Functional Capacity in Congenital Heart Disease: A Systematic Review and Meta-Analysis

Background Children and adolescents with congenital heart disease often have alterations in their exercise capacity that can be evaluated by various functional testing. Objective To evaluate the functional capacity of children and adolescents with congenital heart disease (CHD) with systematic review and meta-analyses. Methods The review included observational studies, data from the first evaluation of randomized clinical trials or observational follow-up periods after clinical trials which evaluated functional capacity by cardiopulmonary exercise test, stress testing, six-minute walk test or step test, in children and adolescents with CHD, aged between six and 18 years, and comparisons with healthy controls in the same age group. The quantitative assessment was performed by meta-analysis, by comparing the maximal oxygen consumption (VO2max) of children and adolescents with CHD and respective control groups. Results Twenty-five of 2.683 studies identified in the search met the inclusion criteria. The VO2max measurement showed that patients with CHD have a decrease of 9.31 ml/Kg/min (95% CI. -12.48 to -6.13; I2, 94.3%, P for heterogeneity < 0.001) compared with the control group. The meta-analysis of the data of maximum heart rate (HR) reached during cardiopulmonary test and stress testing, retrieved from 18 studies, showed a HR value of -15.14 bpm (95% CI. -20.97 to -9.31; I2, 94.3%, P for heterogeneity < 0.001) compared with the control group. Conclusion Children and adolescents with CHD have lower VO2max and HR compared to controls.


Introduction
Children with congenital heart disease (CHD) often have a sedentary lifestyle that may reflect both inherent physiological limitations in addition to overprotection of parents. 1 Such lifestyle pattern is likely to be maintained throughout adulthood, which can result in increased risk for cardiovascular diseases. 1 In children with restriction for physical activity practice, there is an increased risk for overweight and there is an increasing in overweight (RR, 2.51; 95% CI, 1.24-3.52) and obesity (RR, 6.14; 95% CI, 2.54-8.82) at follow-up. 2 Functional capacity may indicate cardiovascular, pulmonary or motor dysfunction. In children with chronic disease, maximal oxygen consumption (VO 2 max) can predict adverse outcomes as well as the greater aerobic fitness is associated with a nearly 10% risk reduction for hospitalization of children with cystic fibrosis. 3 The assessment of functional capacity in patients with heart disease is an important clinical tool for diagnosis, quantification of symptoms, prognosis and evaluation of response to treatment. 4 Several tests are available to assess functional capacity, 5 but their use in children and adolescents can give different information than those obtained from adults due to differences in physiological and metabolic responses to stress. Concerning differences in cardiovascular responses, healthy children showed higher chronotropic and lower inotropic responses during maximal effort. 5 Furthermore, the information of the tests is not standardized in terms of values, which limits the comparison of different studies.
Functional capacity varies according to the type of CHD, surgical outcome, age and gender of the patient. Patients with incomplete repair of heart defects present significant reductions in peak work rate and age-adjusted maximum ventilation as compared with their pairs who undergone complete repair. 6 Most of the published studies have a small sample size and include children, adolescents and adults, with a large range of age of subjects. 7 Thus, the present study aimed to systematically review the literature to summarize the functional capacity of children and adolescents diagnosed with CHD, through a meta-analysis of observational studies.

Methods Eligibility criteria
This review included observational studies (cohort, cross-sectional or case-control studies), data from the first evaluation of randomized or non-randomized clinical trials or observational follow-up periods after clinical trials, in which the sample consisted of children and adolescents with CHD, aged between six and 18 years. Other conditions for inclusion of the studies were evaluation of functional capacity by cardiopulmonary exercise test, stress testing, six-minute walk test (6MWT) or step test.
Studies published in English were included. Only studies published after 1980 were considered, since methods for evaluation of functional capacity were not standardized before that period.

Strategy of search and selection of studies
The following electronic databases were searched in June 2015: MEDLINE (accessed through Pubmed), Cochrane Central Register of Controlled Trials (Cochrane CENTRAL) and EMBASE. In addition, references from published studies were also searched manually. Duplicate reports were deleted in the first step of selection of articles. The MeSH terms and entry terms used are presented in Box 1 (Supplementary File).
The titles and abstracts of all articles identified in the search strategy were assessed in duplicate by independent investigators (C.W.S. and A.C.). All abstracts that did not provide sufficient information regarding the inclusion and exclusion criteria were selected for full-text evaluation. In the second phase, the same reviewers independently evaluated these full-text articles and made their selection in accordance with the eligibility criteria. Any disagreements between reviewers were resolved through consensus and, in cases of persistent disagreement, a third reviewer (G.S.) assessed the publications.

Data extraction
Data were extracted independently by two reviewers (C.W.S and A.C.), using standardized forms comprising methodological characteristics, description of interventions, and outcomes; disagreements were resolved by consensus or by a third reviewer (G.S.).
In order to quantify possible differences on the functional capacity, the primary outcomes were the VO 2 max and the distance walked in the 6MWT. Additionally, maximum heart rate (HR) and other physiological variables taken from the cardiopulmonary exercise test (cardiovascular assessment and gas analyzes with direct measurement of oxygen consumption ), 6MWT and stress testing (cardiovascular assessment, in which symptoms were observed, the behavior of heart rate, blood pressure and electrocardiogram) were also entered into the analyses. Variables extracted from the cardiopulmonary exercise test were the first and second ventilatory thresholds, and from the exercise stress testing we extracted the maximum systolic blood pressure (SBP).

Assessment of risk of bias
The methodological quality of the studies was assessed by two researchers (C.W.S and A.C.), previously trained and qualified. The Newcastle-Ottawa Scale was used for case-control and cohort studies, whereas cross-sectional studies were evaluated with an adaptation of the same scale. The quality score of cohort studies and case-control studies was calculated by the assessment of three components: selection of the study groups (0-4 points), quality of adjustment for confounding (0-2 points) and evaluation of exposure or outcome of interest. The cohort studies evaluation was used for quasi-experimental studies. In the case of cross-sectional studies, the score was calculated in two components: selection of the study groups (0-3 points) and assessment of the outcome of interest (0-4 points). The maximum score could be 9 points for case-control and cohort studies and seven points for cross-sectional studies, representing a high methodological quality. 8 Disagreement between reviewers were resolved by consensus, and, in cases of persistent disagreement, the assessment was made by a third reviewer (G.S).

Data analysis
The quantitative assessment of the included studies was performed by meta-analysis, by comparing the VO 2 max in relation to body mass of children and adolescents with CHD and respective control groups without CHD. Combined estimates of effects were generated through the maximum values obtained in the studies reviewed, and are presented as weighted mean differences. Statistical heterogeneity among the results on functional capacity of the studies was assessed by the Cochran's Q test, with significance level of 0.1, and by the inconsistency I 2 test, in which values above 50% were considered as indicative of high heterogeneity. 9 The heterogeneity among the studies was explored using two strategies. Initially, each study was individually removed from the meta-analysis in order to verify any particular influence on the results. Second, the influence of age and maximum HR during exercise testing was evaluated by univariate meta-regression, and a threshold of p < 0.05 was used to indicate statistical significance.
The analyses were performed using Stata software version 11.0.

Results
Twenty-five of the 2.683 studies identified in the search met the criteria of eligibility and were included in the analysis. Figure 1 shows the flow chart of studies of this review. The age of the participants ranged from six to 18 years. Seventeen cross-sectional studies, three quasi-experimental studies and five cross-sectional studies with follow-up were included, with a total of 770 patients with CHD and 754 healthy controls.
The characteristics of the studies are presented in Table 1.
Most of the studies investigated children who underwent surgical correction for cyanotic CHD, such as tetralogy of Fallot (T4F), transposition of the great vessels (TGV) and univentricular hearts. Only one study evaluated children that were not submitted   13 used the cardiopulmonary exercise test, but the maximum HR of the participants was not reached, and the test was therefore considered as submaximal.
The methodological quality of the cohort studies ranged from two to seven points, with an average of 6.0 ± 1.8 points.
For cross-sectional studies, the score varied from three to seven points, with an average of 5.4 ± 1.0. The cohort study with lowest score (by Pfammater et al. 14 ) did not describe the origin of the cohort, the methods for assessing the outcome of interest, and how losses were controlled. Among cross-sectional studies, the publication by Page et al. 15 had only three points, since it did not present non-response rates and did not inform on the representativeness of the sample, origin of the control group and situation of this group (whether it was disease-free). Among quasi-experimental studies, two had four points and one received five points.
In the meta-analysis including 17 studies that conducted cardiopulmonary exercise tests with measurement of VO 2 max, it was 9.31 ml/kg/min lower in patients with CHD (95% CI, -12.48 to -6.13; I 2 , 94.3%, P for heterogeneity < 0.001), as compared with the control group. As shown in Figure 2, studies were stratified according to the type of ergometer used for the maximal test. Eleven studies used the cycloergometer. In these studies, the difference between VO 2 max in the CHD group and the control group was -9.71 ml/Kg/min (95% CI -14.06 to -5.36; I 2 = 94.2%, P for heterogeneity < 0.00001). Considering the six studies that used the treadmill, the difference between VO 2 max in the CHD group and the control group was -8.58 ml/Kg/min (95% CI -12.73 to -4.44; I 2 91.5%, P for heterogeneity < 0.00001).
The meta-analysis on the anaerobic threshold included six studies, showing that the CHD group presented an anaerobic threshold of -4.27 mL/kg/minute (95% CI, -10.84 to 2.31; I 2 , 97.6%, P for heterogeneity < 0.001) as compared with the control group.  Figure 3 shows the meta-analysis of the maximum HR reached during cardiopulmonary exercise test and stress testing, retrieved from 18 studies. The CHD group presented HR of -15.14 bpm (95% CI, -20.97 to -9.31; I 2 , 94.3%, P for heterogeneity < 0.001) as compared with the control group.
Considering the variable HR according to the type of test, 14 studies evaluated maximal HR though the exercise test. In these studies, the CHD group showed a difference of -17.70bpm (95% CI -24.37 to -11.03; I 2 , 94.4%, P for heterogeneity < 0.00001) in relation to the control group. In the four studies that used stress test for evaluation, all presented data as maximal HR. Meta-analysis of these studies showed that the CHD group had a lower HR when compared to the control group (difference -4.68bpm (95% CI -9.32 to -0.04; I 2 , 43.4%, P for heterogeneity = 0.15) ( Figure 3).
The meta-regression showed that the age (n = 16) was not associated with the heterogeneity observed in VO 2 max (R 2 = 18.43%. p = 0.09). Maximum HR (n = 13), however, had a significant influence on the heterogeneity observed in VO 2 max (R 2 = 69.20%. p = 0.005), as shown in Figure 4. An inverse relationship is por was between the chronotropic deficit and VO 2 max (β = -0.688; p = 0.005).
Since only one study evaluated functional capacity through the 6MWT, distance walked could not be analyzed.
None of the included studies used the step test for evaluation of functional capacity.

Discussion
This systematic review with meta-analysis of observational studies showed that children and adolescents with CHD present a decrease in functional capacity and in the anaerobic threshold during an exercise maximal test as compared with healthy individuals of the same age group, even when treated. In addition, children and adolescents with CHD have a chronotropic deficit that explained 69.20% of the VO 2 max variance observed among the 13 studies analyzed.
Maximal oxygen consumption (VO 2 max) has been widely used as gold standard for evaluation of functional capacity in healthy or ill individuals. There is a difference in cardiorespiratory responses between adults and children. 5 The anatomically smaller heart size in children results in lower venous return, and therefore lower cardiac output, which in turn results in lower VO 2 max when compared with adults. Therefore, the most important compensatory mechanism for children is through the increase in HR. 16 During exercise, the systolic volume increases around 20% in a normal heart, and the further increase in the cardiac output is due to an increase in HR. 17 Although expected, the information that children and adolescents with CHD in fact have lower functional capacity than their peers, even after corrective surgery, is first summarized in the present meta-analysis.
Individuals with CHD have insufficient chronotropic response, which leads to a decreased maximum HR, consequently reducing the VO 2 max in this population. 17 Fedriksen et al. 7 investigated children between eight and 17 years of age with several types of CHD and observed that those aged 10 to 13 years with obstruction of the left ventricular output presented oxygen consumption values above that of those with TGV or T4F. Children with T4F had a natural development of the capacity for physical exercise, which was however lower than that of healthy children; children with TGV showed a decline of VO 2 between the ages of 12 and 13 years, probably due to a reduction of right ventricular function. 7 In the present meta-analysis, maximum HR was diminished in 15.14 bpm in the CHD group as compared to the control group. This chronotropic incompetence implies an inability to increase the HR in response to metabolic demand. 18 The activity of the sympathetic and parasympathetic nervous system, which plays an important role in the modulation of HR during exercise, can be affected by ischemia and/or denervation resulting from surgical procedure or, in cases of cyanotic CHD, by chronic hypoxemia. 19 Ohuchi et al. 20 observed that both SBP at rest or during peak exercise and HR variability were lower in the group of children with univentricular hearts compared with healthy controls, 20 which supports this hypothesis, that the HR directly influences the VO 2 max.
The anaerobic threshold, defined as the maximum intensity of exercise performed by an individual using aerobic metabolism, is inversely related to age. 21 In a study with 17 children with complex CHD, evaluated by cardiopulmonary exercise testing, Ohuchi et al. 22 observed that the anaerobic threshold was lower in these children as compared with the control group. 22 In addition, Paridon et al. 23 also used cardiopulmonary exercise test to assess 411 children who undergone Fontan procedure showing normal maximal oxygen consumption in 28% of the sample. Maximal oxygen consumption (VO 2 max) within the normal range was observed in only 28% of the sample. However, the anaerobic threshold was in the normal predicted range in most individuals (63%), suggesting that this population with univentricular hearts could tolerate a high level of submaximal and non-maximal activity. 23 Most studies showed high methodological quality in the evaluation of both exposure and outcome variables. Cross -sectional studies described more detailed evaluations regarding these variables when compared to cohort studies.
The main study limitation derive from that most studies included patients with different types of heart disease, and used different types of evaluation protocols with heterogeneity of ergometers for functional capacity evaluation, even if these are standardized in the literature. Thus, studies showed important differences in relation to these methodological aspects, although all have fulfilled the inclusion criteria for this meta-analysis. High heterogeneity observed in the meta-analyses partially reflects such

Hovels 41
Ishi 29 Sarubbi 32 Overall (I-squared = 94.3%, p = 0.000) NOTE: Weights are from random effects analysis Maximal heart rate (HRmax) methodological aspects, and we therefore explored it by using meta-regression analyses for factors of interest. In addition, the heterogeneous nature of the congenital heart lesions may also limit wide exploration of studies in this field, since many lesions have different pathophysiological behaviors and a broad spectrum of severity. In this context, it is important to systematically review all the available information in order to establish more detailed and useful evidence for this specific group.

Conclusion
The presence of CHD in children and adolescents is associated with lower functional capacity than in healthy controls, measured by VO 2 max in cardiopulmonary exercise testing, being influenced by the impaired chronotropic response observed in this population, and not by age. In addition, a lower ventilatory threshold was observed in the same group, suggesting a lower ability to perform aerobic Maximal heart rate (% HRmax) difference between controls and patients VO 2 max WMD (ml/Kg/min)