- Email: [email protected]
Comparison of Two Strategies for Aortic Valve-Sparing Root Replacement
Journal Pre-proof A Comparison of Two Strategies for Aortic Valve-sparing Root Replacement Erik Beckmann, MD, Alessandro Leone, MD, Andreas Martens, MD, Carlo Mariani, MD, Heike Krueger, RN, Serghei Cebotari, MD, Roberto Di Bartolomeo, MD, Axel Haverich, MD, Malakh Lal Shrestha, MD, Davide Pacini, MD PII:
S0003-4975(19)31068-9
DOI:
https://doi.org/10.1016/j.athoracsur.2019.07.006
Reference:
ATS 32846
To appear in:
The Annals of Thoracic Surgery
Please cite this article as: Beckmann E, Leone A, Martens A, Mariani C, Krueger H, Cebotari S, Di Bartolomeo R, Haverich A, Shrestha ML, Pacini D, A Comparison of Two Strategies for Aortic Valvesparing Root Replacement, The Annals of Thoracic Surgery (2019), doi: https://doi.org/10.1016/ j.athoracsur.2019.07.006. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 by The Society of Thoracic Surgeons
A Comparison of Two Strategies for Aortic Valve-sparing Root Replacement Running title: straight vs. valsalva graft
Erik Beckmann, MD1, Alessandro Leone, MD2, Andreas Martens, MD1, Carlo Mariani, MD2, Heike Krueger, RN1, Serghei Cebotari, MD1, Roberto Di Bartolomeo, MD2, Axel Haverich, MD1, Malakh Lal Shrestha, MD1, Davide Pacini, MD2
1
Department of Cardiothoracic, Transplantation and Vascular Surgery, Hannover Medical
School, Germany 2
Department of Cardiac Surgery, University of Bologna Hospital, Italy
The manuscript has been selected for presentation at the 54th Annual Meeting of the Society of Thoracic Surgeons in Ft. Lauderdale, FL, January 27 to 31, 2018.
Corresponding author: Erik Beckmann Department of Cardiothoracic Transplantation and Vascular Surgery Carl-Neuberg-Str. 1 Hannover 30625, Germany Email: [email protected]
Word count: 4602
1
Abstract Background. Since its introduction in 1992, multiple variations of aortic valve-sparing David procedure technique have been described. Here, we present the short- and mid-term outcome of two centers using the straight tube graft (David-I) and the Valsalva prosthesis in patients who underwent isolated David procedure.
Methods. Between 03/2002-10/2015, 232 patients underwent David procedure at two European centers. Patients received either straight tube graft (David-I, group A, n=103, 74% male) or Valsalva graft (group B, n=129, 85% male). Mean age was 47±17 in group A and 48±17 in group B (p=0.916).
Results. There were significantly more cusp repairs in group B (n=28, 22%) compared to group A (n=4, 4%, n<0.001). The 30-day mortality rate was 1% (n=1) in group A and 2% (n=2, p=0.698) in group B. Postoperative echocardiography showed aortic insufficiency ≥2° in 0% (n=0) of group A and 17% (n=21) of group B (p<0.001). Follow-up comprised a total of 1530 patient-years, and survival was comparable between the two groups (p=0.799). Follow-up echocardiography showed aortic insufficiency ≥II° in 22% (n=15) of group A and 39% (n=33) of group B (p<0.026). The rates for aortic valve-related re-operation were 8% (n=8) in group A and 13% (n=16) in group B (p=0.241). Logistic Cox regression analysis identified bicuspid aortic valve (OR=3.435, 95%-CI=1.459 – 8.083, p=0.005) and postoperative aortic insufficiency ≥II° (OR=5.988, 95%-CI=2.545 – 14.088, p<0.001) as risk factors for aortic valve related re-operation.
Conclusions. Aortic valve-sparing David procedure has acceptable mid-term results. Our results show that the David-I procedure with straight tube graft is not inferior to those performed with Valsalva prosthesis.
2
Valve-sparing aortic root replacement has been first described by David for the treatment of aortic root aneurysms (1). Multiple studies have shown excellent short-, mid- and long-term outcome of this technique for both elective and emergent settings, as well as in Marfan patients (2-9). The original reimplantation technique has been performed with a straight tube graft, now generally know as the David I procedure (1). In regards to physiology, it has been demonstrated that the Sinus of Valsalva have beneficial effects on stress and hemodynamics of the aorta and the aortic valve (10). Both in vivo and in vitro experiments have raised the concern that the usage of a straight tube graft might result in less favorable hemodynamics (11,12). Another concern that has been raised is the potential contact of the aortic leaflets with the aortic wall, which could theoretically lead to early valve deterioration (12,13). For these reasons, it has been suggested that a prosthetic graft with preformed sinuses of Valsalva might lead to more physiological biomechanics (13-15). Various companies have introduced prosthetic grafts with preformed sinus of Valsalva (13,14). Although the advocates of sinus of Valsalva grafts emphasize more optimal hemodynamics, no study has ever shown a clinical benefit in terms of survival or rate of aortic valve-related re-operation of these grafts when compared to the original David technique with a straight tube graft. Therefore, we have conducted this retrospective analysis with follow-up to analyze the clinical outcome of the straight tube graft technique and the Valsalva graft method in a large patient cohort of two high-volume European centers. We hypothesized that a straight tube graft is not inferior to a Valsalva graft. The primary end point of this study was aortic valve-related re-operation during follow up.
Patients and Methods Study design Between 03/2002-10/2015, 232 patients underwent isolated David procedure at two European centers (Bologna University Hospital, Italy and Hannover Medical School, Germany). Only patients who underwent isolated David procedure without any concomitant cardiac surgery 3
were included in this study. The study was in line with both centers’ ethical standards. Patients received either a straight tube graft (David-I, group A, n=103) or a Valsalva graft (group B, n=129). Of note, all operations using a Valsalva graft have been performed by University of Bologna Hospital, while all except one operations with a straight tube graft have been performed by Hannover Medical School.
Surgical technique The surgical techniques of both groups have been described in previous publications (3,7). The detailed surgical technique can be found in the supplemental section.
Anticoagulation therapy and postoperative follow-up After aortic valve-sparing surgery, patients were anticoagulated with Coumadin for the first 3 months only. Thereafter, the patients received a life-long therapy with aspirin unless there were no other indications. We believe that that initial anticoagulation with coumadin helps to prevent thrombembolic events until a neo-intimal layer has formed on the inside of the prosthetic graft. Echocardiography was performed intraoperatively and before discharge. Individual consent was obtained from patients to allow for follow-up examination. Follow-up was done according to common guidelines (16). Patients were contacted by telephone and/or seen in our clinic. Primary care physicians and/or cardiologists were contacted and examination results with echocardiography data were obtained. Primary endpoint of this study was aortic valve-related re-operation. The indication for valverelated re-operation was a failing aortic valve (not aortic stenosis).
Statistical analysis Data analysis was performed using SPSS 22 Statistics software (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.). Normal distribution of variables was analyzed with the Kolmogorov-Smirnov test. Normally 4
distributed continuous variables are stated as mean ± standard deviation, while continuous variables without normal distribution are stated as median + range. Categorical variables are stated as absolute numbers and proportions. Differences in categorical variables were analyzed using the Χ2test or Fisher’s exact test, depending on sample size. Differences in continuous variables were tested using t-test or Mann-Whitney u-test, depending on sample size. Kaplan-Meier analysis was used for evaluation of both survival and re-operation of the aortic valve, and the log-rank test was used to test for differences. Cox regression analysis was performed using a binary logistic regression model to discriminate independent risk factors for aortic-valve related re-operation. All included variables can be found in the univariate anaysis in supplemental table 2. A value of p<0.05 was considered statistically significant.
Results Descriptive analysis The detailed patient demographics can be found in table 1. The mean age of all patients was 47 ± 17 years and 186 (80%) were male. There were more male in group B (85%, n=110) than in group A (74%, n=76, p=0.029). Hyperlipidemia showed a higher incidence in group A (27%, n=28) than in group B (14%, n=18, p=0.012). Nicotine usage was more common in group B (40%, n=52) than in group A (19%, n=20, p=0.001).
All other pre-operative
characteristics were evenly distributed in both groups. Marfan’s syndrome was present in 30% (n=31, group A) and 24% (n=31, group B, p=0.300). Bicuspid aortic valve (BAV) was present in 9% (n=9, group A) and 13% (n=17, group B, p=0.287). In the entire cohort, the indication for surgery was aortic aneurysm in 93% (n=216) of patients and acute aortic dissection type Stanford A with root involvement in 7% (n=16) of patients. There was no significant difference in indication for surgery betwen the two groups (p=0.957). Preoperative echocardiographic data can be found in table 2. The degrees of aortic insufficiency (AI) was distributed mostly equal between the two groups (p=0.180).
5
The peri-operative data is shown in table 3. The mean aortic cross-clamp times were 112±21 in group A and 127±22 min in group B, respectively (p<0.001). There were significantly more cusp repairs in group B (n=28, 22%) compared to group A (n=4, 4%, n<0.001) (see also supplemental table 1). The 30-day mortality rate was 1% (n=1) in group A and 2% (n=2, p=0.698) in group B. The overall stroke rate was 1% (n=2) and was not significantly different in both groups (p=0.204). Postoperative echocardiography was performed for 93% (n=215, table 4) of all patients. Echocardiography showed AI≥II° in 0% (n=0) of group A and 17% (n=21) of group B (p<0.001).
Follow-up results Follow-up echocardiography was obtained for 66% (n=153) of all patients. The mean echocardiographic follow up time was 5.44 ± 3.2 years for group A and 5.5 ± 4.1 years for group B. Follow-up echocardiography showed AI ≥ II° in 22% (n=15) of group A and 39% (n=33) of group B (p<0.026, table 4). The overall rate for aortic valve-related re-operation was 10% (n=24) in the entire cohort (table 4). The overall rates for aortic-valve-related reoperation were 8% (n=8) in group A and 13% (n=16) in group B (p=0.241). The follow-up was complete for 100% of patients and comprised a total of 1530 patient-years. The mean follow up time was 5.8 ± 3.5 years for group A and 7.2 ± 4.3 years for group B. The Kaplan Meier curve for freedom from aortic-valve related re-operation is given in figure 1. In group A, the estimates for freedom from valve-related re-operation at 1, 5 and 10 years after initial surgery were 100%, 92% and 86%, respectively. In group B, the 1, 5 and 10-year estimates for freedom from valve-related re-operation were 99%, 93% and 87%, respectively. There was no significant difference between the two groups (p=0.420). The reasons for reoperation in these patients were endocarditis in two patients and a failed aortic valve in six patients of group A. In group B, one patient had endocarditis and fourteen had a failed aortic valve.
6
The Kaplan Meier curve for survival is shown in figure 2. The survival estimates at 1, 5 and 10 years were 99%, 99% and 85% in group A. The respective estimates for group B were 98%, 96% and 93%. There was no significant difference between the two groups (p=0.799).
Risk factor analysis We performed a risk factor analysis for aortic valve-related re-operation. The results of the univariate analysis can be found in the supplemental section (supplemental table 2). While univariate analysis showed that BAV (p=0.004), graft type (p=0.009), and center (p=0.010) were significantly associated with aortic valve-related re-operation, this was not confirmed during logistic Cox regression analysis. Logistic Cox regression analysis showed that BAV (OR=3.435, 95%-CI=1.459 – 8.083, p=0.005) and post-operative AI ≥ II° (OR=5.988, 95%CI=2.545 – 14.088, p<0.001) were risk factors for re-operation of the aortic valve (table 5).
Subgroup analysis We performed a subgroup analysis and excluded patients with BAV, those who underwent cusp repair, and/or those who had postoperative AI ≥ II°, in order to exclude patients who could potentially bias the the outcome during follow-up. The detailed results can be found in the supplemental section. The majority of the patient demographics were distributed equally among the two subgroups (supplemental table 3). The preoperative echocardiographic data showed that the straight tube graft subgroup showed a higher rate of AI III° and IV° (p=0.046, supplemental table 4). The early postoperative outcome was mostly equal between the two subgroups (supplemental table 5). During follow-up, the incidence of aortic valve-related reoperation was comparable between the straight tube subgroup (n=5, 5%) and the Valsalva subgroup (n=4, 5%, p=0.841). We have also performed a risk factor analysis for aortic valverelated re-operaton, and no risk factor was identified.
Comment
7
Valve-sparing aortic root replacement with a straight tube graft (David I) has been criticized for its potentially suboptimal hemodynamics when compared to grafts with preformed sinus of Valsalva (11-13). In this study, we have compiled the data of two experienced European centers, and aimed at analyzing the outcome of valve-sparing aortic root replacement using both straight tube grafts and Valsalva grafts. In consideration of the study design’s potential inherent limitations, we could demonstrate that the usage of a straight tube graft does not compromise the short- and mid-term outcome after valve-sparing aortic root replacement and is not inferior when compared to Valsalva grafts.
Early postoperative outcome In this study, all operations with a Valsalva graft have been performed by University of Bologna Hospital, while all except one operations with a straight tube graft have been performed by Hannover Medical School. In both centers and with both graft types, the early postoperative results showed a low and comparable rate of complications. The overall 30-day mortality was 1%, which represents a reasonable number for an aortic (valve-sparing) root replacement. The 30-day mortality rate of our study is similar to the results of other reports or even lower (3,6,9,17). Both the rate for permanent neurological deficit and for acute kidny injury requiring dialysis were 1%, respectively, which represent low rates as well. As the rates for perioperative mortality and morbidity were very low in our cohort and as we did not observe any statstically significant difference between the two groups, we conclude that both techniques (straight tube graft and Valsalva graft) are safe and feasible approaches for valvesparing aortic root replacement.
The postoperative echocardiography results showed significantly more patients with AI ≥ II° in group B (Valsalva graft) than in group A (straight tube graft). This difference can directly be contributed to the different hospital approaches on the tolerance of postoperative AI. At University of Bologna Hospital, where the Valsalva grafts have been used, it is the policy to tolerate AI II°. Only if there is AI II-III or more on postoperative echocardiography, the aorta 8
is reclamped and the aortic valve is revised. In contrast, at Hannover Medical School, it is the policy to not tolerate AI II°. At this center, the policy is to achieve AI ≤ I° and if there is a higher degree of AI, the aorta is reclamped and the aortic valve reconstruction is revised. Second, the rate of cusp repair was significantly higher at University of Bologna Hospital. The more aggressive approach towards cusp repair represents another strategic difference between the two participating centers. Achieving an adequate results with a structurually not ideal aortic valve is more challenging. These differences have to be kept in mind when reviewing the early postoperative results, especially the echocardiographic data.
Follow-up During follow up, we observed no statistically significant difference between the mid-term surival of patients who received a straight tube graft and those who received a valsalva graft. Furthermore, patients who underwent aortic valve-sparing root replacement with a straight tube graft did not have a higher rate of failing aortic valves than patients who received a Valsalva graft. The rates for reoperation of the aortic valve were similar in both groups and there was no statistically significant difference. The slightly higher rate of aortic valve-related re-operation in the Valsalva group (although not statistically significant) might again be explained by the fact that only two centers were involved in this study, involving their different (peri-)operative approaches as mentioned earlier. The rate for cusp repair was significantly higher in patients who received a Valsalva graft. It has been previously described that cusp repair in the setting of aortic valve-sparing root replacement is a risk factor for late adverse outcome (18). At Bologna University, during the early experience cusp repair was performed when necessary as: decalcification of the leaflets, reinforcement of the free edge or plication. However, these techniques have been progressively abandoned, due to the worse results that we observed in our clinical practice, especially in bicuspid aortic valve. This can be regarded as an institutional learning curve. The difference in the rate of cusp repair could explain the higher rate for valve-related reoperation. 9
Furthermore, the higher rate of cusp repairs in the Valsalva group could also explain the higher rate of postoperative AI ≥ II°. As mentioned earlier, there were significantly more cusp repairs performed in the Valsalva group, indicating more complex and challenging repair. Since more cusp repairs were performed in the Valsalva group, and since these operations have been performed at only one institution where it is the policy to accpet AI II°, this could also predispose to potentially more reoperations in the future. This institutional difference on approaching aortic valve repair with its potential for bias has to be kept in mind when reviewing the follow-up data. In the Kaplan Meier curve for valve-related re-operation, there seems to be an evolving divergence at approximately 7 years. This raises concern for emerging divergence between the two groups during the future follow up time. However, the follow up rate for 7+ years is only 12% for group A and 27% for group B, therefore we cannot comment on this issue with certainty. Future studies with a longer follow-up time will be needed to investigate this issue. We conclude that – even if the Valsalva graft may provide better hemodynamics (10-15) – the usage of a straight tube graft for David procedure does not compromise the mid-term performance of the aortic valve and is not inferior to the Valsalva graft.
Risk factor analysis During univariate analysis, we found that the variables BAV, graft type, and center were associated with aortic valve related reoperation. However, during logistic Cox regression analsyis this was not confirmned for all of these three variables. Instead, we found that the variables BAV and postoperative AI ≥ II° were the only two risk factors for aortic valverelated re-operation. This difference between the uni- and multivariate analysis can be explained by the fact that the Valsalva graft implantations have been performed at one center, with a higher rate of postoperative AI. As outlined before, this is due to the higher rate of cusp repair and the hospital policy on the acceptance of AI. Therefore, one has to assume that the variables graft type and center are confounding variables, and only the variable postoperative AI ≥ II° is a true risk factor for future re-operation. 10
In addition to AI ≥ II°, our logistic Cox regression model identified BAV to be a risk factor for aortic valve-related re-operation. The current literature on the fate of a BAV after valvesparing aortic root replacement is not completely consistent. On the one hand, there are articles reporting an adequate outcome after aortic valve-sparing root replacement and a BAV (19). On the other hand, there are studies that observed a higher rate of late adverse outcome after repair of BAV (18).
Since there was concern that the higher rate of postoperative AI ≥ II° in the Valsalva group could potentially bias the analysis, we performed a subgroup analysis and excluded all patients who had BAV, those who underwent cusp repair, and/or those who had postoperative AI ≥ II°. By these means, we were aiming to eliminate all patients who had already a suboptimal aortic valve directly postoperatively. During follow up, we observed a comparable rate of aortic valve-related reoperations between the two groups. We did not find a statistically significant difference. These results corroborate that the two graft types are indeed equal and do not have an impact on the durability of the aortic valve reconstruction.
Limitations One limitation of this study is its retrospective character. Furthermore, there might be selection bias, since there was no random assignment to the two groups like in a randomized trial. However, we aimed at compensating for this issue by including only isolated David procedures and excluded cases with concomitant procedures to prevent bias from additional extensive cardiac procedures. Although our follow up was 100% complete, the completion rate for echocardiographic data was only 66%. This can be regarded as another limitation, and is contributed to the retrospective character of this study with the first operations having been performed more than 15 years ago. Although we have contacted all patients, we were not able to obtain echocardiographic data from all of them. This issue could be addressed in future prospective and randomized studies. 11
Another limitation of this study is the fact that all operations with a Valsalva graft have been performed by University of Bologna Hospital, while all except one operations with a straight tube graft have been performed by Hannover Medical School. We acknowledge that there might be potential institutional bias, including the different institutional strategies on how to approach aortic valve repair and tolerate the degree of postoperative AI. However, both centers have a significant expertise with the David procedure and perform this operation frequently. Indeed, our logistic Cox regression analysis confirmed that the center is not associated with adverse outcome. Furthermore, the operations have not been performed by a single surgeon but multiple surgeons per center.
Conclusions Regardless of the underlying pathology, aortic valve-sparing David procedure is a safe operation with acceptable mid-term outcome. Our results show that the David I procedure with a straight tube graft is not inferior to those performed with a Valsalva prosthesis.
12
References
1. David TE, Feindel CM. An aortic valve-sparing operation for patients with aortic incompetence and aneurysm of the ascending aorta. J Thorac Cardiovasc Surg. 1992; 103(4):617-21.
2. David TE, Armstrong S, Ivanov J, et al. Results of aortic valve-sparing operations. J Thorac Cardiovasc Surg. 2001; 122:39-46.
3. Shrestha M, Baraki H, Maeding I, et al. Long-term Results after aortic valve-sparing operation (David I). Eur J Cardiothorac Surg. 2012; 41: 56-62.
4. Beckmann E, Martens A, Alhadi FA, Ius F, Koigeldiyev N, Fleissner F, et al. Is Bentall Procedure Still the Gold Standard for Acute Aortic Dissection with Aortic Root Involvement? Thorac Cardiovasc Surg. 2016; 64(2):116-23.
5. Beckmann E, Martens A, Pertz J, et al. Valve-sparing David I procedure in acute aortic type A dissection: a 20-year experience with more than 100 patients. Eur J Cardiothorac Surg. 2017; 52(2):319-324.
6. Di Bartolomeo R, Pacini D, Martin-Suarez S, et al. Valsalva prosthesis in aortic valve sparing operations. Interact Cardiovasc Thorac Surg. 2006; 5:294-298.
7. Pacini D, Settepani F, De Paulis R, et al. Early results of valve-sparing reimplantation procedure using the valsalva conduit: a multicenter study. Ann Thorac Surg. 2006; 82:865-72.
13
8. Cameron DE, Alejo DE, Patel ND, et al. Aortic root replacement in 372 Marfan patients: evolution of operative repair over 30 years. Ann Thorac Surg. 2009; 87(5):1344-9.
9. De Paulis R, Scaffa R, Nardella S, et al. Use of the Valsalva graft and long-term follow-up. J Thorac Cardiovasc Surg. 2010; 140(6):S23-7.
10. Robicsek F, Thubrikar MJ. Role of sinus wall compliance in aortic leaflet function. Am J Cardiol. 1999; 944-946.
11. Leyh RG, Schmidtke C, Sievers CC, et al. Opening and closing characteristics of the aortic valve after different types of valve-preserving surgery. Circulation. 1999; 100:21532160.
12. Fries R, Graeter T, Aicher D, et al. In vitro comparison of aortic valve movement after valve-preserving aortic replacement. J Thorac Cardiovasc Surg. 2006, 132:32-7.
13. Schmidtke C, Sievers HH, Frydrychowicz A, et al. First clinical results with the new sinus prosthesis used for valve-sparing aortic root replacement. Eur J Cardiothorac Surg. 2013; 43:585-590.
14. De Paulis R, De Matteis GM, Nardi P, et al. A new aortic Dacron conduit for surgical treatment of aortic root pathology. Ital Heart J. 2000; 1:457-463.
15. De Paulis R, De Matteis GM, Nardi P, et al. Analysis of valve motion after the reimplantation type of valve-sparing procedure (David I) with a new aortic root conduit. Ann Thorac Surg. 2002; 74:53-57.
14
16. Akins CW, Miller DC, Turina MI, et al. Guidelines for reporting mortality and morbidity after cardiac valve interventions. Eur J Cardiothorac Surg. 2008; 33(4):523-528.
17. David TE, Feindel CM, Webb GD, et al. Aortic valve preservation in patients with aortic root aneurysm: results of the reimplantation technique. Ann Thorac Surg. 2007; 83(2):S732-5.
18. Esaki J, Leshnower BG, Binongo JN, et al. Risk factors for late aortic valve dysfunction after David V valve-sparing root replacement. Ann Thorac Surg. 2017; 104(5):1479-1487.
19. Bavaria JE, Desai N, Szeto WZ, et al. Valve-sparing root replacement and leaflet repair in a bicuspid aortic valve: comparison with the 3-cusp David preocedure. J Thorac Cardiovasc Surg. 2015; 149(2 Suppl):S22-8.
15
Table 1: Patient characteristics Entire cohort
Straight tube
Valsalva
p-value
n=232
n=103
n=129
Age (years)
47 ± 17
47 ± 17
47 ± 17
0.916
Male (n, %)
186 (80%)
76 (74%)
110 (85%)
0.029
Height (cm)
176 ± 20
176 ± 27
176 ± 12
0.954
Weight (kg)
81 ± 14
82 ± 16
80 ± 13
0.568
BMI (kg/m2)
26 ± 5
25 ± 6
26 ± 5
0.085
Marfan’s syndrome (n, %)
62 (27%)
31 (30%)
31 (24%)
0.300
Bicuspid aortic valve (n, %)
26 (11%)
9 (9%)
17 (13%)
0.287
Arterial hypertension (n, %)
137 (59%)
62 (60%)
72 (56%)
0.752
Hyperlipidemia (n, %)
46 (20%)
28 (27%)
18 (14%)
0.012
Diabetes mellitus (n, %)
7 (3%)
3 (3%)
4 (3%)
0.934
Nicotine (n, %)
72 (31%)
20 (19%)
52 (40%)
0.001
Re-do (n, %)
5 (2%)
4 (4%)
1 (1%)
0.105
Maximum aortic diameter (mm)
53 ±7
54 ± 8
52 ± 6
0.003
Aortic aneurysm (n, %)
216 (93%)
96 (93%)
120 (93%)
0.957
Acute type A dissection (n, %)
16 (7%)
7 (7%)
9 (7%)
0.957
Characteristic
Indication for surgery:
Abbreviations for table 1: BMI = body mass index
16
Table 2: Pre-operative echocardiographic data Entire cohort
Straight tube
Valsalva
n=232
n=103
n=129
AI 0° (n, %)
27 (12%)
10 (10%)
17 (13%)
AI I° (n, %)
47 (20%)
21 (20%)
26 (20%)
AI II° (n, %)
69 (30%)
27 (26%)
42 (33%)
AI III (n, %)
70 (30%)
39 (38%)
31 (24%)
AI IV° (n, %)
19 (8%)
6 (6%)
13 (10%)
Characteristic
p-value
0.180
Abbreviations for table 1: AI = aortic insufficiency
17
Table 3: Peri-operative data Entire cohort
Straight tube
Valsalva
p-value
n=232
n=103
n=129
159 ± 36
152 ± 30
0.128
Characteristic Cardiopulmonary
bypass
time 155 ± 33
(min) Aortic cross-clamp time (min)
121 ± 23
112 ± 21
127 ± 22
<0.001
Cusp repair (n, %)
32 (14%)
4 (4%)
28 (22%)
<0.001
ICU stay (d)
1 (0-17)
1 (0-17)
1 (0-6)
0.486
Rethoracotomy (n, %)
10 (4%)
3 (3%)
7 (5%)
0.349
New dialysis (n, %)
2 (1%)
0 (0%)
2 (2%)
0.204
Permanent neurological deficit (n,
2 (1%)
0 (0%)
2 (2%)
0.204
3 (1%)
1 (1%)
2 (2%)
0.698
%) Perioperative mortality (n, %)
Abbreviations for table 3: ICU = intensive care unit
18
Table 4: Early and late postoperative aortic valve-related outcome
Entire cohort
Straight tube
Valsalva
n=215
n=94
n=121
AI < II° (n, %)
194 (90%)
94 (100%)
100 (83%)
AI ≥ II° (n, %)
21 (10%)
0 (0%)
21 (17%)
Follow up echocardiography
n=153
n=68
n=85
AI < II° (n, %)
105 (69%)
53 (78%)
52 (61%)
AI ≥ II° (n, %)
48 (31%)
15 (22%)
33 (39%)
Re-operation
n=232
n=103
n=121
Aortic valve-related re-operation
24 (10%)
8 (8%)
16 (13%)
Postoperative
p-value
echocardiography <0.001
0.026
0.241
(n, %)
Abbreviations for table 4: AI = aortic insufficiency
19
Table 5: Risk factor analysis for aortic valve-related re-operation. Factor
p-value
Odds ratio
95%-CI
Bicuspid valve
0.005
3.435
1.459 – 8.083
Postoperative AI ≥ II°
<0.001
5.988
2.545 – 14.088
Abbreviations for table 5: 95%-CI = 95% confidence intervall, AI = aortic insufficiency
20
Figure Legends
Figure 1: Freedom from aortic valve-related re-operation. Kaplan Meier curves for freedom from valve-related re-operation including all patients who underwent isolated David procedure with either a straight tube graft or a valsalva graft. Time origin on x-axis denotes day of surgery.
Figure 2: Survival. Kaplan Meier survival curves for patients who underwent isolated David procedure with either a straight tube graft or a valsalva graft. Time origin on x-axis denotes day of surgery.
21
S0003-4975(19)31068-9
DOI:
https://doi.org/10.1016/j.athoracsur.2019.07.006
Reference:
ATS 32846
To appear in:
The Annals of Thoracic Surgery
Please cite this article as: Beckmann E, Leone A, Martens A, Mariani C, Krueger H, Cebotari S, Di Bartolomeo R, Haverich A, Shrestha ML, Pacini D, A Comparison of Two Strategies for Aortic Valvesparing Root Replacement, The Annals of Thoracic Surgery (2019), doi: https://doi.org/10.1016/ j.athoracsur.2019.07.006. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 by The Society of Thoracic Surgeons
A Comparison of Two Strategies for Aortic Valve-sparing Root Replacement Running title: straight vs. valsalva graft
Erik Beckmann, MD1, Alessandro Leone, MD2, Andreas Martens, MD1, Carlo Mariani, MD2, Heike Krueger, RN1, Serghei Cebotari, MD1, Roberto Di Bartolomeo, MD2, Axel Haverich, MD1, Malakh Lal Shrestha, MD1, Davide Pacini, MD2
1
Department of Cardiothoracic, Transplantation and Vascular Surgery, Hannover Medical
School, Germany 2
Department of Cardiac Surgery, University of Bologna Hospital, Italy
The manuscript has been selected for presentation at the 54th Annual Meeting of the Society of Thoracic Surgeons in Ft. Lauderdale, FL, January 27 to 31, 2018.
Corresponding author: Erik Beckmann Department of Cardiothoracic Transplantation and Vascular Surgery Carl-Neuberg-Str. 1 Hannover 30625, Germany Email: [email protected]
Word count: 4602
1
Abstract Background. Since its introduction in 1992, multiple variations of aortic valve-sparing David procedure technique have been described. Here, we present the short- and mid-term outcome of two centers using the straight tube graft (David-I) and the Valsalva prosthesis in patients who underwent isolated David procedure.
Methods. Between 03/2002-10/2015, 232 patients underwent David procedure at two European centers. Patients received either straight tube graft (David-I, group A, n=103, 74% male) or Valsalva graft (group B, n=129, 85% male). Mean age was 47±17 in group A and 48±17 in group B (p=0.916).
Results. There were significantly more cusp repairs in group B (n=28, 22%) compared to group A (n=4, 4%, n<0.001). The 30-day mortality rate was 1% (n=1) in group A and 2% (n=2, p=0.698) in group B. Postoperative echocardiography showed aortic insufficiency ≥2° in 0% (n=0) of group A and 17% (n=21) of group B (p<0.001). Follow-up comprised a total of 1530 patient-years, and survival was comparable between the two groups (p=0.799). Follow-up echocardiography showed aortic insufficiency ≥II° in 22% (n=15) of group A and 39% (n=33) of group B (p<0.026). The rates for aortic valve-related re-operation were 8% (n=8) in group A and 13% (n=16) in group B (p=0.241). Logistic Cox regression analysis identified bicuspid aortic valve (OR=3.435, 95%-CI=1.459 – 8.083, p=0.005) and postoperative aortic insufficiency ≥II° (OR=5.988, 95%-CI=2.545 – 14.088, p<0.001) as risk factors for aortic valve related re-operation.
Conclusions. Aortic valve-sparing David procedure has acceptable mid-term results. Our results show that the David-I procedure with straight tube graft is not inferior to those performed with Valsalva prosthesis.
2
Valve-sparing aortic root replacement has been first described by David for the treatment of aortic root aneurysms (1). Multiple studies have shown excellent short-, mid- and long-term outcome of this technique for both elective and emergent settings, as well as in Marfan patients (2-9). The original reimplantation technique has been performed with a straight tube graft, now generally know as the David I procedure (1). In regards to physiology, it has been demonstrated that the Sinus of Valsalva have beneficial effects on stress and hemodynamics of the aorta and the aortic valve (10). Both in vivo and in vitro experiments have raised the concern that the usage of a straight tube graft might result in less favorable hemodynamics (11,12). Another concern that has been raised is the potential contact of the aortic leaflets with the aortic wall, which could theoretically lead to early valve deterioration (12,13). For these reasons, it has been suggested that a prosthetic graft with preformed sinuses of Valsalva might lead to more physiological biomechanics (13-15). Various companies have introduced prosthetic grafts with preformed sinus of Valsalva (13,14). Although the advocates of sinus of Valsalva grafts emphasize more optimal hemodynamics, no study has ever shown a clinical benefit in terms of survival or rate of aortic valve-related re-operation of these grafts when compared to the original David technique with a straight tube graft. Therefore, we have conducted this retrospective analysis with follow-up to analyze the clinical outcome of the straight tube graft technique and the Valsalva graft method in a large patient cohort of two high-volume European centers. We hypothesized that a straight tube graft is not inferior to a Valsalva graft. The primary end point of this study was aortic valve-related re-operation during follow up.
Patients and Methods Study design Between 03/2002-10/2015, 232 patients underwent isolated David procedure at two European centers (Bologna University Hospital, Italy and Hannover Medical School, Germany). Only patients who underwent isolated David procedure without any concomitant cardiac surgery 3
were included in this study. The study was in line with both centers’ ethical standards. Patients received either a straight tube graft (David-I, group A, n=103) or a Valsalva graft (group B, n=129). Of note, all operations using a Valsalva graft have been performed by University of Bologna Hospital, while all except one operations with a straight tube graft have been performed by Hannover Medical School.
Surgical technique The surgical techniques of both groups have been described in previous publications (3,7). The detailed surgical technique can be found in the supplemental section.
Anticoagulation therapy and postoperative follow-up After aortic valve-sparing surgery, patients were anticoagulated with Coumadin for the first 3 months only. Thereafter, the patients received a life-long therapy with aspirin unless there were no other indications. We believe that that initial anticoagulation with coumadin helps to prevent thrombembolic events until a neo-intimal layer has formed on the inside of the prosthetic graft. Echocardiography was performed intraoperatively and before discharge. Individual consent was obtained from patients to allow for follow-up examination. Follow-up was done according to common guidelines (16). Patients were contacted by telephone and/or seen in our clinic. Primary care physicians and/or cardiologists were contacted and examination results with echocardiography data were obtained. Primary endpoint of this study was aortic valve-related re-operation. The indication for valverelated re-operation was a failing aortic valve (not aortic stenosis).
Statistical analysis Data analysis was performed using SPSS 22 Statistics software (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.). Normal distribution of variables was analyzed with the Kolmogorov-Smirnov test. Normally 4
distributed continuous variables are stated as mean ± standard deviation, while continuous variables without normal distribution are stated as median + range. Categorical variables are stated as absolute numbers and proportions. Differences in categorical variables were analyzed using the Χ2test or Fisher’s exact test, depending on sample size. Differences in continuous variables were tested using t-test or Mann-Whitney u-test, depending on sample size. Kaplan-Meier analysis was used for evaluation of both survival and re-operation of the aortic valve, and the log-rank test was used to test for differences. Cox regression analysis was performed using a binary logistic regression model to discriminate independent risk factors for aortic-valve related re-operation. All included variables can be found in the univariate anaysis in supplemental table 2. A value of p<0.05 was considered statistically significant.
Results Descriptive analysis The detailed patient demographics can be found in table 1. The mean age of all patients was 47 ± 17 years and 186 (80%) were male. There were more male in group B (85%, n=110) than in group A (74%, n=76, p=0.029). Hyperlipidemia showed a higher incidence in group A (27%, n=28) than in group B (14%, n=18, p=0.012). Nicotine usage was more common in group B (40%, n=52) than in group A (19%, n=20, p=0.001).
All other pre-operative
characteristics were evenly distributed in both groups. Marfan’s syndrome was present in 30% (n=31, group A) and 24% (n=31, group B, p=0.300). Bicuspid aortic valve (BAV) was present in 9% (n=9, group A) and 13% (n=17, group B, p=0.287). In the entire cohort, the indication for surgery was aortic aneurysm in 93% (n=216) of patients and acute aortic dissection type Stanford A with root involvement in 7% (n=16) of patients. There was no significant difference in indication for surgery betwen the two groups (p=0.957). Preoperative echocardiographic data can be found in table 2. The degrees of aortic insufficiency (AI) was distributed mostly equal between the two groups (p=0.180).
5
The peri-operative data is shown in table 3. The mean aortic cross-clamp times were 112±21 in group A and 127±22 min in group B, respectively (p<0.001). There were significantly more cusp repairs in group B (n=28, 22%) compared to group A (n=4, 4%, n<0.001) (see also supplemental table 1). The 30-day mortality rate was 1% (n=1) in group A and 2% (n=2, p=0.698) in group B. The overall stroke rate was 1% (n=2) and was not significantly different in both groups (p=0.204). Postoperative echocardiography was performed for 93% (n=215, table 4) of all patients. Echocardiography showed AI≥II° in 0% (n=0) of group A and 17% (n=21) of group B (p<0.001).
Follow-up results Follow-up echocardiography was obtained for 66% (n=153) of all patients. The mean echocardiographic follow up time was 5.44 ± 3.2 years for group A and 5.5 ± 4.1 years for group B. Follow-up echocardiography showed AI ≥ II° in 22% (n=15) of group A and 39% (n=33) of group B (p<0.026, table 4). The overall rate for aortic valve-related re-operation was 10% (n=24) in the entire cohort (table 4). The overall rates for aortic-valve-related reoperation were 8% (n=8) in group A and 13% (n=16) in group B (p=0.241). The follow-up was complete for 100% of patients and comprised a total of 1530 patient-years. The mean follow up time was 5.8 ± 3.5 years for group A and 7.2 ± 4.3 years for group B. The Kaplan Meier curve for freedom from aortic-valve related re-operation is given in figure 1. In group A, the estimates for freedom from valve-related re-operation at 1, 5 and 10 years after initial surgery were 100%, 92% and 86%, respectively. In group B, the 1, 5 and 10-year estimates for freedom from valve-related re-operation were 99%, 93% and 87%, respectively. There was no significant difference between the two groups (p=0.420). The reasons for reoperation in these patients were endocarditis in two patients and a failed aortic valve in six patients of group A. In group B, one patient had endocarditis and fourteen had a failed aortic valve.
6
The Kaplan Meier curve for survival is shown in figure 2. The survival estimates at 1, 5 and 10 years were 99%, 99% and 85% in group A. The respective estimates for group B were 98%, 96% and 93%. There was no significant difference between the two groups (p=0.799).
Risk factor analysis We performed a risk factor analysis for aortic valve-related re-operation. The results of the univariate analysis can be found in the supplemental section (supplemental table 2). While univariate analysis showed that BAV (p=0.004), graft type (p=0.009), and center (p=0.010) were significantly associated with aortic valve-related re-operation, this was not confirmed during logistic Cox regression analysis. Logistic Cox regression analysis showed that BAV (OR=3.435, 95%-CI=1.459 – 8.083, p=0.005) and post-operative AI ≥ II° (OR=5.988, 95%CI=2.545 – 14.088, p<0.001) were risk factors for re-operation of the aortic valve (table 5).
Subgroup analysis We performed a subgroup analysis and excluded patients with BAV, those who underwent cusp repair, and/or those who had postoperative AI ≥ II°, in order to exclude patients who could potentially bias the the outcome during follow-up. The detailed results can be found in the supplemental section. The majority of the patient demographics were distributed equally among the two subgroups (supplemental table 3). The preoperative echocardiographic data showed that the straight tube graft subgroup showed a higher rate of AI III° and IV° (p=0.046, supplemental table 4). The early postoperative outcome was mostly equal between the two subgroups (supplemental table 5). During follow-up, the incidence of aortic valve-related reoperation was comparable between the straight tube subgroup (n=5, 5%) and the Valsalva subgroup (n=4, 5%, p=0.841). We have also performed a risk factor analysis for aortic valverelated re-operaton, and no risk factor was identified.
Comment
7
Valve-sparing aortic root replacement with a straight tube graft (David I) has been criticized for its potentially suboptimal hemodynamics when compared to grafts with preformed sinus of Valsalva (11-13). In this study, we have compiled the data of two experienced European centers, and aimed at analyzing the outcome of valve-sparing aortic root replacement using both straight tube grafts and Valsalva grafts. In consideration of the study design’s potential inherent limitations, we could demonstrate that the usage of a straight tube graft does not compromise the short- and mid-term outcome after valve-sparing aortic root replacement and is not inferior when compared to Valsalva grafts.
Early postoperative outcome In this study, all operations with a Valsalva graft have been performed by University of Bologna Hospital, while all except one operations with a straight tube graft have been performed by Hannover Medical School. In both centers and with both graft types, the early postoperative results showed a low and comparable rate of complications. The overall 30-day mortality was 1%, which represents a reasonable number for an aortic (valve-sparing) root replacement. The 30-day mortality rate of our study is similar to the results of other reports or even lower (3,6,9,17). Both the rate for permanent neurological deficit and for acute kidny injury requiring dialysis were 1%, respectively, which represent low rates as well. As the rates for perioperative mortality and morbidity were very low in our cohort and as we did not observe any statstically significant difference between the two groups, we conclude that both techniques (straight tube graft and Valsalva graft) are safe and feasible approaches for valvesparing aortic root replacement.
The postoperative echocardiography results showed significantly more patients with AI ≥ II° in group B (Valsalva graft) than in group A (straight tube graft). This difference can directly be contributed to the different hospital approaches on the tolerance of postoperative AI. At University of Bologna Hospital, where the Valsalva grafts have been used, it is the policy to tolerate AI II°. Only if there is AI II-III or more on postoperative echocardiography, the aorta 8
is reclamped and the aortic valve is revised. In contrast, at Hannover Medical School, it is the policy to not tolerate AI II°. At this center, the policy is to achieve AI ≤ I° and if there is a higher degree of AI, the aorta is reclamped and the aortic valve reconstruction is revised. Second, the rate of cusp repair was significantly higher at University of Bologna Hospital. The more aggressive approach towards cusp repair represents another strategic difference between the two participating centers. Achieving an adequate results with a structurually not ideal aortic valve is more challenging. These differences have to be kept in mind when reviewing the early postoperative results, especially the echocardiographic data.
Follow-up During follow up, we observed no statistically significant difference between the mid-term surival of patients who received a straight tube graft and those who received a valsalva graft. Furthermore, patients who underwent aortic valve-sparing root replacement with a straight tube graft did not have a higher rate of failing aortic valves than patients who received a Valsalva graft. The rates for reoperation of the aortic valve were similar in both groups and there was no statistically significant difference. The slightly higher rate of aortic valve-related re-operation in the Valsalva group (although not statistically significant) might again be explained by the fact that only two centers were involved in this study, involving their different (peri-)operative approaches as mentioned earlier. The rate for cusp repair was significantly higher in patients who received a Valsalva graft. It has been previously described that cusp repair in the setting of aortic valve-sparing root replacement is a risk factor for late adverse outcome (18). At Bologna University, during the early experience cusp repair was performed when necessary as: decalcification of the leaflets, reinforcement of the free edge or plication. However, these techniques have been progressively abandoned, due to the worse results that we observed in our clinical practice, especially in bicuspid aortic valve. This can be regarded as an institutional learning curve. The difference in the rate of cusp repair could explain the higher rate for valve-related reoperation. 9
Furthermore, the higher rate of cusp repairs in the Valsalva group could also explain the higher rate of postoperative AI ≥ II°. As mentioned earlier, there were significantly more cusp repairs performed in the Valsalva group, indicating more complex and challenging repair. Since more cusp repairs were performed in the Valsalva group, and since these operations have been performed at only one institution where it is the policy to accpet AI II°, this could also predispose to potentially more reoperations in the future. This institutional difference on approaching aortic valve repair with its potential for bias has to be kept in mind when reviewing the follow-up data. In the Kaplan Meier curve for valve-related re-operation, there seems to be an evolving divergence at approximately 7 years. This raises concern for emerging divergence between the two groups during the future follow up time. However, the follow up rate for 7+ years is only 12% for group A and 27% for group B, therefore we cannot comment on this issue with certainty. Future studies with a longer follow-up time will be needed to investigate this issue. We conclude that – even if the Valsalva graft may provide better hemodynamics (10-15) – the usage of a straight tube graft for David procedure does not compromise the mid-term performance of the aortic valve and is not inferior to the Valsalva graft.
Risk factor analysis During univariate analysis, we found that the variables BAV, graft type, and center were associated with aortic valve related reoperation. However, during logistic Cox regression analsyis this was not confirmned for all of these three variables. Instead, we found that the variables BAV and postoperative AI ≥ II° were the only two risk factors for aortic valverelated re-operation. This difference between the uni- and multivariate analysis can be explained by the fact that the Valsalva graft implantations have been performed at one center, with a higher rate of postoperative AI. As outlined before, this is due to the higher rate of cusp repair and the hospital policy on the acceptance of AI. Therefore, one has to assume that the variables graft type and center are confounding variables, and only the variable postoperative AI ≥ II° is a true risk factor for future re-operation. 10
In addition to AI ≥ II°, our logistic Cox regression model identified BAV to be a risk factor for aortic valve-related re-operation. The current literature on the fate of a BAV after valvesparing aortic root replacement is not completely consistent. On the one hand, there are articles reporting an adequate outcome after aortic valve-sparing root replacement and a BAV (19). On the other hand, there are studies that observed a higher rate of late adverse outcome after repair of BAV (18).
Since there was concern that the higher rate of postoperative AI ≥ II° in the Valsalva group could potentially bias the analysis, we performed a subgroup analysis and excluded all patients who had BAV, those who underwent cusp repair, and/or those who had postoperative AI ≥ II°. By these means, we were aiming to eliminate all patients who had already a suboptimal aortic valve directly postoperatively. During follow up, we observed a comparable rate of aortic valve-related reoperations between the two groups. We did not find a statistically significant difference. These results corroborate that the two graft types are indeed equal and do not have an impact on the durability of the aortic valve reconstruction.
Limitations One limitation of this study is its retrospective character. Furthermore, there might be selection bias, since there was no random assignment to the two groups like in a randomized trial. However, we aimed at compensating for this issue by including only isolated David procedures and excluded cases with concomitant procedures to prevent bias from additional extensive cardiac procedures. Although our follow up was 100% complete, the completion rate for echocardiographic data was only 66%. This can be regarded as another limitation, and is contributed to the retrospective character of this study with the first operations having been performed more than 15 years ago. Although we have contacted all patients, we were not able to obtain echocardiographic data from all of them. This issue could be addressed in future prospective and randomized studies. 11
Another limitation of this study is the fact that all operations with a Valsalva graft have been performed by University of Bologna Hospital, while all except one operations with a straight tube graft have been performed by Hannover Medical School. We acknowledge that there might be potential institutional bias, including the different institutional strategies on how to approach aortic valve repair and tolerate the degree of postoperative AI. However, both centers have a significant expertise with the David procedure and perform this operation frequently. Indeed, our logistic Cox regression analysis confirmed that the center is not associated with adverse outcome. Furthermore, the operations have not been performed by a single surgeon but multiple surgeons per center.
Conclusions Regardless of the underlying pathology, aortic valve-sparing David procedure is a safe operation with acceptable mid-term outcome. Our results show that the David I procedure with a straight tube graft is not inferior to those performed with a Valsalva prosthesis.
12
References
1. David TE, Feindel CM. An aortic valve-sparing operation for patients with aortic incompetence and aneurysm of the ascending aorta. J Thorac Cardiovasc Surg. 1992; 103(4):617-21.
2. David TE, Armstrong S, Ivanov J, et al. Results of aortic valve-sparing operations. J Thorac Cardiovasc Surg. 2001; 122:39-46.
3. Shrestha M, Baraki H, Maeding I, et al. Long-term Results after aortic valve-sparing operation (David I). Eur J Cardiothorac Surg. 2012; 41: 56-62.
4. Beckmann E, Martens A, Alhadi FA, Ius F, Koigeldiyev N, Fleissner F, et al. Is Bentall Procedure Still the Gold Standard for Acute Aortic Dissection with Aortic Root Involvement? Thorac Cardiovasc Surg. 2016; 64(2):116-23.
5. Beckmann E, Martens A, Pertz J, et al. Valve-sparing David I procedure in acute aortic type A dissection: a 20-year experience with more than 100 patients. Eur J Cardiothorac Surg. 2017; 52(2):319-324.
6. Di Bartolomeo R, Pacini D, Martin-Suarez S, et al. Valsalva prosthesis in aortic valve sparing operations. Interact Cardiovasc Thorac Surg. 2006; 5:294-298.
7. Pacini D, Settepani F, De Paulis R, et al. Early results of valve-sparing reimplantation procedure using the valsalva conduit: a multicenter study. Ann Thorac Surg. 2006; 82:865-72.
13
8. Cameron DE, Alejo DE, Patel ND, et al. Aortic root replacement in 372 Marfan patients: evolution of operative repair over 30 years. Ann Thorac Surg. 2009; 87(5):1344-9.
9. De Paulis R, Scaffa R, Nardella S, et al. Use of the Valsalva graft and long-term follow-up. J Thorac Cardiovasc Surg. 2010; 140(6):S23-7.
10. Robicsek F, Thubrikar MJ. Role of sinus wall compliance in aortic leaflet function. Am J Cardiol. 1999; 944-946.
11. Leyh RG, Schmidtke C, Sievers CC, et al. Opening and closing characteristics of the aortic valve after different types of valve-preserving surgery. Circulation. 1999; 100:21532160.
12. Fries R, Graeter T, Aicher D, et al. In vitro comparison of aortic valve movement after valve-preserving aortic replacement. J Thorac Cardiovasc Surg. 2006, 132:32-7.
13. Schmidtke C, Sievers HH, Frydrychowicz A, et al. First clinical results with the new sinus prosthesis used for valve-sparing aortic root replacement. Eur J Cardiothorac Surg. 2013; 43:585-590.
14. De Paulis R, De Matteis GM, Nardi P, et al. A new aortic Dacron conduit for surgical treatment of aortic root pathology. Ital Heart J. 2000; 1:457-463.
15. De Paulis R, De Matteis GM, Nardi P, et al. Analysis of valve motion after the reimplantation type of valve-sparing procedure (David I) with a new aortic root conduit. Ann Thorac Surg. 2002; 74:53-57.
14
16. Akins CW, Miller DC, Turina MI, et al. Guidelines for reporting mortality and morbidity after cardiac valve interventions. Eur J Cardiothorac Surg. 2008; 33(4):523-528.
17. David TE, Feindel CM, Webb GD, et al. Aortic valve preservation in patients with aortic root aneurysm: results of the reimplantation technique. Ann Thorac Surg. 2007; 83(2):S732-5.
18. Esaki J, Leshnower BG, Binongo JN, et al. Risk factors for late aortic valve dysfunction after David V valve-sparing root replacement. Ann Thorac Surg. 2017; 104(5):1479-1487.
19. Bavaria JE, Desai N, Szeto WZ, et al. Valve-sparing root replacement and leaflet repair in a bicuspid aortic valve: comparison with the 3-cusp David preocedure. J Thorac Cardiovasc Surg. 2015; 149(2 Suppl):S22-8.
15
Table 1: Patient characteristics Entire cohort
Straight tube
Valsalva
p-value
n=232
n=103
n=129
Age (years)
47 ± 17
47 ± 17
47 ± 17
0.916
Male (n, %)
186 (80%)
76 (74%)
110 (85%)
0.029
Height (cm)
176 ± 20
176 ± 27
176 ± 12
0.954
Weight (kg)
81 ± 14
82 ± 16
80 ± 13
0.568
BMI (kg/m2)
26 ± 5
25 ± 6
26 ± 5
0.085
Marfan’s syndrome (n, %)
62 (27%)
31 (30%)
31 (24%)
0.300
Bicuspid aortic valve (n, %)
26 (11%)
9 (9%)
17 (13%)
0.287
Arterial hypertension (n, %)
137 (59%)
62 (60%)
72 (56%)
0.752
Hyperlipidemia (n, %)
46 (20%)
28 (27%)
18 (14%)
0.012
Diabetes mellitus (n, %)
7 (3%)
3 (3%)
4 (3%)
0.934
Nicotine (n, %)
72 (31%)
20 (19%)
52 (40%)
0.001
Re-do (n, %)
5 (2%)
4 (4%)
1 (1%)
0.105
Maximum aortic diameter (mm)
53 ±7
54 ± 8
52 ± 6
0.003
Aortic aneurysm (n, %)
216 (93%)
96 (93%)
120 (93%)
0.957
Acute type A dissection (n, %)
16 (7%)
7 (7%)
9 (7%)
0.957
Characteristic
Indication for surgery:
Abbreviations for table 1: BMI = body mass index
16
Table 2: Pre-operative echocardiographic data Entire cohort
Straight tube
Valsalva
n=232
n=103
n=129
AI 0° (n, %)
27 (12%)
10 (10%)
17 (13%)
AI I° (n, %)
47 (20%)
21 (20%)
26 (20%)
AI II° (n, %)
69 (30%)
27 (26%)
42 (33%)
AI III (n, %)
70 (30%)
39 (38%)
31 (24%)
AI IV° (n, %)
19 (8%)
6 (6%)
13 (10%)
Characteristic
p-value
0.180
Abbreviations for table 1: AI = aortic insufficiency
17
Table 3: Peri-operative data Entire cohort
Straight tube
Valsalva
p-value
n=232
n=103
n=129
159 ± 36
152 ± 30
0.128
Characteristic Cardiopulmonary
bypass
time 155 ± 33
(min) Aortic cross-clamp time (min)
121 ± 23
112 ± 21
127 ± 22
<0.001
Cusp repair (n, %)
32 (14%)
4 (4%)
28 (22%)
<0.001
ICU stay (d)
1 (0-17)
1 (0-17)
1 (0-6)
0.486
Rethoracotomy (n, %)
10 (4%)
3 (3%)
7 (5%)
0.349
New dialysis (n, %)
2 (1%)
0 (0%)
2 (2%)
0.204
Permanent neurological deficit (n,
2 (1%)
0 (0%)
2 (2%)
0.204
3 (1%)
1 (1%)
2 (2%)
0.698
%) Perioperative mortality (n, %)
Abbreviations for table 3: ICU = intensive care unit
18
Table 4: Early and late postoperative aortic valve-related outcome
Entire cohort
Straight tube
Valsalva
n=215
n=94
n=121
AI < II° (n, %)
194 (90%)
94 (100%)
100 (83%)
AI ≥ II° (n, %)
21 (10%)
0 (0%)
21 (17%)
Follow up echocardiography
n=153
n=68
n=85
AI < II° (n, %)
105 (69%)
53 (78%)
52 (61%)
AI ≥ II° (n, %)
48 (31%)
15 (22%)
33 (39%)
Re-operation
n=232
n=103
n=121
Aortic valve-related re-operation
24 (10%)
8 (8%)
16 (13%)
Postoperative
p-value
echocardiography <0.001
0.026
0.241
(n, %)
Abbreviations for table 4: AI = aortic insufficiency
19
Table 5: Risk factor analysis for aortic valve-related re-operation. Factor
p-value
Odds ratio
95%-CI
Bicuspid valve
0.005
3.435
1.459 – 8.083
Postoperative AI ≥ II°
<0.001
5.988
2.545 – 14.088
Abbreviations for table 5: 95%-CI = 95% confidence intervall, AI = aortic insufficiency
20
Figure Legends
Figure 1: Freedom from aortic valve-related re-operation. Kaplan Meier curves for freedom from valve-related re-operation including all patients who underwent isolated David procedure with either a straight tube graft or a valsalva graft. Time origin on x-axis denotes day of surgery.
Figure 2: Survival. Kaplan Meier survival curves for patients who underwent isolated David procedure with either a straight tube graft or a valsalva graft. Time origin on x-axis denotes day of surgery.
21