Impact of Acute Kidney Injury on One-Year Survival After Surgery for Aortic Dissection

Transcription

1 Impact of Acute Kidney Injury on One-Year Survival After Surgery for Aortic Dissection ADULT CARDIAC Hsing-Shan Tsai, MD, Feng-Chun Tsai, MD, Yung-Chang Chen, MD, Lung-Sheng Wu, MD, Shao-Wei Chen, MD, Jaw-Ji Chu, MD, Pyng-Jing Lin, MD, and Pao-Hsien Chu, MD, FCCP Division of Cardiology, Department of Internal Medicine, Division of Thoracic and Cardiovascular Surgery, Department of Surgery, and Department of Nephrology, Department of Internal Medicine, Chang Gung Memorial Hospital, Chang Gung University College of Medicine, Taipei, Taiwan Background. Surgical treatment is an option for both type A aortic dissection and complicated type B aortic dissection. Acute kidney injury (AKI) influences the disease course after surgery. Our hypothesis was that AKI should be an important prognostic factor for aortic dissection after surgical treatment. Methods. Between July 2005 and October 2010, 268 patients (mean age years; range, 16 to 88) underwent open surgery for aortic dissection. We reviewed the clinical presentations, surgical variables, and postoperative outcomes to identify the risk factors of death. The 256 patients were divided into groups, with and without AKI, within 24 hours after operation according to the RIFLE (acronym for risk, injury, failure, loss, end stage) criteria. Results. The in-hospital mortality rate was 17.9%, the 1-year mortality rate was 18.7%, and the major adverse cardiac events rate within 1 year was 29.9%. In multivariate analysis, patients more than 70 years of age (hazard ratio [HR] 2.390, p 0.029), cardiogenic shock (HR 2.895, p 0.005), preoperative ventilator use (HR 4.137, p 0.018), operation at midnight (HR 2.295, p 0.028), longer bypass time (HR 1.007, p < 0.001), and AKI (HR 2.552, p 0.041) were clinical predictors of mortality. Kaplan- Meier analysis showed that the survival rate was strongly correlated with the severity of AKI by the RIFLE criteria. The independent predictors of AKI included hypertension (odds ratio 2.340, p 0.027), sepsis (odds ratio 2.594, p 0.043), and lower limb malperfusion (odds ratio 4.558, p 0.022). Conclusions. Our study provides outcomes of postoperative aortic dissection. We found that AKI was a predictor of 1-year mortality by using the RIFLE criteria. Factors associated with increased 1-year mortality and AKI should be taken into consideration for surgery and postoperative care. (Ann Thorac Surg 2012;94: ) 2012 by The Society of Thoracic Surgeons Accepted for publication May 25, Address correspondence to Dr Chu, First Cardiovascular Division, Department of Internal Medicine, Chang Gung Memorial Hospital, 199 Tun-Hwa North Rd, Taipei 105, Taiwan; pchu@adm.cgmh.org.tw. Aortic dissection is a life-threatening disease. Surgical treatment is an option for type A aortic dissection and complicated type B aortic dissection [1]. Despite the development of endovascular interventions for aortic dissection and the improvement of postoperative management, open surgery remains the gold standard for acute type A dissection. A variety of risk factors of in-hospital and long-term mortality of aortic dissection have been elucidated [2 8]. Acute kidney injury (AKI) is an independent outcome predictor in patients undergoing cardiac surgery [9 14]. Methods of preventing AKI using off-pump cardiac surgery and the necessity of hemodialysis before and after cardiac surgery have been discussed. Recently, the RIFLE criteria (risk, injury, failure, loss, end stage) have been developed as a classification system for AKI, and it has been widely applied in intensive and postoperative care [10, 11, 15 23]. A few studies have focused on the postoperative conditions of aortic dissection and AKI; however, none has used the RIFLE criteria [13, 24]. In this retrospective analysis, we used the RIFLE criteria to predict the 1-year prognosis of aortic dissection surgery. Material and Methods Study Population Between July 2005 and October 2010, 268 sequential patients undergoing open surgery for aortic dissection at the Linkou Chang Gung Memorial Hospital were reviewed. The local Institutional Review Board of the hospital approved this retrospective study (CGMH96048B) and waived the need for individual consent. All data in our study were anonymized. Aortic dissection was proven by enhanced computed tomography and defined as type A or type B according to the Stanford classification. The patients underwent an emergency operation if they were diagnosed with Stanford type A aortic dissection or complicated Stanford type B aortic dissection with life threatening conditions (eg, rupture, malperfusion syndrome, progression of dissection, enlarging aneurysm, inability to control blood pressure, or symptoms) after surgical consultation and their agreement. The patients with complicated type B aortic dissection who 2012 by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc

2 ADULT CARDIAC 1408 TSAI ET AL Ann Thorac Surg IMPACT OF AKI AFTER AORTIC DISSECTION SURGERY 2012;94: were at high risk of life-threatening conditions underwent urgent repair, and the remaining patients received elective surgery. Type B aortic dissection without associated complications was managed medically. Endovascular intervention was excluded from this retrospective analysis as we wanted to focus on the effects of traditional surgery. Four patients receiving regular hemodialysis and 8 patients who died on the first day after surgery were excluded from the subgroup analysis, as data on changes in consequent renal function were not available. Subgroup analysis of AKI was performed among the remaining 256 patients. Data Collection This was a retrospective study to review the baseline demographics, clinical data, mortality, and major adverse cardiac events (MACE) such as myocardial infarction, stroke, and cardiac death, for 1 year after surgery [25 27]. The preoperative renal function was defined by the estimated glomerular filtration rate, using the Chinese Modification of Diet in Renal Disease equation [28]. The RIFLE criteria were used to classify patients postoperatively into with AKI and without AKI groups by using the last preoperative and the 24-hour postoperative serum creatinine levels. RIFLE Criteria The RIFLE criteria were defined based on the magnitude of elevation in either serum creatinine or urine output, as listed in Table 1. We used the maximum change in serum creatinine level on the first day after surgery to classify patients according to the RIFLE criteria. Statistical Analysis Categorical variables were expressed as frequencies and percentages, and continuous variables as mean SD. The two-tailed Student s t test was used to compare continuous variables, and Pearson s 2 or Fisher s exact test to compare categorical variables. Significance was defined at a probability value of less than Potential relationships between clinical and surgical variables and death were explored in a time-to-event framework using Cox proportional hazard models. Kaplan-Meier curves Table 1. The RIFLE Criteria for Acute Kidney Injury RIFLE Criteria Serum Creatinine GFR Urine Output Risk 1.5-fold increase Decrease 25% ml kg 1 h for 6 hours Injury Twofold increase Decrease 50% ml kg 1 h for 12 hours Failure Threefold increase Decrease 75% 0.5 ml kg 1 h 1 for 24 hours Loss Complete loss of kidney function for more than 4 weeks ESRD Complete loss of kidney function for more than 3 months ESRD end-stage renal disease; GFR glomerular filtration rate; RIFLE risk, injury, failure, loss, end stage. were used to demonstrate the effect of RIFLE classification on survival at 1 year. Associations between variables and AKI were identified by a logistic regression equation, and in stepwise multivariate analysis. All statistical analyses were performed using Predictive Analytics Software, version 18.0 (SPSS, Chicago, IL). Results Study Population Features The 268 patients in this retrospective analysis were predominantly male (72.8%), with a mean age of years (range, 16 to 88), and 31 patients (13.8%) were older than 70 years of age (Table 2). A majority 222 of 268 patients (82.8%) had diagnosed with hypertension, and 56% (124 of 222) of these patients had never received antihypertensive drugs. Comorbid conditions included 103 patients who had preoperative renal function impairment defined as an estimated glomerular filtration rate of less than 60 ml/min (38.4%), Marfan syndrome (12.3%), and a history of prior aortic surgery (19.0%). On admission, 158 patients (59%) presented with chest pain. The frequency of syncope, branch involvement, cardiac tamponade, and unstable hemodynamic status were demonstrated. Outcomes Type A aortic dissection surgery (166 of 268, 61.9%) and emergency operations (176 of 268, 65.7%) were most commonly performed. Thirty-nine patients (14.6%) needed another operation to check internal bleeding after the aortic surgery, 34 patients (12.7%) had sepsis documented by blood culture results, and 19 patients (7.1%) were complicated with lower limb malperfusion after surgery. Acute kidney injury within 24 hours after operation occurred in more than half of the patients (52.7%) according to the RIFLE criteria, and 30 patients (11.2%) received temporary hemodialysis. The inhospital mortality rate was 17.9%, the 1-year mortality rate was 18.7%, and the MACE rate was 29.9%. A risk prediction model was then developed. In univariate analysis, advanced age (greater than 70 years), impaired preoperative renal function (estimated glomerular filtration rate 60 ml/min), syncope, preoperative ventilator use, cardiogenic shock, branch involvement, acute type A, emergency operation, longer operation time, longer bypass time, prolonged ventilator use after surgery (more than 7 days), rechecking bleeding, sepsis, AKI, and the necessity of temporary hemodialysis were associated with increased mortality. However, Bentall surgery did not increase the postoperative mortality rate. Stepwise multivariate analysis identified patients older than 70 years of age (hazard ratio [HR] 2.390, p 0.029), cardiogenic shock (HR 2.895, p 0.005), preoperative ventilator use (HR 4.137, p 0.018), operation at midnight (HR 2.295, p 0.028), longer bypass time (HR 1.007, p 0.001), and AKI (HR 2.552, p 0.041) to be independent predictors of 1-year mortality (Table 3).

3 Ann Thorac Surg TSAI ET AL ;94: IMPACT OF AKI AFTER AORTIC DISSECTION SURGERY Table 2. Clinical Presentations, Surgery, and Postoperative Results of Postoperative Aortic Dissection At One Year Variable Overall Survived Died p Value ADULT CARDIAC Patients Characteristics Male 195 (72.8) 162 (74.3) 33 (66.0) Age, years Age 70 years 31 (13.8) 25 (11.5) 12 (24.0) Smoking 131 (48.9) 109 (50.0) 22 (44.0) Systemic diseases Hypertension 222 (82.8) 186 (85.3) 36 (72.0) Diabetes mellitus 21 (7.8) 16 (7.3) 5 (10.0) egfr 60 ml/min 103 (38.4) 77 (35.3) 26 (52.0) End-stage renal disease 4 (1.5) 4 (1.8) 0 (0) Marfan syndrome 33 (12.3) 27 (12.4) 6 (12.0) Prior aortic surgery 51 (19.0) 47 (21.6) 4 (8.0) Clinical presentations Chest pain 158 (59) 124 (56.9) 34 (68) Syncope 22 (8.2) 14 (6.4) 8 (16.0) Preoperative ventilator 18 (6.7) 7 (3.2) 11 (22.0) Cardiogenic shock 35 (13.1) 16 (7.3) 19 (38.0) Cardiac tamponade 18 (6.7) 9 (4.1) 9 (18.0) Carotid dissection 5 (1.9) 2 (0.9) 3 (6.0) Coronary dissection 9 (3.4) 4 (1.8) 5 (10.0) Renal infarction 19 (7.1) 17 (7.8) 2 (4.0) Neurologic dysfunction 6 (2.2) 3 (1.4) 3 (6.0) Surgery characteristics Acute type A 166 (61.9) 127 (58.3) 39 (78.0) Emergency operation 176 (65.7) 137 (62.8) 39 (78) Operation time, minutes Operation at midnight 113 (42.2) 78 (35.8) 35 (70.0) Bentall surgery 29 (10.8) 20 (9.2) 9 (18) Bypass time, minutes Clamp time, minutes a Arrest time, minutes a Postoperative events Recheck bleeding 39 (14.6) 25 (11.5) 14 (28.0) Intubation time, hours Tracheostomy 11 (4.1) 7 (3.2) 4 (8.0) Ventilator 7 days 53 (19.8) 32 (14.7) 21 (42.0) Sepsis 34 (12.7) 20 (9.2) 14 (28) Acute kidney injury 135 (52.7) 102 (48.1) 32 (76.2) Temporary hemodialysis 30 (11.2) 12 (5.6) 18 (36.0) Lower limb malperfusion 19 (7.1) 16 (7.3) 3 (6.0) Stroke 43 (16) 31 (14.2) 12 (24.0) ICU stay, days Hospital stay, days MACE 80 (29.9) 31 (14.2) 49 (98) Death in hospital 48 (17.9) Death within 1 year 50 (18.7) a Data on cross-clamp time for 210 patients; data on circulatory arrest time for 239 patients. Data presented as n (%) or mean SD. egfr estimated glomerular filtration rate; ICU intensive care unit; MACE major adverse cardiac events.

4 ADULT CARDIAC 1410 TSAI ET AL Ann Thorac Surg IMPACT OF AKI AFTER AORTIC DISSECTION SURGERY 2012;94: Incidence of AKI and Its Clinical Implications In the subgroup analysis, the 256 patients were divided into four groups according to the serum creatinine elevation 24 hours after operation, excluding 4 patients on preoperative regular dialysis and 8 patients who died on the day of the surgery as data on serum creatinine levels were not available. Among 135 patients in whom AKI developed within 24 hours after aortic surgery, 57 (47.3%) had an episode of AKI classified as RIFLE-R (risk), 56 (21.9%) had RIFLE-I (injury), and 22 (8.6%) had RIFLE-F (failure). The remaining 121 patients (47.3%) who did not have AKI on the first day were classified into to the fourth group. Kaplan-Meier survival analysis of the individual RIFLE categories showed that the 1-year survival rate differed significantly by the magnitude of AKI (p 0.02; Fig 1). The patients were then separated into with AKI and without AKI groups to clarify the risk factors by univariate analysis (Table 4). The AKI group tended to have higher percentages of hypertension, more operations at midnight, longer operation time, longer bypass time, longer postoperative ventilator use, longer intensive care unit and hospital stay, more sepsis, higher incidence of lower limb malperfusion, and higher mortality rate. However, the AKI group was younger, had less preoperative ventilator use, and no significant cardiogenic shock on arrival. Finally, among all patients after aortic surgery, multivariate analysis identified hypertension (odds ratio 2.340, p 0.027), sepsis (odds ratio 2.594, p 0.043), and lower limb malperfusion (odds ratio 4.558, p 0.022) as independent predictors of postoperative AKI (Table 5). Comment Repair for type A aortic dissection or complicated type B aortic dissection saves life. However, the in-hospital mortality and long-term mortality rates remain high despite decades of experience, advances in diagnosis, improved surgical techniques, myocardial protection, and postoperative management [3, 7, 8, 10, 29 31]. In this study, we found an in-hospital mortality rate of 17.9%, 1-year mortality rate of 18.7%, postoperative stroke rate of 16%, and MACE rate of 29.9%. The mortality results Table 3. Multivariate Analysis for Clinical Outcomes of One-Year Mortality Variable Hazard Ratio 95% CI p Value Age 70 years Cardiogenic shock Preoperative ventilator Operation at midnight Bypass time Sepsis Acute kidney injury CI confidence interval. Fig 1. Survival analysis of patients in different stages of acute kidney injury (AKI) showed the prognosis to be related to the severity of renal dysfunction (p 0.02). The number of patients at risk is shown at the bottom of the graph. (RIFLE risk, injury, failure, loss, end stage renal disease.) are comparable to previous studies, with reported rates ranging from 13% to 30% [8, 29, 32 36]. A variety of risk factors, including age, clinical presentation, poor preoperative renal function, false lumen size, a previous cardiac surgery history, prolonged bypass time, and postoperative AKI, have been discussed after cardiac surgery [2 5, 10, 29, 37]. In this study, advanced age (greater than 70 years) and cardiogenic shock could predict a poor clinical outcome. However, a history of prior aortic surgery should not preclude patients from surgical correction of aortic dissection as stated in previous reports [33]. During an operation, a prolonged bypass times increases the possibility of death. After the operation, AKI within 24 hours was also an important predictor of mortality. The reported incidence of AKI after cardiac surgery depends on its definition. The high incidence of AKI in this study is a concern, as it may lead to a higher dialysis rate. Acute kidney injury is an important issue in cardiac surgery, as it can lead to a worse prognosis and death [9, 10]. Our study used the RIFLE criteria to show that the severity of AKI within 24 hours after operation is a good predictor for 1-year mortality after aortic correction. Most patients who died (32 of 42, 76.2%) experienced AKI after surgery during the first 24 hours, and 84% of the patients (27 of 32) underwent temporal hemodialysis postoperatively. The first 24 hours is critical for the prediction of AKI, and AKI may be preventable, thereby leading to a reduction in the mortality rate. It is simple and cost effective to use the maximum changes in serum creatinine levels on the first day after aortic surgery according to the RIFLE criteria to predict the mortality risk.

5 Ann Thorac Surg TSAI ET AL 2012;94: IMPACT OF AKI AFTER AORTIC DISSECTION SURGERY Table 4. Clinical Presentations, Surgery, and Postoperative Results of Patients With and Patients Without Postoperative Acute Kidney Injury Variable Nonacute Kidney Injury Acute Kidney Injury p Value Number of patients Patient characteristics Male 83 (68.6) 107 (79.3) 0.51 Age, years Age 70 years 22 (18.2) 14 (10.4) Smoking 59 (48.8) 70 (51.9) Systemic diseases Hypertension 93 (76.9) 119 (88.1) Diabetes mellitus 7 (5.8) 14 (10.4) egfr 60 ml/min 46 (38.0) 47 (34.8) Clinical presentations Chest pain 65 (53.7) 85 (63.0) Syncope 12 (9.9) 7 (5.2) Preoperative ventilator 10 (8.3) 2 (1.5) Cardiogenic shock 15 (12.4) 13 (9.6) Cardiac tamponade 8 (6.6) 7 (5.2) Carotid dissection 2 (1.7) 2 (1.5) Coronary dissection 2 (1.7) 6 (4.4) Renal infarct 5 (4.1) 14 (10.4) Neurologic dysfunction 2 (1.7) 2 (1.5) Surgery characteristics Acute type A 56 (55.4) 89 (65.9) Emergency operation 73 (60.3) 93 (68.9) Operation time, minutes Operation at midnight 39 (32.2) 68 (50.4) Bentall surgery 12 (9.9) 17 (12.6) Bypass time, minutes Clamp time, minutes a Arrest time, minutes a Postoperative events Recheck bleeding 6 (5.0) 32 (23.7) Intubation time, hours ICU stay, days Hospital stay, days Sepsis 8 (6.6) 24 (17.8) Tracheostomy creation 3 (2.5) 8 (5.9) Ventilator 7 days 11 (9.1) 41 (30.4) Temporary 3 (2.5) 27 (20.0) hemodialysis Lower limb 3 (2.5) 16 (11.9) malperfusion Stroke 10 (8.3) 33 (24.4) Death in hospital 9 (7.4) 31 (23.0) Death within 1 year 10 (8.3) 32 (23.7) MACE 18 (14.9) 54 (40.0) a Data on cross-clamp time for 210 patients; data on circulatory arrest time for 239 patients. Data presented as n (%) or mean SD. egfr estimated glomerular filtration rate; ICU intensive care unit; MACE major adverse cardiac events. Table 5. Multivariate Analysis for Acute Kidney Injury on First Day After Surgery for Aortic Dissection Variable Odds Ratio 95% CI p Value Hypertension Operation time Operation at midnight Bypass time Sepsis Lower limb malperfusion CI confidence interval. After multivariate analysis, hypertension, sepsis, and the occurrence of lower limb malperfusion were independent predictors of AKI. Hence, further prospective studies are needed to investigate strategies (eg, the control of blood pressure, the prevention of sepsis, and the improvement of lower limb perfusion) to decrease AKI and lower the mortality rate. Study Limitations This is a retrospective study, and some data of the cross-clamp time and circulatory arrest time were missing. Among the 268 patients, we only had data on cross-clamp times for 210 patients, and circulatory arrest times for 239. The missing data may have led to statistical inaccuracies. In addition, we did not have complete renal function data after 24 hours, and we were not able to accurately compare the development of AKI beyond 24 hours after surgery. In conclusion, aortic dissection is a disease associated with a high mortality rate. In multivariate analysis, patients older than 70 years of age, with cardiogenic shock, preoperative ventilator use, operation at midnight, prolonged bypass time, and AKI could predict the 1-year mortality rate after surgery for aortic dissection. Acute kidney injury, defined according to the RIFLE criteria, was an accurate early predictor of 1-year mortality, and could help in making decisions about postoperative care. In multivariate analysis, hypertension, sepsis, and the occurrence of lower limb malperfusion were predictors of AKI. With the advances in medical care and standardized surgical techniques, reducing aortic dissection-related complications and comorbidities may represent the most efficacious strategy for improving results. References Nienaber CA, Eagle KA. Aortic dissection: new frontiers in diagnosis and management. Part II: therapeutic management and follow-up. Circulation 2003;108: Nallamothu BK, Mehta RH, Saint S, et al. Syncope in acute aortic dissection: diagnostic, prognostic, and clinical implications. Am J Med 2002;113: Stamou SC, Hagberg RC, Khabbaz KR, et al. Is advanced age a contraindication for emergent repair of acute type A aortic dissection? Interact Cardiovasc Thorac Surg 2010;10: Akutsu K, Nejima J, Kiuchi K, et al. Effects of the patent false lumen on the long-term outcome of type B acute aortic dissection. Eur J Cardiothorac Surg 2004;26: ADULT CARDIAC

6 ADULT CARDIAC 1412 TSAI ET AL Ann Thorac Surg IMPACT OF AKI AFTER AORTIC DISSECTION SURGERY 2012;94: Onitsuka S, Akashi H, Tayama K, et al. Long-term outcome and prognostic predictors of medically treated acute type B aortic dissections. Ann Thorac Surg 2004;78: Gaul C, Dietrich W, Erbguth FJ. Neurological symptoms in aortic dissection: a challenge for neurologists. Cerebrovasc Dis 2008;26: Booher AM, Isselbacher EM, Nienaber CA, et al. Ascending thoracic aorta dimension and outcomes in acute type B dissection (from the International Registry of Acute Aortic Dissection [IRAD]). Am J Cardiol 2011;107: Santini F, Montalbano G, Casali G, et al. Clinical presentation is the main predictor of in-hospital death for patients with acute type A aortic dissection admitted for surgical treatment: a 25 years experience. Int J Cardiol 2007;115: Brown JR, Kramer RS, Coca SG, et al. Duration of acute kidney injury impacts long-term survival after cardiac surgery. Ann Thorac Surg 2010;90: Robert AM, Kramer RS, Dacey LJ, et al. Cardiac surgeryassociated acute kidney injury: a comparison of two consensus criteria. Ann Thorac Surg 2010;90: De Santo LS, Romano G, Amarelli C, et al. Implications of acute kidney injury after heart transplantation: what a surgeon should know. Eur J Cardiothorac Surg 2011;40: Kowalik MM, Lango R, Klajbor K, et al. Incidence- and mortality-related risk factors of acute kidney injury requiring hemofiltration treatment in patients undergoing cardiac surgery: a single-center 6-year experience. J Cardiothorac Vasc Anesth 2011;25: Mariscalco G, Lorusso R, Dominici C, et al. Acute kidney injury: a relevant complication after cardiac surgery. Ann Thorac Surg 2011;92: Coca SG, Jammalamadaka D, Sint K, et al. Preoperative proteinuria predicts acute kidney injury in patients undergoing cardiac surgery. J Thorac Cardiovasc Surg 2012;143: Jenq CC, Tsai MH, Tian YC, et al. RIFLE classification can predict short-term prognosis in critically ill cirrhotic patients. Intensive Care Med 2007;33: Lin CY, Tsai FC, Tian YC, et al. Evaluation of outcome scoring systems for patients on extracorporeal membrane oxygenation. Ann Thorac Surg 2007;84: Chen YC, Jenq CC, Tian YC, et al. Rifle classification for predicting in-hospital mortality in critically ill sepsis patients. Shock 2009;31: Lin CY, Kao KC, Tian YC, et al. The RIFLE score increases the accuracy of outcome prediction in patients with acute respiratory distress syndrome undergoing open lung biopsy. Respiration 2009;77: Chang CH, Lin CY, Tian YC, et al. Acute kidney injury classification: comparison of AKIN and RIFLE criteria. Shock 2010;33: Lin CY, Kao KC, Tian YC, et al. Outcome scoring systems for acute respiratory distress syndrome. Shock 2010;34: Fan PC, Chang CH, Tsai MH, et al. Predictive value of acute kidney injury in medical intensive care patients with sepsis originating from different infection sites. Am J Med Sci 2011 Dec 2 [E-pub ahead of print]. 22. Hoste EA, Clermont G, Kersten A, et al. RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care 2006;10:R Bagshaw SM, George C, Dinu I, et al. A multi-centre evaluation of the RIFLE criteria for early acute kidney injury in critically ill patients. Nephrol Dial Transplant 2008;23: Apaydin AZ, Buket S, Posacioglu H, et al. Perioperative risk factors for mortality in patients with acute type A aortic dissection. Ann Thorac Surg 2002;74: Chu PH, Chiang CW, Hsu LA, et al. Low prevalence of coronary arterial disease in Chinese adults with mitral stenosis. Chang Gung Med J 2001;24: Tang CH, Wu CS, Lee TH, et al. Preeclampsia-eclampsia and the risk of stroke among peripartum in Taiwan. Stroke 2009;40: Lin YS, Tang CH, Yang CY, et al. Effect of pre-eclampsiaeclampsia on major cardiovascular events among peripartum women in Taiwan. Am J Cardiol 2011;107: Ma YC, Zuo L, Chen JH, et al. Modified glomerular filtration rate estimating equation for Chinese patients with chronic kidney disease. J Am Soc Nephrol 2006;17: Trimarchi S, Nienaber CA, Rampoldi V, et al. Contemporary results of surgery in acute type A aortic dissection: the International Registry of Acute Aortic Dissection experience. J Thorac Cardiovasc Surg 2005;129: Acosta S, Blomstrand D, Gottsater A. Epidemiology and long-term prognostic factors in acute type B aortic dissection. Ann Vasc Surg 2007;21: Tanaka M, Kimura N, Yamaguchi A, et al. In-hospital and long-term results of surgery for acute type a aortic dissection: 243 consecutive patients. Ann Thorac Cardiovasc Surg 2012;18: Suzuki T, Mehta RH, Ince H, et al. Clinical profiles and outcomes of acute type B aortic dissection in the current era: lessons from the International Registry of Aortic Dissection (IRAD). Circulation 2003;108(Suppl 1): Collins JS, Evangelista A, Nienaber CA, et al. Differences in clinical presentation, management, and outcomes of acute type a aortic dissection in patients with and without previous cardiac surgery. Circulation 2004;110(Suppl 1): Sinatra R, Melina G, Pulitani I, et al. Emergency operation for acute type A aortic dissection: neurologic complications and early mortality. Ann Thorac Surg 2001;71: Olsson C, Eriksson N, Stahle E, et al. Surgical and long-term mortality in 2634 consecutive patients operated on the proximal thoracic aorta. Eur J Cardiothorac Surg 2007;31: Campbell-Lloyd AJ, Mundy J, Pinto N, et al. Contemporary results following surgical repair of acute type a aortic dissection (AAAD): a single centre experience. Heart Lung Circ 2010;19: Apaydin AZ, Posacioglu H, Islamoglu F, et al. Analysis of perioperative risk factors in mortality and morbidity after modified Bentall operation. Jpn Heart J 2002;43:151 7.