Predictors for all-cause mortality in men after transcatheter aortic valve replacement: A report from the LAPLACE-TAVI registry

Abstract

Background

Information regarding the outcomes of transcatheter aortic valve replacement (TAVR) in men is limited. This study aimed to investigate short- to mid-term outcomes and prognostic predictors in this population.

Method and Results

The data of 519 men were analyzed from 1,693 consecutive patients with symptomatic severe aortic stenosis who underwent TAVR at six hospitals between April 2010 and July 2020. The primary endpoint was all-cause mortality at 30 days after TAVR. The mean age and Society of Thoracic Surgeons (STS) score were 83.7 ± 5.9 years and 6.3 ± 4.7%, respectively. Overall, 23.5% of patients consumed alcohol with a frequency of > 1 drinks/week, and 12.1% consumed alcohol with a frequency of > 8 drinks/week, while 66.1% were former smokers and 4.2% were current smokers. Mortality at 30 days was 0.8%. During the median follow-up period of 448 days, the estimated survival rates at 1 year post-TAVR was 90.7 ± 1.4%. In multivariate analysis, the serum albumin level [hazard ratio (HR): 2.20, 95% confidence interval (CI):1.36–3.62, p = 0.001], atrial fibrillation (HR: 1.79, 95% CI: 1.13–2.82, p = 0.012), and STS score (HR: 1.33, 95% CI: 1.06–1.67, p = 0.015) were independently associated with all-cause mortality following TAVR. Adjusted hazard ratios of current smoking, heavy drinking, and presence of cancer were 1.05 (95% CI: 0.36–2.98),1.37 (95% CI: 0.75–2.48), and 1.13 (95% CI: 0.75–2.48), respectively.

Conclusion

Our study demonstrated that serum albumin levels, atrial fibrillation, and STS score were independently associated with all-cause mortality following TAVR in men.

1. Introduction

Sex differences in cardiovascular diseases have been investigated in many studies [1], [2], [3]. Similar to the pathogenesis of vascular atherosclerosis, that of acquired aortic stenosis (AS) predominantly affects men, driven by endothelial damage, inflammation, and fibrosis [4].

In recent decades, transcatheter aortic valve replacement (TAVR) has been associated with increased survival and symptom improvement than those with medical therapy and has become part of the standard of care for patients with severe AS [5], [6]. In many countries, approximately half of the patients undergoing TAVR are older women, and differences in outcomes between sexes have been largely noted [7], [8], [9], [10]. Although perioperative bleeding complications are more common in women in short-term period, no significant differences have been reported in long-term prognosis between sexes [11], [12], [13], [14], [15].

Particularly, the impact of women-specific parameters, such as smaller body size, hormonal changes, and osteoporosis, on the outcomes after TAVR has been discussed in recent studies [16], [17]. However, the impact of men-specific parameters, including smoking and drinking habit, and presence of cancer on the outcomes after TAVR remains unclear. This study aimed to investigate short- to mid-term outcomes and prognostic factors in men after TAVR as a treatment for severe AS.

2. Methods

This study is a retrospective analysis of a prospective multi-center registry database of patients who underwent TAVR at six centers in Japan, termed The aLliAnce for exPloring cLinical prospects of AortiC valvE disease (LAPLACE-TAVI registry) (https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno = R000035566, UMIN000031133). The study population comprised 1,693 patients with severe AS who consulted at the Sakakibara Heart Institute, Juntendo University, Yamagata University, Mie University, Hirosaki University, and Kawasaki Saiwai Hospital between April 2010 and July 2020. All cases were reviewed and deemed appropriate for TAVR by a multidisciplinary team comprising cardiac surgeons, interventional cardiologists, anesthesiologist, and imaging specialists. Patients who presented with at least one of the following criteria did not undergo TAVR according to the local guidelines: asymptomatic state, dialysis, active endocarditis, other active infections, and noncardiac conditions with a resultant estimated life expectancy of < 12 months. Women were excluded due to the design of the study. Moreover, to investigate the impact of men-specific factors, such as smoking and drinking habit on the outcomes of TAVR in this population, 519 of 568 men had sufficient data, and prognostic predictors were analyzed by multivariate analysis. The primary endpoint was all-cause mortality at 30 days after TAVR, and the Society of Thoracic Surgeons (STS) score was calculated to predict the surgical risk. In all patients who had cancer at the time of TAVR, active cancer, defined as disease undergoing treatment or treatment planning concurrent with TAVR or completed within 1 year before TAVR, was also reported. While, 52 (42.2%) were reported to have inactive cancer, and rest of them (n = 34, 27.6%) do not have sufficient information about cancer activity or type of cancer. Informed consent was obtained from all patients. In addition, all patient information was obtained from the medical records or telephone interviews. Therefore, follow-up and patient care were ensured. This study was performed in accordance with the ethical principles of the 1975 Declaration of Helsinki and was approved by the Human Investigation Committee of the Sakakibara Heart Institute (approval number 17–048 received on Nov 28, 2017).

Echocardiographic findings were analyzed by full-time academic echocardiographers in accordance with the American College of Cardiology/American Heart Association guidelines [18]. All endpoints were defined according to the Valve Academic Research Consortium-2 Criteria [19]. TAVR was performed as described in a previous study [5].

Continuous and categorical variables were expressed as means and standard deviations (SD) and as numbers and percentages, respectively. The Shapiro–Wilk test was used to assess whether continuous variables were normally distributed. Statistical significance was defined as a two-sided p < 0.05. Adjusted variables with p < 0.15 determined via univariate analysis were entered into a multivariate Cox regression analysis to determine the influence of outcome. To better understand the impact of sex-specific parameters on outcomes, hazard ratios of smoking, drinking and cancer parameters were adjusted with parameters that were independently associated with the primary endpoint by multivariate analysis. All analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA).

3. Results

The demographic and clinical characteristics of patients at baseline are summarized in Table 1. The mean age and STS score were 83.7 ± 5.9 years and 6.3 ± 4.8, respectively. Of the 519 patients, 75.3%, 27.0%, and 55.7% had hypertension, diabetes, and dyslipidemia, respectively. In addition, 23.5% of patients consumed alcohol with a frequency of > 1 drinks/week, and 12.1% consumed alcohol with a frequency of > 8 drinks/week, while 66.1% were former smokers and 4.2% were current smokers at the time of the TAVR procedure. The proportion of never smokers and no-drinkers tended to increase with increasing age group (Fig. 1, Fig. 2). Of 123 patients with cancers, 37 (30.0%) were reported to have active cancers. Among all patients with cancers, prostate cancer (n = 10/123, 8.1%) was the most common, followed by colon cancer (n = 4/123, 3.2%), liver cancer (n = 4/123, 3.2%), kidney cancer (n = 4/123, 3.2%), lung cancer (n = 4/123, 3.2%), and gastric cancer (n = 4/123, 3.2%). Rest of them were as follows; pancreatic cancer (n = 2, 5.4%), ureter cancer (n = 1, 2.7%), thyroid cancer (n = 1, 2.7%), malignant soft tissue tumor (n = 1, 2.7%), malignant lymphoma (n = 1, 2.7%), and esophageal cancer (n = 1, 2.7%), respectively.

Table 1.

Baseline patient characteristics.

Characteristic (n = 519)
Age (at TAVR), years	83.7 ± 5.9
Body mass index, kg/m2	22.5 ± 3.2
NYHA classification III/IV, %	265 (51.5)
STS score, %	6.30 ± 4.8
Smoking, %
Never	154 (29.6)
former	343 (66.1)
Current	22 (4.2)
Drinking, %
No drinking	334 (64.4)
>1 drinks/week	122 (23.5)
>8 drinks/week	63 (12.1)
Diabetes mellitus, %	140 (27.0)
Hypertension, %	391 (75.3)
Dyslipidemia, %	289 (55.7)
Prior MI, %	50 (9.6)
PAD (ABI < 0.9), %	96 (18.5)
Cancer, %	123 (23.7)
Prior stroke, %	84 (16.2)
Atrial fibrillation, %	129 (24.9)
COPD (FEV1%<70), %	81 (15.6)
Hemoglobin level, g/dL	12.0 ± 1.75
eGFR, mL/(min·1.73 m2)	54.3 ± 18.6
Serum albumin, g/dL	3.74 ± 0.46
Ejection fraction, %	58.7 ± 11.5
AV area, cm2	0.72 ± 0.18
AV mean pressure gradient, mmHg	47.7 ± 16.8
Mitral regurgitation, %
Trace	300 (57.8)
Mild	160 (30.8)
Moderate	55 (10.6)
Severe	4 (0.8)
Tricuspid regurgitation, %
Trace	359 (69.2)
Mild	114 (22.0)
Moderate	36 (6.9)
Severe	10 (1.9)
TRPG, mmHg	26.7 ± 9.8

Values are presented as mean ± SD, n (%). Abbreviations: TAVR, transcatheter aortic valve replacement; NYHA, New York Heart Association; STS, Society of Thoracic Surgeons; MI, myocardial infarction; CABG, coronary artery bypass graft; PAD, peripheral artery disease; ABI, ankle brachial index; COPD, chronic obstructive pulmonary disease; FEV1%, percent predicted forced Expiratory Volume; eGFR, estimated glomerular filtration rate; AV, aortic valve; TRPG, tricuspid regurgitation pressure gradient.

Fig. 1.

Prevalence of drinking habit in men with transcatheter aortic valve replacement The proportion of no-drinkers tended to increase with increasing age group.

Fig. 2.

Prevalence of smoking in men with transcatheter aortic valve replacement. The proportion of never smokers tended to increase with increasing age group.

All-cause mortality at 30 days was 0.8% (Primary endpoint; cause of death). Four patients died within 30 days (two patients with annular rupture, one patient with severe heart failure, and one patient with sudden death just after discharge). The procedural success was 97.2% with an early safety at 30 days of 86.1% (Table 2). During the median follow-up period of 448 days, the post-TAVR survival rates after 1 and 3 years were 90.7 ± 1.4% and 74.3 ± 2.9%, respectively, as estimated by Kaplan–Meier analysis (Fig. 3).

Table 2.

Procedural outcomes.

Outcome (n = 519)
Transfemoral approach, %	487 (93.8)
New generation valve (SAPIEN3/EvolutPro), %	447 (86.1)
Procedure time, minutes	85.5 ± 56.1
Radiation time, minutes	22.6 ± 10.5
Contrast volume, mL	68.3 ± 46.5
General anesthesia, %	214 (41.2)
Device success, %	506 (97.5)
Stay in intensive care unit, days	2.0 ± 4.0
Stay in hospital after TAVR, days	12.8 ± 39.5
Paravalvular leak more than moderate, %	16 (3.1)
Early safety at 30 days, %	445 (85.7)
Mortality, %	4 (0.8)
All stroke, %	21 (4.0)
Life-threatening bleeding, %	41 (7.9)
Acute kidney injury (AKIN stage 2 or 3), %	18 (3.5)
Coronary obstruction requiring intervention, %	0 (0)
Major vascular complication, %	17 (3.3)

Values are presented as mean ± SD, n(%).

Abbreviations: TAVR, transcatheter aortic valve replacement; AKIN, acute kidney injury network.

Fig. 3.

Kaplan–Meier survival curve after transcatheter aortic valve replacement During the median follow-up period of 448 days, the post-TAVR survival rates after 1 and 3 years were 90.7 ± 1.4% and 74.3 ± 2.9%, respectively, as estimated by Kaplan–Meier analysis.

After performing univariate analyses, baseline variables, including age, body mass index, New York Heart Association classification Ⅲ/Ⅳ, hypertension, dyslipidemia, STS score, peripheral artery disease, atrial fibrillation, serum hemoglobin and albumin level, estimated glomerular filtration rate, mitral and tricuspid regurgitation, and new-generation valve (using SAPIEN 3, Edwards, U.S.A/Evolut PRO, Medtronic, U.S.A) were included in a multivariate Cox regression analysis (Table 3). Cox regression analysis showed that albumin level [hazard ratio (HR):2.20, 95% confidence interval (CI): 1.36–3.62, p = 0.001], AF (HR: 1.79, 95% CI: 1.13–2.82, p = 0.012), and STS score (HR: 1.33, 95% CI: 1.06–1.67, p = 0.015) were independently associated with all-cause mortality. Notably, the adjusted HR of sex-specific parameters such as current smoking, and heavy drinking, and presence of cancer were as follows: (current smoking, HR: 1.05, 95% CI: 0.37–2.98, p = 0.930; heavy drinking (>8 drink/week), HR: 1.37, 95% CI: 0.75–2.48, p = 0.307; and presence of cancer, HR: 1.13, CI: 0.69–1.87) (Table 4). In sensitivity analysis for additionally assessing the relationship between all-cause mortality and TAVR procedure, device success was added in the multivariate analysis. It reveals that device success was additionally independently associated with all-cause mortality after TAVR. Predictors other than device success (STS score, AF, albumin level) were still associated with all-cause mortality, and outcomes did not differ (Supplementary table).

Table 3.

Predictors of all-cause mortality in univariate and multivariate Cox regression analyses.

	Hazard ratio	95% CI	p-value	adjusted Hazard ratio	95% CI	p-value
Age (at TAVR)	1.32	1.07–1.62	0.008			0.236
Body mass index	1.39	0.89–2.18	0.144			0.545
NYHA classification Ⅲ/Ⅳ	1.52	1.00–2.30	0.049			0.754
Diabetes mellitus	0.84	0.53–1.35	0.476
Hypertension	0.56	0.36–0.87	0.010			0.091
Dyslipidemia	0.64	0.42–0.96	0.030			0.227
Prior MI	1.12	0.61–2.05	0.725
PAD (ABI < 0.9)	1.74	1.11–2.74	0.016			0.063
Prior stroke	1.15	0.67–1.98	0.605
Atrial fibrillation/Atrial flutter	2.03	1.31–3.12	0.001	1.79	1.13–2.82	0.012
COPD (FEV1.0%<70)	0.97	0.53–1.79	0.930
Cancer	1.30	0.82–2.05	0.266
Serum hemoglobin level	0.45	0.30–0.68	<0.001			0.074
STS score	1.19	1.08–1.14	<0.001	1.33	1.06–1.67	0.015
Serum albumin level	2.99	1.96–4.56	<0.001	2.20	1.36–3.62	0.001
eGFR [every 30 mL/(min·1.73 m2) decrease]	1.41	1.15–1.71	0.001			0.137
Ejection fraction	1.28	0.94–1.73	0.114
AV area	1.09	0.93–1.28	0.303
AV mean pressure gradient	1.16	0.90–1.49	0.250
Mitral regurgitation	1.38	1.03–1.85	0.029			0.279
Tricuspid regurgitation	1.12	1.00–1.26	0.055			0.874
TRPG	1.42	0.45–4.49	0.556
Transfemoral approach	1.40	0.79–2.48	0.254
New generation valve (SAPIEN3/Evolut Pro)	0.62	0.39–0.99	0.044			0.088

Abbreviations: TAVR, transcatheter aortic valve replacement; CI, confidence interval; NYHA, New York Heart Association; MI, myocardial infarction; CABG, coronary artery bypass graft; PAD, peripheral artery disease; ABI, ankle brachial index; COPD, chronic obstructive pulmonary disease; STS, Society of Thoracic Surgeons; eGFR, estimated glomerular filtration rate; AV, aortic valve; TRPG, tricuspid regurgitation pressure gradient.

Table 4.

Adjusted hazard ratio for all-cause mortality after TAVR in men.

	Adjusted hazard ratio	95% CI
Smoking (current)	1.05	0.37–2.98
Heavy drinking (>8 drink/week)	1.37	0.75–2.48
Presence of cancer	1.13	0.69–1.87

Adjusted for atrial fibrillation, serum albumin, and STS score.

Abbreviations: STS, Society of Thoracic Surgeons; CI, confidence interval.

4. Discussion

This study’s findings are as follows: (1) procedural success and all-cause mortality at 30-day after TAVR were 97.2% and 0.8%, which represents excellent outcomes; (2) the factors affecting all-cause mortality after TAVR in men were STS score, AF, and serum albumin level; and (3) we focused on the prevalence and impact of smoking and drinking habit, and presence of cancer as men-specific factors.

All-cause mortality rate at 30 days in men were comparable with those in previous study with similar age population and the STS scores [20]. In terms of cause of death, all patients who died within 30 days, had cardiovascular related mortality, and this is also consistent with previous study [20], suggesting that there is more frequent atherosclerotic disease in men than women [9].

The STS score was originally designed to predict the prognosis after surgical aortic valve replacement. However, similar to this study, some reports suggested that the STS score is associated with prognosis after TAVR and is more effective in predicting the perioperative and long-term prognosis after TAVR than EuroSCORE [10], [21]. A significant association with a prognosis was demonstrated, indicating that mortality risk can be assessed not only by subjective indices, such as frailty, but also by objective indices, which may be clinically useful in assessing mortality risk after TAVR.

Serum albumin level has been associated with prognosis after TAVR and was used as a prognostic factor in this study [22]. Low serum albumin levels are considered part of frailty, and the level is a relevant prognostic marker for frailty [21], [23]. All patients were expected to have greater nutritional status by receiving TAVR by the multidisciplinary heart team approach, however, the reality is that some patients might have poor nutritional status regardless of severe aortic stenosis, and subsequently died after TAVR.

Compared with the Women’s INternational Transcatheter Aortic Valve Implantation (WIN-TAVI) studies, which show that sex-specific factors such as the history of pregnancy and number of prior pregnancies were incremental predictors of the 30-day primary safety endpoint, men-specific risk factors such as smoking and drinking habit were not independent prognostic factors in this study, irrespective of the degree [16], [17]. Regarding the cancer, presence of gynecological or breast cancer in WIN-TAVI study was not associated with poor outcomes, and this is in line with our results despite lacking of information about metastasis in each cancer in our registry data. [16], [24]. Nonetheless, their impact on prognosis after TAVR has been less well reported. To the best of our knowledge, this is the first study to specifically investigate the prevalence and impact of smoking and drinking habit, and presence of cancer in TAVR outcomes.

Tobacco smoke includes numerous toxic components that can cause injury via inflammation and irritation, asphyxiation, carcinogenesis, and other mechanisms [25]. Notably, smoking contributes to an increased risk of heart failure and myocardial infarction. In addition, smoking is considered a risk factor for AF, which was a prognostic determinant in this study, and this association was particularly strong in men [26]. One possible reason that smoking was not a significant prognostic factor on all-cause mortality in this study was that smokers may have well-controlled comorbidities after prior cardiovascular events at a younger age than never smokers; therefore, smoking may not have caused the worse prognosis in this study, although it is a matter of speculation. Notably, this study did not strictly distinguish chronic obstructive pulmonary disease (COPD) from inflammation due to smoking because COPD was solely defined as a decrease in forced expiratory volume% during 1 s of spirometry and this might be a reason why smoking and COPD were not identified as prognostic determinants.

Alcohol adversely affects all organs (e.g., liver disease, neurological disorders, poor nutrition, incontinence, diarrhea, myopathy, fractures, and adverse reactions to medications). A relationship between drinking habit and heart disease has also been previously reported [27]. Consuming < 7 drinks per week reduces the incidence of heart failure, particularly in men, whereas drinking >21 drinks per week is considered sufficient to cause myocardial damage. However, it is assumed that the number of such patients with heavy drinking is relatively small, particularly in this study, which are included older men. Conversely, AF is closely related to alcohol intake, and a previous study has reported that the incidence of AF increases when alcohol consumption increases, even if it does not lead to heavy drinking [28]. In this study, AF was detected as a prognostic determinant after TAVR, and it is possible that drinking habit as well as smoking habit indirectly impacted the prognosis negatively after TAVR.

4.1. Limitations

This study has some limitations. First, the number of patients affected by current smoking and heavy drinking is limited, and this may have been underpowered to show a statistical difference. Moreover, other men-specific prognostic factors may be identified in the larger study. Second, information about drinking habit was obtained by weekly basis, not by daily basis. In addition, there is a lack of information about presence of metastasis in this registry data. Third, it is possible to have a bias in outcomes due to the postoperative lifestyle change.

5. Conclusion

Our study demonstrated that the results of TAVR in men were excellent. Serum albumin levels, AF, and STS scores were independently associated with all-cause mortality after TAVR in men. Larger study is recommended to further examine the effects of men-specific prognostic factors such as smoking and drinking habit, and presence of cancer in men.

Author contributions

All the authors contributed to the conception and design of the study. The material preparation, data collection, and analysis were performed by Kota Nishida, Ryosuke Higuchi, and Mike Saji. The first draft of the manuscript was written by Kota Nishida, and all the authors commented on the previous versions. All authors have read and approved the final manuscript. All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.

Registration number of clinical studies

This registry was registered in the clinical research database (UMIN 000031133).

Acknowledgment of grant support/Funding

This work was supported by a research grant from the Sakakibara Heart Foundation.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

The following are the Supplementary data to this article: