Marc Hendrikx, MD, PhD, Dept. of Cardiothoracic Surgery, Jessa Hospital, Campus Virga Jesse, Stadsomvaart 11, B-3500 Hasselt, Belgium. E-mail: marc.hendrikx@jessazh.be Received 21 November 2012; revision accepted for publication 25 June 2013. INTRODUCTION The prevalence of thoracic aortic aneurysms has increased over the last 30 years. Actually, the incidence Elective reconstruction of the ascending aorta for aneurysmal disease restores normal life expectancy. An analysis of risk factors for early and late mortality Dries VAN DUFFEL 1* , BSc; Ruben VAN GEMERT 1* , BSc; Pascal STARINIERI 2 MSc; Jean-Louis PAUWELS 2 , MSc; Agnes NATUKUNDA 3 , BSc; Trias Wahyuni RAKHMAWATI 3 , BSc; Maxwell Tawanda CHIREHWA 3 , BSc; James ORWA 3 , BSc; Herbert THYS 3 , PhD; Patrick DEBOOSERE 4 , PhD; Boris ROBIC 1,2 , MD; Urbain MEES 2 , MD; Marc HENDRIKX 1,2 , MD, PhD 1 Faculty of Medicine and Life Sciences, Hasselt University, Hasselt, Belgium; 2 Dept. of Cardiothoracic Surgery, Jessa Hospital, Hasselt, Belgium; 3 CENSTAT, Hasselt University, Hasselt, Belgium; 4 Faculty of Sociology, Vrije Universiteit Brussels, Brussels, Belgium. *Tese authors contributed equally to the paper Objective We investigated the survival of patients who had undergone elective reconstruction of the ascending aorta for degenerative aneurysms. The long-term survival was compared to an age- and sex-matched case-control population. An analysis of risk factors, inuencing survival was made. Methods and results From May 1998 to January 2012, 72 patients underwent elective reconstruction of the ascending aorta for degen- erative disease at the department of Cardiothoracic Surgery of the Jessa Hospital, Hasselt, Belgium. Sixty patients were treated with Bentall procedures, whereas 12 received valve-sparing procedures. The average age of the patient group was 65.5 years (range 24-80), with 64% males. Thirty-day mortality was 9.7% (consistent with calculated Euroscore II: 9.2%). The long-term survival was 80.9% at 3, 5 and 10 years. No deaths occurred between 3 and 10 years postoperatively. In an age- and sex case-matched Belgian population, 3-, 5- and 10-year survival were 95.7%, 94.7% and 85.2%, respectively. Long-term survival was not signicantly dierent between both groups. Poor NYHA class at the time of surgery (P = 0.041) and COPD (P = 0.028) had a signicant impact on global survival. Valve-sparing operations provide similar long-term survival, avoiding thrombo-embolic complications. Conclusions Reconstruction of the ascending aorta for degenerative aneurysmal disease restores normal life expectancy, compared with an age- and sex-matched case-control population. Early mortality is consistent with the Euroscore II risk calculation. Whereas late survival progressively declines in the average population, it remains constant in the treated group after 3 years. COPD and poor functional class signicantly impair survival. Valve-sparing procedures confer a similar long-term survival as valve replacement. Keywords: Aorta ascending aneurysm survival risk factor. in individuals over 65 years is estimated around 3-4%. The condition negatively impacts on the survival of patients, mainly due to the occurrence of two dreaded complications: dissection or rupture, both occurring at an incidence of approximately 3.5 in 100,000. Elective reconstruction of the ascending aorta by either the clas- sical Bentall operation or more recently by valve-sparing procedures, such as the Yacoub or David procedures, aims at preventing those complications. Whether this surgery restores life expectancy to normal, remains a matter of debate. This paper describes our clinical expe- rience with these procedures and the long-term out- come. Acta Cardiol 2013; 68(4): 349-353 doi: 10.2143/AC.68.4.2988887 [ Original article ] D. Van Duel et al. 350 Twelve patients underwent valve-sparing procedures: 7 reimplantation or David procedures and 5 remodelling or Yacoub procedures. Duration of extracorporeal circulation was 172 73 min for a clamp time of 113 43 min. In 5 patients, circulatory arrest had to be carried out in order to reconstruct the aortic arch during the same procedure. This was carried out under deep hypothermic circulatory arrest (equaliza- tion of all temperatures at 18C) combined with selective antegrade cerebral perfusion. The average duration of circulatory arrest was 24 8 min. STATISTICAL ANALYSIS A Kaplan-Meier curve was constructed to calculate survival in the treatment group. The common closing date method was used and the survival status of each patient was assessed on August 31, 2012. To compare the survival of those patients with the expected survival of an average Belgian population, a randomly selected (age and sex) case-matched population was studied. The two groups were compared, using the log-rank test. The Cox proportional hazard models were adapted to estimate the relative risk of death for covariates in the dataset. Two-sided P-values of 0.05 were considered statistically significant. The statistical analysis was car- ried out using STATA and R-software. RESULTS Perioperative and in-hospital mortality and morbidity There were seven 30-day or in-hospital deaths (9.7%). No patient died during surgery. The patients died from the following causes: two as a consequence of sepsis, two due to multi-organ failure, one as a consequence of pneumonia, one because of right ventricular failure and one following rupture of an abdominal aneurysm. Postoperative complications can be summarized as follows: 14 patients (19%) suffered pulmonary compli- cations, three needed a revision for bleeding or tampon- ade (4%), 24 patients (33%) presented with one or more episodes of atrial fibrillation, high grade AV-block devel- oped in three patients (4%). Transient renal function impairment (GFR < 60 ml/min) occurred in 18 patients (25%), none of whom required dialysis. Time-related survival Within three months of discharge from the hospital, another four patients died: one from pneumonia, one as a consequence of CVA, one patient committed suicide PATIENTS AND METHODS From May 1998 to January 2012, 72 patients under- went elective reconstruction of the ascending aorta. Data from this patient cohort were analysed, using the data- base from the department of Cardiothoracic Surgery at the Jessa hospital. Follow-up information was obtained using patient files, telephone contact with the patient or his/her close relatives or through the general practi- tioner. Follow-up was 100% complete. Patient characteristics are shown in table I. The aetiology of the aortic disease was degenerative in 25 patients (35%). There was annulo-aortic ectasia in 13 patients (18%) and 17 patients presented with a bicus- pid valve (24%). Less common indications for recon- struction were aneurysm of the sinus of Valsalva (4 patients, 6%) and endocarditis (4 patients, 6%). One patient had a dysfunctional prosthesis and 8 patients (11%) were treated for chronic dissection of the ascend- ing aorta. As a rule, a diameter of 5.5 cm was taken as cut-off for surgery, except for Marfan patients and patients with bicuspid valves, in which case surgery was performed at 5 cm or more. In 44 patients, a Bentall procedure was performed using a valved conduit with a St. Jude mechanical pros- thesis. Sixteen patients underwent a modified Bentall procedure, using a bioprosthesis (8 Carpentier-Edwards pericardial valves, 2 Medtronic-Mosaic porcine valves, 2 Mitroflow pericardial valves, 1 biovalsalva graft, 2 Free- style full root prostheses and 1 full root homograft). Table 1 Patient characteristics and risk prole Variable (N = 72) Value % Gender (M/F) 46/26 64/36 Median age (years; Q1-Q3) 65.5 (57;71) Age distribution (years) 24 - 80 Cardiovascular risk factors Hypertension 35 48.6 Coronary artery disease 26 36.1 Diabetes mellitus 7 9.7 COPD 15 20.8 Bicuspid valve 15 20.8 NYHA class I & II 38 52.7 III & IV 34 47.2 Marfan syndrome 2 2.8 Ejection fraction 60% 42 58.3 > 60% 30 41.6 Creatinine (mg/dl; mean SD ) 1.09 0.42 Survival after ascending aortic reconstructive surgery 351 3.39 times higher than that of non-COPD patients. Patients with a good functional status (NYHA I or II) at the time of surgery survived better than those with a poor functional status (NYHA III and IV; P = 0.041). Other covariates could not be identified to significantly influence survival after reconstructive surgery. This, however, could be the consequence of the small sample size, combined with a limited number of events. No mortality occurred in the treated group from three years after surgery onwards. Six out of 43 patients who received a mechanical prosthesis, died. In the biological group, 3 out of 17 died and in the valve-sparing group 4 out of 12. Long-term survival was not significantly different between patients with a mechanical or a biological prosthesis (P = 0.62). Also valve-sparing surgery had a comparable survival (P = 0.33 vs biological prosthesis). and one elderly patient died in a nursing home as a con- sequence of progressive deterioration of his general con- dition. In the second year after surgery, one patient died of the consequences of diabetes type II. In year 3 after surgery, another two patients died, one following a road traffic accident and one from unknown causes. Three-, five- and ten-year survival remained constant at 80.9%. To compare the survival of these patients with the expected survival of the average Belgian population, a randomly selected age- and sex-matched case-control population was studied. The Kaplan-Meier survival curve of both groups is shown in figure 1. Three-, 5- and 10-year survival of this group was 95.7%, 94.7% and 85.2%, respectively. The case-matched population shows a better initial survival, due to the expected per operative mortality in the aortic surgery group. Later on, however, there is an incremental loss in the case-matched popula- tion, resulting in a similar long-term survival. Incremental risk factors for premature death To assess which covariates influence global survival of elective aortic reconstructive surgery, a Cox propor- tional hazard model was applied to the population. As preoperative variables, age of the patient at surgery, gender, left ventricular function, hypertension, presence of a bicuspid aortic valve, associated coronary disease, diabetes, COPD, renal function impairment and NYHA class were taken into account. As perioperative variables, surgeon experience and type of surgery were built into the model. The data of the proportional hazard model are shown in table 2. The presence of COPD at surgery revealed to be an unfavourable factor for global survival (P = 0.028). The risk of death for COPD patients in this series was Fig. 1 Kaplan-Meier survival curve in aortic surgery and case-matched population Table 2 Proportional hazard model Characteristic Unadjusted Adjusted HR (SE) P -value HR (SE) P -value Age 1.005 (0.026) 0.831 1.006 (0.031) 0.854 Gender 1.760 (0.981) 0.310 2.694 (1.800) 0.138 Hypertension 0.474 (0.285) 0.215 0.410 (0.278) 0.189 Bicuspid valve 0.674 (0.518) 0.608 1.099 (0.968) 0.915 Coronary disease 1.464 (0.707) 0.708 1.464 (0.927) 0.547 Diabetes mellitus 0.745 (0.776) 0.778 0.873 (0.974) 0.903 COPD 3.394 (1.889) 0.028 4.542 (2.875) 0.017 Renal failure 1.012 (0.011) 0.275 1.013 (0.015) 0.382 Surgeon experience (years) 0.911 (0.048) 0.076 0.931 (0.077) 0.390 NYHA (I or II vs III or IV) 0.112 (0.120) 0.041 Ejection fraction 0.634 (0.449) 0.51 D. Van Duel et al. 352 86% and 81.5%, respectively. The 5- and 10-year survival reported by Maureira 5 et al. is 86.3 2.78% and 73.7 4.23%. Tamura et al. 6 report a survival of 84.2% at 5 years and 64.3% at 15 years. Verbakel et al. 7 report a 78% 4% overall survival at 5 years and 66% 10% at 10 years (with an early mortality of 11.2%). In the previously mentioned National UK Registry 5 year sur- vival was 77.1% and 10 year survival 70% 3 . Because it is difficult to compare absolute survival data in different series, given the fact that risk factors may considerably vary between groups, we decided to compare the survival of the treatment group with the expected survival of an average Belgian population. A randomly selected age- and sex-matched case-control study population was obtained from the national Belgian census. Due to the expected early mortality, the case- matched population had a better initial survival. How- ever, due to a continued attrition rate in the case-matched population, long-term survival was not significantly different in both groups. Therefore, this study suggests that patients with degenerative disease of the ascending aorta who survive reconstructive surgery, have a low risk of complications and a normal life expectancy thereafter. Risk factor analysis Cox proportional hazard analysis identified COPD as an independent risk factor for overall mortality with an adjusted hazard ratio of 4.542 (SE 2.875, P = 0.017). The risk of premature death for COPD patients was 3.39 times higher than for non-COPD patients. Due to the limited number of patients and events, this study may, however, not have the mathematical power to identify additional variables. In our study, for example, age could not be identified to have a predictive value for survival. Although this observation is confirmed by other authors 8,9 , Prifti et al. 4
and David et al. 10 find a relation between age of the patient and survival. It should be pointed out, however, that the average age of this patient group was relatively high (65.5 years). In younger patients, survival is sig- nificantly superior. David et al. obtained a 5-year sur- vival of 94% and a 10-year survival of 88% in a patient population with an average age of 45 years at surgery (40% Marfan patients). The choice of valve for aortic root reconstruction seems to have no influence on long-term outcome. There was no difference between long-term survival of patients with a mechanical or a biological prosthesis (P = 0.62). This observation confirms previous reports by Etz et al. 11 and Lehr et al. 12 . However, in the context of aortic root reconstruction, all biological prostheses may not perform equally. A recent report on the Hancock bio- prosthetic porcine conduit for aortic root reconstruction Morbidity Forty-four patients were treated with coumadin for a mechanical prosthesis. In this group, two serious cer- ebral bleeding events occurred: one resulted in a per- manent deficit and one in a transient ischaemic attack. One patient with a biological Bentall procedure devel- oped an ischaemic CVA (4%). There were no thrombo- embolic complications in the valve-sparing group. One patient developed paraparesis following spinal chord ischaemia as a consequence of circulatory arrest for an associated replacement of the aortic arch. There were no reinterventions for failed Bentall pro- cedure as a consequence of pseudoaneurysma formation at the level of the coronary buttons. No significant aor- tic incompetence occurred in the valve-sparing group. DISCUSSION Hospital mortality Hospital mortality in our patient population of elec- tive ascending aortic surgery reconstruction was 9.7%. The calculated operative risk according to Euroscore II for this group was 9.2%. This is comparable to reported mortality in other papers. In a mixed population of elec- tive and urgent cases, Pacini et al. report a hospital mortality of 6.9% with 6.6% acute dissections 1 . Kou- choukos et al. report 5.2% mortality with 10% acute dissections 2 . In a recent report from a national cohort in the United Kingdom, 30-day mortality for non-urgent composite valved grafts in aortic position is 6.9% (1,379 patients) 3 . Our single-centre study confirms pre- vious reports that ascending aortic reconstruction can be performed with a reasonable operative risk. Neverthe- less, reconstruction of the ascending aorta remains a complex procedure and carries a significant early risk even in elective patients. Whether this early risk can be reduced, remains a matter of debate. However, in this study we identified functional class as a significant predictor of early mortal- ity (P = 0.041). Therefore we speculate that early outcome could be improved by earlier referral of patients, par- ticularly since poor functional class seems associated with aortic expansion and/or valvular regurgitation, resulting in left ventricular dysfunction. Long-term survival Five-year survival in the treated group was 80.9% and no further attrition was observed until 15 years after surgery. Although this is a limited number of patients, this long-term follow-up is comparable to other series. Prifti et al. 4 report a 3-, 5- and 10-year survival of 91.8%, Survival after ascending aortic reconstructive surgery 353 of late aortic incompetence. In this series we preserved the aortic valve only in patients with tricuspid aortic valve and normal cusps. With increasing confidence in this technique, indications have broadened. Several groups have shown that reconstruction of bicuspid aor- tic valves in the presence of proximal aortic dilatation results in a good short-term outcome, regardless whether the remodelling technique 16 or the reimplantation tech- nique is used 17 . In the younger population, valve-sparing surgery shows distinct advantages: in this patient group, no thrombo-embolic complications were observed, whereas in the mechanical prosthesis group one CVA and one TIA occurred and in the bioprosthesis group one ischae- mic CVA. Despite these promising evolutions in aortic recon- structive surgery, even the surgeons most dedicated to valve-sparing surgery, will admit that there are a number of severely diseased aortic valves, in which the Bentall procedure remains the treatment of choice. It is reassur- ing to see that this procedure can be carried out with a good and predictable short- and long-term survival and that for the hospital survivors we are able to restore life expectancy to normal. CONFLICT OF INTEREST: none declared.. reported low actuarial survival rates at 5 (77.0 5.3%) and 10 (54.0 7.5%) years (excluding operative and in- hospital deaths 14%). There was a high rate of reinter- vention in this group for structural valve degeneration (freedom from reoperation at 10 years 64 10.2%) 13 . It should be pointed out that in our series no reinterven- tions for structural degeneration had to be performed. This series includes 12 aortic valve-sparing proce- dures, designed to preserve the aortic valve cusps. David et al. 14 and Yacoub et al. 15 have developed different tech- niques in which the native valve can be preserved and lifelong anticoagulation can be avoided or at least post- poned. Remodelling of the aortic root, as first described by Yacoub et al. 15 is probably the more physiologic operation because it recreates the aortic sinuses and sino-tubular junction allowing for near-physiologic motion of the cusps and annulus. However, particularly in patients with annuloaortic ectasia, lack of support of the aortic annulus may result in progressive dilatation and subsequent aortic incompetence. Therefore, we did not use this technique in annuloaortic ectasia and man- aged to avoid significant aortic incompetence. Reim- plantation of the aortic valve, as described by David et al. 14 is a more demanding procedure, since it implies suturing the entire annulus and subcommissural trian- gles into a prosthetic graft. However, this technique stabilizes the annulus and therefore reduces the incidence 1. Pacini D, Ranocchi F, Angeli E, Settepani F, Pagliaro M, Martin-Suarez S, Di Bartolomeo R, Pierangeli A. Aortic root replacement with composite valve graft. Ann Thorac Surg 2003; 76: 90-8. 2. Kouchoukos NT, Wareing TH, Murphy SF, Perrillo JB. Sixteen-year experience with aortic root replacement. Results of 172 operations. Ann Surg 1991; 214: 308-18; discussion 18-20. 3. Kalkat MS, Edwards M-B, Taylor KM, Bonser RS. Composite aortic valve graft replacement: mortality outcomes in a national registry. Circulation 2007; 116: 301-6. 4. Prifti E, Bonacchi M, Frati G, Proietti P, Giunti G, Babatasi G, Massetti M, Sani G. Early and long-term outcome in patients undergoing aortic root replacement with composite graft according with composite graft according to Bentalls technique. Eur J Cardiothorac Surg 2002; 21: 15-21. 5. Maureira P, Vanhuyse F, Martin C, Lekehal M, Carteaux JP, Tran N, Villemot JP. Modified Bentall procedure using two short grafts for coronary reimplantation: long-term results. Ann Thorac Surg 2012; 93: 443-9. 6. Tamura K, Arai H, Kawaguchi S, Makita S, Miyagi N, Watanabe T, Fujiwara T. Long-term results of modified Bentall procedure using flanged composite aortic prosthesis. Ann Thorac Cardiovasc Surg 2013; 19: 126-30. 7. Verbakel KM, Van Straten AH, Hamad MA, Tan ES, ter Woorst JF. Results of one-hundred and seventy patients after elective Bentall operation. Asian Cardiovasc Thorac Ann 2012; 20: 418-25. 8. Svensson L, Crawford S, Hess K, Coselli J, Safi H. Composite valve graft replacement of the proximal aorta: comparison of techniques in 348 patints. Ann Thorac Surg 1992; 54: 427-37. 9. Tambeur L, Tirone DE, Unger M, Armstrong S, Ivanov J, Webb G. Results of surgery for aortic root aneurysm in patients with the Marfan syndrome. Eur J Cardiothorac Surg 2000; 17: 415-9. 10. David TE, Feindel CM, Webb GD, Colman JM, Armstrong S, Maganti M. Long-term results of aortic valve-sparing operations for aortic root aneurysm. J Thorac Cardiovasc Surg 2006; 132: 347-54. 11. Etz CD, Bischoff MS, Bodian C, Roder F, Brenner R, Griepp RB, Di Luozzo G. The Bentall procedure: is it the gold standard? A series of 597 consecutive cases. J Thorac Cardiovasc Surg 2010; 140(6 Suppl): S64-70; discussion S86-91. 12. Lehr EJ, Wang PZ, Oreopoulos A, Kanji H, Norris C, Macarthur R. Midterm outcomes and quality of life of aortic root replacement: mechanical vs biological conduits. Can J Cardiol 2011; 27: 262.e15-20. 13. Badiu CC, Bleiziffer S, Eichinger WB, Hettich I, Krane M, Bauernschmitt R, Lange R. Long-term performance of the Hancock bioprosthetic valved conduit in the aortic root position. J Heart Valve Dis 2011; 20: 191-8. 14. David TE, Armstrong S, Ivanov J, Webb GD. Aortic valve sparing operations: an update. Ann Thorac Surg 1999; 67: 1840-56. 15. Yacoub MH, Gehle P, Chandrasekaran V, Birks EJ, Child A, Radley-Smith R. Late results of a valve-sparing operation in patients with aneurysms of the ascending aorta and root. J Thorac Cardiovasc Surg 1998; 115: 1080-90. 16. Schfers HJ, Kunihara T, Fries P, Brittner B, Aicher DJ. Valve-preserving root replacement in bicuspid aortic valves. J Thorac Cardiovasc Surg 2010; 140(6 Suppl): S36-40; discussion S45-51. 17. de Kerchove L, Boodhwani M, Glineur D, Vandyck M, Vanoverschelde J-L, Noirhomme P, El Khoury G. Valve sparing-root replacement with the reimplantation technique to increase the durability of bicuspid aortic valve repair. J Thorac Cardiovasc Surg 2011; 142: 1430-8 REFERENCES