Sex Does Not Predict Death After Transcatheter Aortic Valve Implantation, Single-Center Study Finds

Women and men who undergo transcatheter aortic valve implantation (TAVI) for severe aortic stenosis appear to face similar risks of dying in the hospital, at 30 days, and within the first year after the procedure, according to a retrospective single-center study of 1,065 consecutive patients treated in Vigo, Spain, between 2010 and 2023.

Of the 1,065 patients included, 519 (48.7 percent) were women and 546 (51.3 percent) were men.

The baseline profiles of the two sexes differed in ways that matter for interpreting the outcomes. Women were slightly older, with a median age of 84 years compared with 83 years in men, and they carried higher surgical risk: their median EuroSCORE II was 4.4 versus 3.4, and their median Society of Thoracic Surgeons (STS) score was 5.9 versus 4.6, both differences statistically significant. Frailty was also more common among women, affecting 28.3 percent versus 23.1 percent of men. Men, by contrast, had higher rates of diabetes mellitus (38.5 percent versus 31.8 percent) and chronic obstructive pulmonary disease (17.8 percent versus 10.8 percent), along with higher baseline serum creatinine levels. Echocardiography showed that women had higher left ventricular ejection fractions and higher mean transaortic gradients, while men had slightly larger aortic valve areas.

Despite these contrasting risk profiles, mortality outcomes were strikingly similar. In-hospital death occurred in 2.0 percent of men and 3.5 percent of women, a difference that did not reach statistical significance. In the date-based survival population, 30-day mortality was 2.4 percent in men (13 of 543) and 3.7 percent in women (19 of 518), again not significant. The time-to-event analysis included 1,061 of the 1,065 patients; four patients with documented death but no ascertainable date of death were excluded. One-year vital status was confirmed in 94.7 percent of the cohort, and 121 deaths occurred within the first year. The crude one-year death proportion was 11.6 percent in men and 11.2 percent in women, and the Kaplan-Meier estimated cumulative mortality at one year was 11.9 percent (95 percent confidence interval 9.4 to 14.9) in men versus 11.4 percent (95 percent confidence interval 9.0 to 14.6) in women, with a log-rank p value of 0.879.

Statistical modeling reinforced the picture of equivalence. In a multivariable Cox proportional hazards regression that included age, sex, frailty, hemoglobin, estimated glomerular filtration rate, atrial fibrillation, STS score, left ventricular ejection fraction, and chronic obstructive pulmonary disease, female sex was not independently associated with one-year mortality, with an adjusted hazard ratio of 1.07 (95 percent confidence interval 0.69 to 1.66; p = 0.772). That model, restricted to the 748 patients with complete covariate information and 86 deaths, did identify higher baseline hemoglobin, higher eGFR, and higher ejection fraction as predictors of better one-year survival. A parsimonious logistic model for 30-day mortality containing sex, age, and STS score likewise found no independent association for female sex, with an adjusted odds ratio of 1.79 (95 percent confidence interval 0.80 to 3.98; p = 0.154). Sensitivity analyses that added procedure year as a continuous covariate left the sex estimates materially unchanged.

Where the sexes did diverge was in procedural complications. Any bleeding, as recorded in VARC-2 registry categories, occurred in 9.1 percent of women versus 4.8 percent of men (p = 0.006), and major or life-threatening/disabling bleeding affected 3.5 percent of women versus 1.3 percent of men (p = 0.018). Vascular complications showed a non-significant trend in the same direction, at 6.5 percent in women versus 3.9 percent in men (p = 0.059). However, after adjustment for age, body mass index, baseline hemoglobin, and eGFR, the association between female sex and any bleeding was attenuated and no longer statistically significant, with an adjusted odds ratio of 1.61 (95 percent confidence interval 0.92 to 2.81; p = 0.093). The authors attribute the unadjusted excess largely to anatomical factors, noting that smaller iliofemoral vessel dimensions in women can raise the sheath-to-artery ratio and contribute to access-site injury.

The opposite pattern emerged for conduction complications. Among the 932 patients without a pre-existing pacemaker and with available post-procedural pacemaker status, new permanent pacemaker implantation was required in 26.9 percent of men versus 17.8 percent of women (p = 0.001). This association survived adjustment for age, prosthesis category, and prosthesis size, with male sex carrying an adjusted odds ratio of 1.49 (95 percent confidence interval 1.03 to 2.17; p = 0.036). A sex-by-prosthesis-category interaction test was not significant (p = 0.953), meaning the data provided no evidence that the sex association differed between self-expanding and balloon-expandable valves. The overall pacemaker rate of 22.5 percent in the cohort is higher than the 15.6 percent reported in a large meta-analysis but is consistent with contemporary European registries, including the French TAVI Registry, which reported 24.7 percent in men and 20.5 percent in women.

Mechanistically, the authors point to established differences in conduction system anatomy and disease. Men more frequently present baseline conduction abnormalities, particularly right bundle branch block, which increases susceptibility to post-procedural atrioventricular block. Larger annular dimensions in men also often require larger prostheses, which may exert greater mechanical stress on the conduction tissue, whereas women typically receive smaller devices, as reflected in the median prosthesis size of 26 millimeters in women versus 29 millimeters in men in this cohort. The authors caution, however, that baseline conduction variables and implantation-depth measurements were not consistently available in the registry, so residual confounding by these factors cannot be excluded.

Other mechanical and ischemic complications were rare and broadly similar between sexes. Coronary occlusion occurred in 0.3 percent of patients overall and was observed only in women, though the difference was not statistically significant. Cardiac tamponade was numerically more frequent in women (1.7 percent versus 0.5 percent; p = 0.067) without reaching significance, annular rupture was rare at 0.5 percent overall, stroke occurred in 1.8 percent of men and 1.2 percent of women, and periprocedural myocardial infarction was uncommon at 0.4 percent overall. Acute kidney injury incidence was similar between sexes across severity stages. Procedural success was high in both groups, though slightly higher in men (98.9 percent versus 97.3 percent; p = 0.044), and women received smaller devices with no significant differences in device type or expansion mechanism.

The authors frame their results as descriptive of sex-related risk profiles rather than as a basis for sex-based treatment decisions, emphasizing the limitations inherent in a retrospective, single-center design that spans a period of major change in TAVI technology, imaging planning, and operator experience. They also note that a recent meta-analysis of 48 studies including 134,935 patients reported a time-dependent pattern in which women experienced a long-term survival advantage after TAVI, so the neutral one-year result in this cohort should not be read as evidence that sex-related survival differences cannot emerge with longer follow-up. Prospective, multicenter studies incorporating contemporary devices, detailed computed tomography anatomy, baseline conduction status, and implantation-depth measurements are, in their view, the next step toward understanding the mechanisms behind the differences that do persist, particularly in bleeding and pacemaker implantation.

Subject of Research: Sex differences in mortality and procedural complications after transfemoral transcatheter aortic valve implantation in a real-world single-center cohort.

Article Title: Sex differences in clinical outcomes after transcatheter aortic valve implantation: a real-world cohort

Article References: Estévez Táboas, M., Echarte-Morales, J., Ruiz Golcher, E. J., Sotomayor-Casillas, C. E., González García, A., Neuhalfen, D., Iglesias-Otero, C., García-Campos Losada, E., Vázquez-Vázquez, N., González Fernández, D., Prado Prado, H., Cuiñas Fariña, C., Antelo-Abeijón, A., Juan-Salvadores, P., Estevez-Loureiro, R., Baz Alonso, J. A., Iñiguez-Romo, A., & Jímenez-Díaz, V. A. (2026). Sex differences in clinical outcomes after transcatheter aortic valve implantation: a real-world cohort. Frontiers in Cardiovascular Medicine, 13, Article 1941651. https://doi.org/10.3389/fcvm.2026.1941651

Image Credits: AI Generated

DOI: 10.3389/fcvm.2026.1941651

Keywords: aortic stenosis, TAVI, sex differences, mortality, bleeding, pacemaker implantation, vascular complications, retrospective cohort, structural heart disease, VARC-2, Spain, differences