Effect of body mass index on clinical outcome and all-cause mortality in patients undergoing transcatheter aortic valve implantation

Overweight and obesity are the fifth leading modifiable cause of death in the world, accounting nearly 3.4 million deaths annually [1]. The prevalence of obesity, often defined as body mass index (BMI), has been increasing dramatically [1]. According to available data, more than two-thirds of adults in the United States, and more than 2.1 billion people worldwide, suffer from obesity [1, 2]. Although it has been suggested that obesity occurs because of an energy imbalance between caloric intake and expenditure, the resulting energy excess and associated weight gain reflects a complex interaction between genes, epigenetic markers, environment, and lifestyle [2‐4].

According to the Framingham Heart study, conducted among participants (age: 30–49 years) with no cardiovascular disease at baseline, overweight and obesity were associated with a decrease in life expectancy and increased early mortality during the follow-up period of ≥4 years [5]. Accordingly, another population-based cohort reported an increased risk of all-cause mortality among elderly (≥85 years) obese participants [6]. In addition to the risk of mortality, obesity is an underlying promotor of systemic metabolic dysfunction, i. e., dyslipidaemia, decreased insulin sensitivity, hyperinsulinaemia, hyperglycaemia, and hypertension [2, 4].

However, despite the well-documented adverse effects of overweight or obesity on general health status, being overweight or obese is associated with better survival in patients undergoing medical interventions [7], vascular surgery [8], cardiovascular intervention [9], and in patients who are hospitalised for acute decompensated heart failure [10, 11]. These observations led to the concept of reverse epidemiology, also known as the obesity paradox. The obesity paradox states that a higher BMI may, counter-intuitively, be linked to improved survival in certain patient groups. However, these observations do not support common practice where weight loss is recommended prior to cardiac treatments.

Contradicting data exist regarding the effect of BMI on outcome in patients with aortic stenosis (AS). In one study (n = 1664) overweight and obese patients with AS were at increased risk for mortality, whereas another study (n = 400) found contradictory results [12, 13]. Although there are few data regarding the effect of BMI on outcome among patients with AS who undergo transcatheter aortic valve implantation (TAVI), several cohort studies showed contradicting or supporting results [14‐20]. For instance, a large French Aortic National CoreValve and Edwards (FRANCE-2) registry showed improved survival outcome among overweight and obese individuals undergoing TAVI [14]. Overweight or obesity was associated with improved survival following TAVI in other cohort studies as well [21]. However, a recent study did not find such a paradoxical relationship [19]. Therefore, the current study aimed to assess the effects of body mass index on short and long-term all-cause mortality in patients undergoing TAVI in the current era.

We performed a retrospective observational cohort study encompassing all eligible consecutive patients who underwent TAVI between September 2008 and October 2016 at the University Medical Center Utrecht, Utrecht, the Netherlands. All demographic and peri-procedural data were prospectively collected in our dedicated database and retrospectively analysed in this study. All patients gave informed consent for the procedure and due to the retrospective nature of the study design, ethics committee approval was waived.

In the present study, we aimed to evaluate the impact of body mass index on all-cause mortality and clinical outcome in patients undergoing TAVI. After adjustment, being overweight was associated with decreased 30-day and 1‑year all-cause mortality, while there was no association found between obesity and mortality outcomes following TAVI. Furthermore, there were no differences observed in postoperative bleeding or vascular complications between the BMI categories.

Considering the aging population, the global prevalence of overweight and obesity is expected to rise [12]. AS is the predominant type of valvular heart disease among elderly and associated with poor prognosis [24]. Prevalence of AS is 3%, increasing with age up to 10% in adults ≥80 years [25]. Currently, TAVI has emerged as a valuable option to treat severe AS in elderly patients considered to be inoperable or at high surgical risk for SAVR [26]. However, fewer data exist regarding body weight management in patients undergoing TAVI.

Interestingly, we observed decreased short-term and long-term mortality outcomes after TAVI among overweight patients. According to our knowledge, this is the first time such a short-term effect of overweight on mortality outcomes after TAVI has been shown. Our findings are in line with literature findings including patients who were admitted for first-time coronary artery bypass (CABG) or combined CABG/aortic valve replacement surgery, patients after coronary angiography for diagnosis of acute coronary syndrome, and among patients undergoing cardiac surgery [27‐29]. While some studies reported no effect of being overweight on mortality outcome after TAVI [15, 19], others reported long-term positive effect of overweight on mortality outcome after TAVI [14, 17, 18]. Although current guidelines are advising weight loss and prevention of overweight and obesity, these counter-intuitive findings regarding the positive association between overweight/obesity and mortality may create the impression that an intentional weight loss may not always be favourable.

Generally, obesity has been associated with adverse clinical health status [4], however, we observed no association between obesity and mortality after TAVI. Our results are in line with some previous reports [15, 18, 19], however, several other studies including a meta-analysis, found a beneficial effect of obesity on mortality outcome after TAVI [14, 17, 21]. In addition, inconstancies in these contradictive observations could be explained by unhealthy metabolic profile (i. e., hypertension, dyslipidaemia, diabetes) of obese individuals included in our study that may have influenced the results [30].

Although we observed no association between BMI as a continuous variable and mortality, several other studies reported a gradual reduction in death rate for every increment in BMI unit (kg/m²) during short-term or long-term follow-up after TAVI [15‐17]. However, a recent study among patients (n = 4571) undergoing TAVI demonstrated that an increase in BMI was associated with higher risk of mortality in patients with elevated BMI (>32 kg/m²) [20]. Moreover, a ‘U’ shape association between BMI a continuous variable and mortality was found among patients with diabetes, acute heart failure and in patients undergoing cardiac surgery [29, 31, 32].

While performing TAVI in overweight or obese individuals may be challenging due to vascular access site and fluoroscopic visualisation [33], there were no differences in vascular and bleeding complications observed between the BMI groups in our cohort. These findings are in line with a recent meta-analysis evaluating the effect of BMI on outcome after TAVI [21]. This could be explained by improved TAVI technique and sustained efficacy of TAVI. However, previous studies using early-generation transcatheter aortic valves and techniques reported higher postoperative complication rates among overweight/obese and underweight individuals after TAVI. For instance, in a multi-centre study (n = 940), higher rates of postoperative minor stroke, minor vascular complications and acute kidney injury stage 1 were observed among obese individuals following TAVI [15]. Consistent with these findings, another study (n = 409) reported higher incidence of major postoperative vascular complications and a trend toward more major and life-threatening bleeding events among obese patients after TAVI [17]. Furthermore, a higher rate of major vascular complications was observed in patients with underweight. However, according to another study, BMI < 20 compared with BMI > 20 was not associated with adverse events following TAVI [34]. Inconsistencies in reported results could be explained by different definitions of BMI, i. e., BMI either as a categorical or continuous variable, which could lead to uncertainty in defining the cut-off points and interpretation of results. Therefore, studies should report their results according to the standardised BMI classification, i. e., WHO classification.

The mechanism behind the obesity paradox remains unclarified [35]. However, there are several possible factors that could explain the paradoxical effect of BMI on clinical outcomes. For instance, excess body weight may increase metabolic reserve and counteract the negative effects of acute injuries. Furthermore, patients with ischaemic heart failure appear to have a higher level of TNF-a concentrations compared with those with a non-ischaemic aetiology [36]. Moreover, adipose tissue has been shown to produce TNF-a receptors [37], therefore, overweight and obese patients may have a protective buffer from the negative effect of increasing TNF-a by producing higher levels of these receptors. Furthermore, several other investigators argue that the obese group, consisting of younger individuals, seeks medical care earlier, is treated medically more aggressively, and therefore benefits more from medical and interventional treatments [38]. However, these differences did not affect the outcome after adjustments in the multivariate model, even though obese individuals in our cohort were younger and used more baseline medications, i. e., diuretics, angiotensin II receptor antagonists, lipid-lowering agents, or insulin, compared with the normal weight group.

BMI either as a continuous or categorical variable has been frequently used to define body weight. However, BMI does not discriminate between the component of body fat, the type and location of fat in the body, or the degree of metabolic diseases that it can cause. In the clinical setting, high muscle mass/low fat has been associated with improved survival in patients with cardiovascular disease [39]. Accordingly, loss of muscle mass has been associated with increased mortality in patients undergoing TAVI [40]. That is why future studies are necessary to determine the most favourable body weight to improve outcome after TAVI.

This study has several important limitations. First, this retrospective, single-centre analysis is subjected to the limitations common to this type of analysis. An observational analysis, including the current study, cannot prove or disprove the existence of a paradoxical relationship between BMI and mortality. Second, we used BMI as a surrogate of body weight. However, combining BMI and measures of central obesity, such as waist circumference and waist-to-hip ratio, may be more valuable in the assessment of mortality risk after TAVI, since central obesity predicts mortality more reliably than BMI alone in patients with coronary heart disease [41]. Third, in contrast to other findings, only overweight was associated with decreased mortality in patients in our cohort. However, it is possible that patients with a more severe profile of comorbidities and a high surgical risk for TAVI were refused to undergo TAVI, which could have affected current results. Fourth, we may not have included all possible (unknown) confounding factors that may have influenced the results. For instance, to address the likely bias attributed to patients with cachexia, we excluded patients with BMI < 18.5. However, as most elderly suffer from lower muscle mass, this will introduce bias attributable to unassured confounders [40]. Fifth, we could not address the effect of BMI changes over time, which may have influenced our results. Finally, although late mortality after TAVI may be attributed to the non-cardiac causes [42], we could not address the cause of death in patients included in this study. Future studies are needed to evaluate cause of death to provide a better understanding of the mechanism of the observed association between BMI and mortality.

Being overweight is associated with improved survival after TAVI. Furthermore, TAVI is safe in different BMI groups with respect to the postoperative complications rate.