Original Article
The outcome of surgical management of type A aortic dissection
Asian Cardiovascular & Thoracic Annals 2014, Vol. 22(6) 687–693 ß The Author(s) 2013 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0218492313506921 aan.sagepub.com
Edvard Skripochnik1, Patricia Friedman1, Robert E Michler1 and Siyamek Neragi-Miandoab2
Abstract Background: Type A aortic dissection is a serious and fatal condition. Methods: We retrospectively evaluated the outcome following repair of ascending aortic dissection in a contemporary cohort of 108 patients with a mean age of 59.5 13.8 years, who were treated between 2006 and 2011. Most patients were male (70, 65%). Results: Circulatory arrest with a mean duration of 22 16 min was performed in 42 (38.9%) patients. Perioperative mortality was 15.7% (n ¼ 17). Univariate analysis showed that cardiopulmonary bypass time (p ¼ 0.0006), age >60 years (p ¼ 0.028), cardiogenic shock at presentation (p ¼ 0.02), New York Heart Association class II–IV (p ¼ 0.038), hemopericardium (p ¼ 0.0035), and preoperative cerebrovascular accident (p ¼ 0.02) were predictors of mortality. Multivariate analysis indicated that age >60 years (odds ratio 7.7, 95% confidence interval: 1.52–38.96, p ¼ 0.0136), preoperative cerebrovascular accident (odds ratio 25.2, 95% confidence interval: 2.45–258.9, p ¼ 0.0066), hemopericardium (odds ratio 41.6, 95% confidence interval: 5.38–320.7, p ¼ 0.0003), and cardiopulmonary bypass time (odds ratio 1.85, 95% confidence interval: 1.32–2.57, p ¼ 0.0003) were independent predictors of perioperative mortality. The 1- and 4-year survival was 80% 3.8% and 69% 5.7%, respectively. Age >60 years (hazard ratio 3.3, 95% confidence interval: 1.4–7.9, p ¼ 0.0064) was the only independent predictor of long-term mortality. Conclusion: Our results identify the major risk factors for perioperative and long-term mortality. Age is an independent risk factor for mortality.
Keywords Aneurysm, dissecting, aortic aneurysm, thoracic, cardiopulmonary bypass, risk factors, survival rate, vascular surgical procedures
Introduction Surgical treatment of acute type A aortic dissection is still associated with a high incidence of adverse outcomes. While the outcome has improved significantly in recent years, patients who are hemodynamically unstable on presentation have high perioperative mortality.1 Early diagnosis, prior to development of cardiac tamponade and hemodynamic compromise, is critical to improve the operative outcome.2 Extended dissection into the aortic root or arch is reported to carry high risks of morbidity and mortality.3 However, in the last 2 decades, surgery of the aorta has undergone significant modification leading to improved survival and a reduced incidence of adverse events. Hypothermic brain perfusion through the right axillary
artery has resulted in a low incidence of neurologic complications.4 Our study was undertaken to identify some of the preoperative patient characteristics that may predict postoperative outcomes such as survival and adverse events. 1 Department of Cardiovascular and Thorscic Surgery, Montefiore Medical Center, Albert Einstein College of Medicine, New York, USA 2 Staten Island University Hospital, Department of Cardiothoracic Surgery, New York, USA
Corresponding author: Siyamek Neragi-Miandoab, MD, PhD, Staten Island University Hospital, Department of Cardiothoracic Surgery, SUNY Downstate, New York, USA. Email: [email protected]
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
688
Asian Cardiovascular & Thoracic Annals 22(6)
Patients and methods The study was approved by the institutional review board at Montefiore Medical Center. We retrospectively evaluated the data of 108 patients who underwent repair of aortic dissection between 2006 and 2011 at our institution. The diagnosis was obtained by computed tomography. An echocardiogram was performed at presentation, based on availability and the hemodynamic stability of the patient. Arterial cannulation was carried out through the axillary artery in all cases. Moderate hypothermia (22 C–26 C) was used in all patients who required partial or full circulatory arrest. The technical aspects of surgery have been well documented and are beyond the scope of this paper. Cerebral perfusion in patients with circulatory arrest was maintained via the axillary artery as well as an intraluminal catheter in the left carotid artery. The impact of pre- and perioperative condition on early and late outcome was examined. Hospital mortality was defined as death for any reason occurring within 30 days after the operation or anytime during the same hospitalization, regardless of length. Neurologic morbidities were defined as stroke or a new neurologic dysfunction that persisted for 72 h. Survival curves were generated using the KaplanMeier method and compared using a logrank test. Univariate analysis was used to determine predictors of morbidity and mortality. Multivariable logistic and Cox regression methods were used to identify independent risk factors. The 95% confidence intervals (95%CI) were calculated for each risk interval. A p value < 0.05 was considered statistically significant. Data analysis was performed using SPSS software (SPSS, Inc., Chicago, IL, USA).
Results The mean age was 59.5 13.8 years and 52 (48%) patients were older than 60 years. The location of the intimal tear was in the aortic root in 16 (14.8%) patients and the aortic arch in 9 (8.3%), requiring root or arch replacement. Preoperative risk factors and patient characteristics are listed in Table 1. Some additional risk factors included current cocaine use in 8 (7.4%) patients and serum creatinine >2 mgdL1 in 11 (10.2%). The New York Heart Association (NYHA) functional status included 48% in class I, 26% in class II, 13.5% in class III, and 12.5% in class IV (52% in classes II–IV). The mean diameter of ascending aorta was 46.1 9.3 mm; 24 (22.2%) patients had an aortic diameter >50 mm. Twenty-seven (25%) patients presented with pleural effusion, and 25 (23.2%) had moderate to severe hemopericardium with tamponade-like physiology on presentation. In addition to replacement of the ascending aorta, concurrent procedures included
Table 1. Preoperative characteristics of 108 patients who underwent repair of aortic dissection. Variable
No. of patients
Age (years) Male sex Body surface area (m2) Body mass index (kgm2) NYHA class III/IV Ejection fraction Asthma/COPD Atrial fibrillation Arrhythmia Coronary artery bypass Aortic repair Aortic valve replacement Root aneurysm Dialysis Cerebrovascular disease Hypercholesterolemia Angina Myocardial infarction Cerebrovascular accident Diabetes mellitus Cardiogenic shock Hypertension Congestive heart failure Peripheral vascular disease
59.52 13.86 70 (65%) 1.97 0.25 28.57 5.48 27 (25%) 0.54 0.11 13 (12%) 6 (6%) 13 (12%) 2 (2%) 11 (10%) 6 (6%) 16 (15%) 5 (5%) 8 (7%) 15 (14%) 7 (6%) 12 (11%) 9 (8.3%) 15 (14%) 5 (5%) 92 (85%) 23 (21%) 16 (15%)
COPD: chronic obstructive pulmonary disease; NYHA: New York Heart Association.
arch replacement in 9 (8.3%) patients, root replacement in 16 (14.2%), and coronary artery bypass grafting in 15 (13.9%). Thirty-eight (35.2%) patients had resuspension of the aortic valve without root replacement, 28 (25.9%) had hemiarch replacement, and 9 (8.3%) had a full arch replacement, of whom 3 (2.7%) had the first stage of the elephant trunk operation. Circulatory arrest was used in 42 (39%) patients for a mean of 22 16 min. Mean perfusion time was 227 90 min; 15 patients (13.9%) had a perfusion time >150 min. Mean crossclamp time was 123 59.6 min. Perioperative mortality was 15.7% (n ¼ 17). Three patients required cardiopulmonary resuscitation on admission and were categorized as salvage operations. Subsequently, these 3 patients had severe left heart failure and required a left ventricular assist device (TandemHeart); all 3 expired on postoperative day 1. The initial cannulation in patients with cardiogenic shock included emergency cannulation of the femoral artery and immediate cooling (22 C–26 C) followed by cannulation of axillary artery. The results of univariate analysis of
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
Skripochnik et al.
689
Table 2. Univariate analysis of risk factors for perioperative mortality. Significant
Nonsignificant
Variable
p value
Variable
p value
Perfusion time Perioperative lactate >4 mgdL1 Duration of ventilation Age >60 years Cardiogenic shock at presentation NYHA class II, III, IV vs. class I Hemopericardium Preoperative cerebrovascular accident
0.0006 0.0004 0.006 0.028 0.02 0.038 0.0035 0.02
Circulatory arrest Crossclamp time Body mass index Body surface area Units of transfusion Concomitant CABG Elevated creatinine Diabetes mellitus
0.73 0.19 0.97 0.57 0.30 0.24 0.2 1.0
CABG: coronary artery bypass grafting; NYHA: New York Heart Association.
perioperative mortality are given in Table 2. All risk factors with a p value < 0.05 were included in the multivariate analysis, which demonstrated that age >60 years (odds ratio [OR] 7.7, 95%CI: 1.52–38.960, p ¼ 0.0136), preoperative cerebrovascular accident (CVA; OR 25.2, 95%CI: 2.45–258.9, p ¼ 0.0066), hemopericardium (OR 41.6, 95%CI: 5.38–320.7, p ¼ 0.0003), and perfusion time (OR 1.85 for every 30 min longer than 150 min, 95%CI: 1.32–2.57, p ¼ 0.0003) were independent risk factors for perioperative mortality. Aortic root dissection and repair (p ¼ 0.7), aortic arch replacement (p ¼ 1.0), and the diameter of the aorta (p ¼ 0.78) were not statistically significant risk factors for perioperative mortality. The median postoperative ventilation time was 14.8 h (range 2.2–136.7 h), initial length of stay in the coronary care unit was 70 h (range 9–515 h) and total coronary care unit stay was 71 h (range 9–559.5 h). Prolonged ventilation was required in 22 (20.4%) patients and 2 (1.85%) developed pneumonia. Table 3 lists some additional postoperative complications. The Kaplan-Meier survival analysis showed overall 1-, 3-, and 4-year survival of 80% 3.8%, 74% 4.6%, 69% 5.7%, respectively (Figure 1a). Despite the high perioperative mortality, once patients survived the perioperative phase, they had a favorable survival rate. Preoperative CVA seemed to be a risk factor for survival; the 9 patients with preoperative CVA had inferior survival compared to those who did not have preoperative CVA: 50% 18% vs. 72% 6% at 4 years, respectively (Figure 1b). However, considering the small number of patients with preoperative CVA, the difference between these 2 groups remained nonsignificant (p ¼ 0.2). Extension of the dissection into the aortic root was treated with root replacement in 16 patients (Figure 1c), of whom 9 had a Bentall procedure, 2 had a bio-Bentall, and 5 had valve-sparing root replacement. The 1-year survival of patients with root
Table 3. Postoperative adverse events. Complication
No. of patients
Blood bank products Exploration for bleeding Permanent stroke Paraplegia Reintubation Prolonged ventilation Pneumonia Sepsis and multiorgan failure New-onset renal failure Readmission within 30 days Sternal wound infection
61 (57%) 8 (7.4%) 2/91 (2.2%) 0 7 (6.5%) 22 (20.4%) 2 (1.85%) 1 (0.93%) 1 (0.93%) 7 (15.7%) 1 (0.93%)
replacement (73% 12%) was lower compared to those who did not need root replacement (82% 4%), but the difference was not significant (p ¼ 0.5). The 9 patients with full arch replacement had inferior survival compared to those who did not require a full arch replacement (Figure 1d). The 1- and 4-year survival was 66% 20% and 33% 25% following arch replacement, compared to 1- and 4-year survival of 80% 4% and 72% 5% in patients who did not require arch replacement (p ¼ 0.5). Age was a major determinant of long-term survival: patients 60 years, p ¼ 0.0005 (Figure 2a). Peripheral vascular disease (PVD), as a sign of advanced vascular abnormality, was a predictor of poor outcome. Patients with PVD had worse long-term survival compared to those with no significant PVD: 54% 15% at 20 months and 36% 18% at 3 years vs. 83% 4% at 1 year and 73% 6% at 4 years, p ¼ 0.012 (Figure 2b). Advanced NYHA class was associated with poor
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
690
Asian Cardiovascular & Thoracic Annals 22(6)
Figure 1. (a) Overall survival of 108 patients who underwent repair of aortic dissection. (b) Patients with preoperative cerebrovascular accident (n ¼ 9) had inferior survival compared to those who did not have preoperative cerebrovascular accident. (c) The extension of dissection into the root and root replacement did not have a significant impact on survival. (d) Aortic arch replacement did not impact on survival.
long-term survival: patients in NYHA class I had a 4-year survival of 78% 8%, and those in classes II, III, and IV had a survival rate of 62% 8% at 4 years (p ¼ 0.02; Figure 2c). In the multivariate analysis, age >60 (hazard ratio 3.3, 95%CI 1.4–7.9, p ¼ 0.0064) was the only independent predictor of long-term postoperative mortality. Patients with serum creatinine >1.5 mgdL1 had a worse outcome after 1 year than those with creatinine 30 kgm2 (p ¼ 0.73), body surface area >2 m2 (p ¼ 0.58), cocaine abuse (p ¼ 0.29), and concomitant coronary artery bypass grafting (p ¼ 0.18) did not have a statistically significant impact on long-term survival in our series, which may be due to the small sample size.
Discussion The management of type A dissection has evolved in the last few years and the outcome has improved
significantly. In our series, the perioperative mortality was 15.7%, which is in line with current data in the literature. Encouragingly, patients who survived the perioperative period had a favorable long-term survival; 1- and 4-year survival was 80% 3.8% and 69% 5.7%, respectively, which also agrees with the current literature.2 The impact of age on survival has been documented by others,5–8 and also supported by our study. We found that age was the only independent risk factor for perioperative and long-term mortality. We dichotomized our group at 60 years of age to have an equal distribution, but others have used 70 or 80 years of age as cutoffs and found higher mortality rates in their patients.9 Although age has been shown to be a risk factor for mortality,5–8 studies have not shown it to be significant for neurologic adverse events following ascending aortic and arch repair.5 Patel and colleagues8 found in a large series of 721 patients, that arch reconstruction with hypothermic circulatory arrest, advanced age, prolonged bypass time, and advanced heart failure were independent predictors of early mortality. Thirty-day morbidity included death 5%, stroke 4.7%, and permanent dialysis 1.9%. Actuarial survival at 10 years was 65%, and prolonged lower-body circulatory arrest time and increasing serum creatinine were significant risk factors for
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
Skripochnik et al.
691
Figure 2. (a) Age was a major predictor of long-term survival. (b) Patients with peripheral vascular disease had worse survival compared to patients who did not have peripheral vascular disease. (c) New York Heart Association class as an indicator of myocardial function was a significant predictor of long-term survival.
long-term mortality.8 We believe that prolonged circulatory arrest may have an impact on perioperative mortality, but its impact on long-term survival may be a bias, as a reflection of perioperative mortality. Further, in Patel’s series, cardiopulmonary bypass time was a significant risk factor for perioperative mortality, in accordance with the literature.8 Although consistently documented, it should be interpreted cautiously because various factors may have an effect on prolonging bypass, including the extent of the patient’s pathology. On the other hand, prolonged bypass time itself is a risk factor for perioperative mortality, independent of patient pathology. We found hemopericardium at the time of presentation to be an independent risk factor for perioperative mortality. It serves as an indication of extravasation and pending rupture of the aorta causing pericardial effusion and increased risk of tamponade. In our series, dissection involved the aortic root in 16 (14.8%) patients and the aortic arch in 9 (8.3%); however, replacement of the aortic arch and root was not associated with increased morbidity and mortality. Although these results are in agreement with the current literature,10–12 we cannot make a statement in this regard due to our small sample size. Some other studies have reported higher morbidity and mortality for arch replacement compared to ascending replacement only.3,13
Hypothermic brain perfusion through the right axillary artery has been successful in reducing the incidence of neurologic complications.4,6 In our series, 2 patients suffered permanent stroke, both had undergone surgery under circulatory arrest. One of these patients had replacement of the aortic root using a biologic valve (bio-Bentall). Considering the small number of patients who had a stroke, we cannot conclude a link between circulatory arrest, extension of repair, and postoperative stroke. The incidence of postoperative neurologic adverse events was low in our series (1.86%) and we did not observe any paraplegia, but preoperative cerebrovascular disease was associated with higher perioperative mortality. Mild hypothermic (28 C) circulatory arrest without any cerebral perfusion, followed by rapid rewarming, was reported by Hata and colleagues14 in a series of 91 patients with acute aortic dissection. With a short circulatory arrest time (19 5 min) and cardiopulmonary bypass time (86.2 17.8 min), they reported a stroke rate of 4.4%, and hospital mortality of 3.3%.14 In a series of 119 patients, Matsuyama and colleagues15 reported a low mortality rate of 3.4% (1.6% for elective cases) and a favorable 4-year survival of 86.5%; incidences of permanent neurologic deficit (5%) and paraplegia (1.7%) were acceptable. Wiedemann and colleagues6 reported high overall mortality (19%) and neurologic adverse events (26%) in a series of
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
692
Asian Cardiovascular & Thoracic Annals 22(6)
329 patients; the lowest mortality (13%) and lowest neurologic adverse event rate (12%) were observed with antegrade cerebral perfusion. Considering the time period of the study, extending over decades, and a nonhomogeneous patient population, Wiedemann and colleagues6 did not identify any independent predictors of adverse neurologic outcome. We had a small number of patients who underwent a redo sternotomy and perhaps because of the small sample size, it was not a risk factor for mortality; others have reported no increased risk of mortality in reoperations for aortic dissection.16,17 Klodell and colleagues16 found in a series of 190 patients with acute type A aortic dissection (159 first operations and 31 redo operations) an overall mortality of 16.8%, and after propensity matching of both groups, there was no difference in mortality or postoperative complications. Our patients with PVD had inferior long-term survival compared to those without PVD (p ¼ 0.012). Stevens and colleagues7 also found that PVD was an independent factor for mortality. Postoperative morbidities included cardiac arrest, prolonged ventilation, renal failure, exploration for bleeding, readmission to the intensive care unit, and reintubation. Two patients in the circulatory arrest group (2/42), suffered a permanent stroke. These results compare favorably with the current literature.2,8,18 Our results identified the major risk factors for perioperative and long-term mortality. Circulatory arrest and replacement of the aortic arch and root did not increase the risk of mortality. Pericardial effusion at the time of surgery is a major predictor of mortality. Once patients survive the perioperative period, they have a favorable long-term survival. Funding This research received no specific grant from any funding agency in the public, commerical, or not-for-profit sectors.
Conflict of interest statement None of the authors report any kind of conflict of interest.
References 1. Schwartz JP, Bakhos M, Patel A, Botkin S and NeragiMiandoab S. Impact of pre-existing conditions, age and the length of cardiopulmonary bypass on postoperative outcome after repair of the ascending aorta and aortic arch for aortic aneurysms and dissections. Interact Cardiovasc Thorac Surg 2008; 7: 850–854. 2. Ehrlich MP, Schillinger M, Grabenwoger M, et al. Predictors of adverse outcome and transient neurological dysfunction following surgical treatment of acute type A dissections. Circulation 2003; 108(Suppl 1): 1318–1323.
3. Pochettino A, Brinkman WT, Moeller P, et al. Antegrade thoracic stent grafting during repair of acute DeBakey I dissection prevents development of thoracoabdominal aortic aneurysms. Ann Thorac Surg 2009; 88: 482–489. 4. Kulik A, Castner CF and Kouchoukos NT. Outcomes after total aortic arch replacement with right axillary artery cannulation and a presewn multibranched graft. Ann Thorac Surg 2011; 92: 889–897. 5. Czerny M, Fleck T, Zimpfer D, et al. Risk factors of mortality and permanent neurologic injury in patients undergoing ascending aortic and arch repair. J Thorac Cardiovasc Surg 2003; 126: 1296–1301. 6. Wiedemann D, Kocher A, Dorfmeister M, et al. Effect of cerebral protection strategy on outcome of patients with Stanford type A aortic dissection. J Thorac Cardiovasc Surg 2013; 146: 647–655. 7. Stevens LM, Madsen JC, Isselbacher EM, et al. Surgical management and long-term outcomes for acute ascending aortic dissection. J Thorac Cardiovasc Surg 2009; 138: 1349–1357. 8. Patel HJ, Nguyen C, Diener AC, Passow MC, Salata D and Deeb GM. Open arch reconstruction in the endovascular era: analysis of 721 patients over 17 years. J Thorac Cardiovasc Surg 2011; 141: 1417–1423. 9. Vanhuyse F, Maureira P, Laurent N, Lekehal M, Grandmougain D and Villemot JP. Surgery for acute type A aortic dissection in octogenarians. J Card Surg 2012; 27: 65–69. 10. Okita Y, Okada K, Omura A, et al. Total arch replacement using antegrade cerebral perfusion. J Thorac Cardiovasc Surg 2013; 145: S63–S71. 11. Montalvo J, Razzouk A, Wang N, et al. Aortic root surgery does not increase the operative risk of acute type A aortic dissection. Am Surg 2011; 77: 88–92. 12. Yamanaka K, Hori Y, Ikarashi J, et al. Durability of aortic valve preservation with root reconstruction for acute type A aortic dissection. Eur J Cardiothorac Surg 2012; 41: e32–e36. 13. Kim JB, Chung CH, Moon DH, et al. Total arch repair versus hemiarch repair in the management of acute DeBakey type I aortic dissection. Eur J Cardiothorac Surg 2011; 40: 881–887. 14. Hata M, Akiyama K, Hata H, et al. Early and midterm outcomes of quick proximal arch replacement with mild hypothermia and rapid rewarming for type A acute aortic dissection. J Thorac Cardiovasc Surg 2013; 146: 119–123. 15. Matsuyama S, Tabata M, Shimokawa T, Matsushita A, Fukui T and Takanashi S. Outcomes of total arch replacement with stepwise distal anastomosis technique and modified perfusion strategy. J Thorac Cardiovasc Surg 2012; 143: 1377–1381. 16. Klodell CT, Karimi A, Beaver TM, Hess PJ and Martin TD. Outcomes for acute type A aortic dissection: effects of previous cardiac surgery. Ann Thorac Surg 2012; 93: 1206–1212.
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
Skripochnik et al.
693
17. Luciani GB, Casali G, Faggian G and Mazzucco A. Predicting outcome after reoperative procedures on the aortic root and ascending aorta. Eur J Cardiothorac Surg 2000; 17: 602–607. 18. Schwartz JP, Bakhos M, Patel A, Botkin S and NeragiMiandoab S. Repair of aortic arch and the impact of
cross-clamping time, New York Heart Association stage, circulatory arrest time, and age on operative outcome. Interact Cardiovasc Thorac Surg 2008; 7: 425–429.
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
The outcome of surgical management of type A aortic dissection
Asian Cardiovascular & Thoracic Annals 2014, Vol. 22(6) 687–693 ß The Author(s) 2013 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0218492313506921 aan.sagepub.com
Edvard Skripochnik1, Patricia Friedman1, Robert E Michler1 and Siyamek Neragi-Miandoab2
Abstract Background: Type A aortic dissection is a serious and fatal condition. Methods: We retrospectively evaluated the outcome following repair of ascending aortic dissection in a contemporary cohort of 108 patients with a mean age of 59.5 13.8 years, who were treated between 2006 and 2011. Most patients were male (70, 65%). Results: Circulatory arrest with a mean duration of 22 16 min was performed in 42 (38.9%) patients. Perioperative mortality was 15.7% (n ¼ 17). Univariate analysis showed that cardiopulmonary bypass time (p ¼ 0.0006), age >60 years (p ¼ 0.028), cardiogenic shock at presentation (p ¼ 0.02), New York Heart Association class II–IV (p ¼ 0.038), hemopericardium (p ¼ 0.0035), and preoperative cerebrovascular accident (p ¼ 0.02) were predictors of mortality. Multivariate analysis indicated that age >60 years (odds ratio 7.7, 95% confidence interval: 1.52–38.96, p ¼ 0.0136), preoperative cerebrovascular accident (odds ratio 25.2, 95% confidence interval: 2.45–258.9, p ¼ 0.0066), hemopericardium (odds ratio 41.6, 95% confidence interval: 5.38–320.7, p ¼ 0.0003), and cardiopulmonary bypass time (odds ratio 1.85, 95% confidence interval: 1.32–2.57, p ¼ 0.0003) were independent predictors of perioperative mortality. The 1- and 4-year survival was 80% 3.8% and 69% 5.7%, respectively. Age >60 years (hazard ratio 3.3, 95% confidence interval: 1.4–7.9, p ¼ 0.0064) was the only independent predictor of long-term mortality. Conclusion: Our results identify the major risk factors for perioperative and long-term mortality. Age is an independent risk factor for mortality.
Keywords Aneurysm, dissecting, aortic aneurysm, thoracic, cardiopulmonary bypass, risk factors, survival rate, vascular surgical procedures
Introduction Surgical treatment of acute type A aortic dissection is still associated with a high incidence of adverse outcomes. While the outcome has improved significantly in recent years, patients who are hemodynamically unstable on presentation have high perioperative mortality.1 Early diagnosis, prior to development of cardiac tamponade and hemodynamic compromise, is critical to improve the operative outcome.2 Extended dissection into the aortic root or arch is reported to carry high risks of morbidity and mortality.3 However, in the last 2 decades, surgery of the aorta has undergone significant modification leading to improved survival and a reduced incidence of adverse events. Hypothermic brain perfusion through the right axillary
artery has resulted in a low incidence of neurologic complications.4 Our study was undertaken to identify some of the preoperative patient characteristics that may predict postoperative outcomes such as survival and adverse events. 1 Department of Cardiovascular and Thorscic Surgery, Montefiore Medical Center, Albert Einstein College of Medicine, New York, USA 2 Staten Island University Hospital, Department of Cardiothoracic Surgery, New York, USA
Corresponding author: Siyamek Neragi-Miandoab, MD, PhD, Staten Island University Hospital, Department of Cardiothoracic Surgery, SUNY Downstate, New York, USA. Email: [email protected]
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
688
Asian Cardiovascular & Thoracic Annals 22(6)
Patients and methods The study was approved by the institutional review board at Montefiore Medical Center. We retrospectively evaluated the data of 108 patients who underwent repair of aortic dissection between 2006 and 2011 at our institution. The diagnosis was obtained by computed tomography. An echocardiogram was performed at presentation, based on availability and the hemodynamic stability of the patient. Arterial cannulation was carried out through the axillary artery in all cases. Moderate hypothermia (22 C–26 C) was used in all patients who required partial or full circulatory arrest. The technical aspects of surgery have been well documented and are beyond the scope of this paper. Cerebral perfusion in patients with circulatory arrest was maintained via the axillary artery as well as an intraluminal catheter in the left carotid artery. The impact of pre- and perioperative condition on early and late outcome was examined. Hospital mortality was defined as death for any reason occurring within 30 days after the operation or anytime during the same hospitalization, regardless of length. Neurologic morbidities were defined as stroke or a new neurologic dysfunction that persisted for 72 h. Survival curves were generated using the KaplanMeier method and compared using a logrank test. Univariate analysis was used to determine predictors of morbidity and mortality. Multivariable logistic and Cox regression methods were used to identify independent risk factors. The 95% confidence intervals (95%CI) were calculated for each risk interval. A p value < 0.05 was considered statistically significant. Data analysis was performed using SPSS software (SPSS, Inc., Chicago, IL, USA).
Results The mean age was 59.5 13.8 years and 52 (48%) patients were older than 60 years. The location of the intimal tear was in the aortic root in 16 (14.8%) patients and the aortic arch in 9 (8.3%), requiring root or arch replacement. Preoperative risk factors and patient characteristics are listed in Table 1. Some additional risk factors included current cocaine use in 8 (7.4%) patients and serum creatinine >2 mgdL1 in 11 (10.2%). The New York Heart Association (NYHA) functional status included 48% in class I, 26% in class II, 13.5% in class III, and 12.5% in class IV (52% in classes II–IV). The mean diameter of ascending aorta was 46.1 9.3 mm; 24 (22.2%) patients had an aortic diameter >50 mm. Twenty-seven (25%) patients presented with pleural effusion, and 25 (23.2%) had moderate to severe hemopericardium with tamponade-like physiology on presentation. In addition to replacement of the ascending aorta, concurrent procedures included
Table 1. Preoperative characteristics of 108 patients who underwent repair of aortic dissection. Variable
No. of patients
Age (years) Male sex Body surface area (m2) Body mass index (kgm2) NYHA class III/IV Ejection fraction Asthma/COPD Atrial fibrillation Arrhythmia Coronary artery bypass Aortic repair Aortic valve replacement Root aneurysm Dialysis Cerebrovascular disease Hypercholesterolemia Angina Myocardial infarction Cerebrovascular accident Diabetes mellitus Cardiogenic shock Hypertension Congestive heart failure Peripheral vascular disease
59.52 13.86 70 (65%) 1.97 0.25 28.57 5.48 27 (25%) 0.54 0.11 13 (12%) 6 (6%) 13 (12%) 2 (2%) 11 (10%) 6 (6%) 16 (15%) 5 (5%) 8 (7%) 15 (14%) 7 (6%) 12 (11%) 9 (8.3%) 15 (14%) 5 (5%) 92 (85%) 23 (21%) 16 (15%)
COPD: chronic obstructive pulmonary disease; NYHA: New York Heart Association.
arch replacement in 9 (8.3%) patients, root replacement in 16 (14.2%), and coronary artery bypass grafting in 15 (13.9%). Thirty-eight (35.2%) patients had resuspension of the aortic valve without root replacement, 28 (25.9%) had hemiarch replacement, and 9 (8.3%) had a full arch replacement, of whom 3 (2.7%) had the first stage of the elephant trunk operation. Circulatory arrest was used in 42 (39%) patients for a mean of 22 16 min. Mean perfusion time was 227 90 min; 15 patients (13.9%) had a perfusion time >150 min. Mean crossclamp time was 123 59.6 min. Perioperative mortality was 15.7% (n ¼ 17). Three patients required cardiopulmonary resuscitation on admission and were categorized as salvage operations. Subsequently, these 3 patients had severe left heart failure and required a left ventricular assist device (TandemHeart); all 3 expired on postoperative day 1. The initial cannulation in patients with cardiogenic shock included emergency cannulation of the femoral artery and immediate cooling (22 C–26 C) followed by cannulation of axillary artery. The results of univariate analysis of
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
Skripochnik et al.
689
Table 2. Univariate analysis of risk factors for perioperative mortality. Significant
Nonsignificant
Variable
p value
Variable
p value
Perfusion time Perioperative lactate >4 mgdL1 Duration of ventilation Age >60 years Cardiogenic shock at presentation NYHA class II, III, IV vs. class I Hemopericardium Preoperative cerebrovascular accident
0.0006 0.0004 0.006 0.028 0.02 0.038 0.0035 0.02
Circulatory arrest Crossclamp time Body mass index Body surface area Units of transfusion Concomitant CABG Elevated creatinine Diabetes mellitus
0.73 0.19 0.97 0.57 0.30 0.24 0.2 1.0
CABG: coronary artery bypass grafting; NYHA: New York Heart Association.
perioperative mortality are given in Table 2. All risk factors with a p value < 0.05 were included in the multivariate analysis, which demonstrated that age >60 years (odds ratio [OR] 7.7, 95%CI: 1.52–38.960, p ¼ 0.0136), preoperative cerebrovascular accident (CVA; OR 25.2, 95%CI: 2.45–258.9, p ¼ 0.0066), hemopericardium (OR 41.6, 95%CI: 5.38–320.7, p ¼ 0.0003), and perfusion time (OR 1.85 for every 30 min longer than 150 min, 95%CI: 1.32–2.57, p ¼ 0.0003) were independent risk factors for perioperative mortality. Aortic root dissection and repair (p ¼ 0.7), aortic arch replacement (p ¼ 1.0), and the diameter of the aorta (p ¼ 0.78) were not statistically significant risk factors for perioperative mortality. The median postoperative ventilation time was 14.8 h (range 2.2–136.7 h), initial length of stay in the coronary care unit was 70 h (range 9–515 h) and total coronary care unit stay was 71 h (range 9–559.5 h). Prolonged ventilation was required in 22 (20.4%) patients and 2 (1.85%) developed pneumonia. Table 3 lists some additional postoperative complications. The Kaplan-Meier survival analysis showed overall 1-, 3-, and 4-year survival of 80% 3.8%, 74% 4.6%, 69% 5.7%, respectively (Figure 1a). Despite the high perioperative mortality, once patients survived the perioperative phase, they had a favorable survival rate. Preoperative CVA seemed to be a risk factor for survival; the 9 patients with preoperative CVA had inferior survival compared to those who did not have preoperative CVA: 50% 18% vs. 72% 6% at 4 years, respectively (Figure 1b). However, considering the small number of patients with preoperative CVA, the difference between these 2 groups remained nonsignificant (p ¼ 0.2). Extension of the dissection into the aortic root was treated with root replacement in 16 patients (Figure 1c), of whom 9 had a Bentall procedure, 2 had a bio-Bentall, and 5 had valve-sparing root replacement. The 1-year survival of patients with root
Table 3. Postoperative adverse events. Complication
No. of patients
Blood bank products Exploration for bleeding Permanent stroke Paraplegia Reintubation Prolonged ventilation Pneumonia Sepsis and multiorgan failure New-onset renal failure Readmission within 30 days Sternal wound infection
61 (57%) 8 (7.4%) 2/91 (2.2%) 0 7 (6.5%) 22 (20.4%) 2 (1.85%) 1 (0.93%) 1 (0.93%) 7 (15.7%) 1 (0.93%)
replacement (73% 12%) was lower compared to those who did not need root replacement (82% 4%), but the difference was not significant (p ¼ 0.5). The 9 patients with full arch replacement had inferior survival compared to those who did not require a full arch replacement (Figure 1d). The 1- and 4-year survival was 66% 20% and 33% 25% following arch replacement, compared to 1- and 4-year survival of 80% 4% and 72% 5% in patients who did not require arch replacement (p ¼ 0.5). Age was a major determinant of long-term survival: patients 60 years, p ¼ 0.0005 (Figure 2a). Peripheral vascular disease (PVD), as a sign of advanced vascular abnormality, was a predictor of poor outcome. Patients with PVD had worse long-term survival compared to those with no significant PVD: 54% 15% at 20 months and 36% 18% at 3 years vs. 83% 4% at 1 year and 73% 6% at 4 years, p ¼ 0.012 (Figure 2b). Advanced NYHA class was associated with poor
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
690
Asian Cardiovascular & Thoracic Annals 22(6)
Figure 1. (a) Overall survival of 108 patients who underwent repair of aortic dissection. (b) Patients with preoperative cerebrovascular accident (n ¼ 9) had inferior survival compared to those who did not have preoperative cerebrovascular accident. (c) The extension of dissection into the root and root replacement did not have a significant impact on survival. (d) Aortic arch replacement did not impact on survival.
long-term survival: patients in NYHA class I had a 4-year survival of 78% 8%, and those in classes II, III, and IV had a survival rate of 62% 8% at 4 years (p ¼ 0.02; Figure 2c). In the multivariate analysis, age >60 (hazard ratio 3.3, 95%CI 1.4–7.9, p ¼ 0.0064) was the only independent predictor of long-term postoperative mortality. Patients with serum creatinine >1.5 mgdL1 had a worse outcome after 1 year than those with creatinine 30 kgm2 (p ¼ 0.73), body surface area >2 m2 (p ¼ 0.58), cocaine abuse (p ¼ 0.29), and concomitant coronary artery bypass grafting (p ¼ 0.18) did not have a statistically significant impact on long-term survival in our series, which may be due to the small sample size.
Discussion The management of type A dissection has evolved in the last few years and the outcome has improved
significantly. In our series, the perioperative mortality was 15.7%, which is in line with current data in the literature. Encouragingly, patients who survived the perioperative period had a favorable long-term survival; 1- and 4-year survival was 80% 3.8% and 69% 5.7%, respectively, which also agrees with the current literature.2 The impact of age on survival has been documented by others,5–8 and also supported by our study. We found that age was the only independent risk factor for perioperative and long-term mortality. We dichotomized our group at 60 years of age to have an equal distribution, but others have used 70 or 80 years of age as cutoffs and found higher mortality rates in their patients.9 Although age has been shown to be a risk factor for mortality,5–8 studies have not shown it to be significant for neurologic adverse events following ascending aortic and arch repair.5 Patel and colleagues8 found in a large series of 721 patients, that arch reconstruction with hypothermic circulatory arrest, advanced age, prolonged bypass time, and advanced heart failure were independent predictors of early mortality. Thirty-day morbidity included death 5%, stroke 4.7%, and permanent dialysis 1.9%. Actuarial survival at 10 years was 65%, and prolonged lower-body circulatory arrest time and increasing serum creatinine were significant risk factors for
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
Skripochnik et al.
691
Figure 2. (a) Age was a major predictor of long-term survival. (b) Patients with peripheral vascular disease had worse survival compared to patients who did not have peripheral vascular disease. (c) New York Heart Association class as an indicator of myocardial function was a significant predictor of long-term survival.
long-term mortality.8 We believe that prolonged circulatory arrest may have an impact on perioperative mortality, but its impact on long-term survival may be a bias, as a reflection of perioperative mortality. Further, in Patel’s series, cardiopulmonary bypass time was a significant risk factor for perioperative mortality, in accordance with the literature.8 Although consistently documented, it should be interpreted cautiously because various factors may have an effect on prolonging bypass, including the extent of the patient’s pathology. On the other hand, prolonged bypass time itself is a risk factor for perioperative mortality, independent of patient pathology. We found hemopericardium at the time of presentation to be an independent risk factor for perioperative mortality. It serves as an indication of extravasation and pending rupture of the aorta causing pericardial effusion and increased risk of tamponade. In our series, dissection involved the aortic root in 16 (14.8%) patients and the aortic arch in 9 (8.3%); however, replacement of the aortic arch and root was not associated with increased morbidity and mortality. Although these results are in agreement with the current literature,10–12 we cannot make a statement in this regard due to our small sample size. Some other studies have reported higher morbidity and mortality for arch replacement compared to ascending replacement only.3,13
Hypothermic brain perfusion through the right axillary artery has been successful in reducing the incidence of neurologic complications.4,6 In our series, 2 patients suffered permanent stroke, both had undergone surgery under circulatory arrest. One of these patients had replacement of the aortic root using a biologic valve (bio-Bentall). Considering the small number of patients who had a stroke, we cannot conclude a link between circulatory arrest, extension of repair, and postoperative stroke. The incidence of postoperative neurologic adverse events was low in our series (1.86%) and we did not observe any paraplegia, but preoperative cerebrovascular disease was associated with higher perioperative mortality. Mild hypothermic (28 C) circulatory arrest without any cerebral perfusion, followed by rapid rewarming, was reported by Hata and colleagues14 in a series of 91 patients with acute aortic dissection. With a short circulatory arrest time (19 5 min) and cardiopulmonary bypass time (86.2 17.8 min), they reported a stroke rate of 4.4%, and hospital mortality of 3.3%.14 In a series of 119 patients, Matsuyama and colleagues15 reported a low mortality rate of 3.4% (1.6% for elective cases) and a favorable 4-year survival of 86.5%; incidences of permanent neurologic deficit (5%) and paraplegia (1.7%) were acceptable. Wiedemann and colleagues6 reported high overall mortality (19%) and neurologic adverse events (26%) in a series of
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
692
Asian Cardiovascular & Thoracic Annals 22(6)
329 patients; the lowest mortality (13%) and lowest neurologic adverse event rate (12%) were observed with antegrade cerebral perfusion. Considering the time period of the study, extending over decades, and a nonhomogeneous patient population, Wiedemann and colleagues6 did not identify any independent predictors of adverse neurologic outcome. We had a small number of patients who underwent a redo sternotomy and perhaps because of the small sample size, it was not a risk factor for mortality; others have reported no increased risk of mortality in reoperations for aortic dissection.16,17 Klodell and colleagues16 found in a series of 190 patients with acute type A aortic dissection (159 first operations and 31 redo operations) an overall mortality of 16.8%, and after propensity matching of both groups, there was no difference in mortality or postoperative complications. Our patients with PVD had inferior long-term survival compared to those without PVD (p ¼ 0.012). Stevens and colleagues7 also found that PVD was an independent factor for mortality. Postoperative morbidities included cardiac arrest, prolonged ventilation, renal failure, exploration for bleeding, readmission to the intensive care unit, and reintubation. Two patients in the circulatory arrest group (2/42), suffered a permanent stroke. These results compare favorably with the current literature.2,8,18 Our results identified the major risk factors for perioperative and long-term mortality. Circulatory arrest and replacement of the aortic arch and root did not increase the risk of mortality. Pericardial effusion at the time of surgery is a major predictor of mortality. Once patients survive the perioperative period, they have a favorable long-term survival. Funding This research received no specific grant from any funding agency in the public, commerical, or not-for-profit sectors.
Conflict of interest statement None of the authors report any kind of conflict of interest.
References 1. Schwartz JP, Bakhos M, Patel A, Botkin S and NeragiMiandoab S. Impact of pre-existing conditions, age and the length of cardiopulmonary bypass on postoperative outcome after repair of the ascending aorta and aortic arch for aortic aneurysms and dissections. Interact Cardiovasc Thorac Surg 2008; 7: 850–854. 2. Ehrlich MP, Schillinger M, Grabenwoger M, et al. Predictors of adverse outcome and transient neurological dysfunction following surgical treatment of acute type A dissections. Circulation 2003; 108(Suppl 1): 1318–1323.
3. Pochettino A, Brinkman WT, Moeller P, et al. Antegrade thoracic stent grafting during repair of acute DeBakey I dissection prevents development of thoracoabdominal aortic aneurysms. Ann Thorac Surg 2009; 88: 482–489. 4. Kulik A, Castner CF and Kouchoukos NT. Outcomes after total aortic arch replacement with right axillary artery cannulation and a presewn multibranched graft. Ann Thorac Surg 2011; 92: 889–897. 5. Czerny M, Fleck T, Zimpfer D, et al. Risk factors of mortality and permanent neurologic injury in patients undergoing ascending aortic and arch repair. J Thorac Cardiovasc Surg 2003; 126: 1296–1301. 6. Wiedemann D, Kocher A, Dorfmeister M, et al. Effect of cerebral protection strategy on outcome of patients with Stanford type A aortic dissection. J Thorac Cardiovasc Surg 2013; 146: 647–655. 7. Stevens LM, Madsen JC, Isselbacher EM, et al. Surgical management and long-term outcomes for acute ascending aortic dissection. J Thorac Cardiovasc Surg 2009; 138: 1349–1357. 8. Patel HJ, Nguyen C, Diener AC, Passow MC, Salata D and Deeb GM. Open arch reconstruction in the endovascular era: analysis of 721 patients over 17 years. J Thorac Cardiovasc Surg 2011; 141: 1417–1423. 9. Vanhuyse F, Maureira P, Laurent N, Lekehal M, Grandmougain D and Villemot JP. Surgery for acute type A aortic dissection in octogenarians. J Card Surg 2012; 27: 65–69. 10. Okita Y, Okada K, Omura A, et al. Total arch replacement using antegrade cerebral perfusion. J Thorac Cardiovasc Surg 2013; 145: S63–S71. 11. Montalvo J, Razzouk A, Wang N, et al. Aortic root surgery does not increase the operative risk of acute type A aortic dissection. Am Surg 2011; 77: 88–92. 12. Yamanaka K, Hori Y, Ikarashi J, et al. Durability of aortic valve preservation with root reconstruction for acute type A aortic dissection. Eur J Cardiothorac Surg 2012; 41: e32–e36. 13. Kim JB, Chung CH, Moon DH, et al. Total arch repair versus hemiarch repair in the management of acute DeBakey type I aortic dissection. Eur J Cardiothorac Surg 2011; 40: 881–887. 14. Hata M, Akiyama K, Hata H, et al. Early and midterm outcomes of quick proximal arch replacement with mild hypothermia and rapid rewarming for type A acute aortic dissection. J Thorac Cardiovasc Surg 2013; 146: 119–123. 15. Matsuyama S, Tabata M, Shimokawa T, Matsushita A, Fukui T and Takanashi S. Outcomes of total arch replacement with stepwise distal anastomosis technique and modified perfusion strategy. J Thorac Cardiovasc Surg 2012; 143: 1377–1381. 16. Klodell CT, Karimi A, Beaver TM, Hess PJ and Martin TD. Outcomes for acute type A aortic dissection: effects of previous cardiac surgery. Ann Thorac Surg 2012; 93: 1206–1212.
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
Skripochnik et al.
693
17. Luciani GB, Casali G, Faggian G and Mazzucco A. Predicting outcome after reoperative procedures on the aortic root and ascending aorta. Eur J Cardiothorac Surg 2000; 17: 602–607. 18. Schwartz JP, Bakhos M, Patel A, Botkin S and NeragiMiandoab S. Repair of aortic arch and the impact of
cross-clamping time, New York Heart Association stage, circulatory arrest time, and age on operative outcome. Interact Cardiovasc Thorac Surg 2008; 7: 425–429.
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 5, 2015
