Overweight and Obesity in Patients with Congenital Heart Disease: A Systematic Review

Background: Overweight and obesity have become a major public health concern in recent decades, particularly in patients with chronic health conditions like congenital heart disease (CHD). This systematic review elaborates on the prevalence and the longitudinal development of overweight and obesity in children and adults with CHD. Methods: A systematic literature search was conducted in PubMed, Cochrane, and Scopus from January 2010 to December 2020 on overweight and obesity prevalence in children and adults with CHD. Results: Of 30 included studies, 15 studies evaluated 5680 pediatric patients with CHD, 9 studies evaluated 6657 adults with CHD (ACHD) and 6 studies examined 9273 both pediatric patients and ACHD. Fifteen studies received the quality rating “good”, nine studies “fair”, and six studies “poor”. In children with CHD, overweight prevalence was between 9.5–31.5%, and obesity prevalence was between 9.5–26%; in ACHD, overweight prevalence was between 22–53%, and obesity was between 7–26%. The prevalence of overweight and obesity was thereby similar to the general population. Overweight and obesity have been shown to increase with age. Conclusion: The prevalence of overweight and obesity in children and adults with CHD is similar to the general population, demonstrating that the growing obesity pandemic is also affecting the CHD population.


Introduction
Overweight and obesity have become major public health concerns in recent decades [1] and are referred to as a global pandemic [2]. The prevalence of childhood obesity has more than doubled in the last three decades [3], which is particularly concerning as childhood obesity tracks into adulthood [4]. Overweight and obesity, in particular, are independent risk factors for cardiovascular diseases and are associated with increased risk of morbidity and mortality in the general population [5]. As the majority of children with congenital heart disease (CHD) nowadays survive into adulthood [6], environmental and behavioral health risk factors are also of concern in this patient cohort [7,8]. As patients with preexisting heart conditions are prone to cardiovascular events [9] due to underlying anatomic and functional abnormalities [10], overweight and obesity need to be considered here too.
At first glance, one would assume that patients with CHD are at increased risk for overweight and obesity because, as cardiac patients, physical activity may be reduced due to exercise restriction or overprotection, and patients participate less in leisure and competitive sports [11]. On the other hand, surgeries, hospitalization, and rehabilitation early in life affect the development of children with CHD, which can be linked to underweight and 1.

Data Collection
Two reviewers screened the relevant articles for title and abstract that had to fulfill the basic inclusion criteria: diagnosed CHD and the measurement of overweight and obesity. At least one of the reviewers had to consider a reference eligible; in case of disagreement, a third reviewer was consulted for a majority decision before full-text analysis.
Documenting critical appraisal of the included literature, reviewers used the "Study Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies" of the National Heart, Lung and Blood Institute (NHLBI), consisting of a 14-item list assessing potential risk for bias. Accordingly, included studies were categorized as good, fair, or poor [16].

Study Inclusion
In total, 982 potential studies were identified in the initial search, of which 905 remained after duplicates had been removed. After screening titles and abstracts, 51 potential studies were retrieved for full-text analysis, of which 21 studies were excluded due to: main focus on the impact of overweight and obesity (n = 12), lacking focus on overweight and obesity (n = 5), focus on underweight (n = 1) and not exclusively assessing patients with CHD (n = 3). Finally, 30 studies with a total of 21,610 CHD patients met the inclusion criteria for this systematic review. The search algorithm and selection process is displayed in Figure 1.
Twenty-six of the included studies were cross-sectional studies that examined the prevalence of overweight and obesity in patients with CHD , and four were cohort studies that investigated the longitudinal change of the prevalence across the lifespan [10,[43][44][45]. The body composition of patients with CHD was compared to a healthy reference cohort in 13 of the included studies [20][21][22][23][24][28][29][30]33,35,40,41,45], whereas 9 studies merely investigated the prevalence of overweight and obesity in patients with

Study Quality
According to the NHLBI study quality assessment tool [16], 15 studies received the quality rating "good", while 9 studies were rated as "fair". Six studies showed a substantial risk of bias and were rated as "poor". Various contributing factors have an impact on weight status and should therefore be considered when investigating the prevalence of overweight and obesity. The included studies in this review controlled for age [21,24,35,37,39], age and lesion severity [27], age and sex [10,17,18,20,23,28,31,33,36,38,40,41,43,44], race/ ethnicity [29,45], family history, parent's nutritional status [19], marital status, educational level, and geographic region [34]. Six studies did not take confounders into account and were therefore rated as "poor" [8,22,25,26,37,42]. Comprehensive information on the quality rating can be found in Table 2. In the first 5 y of age and in the first 5 y postoperatively, the ToF cohort had a significantly lower BMI compared with the control group (p = 0.042 and p =0.028). Afterward, no sig. difference between CHD & healthy controls (p =0.079).
9.5/26.2 % of children who were overweight/obese ranged from 22% to 44%, with the lowest incidence in those with moderate disease or that were surgically corrected and the highest incidence in those with mild disease.

31.5/16.4
Children with overweight/obesity were older (p < 0.001), had lower median household income (p = 0.031), and more often complex CHD (p = 0.008). Children with CHD have an increased risk of becoming overweight & obese in early childhood. (14.4) BMI did not differ between CHD and children in the general population. Greater obesity in children with mild heart disease (p = 0.04). Sedentary behaviors, cardiomegaly, and the NYHA class II-IV were associated with being overweight/obese.
At age 5 y, patients with CoA had significantly greater BMI z-scores compared with age-sex matched normal data (p < 0.001). The proportion of excess weight in COA significantly increased over time (p < 0.001). Adults with repaired COA developed obesity at a greater rate than those with either AS (p = 0.004) or with PS or ASD (p < 0.001).
Adults: 22/17 (Children: 15) The likelihood of being overweight/obese as an adult was three times higher if there was a BMI ≥ 85th percentile in childhood (p < 0.01). Pediatric rates of overweight/obesity comparable to healthy controls. No race or gender differences between overweight/ obese. Overweight/obesity in adulthood was associated with lower heart failure rates (4 vs.19%, p = 0.03).  Mean BMI 18.9 ± 4.7 kg/m 2 CHD participants had a significantly greater waist circumference than controls when controlling for sex, birth weight, physical activity, and total lean mass. CHD and control groups were similar in BMI, total fat mass, total lean mass, percent fat mass, percent lean mass. In the first 5 y of age and in the first 5 y postoperatively, the ToF cohort had a significantly lower BMI compared with the control group (p = 0.042 and p =0.028). Afterward, no sig. difference between CHD & healthy controls (p =0.079).
9.5/26.2 % of children who were overweight/obese ranged from 22% to 44%, with the lowest incidence in those with moderate disease or that were surgically corrected and the highest incidence in those with mild disease.

31.5/16.4
Children with overweight/obesity were older (p < 0.001), had lower median household income (p = 0.031), and more often complex CHD (p = 0.008). Children with CHD have an increased risk of becoming overweight & obese in early childhood. (14.4) BMI did not differ between CHD and children in the general population. Greater obesity in children with mild heart disease (p = 0.04). Sedentary behaviors, cardiomegaly, and the NYHA class II-IV were associated with being overweight/obese.
At age 5 y, patients with CoA had significantly greater BMI z-scores compared with age-sex matched normal data (p < 0.001). The proportion of excess weight in COA significantly increased over time (p < 0.001). Adults with repaired COA developed obesity at a greater rate than those with either AS (p = 0.004) or with PS or ASD (p < 0.001).
Adults: 22/17 (Children: 15) The likelihood of being overweight/obese as an adult was three times higher if there was a BMI ≥ 85th percentile in childhood (p < 0.01). Pediatric rates of overweight/obesity comparable to healthy controls. No race or gender differences between overweight/ obese. Overweight/obesity in adulthood was associated with lower heart failure rates (4 vs.19%, p = 0.03). The prevalence of overweight/obesity in CHD adolescents close to controls. In ♀ fewer CHD children were overweight/obese (12.2% vs. 18.7%, p = 0.002). ♂ with moderate to severe CHD had a lower prevalence of overweight/obesity (p = 0.025). Children with cyanotic CHD had significantly lower prevalence of overweight/ obesity (1.5% vs. 15.5%, p = 0.003) than those with non-cyanotic. The likelihood of being overweight/obese as an adult was three times higher if there was a BMI ≥ 85th percentile in childhood (p < 0.01). Pediatric rates of overweight/obesity comparable to healthy controls. No race or gender differences between overweight/ obese. Overweight/obesity in adulthood was associated with lower heart failure rates (4 vs.19%, p = 0.03).  (7), ToF (5) (7), ToF (5) 10.9 ± 2.6 -

WC & DXA
Mean BMI 18.9 ± 4.7 kg/m 2 CHD participants had a significantly greater waist circumference than controls when controlling for sex, birth weight, physical activity, and total lean mass. CHD and control groups were similar in BMI, total fat mass, total lean mass, percent fat mass, percent lean mass.
Welisch et al. 2013 [40] 1080 (483)   In the first 5 y of age and in the first 5 y postoperatively, the ToF cohort had a significantly lower BMI compared with the control group (p = 0.042 and p =0.028). Afterward, no sig. difference between CHD & healthy controls (p =0.079).
9.5/26.2 % of children who were overweight/obese ranged from 22% to 44%, with the lowest incidence in those with moderate disease or that were surgically corrected and the highest incidence in those with mild disease.

31.5/16.4
Children with overweight/obesity were older (p < 0.001), had lower median household income (p = 0.031), and more often complex CHD (p = 0.008). Children with CHD have an increased risk of becoming overweight & obese in early childhood.
BMI did not differ between CHD and children in the general population. Greater obesity in children with mild heart disease (p = 0.04). Sedentary behaviors, cardiomegaly, and the NYHA class II-IV were associated with being overweight/obese.
At age 5 y, patients with CoA had significantly greater BMI z-scores compared with age-sex matched normal data (p < 0.001). The proportion of excess weight in COA significantly increased over time (p < 0.001). Adults with repaired COA developed obesity at a greater rate than those with either AS (p = 0.004) or with PS or ASD (p < 0.001).
Adults: 22/17 (Children: 15) The likelihood of being overweight/obese as an adult was three times higher if there was a BMI ≥ 85th percentile in childhood (p < 0.01). Pediatric rates of overweight/obesity comparable to healthy controls. No race or gender differences between overweight/ obese. Overweight/obesity in adulthood was associated with lower heart failure rates (4 vs.19%, p = 0.03). The prevalence of overweight/obesity in CHD adolescents close to controls. In ♀ fewer CHD children were overweight/obese (12.2% vs. 18.7%, p = 0.002). ♂ with moderate to severe CHD had a lower prevalence of overweight/obesity (p = 0.025). Children with cyanotic CHD had significantly lower prevalence of overweight/ obesity (1.5% vs. 15.5%, p = 0.003) than those with non-cyanotic. The likelihood of being overweight/obese as an adult was three times higher if there was a BMI ≥ 85th percentile in childhood (p < 0.01). Pediatric rates of overweight/obesity comparable to healthy controls. No race or gender differences between overweight/ obese. Overweight/obesity in adulthood was associated with lower heart failure rates (4 vs.19%, p = 0.03). In the first 5 y of age and in the first 5 y postoperatively, the ToF cohort had a significantly lower BMI compared with the control group (p = 0.042 and p =0.028). Afterward, no sig. difference between CHD & healthy controls (p =0.079).
(Child: 14.5, Adolescent: 26.5%) The prevalence of overweight/obesity in CHD adolescents close to controls. In ♀ fewer CHD children were overweight/obese (12.2% vs. 18.7%, p = 0.002). ♂ with moderate to severe CHD had a lower prevalence of overweight/obesity (p = 0.025). Children with cyanotic CHD had significantly lower prevalence of overweight/ obesity (1.5% vs. 15.5%, p = 0.003) than those with non-cyanotic. BMI did not differ between CHD and children in the general population. Greater obesity in children with mild heart disease (p = 0.04). Sedentary behaviors, cardiomegaly, and the NYHA class II-IV were associated with being overweight/obese.
At age 5 y, patients with CoA had significantly greater BMI z-scores compared with age-sex matched normal data (p < 0.001). The proportion of excess weight in COA significantly increased over time (p < 0.001). Adults with repaired COA developed obesity at a greater rate than those with either AS (p = 0.004) or with PS or ASD (p < 0.001).
Adults: 22/17 (Children: 15) The likelihood of being overweight/obese as an adult was three times higher if there was a BMI ≥ 85th percentile in childhood (p < 0.01). Pediatric rates of overweight/obesity comparable to healthy controls. No race or gender differences between overweight/ obese. Overweight/obesity in adulthood was associated with lower heart failure rates (4 vs.19%, p = 0.03).  The prevalence of overweight/obesity in CHD adolescents controls. In ♀ fewer CHD children were overweight/obese ( 18.7%, p = 0.002). ♂ with moderate to severe CHD had a low alence of overweight/obesity (p = 0.025). Children with cyano had significantly lower prevalence of overweight/ obesity ( 15.5%, p = 0.003) than those with non-cyanotic. Increasing exercise duration was associated with lower BMI (p = 0.01). Restriction to mild exertion and participation in low-intensity exercise were both associated with increased BMI.  Mean BMI% did not differ between CHD sample and paired controls over a 5 y period. Significant increased BMI% change in the age cohort of 5-10 y (p = 0.04), in   The proportion of individuals with overweight/obesity increased with age (p < 0.001). A higher proportion of individuals with moderate lesion severity (29%) had obesity compared to simple (24%) and complex (18%, p < 0.001) lesion severity In the first 5 y of age and in the first 5 y postoperatively, the ToF cohort had a significantly lower BMI compared with the control group (p = 0.042 and p =0.028). Afterward, no sig. difference between CHD & healthy controls (p =0.079).

9.5/26.2
% of children who were overweight/obese ranged from 22% to 44%, with the lowest incidence in those with moderate disease or that were surgically corrected and the highest incidence in those with mild disease.

31.5/16.4
Children with overweight/obesity were older (p < 0.001), had lower median household income (p = 0.031), and more often complex CHD (p = 0.008). Children with CHD have an increased risk of becoming overweight & obese in early childhood. (14.4) BMI did not differ between CHD and children in the general population. Greater obesity in children with mild heart disease (p = 0.04). Sedentary behaviors, cardiomegaly, and the NYHA class II-IV were associated with being overweight/obese.
At age 5 y, patients with CoA had significantly greater BMI z-scores compared with age-sex matched normal data (p < 0.001). The proportion of excess weight in COA significantly increased over time (p < 0.001). Adults with repaired COA developed obesity at a greater rate than those with either AS (p = 0.004) or with PS or ASD (p < 0.001).

Adults: 22/17 (Children: 15)
The likelihood of being overweight/obese as an adult was three times higher if there was a BMI ≥ 85th percentile in childhood (p < 0.01). Pediatric rates of overweight/obesity comparable to healthy controls. No race or gender differences between overweight/ obese. Overweight/obesity in adulthood was associated with lower heart failure rates (4 vs.19%, p = 0.03). Children with overweight/obesity were older (p < 0.001), had lower median household income (p = 0.031), and more often complex CHD (p = 0.008). Children with CHD have an increased risk of becoming overweight & obese in early childhood. The likelihood of being overweight/obese as an adult was three times higher if there was a BMI ≥ 85th percentile in childhood (p < 0.01). Pediatric rates of overweight/obesity comparable to healthy controls. No race or gender differences between overweight/ obese. Overweight/obesity in adulthood was associated with lower heart failure rates (4 vs.19%, p = 0.03). Increasing exercise duration was associated with lower BMI (p = 0.01). Restriction to mild exertion and participation in low-intensity exercise were both associated with increased BMI.
10.9 ± 2.6 -WC & DXA Mean BMI 18.9 ± 4.7 kg/m 2 CHD participants had a significantly greater waist circumference than controls when controlling for sex, birth weight, physical activity, and total lean mass. CHD and control groups were similar in BMI, total fat mass, total lean mass, percent fat mass, percent lean mass. Excess weight was more common among ♂ (60%). Family history of obesity was associated with excess weight (p = 0.001). In patients with cyanotic lesions, overweight was less frequent than in acyanotic (23.3% vs. 27.7%).  The prevalence of overweight/obesity in CHD adolescents close to controls. In ♀ fewer CHD children were overweight/obese (12.2% vs. 18.7%, p = 0.002). ♂ with moderate to severe CHD had a lower prevalence of overweight/obesity (p = 0.025). Children with cyanotic CHD had significantly lower prevalence of overweight/ obesity (1.5% vs. 15.5%, p = 0.003) than those with non-cyanotic.  The obesity rate was similar between matched ACHD and healthy controls. Metabolic syndrome was more common in ACHD patients than in controls (15.0% versus 7.4%). Obese ACHD patients were more likely to have metabolic syndrome than obese controls (93.1% vs. 44%). In the first 5 y of age and in the first 5 y postoperatively, the ToF cohort had a significantly lower BMI compared with the control group (p = 0.042 and p =0.028). Afterward, no sig. difference between CHD & healthy controls (p =0.079).
9.5/26.2 % of children who were overweight/obese ranged from 22% to 44%, with the lowest incidence in those with moderate disease or that were surgically corrected and the highest incidence in those with mild disease.

31.5/16.4
Children with overweight/obesity were older (p < 0.001), had lower median household income (p = 0.031), and more often complex CHD (p = 0.008). Children with CHD have an increased risk of becoming overweight & obese in early childhood. (14.4) BMI did not differ between CHD and children in the general population. Greater obesity in children with mild heart disease (p = 0.04). Sedentary behaviors, cardiomegaly, and the NYHA class II-IV were associated with being overweight/obese.
At age 5 y, patients with CoA had significantly greater BMI z-scores compared with age-sex matched normal data (p < 0.001). The proportion of excess weight in COA significantly increased over time (p < 0.001). Adults with repaired COA developed obesity at a greater rate than those with either AS (p = 0.004) or with PS or ASD (p < 0.001).
Adults: 22/17 (Children: 15) The likelihood of being overweight/obese as an adult was three times higher if there was a BMI ≥ 85th percentile in childhood (p < 0.01). Pediatric rates of overweight/obesity comparable to healthy controls. No race or gender differences between overweight/ obese. Overweight/obesity in adulthood was associated with lower heart failure rates (4 vs.19%, p = 0.03).
One study reported decreased BMI in CHD children [23], whereas another study reported increased obesity rates in CHD children compared to healthy controls [33]. CHD participants had a significantly greater waist circumference than controls when controlling for sex, birth weight, physical activity score, and total lean mass [20].
Overweight and obesity occurred more frequently in CHD boys than in girls [19,23]. There was no association between overweight and obesity prevalence and CHD severity or surgical status in CHD children [26,40]. Children with cyanotic CHD had a lower prevalence of overweight and obesity than those with non-cyanotic CHD [19,23]. Significantly lower BMI values were found in children with tetralogy of Fallot five years after surgery compared to age-matched controls [21]. Patients with Transposition of the Great Arteries and Coarctation of the Aorta showed significantly higher BMI compared with norm data and other cardiac diagnoses [38,44]. Studies on Fontan children reported a lower prevalence of overweight and obesity at the time of Fontan procedure, which increased to 30% five years post-surgery [10], but showed no significant difference to healthy controls [35].

Overweight and Obesity in Adults with CHD
In adults with CHD, overweight prevalence was found in a range from 22% [24] to 53% [30], while obesity was reported in a range between 7% [22] and 26% [29]. Studies comparing overweight and obesity prevalence with healthy controls found either similar overweight and obesity rates [22,29,43] or lower prevalence compared to the reference cohort [31,42]. However, ACHD showed a decreased prevalence of morbid obesity compared to matched controls [29].
Two studies found no association between CHD severity and prevalence of overweight and obesity [26,29], whereas two other studies reported a reduced prevalence in the more complex classes like total cavopulmonary connection compared to the reference population [31,42]. One study on ACHD reported men being more overweight than women [31], while another study found no sex differences [34]. Hispanic patients were more likely to be obese than white ACHD after adjusting for age and gender [29]. Non-Hispanic Blacks with CHD had a 58% increased risk of obesity in young adulthood and 33% in late adulthood compared to white CHD [28].
BMI was positively associated with impaired NYHA class and cardiovascular medication intake [42]. Metabolic syndrome was more common in ACHD patients than in controls, and obese ACHD patients were more likely to have metabolic syndrome than obese controls [43].

Longitudinal Development of Overweight and Obesity
The number of overweight and obese children increased during adolescence [27,44]. Also, in adults with CHD, age correlated with increased BMI [29,31,34,42], with similar results found in the general population [29,31]. Generally, adults had significantly higher BMI scores and were more likely to be overweight or obese compared with younger samples [26]. The odds of being overweight or obese as an adult were found to be three times higher if their childhood BMI was already above the 85th percentile [24].

Discussion
This systematic review shows that the overall distribution of overweight and obesity is consistent with the prevalence in the general population, in which overweight and obesity rates have increased dramatically in recent decades [37]. Particularly, CHD patients with increasing age, male gender, and Hispanic origin are at risk for overweight and adiposity. These findings are worrisome, as the CHD population is at higher risk of developing metabolic syndrome and premature morbidity and mortality compared to their healthy peers [43]. Obesity is associated with an increased risk of cardiovascular diseases, like heart failure and coronary artery disease [46]. Excessive body adipose tissue impacts the vessel wall by changes in blood pressure, glucose level, lipid metabolism, and systemic inflammation [47]. In children and adults with CHD, overweight and obesity were shown to be associated with cardiac comorbidities, increased cardiac medication intake, and higher systolic and diastolic blood pressure compared to normal-weight patients. According to this recently published study, the elevated risk of morbidity and mortality in the CHD population is exacerbated by obesity [27]. Obesity has been associated with adverse perioperative outcomes [48] as well as adverse short-term outcomes after cardiothoracic surgery in patients with CHD [49]. In contrast, another study on ACHD reported that overweight and moderate obesity were associated with reduced mortality rates, especially in symptomatic ACHD patients and those with complex underlying cardiac defects, replicating the so-called "obesity paradox" in the general population [14]. It was also shown that overweight and obesity are associated with lower heart failure rates in ACHD [24]. In the elderly, this may indeed be true, as a slightly elevated weight above normal BMI can be a resource to build upon when facing surgical procedures or beginning aggressive drug therapy. However, overweight and obesity are among the most significant contributors to illness and adverse health outcomes and should be prevented or reduced at all costs, at least in younger patients with CHD.
Underlying causes and effects of the high overweight and obesity prevalence in the CHD population are multifactorial. Heart failure-associated reasons such as medication intake were identified along with the already mentioned behavioral factors. Two major factors influencing the CHD population are physical inactivity due to restrictions or overprotection by physicians and parents, as well as weight gain interventions in infancy, which can develop into excess weight later in life [17]. Part of the included studies examined the association between overweight and obesity and contributing factors. Yang et al. reported an association between sedentary behavior and increased overweight and obesity prevalence in CHD children [41]. BMI has been shown to be negatively associated with exercise duration [32] and exercise capacity in the CHD population [42].
Continuous monitoring of weight status and interventions to reduce the prevalence of overweight and obesity are required in this patient cohort at high risk for acquired cardiovascular disease [37]. Interventions such as lifestyle and nutritional counseling to reduce risk factors for obesity should begin in early childhood and include parental education.

Limitations and Further Research
The classification of overweight and obesity in the majority of included studies refers to the BMI, which, however, does not analyze fat distribution and body composition. Abdominal fat, in particular, has been associated with various pathologies [45]. Therefore, further studies should specifically address abdominal fat in order to quantify the visceral component of adipose tissue [31]. Moreover, BMI is based on both fat mass and lean mass and therefore carries the risk of classifying muscular patients as overweight [37].
The majority of included studies have not considered race and ethnicity as potential confounders, even though their influence on weight status is well established [44].

Conclusions
The prevalence of overweight and obesity in children and adults with CHD reflects that of the general population, demonstrating that the growing obesity pandemic is also affecting the CHD population. Such a high prevalence of overweight and obesity is particularly worrying in the context of CHD. We hope that these results will raise awareness of this issue and encourage appropriate health-promoting interventions and individual consultations.