Gender Differences in Mortality Following Tanscatheter Aortic Valve Replacement (TAVR): a Single-Centre Retrospective Analysis from China.
Qi Liu,Yali Wang,Yijian Li,Tianyuan Xiong,Fei Chen,Yuanweixiang Ou,Xi Wang,Yijun Yao,Kaiyu Jia,Yujia Liang,Xin Wei,Xi Li,Yong Peng,Jiafu Wei,Sen He,Qiao Li,Wei Meng,Guo Chen,Wenxia Zhou,Mingxia Zheng,Xuan Zhou,Zhengang Zhao,Chen Mao,Feng Yuan
DOI: https://doi.org/10.1097/cm9.0000000000002779
IF: 6.133
2023-01-01
Chinese Medical Journal
Abstract:To the Editor: Transcatheter aortic valve replacement (TAVR) has become an established treatment for patients with severe aortic stenosis (AS) at all levels of surgical risk.[1,2] With the exploration of TAVR and technological developments, TAVR is an alternative for patients with intermediate or low surgical risk.[3] There are some studies regarding gender differences in TAVR, but there are few studies on it in Asia.[4] However, the studies of outcomes based on gender differences in TVAR are still lacking, especially in China and other Asian countries. Patients' data were collected in the West China Hospital of Sichuan university Transcatheter Aortic Valve Replacement (WATCH TAVR) registration study (registered number: ChiCTR2000033419). The whole study was approved by the Ethics Committee of Sichuan University West China Hospital (No. 2020-470). This study was retrospective and the patients' informed consent was waived. We consecutively enrolled patients with symptomatic severe AS at Sichuan University West China Hospital from April 2012 to April 2019. All enrolled patients were treated with TAVR. Our multidisciplinary heart team discussed the indication for TAVR in all cases. The primary endpoint was all-cause mortality at 1 year, 3 years, and over 3 years. We also investigated stroke, permanent pacemaker implantation (PPI), and other major adverse cardiac events (MACEs). The endpoints were adjudicated using the standardized VARC-2 criteria.[5] Continuous variables are described as median (interquartile range [IQR]). Categorical variables are expressed as frequency (percentage). Continuous variables were compared through Student's t test or the nonparametric Mann–Whitney U test, as appropriate, and categorical variables were compared through the χ2 test or Fisher's exact test, as appropriate. The analyses of time-to-first events were described using Kaplan–Meier curves and were compared between the groups with the log-rank test. Logistic regression analyses were imputed for the univariate analysis to find predictors for outcomes. All computations relied on commercially available software STATA (v16.0 for Windows; StataCorp LLC., Texas, USA), with statistical significance set at two-tailed P <0.05. A total of 462 patients were included, and 18 were excluded. A total of 444 patients (mean age: 73.9 ± 6.3 years) were finally included in the analysis, with a median follow-up of 1.88 (IQR: 1.03–3.75) years. Fifty-five percent (n = 243) of the patients were male, and 45% (n = 201) were female. Most of the patients underwent TAVR through a transfemoral access via an arterial puncture. The prosthesis used was larger in male patients compared with female patients (26.5 mm [IQR: 26.0–29.0] vs. 26.0 mm [IQR: 23–36], P <0.001). The transcatheter approach and prosthesis type showed no statistically significant difference. The female patients had more vascular complications compared with male patients (29 [14.43%] vs. 13 [5.35%], P = 0.001). Intraprocedural bleeding occurred in 47 (10.63%) patients but happened more in the female patients (19 [7.88%] vs. 28 [13.93%], P = 0.040). Paravalvular leak happened in 60.81% (270/444) patients, which was not statistically significant different between the genders (P = 0.352). A total of 80 (80/444, 18.18%) patients had permanent pacemaker implantation, but no statistically significant difference in gender was detected. Notably, 69 patients died during the follow-up. The one-year mortality and one-to-three-year mortality were not statistically significant different between the genders. However, the long-term (over three years) mortality of males was higher than that of females (10 [4.12%] vs. 1 [0.5%], P = 0.015). Other clinical outcomes, including post-TAVR rehospitalization, cardiac rehospitalization, new-onset left bundle branch block (LBBB), neurological symptoms, thrombotic event, arrhythmia, stroke, and myocardial infarction, showed no statistically significant difference in our study [Table 1]. Table 1 - Procedural Details and Clinical Outcomes of patients underwent TAVR. Characteristics Total (n = 444) Male (n = 243) Female (n = 201) P values Puncture 391 (88.06) 214 (88.07) 177 (88.06) 0.998 Predilation 181 (40.77) 94 (38.68) 87 (43.28) Post-dilation 22 (4.95) 12 (4.94) 10 (4.98) Contrast agent dose, mL 240 (230–330) 240 (220–300) 265 (230–330) Prothesis size, mm 26.0 (26.0–29.0) 26.5 (26.0–29.0) 26.0 (23.0–26.0) <0.001 Transcatheter approach 0.849 TFA 441 (99.32) 242 (99.59) 199 (99) TCA 2 (0.45) 0 2 (1) TSA 1 (0.23) 1 (0.41) 0 Prothesis type 0.066 CoreValve 40 (9.01) 25 (10.29) 15 (7.46) Venus A 245 (55.18) 141 (58.02) 104 (51.74) Venus–A Plus 23 (5.18) 11 (4.53) 12 (5.97) Lotus 29 (6.53) 13 (5.35) 16 (7.96) Vitaflow Ⅰ 28 (6.31) 14 (5.76) 14 (6.97) Vitaflow Ⅱ 2 (0.45) 1 (0.41) 1 (0.50) Edward S3 9 (2.03) 6 (2.47) 3 (1.49) Edward XT 18 (4.05) 7 (2.88) 11 (5.47) Taurasone 47 (10.59) 24 (9.88) 23 (11.44) Clinical outcomes Vascular complications 42 (9.46) 13 (5.35) 29 (14.43) 0.001 Major vascular complications 12 (2.70) 2 (0.82) 10 (4.98) Minor vascular complications 30 (6.76) 11 (4.53) 19 (9.45) Intra-TAVR bleeding 47 (10.63) 19 (7.88) 28 (13.93) 0.040 Minor bleeding 26 (5.88) 11 (4.56) 15 (7.46) Major bleeding 21 (4.75) 8 (3.32) 13 (6.47) Paravalvular leak 270 (60.81) 143 (58.85) 121 (63.18) 0.352 Mild PVL 165 (37.16) 85 (34.98) 80 (39.80) Moderate PVL 105 (23.65) 58 (23.87) 47 (23.38) Thrombotic event 21 (4.73) 10 (4.12) 11 (5.47) 0.502 Arrythmia 8 (1.80) 5 (2.06) 3 (1.49) 0.656 Stroke 8 (1.80) 4 (1.65) 4 (1.99) 0.786 PPI 80 (18.18) 46 (19.17) 34 (16.91) 0.557 New onset LBBB 127 (28.93) 73 (30.17) 54 (27.41) 0.527 Myocardial infarction 1 (0.23) 0 1 (0.50) 0.271 1-year death 35 (7.88) 20 (8.23) 15 (7.46) 0.765 1-to-3-year death 23 (5.18) 14 (5.76) 9 (4.48) 0.544 Over 3-year death 11 (2.48) 10 (4.12) 1 (0.50) 0.015 Data are presented as n (%) or median (interquartile range). LBBB: Left bundle branch block; PVL: Paravalvular leak; PPI: Permanent pacemaker implantation; TAVR: Transcatheter aortic valve replacement; TCA: Transcarotid approach; TFA: Transfemoral approach; TSA: Trans-subclavian approach. The cumulative survival curve showed no statistically significant difference between the gender after TAVR (P = 0.198, Supplementary Figure 1, https://links.lww.com/CM9/B643). We further analyzed the three-year landmark of cumulative survival curves according to TAVR mortality [Supplementary Figure 2, https://links.lww.com/CM9/B643], showing a statistically significant difference between the gender (P = 0.043). Patients who died within three years after TAVR were excluded from the analysis. Among the enrolled factors based on the characteristics we presented above, being male (HR 9.88, 95% CI: 1.1 to 90.6, P = 0.043) was the only predictor of long-term death for patients who survived three years after TAVR [Supplementary Table 1, https://links.lww.com/CM9/B643]. Our study represented a large single-center cohort of TAVR patients in China. The main findings in our study are as follows: (i) women were more likely to have bleeding events during TAVR but had a similar rate of thrombotic events, arrhythmia, stroke, PPI, and myocardial infarction compared with male patients; (ii) the long-term (over 3 years) mortality rate was higher in men than in women, while the 1-year and 1-to-3-year mortality were not statistically significantly different in the two gender groups; and (iii) there was a similar rate of new LBBB and neurological symptoms between the two gender groups. Detailed baseline characteristics and anatomy parameters are available in the Supplementary Tables 2 and 3, https://links.lww.com/CM9/B643. In this study, we found that women were more likely to have procedural bleeding events. However, the results of TAVR will get better with experience. We grouped patients according to whether they were admitted before 2016. Subgroup analysis showed females had more procedural bleeding only before 2016 (20 [40] vs. 14 [18.92], P = 0.01). And the main cause for bleeding was vascular complications (odds ratio [OR] = 8.39, P <0.001). TAVR complications are closely related to prognosis. Appropriate dealing of TAVR complications is quite important.[6,7] In the survival analysis, thirty-day mortality showed no statistical difference between the gender groups (Male vs. Female: 7 [2.88%] vs. 9 [4.48%], P = 0.396). From the survival curve, we can determine the difference in mortality over the 3-year follow-up with a hazard ratio of 9.88 (P = 0.043), Multivariate Cox regression analysis showed that male was the only predictor for higher mortality over 3 years after TAVR. This finding is similar to some studies in European countries.[8,9] However, the difference might be related to the natural life expectancy of human beings. In China, women have six more years of life expectancy on average than men, according to the World Health Organization (WHO) report.[10] The study has several limitations. First, the basic data was collected retrospectively. Second, this single-center study might not best represent the situations in other parts of China and in other Asian countries. Third, in the early exploration of TAVR, we only used the STS score. A comparison between the STS score and EuroSCORE was not performed. Funding This stydy was supported by grants from the West China Hospital "1·3·5" Discipline of Excellence Project-"Mechanisms of aortic stenosis and the clinical applications", National Natural Science Foundation of China (No.82001899) and the National Natural Science Foundation of China (No.81901825).