Severe acute postoperative renal failure (defined as

Transcription

1 On-Pump Versus Off-Pump Coronary Artery Bypass Grafting: Impact on Postoperative Renal Failure Requiring Renal Replacement Therapy Jan Bucerius, MD, Jan F. Gummert, MD, PhD, Thomas Walther, MD, PhD, Dierk V. Schmitt, MD, Nicolas Doll, MD, Volkmar Falk, MD, PhD, and Friedrich W. Mohr, MD, PhD Department of Cardiac Surgery, Heart Center, University of Leipzig, Leipzig, Germany Background. Despite refinements in perioperative patient management postoperative renal failure requiring hemofiltration or dialysis is still a common complication after coronary artery bypass grafting associated with impaired patient outcome. Methods. Prospective data on 9,631 patients receiving myocardial revascularization with (coronary artery bypass grafting [n 8,870]) or without cardiopulmonary bypass (off-pump coronary artery bypass grafting [n 761]) between April 1996 and August 2001 were evaluated by univariate and multivariate logistic regression analysis. Results. Overall prevalence of postoperative continuous renal replacement therapy was 4.1% (coronary artery bypass grafting, 4.3%; off-pump coronary artery bypass grafting, 1.8%; p 0.001). Thirty of 40 selected preoperative and intraoperative patient and treatment related variables had a high association with the requirement for postoperative renal replacement therapy; fifteen of these variables were independent predictors in the whole Severe acute postoperative renal failure (defined as renal failure requiring continuous renal replacement therapy) is a major complication of cardiac surgery associated with a very high mortality (40% to 100%) [1 4]. Contemporary practice of cardiac surgery encounters an increasing number of patients who are both older and afflicted with substantial additional risk factors as well as those requiring repeat cardiac surgery than previously. In parallel there are improvements in medical treatments such as off-pump coronary artery bypass grafting (OPCAB). Patients with certain concomitant diseases are at increased risk for a higher mortality and morbidity after cardiac surgery [5]. Care of these high-risk patients mandates the development of strategies to minimize the harmful effects of cardiac surgery and, in addition, cardiopulmonary bypass (CPB) with its undesirable effects that may contribute to an increased risk of postoperative patient s morbidity. Accepted for publication July 10, Address reprint requests to Dr Bucerius, University of Leipzig, Heart Center, Department of Cardiac Surgery, Strümpellstr. 39, Leipzig D-04289, Germany; bucerj@medizin.uni-leipzig.de. study population. Off-pump coronary artery bypass surgery was identified as having a significantly lower predictive value for postoperative continuous renal placement therapy. In the subgroup of patients undergoing off-pump coronary artery bypass grafting surgery, a second multivariate logistic regression model revealed preoperative cardiogenic shock, urgent operation, intraoperative low cardiac output, and high transfusion requirement as independent predictors for postoperative renal replacement therapy. Conclusions. Patients with preoperative nondialysis dependent renal insufficiency are at a high risk for further decline in renal function requiring postoperative continuous renal replacement therapy. Off-pump coronary artery bypass surgery is associated with a lower prevalence of postoperative renal replacement therapy after coronary artery bypass grafting. (Ann Thorac Surg 2004;77:1250 6) 2004 by The Society of Thoracic Surgeons Prevalence of acute renal failure requiring renal replacement therapy after cardiac surgery has been previously reported to vary from 1.0% to 7.0% in patients undergoing cardiac surgery with CPB [3]. However, there is a lack of data regarding prevalence and predictors of this severe postoperative complication in patients undergoing coronary artery bypass grafting, especially on the beating heart. This increasingly used surgical technique may be beneficial with regard to prevalence of postoperative renal failure requiring continuous renal replacement therapy (CRRT) even in high-risk subgroups. Aim of the present study was to systematically investigate the prevalence and possible predictors, as well as variables leading to acute renal failure requiring CRRT after coronary artery bypass grafting both with CPB and on the beating heart. Material and Methods Study Population Nine thousand, six hundred, thirty-one consecutive adult patients undergoing isolated coronary artery bypass grafting between April 1996 and August 2001 were in by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /s (03)

2 Ann Thorac Surg BUCERIUS ET AL 2004;77: CRRT AFTER CABG 1251 Table 1. Preoperative Patient Characteristics n Age [years] a Gender Male (%) CCS 3 (%) b NYHA 3 (%) c LVEF (%) d Preoperative Non-Dialysis Renal Failure e No. of Grafts f Total 9, ,463 (77.5) % CABG 8, ,853 (77.2) % OPCAB (80.2) % a p 0.013; b p ; c p 0.001; d p 0.06; e p 0.53; f coronary artery bypass grafts, p CABG conventional coronary artery bypass grafting; CCS Canadian Cardiovascular Society angina classification; LVEF left ventricular ejection fraction; NYHA New York Heart Association heart failure association; OPCAB off-pump coronary artery bypass grafting. cluded in this study. In 8,870 patients, conventional coronary artery bypass grafting (CABG) surgery with cardiopulmonary bypass was performed, 761 patients underwent CABG on the beating heart without CPB through a median sternotomy (OPCAB). Patients were assigned to one of the surgical techniques according to clinical indications as well as to the surgeon s decision to use the OPCAB or the conventional CABG approach. Patients undergoing minimally invasive direct coronary artery bypass grafting (MIDCAB) through a left lateral mini-thoracotomy were excluded from this analysis. Definition of Postoperative CRRT Postoperative CRRT was defined as any postoperative renal insufficiency requiring first time hemofiltration, dialysis or any other renal replacement therapy at our institution. Oligo-anuria or serum urea levels greater than or equal to 160 mg/dl, or both, have been indications for CRRT. A sole elevation of the serum creatinine has not been an indication for CRRT in most cases. However, the majority of patients requiring CRRT fulfilled more than one of the previously mentioned criteria. Postoperative time period in the intensive care unit in which renal failure requiring CRRT was assessed lasted for days in the study population. Patients suffering from preoperative renal failure requiring acute or chronic CRRT have been excluded from this analysis. Technique of CRRT The technique of CRRT consisted of a double lumen catheter that was placed in the venous system. The blood flow rate was kept as high as possible. Different types of hemofiltration and dialysis machines like Fresenius Multifiltrate, ADM/ABM 08 (Fresenius Medical Care, Bad Homburg, Germany) or Hospal Prisma (Hospal Medizintechnik GmbH, Planegg- Martinsried, Germany) were used. Replacement fluid was administered pre-filter at a dynamically adjusted rate chosen to achieve the desired fluid therapy goals for any time period. Systemic anticoagulation was carried out according to the standard anticoagulation protocol adapted to the performed surgical intervention and the circuit duration [1]. Data Collection Perioperative data were recorded prospectively using an online database system as previously described (Medwork database software (Lenz Partner GmbH, Germany) [6]. All variables analyzed were entered prospectively to accomplish a complete and valid data set for each patient. The validity of the data was routinely ensured by using this information for generating text documents, thus resulting in a meticulous confirmation of the entered data by the user. Preoperative and intraoperative risk factors included in a univariate analysis and consecutively in a stepwise logistic regression model are listed in the Appendix. Statistical Analysis Continuous variables are expressed as mean standard deviation, categorical data as proportions. Continuous variables between patients with and without postoperative CRRT were compared using the student s unpaired t test for variables, which were normally distributed or used the Mann Whitney U test for continuous variables without normal distribution. Categorical variables were compared by using 2 analysis. Univariate analysis of risk factors was performed by a calculating odds ratio (OR) with a 95% confidence interval. Variables with a p value of less than 0.05 were consecutively examined to a multivariate logistic regression model to assess the independent impact of the risk factors on postoperative CRRT. A stepwise procedure (the backward Wald test) was used. A p value less than 0.05 was used to enter and eliminate variables [7]. To identify potential risk factors for postoperative CRRT in the subgroup of patients undergoing OPCAB surgery, a second multivariate logistic regression model was performed with those preoperative and intraoperative risk factors in the Appendix marked with an asterisk. The design of this second multivariate regression analysis was the same as that previously stated. All statistical analyses were performed using the SPSS statistical package 9.0 (SPSS Corp, Chicago, IL). Results Patient characteristics are shown in Table 1. There were significant differences between all patients in the CABG and in the OPCAB group with regard to patient s age (p 0.013), preoperative New York Heart Association func-

3 1252 BUCERIUS ET AL Ann Thorac Surg CRRT AFTER CABG 2004;77: Table 2. Independent Predictors of Postoperative CRRT in the Whole Study Population Fig 1. Prevalence of postoperative continuous renal replacement therapy (postop. CRRT) according to the type of coronary artery bypass grafting in the total study population and in patients with preoperative renal disease. f total study population; e patients with preoperative renal disease. (CABG conventional coronary artery bypass grafting; OPCAB off-pump coronary artery bypass grafting.) tional class (p 0.001), and number of coronary artery bypass grafts (p ) all being higher in the CABG group. However, preoperative impaired CCS classification was significantly higher in the OPCAB group (p ). Preoperative left ventricular ejection fraction tended to be higher in the OPCAB group without reaching a statistically significant difference (p 0.06). Prevalence of preoperative nondialysis dependent renal failure was higher in the CABG group, but failed to reach statistical significance (p 0.53). Patients in the CABG group had a significant longer ICU stay (4.0 days 7.0 vs 2.8 days 5.8; p ) as well as total hospital stay (11.4 days 9.5 vs ; p ) as those undergoing OPCAB surgery. Furthermore, 30-day mortality was significantly higher in the CABG group (3.8% vs 2.1%; p 0.016). The prevalence of postoperative renal failure requiring CRRT was 4.1% for the total study population. Prevalence varied among both groups as shown in Figure 1 (p 0.001). In patients with preoperative renal disease (history of renal failure or pathologic elevated serum creatinine 2.0 mg/dl or urea treated medically without renal replacement therapy, or both; see Appendix), prevalence of postoperative CRRT was higher in those undergoing CABG surgery as compared with those in the OPCAB group reaching close to statistical significance (p 0.055) (Fig 1). Patients requiring postoperative CRRT had a significantly lower left ventricular ejection fraction (42.1% 23.7 vs 53.6% 20.3; p ) as patients without this postoperative complication. Furthermore, patients with postoperative CRRT suffered significantly more often from impaired preoperative New York Heart Association functional classification ( 3 [90.6% vs 78.9%; p ]) and from impaired CCS-classification ( 3 [50.4% vs 36.1%; p ]). However, no statistically significant differences with regard to patient s age have been found between patients with and without postoperative CRRT Odds- 95% Confidence Ratio Interval p Value Variables associated with a higher prevalence of postop CRRT a Preoperative variables: History of renal disease Diabetes mellitus Peripheral vascular disease CCS Myocardial infarction Previous CABG surgery Age 80 years Cardiogenic shock Urgent operation Intraoperative variables: Perfusion time 2 hours Hemofiltration RBC-transfusion 1000 ml IABP-support Assist device Low cardiac output Variables associated with a lower prevalence of postop CRRT a OPCAB a Analyzed by multivariate logistic regression analysis. CABG coronary artery bypass grafting; CCS Canadian Cardiovascular Society angina classification; CRRT continuous renal replacement therapy; IABP intra-aortic balloon pump; postop postoperative; RBC red blood cell. (65.4 years 13.4 vs 65.3 years 9.1; p 0.838). Number of bypass grafts was significantly lower in patients with postoperative renal replacement therapy (2.0 grafts 1.1 vs 2.5 grafts 0.9; p ), and 75.1% of those patients were male. By using univariate analysis, 30 of 40 variables were identified as having a high association with postoperative CRRT (see bold type in Appendix). By using stepwise logistic regression analysis 15 variables were identified as independent predictors of postoperative renal failure requiring renal replacement therapy. One variable was found to be associated with a significantly lower prevalence of this postoperative complication (Table 2). Prevalence and statistically significant differences of independent predictors as related to the different surgical procedures are shown in Table 3. By using a second stepwise logistic regression model, 4 of 33 preoperative and intraoperative risk factors were identified to be independently associated with postoperative CRRT in the subgroup of patients undergoing OPCAB surgery (Table 4). Patients requiring postoperative CRRT had a longer ICU stay (20.0 days 19.2 vs 3.2 days 4.7; p ) as well as a longer total postoperative hospital stay (24.2 days 24.5 vs 10.7 days 7.7; p ) than patients

4 Ann Thorac Surg BUCERIUS ET AL 2004;77: CRRT AFTER CABG 1253 Table 3. Prevalence of Independent Predictors of Postoperative CRRT According to the Type of Coronary Artery Bypass Grafting without CRRT. Overall perioperative (30-day) mortality rate in all patients was 3.7%. Postoperative renal failure requiring CRRT led to a more than 20-fold increase in mortality as compared with patients without this severe postoperative complication (42.7% vs 2.0%; p ) (Fig 2). Comment Total CABG OPCAB Variables associated with higher Prevalence (%) p Value risk of CRRT Preoperative variables: History of renal disease Diabetes mellitus Peripheral vascular disease CCS Myocardial Infarction Previous CABG surgery Age 80 years Cardiogenic shock Urgent operation Intraoperative variables: Perfusion time 2 hours Hemofiltration RBC - transfusion 1000 ml IABP - support Assist device Low cardiac output CABG coronary artery bypass grafting; CCS Canadian Cardiovascular Society angina classification; CRRT continuous renal replacement therapy; IABP intra-aortic ballon pump; RBC red blood cells. A significant proportion of patients undergoing CPB had a degree of renal dysfunction develop postoperatively [8]. Both the underlying cardiac disease with the subsequent Table 4. Independent Predictors of Postoperative CRRT in the Subgroup of Patients Undergoing OPCAB Surgery Odds- 95% Confidence Ratio Interval p Value Variables associated with a higher prevalence of postoperative CRRT a Preoperative variables: Cardiogenic shock Urgent operation Intraoperative variables: Low cardiac output RBC - transfusion 1000 ml a Analyzed by multivariate logistic regression analysis. CRRT continuous renal replacement therapy; RBC red blood cells. Fig 2. Intensive care unit (ICU) stay (days), total postoperative hospital stay (days), and 30-day mortality rate in patients with and without postoperative continuous renal replacement therapy (CRRT). Dot 30-day mortality rate. surgical intervention, as well as preoperative risk factors are important determinants of postoperative renal function. Beating heart surgery in patients undergoing coronary artery bypass grafting is an increasingly used surgical technique. Potentially undesirable effects of CPB that negatively affect postoperative renal function, such as inadequate perfusion pressure, renal hypoperfusion, nonpulsatile flow, or inflammatory reactions can thus be avoided [9 11]. In the current series comprised of more than 9,000 adult patients undergoing coronary artery bypass grafting using CPB or OPCAB technique, several preoperative and intraoperative risk factors significantly associated with CRRT-dependent postoperative renal failure have been identified. As reported in other series, a significant relationship has been found between the severity of the underlying cardiac disease and renal replacement therapy requiring renal failure. This association was revealed by a significant odds ratios of CCS greater than or equal to 3 (OR, 1.41), cardiogenic shock (OR, 3.28), and preoperative myocardial infarction (OR, 1.31) with all of them frequently resulting in the need for urgent operation (OR, 2.36), and postoperative CRRT-dependent renal failure [9, 12]. In addition, indication for repeat coronary artery bypass grafting has been significantly associated with postoperative CRRT in this series (OR, 1.95) as recently reported, probably due to severe underlying coronary artery disease [12, 13]. The effect of diabetes mellitus (OR, 1.38) on postoperative renal failure may be the result of renal parenchymal disease, such as glomerulonephritis or glomerulosclerosis. Furthermore, renal artery stenosis in diabetic patients may further compromise renal function. As previously stated, inadequate perfusion pressure during CPB and postoperative ICU periods may even elevate this risk [3, 9, 14]. In this context, it was not surprising that peripheral vascular disease (OR, 1.62) as a systemic disease frequently affecting both renal arteries was significantly

5 1254 BUCERIUS ET AL Ann Thorac Surg CRRT AFTER CABG 2004;77: associated with postoperative CRRT in this series as well [12]. Higher patient s age is one of the most reported preoperative risk factors for postoperative renal failure requiring CRRT [9, 12, 15, 16]. This finding was confirmed in this series as age of 80 years or older (OR, 1.99) and has been found to be significantly associated with postoperative renal failure. It was not surprising that preoperative nondialysis dependent renal dysfunction predicts a further decline in renal function leading to postoperative mechanical renal support as reported by several previous series (OR, 2.44) [3, 8, 9, 12, 14 16]. This may be due to renal parenchymal disorders being more susceptible to have postoperative renal failure develop after cardiac surgery. Renal cord ischemia may be a result of an intraoperative drop in renal perfusion pressure during onset of CPB or during mechanical lifting of the heart in OPCAB surgery. Renal failure in those predisposed patients is further depressed even during postoperative periods [9]. Intraoperative low cardiac output (OR, 2.8) leading to impaired perfusion pressure frequently after intraaortic balloon pump or assisted device support, or both (OR, 1.88 and 3.23, respectively) was significantly associated with postoperative CRRT in this series. This association has been reported by several previous studies [8, 13, 17]. It is obvious that the named devices are inserted in order to assist in the maintenance of cardiac output in unstable patients undergoing cardiac surgery. The use of both devices may also be associated with the development of postoperative renal failure if insertion of the cannulas or the balloon leads to limb ischemia and rhabdomyolysis [18]. Intraoperative hemofiltration (OR, 2.25) and prolonged perfusion time (OR, 2.62) are both variables inalienably associated with the use of cardiopulmonary bypass and have been shown as significant predictors of postoperative renal failure [13, 16, 19]. Both of them may indicate nonphysiological conditions leading to an impaired renal function during the intraoperative as well as the postoperative period. Need for intraoperative hemofiltration seems to be a marker for an existing preoperative renal insufficiency, a significant risk factor itself as previously mentioned. Intraoperative transfusion of a significant amount of red blood cells ( 1000 ml; OR, 1.86) most likely indicates unstable intraoperative conditions leading to an impaired renal function. Off-pump coronary artery bypass surgery was the only independent predictor significantly associated with a lower prevalence of postoperative CRRT requiring renal failure (OR, 0.55). In several previous studies, lower prevalence of postoperative complications (eg, cerebral deficits [stroke or postoperative delirium] or renal insufficiency), and a lower mortality rate have been documented for patients undergoing beating heart surgery as compared with patients undergoing conventional CABG [20 24]. Furthermore, in a series comprising more than 3,000 consecutive patients, Ascione and associates [25] demonstrated higher postoperative serum creatinine and urea levels in patients with preoperative nondialysisdependent renal insufficiency undergoing on-pump CABG with a significant difference at 12 hours postoperatively as compared with those undergoing OPCAB surgery. In addition, in a further study they reported a significantly impaired renal tubular function as assessed by increased N-acetyl glucosaminidase activity in the CABG group [26]. Loef and associates [27] found significantly less changes in microalbuminuria, free hemoglobin, fractional excretion of sodium, and free water clearance as well as N-acetyl- -D glucosaminidase as a marker for tubular function and damage, respectively, in patients undergoing OPCAB as compared with CABG patients. They conclude that off-pump coronary surgery attenuates renal injury after surgical myocardial revascularization. Hayashida and associates [28] found a significantly less increase in creatinine levels and a greater creatinine clearance in OPCAB patients as compared with the CABG group. Postoperative recovery of free water clearance was more prompt in the OPCAB group. In contrast to these findings, in a recent study by Gamoso and associates [15] including 690 patients, no significant reduction of perioperative renal dysfunction in OPCAB patients could been found. Avoiding CPB is beneficial even in patients with an existing preoperative renal insufficiency undergoing CABG as confirmed in this study (Fig 1) [25]. This benefit may be due to the avoidance of nonpulsatile flow, renal hypoperfusion, hypothermia, and prolonged duration of CPB for all of them thought to have adverse effects on renal function in patients undergoing off-pump coronary artery bypass grafting. When narrowing the analysis to the subgroup of patients receiving OPCAB surgery, the logistic regression model revealed preoperative cardiogenic shock (OR, 3.93) and urgent operation (OR, 9.8) as well as intraoperative low cardiac output (OR, 9.85) and high transfusion requirement (OR, 36.18) as independent predictors for postoperative renal failure requiring renal replacement therapy. As discussed in context with the risk factors for postoperative renal failure in the whole study population, all of these four predictors indicate an impaired preoperative or intraoperative cardiac function or unstable intraoperative conditions potentially leading to postoperative renal failure with the subsequent need for CRRT. It is obvious that this severe cardiac impairment can not be overcome totally by avoiding cardiopulmonary bypass in OPCAB surgery. Interestingly, patients with concomitant diseases or risk factors seem to benefit from the beating heart approach as all of the independent risk factors identified (except those previously named) in the whole study population did not show a significant association with postoperative renal failure requiring CRRT in the analyzed subgroup of patients undergoing OPCAB surgery. In this series, preoperative patient characteristics revealed significant differences regarding age, New York Heart Association functional class, prevalence of preoperative nondialysis dependent renal failure, and number of coronary bypass grafts between both surgical groups. This reflects that initially only selected patients received OPCAB surgery. The possibility of selection bias (as for

6 Ann Thorac Surg BUCERIUS ET AL 2004;77: CRRT AFTER CABG 1255 all nonrandomized studies) is the main limitation of the current study. However, it is important to note that the OPCAB group did not have the lowest prevalence of all independent predictors for postoperative CRRT as shown in Table 3. Furthermore, prevalence of impaired CCS classification was significantly higher in the OPCAB group, and prevalence of preoperative nondialysis dependent renal failure being one major independent risk factor for postoperative CRRT failed to be statistically significant between both groups. In addition, multivariable logistic regression analysis revealed that OPCAB surgery was associated with lower risk of postoperative renal replacement therapy, even after accounting for other risk factors. In addition, prevalence of postoperative CRRT was lower in patients with preoperative renal disease undergoing OPCAB surgery reaching close to statistical significance. Therefore we conclude that OPCAB surgery with avoidance of CPB seems to be associated with a significantly lower risk for postoperative renal failure requiring CRRT. In this series of patients after coronary artery bypass surgery either with cardiopulmonary bypass or on the beating heart, several preoperative and intraoperative patient or treatment variables have been shown to be significantly associated with postoperative CRRT. Understanding and analysis of the multivariate regressions presented in this study, possibly leading to a preoperative scoring-system estimating the risk of postoperative renal failure and subsequent continuous renal replacement therapy should allow the identification of high-risk subsets of patients undergoing coronary artery bypass graft surgery. This may lead to preoperative and intraoperative interventions, such as OPCAB surgery, to reduce severe postoperative renal failure. References 1. Bent P, Khim Tan H, Bellomo R, et al. Early and intensive continuous hemofiltration for severe renal failure after cardiac surgery. Ann Thorac Surg 2001;71: Lange HW, Aeppli DM, Brown DC. Survival of patients with acute renal failure requiring dialysis after open-heart surgery: early prognostic indicators. Am Heart J 1987;113: Baudouin SV, Wiggins J, Keogh BF, Morgan CJ, Evans TW. Continuous veno-venous haemofiltration following cardiopulmonary bypass. Intensive Care Med 1993;19: Mazzarella V, Galluci T, Tozzo C. Renal function in patients undergoing cardiopulmonary bypass operations. J Thorac Cardiovasc Surg 1992;104: Kiziltepe U, Uysalel A, Corapcioglu T, Dalva K, Akan H, Akalin H. Effects of combined conventional and modified ultrafiltration in adult patients. Ann Thorac Surg 2001;71: Gummert JF, Kluge M, Rei mann EM, Bung J, Mohr FW. Einführung eines komplexen medizinischen Dokumentationssystems am Herzzentrum Leipzig. In: Krian A, Scheld HH, eds. Dokumentationsverfahren in der Herzchirurgie III. Darmstadt: Steinkopff, 1998: Rosner BA. Fundamentals of biostatistics, 4th ed. Belmont, California: Duxbury Press, Corwin HL, Sprague SM, DeLaria GA, Norusis MJ. Acute renal failure associated with cardiac operations. A case control study. J Thorac Cardiovasc Surg 1989;98: Suen WS, Mok CK, Chiu SW, et al. Risk factors for development of acute renal failure (ARF) requiring dialysis in patients undergoing cardiac surgery. Angiology 1998;49: Bellomo R, Raman J, Ronco C. Intensive care unit management of the critically ill patient with fluid overload after open heart surgery. Cardiology 2001;96: Hall RI, Stafford Smith M, Rocker G. The systemic inflammatory response to cardiopulmonary bypass: pathophysiological, therapeutic and pharmacological considerations. Anesth Analg 1997;85: Chertow GM, Lazarus JM, Christiansen CL, et al. Preoperative renal risk stratification. Circulation 1997;95: Fortescue EB, Bates DW, Chertow GM. Predicting acute renal failure after coronary bypass surgery: cross-validation of two risk-stratification algorithms. Kidney Int 2000;57: Conlon PJ, Stafford-Smith M, White WD, et al. Acute renal failure following cardiac surgery. Nephrol Dial Transplant 1999;14: Gamoso MG, Phillips-Bute B, Landolfo KP, Newman MF, Stafford-Smith M. Off-pump versus on-pump coronary artery bypass surgery and postoperative renal dysfunction. Anesth Analg 2000;91: Mangos GJ, Brown MA, Chan WY, Horton D, Trew P, Whitworth JA. Acute renal failure following cardiac surgery: incidence, outcomes and risk factors. Aust N Z J Med 1995;25(4): Zanardo G, Michielon P, Paccagnella A, et al. Acute renal failure in the patient undergoing cardiac operation. Prevalence, mortality rate, and main risk factors. J Thorac Cardiovasc Surg 1994;107: Maccario M, Fumagalli C, Dottori V, et al. The association between rhabdomyolysis and acute renal failure in patients undergoing cardiopulmonary bypass. J Cardiovasc Surg Torino 1996;37: Llopart T, Lombardi R, Forselledo M, Andrade R. Acute renal failure in open heart surgery. Renal failure 1997;19(2): Murkin JM, Boyd WD, Ganapathy S, et al. Beating heart surgery: why expect less central nervous system morbidity? Ann Thorac Surg 1999;68: Kshettry VE, Flavin TF, Emery RW, Nicoloff DM. Does multi-vessel off pump coronary artery bypass (OPCABG) reduce postoperative morbidity? Ann Thorac Surg 2000;69: Mack MJ, Osborne JA, Shennib H. Arterial graft patency in coronary artery bypass grafting: what do we really know? Ann Thorac Surg 1998;66: Bull DA, Neumayer LA, Stringham JC, Meldrum P, Affleck DG, Karwande SV. Coronary artery bypass grafting with cardiopulmonary bypass versus off-pump cardiopulmonary bypass grafting: does eliminating the pump reduce morbidity and cost? Ann Thorac Surg 2001;71: Plomondon ME, Cleveland JC, Ludwig ST, et al. Off-pump coronary artery bypass is associated with improved riskadjusted outcomes. Ann Thorac Surg 2001;72: Ascione R, Nason G, Al-Ruzzeh S, Ko G, Ciulli F, Angelini GD. Coronary revascularization with or without cardiopulmonary bypass in patients with preoperative nondialysisdependent renal insufficiency. Ann Thorac Surg 2001;72: Ascione R, Lloyd CT, Underwood MJ, Gomes WJ, Angelini GD. On- pump versus off-pump coronary revascularization: evaluation of renal function. Ann Thorac Surg 1999;68: Loef BG, Epema AH, Navis G, Ebels T, van Oeveren W, Henning RH. Off-pump coronary revascularization attenuates transient renal damage compared with on-pump coronary revascularization. Chest 2002;121(4): Hayashida N, Teshima H, Chihara S, et al. Does off-pump coronary artery bypass grafting really preserve renal function? Circ J 2002;66:921 5.

7 1256 BUCERIUS ET AL Ann Thorac Surg CRRT AFTER CABG 2004;77: Appendix Perioperative Variables Analyzed by Univariate Analysis (Factors That Had a Statistically Significant Association With Postoperative CRRT are Printed in Bold) Univariate analysis of the whole study population (n 9,631) Patients with postop CRRT (n 393) Patients without postop CRRT (n 9,238) Prevalence % (n) p Value OR Preoperative variables Age > 80 years a 6.1 (24) 3.0 (277) History of embolism a 5.3 (21) 2.5 (231) History of syncope a 12.9 (51) 7.4 (683) History of cardiogenic shock a 32.5 (128) 6.7 (616) Diabetes mellitus a 43.5 (171) 36.8 (3,396) (Glucose intolerance treated with diet, oral hypoglycemics or insulin) History of renal disease a 12.2 (48) 3.0 (278) (History of renal failure or pathological elevated serum creatinine ( 2.0 mg/dl) treated medically without hemofiltration and/or dialysis) Preoperative infection a 2.5 (10) 0.9 (79) (Infectious disease including endocarditis) NYHA > 3 a 90.6 (356) 78.9 (7,292) LVEF < 30 % a 17.3 (68) 7.9 (738) (Assessed by angiography or 2D echocardiography) History of peripheral vascular disease a 33.1 (130) 21.0 (1,940) Atrial fibrillation a 7.9 (31) 3.9 (368) (History of preoperative atrial fibrillation) Urgent operation a 56.2 (221) 20.0 (1,849) Emergency surgery a 2.5 (10) 0.7 (66) (Emergent surgery due to complications during coronary angiography and/or PTCA) Prior myocardial infarction a 67.2 (264) 54.9 (5,068) CCS > 3 a 50.4 (198) 36.1 (3,333) Prior cardiac surgery a 8.9 (35) 3.7 (341) Prior CABG a 8.1 (32) 3.1 (288) Age 50 years a 7.1 (28) 6.6 (613) Age 60 and 70 years a 38.4 (151) 40.5 (3,737) Age 70 and 80 years a 36.9 (145) 32.4 (2,995) Arterial hypertension a 72.3 (284) 73.9 (6,834) (Patient taking antihypertensive medication preoperatively) Sex (Male) a 75.1 (295) 77.6 (7,168) Hyperlipidemia a 50.1 (197) 53.6 (4,953) Prior aortic valve surgery a 0.3 (1) 0.3 (29) Prior mitral valve surgery a 0.5 (2) 0.2 (16) Intraoperative variables Duration of surgery > 3 hours a 35.1 (138) 19.0 (1,758) Total CPB time > 2 hours 30.3 (119) 6.3 (582) Intraoperative hemofiltration 24.9 (98) 5.8 (540) Intraoperative hypothermia < 32 C 33.8 (133) 26.3 (2,428) Intraoperative RBC-transfusion > 1000 ml a 12.5 (49) 1.4 (125) Intraoperative low cardiac output a 20.1 (79) 1.5 (141) Intraoperative IABP-support a 18.6 (73) 1.8 (166) Intraoperative assist device a 10.2 (40) 0.3 (31) (ECMO, Berlin heart ) Intraoperative need for pacemaker stimulation a 30.0 (118) 15.0 (1,388) Coronary bypass grafts > 2 a 69.9 (275) 87.1 (8,047) Use of cardioplegia 61.8 (243) 80.1 (7,402) OPCAB 3.6 (14) 8.1 (747) Intraoperative blood loss > 500 ml a 2.3 (9) 0.4 (35) Ischemic time 1 hour 13.5 (53) 10.7 (992) a Variables included in the logistic regression analysis to identify risk factors for postoperative renal failure requiring CRRT in the subgroup of patients undergoing OPCAB surgery (n 761). CABG coronary artery bypass grafting; CCS Canadian Cardiovascular Society angina classification; CPB cardiopulmonary bypass; CRRT continuous renal replacement therapy; ECMO extracorporal membrane oxygenation; IABP intra-aortic ballon pump; LVEF left ventricular ejection fraction; NYHA New York Heart Association heart failure association; OPCAB off-pump coronary artery bypass grafting; OR Odds ratio; postop postoperative; PTCA percutaneous trans coronary angioplasty; RBC red blood cells.