Body composition in patients with Fontan physiology: a systematic review

This study was conducted according to the PRISMA guidelines [13]. A systematic search was performed in online databases on 13 April 2023. The search is presented in Supplement 1. The searched databases were Embase, Medline Ovid, Web of Science, and Cochrane CENTRAL. The search terms consisted of keywords related to Fontan circulation and body composition values. No filters were applied.

All studies were screened on title and abstract by two independent reviewers (JNP and RJB). Studies were included when they described body composition measurements by compartment in patients with a Fontan circulation. The full text of relevant articles was assessed for inclusion in the systematic review. The exclusion criteria were as follows: (a) studies without patients with a Fontan circulation, (b) body composition values of interest (body fat percentage, fat mass (index), lean mass (index), appendicular lean mass (index), or leg lean mass (index)) were not mentioned, (c) body composition values not specifically matched to patients with a Fontan circulation, and (d) results from conference papers and abstracts without a full text available. Discrepancies between the reviewers were discussed and resolved together with a third reviewer (LES).

From the resulting articles, two reviewers (JNP and RJB) extracted all data needed independently using Excel (version 2019). The extracted data from each reviewer was verified by the extracted data from the other reviewer; any discrepancies were discussed together with a third author (LES). The following data were extracted: study characteristics—journal, year of publication, study design, study size (of the patient group and, if present, of the control group), time frame of inclusion, and location or database; patient characteristics—age, age at Fontan completion, sex, Fontan type, ventricle type, and the method used to compare the body composition of the patients with a Fontan circulation (control group or pre-existing reference values); methodology used for measuring the body composition; measurement outcomes—body mass index (BMI), body fat percentage, appendicular lean mass (index), leg lean mass (index), fat mass (index), and lean mass (index); conclusions of studies related to the relationship between body composition and (adverse) outcomes (such as re-interventions and cardiac failure); and conclusions related to the body composition measurements. Studies were divided into groups and summarized based on the age of the study population—adults only, children only, and both adults and children.

The Strengthening Reporting of Observational Studies in Epidemiology (STROBE) checklist was used to assess the quality of cohort and cross-sectional studies [14].

A total of 1392 studies were identified through the search. After removing duplicates and screening on title and abstract, the full text of 29 articles was screened. A total of 18 studies were included in this systematic review [5, 15‐31]. This process is presented in Fig. 1.

None of the studies were excluded based on quality. All studies met over 75% of the STROBE checklist criteria. The quality assessment is presented in Supplement 2.

The study designs of all included studies are described in Table 1, and patient characteristics in Table 2. In total, body composition was measured in 774 patients with a Fontan circulation, and 298 healthy participants served as controls. Five studies included children only [15, 16, 24, 27, 28], five studies included adults only [5, 17, 18, 25, 32], and eight studies included both children and adults [19‐23, 26, 30, 31]. The mean age of the patients between studies ranged from 1.5 to 30 years; 45.0% of the participants was female. The mean age at Fontan completion was 4.44 years. The percentage of a dominant left ventricle was 52.6%, and the percentage of patients with an open fenestration was 49.8%.

The measurement methods are summarized in Table 1. Twelve studies assessed the body composition using dual-energy X-ray absorptiometry (DXA) [5, 15, 16, 18‐22, 27, 28, 30‐32]. Five studies estimated the body composition using bioelectrical impedance analysis (BIA) [17, 23, 25, 28, 32]. One study assessed the body composition using peripheral quantitative computed tomography (CT) taken at a specific location on the left tibia, corresponding to 66% site of its total length [24]. One study assessed the body composition using liver or abdominal magnetic resonance (MR) at levels thoracic vertebrae 12 (T12) and lumbar vertebrae 3 (L3) [26]. Seven studies recruited control patients [15, 16, 18, 23‐25, 30]. Nine studies used reference values to compare their patient group [5, 19‐22, 24, 27, 31]. Two studies did not compare the body composition with control patients or reference values [17, 32]. Eight out of eighteen studies [5, 18‐21, 23, 27] used Z-scores. The studies indexed their results on sex and age [5, 18, 19], sex, age, race, height, BMI, and leg-length [20‐22, 27] or did not mention how they normalized their results [23].

Out of the 18 included studies, 17 calculated the BMI of Fontan patients and controls or reference values [5, 15, 16, 18‐28, 30‐32]. These outcomes are documented in Table 3. Nine studies measured fat mass in patients with a Fontan circulation [5, 15, 17, 19, 21, 23, 25, 30, 32], and 15 measured lean/muscle mass in patients with a Fontan circulation [5, 15, 16, 18‐28, 31]. Outcomes regarding fat, lean, and muscle mass can also be found in Table 3. All measurement outcomes are summarized in Table 4. Measured BMI was within normal range and not significantly different from control patients or reference values in fifteen out of seventeen studies [5, 15, 16, 18‐28, 30‐32]. Five out of nine studies concluded that patients with a Fontan circulation had a higher fat mass [5, 15, 19, 23, 31]. Four studies concluded that Fontan patients had a comparable lean/muscle mass to controls/reference values [16, 24, 26, 30]. Twelve out of fifteen studies concluded that patients with a Fontan circulation had a reduced (skeletal) muscle mass and/or lower lean mass [5, 15, 18‐23, 25‐27, 31].

Five studies investigated associations between body composition and cardiac function [5, 19, 20, 23, 25]. Tran et al. showed that the degree of muscle deficit is associated with ventricular systolic impairment [5]. This corresponds to the findings of the other four studies. A higher leg lean mass was associated with larger increases of cardiac index and indexed systemic flow during exercise [20]. Higher body fat was associated with poorer New York Heart Association class (III or IV) in study of Shiina et al. [25]. Also, various Fontan complications including cardiac failure (New York Heart Association class III/IV), protein-losing enteropathy, plastic bronchitis, heart transplant, and death were associated with increased adiposity and/or lower muscle mass in the studies of Cao et al. and Powell et al. [19, 23]. Cao et al. demonstrated that every 1% increase in body fat percentage was associated with a 10% increased risk of reaching a Fontan complication. In total, eight studies investigated the relation between body composition and exercise capacity [5, 17, 18, 20, 23, 26]. Higher muscle mass was correlated with higher exercise capacity (Peak VO2) in three studies; higher leg lean mass showed the same association [5, 18, 23, 26]. Avitabile et al. found a correlation between skeletal muscle deficits and a reduced exercise capacity [20]. Patients with an improving exercise capacity over time had a lower body fat percentage compared to patients with a Fontan circulation that diminished in exercise capacity over time [17]. Lastly, Vaikunth et al. also showed that lower leg lean mass was associated with a lower whole body bone mass density [22]. Two studies investigated the effect of exercise training on (leg) lean muscle mass, both showing no change [27, 29].

The findings of this systematic review are heterogeneous. Reviewed studies included both children and adults, and different techniques were used to measure body composition. Most studies used DXA or BIA to measure the body composition. The use of DXA to measure different body compartments has been described and validated against the gold standard (whole-body potassium counting) in multiple clinical studies, showing comparable outcomes [34]. As DXA scans are easier, cheaper, and less stressful compared to whole-body potassium counting for the patient, DXA is the recommended measuring methods for assessing body composition in both research and clinical setting [34]. However, the use of DXA is limited to the research/ clinical context due to equipment cost, exposure to radiation (although very low), and lack of portability [35]. BIA has been proven to be one of the most practical methods to estimate body composition, as costs are low, its quick assessment procedures, and accessibility [36]. However, measurements obtained with BIA are less accurate and influenced by sex, ethnicity, weight, and age [37]. Also, diseases which change serum electrolytes, blood flow, and fluid distribution affect the accuracy of BIA measurements, ultimately making BIA less accurate for Fontan patients [37]. MRI and CT were used in only two studies [23, 24]. MRI and CT can measure both whole-body composition and body composition of a certain area. A study from Kullberg et al. showed strong correlations between MRI, CT, and DXA; however, both MRI and CT have major disadvantages [38]. MRI measurements might be highly dependent on used scanning protocols and are very expensive and time-consuming. Although CT is much faster, exposure to radiation is high; in addition, both techniques are very costly. Surprisingly, our systematic review did not find any reporting on the use of air-displacement plethysmography or skinfold measurements in patients with a Fontan circulation.

As mentioned earlier, BMI is the most commonly used measurement to define adiposity [39]. However, BMI can be misleading as a combination of reduced lean mass and increased adiposity can result in a “normal” BMI, as was confirmed by this systematic review. In most included studies, patients with a Fontan circulation had an unfavorable body composition profile characterized by reduced skeletal muscle mass and increased adiposity, despite a normal BMI. The only two studies reporting normal muscle mass compared to the healthy population were performed in children. These findings may have important diagnostic and prognostic implications. In the general population, an unfavorable body composition is associated with various morbidities, including diabetes, cardiovascular disease, cancer, and early death [40‐42]. Studies included in this systematic review also found an association between unfavorable body composition in Fontan patients and adverse outcomes, including cardiac failure [19, 23]. This is particularly relevant since several studies have shown the importance of the skeletal muscle pump to increase venous return during exercise [18, 20]. The study by Avitabile et al. showed that having a higher leg lean mass was associated with increases in cardiac index during exercise, suggesting that Fontan patients with higher leg lean mass may indeed be better able to augment systemic output during exercise [20]. In addition, all studies investigating the association between (leg) lean mass and exercise capacity included in this systematic review found that a reduced lean mass was associated with worse exercise capacity [5, 18, 23, 26]. While this may be intuitive, these findings stress the importance of maintaining and improving muscle mass in these patients. Causes of muscle mass deficits are poorly defined in patients with a Fontan circulation. Relative deconditioning, chronically elevated central venous pressure, decreased physical activity levels and neurohormonal activation, and malnutrition might be contributing factors [43‐45]. As prospective studies are currently lacking, the question also remains whether worse Fontan outcomes are caused by less physical activity leading to an unfavorable body composition, or the other way around. A way to improve body composition in the healthy population is by performing moderate to vigorous physical activity [46, 47]. Exercise in patients with a Fontan circulation has been shown to improve exercise capacity, cardiac function and quality of life [9]. While there are few existing data supporting any particular approach, establishing a healthy lifestyle in Fontan patients, is likely to have a substantive positive impact. Recent studies from both Avitabile et al. and Chemello et al. measured body composition respectively before and after lower leg-focused muscle exercise and an 6-month lasting exercise intervention; both studies did not find a significant change [27, 29].

Several limitations of the studies included have to be discussed. Patient characteristics were not always reported in detail, and information regarding physical activity levels of the included patients was lacking. The majority of the studies did not include a healthy control group but compared the body composition outcomes of the patients to reference values. Information regarding the exact origin of these reference values and methods used to index mass was often lacking, some studies did not correct their results for height at all which is especially important in studies including children. Some results were indexed very well by certain notable exceptions [20‐22, 27]. Two studies did not compare body composition outcomes to their healthy population, making it hard to draw definite conclusions from their data [17, 28]. Body composition outcomes of children and adults (in the mixed studies) and males/females were not reported separately; also cohorts were homogeneous. We could not rule out overlap of patients between studies for all studies included in our review. Hansson et al. and Sandberg et al. included patients from the Northern part of Sweden and Stockholm [15, 16]. The patient group of four studies consisted of patients from the Children’s Hospital of Philadelphia [20‐22, 27]. Three studies used the same registry to collect patient data [5, 19, 31]. Lastly, there was a high inter-study variability for measured outcomes; studies reported raw values, indexed values for height squared, or Z-scores, making absolute values hard to compare.

Future studies should be prospective to not only demonstrate associations but also investigate causes and include homogenous cohorts to minimalize confounders. Patients should be followed longitudinally, as this can provide a better insight into the development of the body composition over time and its role as a predictor for an adverse event. Importantly, means to improve the unfavorable body composition in patients with a Fontan circulation should be explored further.