Distribution, recognition and clinical implications of syndromic conditions in adults with congenital heart disease

Abstract

Objective

Syndromic conditions are frequently associated with congenital heart disease (CHD). Their distribution, recognition and prognostic impact in adults with CHD (ACHD) remain poorly characterised due to limited assessment beyond childhood.

Methods

We retrospectively analysed 9826 patients with ACHD from a tertiary ACHD centre. Syndromic conditions were identified from clinical records and categorised as chromosomal, single-gene or other syndromes. Clinical characteristics, comorbidities and CHD complexity were compared between syndromic and non-syndromic patients. Multivariable Cox regression was used to evaluate associations with all-cause mortality.

Results

Of 9826 patients, 1069 (10.9%) had a syndromic diagnosis. Marfan syndrome (4.1%) and Down syndrome (3.6%) were the most common, followed by Noonan syndrome (0.7%) and 22q11.2 deletion syndrome (0.7%). Syndromic patients were younger (median 34 vs 39 years), had higher rates of pulmonary arterial hypertension (15% vs 5.9%) and sleep apnoea (7.4% vs 2.8%), but fewer traditional cardiovascular risk factors. Down syndrome clustered with atrioventricular septal defect and 22q11.2 deletion with conotruncal defects. In Tetralogy of Fallot, 4.3% had 22q11.2 deletion, mostly diagnosed in childhood; among those without prior diagnosis, most were neither clinically suspected nor tested in adulthood, suggesting under-recognition compared with paediatric cohorts. A syndromic diagnosis was independently associated with higher all-cause mortality (HR 2.06, 95% CI 1.76 to 2.42, p<0.001).

Conclusions

We present herewith our institutional ACHD data showing that approximately 1 in 10 patients had a syndromic condition, which was clinically relevant in many respects, cardiovascular or other, including overall mortality. We submit that improved recognition and expanded access to genetic screening may provide additional information relevant to risk stratification and long-term outcome in this growing patient population.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Syndromic conditions are common in paediatric congenital heart disease, but their recognition and documentation in adulthood are variable, and long-term implications in adults with congenital heart disease (ACHD) are not well characterised.

WHAT THIS STUDY ADDS

• In a contemporary cohort of 9826 patients with ACHD, 10.9% had a syndromic diagnosis. Several conditions, particularly 22q11.2 deletion in Tetralogy of Fallot, were frequently unrecognised in adulthood. Syndromic diagnoses were associated with differences in long-term outcomes, including all-cause mortality.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• These findings support more systematic consideration of syndromic conditions in ACHD practice, including genetic evaluation and discussions of recurrence risk and reproductive planning within routine follow-up.

Introduction

Congenital heart disease (CHD) is the most common birth defect, affecting approximately 1% of newborns.¹ CHD encompasses a broad spectrum of anatomical defects, ranging from simple lesions requiring minimal follow-up to complex malformations necessitating staged surgical interventions and lifelong surveillance.2,5 Some patients with CHD have also genetic syndromes such as Down syndrome or 22q11.2 deletion, where the cardiac anomaly is part of a broader multisystem disorder.^{6 7} In these individuals, the presence of extracardiac manifestations ranging from immunodeficiencies, developmental delay to endocrine and renal anomalies poses additional challenges in long-term care.⁸ Such extracardiac manifestations are clinically relevant and have been associated with increased healthcare needs and more complex long-term management.⁹

While syndromic conditions are frequently recognised and documented during paediatric care,10,12 their identification and documentation in adulthood remain inconsistent and potentially incomplete. Many patients with adults with CHD (ACHD) transition into adult care without a comprehensive genetic evaluation or with limited historical data regarding syndromic diagnoses.^{13 14} Variability in diagnostic evaluation during transition may result in incomplete documentation of syndromic features in adulthood, potentially affecting clinical assessment and long-term follow-up. Furthermore, awareness of syndromic diagnoses is critical for informing genetic counselling, including assessment of recurrence risk, reproductive planning and maternal-fetal risk assessment during pregnancy.⁶

This study aims to assess the current state of syndromic condition recognition in patients with ACHD from a clinical perspective. By comparing the clinical identification of syndromic features in adulthood against expectations based on paediatric patterns and known associations, we aim to highlight potential gaps in diagnostic recognition, clarify the extent of potential underdiagnosis and identify areas for improvement in multidisciplinary care and genetic evaluation practices for adults living with CHD.

Methods

Study population and data collection

We conducted a retrospective analysis of ACHD followed at our tertiary ACHD centre from January 2000 to February 2025. Diagnostic classifications were based on the underlying cardiac anatomy and assigned according to standard morphological criteria. Patients were grouped by congenital diagnosis and further categorised into anatomical complexity tiers: simple, moderate and complex, following the Bethesda classification.¹⁵ Syndromic conditions such as Marfan, Loeys-Dietz and Ehlers-Danlos syndromes, which are not typically classified as CHD, were grouped under ‘Other’, unless accompanied by a specific CHD diagnosis. Patients or members of the public were not involved in the design, conduct, reporting or dissemination plans of this research.

Classification of genetic abnormalities

Genetic abnormalities were determined based on the presence of explicitly stated diagnoses in clinical documentation, including clinical letters, reports and discharge summaries. Each patient was assigned to one of four categories: normal (no diagnosed genetic syndrome), single gene disorder (pathogenic variants in a single gene, eg, FBN1, PTPN11), chromosomal abnormality (chromosomal deletions, duplications or aneuploidy confirmed by cytogenetic or molecular testing, eg, 22q11.2 deletion, Down syndrome) and other (syndromic conditions not falling into the above categories, eg, Scimitar syndrome). In a subgroup analysis of patients with Tetralogy of Fallot (TOF), we further classified individuals into four diagnostic subgroups: (1) diagnosed in childhood, (2) not clinically suspected, (3) clinically not excluded but not tested and (4) clinically not excluded and tested in adulthood.

Cross-classification of genetic abnormalities by cardiac diagnosis

To explore genotype–phenotype relationships, we calculated the percentage of patients with selected genetic abnormalities within each cardiac diagnosis group. Heatmaps were generated to visualise the distribution of genetic abnormalities across diagnostic categories. To accommodate wide variation in prevalence, a pseudo-logarithmic transformation was applied to the colour scale.

Statistical analysis

Descriptive statistics were used to calculate the frequency and proportion of each genetic abnormality in the overall cohort. Continuous variables are reported as mean±SD or as median with IQR, while categorical variables are presented as counts and percentages. The primary analyses, which were prespecified, focused on describing the distribution and clinical characteristics of syndromic conditions within the overall ACHD cohort. The association between syndromic conditions and all-cause mortality was evaluated as a secondary analysis using multivariable Cox proportional hazards regression. Models were adjusted for basic demographic factors, congenital complexity, and New York Heart Association (NYHA) functional class. The proportional hazards assumption was assessed using Schoenfeld residuals. Hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated. Missing data were handled using available-case analysis; no imputation was performed. Subgroup analyses in TOF and diagnostic pathway assessments were exploratory. All statistical analyses were performed using R software V.4.5.2 (R Foundation for Statistical Computing, Vienna, Austria). A p value of <0.05 was considered statistically significant.

Results

Distribution and classification of syndromic conditions

A total of 9826 patients with ACHD were included in the analysis. Among them, 1069 patients (10.9%) had a documented syndromic diagnosis based on clinical notes or reference letters, while the remaining 8757 patients (89.1%) were classified as non-syndromic (table 1). Genetic abnormalities were classified into three major categories: single-gene disorders (555 patients, 51.9%), chromosomal abnormalities (492 patients, 46.0%) and other syndromes (22 patients, 2.1%) (figure 1). These classifications were based on established molecular or cytogenetic diagnoses, when available.

Table 1. Distribution of genetic abnormalities in syndromic adult congenital heart disease patients.

Syndromic conditions	n	Percentage	Classification
Normal	8756	89.12	Normal
Marfan syndrome	401	4.08	Single gene disorder
Down syndrome	349	3.57	Chromosomal abnormality
Noonan syndrome	73	0.74	Single gene disorder
22q11.2 deletion	71	0.69	Chromosomal abnormality
Ehlers-Danlos syndrome	43	0.44	Single gene disorder
Williams syndrome	39	0.40	Chromosomal abnormality
Turner syndrome	30	0.31	Chromosomal abnormality
Scimitar syndrome	22	0.22	Other
Loeys-Dietz syndrome	10	0.11	Single gene disorder
Kartagener syndrome	9	0.09	Single gene disorder
Charge syndrome	7	0.07	Single gene disorder
Neurofibromatosis type 1	3	0.03	Single gene disorder
Others1	8	0.08	Single gene disorder
Others2	5	0.05	Chromosomal abnormality

Percentages represent the proportion within each column category. ‘Normal’ refers to patients without a diagnosed genetic syndrome. ‘Syndromic’ refers to those with a confirmed genetic syndrome. ‘Others¹’ (single gene disorders) include rare conditions (each n=1): Robinow syndrome, Cornelia de Lange syndrome, Mowat-Wilson syndrome, Frank-ter Haar syndrome, Sotos syndrome, SMAD6 variant, Klippel-Feil syndrome and lymphoedema-distichiasis syndrome. ‘Others²’ (chromosomal abnormalities) include rare conditions (each n=1): trisomy 10, trisomy 14, trisomy 18, 8p23.1 microdeletion and 17q12 deletion.

Figure 1. Distribution of syndromic subtypes among patients with genetic abnormalities. The pie chart illustrates the proportion of syndromic subtypes in the cohort. Percentages indicate the proportion of each category relative to all syndromic patients.

The most frequently identified syndromic condition in our centre was Marfan syndrome, observed in 401 patients (4.1%), followed by Down syndrome in 350 (3.6%), Noonan syndrome in 73 (0.7%) and 22q11.2 deletion in 68 patients (0.7%) (table 1). Other less common conditions included Ehlers-Danlos syndrome (0.4%), Williams syndrome (0.4%) and Turner syndrome (0.3%). A total of 25 distinct syndromic conditions were recognised and classified into chromosomal abnormalities, single-gene disorders and other rare syndromes based on presumed underlying aetiology.

Patients in the syndromic group were slightly younger than those in the non-syndromic group (median age: 34 vs 39 years) (table 2). The sex distribution was balanced across groups, with approximately equal percentages of male and female patients. Simple lesions were more prevalent among non-syndromic patients (51%) than among syndromic patients (19%). Functional status was generally preserved in both groups, with the majority of patients classified as NYHA class I. However, NYHA class III/IV was more common in the syndromic group (8.0%) than in the remainder (5.3%). Syndromic patients also had a lower reported prevalence of several acquired cardiovascular risk factors, including hypertension (9.4% vs 16%), hypercholesterolaemia (5.8% vs 12%) and diabetes mellitus (3.0% vs 4.6%). In contrast, sleep apnoea was more frequently observed in syndromic patients (7.4% vs 2.8%). Notably, pulmonary arterial hypertension was observed in 15% of syndromic patients.

Table 2. Baseline characteristics of patients with adult congenital heart disease with and without genetic syndromes.

Characteristics	Overall N=9826*	Normal N=8757*	Syndromic N=1069*	P value†
Age (years)	39 (±16)	39 (±16)	34 (±14)	<0.001
Sex				0.54
Female	4885 (50)	4363 (50)	522 (49)
Male	4941 (50)	4394 (50)	554 (51)
Complexity				<0.001
Simple	4653 (47)	4448 (51)	205 (19)
Moderate	2752 (28)	2511 (29)	240 (22)
Complex	1591 (16)	1423 (16)	168 (16)
Other	830 (8.4)	374 (4.3)	455 (43)
BMI	25.9 (±5.7)	26.0 (±5.6)	25.5 (±6.0)	0.011
NYHA				0.001
I	7074 (76)	6320 (76)	754 (75)
II	1762 (19)	1587 (19)	175 (17)
III/IV	522 (5.6)	441 (5.3)	81 (7.9)
Systemic hypertension	1460 (15)	1360 (16)	100 (9.4)	<0.001
Hypercholesterolaemia	1073 (11)	1011 (12)	62 (24)	0.001
Diabetes mellitus	436 (4.4)	404 (4.6)	32 (3.2)	0.015
Sleep apnoea	327 (3.3)	248 (2.8)	79 (7.5)	<0.001
Coronary artery stenosis	343 (3.5)	323 (3.7)	20 (1.9)	0.002
Pulmonary arterial hypertension	611 (6.9)	463 (5.9)	148 (15)	<0.001

Values are presented as mean (±SD) or number (%). ‘Normal’ refers to patients without a diagnosed genetic syndrome. ‘Syndromic patients’ refers to those with a confirmed genetic syndrome. Percentages represent the proportion within each column category.

^*Mean (±SD); n (%).

^†Wilcoxon rank-sum test; Pearson’s χ2 test.

BMI, body mass index; NYHA, New York Heart Association.

Distribution of syndromic conditions by cardiac diagnosis

Among our ACHD cohort, the prevalence of syndromic conditions varied substantially across diagnostic subgroups. As shown in the heatmap (figure 2), Down syndrome was predominantly observed in patients with atrioventricular septal defect. In contrast, 22q11.2 deletion syndrome was most frequently associated with conotruncal defects, particularly TOF and truncus arteriosus. Other syndromes, such as trisomy 18, 8p23.1 microdeletion and a Smad6 variant, were observed in fewer patients and exhibited a more heterogeneous distribution.

Figure 2. Distribution of genetic abnormalities across congenital heart disease diagnostic groups. The heatmap shows the percentage of patients with each genetic abnormality within each congenital heart disease diagnostic group. Darker shades of red indicate a higher proportion of patients in that group with the specified genetic abnormality. Percentages are calculated as the number of patients with the genetic abnormality divided by the total number of patients in each diagnostic group. ASD, atrial septal defect; AVSD, atrioventricular septal defect; CCTGA, congenitally corrected transposition of the great arteries; PDA; VSD; ventricular septal defect.

In a focused analysis of 1024 patients with TOF, 44 (4.3%) had a diagnosis of 22q11.2 deletion syndrome, including 34 diagnosed in childhood and 10 identified after referral to ACHD care (figure 3a). Of the 990 patients without a prior diagnosis at adult entry, 771 (77.9%) were not clinically suspected, whereas 219 (22.1%) were considered not clinically excluded. Among these, 63 (28.8%) underwent genetic testing, of whom 10 (15.9%) were positive and 53 (84.1%) were negative. The remaining 156 patients (71.2%) did not undergo testing. Similar findings were observed among the right aortic arch subgroup. Among 47 patients with TOF with a right-sided aortic arch, 8 (17.0%) had 22q11.2 deletion (figure 3b). Demographic comparison showed that patients diagnosed in childhood were the youngest (27.9±9.0 years), followed by those not tested (32.9±11.6 years), those tested in adulthood (33.3±8.7 years) and those not clinically suspected (38.7±15.0 years) (figure 4a). Regarding sex distribution, females accounted for 55.9% of the childhood-diagnosed group, 44.1% of the not clinically suspected group, 51.6% of the not-tested group and 60.3% of the tested group. Female predominance was observed in the tested group, whereas males predominated among those not clinically suspected (figure 4b).

Figure 3. Diagnostic status of 22q11.2 deletion in adults with Tetralogy of Fallot. (a) Among all adults with TOF, the panel shows the proportions diagnosed in childhood, those not clinically suspected at adult entry and those in whom the condition could not be excluded and who were subsequently tested. (b) The same diagnostic categories are shown for the subgroup with a right-sided aortic arch, highlighting the higher diagnostic yield in this anatomical subgroup. ACHD, adults with congenital heart disease; TOF, Tetralogy of Fallot.

Figure 4. Age and sex distribution by diagnostic group in patients with Tetralogy of Fallot. (a) Mean age is compared across diagnostic subgroups, illustrating younger age among patients diagnosed in childhood and older age among those not clinically suspected. (b) Proportion of female patients across the same diagnostic subgroups, showing sex-related differences in recognition and testing patterns.

Prognostic impact of syndromic conditions

In a secondary analysis, we examined the relationship between syndromic conditions and all-cause mortality using available follow-up data. During the follow-up period, 199 deaths out of 1.026 syndromic patients were recorded. After adjustment for age, sex, NYHA class and CHD complexity, the presence of a syndromic diagnosis remained independently associated with increased mortality risk compared with the non-syndromic subgroup (HR 2.06, 95% CI 1.76 to 2.42, p<0.001) (figure 5). This finding underscores the clinical relevance of syndromic conditions beyond anatomical diagnosis alone.

Figure 5. Multivariable Cox proportional hazards analysis of syndromic condition for all-cause mortality. The forest plot shows adjusted HRs with 95% CIs for age, sex, congenital heart disease complexity, New York Heart Association class and the presence of a syndromic condition. NYHA, New York Heart Association.

Discussion

We provide a comprehensive, contemporary description of syndromic conditions in a large cohort of patients with ACHD, demonstrating that 10.9% had a documented syndromic diagnosis. Genetic abnormalities are nearly evenly divided between chromosomal abnormalities and single-gene disorders. Importantly, we found clear evidence of under-recognition, particularly of 22q11.2 deletion syndrome in TOF, where most undiagnosed cases had never been clinically suspected or tested in adulthood. These findings highlight a persistent diagnostic gap in adult CHD care and suggest that syndromic diagnoses are associated with differences in long-term outcomes, including an independent association with higher mortality.

Our observed syndromic prevalence of 10.9% among patients with ACHD is broadly consistent with previous reports from adult cohorts, such as 13–14% in studies by Edwards et al¹³ and Bracher et al.¹⁶ These findings collectively suggest that a substantial percentage of patients with ACHD have an underlying syndromic condition, despite the lack of routine genetic evaluation in adulthood. Notably, our cohort showed lower frequencies of 22q11.2 deletion and other microdeletions than expected in paediatric series,^{17 18} likely reflecting underdiagnosis in earlier eras and potentially attrition before reaching middle adulthood.¹⁹

Our findings highlight significant differences in clinical characteristics between syndromic and non-syndromic patients. Syndromic individuals were generally younger, more likely to have complex congenital anatomy and had a higher prevalence of pulmonary arterial hypertension and sleep apnoea. Consistent with the study by Oliver et al,²⁰ our study also found that syndromic diagnoses were associated with increased all-cause mortality after adjustment for clinical and anatomical factors, despite fewer traditional cardiovascular risk factors being present among them. These results suggest that syndromic status may reflect additional biological vulnerability not fully explained by conventional risk factors. Syndromic status may represent an additional marker of clinical complexity, with associations extending beyond cardiovascular anatomy to extracardiac comorbidities, frailty and healthcare utilisation.^{21 22}

The TOF subgroup analysis revealed that nearly one-quarter of patients without a prior 22q11.2 diagnosis were clinically not excluded, yet only one-third underwent formal testing, an area that we must improve on. The higher diagnostic yield among patients with right-sided aortic arch underscores the potential value of using anatomical markers to guide genetic evaluation in adulthood. Furthermore, sex and age differences across subgroups suggest evolving recognition patterns and potentially missed opportunities in earlier decades.

Recent retrospective studies have also highlighted the clinical significance of 22q11.2 deletion syndrome in adult TOF populations. van Mil et al²³ reported a prevalence of 13.1% in a cohort of 612 adults with repaired TOF, showing that individuals with the deletion had significantly higher all-cause mortality and died nearly two decades earlier than those without deletion. Similarly, Kauw et al²⁴ found a 7% prevalence in a national cohort of 479 adults with TOF, with 22q11.2DS independently predicting worse long-term survival. Notably, more than half of the deletion cases had not been recognised prior to adulthood. These findings align with our observation that 22q11.2 deletion is frequently underdiagnosed in adult TOF, consistent with reports linking the deletion to differences in long-term outcomes. Our observed prevalence of 4.3% is lower than that in the Toronto and Dutch cohorts.^{23 24} Part of this difference may relate to disease classification: patients with univentricular or other complex forms of CHD, in whom 22q11.2 deletion also occurs, were categorised as ‘complex’ rather than ‘Tetralogy of Fallot’ and therefore excluded from the TOF-specific prevalence estimate. Additional factors, such as variation in case ascertainment, testing practices and genetic referral infrastructure, may also contribute. London’s highly international population may further influence the observed rate, as incomplete medical histories are common among migrants. Incomplete transition from paediatric to adult ACHD services, particularly among socially disadvantaged or migrant populations, may contribute to under-recognition of syndromic conditions such as 22q11.2 deletion in adulthood, consistent with findings from a recent national cohort in England showing lower transition rates in these groups.²⁵ Nonetheless, the consistent association between 22q11.2 deletion and adverse outcomes across studies reinforces the importance of routine genetic evaluation in this population,^{26 27} an area that clearly warrants attention and further investment. However, in current ACHD practice, genetic testing is not uniformly implemented according to standardised criteria and is often performed at the treating physician’s discretion. This may contribute to variability in diagnostic practices and potential under-recognition of syndromic conditions, as reflected in our findings.

Our findings, therefore, reinforce the need for greater attention to syndromic diagnoses in ACHD care. Recognition of syndromic conditions may help clinicians consider broader multidisciplinary assessment, including screening for extracardiac comorbidities and discussions relevant to reproductive counselling.^{28 29} In addition, many syndromic patients with ACHD present with physical disabilities and neurocognitive impairment, necessitating a multidisciplinary approach to care. In particular, as more patients with complex CHD survive into adulthood, the clinical implications of syndromic diagnoses, including their impact on neurocognitive function, fertility and long-term survival, must be more thoroughly integrated into ACHD care models.³⁰ This includes collaboration with neurology, psychiatry, social services and specialised support systems. Structured transition programmes and long-term social support are essential to optimise outcomes and quality of life in this population.

This study has several strengths, including its large sample size, real-world contemporary clinical data and structured classification of genetic abnormalities. However, limitations must be acknowledged. First, syndromic diagnoses were primarily based on clinical documentation rather than on systematic genetic testing, and detailed genetic data were available only for selected subgroups, such as patients with tetralogy of Fallot. Given the retrospective design and reliance on routinely collected clinical data, both genetic conditions and comorbidities may have been underdiagnosed or incompletely captured, leading to potential misclassification and under-ascertainment and resulting in incomplete characterisation of these conditions. Second, the mortality analysis, while adjusted for significant covariates, was exploratory and may be subject to residual confounding. Finally, the generalisability of our findings may be limited by referral bias and variation in diagnostic practices across centres.

In conclusion, our institutional ACHD cohort demonstrates that approximately 1 in 10 patients has an associated syndromic condition, with clinically meaningful implications across both cardiovascular and non-cardiovascular domains, including overall mortality. These findings highlight that enhanced clinical recognition, together with broader access to genetic evaluation, may refine risk stratification and improve long-term outcome assessment in this expanding patient population.

Data availability statement

Data are available upon reasonable request.