Clinical features of non-compaction cardiomyopathy across age groups: a retrospective study of 415 patients

Study population and clinical data collection

We conducted a systematic electronic medical record review to identify patients diagnosed with NCM at the Guangdong Provincial People’s Hospital between January 2013 and January 2023. Potential cases were initially identified through the hospital’s electronic health record system using the International Classification of Diseases, Tenth Revision diagnostic codes for NCM. A total of 491 potential cases were identified through this process. Each identified case was then manually reviewed by two independent investigators to confirm the diagnosis based on established Jenni and Chin diagnostic criteria. After excluding patients who did not meet the diagnostic criteria (n=76) and those with incomplete clinical documentation (n=8), a final cohort of 415 patients was established for analysis.

Echocardiography

As part of our retrospective study design, comprehensive echocardiographic examinations were performed using a Philips EPIQ CVx ultrasound system (Philips Healthcare, Best, The Netherlands) with an S8-3 transducer during routine clinical care between January 2013 and January 2023. Left ventricular ejection fraction (LVEF) was calculated using the modified Simpson’s biplane method from apical four-chamber and two-chamber views, following the recommendations of the American Society of Echocardiography. All echocardiographic images from each patient were independently analysed by two experienced cardiac imaging specialists. Any discrepancies in measurements or interpretations were resolved through consensus discussion. The diagnosis of NCM was established based on the modified Chin criteria, requiring the presence of all three of the following echocardiographic features: (1) multiple trabeculae on echocardiography; (2) colour Doppler imaging showed multiple deep trabecular recesses communicating with the ventricle, visible in the apex or middle of the ventricle; (3) the endocardium was a two-layer structure, with a non-dense layer/dense layer >2.

Cardiac electrophysiology

All ECG recordings were independently evaluated by two cardiac electrophysiologists. In cases of disagreement, a third expert electrophysiologist was consulted for final adjudication. Cardiac arrhythmias were systematically categorised into four distinct groups: (1) atrial tachycardia (AT), including atrial flutter, atrial fibrillation, pre-excitation syndrome and supraventricular tachycardia, etc; (2) ventricular tachycardia (VT), including VT, ventricular fibrillation, etc; (3) heart block, including second-degree/third-degree atrioventricular (AV) block, bundle branch block, sick sinus syndrome, etc; (4) others, including ST segment change, dextral cardiac rhythm, etc.

Statistics analysis

Continuous variables were presented as mean±SD for normally distributed data, and as median (25th, 95th percentiles) for non-normally distributed data. Statistical comparisons were performed using one-way analysis of variance or independent t-tests for normally distributed variables, and Wilcoxon signed-rank test for non-parametric data. Categorical variables were analysed using either the χ² test or the McNemar-Bowker test as appropriate. The relationships between continuous variables were assessed using Pearson’s correlation coefficient for normally distributed data or Spearman’s rank correlation coefficient for non-parametric data. All statistical tests were two-sided, with statistical significance set at p<0.05. Data analyses were conducted using R software (V.4.0) and SPSS (V.24.0, IBM, Armonk, New York, USA).

Patient population and demographics

Based on established diagnostic criteria, we identified a final cohort of 415 patients from an initial screening population of 491 individuals. The cohort comprised three distinct phenotypes of NCM: LVNCM (n=316, 76.1%), right ventricular NCM (n=33, 8.0%) and biventricular NCM (n=66, 15.9%). The study population included 229 females (44.1%) and 232 males (55.9%). Age-stratified analysis revealed three groups: infants (0–1 year; n=169), consisting of 94 males (55.6%) and 75 females (44.4%); children and adolescents (1–18 years; n=149), comprising 68 males (45.6%) and 81 females (54.4%) and adults (>18 years; n=97), including 70 males (72.7%) and 27 females (27.8%). Detailed demographic characteristics are presented in table 1, with distribution patterns illustrated in figure 1 and online supplemental figure S3.

Clinical presentation

The spectrum of clinical manifestations is detailed in online supplemental table S1. Heart failure symptoms, primarily presenting as dyspnoea and chest tightness, were observed in 112 patients (27.0%). Cardiac murmur was documented in 102 patients (24.6%). Less common presentations included cyanosis, syncope, chest pain and failure to thrive. Notably, 43 patients (10.4%) remained asymptomatic, while 56 patients (13.5%) presented with respiratory symptoms (cough, sputum production and fever) secondary to pulmonary infections. Cardiac arrhythmias, detected either through symptoms of palpitations or ECG abnormalities, were present in 29 patients (7.0%). Age-stratified analysis revealed that heart failure predominated in adults (>18 years), whereas cardiac murmur was more frequently observed in paediatric groups (online supplemental table S1). Advanced heart failure, defined as New York Heart Association functional class III–IV, was present in 164 patients (39.5%), with the highest prevalence in the adult group (51/97, 52.6%). During hospitalisation, eight patients died. The causes of mortality included refractory heart failure (n=4), septic shock (n=1), disseminated intravascular coagulation (n=1) and acute cardiac allograft rejection (n=2) (table 2).

NCM with CHD

In our cohort, 169 patients presented with isolated NCM, while 246 patients (50.3%) had CHD. The most common associated CHD was isolated patent ductus arteriosus or patent foramen ovale (PDA/PFO), occurring in 91 patients (21.9%). Isolated atrial septal defect (ASD) and ventricular septal defect (VSD) were observed in 29 (7.0%) and 37 (8.9%) patients, respectively. Complex CHD, comprising multiple cardiac anomalies, was identified in 76 patients (18.3%). These complex cases included various combinations of ASD, VSD, tetralogy of Fallot, double outlet right ventricle, transposition of the great arteries and functionally univentricular heart, among others (table 1, figure 1 and online supplemental figure S1).

Cardiac electrophysiology

ECG abnormalities were documented in 159 patients (61.7%). The most frequent cardiac conduction disturbance was complex heart block, affecting 69 patients (16.7%). AT was observed in 44 patients (10.7%), while VT occurred in 29 patients (4.1%). Age-stratified analysis revealed that cardiac arrhythmias were most prevalent in adults (>18 years; 67/97, 69.1%), followed by children and adolescents (1–18 years; 72/149, 48.3%), with the lowest frequency observed in infants (0–1 year; 20/169, 11.8%) (table 1).

NCM with valvular regurgitation

Valvular regurgitation was identified in 310 patients (74.7%) with NCM, while 105 patients (25.3%) demonstrated normal valvular function. Multiple valve regurgitation, affecting more than two valves, was the predominant pattern, observed in 162 patients (39.0%). The prevalence of multivalvular involvement was highest in adults (>18 years; 61/97, 62.9%), followed by children and adolescents (1–18 years; 55/149, 36.9%), with the lowest frequency in infants (0–1 year; 46/169, 27.3%) (table 1 and figure 2).

MR was the second most common valvular manifestation, affecting 75 patients (18.1%). Comparative analysis between patients with NCM with isolated MR and those without valvular regurgitation revealed significant differences in cardiac parameters. In infants (0–1 year), patients with isolated MR showed reduced LVEF (70.00 (65.00, 77.00)% vs 63.50 (40.00, 71.00)%, p=0.012) and fractional shortening (FS) (37.00 (33.00, 44.00)% vs 33.00 (18.50, 39.50)%, p=0.018). Additionally, significant chamber dilation was observed in LA diameter (18.60 (15.50, 21.00) vs 22.50 (19.00, 27.00) mm, p<0.001), LVES diameter (16.00 (12.50, 18.50) vs 21.00 (17.00, 27.10) mm, p<0.001) and LVED diameter (24.00 (20.10, 30.00) vs 32.50 (26.00, 36.00) mm, p<0.001) (online supplemental table S1, figure 3).

In the 1–18 years age group with isolated MR, similar patterns of ventricular dysfunction were observed, with decreased LVEF (70.00 (65.00, 77.00)% vs 63.50 (40.00, 71.00)%, p=0.012) and FS (37.00 (33.00, 44.00)% vs 33.00 (18.50, 39.50)%, p=0.018). Significant chamber enlargement was noted in LA diameter (18.60 (15.50, 21.00) vs 22.50 (19.00, 27.00) mm, p<0.001), LVES diameter (16.00 (12.50, 18.50) vs 21.00 (17.00, 27.10) mm, p<0.001) and LVED diameter (24.00 (20.10, 30.00) vs 32.50 (26.00, 36.00) mm, p<0.001) (online supplemental table S2, figure 3).

Adult patients (>18 years) with isolated MR demonstrated marked systolic dysfunction, with reduced LVEF (55.69±16.50% vs 37.59±13.34%, p=0.008) and FS (29.85±10.79% vs 17.82±7.75%, p=0.016). Significant ventricular dilation was evident in LVES diameter (30.00 (27.00, 40.00) vs 50.00 (43.00, 57.00) mm, p=0.003) and LVED diameter (50.00 (45.00, 52.00) vs 61.00 (57.00, 68.00) mm, p=0.002) (online supplemental table S3, figure 3).

Correlation analysis between MR area and cardiac parameters revealed age-specific patterns. In infants, MR area correlated significantly with chamber dimensions (LVED: r=0.434, p=0.008; LVES: r=0.404, p=0.015; LA: r=0.522, p=0.001) but not with systolic function parameters (LVEF: p=0.448; FS: p=0.365). In the paediatric group (1–18 years), strong correlations were observed with both structural (LVED: r=0.691, p<0.001; LVES: r=0.784, p<0.001; LA: r=0.894, p<0.001) and functional parameters (FS: r=0.584, p=0.004; LVEF: r=0.546, p<0.001). Notably, only this age group showed a significant correlation between MR area and NC ratio (r=−0.441, p<0.04). In adults, MR area correlated with chamber dimensions (LVED: r=0.512, p=0.035; LVES: r=0.508, p=0.037; LA: r=0.748, p<0.001) but not with LVEF (p=0.448) (table 3).

Isolated tricuspid regurgitation (TR) was observed in 60 patients (14.5%). Among patients with isolated TR, significant right heart enlargement was documented only in infants, with increased RV diameter (16.00 (13.00, 18.00) vs 18.00 (16.00, 21.75) mm, p=0.005) and right atrium diameter (16.00 (14.00, 19.00) vs 19.00 (16.00, 22.00) mm, p=0.005). No significant differences were observed in other age groups (online supplemental table S1 and figure S2).

Clinical symptom

Heart failure consistently emerges as the predominant clinical manifestation in patients with NCM across various studies, aligning with our findings.1 6 11 The extensive paediatric cohort study by Waning et al reported heart failure as the most prevalent manifestation (27%).11 Interestingly, our study revealed that among patients under 18 years, heart murmur was the most common presentation (31.1%), with heart failure being least common (18.6%). This divergence likely reflects our centre’s high proportion of patients with NCM with left-to-right shunt CHD, which typically presents with heart murmurs. In adults, while Waning et al identified arrhythmia as the predominant symptom (26%), our study found heart failure most prevalent (54.6%).11 Thromboembolism is traditionally considered to be one of the three major clinical symptoms in patients with NCM,1 10 but in our study, thromboembolism occurred in very few patients, and only in adult patients. In other relevant studies, the proportion of thromboembolism in patients with NCM was also low.7 11

Congenital heart disease

The European Society of Cardiology classifies NCM as an unclassified hereditary cardiomyopathy, distinguishing between isolated NCM and NCM with concomitant congenital heart disease.12 NCM manifests as left ventricular, right ventricular or biventricular involvement, with left ventricular type predominating and isolated right ventricular involvement being rare.12 13 Our study revealed frequent NCM occurrence with left-to-right shunt congenital heart disease, consistent with previous findings.14–16 In this study, most patients with left-to-right shunt congenital heart disease combined with left heart volume overload were found, which was similar to the results of previous studies.16 17 Isolated NCM showed higher prevalence in patients aged >1 year, with PDA/PFO being the most common concurrent congenital heart disease. Notably, 34.3% of infants exhibited multiple congenital heart defects, possibly reflecting earlier hospital presentation due to more severe clinical manifestations.

Arrhythmia

The characteristic deep recessions in NCM may facilitate Purkinje fibre infiltration into the myocardium, potentially causing irregular depolarisation and delayed repolarisation, leading to arrhythmias.18 Previous studies indicate arrhythmias in over 50% of patients with NCM.18–20 Our findings showed similar prevalence in patients >1 year of age, while infants demonstrated a notably lower occurrence (11.8%), possibly reflecting their immature cardiac conduction system. Patients with NCM exhibit various arrhythmias, including ventricular premature beats, AV block, bundle branch block and atrial fibrillation, often with multiple types coexisting.21 22 While previous studies emphasised ventricular arrhythmias,18 22 we found atrial arrhythmias predominant in children and adolescents, with ventricular arrhythmias slightly more common in adults.

Valvular regurgitation

Recent attention has focused on valvular regurgitation in NCM. While valvular disease generally increases with age,23 our study showed comparable prevalence between infant and paediatric groups, suggesting age may not strongly correlate with valvular disease in younger patients with NCM. MR emerged as the most prevalent valvular condition, consistent with previous reports.4 11 20 Zou et al demonstrated strong associations between MR and impaired left ventricular function with geometric remodelling,4 while Gao et al identified increased adverse events in patients with NCM with left ventricular dilation.13 Our analysis confirmed that patients with NCM with MR across all age groups showed significantly impaired left heart function and increased chamber size. This may reflect ventricular remodelling affecting mitral valve mechanics24 and increased left heart volume load.25 Regarding TR, right heart enlargement was observed only in infants, primarily attributed to tricuspid valve deformity.

Limitation

This study has several important limitations that merit consideration. First, its retrospective design inherently introduces potential biases and necessitates cautious interpretation of the findings. Second, our investigation did not address therapeutic interventions in patients with NCM, primarily due to the current absence of standardised treatment protocols for this condition. The lack of consensus regarding optimal management strategies for patients with NCM limited our ability to evaluate treatment outcomes systematically.