Body Mass Index in TAVR

Corresponding Author

Severe aortic stenosis, once a surgical disease, is now routinely treated percutaneously with transcatheter aortic valve replacement (TAVR) across the entire risk spectrum.^1,2 The procedure’s growing safety profile has elevated patient-level factors—frailty, nutritional status, and body composition—as critical determinants of outcomes. Among these, body mass index (BMI) has emerged as a paradoxical clinical metric. Earlier registry and trial analyses suggested that higher BMI conferred a protective benefit in elderly cardiovascular patients.³ Conversely, underweight patients consistently exhibited poorer outcomes, likely reflecting underlying frailty and diminished physiological reserve. For instance, the French FRANCE-2 (FRench Aortic National CoreValve and Edwards-2) registry⁴ (n = 3,072) reported 1-year survival climbing from 67.9% in patients with BMI <18.5 kg/m² to 80.3% in those ≥30 kg/m², with obesity independently protective (adjusted HR: ≈0.74 vs normal weight). Similarly, in the 31,900-patient STS/ACC TVT Registry,⁵ underweight status carried a 41% excess adjusted risk of 1-year death (HR: 1.41), whereas class I to II obesity conferred a 12% to 20% relative risk reduction (HR: 0.88-0.80). This so-called “obesity paradox”^6,7 has provoked debate—especially given obesity’s rising prevalence and complex relationship with cardiovascular outcomes.

In this issue of JACC: Asia, Cantey et al⁸ present findings from the Trans-Pacific TAVR Registry, assessing the impact of BMI in contemporary TAVR practice. Among 1,339 patients with severe aortic stenosis, BMI had no discernible effect on survival or major complications after transfemoral TAVR. At 4 years, overall survival was 76.6%, with no differences among underweight (BMI <18.5 kg/m²), normal (>18 and <25 kg/m²), overweight (>25 and <30 kg/m²), or obese (>30 kg/m²) strata (log-rank P = 0.69). The 30-day mortality was 1.9%; 1-year mortality ranged from 10% (underweight) to 18% (obese), but differences vanished after multivariable adjustment. Obese patients carried more atherosclerotic cardiovascular risk, whereas the underweight group was frailer yet showed the greatest left ventricular mass regression at 1 year. Taken together, this study challenges the long-debated “obesity paradox” in TAVR and suggests that nutritional/frailty status, not BMI per se, deserves center-stage in preprocedural risk discussion.

The Trans-Pacific TAVR Registry uniquely bridges anthropometric divides. The study implicitly recognizes that adiposity and cardiovascular risk vary across ethnicities—moving BMI interpretation beyond simplistic numeric thresholds. Centralized VARC-2⁹ adjudication and standardized echocardiography enhance internal validity, minimizing intercenter variability. Moreover, inclusion of biochemical (albumin, creatinine) and procedural (sedation strategy, valve generation) covariates reveals nuanced confounders traditionally overlooked in BMI analyses. Furthermore, by assessing cardiac remodeling, the study highlights paradoxical hemodynamic improvements among underweight patients, challenging conventional assumptions about BMI’s utility and shifting emphasis toward nutritional and frailty assessments.

This study’s observational design does inherently risk residual confounding. Lack of direct frailty measures—sarcopenia, grip strength—or advanced body composition imaging leaves BMI’s known limitations unaddressed. The underweight subgroup is small (n = 45),⁸ limiting statistical power and stability of conclusions. Additionally, incomplete echocardiographic follow-up (42% at 1 year) introduces selection bias favoring healthier survivors. Reliance on WHO BMI cutoffs rather than ethnic-specific thresholds or advanced adiposity metrics further constrain interpretability, especially given distinct Asian vs Western body-composition patterns.

What can we deduce from these observations? First, BMI is a crude anthropometric proxy. It ignores sarcopenia, visceral adiposity, and unintentional weight loss—features that track more closely with resilience or frailty. Second, paradoxes often stem from bias, suggesting how collider stratification, reverse causation, and measurement error can flip harmful exposures into seemingly protective ones. For practical purposes, device choice or access planning need not be reflexively altered by weight alone; contemporary transfemoral systems accommodate sheath-to-femoral-artery ratios safely across a vast BMI range. Third, the modest pacemaker and bleeding differences among obese patients in prior analyses have largely faded in the sedation-first, ultrasound-guided access era. Hence, clinicians should pivot from BMI triage toward actionable domains—albumin, hemoglobin, quadriceps circumference—that genuinely map onto functional recovery.

Much uncharted terrain remains. Which matters most: subcutaneous vs visceral fat, or skeletal muscle mass? Early evidence suggests that computed tomography–derived psoas cross-sectional area or fat radiodensity may outperform gait speed in long-term risk modelling,10, 11, 12 but prospective validation is lacking. In the Montreal-Munich cohort of 208 TAVR patients, each additional 1 cm² of psoas area was associated with lower adjusted all-cause mortality in women by 12% (HR: 0.88; 95% CI: 0.78-0.99).¹³ Another Japanese multicenter registry of 1,372 patients showed that high visceral fat radiodensity—a marker of myosteatosis—independently increased 5-year mortality (HR: 1.34; 95% CI: 1.03-1.76).¹⁴ The ongoing Protein and Exercise to Reverse Frailty in Older Men and Women Undergoing Transcatheter Aortic Valve Replacement (PERFORM-TAVR) Trial¹⁵—a 200-patient multicenter randomized study—is testing whether combining preprocedural protein supplementation with post-TAVR exercise improves functional status and quality of life, potentially yielding the first causal evidence that body composition interventions change outcomes. Meanwhile, long-term follow-up of the transpacific cohort will help clarify valve durability and clinical trajectories at BMI extremes. Advanced analytics—including AI-driven computed tomography assessments, metabolomic and proteomic biomarkers, and machine-learning frailty engines—promise to shift risk stratification from crude BMI to truly personalized phenotypes. Together, these innovations herald a future where scale-based metrics give way to precision resilience profiling in structural heart disease.

In conclusion, Cantey et al⁸ demonstrate that in a modern, sedation-first, multinational practice, BMI fails to discriminate risk. The "obesity paradox" dissolves when viewed through ethnically diverse, frailty-adjusted lenses. Frailty—not thinness per se—is the real foe. Guidelines^16,17 should therefore pivot away from numeric BMI triggers toward multidimensional frailty batteries that capture nutrition, muscle, cognition, and psychosocial reserve. For clinicians, this means focusing on modifiable resilience markers: albumin >3.5 g/dL, hemoglobin >11 g/dL, and preserved mobility. As TAVR enters its third decade, the scale gives way to the dynamometer—muscle, not mass, defines survivorship

Funding Support and Author Disclosures

The authors have reported that they have no relationships relevant to the contents of this paper to disclose.