Approximately eight in 1000 children are born with congenital heart disease, with approximately two-thirds requiring treatment. Owing to advances in treatment, ∼90% are now expected to survive into adulthood compared with 20% 50 years ago. 1 This has resulted in an estimated current population of 135,000 adult congenital heart patients in England. 2
Traditionally, the outcome in congenital heart disease has been measured in terms of mortality and functional status. There is an increasing recognition that reliance on clinical measures in determining outcomes is not sufficient as it fails to capture the patients’ perspective.Reference Skevington, Lotfy and O'Connell 3 Consequently, interest has turned to examining the quality of life of adult congenital heart disease patients. Although a universally accepted definition of quality of life does not exist, it is usefully conceptualised in three broad domains, namely, physical, psychosocial, and environmental.
Research suggests that adult patients with congenital heart disease, like other patients with chronic conditions, are faced with physical health issues and psychosocial challenges.Reference Kovacs, Sears and Saidi 4 A clear perspective of the evidence on adult congenital heart disease patients’ quality of life is lacking. In order to clarify evidence regarding the quality of life of adult congenital heart disease patients, a systematic review of the literature was conducted. The purpose of the systematic review was fourfold:
-
• Describe the quality of life of adult patients with congenital heart disease.
-
• Examine the evidence on the relationship between disease severity variables and quality of life.
-
• Identify the quality of the existing literature.
-
• Suggest areas for future investigations.
Methodology
In all, seven electronic databases – Science Direct, Pubmed, PsycInfo, Amed, Embase, CINAHL Plus, and Medline – were searched (see Supplementary Figure S1). Owing to the fact that the research domain is relatively new, a start date criterion was not applied in the search, and all relevant articles published until November, 2011 were reviewed. A language criterion was applied as the team could only review articles written in the English language. Initial search terms used were specific, for example “quality of life AND GUCH”, but as the search did not retrieve many relevant articles it was broadened by using various word combinations, that is, (“quality of life” OR “life satisfaction” OR “well being”) AND (GUCH OR ACHD OR congenital heart disease OR congenital cardiac disease OR congenital heart defect OR congenital cardiac defect). Where the search yielded too many results, additional filter terms were used, for example AND adult. In all, 31 articles met the following inclusion/exclusion criteria, including three identified from the reference lists.
Exclusion criteria
-
• Qualitative studies.
-
• Sample solely consisting of Marfan syndrome patients.
-
• Abstract quality of life measures, for example non-validated single-item measures or question sets.
-
• Study of the effects of individual psychosocial variables, for example anxiety, depression, coping, on quality of life.
-
• Reviews and meta-analyses.
-
• Opinion articles, commentaries, and reports.
Inclusion criteria
-
• Quantitative studies.
-
• Adult congenital heart disease patients.
-
• Intention to measure quality of life concept.
-
• Quality of life description and/or how disease severity variables influence quality of life.
-
• Validated measures of quality of life.
-
• English language.
These studies were reviewed by means of data extraction forms (see Supplementary Table S1).
Results
A total of 31 studies were reviewed; 26 examined the patients’ quality of life in comparison with the general population and 22 examined the relationship of disease variables with quality of life.
Demographics and methodology
The number of participants ranged from 22 to 912, age ranged from 14 to 85 years, and gender was roughly balanced (see Table 1). The majority of the studies (23/31) included patients with a range of diagnoses, seven included patients with a single diagnosis, and one categorised patients in terms of the treatment they received. Most studies had a cross-sectional design, with two out of 31 studies using a longitudinal design. All but four studies included a comparison group of either normative data or matched controls.
AA = aortic atresia; AIn = aortic insufficiency; AO = arch obstruction; APVD = abnormal pulmonary venous drainage; AS = aortic stenosis; ASD = atrial septal defect; ASDII = secundum ASD; AVD = atrial valve diseases; AVP = aortic valve pathology; AVSD = atrioventricular septal defect; BAV = bifoliate aortic valve; BH = biventricular heart; CAT = common arterial trunk; CAVCD = complete atrioventricular canal defect; CCTGA = congenitally corrected TGA; CHD-TAAQOL = congenital heart disease-TAAQOL; COA = coarctation of the aorta; DILV = double-inlet left ventricle; DIV = double-inlet ventricle; DORV = double-outlet right ventricle; DVAC = discordant ventriculoarterial connection; ES = Eisenmenger syndrome; GBB = Giessener Complaints Questionnaire; HLHS = hypoplastic left heart syndrome; IO = inoperable; IS = intact septum; LAS = Linear Analogue Scale; MA = mitral atresia; ME = medical; Min = mitral insufficiency; MP = mechanical prosthesis; MVP = mitral valve prolapse; PA = pulmonary atresia; PDA = patent ductus arteriosus; POS = palliative operation shunt; PR = pulmonary regurgitation; PS = pulmonary stenosis; PVD = pulmonary valve diseases; PVS = pulmonary valve stenosis; SC = surgically cured; SCor = surgically corrected; SEIQoL-DW = schedule for the evaluation of individual QoL-direct weighting; SF-36 = Short Form-36; SIP = sickness impact profile; SP = surgically palliated; SQoL = subjective quality of life; SRV = systemic right ventricle; SV = single ventricle; SWLS = Satisfaction With Life Scale; TA = tricuspid atresia; TAAQOL = TNO-AZL Questionnaire for adult's health-related quality of life; TGA = transposition of the great arteries; TOF = tetralogy of Fallot; VSD = ventricular septal defect; WHOQOL-Bref = World Health Organization Quality of Life-Bref.
*Only conditions diagnosed in more than 10% of the sample were listed and compared
**Studies with common sample
***Diagnostic groups age- and gender-matched at operation and follow-up
Quality of life assessment
A variety of quality of life instruments were used (Table 1). Generic instruments included the Short Form-36 Health Survey (SF-36, 17 studies), TNO-AZL (TAAQOL, four studies), Sickness Impact Profile (SIP, two studies), Brief World Health Organisation Quality of Life assessment (WHOQOL-Bref, two studies), Duke Health Profile (one study), and Subjective Quality of Life (SQoL, one study). Single-item generic quality of life was assessed by the Linear Analogue Scale (LAS, one study) and Satisfaction with Life Scale (SWLS, one study). Disease-specific and individual measures included the TNO-AZL Congenital Heart Disease version (CHD-TAAQOL, two studies) and the Schedule for the Evaluation of Individual Quality of Life-Direct Weighting (SEIQoL-DW, one study), respectively.
The heterogeneous nature of the measures hinders the combination of findings. For the purpose of describing the findings, the scales or sub-scales of each instrument were categorised into three broad domains, that is, physical, psychosocial, and environmental/occupational quality of life (see Supplementary Table S2). These domains included all relevant scales from the instruments that were measured in two or more studies.
Quality of life in adult congenital heart disease patients
Table 2 presents the findings of 26 studies that compared the quality of life of adult congenital heart disease patients with the general healthy population.
↓ = poorer QoL; ↑ = better QoL; = = similar QoL (in relation to controls); blank = not measured; QoL = quality of life
Physical domain
Of the 22 studies, 17 reported poorer physical functioning, whereas five studies reported no differences between the patients and the general population. Risk factors for poorer functioning included younger age,Reference Bruto, Harrison and Fedak 7 , Reference Mokhles, van de Woestijne and de Jong 22 female gender,Reference Chen, Liao and Wang 8 disease severity,Reference Gratz, Hess and Hager 12 and the presence of arrhythmias.Reference Irtel, Vetter and Stuber 15 In all, eight out of 18 studies found more role limitations due to physical problems, whereas 10 out of 18 studies found no differences between patients and the general population. Risk factors for greater limitations were specific treatment groups, that is, cured, corrected, and inoperable – versus palliated and medicalReference Lane, Lip and Millane 19 – and arrhythmias.Reference Irtel, Vetter and Stuber 15 In all, four out of 21 studies reported more pain in patients, 14 out of 21 reported no differences, and three out of 21 reported less pain than the general population. Risk factors for more pain were being inoperableReference Lane, Lip and Millane 19 and arrhythmias.Reference Irtel, Vetter and Stuber 15 In all, 15 out of 21 studies showed poorer general health perception in patients, five found no difference, and one found better general health perception in patients than in the general population. Risk factors were disease severity – cyanosis, Fontan physiologyReference Gratz, Hess and Hager 12 – and arrhythmias.Reference Irtel, Vetter and Stuber 15 All (2/2) studies reported more symptoms in patients than in controls. There was one (1/4) study that indicated poorer gross motor functioning, whereas three out of four found no difference. All (4/4) studies reported no differences between patients and controls in fine motor functioning. There was one (1/4) study that reported poorer cognitive functioning, whereas three out of four reported no differences between patients and controls. With regard to sleep, two out of five studies reported poorer results for patients, whereas three out of five found no difference.
Psychosocial domain
With regard to this domain, seven out of 23 studies indicated compromised psychological/mental functioning in patients, whereas 16 showed no difference. Risk factors for poorer functioning were female gender,Reference Daliento, Mapelli and Russo 10 arrhythmias,Reference Irtel, Vetter and Stuber 15 and specific treatment groups, that is, being inoperable and cured.Reference Lane, Lip and Millane 19 More role limitations due to emotional problems were reported in two out of 17 studies, less limitations in one out of 17, and no differences in 14 out of 17. Risk factors for more limitations were arrhythmiasReference Irtel, Vetter and Stuber 15 and being surgically cured.Reference Lane, Lip and Millane 19 Out of 20 studies, seven indicated lower vitality in patients, whereas 13 showed no difference. Risk factors included arrhythmiasReference Irtel, Vetter and Stuber 15 and being inoperable and cured.Reference Lane, Lip and Millane 19 All (4/4) studies indicated no differences in happiness and sexual functioning between patients and controls. In all, two out of 23 studies showed reduced social functioning in patients, whereas 21 out of 23 showed no differences. Risk factors included arrhythmiasReference Irtel, Vetter and Stuber 15 and being inoperable, cured, or corrected.Reference Lane, Lip and Millane 19
Environmental/occupational domain
Out of six studies, one reported more disability in environment and daily activities in patients, four showed no difference, and one indicated better environmental circumstances compared with the controls.
Disease severity and quality of life
In all, 22 studies examined the association between disease severity variables and quality of life (see Table 3).
Y = association with QoL; – = no association with QoL; blank = not measured; QoL = quality of life
Cyanosis
In all, six out of nine studies reported poorer quality of life in cyanotic patients, mainly in physical quality of life, that is, physical functioning and general health, whereas three showed no difference.
Diagnosis
Of the seven studies assessing quality of life by diagnosis, only one studyReference Winter, Bouma and van Dijk 34 reported better quality of life in the severe group, for example tetralogy of Fallot, than the structurally simpler group, for example atrial septal defect, whereas the majority (6/7) reported diminished quality of life in the more complex diagnoses, for example single ventricle, Fontan, Marfan syndrome, mainly in relation to physical and environmental quality of life.
Exercise capacity
Of the six studies, five reported an association between exercise capacity as measured by peak oxygen uptake (VO2) and physical quality of life that is, physical functioning, general health, whereas one found no association.
Functional status
In all, four out of seven studies found an association between functional status – New York Heart Association, Ability, Warnes & Somerville Indexes – and quality of life, that is, physical functioning, environmental, whereas three out of seven found no association.
Ventricular dysfunction
Ventricular dysfunction was not found to be associated with quality of life in four out of five studies, whereas one study reported an association with physical quality of life.
Arrhythmias
Out of four studies, two reported an association between arrhythmias and quality of life – general health, role limitations due to emotional problems – whereas two studies found no association.
Treatment type
Of the two studies assessing quality of life by treatment type, one reported better quality of life in patients who were treated with recent surgical techniques, that is, recent arterial switch versus Mustard & Senning. Contrary to expectations, the second study reported that “cured” (curative surgery) patients have poorer quality of life, that is, mental health, social functioning, than patients with corrections, that is, further operation possibility, and medical patients, that is, no surgery. Palliated – further operation possibility – and medical patients were similar. Inoperable patients had poorer quality of life than all groups except for similar social functioning and vitality to “cured” patients.
Severity
Few of the above and other studies categorised disease severity in different ways (see Table 4). In all, four out of eight studies reported significant associations between severity – assessed by diagnosis, objective physical index, cyanosis – and mainly physical quality of life and four reported no association, the latter measured severity by diagnosis.
Discussion
The first objective of this review was to investigate the quality of life of adult congenital heart disease patients. The pattern of results indicated that there is strong evidence that adult patients with congenital heart disease experience reduced quality of life in the physical domains, that is, reduced physical functioning, poorer general health, and more symptoms than the general population. These conclusions are similar to a previous literature review.Reference van Rijen, Utens and Roos-Hesselink 36 Although these findings are likely to reflect the position of adult congenital heart disease patients, it raises an important issue regarding expectations. Given that quality of life is patient-reported, it is possible that the findings may reflect a perceived lack of understanding, leading patients to develop unrealistic expectations and misconceptions about their physical functioning. An unclear disease course may leave patients feeling uncertain and insecure about their condition and prognosis.Reference Claessens, Moons, de Casterlé, Cannaerts, Budts and Gewillig 37 Although predicting a clear prognosis is difficult, it may be important to review the information that is provided by healthcare professionals to patients.
It is also important that in some domains of physical quality of life, for example, pain, sleep, gross/fine motor functioning, as well as psychosocial and environmental/occupational quality of life, results indicated that patients are not different from the healthy population. The finding of no impact on psychosocial functioning in the adult patients with congenital heart disease has a number of potential explanations. One is patients having a strong sense of coherence about their condition because growing up with congenital heart disease made them understand, manage, and find meaning in their experiences.Reference Moons and Norekvai 38 Another possible explanation is change in internal standards, values, and priorities,Reference Schwartz and Sprangers 39 that is, response shift. Patients may develop different internal health values from healthy individuals.Reference Moons, van Deyk, Marquet, De Bleser, De Geest and Budts 40 In one of the reviewed studies,Reference Bruto, Harrison and Fedak 7 younger patients reported poorer quality of life compared with their peers, whereas older patients reported normal quality of life. This may signify redefinition of standards, values, and priorities and modification of quality of life expectations with increasing age. The lack of differences in environmental/occupational quality of life may signify that patients have no problems in work, access to healthcare, or daily activities.
A few studies grouped patients based on various severity factors. The findings indicate no differences in quality of life between the severity groups that used composite severity ratings, suggesting that categorising patients in this way may reflect clinical issues but may mask significant effects of individual severity indicators. For instance, when cyanosis was combined with other factors into a severity classification, it did not associate with quality of life, but when considered independently it was found to associate with most quality of life domains.Reference Cohen, Daniela, Yalonetsky, Gagin and Lorber 9 Considering each variable individually, the severity indicators that were associated with quality of life were diagnosis, cyanosis, exercise capacity, arrhythmias, and functional status. It appears more valuable to assess the independent impact of specific variables and their relation to quality of life, rather than constructing a broader-based categorisation such as severity.
There are numerous methodological issues that are apparent in the research reviewed. One difficulty is that by definition adult congenital heart disease patients as a group will be heterogeneous. This is exacerbated in some studies by the failure to distinguish between different diagnostic categories. Whereas some studies focused on single diagnoses, others included different diagnoses. Bruto et alReference Bruto, Harrison and Fedak 7 argued that it is over-simplistic to categorise patients on the basis of initial diagnosis and native anatomy, and that emphasis should be given to disease course and patient characteristics. In general, the heterogeneity makes for large variability and difficulty in comparing studies. Mixed and unexpected findings can also be attributed to methodological weaknesses such as small sample sizes, cross-sectional designs, and different quality of life measures and diagnosis categorisations. Convenience sampling may have increased type II error and selection biases.
The absence of a quality of life definition in the reviewed studies is important because previous studies report that patients distinguish between quality of life – predominantly affected by psychological functioning – and perceived health status – affected by physical functioning.Reference Smith, Avis and Assmann 41 Evidence suggests that quality of life should also be distinguished from health-related quality of life, a term to be used when illness is relevant, but this is not always applied in the literature.Reference Pike, Evangelista, Doering, Koniak-Griffin, Lewis and Child 42 A clear definition of quality of life is needed to guide the selection of the most appropriate measures. Moons et alReference Moons, van Deyk, Budts and De Geest 43 provide a review of methodological problems around quality of life measurement in adult congenital heart disease.
To address these methodological limitations, future studies should include larger patient groups, thus allowing generalisability of the findings. Longitudinal studies will allow investigation of quality of life changes at different time-points and detection of associations between quality of life and its determinants, as well as potential age differences. Sound quality of life conceptualisation and measurement is also suggested. Finally, more research needs to be conducted towards a consensus regarding disease severity classification.
Limitations
The studies’ heterogeneity challenged the synthesis and comparison of findings. Application of stricter inclusion criteria, however, would have subsequently limited the number of studies, thus compromising the generalisability of findings. Other psychosocial factors that may influence quality of life, for example coping, anxiety, depression, have not been reviewed, as the plethora of evidence suggests the potential for an independent review in this area.
Conclusion
The quality of life of adult patients with congenital heart disease appears to be compromised in physical domains and there are indicative disease severity variables that may be included in a standardised classification system of disease severity. Future research needs to address important methodological limitations of previous studies. Knowledge about the patients’ quality of life and the influencing factors can potentially lead to the development of targeted interventions for the improvement of quality of life in the adult population of congenital heart disease patients.
Acknowledgements
This work was supported by the UCLH/UCL National Institute for Health Research, Comprehensive Biomedical Research Centre.