Differences in recurrence rate of aortic regurgitation (AR) and extent of left ventricular (LV) remodeling across the different surgical options in patients operated for type A aortic dissection remain unknown. The present evaluation compared the AR recurrence rate and changes in LV volumes and systolic function in valve-sparing aorta replacement (VSAR), supracoronary ascending aorta replacement (SCAR), and aortic valve and aorta replacement (AVAR). A total of 97 patients (58 ± 12 years, 62% men) with acute type A aortic dissection who underwent VSAR (n = 24), SCAR (n = 43), or AVAR (n = 30) were evaluated. Changes in LV volumes and function between postoperative and follow-up were compared using linear mixed models. Postoperative AR grades were not significantly different between groups. However, after median follow-up of 47 months, AR grade ≥2 was significantly more often observed in SCAR (55%) and VSAR (25%) compared to AVAR (0%, p <0.001). LV volumes remained stable in VSAR and AVAR but increased significantly in SCAR (LV end-diastolic volume: from 99 ± 4 to 131 ± 6 ml; p <0.001; LV end-systolic volume: from 49 ± 3 to 66 ± 5 ml; p = 0.002). In patients with recurrent AR grade ≥2 at follow-up, LV volumes increased, whereas patients without recurrent AR did not show significant LV dilatation. In conclusion, patients with acute type A aortic dissection who underwent SCAR or VSAR showed more frequently AR grade ≥2 recurrence compared to AVAR. However, only patients who underwent SCAR experienced adverse LV remodeling at follow-up. Recurrence of AR grade ≥2 was associated with adverse LV remodeling.
Acute type A aortic dissection is a life-threatening condition with 50% mortality within the first 48 hours if not operated. Resection of the primary intimal tear, stabilization of the aortic wall and prevention of aortic rupture are the surgical goals and can be achieved by performing a valve-sparing aorta replacement (VSAR), supracoronary ascending aorta replacement (SCAR), or aortic valve and aorta replacement (AVAR). Previous studies showed no difference in perioperative and midterm survival between these surgical procedures. However, SCAR is associated with dilatation of the aortic sinuses and recurrence of aortic regurgitation (AR) at follow-up which may warrant a relatively high risk on reoperation. Furthermore, recurrence of AR at follow-up may lead to left ventricular (LV) dilation and systolic dysfunction. However, the effects of the type of surgery for acute type A aortic dissection on LV volumes and function during follow-up have not been evaluated. The aim of the present study was to assess differences in LV remodeling during follow-up for the several surgical procedures in patients with acute type A aortic dissection taking into consideration the differences in AR recurrence rates.
Methods
Patients with acute type A aortic dissection who underwent surgery at the Leiden University Medical Center from July 1, 1994, to July 1, 2013, and who survived the initial hospitalization were evaluated. Patients were included if postoperative transthoracic echocardiography was available. Ninety-seven patients were divided into 3 groups according to the surgical procedure performed: VSAR (n = 24), SCAR (n = 43), or AVAR (n = 30). Patients with connective tissue disease were excluded.
Clinical and surgical characteristics were prospectively collected in the departmental Cardiology Information System (EPD-Vision; Leiden University Medical Center, Leiden, The Netherlands) and retrospectively analyzed. LV volumes and function were evaluated with 2-dimensional transthoracic echocardiography postoperatively and during follow-up (≥6 months after surgery, available in 53 patients). The institutional ethical committee approved this retrospective study. Written informed consent was obtained when applicable (a waiver was obtained for retrospective analysis of clinically acquired data at the Leiden University Medical Center. For analysis of data acquired in other centers, patient written informed consent was requested). Changes in LV volumes and function over time were assessed and compared between the 3 different surgical procedures. In addition, the incidence of recurrent AR over time was assessed.
Transthoracic echocardiography was performed with commercially available ultrasound systems (Vivid 7, E9 or System 5; General Electric Healthcare, Vingmed, Horten, Norway) equipped with 3.5-MHz or M5S transducers. The echocardiographic data were digitally stored in cine-loop format, and data analysis was retrospectively performed using EchoPac (112.0.1; GE Medical Systems, Horten, Norway). LV volumes were quantified at end-diastole and end-systole in the apical 2- and 4-chamber views using the Simpson biplane method, and LV ejection fraction was calculated. AR grade was assessed using a multiparametric approach that included the measurement of the jet width relative to the LV outflow tract width, vena contracta, and/or the pressure halftime of the regurgitant flow (if feasible) according to current recommendations.
The decision to perform VSAR, SCAR, or AVAR was left at the discretion of the surgeon on duty. During VSAR, the native sinuses of Valsalva were resected, and a graft was implanted using the reimplantation technique (modified David procedure, n = 19) or the remodeling technique (Yacoub technique, n = 5), as previously described. Concomitant procedures (leaflet triangular resection, leaflet resuspension, and plication of the free edge of the leaflet) were performed if needed. For SCAR, the ascending aorta was resected until the sinotubular junction and replaced by a Hemashield tubular graft. If necessary, resuspension of the commissures (n = 15) and/or restoration of the sinuses of Valsalva using bioglue (n = 24) or gelatin-resorcin-formalin glue (n = 5) was performed. During AVAR, the native sinuses of Valsalva and valve were excised and replaced by either a biologic (n = 18) or mechanical prosthesis (n = 12). In every patient, the distal ascending aorta and arch were inspected under deep hypothermic circulatory arrest. If a (re)entry tear was present in the arch, concomitant (hemi-)arch replacement was performed.
All patients underwent transthoracic echocardiography postoperatively before discharge. Transthoracic echocardiography at follow-up was performed at the discretion of the treating cardiologist. Follow-up echocardiography was available in 53 patients and was included in the present study when it was performed at least 6 months after surgery. The median echocardiographic follow-up duration was 47 months (interquartile range 18 to 76 months) and comparable between the 3 groups (VSAR 49 months, interquartile range 19 to 74 months; SCAR 55 months, interquartile range 31 to 77 months; AVAR 24 months, interquartile range 12 to 56 months; p = 0.150).
Data analysis was performed using SPSS software version 20 (SPSS, Chicago, Illinois). Continuous variables were reported as mean ± standard deviation or median and interquartile range (IQR) when appropriate. Categorical variables were reported as numbers and percentages. Differences between the 3 different surgical procedures were analyzed using analysis of variance test, the Kruskal–Wallis test, or the chi-square test. Survival and freedom from reoperation were analyzed using the Kaplan–Meier curves, and differences among surgical procedures were assessed with the log-rank test. Linear mixed model analysis was used to assess the differences in change in LV volumes and LV ejection fraction over time among the groups. Type of surgery (VSAR, SCAR, or AVAR) and timing of transthoracic echocardiography (postoperative or late follow-up) were incorporated in the model as fixed variables as well as the interaction between type of surgery and timing of transthoracic echocardiography. An unstructured covariance matrix was applied. The estimated marginal mean ± standard error of the mean was presented. Post hoc analyses were performed using the Bonferroni test to correct for multiple comparisons. Subgroup analysis was performed to compare LV remodeling in patients with and without recurrent AR grade ≥2. All statistical tests were 2 sided. A p value<0.05 was considered statistically significant.
Results
A total of 97 patients (mean age 58 ± 12 years, 62% men) who underwent emergent surgery for acute type A aortic dissection and survived the index hospitalization were evaluated. Table 1 provides the baseline clinical and surgical characteristics of the patients. Patients who underwent VSAR were significantly younger and more often men than patients who underwent SCAR or AVAR. Hypertension was more often present in patients who underwent SCAR compared to patients treated with VSAR or AVAR. The EuroSCORE II was slightly greater in patients who underwent AVAR compared to VSAR and SCAR. In SCAR, the cardiopulmonary bypass and aortic cross clamp times were significantly shorter compared to VSAR and AVAR.
| VSAR (n=24) | SCAR (n=43) | AVAR (n=30) | p-value | |
|---|---|---|---|---|
| Age (years) | 50±7 | 62±11 | 58±14 | <0.001 |
| Men | 20 (83%) | 22 (51%) | 18 (60%) | 0.033 |
| Diabetes mellitus | 0 | 0 | 2 (7%) | 0.087 |
| Hypertension | 6 (25%) | 26 (60%) | 12 (40%) | 0.018 |
| Dyslipidemia | 0 | 4 (9%) | 2 (7%) | 0.374 |
| Critical preoperative state | 0 | 2 (5%) | 2 (7%) | 0.460 |
| EuroSCORE II (%) | 4.7 (4.1-6.4) | 5.3 (3.4-7.2) | 6.0 (4.8-8.0) | 0.069 |
| Bicuspid aortic valve | 1 (4%) | 0 | 6 (20%) | 0.004 |
| CPB time (minutes) | 267±75 | 191±48 | 253±66 | <0.001 |
| AoX time (minutes) | 209±66 | 119±38 | 178±48 | <0.001 |
| Aortic (hemi-)arch replacement | 14 (58%) | 16 (37%) | 15 (50%) | 0.299 |
| Mitral valve surgery | 0 | 1 (2%) | 0 | 0.530 |
| Coronary bypass | 0 | 1 (2%) | 2 (7%) | 0.345 |
The 5-year survival rate in this cohort was 91 ± 4% and was not significantly different between the surgical procedures (VSAR 100%, SCAR 90 ± 6%, AVAR 82 ± 10%; log-rank p = 0.653; Figure 1 ). Reoperation at follow-up on the proximal and/or distal thoracic aorta was performed in 2 patients who underwent VSAR, 8 SCAR, and 4 AVAR. The freedom from reoperation on the proximal and or distal aorta after 5-year follow-up was 86 ± 5% and comparable between the groups (VSAR 95 ± 5%, SCAR 83 ± 8%, AVAR 77 ± 13%; log-rank p = 0.516; Figure 1 ). However, when considering only reoperation on the proximal aorta, aortic valve replacement was performed in 2 and 7 patients treated initially with VSAR and SCAR, respectively, whereas none of the patients treated with AVAR required reoperation of the proximal aorta. Reasons for reoperation were severe recurrent AR in 7 patients, dilatation of the sinuses of Valsalva without AR in 1 patient, and aortic valve stenosis in 1 patient. Therefore, the 5-year freedom from proximal reoperation after SCAR (88 ± 7%) was slightly less favorable compared with VSAR and AVAR (95 ± 5% and 100%, respectively, log-rank p = 0.060; Figure 1 ).
The prevalence of significant AR directly postoperatively and during follow-up is displayed in Figure 2 . Postoperative AR grade ≥2 was present in 13% of patients who underwent VSAR compared to 8% in patients who underwent SCAR and 4% of patients who underwent AVAR (p = 0.136). In contrast, at long-term follow-up, there was a significant difference in AR grade between the surgical procedures: in patients who underwent VSAR or SCAR, AR grade ≥2 was observed in 25% and 55% of patients, respectively, whereas none of the patients who underwent AVAR showed AR grade ≥2 (p <0.001).
The immediately postoperative LV end-diastolic volume, LV end-systolic volume, and LV ejection fraction were comparable among the 3 groups ( Figure 3 ). However, there was a significant difference in the LV end-diastolic volume and LV end-systolic volume at late follow-up in the surgical procedures. In the VSAR group, the LV end-diastolic volume (108 ± 9 vs 105 ± 9 ml; p = 0.756) and LV end-systolic volume (54 ± 7 vs 47 ± 6 ml; p = 0.387) remained stable. In contrast, the LV end-diastolic volume increased during follow-up in SCAR (99 ± 4 vs 131 ± 6 ml; p <0.001). The LV end-systolic volume also increased significantly in the SCAR group from 49 ± 3 to 66 ± 5 ml (p = 0.002). After AVAR, the volumes remained stable. The LV ejection fraction tended to improve in patients who underwent VSAR (53 ± 2% vs 57 ± 2%; p = 0.074), whereas it remained stable in both SCAR (52 ± 1% vs 51 ± 2%; p = 0.546) and AVAR (52 ± 1% vs 54 ± 2%; p = 0.489). The group–time interaction effect on LV end-diastolic volume (p = 0.008) and LV end-systolic volume (p = 0.018) indicated a significant effect of the type of surgery on the change in LV volumes over time.
