Introduction

Adults with congenital heart disease (ACHD) are at a higher risk of developing coronary artery disease (CAD) due to coronary anomalies, endothelial dysfunction, and surgical methods of CHD repair.1 Aging of the ACHD population observed in the era of modern cardiac surgery results in a higher burden of traditional cardiovascular (CV) risk factors and acquired heart diseases, including CAD.2-4 Atherosclerosis of coronary arteries superimposed on ACHD may have detrimental health consequences.5,6 To date, several studies have assessed the epidemiology of CAD in ACHD, and its relationship with the lesion type and CV risk factors.7-12 According to previous research, certain specific congenital conditions associated with widespread vasculopathy (eg, bicuspid aortic valve and coarctation of the aorta) predispose to premature development of CAD.12 On the other hand, the protective effect of cyanosis was also investigated.13 However, recent analyses have shown that classic atherosclerotic CV risk factors, such as systemic hypertension or hyperlipidemia, but not the underlying congenital abnormalities, are strong predictors of CAD in the ACHD population.8,10 Moreover, according to Bauer et al,14 the guideline-directed medical therapy of acquired cardiovascular disease in the ACHD population is unsatisfactory. Data reporting the prevalence of CAD and its determinants exclusively in older individuals with ACHD are scarce, even though this issue is of importance for several reasons. First, older patients with CAD are more likely to present with atypical symptoms. Furthermore, limited exercise capacity often observed in patients with CHD may not lead to angina, and thus postpone the diagnostic workup. In a study by Tutarel et al,15 CAD turned out to be an independent predictor of all-cause mortality in ACHD patients aged 60 years or older. Therefore, the aim of the present study was to assess the prevalence of CAD, its risk factors, and the use of secondary prevention pharmacotherapy among older patients with ACHD.

Patients and methods

We performed a retrospective analysis of patients hospitalized in our department on an elective or urgent basis between July 2013 and March 2020. If a patient was admitted more than once during the given timeframe, data from the first admission following their 60th birthday were considered. The inclusion criteria were 1) age of at least 60 years at the time of hospitalization, 2) previously or newly diagnosed CHD, and 3) known coronary artery status. Congenital heart defects were classified as mild, moderate, or severe, according to the European Society of Cardiology guidelines.16 The prevalence of classic CAD risk factors (systemic hypertension, dyslipidemia, smoking, overweight / obesity, diabetes, and chronic kidney disease [CKD]) was estimated as described previously.4 The patients with a body mass index (BMI) between 25 and 29.9 kg/m2 were classified as overweight, and those with a BMI greater than or equal to 30 kg/m2 were considered obese.17 CKD was defined as an estimated glomerular filtration rate below 60 ml/min/1.73 m2. Smoking status was determined based on current or previous use of tobacco products, self-reported by the patients.

CAD was defined as a history of acute coronary syndrome (ACS; ST-segment elevation myocardial infarction, non–ST-segment elevation myocardial infarction, or unstable angina), a history of planned coronary revascularization (coronary artery bypass grafting [CABG] or percutaneous coronary intervention [PCI]), or more than 50% diameter stenosis in any vessel on invasive coronary angiography or computed tomography (CT) angiography. Data on guideline-recommended pharmacotherapy were also collected.18,19

Ethics

The study was performed in accordance with the principles of the Declaration of Helsinki, and approved by the institutional medical ethics committee (1890/20). The informed consent requirement was waived because of the retrospective nature of the study.

Statistical analysis

Unless stated otherwise, continuous variables were presented as mean (SD), and categorical variables were expressed as numbers and frequencies. The Shapiro–Wilk test was used to assess the normality of variable distribution. Differences between the groups were compared using the t test, and in the case of categorical variables, the Z test was used. A univariable logistic regression analysis was performed to determine the odds ratios (ORs) and 95% CIs for risk factors associated with CAD. A multivariable logistic regression model including demographic, anthropometric, and clinical parameters was constructed, with CAD as a dependent variable. A 2-sided P value below 0.05 was considered significant. All data were analyzed with R software package, version 4.3.1 (The R Foundation for Statistical Computing, Vienna, Austria).

Results

During the study period, we identified 322 patients with ACHD aged 60 years or older at any hospitalization in our department. Of those, 198 individuals with known coronary artery status were included in the analysis. The most common congenital cardiac abnormality in the study population was atrial septal defect, followed by bicuspid aortic valve lesions. CAD was found in 54 participants (27.3%); 8 patients had a history of ACS, 18 patients had previously undergone planned CABG/PCI procedures, and 31 individuals had documented significant stenosis on coronary angiography. More detailed information regarding baseline characteristics of the CAD and non-CAD patients is provided in Table 1.

Table 1. Patient clinical characteristics and prevalence of cardiovascular risk factors in the study population

Parameter

Total (n = 198)

CAD (n = 54)

Non-CAD (n = 144)

P value

Age, y, mean (SD)

66.2 (5.3)

67.2 (5.5)

65.8 (5.2)

0.1

Male sex

86 (43.4)

35 (64.8)

51 (35.4)

<⁠0.001

History of cardiac surgery / percutaneous intervention

106 (53.5)

25 (46.3)

81 (56.2)

0.21

CHD complexity

Mild

156 (78.8)

49 (90.7)

107 (74.3)

0.02a

Moderate

33 (16.7)

4 (7.4)

29 (20.1)

Severe

9 (4.5)

1 (1.9)

8 (5.6)

CHD type

ASD II

60 (30.3)

17 (31.5)

43 (29.9)

0.83

ASD sv

18 (9.1)

3 (5.6)

15 (10.4)

0.3

ASD I

9 (4.5)

0

9 (6.3)

0.06

VSD

8 (4)

4 (7.4)

4 (2.8)

0.14

BAV lesions

57 (28.8)

20 (37)

37 (25.7)

0.12

CoA

4 (2)

1 (1.8)

3 (2.1)

0.89

Ebstein anomaly

8 (4)

0

8 (5.5)

0.08

Tetralogy of Fallot

7 (3.5)

0

7 (4.8)

0.1

ccTGA

7 (3.5)

2 (3.7)

5 (3.5)

0.95

Other

20 (10.1)

7 (13)

13 (9)

0.4

CV risk factors

Systemic hypertension

126 (63.4)

43 (79.6)

83 (57.6)

0.004

Dyslipidemia

143 (72.2)

48 (88.9)

95 (66)

0.001

Overweight / obesity

136 (68.7)

34 (63)

102 (70.8)

0.29

Diabetes mellitus

47 (23.7)

15 (27.8)

32 (22.2)

0.41

Chronic kidney disease

39 (19.7)

12 (22.2)

27 (18.8)

0.73

Tobacco smoking

58 (29.3)

23 (42.6)

35 (24.3)

0.01

Sum of CV risk factors, median (IQR)

3 (2–4)

3 (2.25–4)

3 (2–3)

0.002

Data are presented as number (percentage) of patients unless indicated otherwise.

a Mild vs moderate / severe CHD

Abbreviations: ASD, atrial septal defect; BAV, bicuspid aortic valve; CAD, coronary artery disease; ccTGA, congenitally corrected transposition of the great arteries; CHD, congenital heart disease; CoA, coarctation of aorta; CV, cardiovascular; IQR, interquartile range; sv, sinus venosus; VSD, ventricular septal defect

The CAD patients were more often men, and they were more likely to be diagnosed with mild congenital heart defects (Table 1). Among the individuals with CAD, 89% presented dyslipidemia, 80% had systemic hypertension, and 43% were former or current smokers, as compared with 66%, 58%, and 24% of the non-CAD patients, respectively. Also, the total number of analyzed traditional CV risk factors was significantly higher in the CAD group. Factors associated with CAD in the univariable analysis are listed in Table 2.

Table 2. Univariable analysis of factors associated with coronary artery disease in patients with congenital heart disease aged 60 years or older

Variable

OR

95% CI

P value

Age (per year)

1.05

0.99–1.11

0.09

Male sex

3.36

1.76–6.56

<⁠0.001

Mild CHD (vs moderate / severe CHD)

3.39

1.36–10.31

0.02

History of cardiac surgery / percutaneous intervention

0.68

0.35–1.25

0.21

CV risk factors

Systemic hypertension

2.87

1.41–6.27

0.005

Dyslipidemia

4

1.71–11.01

0.003

Overweight / obesity

0.74

0.38–1.45

0.37

Diabetes mellitus

1.35

0.65–2.72

0.41

Chronic kidney disease

1.24

0.56–2.62

0.59

Tobacco smoking

2.31

1.19–4.48

0.01

Number of CV risk factors

1.58

1.2–2.1

0.001

Abbreviations: OR, odds ratio; others, see Table 1

In the multivariable analysis, male sex, systemic hypertension, and dyslipidemia remained significantly positively associated with CAD, whereas overweight / obesity was significantly negatively associated with the outcome (Table 3).

Table 3. Multivariable analysis of factors associated with coronary artery disease in patients with congenital heart disease aged 60 years or older

Variable

OR

95% CI

P value

Male sex

3.19

1.59–6.62

0.001

Dyslipidemia

4.26

1.71–12.4

0.003

Systemic hypertension

2.33

1.05–5.48

0.04

Overweight / obesity

0.45

0.2–0.97

0.04

Abbreviations: see Table 2

Additionally, we analyzed the use of guideline-recommended CAD pharmacotherapy in the study population. β-Blockers were prescribed to 87% of the patients, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers to 74%, and lipid-lowering agents to 96% of the participants (Figure 1A). The proportion of CAD patients on antiplatelet or anticoagulant drugs was 89% (39% of them were prescribed at least 1 antiplatelet drug, 30% were only on oral anticoagulants, and 20% received both types of treatment) (Figure 1B).

Figure 1. Guideline-recommended pharmacotherapy for coronary artery disease in patients with congenital heart disease aged 60 years or older; A – proportion of patients taking each class of drugs; B – distribution of antithrombotic drug prescription in the patients with coronary artery disease

Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker

Discussion

Our study showed a high prevalence of CAD among ACHD patients, as compared with previous research, in which the reported rates ranged from 1% among all adults with CHD12 to 22% among patients older than 35 years.10 This can be explained by more advanced age of our population. However, in a large Canadian analysis,3 CAD was diagnosed in only 10% of ACHD patients aged 65 years and older. The aforementioned study might have underestimated the prevalence of the disease, as it was based on diagnostic codes for comorbid conditions. Other research demonstrated that over one-third of ACHD patients with significant CAD confirmed on angiography were asymptomatic.12

The prevalence of CAD in our analysis was similar to the overall CAD prevalence among other patients in this age group. In a study on the predictors of adverse outcomes in healthy aging adults randomized from among community-dwelling individuals aged 60 to 74 years, CAD was diagnosed in 26.2% of the participants.20 This observation confirms the findings of a large study by Kuijpers et al,21 involving a general population of the Netherlands. The authors showed an increased risk for CAD within the ACHD subgroup, but the relative risk declined with age, remaining significant up to the age of approximately 65 years in women and 50 years in men. It may be speculated that as the patients age, the role of CHD-related predisposing factors (coronary anomalies, surgical interventions, and endothelial dysfunction) diminishes, and the role of conventional CV risk factors grows.

In our study, the incidence of CAD was higher in the older patients with mild CHD than in those with moderate / severe disease. Similarly, Giannakoulas et al7 reported lower prevalence of CAD in patients with intermediate / complex lesions. This may result from a high percentage of cyanotic patients included in their analysis, as some studies showed a decreased burden of atherosclerosis in patients with cyanotic CHD.13 In our study, the percentage of cyanotic CHD was very low; and CAD in patients with mild heart defects seems to be predominantly a lifestyle-related disease. In our previous work,4 we showed a higher burden of conventional CV risk factors among older individuals with mild CHD, as compared with those with more complex heart defects.

Similarly to the observation made in the general population,22 male sex turned out to be an independent predictor of CAD in older patients. This finding holds true also in younger ACHD individuals who underwent selective coronary angiography or CT angiography prior to CHD surgery or as part of a diagnostic hemodynamic assessment.7,10

Well-established modifiable risk factors for CAD were significantly associated with greater prevalence of the disease in our cohort. These results are consistent with other studies. A multicenter case-control analysis based on data from the CONCOR registry8 revealed that traditional atherosclerotic risk factors, that is, hypertension, hypercholesterolemia, and smoking, were independently associated with a CAD diagnosis in ACHD patients. It should be pointed out that the authors did not assess patients screened for CAD but individuals with a history of ACS or revascularization due to significant CAD.

No difference in BMI values was found between the CAD group and controls in the abovementioned analysis.8 Other studies did not address the issue of overweight and obesity as risk factors for CAD in the ACHD population. Our data showed that overweight / obesity was a significant negative independent predictor of CAD. We believe that our observation is not due to collinearity of the independent variables (the variance inflation factors for all independent variables were <⁠1.13). Instead, we attribute the result to the suppression effect: overweight / obesity was not a predictor of CAD in the univariable analysis, while adding this variable to the multivariable model including male sex, systemic hypertension, and dyslipidemia increased the model fit. Our results are consistent with the findings of some studies in the general population suggesting a phenomenon called the obesity paradox, that is, a lower risk of CAD in more obese patients.23,24 In older patients with CV diseases, an individual approach to weight management, considering comorbidities and functional status, is suggested.25 In our opinion, the same mode of treatment should be applied in older ACHD patients.

The use of guideline-recommended pharmacotherapy for CAD in our study seems to be satisfactory. Over 95% of the ACHD patients received a lipid-lowering agent for secondary prevention of CAD. Our data are similar to the results of studies assessing pharmacotherapy in the general population, in which the percentage of patients receiving lipid-lowering drugs after hospitalization for CAD ranged from 91% to 94%.26-28 In contrast, data from the German National Register of Congenital Heart Defects14 showed a very small percentage of ACHD patients with CAD receiving statins (18.8%). Almost 90% of the patients in our cohort were treated with antithrombotic drugs, and this percentage was also close to the rate observed in the population of CAD patients without congenital heart defects. The use of antithrombotic treatment in the German population of ACHD patients was lower, but exceeded 80%.14 Among the antithrombotic agents prescribed to our study patients, we observed a high percentage of anticoagulants, which might be associated with the presence of other indications for this mode of treatment.

In conclusion, the study demonstrated a high burden of CAD in older patients with ACHD. Modifiable classic CV risk factors turned out to be independent predictors of the disease. Our observation is of clinical importance for several reasons. It draws attention to the fact that aging patients with ACHD may suffer from acquired heart disease, and highlights the role of preventive actions in patients with congenital heart defects. Secondary preventive therapy should be part of an overall surveillance in older individuals with ACHD. On the other hand, we documented the presence of the obesity paradox in ACHD patients with CAD. Such an observation has not been reported yet.

Study limitations

Some limitations should be considered while interpreting our findings. First, the prevalence of CAD might be underestimated, as only the patients with suspected CAD, those with a history of ACS, and the individuals undergoing planned cardiac surgery were analyzed. Secondly, as the study was conducted in a tertiary care hospital, it might not well reflect the use of recommended pharmacologic therapy in all ACHD patients with CAD. Finally, the cross-sectional design of the study did not allow us to assess the impact of CAD on all-cause mortality among older ACHD patients.

Conclusions

CAD is common in patients with ACHD aged 60 years or older. Our results underline the importance of identification and treatment of modifiable cardiovascular risk factors in this population. The obesity paradox might also play a role, and requires further investigation. The prescription rate of guideline-recommended medications in older ACHD patients with CAD seems to be satisfactory.