Does the “obesity paradox” exist after transcatheter aortic valve implantation?

Consecutive patients with severe symptomatic aortic stenosis (AS) who underwent TAVI between May 2018 and April 2021 at our institute and were considered inoperable or at high risk for a traditional SAVR were included [6]. The medical staff measured the height and weight of all patients on admission. To improve the sample sizes in each group, patients were classified into three groups according to 25th, 26th–75th, and 76th–99th percentiles of body mass index (BMI). Patients were considered low weight if their BMI was < 21.9 kg/m², middle weight if their BMI was 21.9–27.0 kg/m², and high weight if their BMI was > 27 kg/m², in contrast to the previous BMI classification system used for adults in China [28]. Patients were also classified into two groups according to vascular access: through the transfemoral artery route (TF-TAVI) and through the transapical route (TA-TAVI). We determined the method that was best for patients based on current guidelines [6] and took decisions by consulting the local heart team comprising cardiac surgeons and cardiologists. Device endpoints, success rate, and related adverse events were recorded according to the Valve Academic Research Consortium (VARC)-2 definitions [29]. Patients matching one or more of the following pathological characteristics were excluded: improper aortic valve size; thrombosis of the left ventricle; active endocarditis; high risk of coronary ostia obstruction; plaque in the ascending aorta or aortic arch with active thrombus.

The related cardiovascular risk factors were recorded and defined as follows: hypertension as systolic blood pressure ≥ 140 mm Hg or diastolic blood pressure ≥ 90 mm Hg [30]; diabetes mellitus as fasting blood glucose ≥ 7 mmol/L or a previous diagnosis [31]; stroke based on a clinical diagnosis or CT scan [32]; hypoxemia as oxygen saturation of < 90% by pulse oximetry or available arterial oxygen < 60 mm Hg [33, 34]. Frailty was evaluated based on serum albumin levels, gait speed, grip strength, and the number of independent activities of daily living.

Informed consent was obtained from all study subjects and the anonymity of patients was preserved. The study was carried out in accordance with the ethical principles of the 1975 Declaration of Helsinki with later amendments.

All procedures were performed under general anesthesia and endotracheal intubation to prevent interferences in the surgical process due to poor patient compliance. TF-TAVI (VitaFlow®Transcatheter Aortic Valve, Micro Port, Shanghai, China) was performed in 94 patients by percutaneous access through the femoral artery, whereas TA-TAVI was performed via a small left anterior thoracotomy (J-Valve, JC Medical, China) in 15 patients. In both types of surgery, 100 I.E./kg body weight of heparin was used for anticoagulation, which was subsequently antagonized by protamine at a 1:1 ratio. Surgeries were performed on average by 5 well-trained surgeons. The detailed procedure is described in the Additional file 1: Video S1.

The primary outcomes were 30-day mortality and a composite of mortality to a maximum follow-up of 35 months. The secondary endpoints were perioperative pulmonary complications. All surgical survivors were monitored via telephone conversations, emails, or letters.

In the general population, as the BMI increases above 30 kg/m², the risk of developing CVD increases and the risk of mortality is higher [35]. Nonetheless, it is often observed that obese individuals with chronic diseases have a reduced risk of mortality compared with nonobese individuals with similar disease characteristics. The positive effect of increased BMI following these interventions has been called the “obesity paradox.” This paradox has been described in several disease categories, including CVD, diabetes mellitus, and chronic kidney disease [36‐38]. The “obesity paradox” has been reported mainly in patients with heart failure, acute coronary syndrome, and following percutaneous and surgical coronary interventions [18‐20, 36, 38]. Some studies have explained that this protective effect is because the soluble tumor necrosis factor (TNF)-α receptors in adipose tissues neutralize the adverse effects of TNF-α and decreases the mortality of patients with chronic inflammatory diseases such as CVD [39]. In addition, higher lipoproteins in circulation may also bind and reduce the role of lipopolysaccharides in stimulating the release of inflammatory cytokines [38].

There is also evidence for the obesity paradox in high-risk patient groups after TAVI [21, 26]. Konigstein et al. reported that after adjusting baseline characteristics, increased BMI was independently related to the improvement in survival after TAVI [21]. Conclusions of the data analysis of the large FRANCE 2 Registry that included 3072 patients were also consistent with the obesity paradox after TAVI [26]. In our univariate model, high weight was found to be closely related to increased survival rates, which was in line with the obesity paradox.

In the multivariate model, the overall mortality rate was found to be similar among the three BMI groups after adjusting for different baseline characteristics including diabetes mellitus, GFR, and frailty. Our findings were contradictory to those of previous studies; according to our multivariable model, there was no obesity paradox after TAVI and the overall midterm survival rates were similar among all BMI groups (Fig 1). Previous studies have explained this discrepancy because the baseline frailty factor was not included in their research and analysis models. The updated VARC-2 document defines frailty as slowness, weakness, exhaustion, wasting and malnutrition, inactivity, and loss of independence, and emphasizes that frailty is the most important feature in current risk models [29] and is a strong predictor of mortality and adverse effects after cardiac surgery and TAVI [40‐42]. Indeed, in this study, frailty was a strong predictor of overall mortality in both the univariate and multivariate models (HR 3.825; p = 0.003; Table 4). Meanwhile, 48.1% of high-weight patients were frail versus 18.5% of low-weight patients (p = 0.032), which might be a possible confounder in resisting the occurrence of the obesity paradox. In addition, high-weight patients in the current study were significantly older, had lower GFR and higher incidences of previous percutaneous coronary intervention, and prevalence of diabetes mellitus, all of which are possible confounding factors that might provide an explanation for the loss of the protective effect of high weight on mortality.

Many studies have consistently reported that patients who are obese are more likely to develop severe hypoxemia after surgery due to the significant decline in lung compliance. Due to their abnormal BMI, certain clinical features of patients who are obese may require special management during the perioperative period, such as higher doses of medications, greater tidal volume, and higher airway pressure to ensure adequate postoperative ventilation. Therefore, there is a significant increase in dyspnea in individuals who are obese people. Additionally, respiratory resistance increases in high-weight individuals, and the mechanism of acute respiratory distress syndrome may be attributed to an imbalance in anti-inflammatory and pro-inflammatory cytokine levels and in the oxidant and antioxidant levels [43‐46]. Most patients who are obese suffer from chronic and excessive inflammation and oxidative stress [43, 44]. When pro-inflammatory signaling pathways are significantly upregulated, patients who are obese are prone to produce more abnormal cytokine products and acute-phase reactants. Furthermore, the production of pro-inflammatory cytokines and mediators is known to increase with weight gain [45, 46]. Moreover, high weight can increase oxidative stress and lead to an increase in reactive oxygen products, which may lead to direct damage of cellular membranes, cause monocyte cellular adhesion, and the release of chemotactic factors and vasoactive substances. The above processes are obvious especially while undergoing CPB. However, in this study, although patients with high BMI had lower PaO₂/FiO₂ than low-weight patients at pre- and early postoperative time (baseline, transitional extubation, and post extubation 12th hour), high weight did not have a significant effect on severe postoperative hypoxemia and longer intubation time in our multiple regression model. An explanation for this divergence may be that patients with TAVI without CPB did not have a marked activation of the signaling pathways discussed above.