sex distribution, and preoperative serum creatinine levels were similar between the two groups. Patients operated on with mini-CPB were more likely to receive clopidogrel within 5 days from surgery. Patients operated on with conventional CPB were more likely to have chronic obstructive pulmonary disease, preopera-tive NYHA class III/IV, and emergency surgery. Patient risk profile according to the EuroSCORE was similar between the two groups.

The distribution of primary surgeon did not differ Table 2. Propensity Score Matched Patient Demographic, Clinical Data, and Procedural Characteristics

Variables

Mini CPB

Yes (n⫽ 104) No (n⫽ 104) p Value

Demographics

Age median (years), median [IQR] 68 [60.0–74.0] 69 [60.0–75.0] 0.5

Female, % 22.1 18.3 0.6

Body mass index (kg/m), median [IQR] 26.6 [24.4–29.4] 26.4 [24.7–28.4] 0.5

Clinical risk factors

Preoperative GFR (mL · min⫺1· 1.73 m⫺2), median [IQR] 64 [49–78] 66 [53–76] 0.43

Diabetes mellitus, % 35.6 29.8 0.4

Chronic obstructive pulmonary disease, % 7.6 4.8 0.5

Hypertension, % 95.2 92.3 0.5

Peripheral vascular disease, % 10.6 19.2 0.1

Preoperative NHYA functional class III–IV, % 8.6 11.5 0.6

Left ventricular ejection fraction , median [IQR] 55 [47.0–60.0] 55 [45.2–60.0] 0.7

Prior PCI, % 20.2 22.1 0.8

Prior myocardial infarction within 30 days, % 45.2 42.3 0.7

Emergency surgery, % 7.6 6.7 1

EuroSCORE, median [IQR] 5.0 [3.0–8.0] 6.0 [4.0–8.0] 0.3

Preoperative medications

Aspirin within 3 days, % 52.9 51 0.8

Clopidogrel within 3 days, % 32.7 35.6 0.7

ACEI/ARB, % 63.5 63.5 0.8 ␤-blockers, % 64.4 55.8 0.2 Procedural characteristics Primary surgeon 0.9 1, % 27.9 (29) 25.1 (26) 2, % 21.1 (22) 20.2 (21) 3, % 18.3 (19) 22.1 (23) 4, % 20.2 (21) 21.1 (22) 5, % 12.5 (13) 11.5 (12)

Number grafts/patient, median [IQR] 3 [2.0–3.0] 3 [2.0–3.0] 0.2

CPB time (minutes), median [IQR] 104 [84.5–124.5] 99 [87.0–124.0] 0.6

Postoperative variables

Low cardiac output syndrome, % 8.2 9.3 0.9

Red blood cell transfusion rate, % 48.1 56.7 0.2

ACEI/ARB⫽ angiotensin-converting enzyme inhibitors/angiotensin receptor blockers; CPB⫽ cardiopulmonary bypass; GFR⫽ glomerular filtration rate; IQR⫽ interquartile range; NYHA⫽ New York Heart Association; PCI⫽ percutaneous coronary intervention.

ADULT

between the two groups nor did the number of grafts per patient and cardiopulmonary bypass time. Conventional CBP was associated with an increased occurrence of red blood cell transfusion postoperatively.

Renal Outcomes

Baseline serum creatinine concentration was 1.1 mg/dL (0.9 to 1.3) versus 1.1 mg/dL (0.9 to 1.2) in the mini-CPB group versus conventional CPB group, respectively (p⫽ 0.33). Postoperative peak serum creatinine concentration was achieved after 2 days (1 to 3), and it was 1.27 mg/dL (1.1 to 1.5) versus 1.37 mg/dL (1.1 to 1.7) in the mini-CPB and conventional CPB groups, respectively (p ⫽ 0.04). Overall, AKI developed in 274 of 705 patients (38.8%). The majority of patients had stage 1 AKI (n⫽ 215), with few patients in stages 2 (n⫽ 31) and 3 (n ⫽ 28). A total of 27 of 705 patients (3.8%) required renal replacement therapy. The median postoperative length of intensive care unit stay in survivors having AKI was 5.4 days (3.9 to 6.8) compared with 2.0 days (1.0 to 3.0) for patients without AKI (p ⫽ 0.0002). The median postoperative length of intensive care unit stay for survivors requiring renal replacement therapy was 16.5 days (9 to 48) com-pared with 2.0 days (1.0 to 3.0) for patients who did not

(p⬍ 0.0001).

The overall incidence of AKI among patients undergo-ing mini-CPB was 30 of 104 (28.8%) compared with 244 of 601 (40.5%) among patients undergoing conventional CPB (p⫽ 0.03); stage 1 AKI occurred in 25 of 104 patients (24%) versus 190 of 601 patients (31.6%), respectively; stage 2 AKI occurred in 3 of 104 (2.8%) versus 28 of 601 (4.65%), respectively; and stage 3 AKI occurred in 2 of 104 (1.9%) versus 26 of 601 (4.3%), respectively (Fig 1). The incidence of renal replacement therapy was 1 of 104 (0.9%) and 26 of 601 (4.3%), respectively (p⫽ 0.16).

Operative Mortality

Overall, the operative mortality rate was 2.5%. The oper-ative mortality rate was 17 of 274 (6.2%) for patients with AKI compared with 1 of 431 (0.2%) for patients without AKI (p ⬍ 0.0001); and 10 of 27 (37%) for patients who required postoperative dialysis compared with 8 of 678 (1.1%) for patients who did not require postoperative dialysis (p⬍ 0.0001). Operative mortality rate was 1 of 104 (0.9%) and 17 of 601 (2.8%) for patients undergoing mini-CPB versus conventional CBP, respectively (p ⫽ 0.49).

Propensity Score Matched Analysis for AKI

In the mini-CPB group, the median propensity score was 0.18 (interquartile range, 0.12 to 0.25); in the conventional CPB group, the median score was 0.12 (interquartile range, 0.17 to 0.24). The C-statistic for the propensity score model was 0.68 (95% CI: 0.65 to 0.72), indicating a good discrimination. Each of mini-CPB patients was successfully matched with a patient operated on with conventional CPB. Clinical patient characteristics for the mini-CPB group and the conventional CPB group after matching are listed in Table 2. All characteristics that

differed significantly between two groups before match-ing became similar.

In the matched-pair statistical analysis, mini-CPB was found to be significantly associated with a decreased occurrence of AKI (28.8% versus 42.3%; unadjusted OR 0.55; 95% CI: 0.31 to 0.98; p ⫽ 0.04). Post hoc power analysis for ␹2 _{test showed that study was adequately} powered to detect significant differences for the inci-dence of AKI (1-␤: 0.89; critical ␹2_{: 11; effect size w: 0.27)} for an␣ value of 0.05.

After adjusting for other significant covariates, mini-CPB was independently associated with a decreased incidence of AKI (adjusted OR 0.61; 95% CI: 0.38 to 0.97). Other variables independently associated with AKI were preoperative glomerular filtration rate (OR 0.988 for 1 mL · min⫺1· 1.73⫺2increase; 95% CI: 0.98 to 0.99), post-operative red blood cell transfusion (OR 1.58; 95% CI: 1.12 to 2.23); CPB time (OR 1.005 for 1-minute increase; 95% CI: 1.0 to 1.009), and postoperative low cardiac output syndrome (OR 1.72; 95% CI: 1.23 to 2.41) (Table 3). The multivariate model significantly predicted the occurrence of AKI (model␹2

: 33; p⬍ 0.0001). The model was well calibrated among deciles of observed and expected risk (Hosmer-Lemeshow␹2

: 12; p⫽ 0.17).

Comment

The present study showed that in patients undergoing isolated CABG, mini-CPB is significantly associated with a decreased incidence of postoperative acute kidney injury when compared with conventional CPB.

Acute kidney injury is one of the most important complications after CABG [1, 2]. Consequences of AKI include an increase in mortality risk, which can exceed 60% among patients requiring dialysis. Even when serum creatinine values remain within the normal range, mod-est increases from baseline values are associated with increased odds of death as well as longer hospital stay[7, 10]. However, there have neither been any associated improvements in incidence nor in mortality, despite many recent advances in our understanding of the etiol-ogy and pathophysioletiol-ogy of AKI[2].

Table 3. Multivariate Logistic Regression Analysisa,b

for Risk Factors Associated With Acute Kidney Injury in the Propensity Score Matched Cohort

Variables

Adjusted

Odds Ratio 95% CI Preoperative glomerular filtration

ratec 0.988 0.98–0.99

Miniaturized cardiopulmonary bypass 0.61 0.38–0.97 Cardiopulmonary bypass timed _{1.005 1.0–1.009} Postoperative red blood cell

transfusion

1.58 1.12–2.23 Postoperative low cardiac output

syndrome

1.72 1.23–2.41

a_Model␹2_{: 33; p⬍ 0.0001.} b_{Hosmer-Lemeshow␹}2_{: 12; p⫽ 0.17.} c_For

1 mL · min⫺1· 1.73 m⫺2. d_{Per 1-minute increase.}

CI⫽ confidence interval.

ADULT

A major obstacle to finding successful management for AKI has been the conduct of trials adequately powered to show a benefit. Acute kidney injury requiring renal replacement therapy is relatively rare in postoperative cardiac surgical patients [11]. To power a study for an absolute reduction in renal replacement therapy require-ment, the necessary sample size would be very large (n⫽ 1,092, assuming an absolute reduction of 2%, a value⫽ 0.01 and 1-␤ ⫽ 0.90). Such a study would require a large multicenter design and would be very costly. Alterna-tively, clinical endpoints could include surrogate out-comes showing consistent association with relevant clin-ical outcomes[7, 10]. In this regard, the inconsistencies in the current literature with respect to the definition of AKI need to be resolved so that future research can be meaningful, reproducible, and comparable.

The most promising classification today is based on the Acute Kidney Injury Network definition[7]. This defini-tion had been strongly associated with adverse outcomes such as overall mortality and intensive care unit stay [12–14], and it should be routinely adopted in evaluating potential strategies for AKI. Mini-CPB has been shown to provide a better biological profile than conventional CPB, reducing the hemodilution and attenuating the systemic inflammatory response syndrome[4]. However, whether it may result in reduced AKI remains to be determined. The two largest randomized controlled trials reported discordant conclusions with regard to renal advantages of the mini-CPB system when compared with conven-tional CPB[5, 6]. It should be pointed out that neither study used AKI as the renal endpoint, and the number of patients with postoperative renal failure requiring renal replacement therapy was too low to reach a statistically significant difference. A potential role of mini-CPB in reducing AKI after CABG has been supported by studies that showed a decreased release of specific renal injury biomarkers such as N-acetyl-glucosaminidase and

uri-nary interleukin-6[4]. However, to draw definitive evi-dences from trials using changes in biomarker levels as surrogate outcome, biomarkers should be first demon-strate strong, independent, graded, and consistent asso-ciations with relevant clinical outcomes. In addition, the biomarker should have strong discriminatory character-istics that include a large difference between patients who present the clinical outcome of interest and those who do not.

In the present risk-adjusted analysis, we defined AKI according to the Acute Kidney Injury Network guide-lines. We observed postoperative AKI in 40.5% of pa-tients undergoing conventional CBP, and this result was quite similar to that recently reported by others[12]. In the propensity score matched cohort, minimized CPB was associated with an absolute reduction in AKI occur-rence of 13.5%. As expected, stage 1 AKI developed in the majority of patients after surgery. However, it has been clearly shown that even this early stage of renal injury may negatively affect patient outcomes[10, 12–14], and mini-CPB should be expected to improve surgical results, reducing its occurrence.

Several mechanisms may contribute to reduce AKI in patients operated on using mini-CPB. The limited he-modilution provided by the miniaturized CPB system permits a reduction of postoperative red blood cell trans-fusion, which is a well-known risk factor for AKI[15]. In addition, the low hematocrit values during extracorpo-real circulation have been shown to correlate signifi-cantly with high postoperative kidney proximal tubular damage assessed by N-acetylglucosaminidase and uri-nary interleukin-6 response [4]. A significant indepen-dent association was found between the lowest hemato-crit during bypass and AKI[16], with significant benefits on renal function after reduction of the bypass prime volume. Moreover, mini-CPB has been shown to provide a better tissue perfusion, which accounts for better kid-ney protection. That might be, in part, the effect of the volume-constant perfusion featured in the closed-loop miniaturized CPB system, regularly resulting in an

ele-Fig 1. Incidence of three degrees of acute kidney injury (AKI) in the mini-cardiopulmonary bypass (CPB) group versus conventional CPB group. The overall incidence of AKI was lower (28.8% versus 40.5%; p⫽ 0.03) among patients undergoing mini-CPB as well as among those with stage 1 AKI (24% versus 31.6% [white bars]); stage 2 AKI (2.8% versus 4.65% [light gray bars]), and stage 3 AKI (1.9% versus 4.3% [dark gray bars]).

Fig 2. Required study sample size to compare mini-cardiopulmonary bypass (CPB) versus conventional CPB in preventing acute kidney injury for different␣ and ␤ values. Assuming an absolute risk re-duction of 13.5%, the sample size required is 262 for an␣ value ⫽ 0.05 and a study power (1-␤) ⫽ 0.80. Df ⫽ 5, effect size ⫽ 0.27501.

ADULT

vated mean arterial pressure compared with conven-tional cardiopulmonary bypass. The increased perfusion pressure and the greater intravascular volume resulting from the removal of a venous reservoir may provide better capillary perfusion of all organs, including the kidney [17]. The impact of the inflammatory response induced by CPB within the kidney is not completely understood. It is interesting that animal models of renal ischemia-reperfusion injury have clearly demonstrated the pathologic role of interstitial inflammation and the elaboration of proinflammatory cytokines and reactive oxygen species in the production of tubular injury[18 – 21]. This local inflammatory response in experimental models is identical to that seen on a more global scale during CPB. Thus, it is likely that mini-CPB may addi-tionally reduce kidney injury by attenuating such an inflammatory response.

This study has some limitations. It was retrospective in design, and the use of the mini-CPB system was not randomized; therefore, our results may be influenced by treatment bias. We performed statistical adjustment in-cluding the use of propensity scores in an attempt to account for the nonrandomized nature of the analysis. However, our propensity analysis was only modestly predictive, which is perhaps not surprising because of few baseline differences between the two groups.

Although we identified the association between mini-CPB CABG and a decreasing risk of AKI, causality cannot be proven. Further randomized controlled trials ade-quately powered are needed to confirm whether the mini-CPB system improves renal outcome after CABG. The present study confirmed that the recent AKI defini-tion should be used as the renal endpoint in such trials. In fact, it strongly correlates with a patient’s outcome, and at the same time, allows one to adequately power a study comparing mini-CPB versus conventional CPB CABG with a reasonable sample size (Fig 2).

References

1. Thakar CV, Worley S, Arrigain S, Yared JP, Paganini EP. Influence of renal dysfunction on mortality after cardiac surgery: modifying effect of preoperative renal function. Kidney Int 2005;67:1112–9.

2. Rosner MH, Okusa MK. Acute kidney injury associated with cardiac surgery. Clin J Ame Nephrol 2006;1:19 –32.

3. Takagi H, Tanabashi T, Kawai N, Kato T, Umemoto T. Off-pump coronary artery bypass sacrifices graft patency: meta-analysis of randomized trials. J Thorac Cardiovasc Surg 2007;133:e2–3.

4. Huybregts RA, Morariu AM, Rakhorst G, et al. Attenuated renal and intestinal injury after use of a mini-cardiopulmonary bypass system. Ann Thorac Surg 2007;83: 1760 – 6.

5. Abdel-Rahman U, Ozaslan F, Risteski PS, et al. Initial expe-rience with a minimized extracorporeal bypass system: is there a clinical benefit? Ann Thorac Surg 2005;80:238 – 44. 6. Remadi JP, Rakotoarivelo Z, Marticho P, Benamar A.

Pro-spective randomized study comparing coronary artery by-pass grafting with the new mini-extracorporeal circulation Jostra System or with a standard cardiopulmonary bypass. Am Heart J 2006;151:198.e1–7.

7. Mehta RL, Kellum JA, Shah SV, et al. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007;11:R31.

8. Stevens LA, Coresh J, Greene T, Levey AS. Assessing kidney function—measured and estimated glomerular filtration rate. N Engl J Med 2006;354:2473– 83.

9. Nashef SA, Roques F, Hammill BG, et al. Validation of European System for Cardiac Operative Risk Evaluation (EuroSCORE) in North American cardiac surgery. Eur J Car-diothorac Surg 2002;22:101–5.

10. Lassnigg A, Schmid ER, Hiesmayr M, et al. Impact of minimal increases in serum creatinine on outcome in pa-tients after cardiothoracic surgery: do we have to revise current definitions of acute renal failure? Crit Care Med 2008;36:1129 –37.

11. Mehta RH, Grab JD, O’Brien SM, et al, for the Society of Thoracic Surgeons National Cardiac Surgery Database In-vestigators. Bedside tool for predicting the risk of postoper-ative dialysis in patients undergoing cardiac surgery. Circu-lation 2006;114:2208 –16.

12. Arora P, Rajagopalam S, Ranjan R, et al. Preoperative use of angiotensin-converting enzyme inhibitors/angiotensin re-ceptor blockers is associated with increased risk for acute kidney injury after cardiovascular surgery. Clin J Am Soc Nephrol 2008;3:1266 –73.

13. Ostermann M, Chang R, for the Riyadh ICU Program Users Group. Correlation between the AKI classification and out-come. Crit Care 2008;12:R144.

14. Lopes JA, Fernandes P, Jorge S, et al. Acute kidney injury in intensive care unit patients: a comparison between the RIFLE and the Acute Kidney Injury Network classifications. Crit Care 2008;12:R110.

15. Koch CG, Li L, Duncan AI, et al. Morbidity and mortality risk associated with red blood cell and blood-component trans-fusion in isolated coronary artery bypass grafting. Crit Care Med 2006;34:1608 –16.

16. Ranucci M, Menicanti L, Frigiola A. Acute renal injury and lowest hematocrit during cardiopulmonary bypass: not only a matter of cellular hypoxemia. Ann Thorac Surg 2005;78: 1880 –1.

17. Wiesenack C, Liebold A, Philipp A, et al. Four years’ expe-rience with a miniaturized extracorporeal circulation system and its influence on clinical outcome. Artif Organs 2004;28: 1082– 8.

18. Burne-Taney MJ, Rabb H. The role of adhesion molecules and T cells in ischemic renal injury. Curr Opin Nephrol Hypertens 2003;12:85–90.

19. Sheridan AM, Bonventre JV. Cell biology and molecular mechanisms of injury in ischemic acute renal failure. Curr Opin Nephrol Hypertens 2000;9:427–34.

20. Donnahoo KK, Meng X, Ayala A, Cain MP, Harken AH, Meldrum DR. Early kidney TNF expression mediates neu-trophil infiltration and injury after renal ischemia-reperfusion. Am J Physiol 1999;277:R922–9.

21. McCoy RN, Hill KE, Ayon MA, Stein JH, Burk RF. Oxidant stress following renal ischemia: changes in the glutathione redox ratio. Kidney Int 1988;33:812–7.

INVITED COMMENTARY

Cardiopulmonary bypass (CPB) has been germane to the development of cardiac surgery. Although its use is often necessary, CPB is associated with the development of a

well-described systemic inflammatory response. Benedetto and colleagues [1] examined the hypothesis that the Medtronic Resting Heart System mini-CPB

cir-© 2009 by The Society of Thoracic Surgeons 0003-4975/09/$36.00

Published by Elsevier Inc doi:10.1016/j.athoracsur.2009.04.031

ADULT