Abstract
Background:
The prognostic role of the congenital heart disease (CHD) anatomic/physiologic classification has not been systematically studied. The purpose of this study was to determine whether CHD physiologic stage provided improvement in prognostic power (to predict all-cause mortality) beyond conventional clinical risk models.
Method:
Retrospective study of adults with CHD at Mayo Clinic (2003–2019). The CHD physiologic stage was assessed at baseline and 36 (24–48) months, and patients were classified into stages A-D at these time points. Clinical stability (remaining in the same stage), clinical improvement (moving to less advanced stage), and clinical deterioration (moving to more advanced stage) were determined at 36 months. We defined conventional clinical risk indices as: age/sex, functional class, comorbidities, cardiac procedures, hepatorenal dysfunction, and ventricular/valvular dysfunction.
Results:
Of 5,321 patients, 1,649 (31%), 1,968 (37%), 1,224 (23%), and 480 (9%) were in stages A, B, C and D at baseline. Of 5,321 patients, 4,588 (86%) also had assessments at 36 months, and of these patients 3,347 (73%), 386 (8%), and 855 (19%) had clinical stability, deterioration, and improvement, respectively. Patients with clinical improvement were more likely to have undergone cardiac procedures between both assessments. Both baseline CHD physiologic stage (hazard ratio [HR] 1.13, 95% confidence interval [CI] 1.09–1.17, p<0.001, per unit increase in stage) and change in CHD physiologic stage (HR 1.46, 95 CI 1.32–1.61, p=0.007, per unit increase in stage) were associated with mortality after adjustments for conventional risk indices and provided incremental improvement in prognostic power beyond conventional clinical risk models as evidence by an increase in c-statistic from 0.702 (0.681–0.724 to 0.769 (0.754–0.787).
Conclusions:
The CHD physiologic stage can potentially be used for risk stratification, as well as to monitor disease progression, and response to therapy.
Keywords: Congenital Heart Disease, Mortality, Prognostication
INTRODUCTION
The adult congenital heart disease (CHD) population is unique and heterogeneous, and is characterized by significant anatomic and physiologic differences within the population.1, 2 There are, sometimes, significant variations in the treatment pathways and options for a given CHD lesion, and the choice of therapy, would influence the clinical presentation, natural history, and outcomes for the given lesion.3–7 These differences in the underlying structural lesions, and the physiologic adaptations to therapies, create significant heterogeneity which makes risk stratification very challenging in this population.
The 2018 American College of Cardiology/American Heart Association (ACC/AHA) guidelines for the management of adults with CHD proposed the CHD anatomic and physiologic classification, and this framework provides separate classification for the anatomic lesions (anatomic groups) and the physiologic adaptations to the underlying anatomy (physiologic stages).8 While the anatomic group (simple, moderate, and complex CHD groups) for a given patient does not typically change over time, the physiologic stages (stages A, B, C and D) would change over time depending on the natural history of the disease and response to interventions. The criteria for determining the physiologic stages are based on hemodynamic factors, functional status, aerobic capacity, and end-organ function.8 These variables are not novel but are conventional clinical indices that have been used in previous risk models in the literature.2, 7, 9, 10 What is not known, is whether the CHD physiologic stages provide incremental improvement in prognostic power beyond that of conventional clinical risk models. The purpose of this study was, therefore, to determine whether CHD physiologic stage at baseline, and temporal change in CHD physiologic stage during follow-up was associated with all-cause mortality and provided incremental improvement in prognostic power beyond conventional clinical risk models.
METHODS
Study Population
All data and supporting materials have been provided with the published article. The Mayo Clinic Institutional Review Board approved the study and waived informed consent for patients that provided research authorization. This is a retrospective cohort study of adults (age ≥18 years) with CHD that received care at Mayo Clinic from January 1, 2003 and December 31, 2019. The Mayo Clinic Institutional Review Board approved the study, and the patients were identified through the MACHD (Mayo Adult Congenital Heart Disease) Registry. The medical records, including clinic notes, hospital discharge notes, procedure notes, laboratory data, and echocardiograms, were reviewed in all patients.
CHD Physiologic Stages
Based on the 2018 ACC/AHA guidelines for the management of adults with CHD, patients were categorized into 4 physiologic stages (stage A-D). Table 1 and Supplementary Table S1 shows the criteria for determining the CHD physiologic stage, while Supplementary Table S2 shows the criteria for determining the CHD anatomic groups. In patients with clinical characteristics meeting the diagnostic criteria for >1 of the physiologic stages, the patients were classified into the highest stage as recommended in the guidelines.8 Of note, patients do not have to meet all the diagnostic criteria for a particular physiologic stage to be classified into that stage.
Table 1:
ACC/AHA Congenital Heart Disease Physiologic Stages
| Stage A |
| NYHA FC I symptoms |
| No hemodynamic or anatomic sequelae |
| No arrhythmias |
| Normal exercise capacity |
| Normal renal/hepatic/pulmonary function |
| Stage B |
| NYHA FC II symptoms |
| Mild hemodynamic sequelae (mild aortic enlargement, mild ventricular enlargement, mild ventricular dysfunction) |
| Mild valvular disease |
| Trivial or small shunt (not hemodynamically significant) |
| Arrhythmia not requiring treatment |
| Abnormal objective cardiac limitation to exercise |
| Stage C |
| NYHA FC III symptoms |
| Significant (moderate or greater) valvular disease; moderate or greater ventricular dysfunction (systemic, pulmonic, or both) |
| Moderate aortic enlargement |
| Venous or arterial stenosis |
| Mild or moderate hypoxemia/cyanosis |
| Hemodynamically significant shunt |
| Arrhythmias controlled with treatment |
| Pulmonary hypertension (less than severe) |
| End-organ dysfunction responsive to therapy1 |
| Stage D |
| NYHA FC IV symptoms |
| Severe aortic enlargement |
| Arrhythmias refractory to treatment |
| Severe hypoxemia (almost always associated with cyanosis) |
| Severe pulmonary hypertension |
| Eisenmenger syndrome |
| Refractory end-organ dysfunction |
Abbreviations: ACHD indicates adult congenital heart disease; AP, anatomic and physiological; ASD, atrial septal defect; AVSD, atrioventricular septal defect; CCTGA, congenitally corrected transposition of the great arteries; CHD, congenital heart disease; d-TGA, dextro-transposition of the great arteries; FC, functional class; HCM, hypertrophic cardiomyopathy; l-TGA, levo-transposition of the great arteries; NYHA, New York Heart Association; TGA, transposition of the great arteries; and VSD, ventricular septal defect
We assessed the CHD physiologic stages at 2 time points. (1) CHD physiologic stage at baseline: This assessment was based on clinical indices obtained within 12 months from the baseline encounter in the adult CHD clinic. (2) CHD physiologic stage at follow-up: This assessment was based on clinical indices obtained at 36 (24–48) months from the baseline encounter in the adult CHD clinic. Because functional status was an important part of the CHD physiologic stage, we excluded assessments performed within 90 days following hospitalization for any indication.
To facilitate comparison of CHD physiologic stages between baseline and follow-up, we treated the physiologic stages as continuous variables, assigning the numbers 1, 2, 3, and 4 to stages A, B, C, and D, respectively. Temporal change in CHD (Δ CHD) physiologic stage was calculated as CHD stage at baseline minus stage at follow-up. For instance, a patient in stage B at baseline, and then advanced from stage B to C at 36 months assessment, would have Δ CHD stage of −1 (calculated as 2 minus 3), signifying clinical deterioration. If a patient advanced from stage B to A during follow-up, then the Δ CHD stage would be +1 (calculated as 2 minus 1), signifying clinical improvement. If a patient remained in the same physiologic stage at baseline and during follow-up, the Δ CHD stage would be 0 (calculated as 2 minus 2), signifying clinical stability.
Conventional Clinical Risk Indices
Based on previous studies describing risk stratification in adults with CHD,2, 7, 9–14 we considered the following variables as conventional clinical risk indices: (1) Demographics (age, sex). (2) Functional class (New York Heart Association I-IV). (3) Prior cardiac procedures (the average number of cardiac surgeries, catheter interventions and electrophysiologic procedures per patient). (4) Comorbidities (hypertension, hyperlipidemia, type 2 diabetes, obesity, coronary artery disease, and atrial fibrillation). (5) End-organ function (estimated glomerular filtration rate, and model for end-stage liver disease excluding international normalized ratio). (6) Echocardiographic indices ventricular and valvular function (systemic ventricular ejection fraction, nonsystemic ventricular fractional area change, systemic atrioventricular valve regurgitation, and nonsystemic atrioventricular valve regurgitation). Offline analysis of echocardiographic images was performed in all patients, and assessment of ventricular systolic function was performed using standard technique.15, 16 Hemodynamically significant atrioventricular valve regurgitation was defined as ≥moderate regurgitation assessed using qualitative Doppler echocardiography.
All-cause mortality was ascertained by review of medical records and the Accurint mortality database, and patients that did not reach this endpoint were censored at the last clinical encounter. Cardiovascular death was defined as death due to myocardial infarction, sudden cardiac death, heart failure, stroke, cardiovascular hemorrhage, and cardiovascular procedures.
Statistical Analysis
Data were presented as mean ± standard deviation, median (interquartile range), and count (%). Between-group comparisons were performed using chi square test, analysis of variance test, unpaired t-test, and Wilcoxon rank sum test as appropriate.
The relationship between CHD physiologic stage and all-cause mortality was assessed using Cox regression. The models were adjusted for conventional clinical risk indices (demographics, functional class, cardiac procedures, comorbidities, end-organ function, and echocardiographic indices). Prior to multivariable adjustments, we first created univariable Cox models assessing the relationship between conventional clinical risk indices and all-cause mortality, and covariates with p<0.1 in the univariable models were included in the model assessing the relationship between CHD physiologic stage and mortality. Stepwise backwards selection was used for covariate in the multivariable model, and a p<0.05 was required for a covariate to remain in the model. Age, sex, and CHD anatomic groups were forced into the models. Missing data were addressed using conditional imputation. Similar analyses were performed to assess the relationship between temporal change in CHD physiologic stage and mortality. In these analyses, temporal change in physiologic stage was modeled as time dependent covariate. Exploratory analysis was performed to assess the relationship between CHD physiologic stage and cardiovascular mortality.
To assess the prognostic power of CHD physiologic stage over conventional clinical risk indices, first we created a multivariable Cox model (base model) using the conventional clinical risk indices (demographics, functional class, cardiac procedures, comorbidities, end-organ function, and echocardiographic indices), and this analysis was based on the subset of patients that had CHD physiologic stage assessments at baseline and follow-up. Next, we created a second model by adding the baseline CHD physiologic stage to the base model, and then created a third model by adding both baseline CHD physiologic stage and Δ CHD physiologic stage to the base model. Survival c-statistic and differences in c-statistic were calculated using 1,000 bootstrap samples, and used to quantify prognostic performance of the different models. All statistical analyses were performed with SAS version 9.4 (SAS Institute Inc, Cary NC), and BlueSky Statistics software (version. 7.10; BlueSky Statistics LLC, Chicago, IL, USA), and p value <0.05 was considered to be statistically significant for all analyses.
RESULTS
Baseline CHD Physiologic Stage
We identified 5,321 patients from the registry. The mean age at the time of baseline encounter was 38±16 years, and 2,735 (52%) were men. Of the 5,321 patients, 1,649 (31%), 1,968 (37%), 1,244 (23%), and 480 (9%) were assigned to stages A, B, C and D, respectively based on pre-defined criteria (Supplementary Table S3). The CHD diagnoses of the patients in the cohort are shown in Supplementary Table S3, while the number of patients with available data and the number (percentage) of patients that met the criteria for the different CHD physiologic stages are shown in Supplementary Table S4. Table 2 shows a comparison of the baseline clinical characteristics of patients in the different physiologic stages. The patients in the more advanced physiologic stages were older, had more cardiac procedures and comorbidities, and had worse ventricular function and hepatorenal function (Table 2).
Table 2:
Baseline Characteristics
| All pts (n=5,321) |
Stage A (n=1,649;31%) |
Stage B (n=1,968; 37%) |
Stage C (n=1,224;23%) |
Stage D (n=480; 9%) |
p | |
|---|---|---|---|---|---|---|
|
| ||||||
| Age, years | 38±16 | 35±11 | 38±10 | 43±13 | 44±9 | 0.004 |
| Male sex | 2,735 (52%) | 854 (52%) | 1,007 (51%) | 623 (51%) | 251 (52%) | 0.7 |
| Body mass index, kg/m2 | 27±6 | 26±5 | 28±7 | 28±6 | 27±7 | 0.3 |
| CHD anatomic group * | ||||||
| Simple | 813 (15%) | 508 (31%) | 305 (16%) | --- | --- | --- |
| Moderate | 3,210 (60%) | 1,114 (69%) | 1,417 (72%) | 481 (39%) | 171 (36%) | --- |
| Complex | 1,298 (24%) | 0 | 246 (13%) | 743 (61%) | 309 (64%) | --- |
| Cardiac Procedures | ||||||
| # Cardiac surgeries/pt | 1.6± 0.6 | 1.3± 0.7 | 1.4± 0.5 | 1.9± 0.7 | 2.1± 1.1 | 0.007 |
| # Catheter intervention/pt | 0.4± 0.2 | 0.3± 0.1 | 0.5± 0.3 | 0.4± 0.2 | 0.4± 0.3 | 0.1 |
| # EP intervention/pt | 0.2± 0.1 | 0.1± 0.1 | 0.2± 0.1 | 0.4± 0.2 | 0.3± 0.2 | 0.03 |
| Comorbidities | ||||||
| Hypertension | 1,323 (25%) | 329 (20%) | 442 (23%) | 406 (33%) | 146 (30%) | <0.001 |
| Hyperlipidemia | 948 (18%) | 267 (16%) | 381 (19%) | 219 (18%) | 81 (17%) | 0.1 |
| Diabetes | 354 (7%) | 103 (6%) | 109 (6%) | 101 (8%) | 41 (9%) | 0.4 |
| Obesity | 997 (19%) | 284 (17%) | 369 (19%) | 262 (21%) | 82 (17%) | 0.3 |
| Coronary artery disease | 323 (6%) | 41 (3%) | 104 (5%) | 129 (11%) | 49 (10%) | <0.001 |
| Atrial fibrillation | 811 (15%) | 0 | 121 (6%) | 442 (36%) | 208 (43%) | <0.001 |
| Medications | ||||||
| Loop diuretics | 1,099 (21%) | 16 (1%) | 404 (21%) | 378 (31%) | 301 (63%) | <0.001 |
| Beta blockers | 1,310 (25%) | 109 (7%) | 386 (20%) | 517 (42%) | 298 (62%) | <0.001 |
| ACEI/ARB | 1,378 (26%) | 254 (15%) | 403 (21%) | 486 (40%) | 235 (49%) | <0.001 |
| MRA | 295 (6%) | 2 (0.1%) | 88 (5%) | 108 (9%) | 97 (20%) | <0.001 |
| End-organ function | ||||||
| eGFR, ml/min/1.73 m2 | 98±28 | 113±15 | 105±13 | 76±27 | 64±33 | <0.001 |
| MELD-XI | 9.5 (9.4–12.8) | 9.4 (9.4–10.9) | 9.5 (9.4–11.6) | 10.9(9.8–13.7) | 11.2(10.1–14.3) | <0.001 |
| Echocardiography | ||||||
| S-ventricular EF, % | 57±16 | 62±8 | 54±11 | 56±10 | 49±9 | 0.04 |
| ≥Moderate S-AVVR | 291 (6%) | 0 | 0 | 193 (16%) | 98 (20%) | --- |
| NS-ventricular FAC, % | 34±11 | 46±7 | 33±9 | 29±10 | 24±8 | <0.001 |
| ≥Mod NS-AVVR | 903 (19%) | 0 | 0 | 687 (69%) | 216 (77%) | -- |
Abbreviations: ACEI: Angiotensin converting enzyme inhibitor; ARB: Angiotensin-II receptor blockers; AVV: atrioventricular valve; CHD: congenital heart disease; EP: electrophysiology; EF: ejection fraction; FAC: fractional area change; GFR: glomerular filtration rate; MRA: mineralocorticoid receptor antagonist; MELD-XI: Model for end-stage liver disease the excluding international normalized ratio; NS: non-systemic; S: systemic
Footnote:
CHD anatomic group was based on the anatomic and physiology classification proposed in the 2018 American College of Cardiology/American Heart Association guidelines for the management of adults with congenital heart disease
Note that we excluded patients with Fontan palliation from the assessment of non-systemic ventricular function and atrioventricular valve regurgitation. Hence, the analyses were based on 4, 921 patients. Severity of atrioventricular valve regurgitation was based on qualitative Doppler assessment. The calculation of ejection fraction and fractional area change were based on 2D echocardiography
Outcomes
The median follow-up for the 5,321 patients was 79 (39 −139) months, and during this period 737 (14%) died. The CHD physiologic stages were associated with all-cause mortality. Using stage A as the reference group, stages B, C, and D were associated with 18%, 33%, and 72% increase in the risk of mortality after adjustment for conventional clinical risk indices (Table 3 and Supplementary Table S5). We observed similar results, when the CHD physiologic stages were remodeled as continuous variables, with 21% increase in the risk of mortality for every unit increase in CHD physiologic stage (hazard ratio [HR] 1.21, 95% confidence interval [CI] 1.14–1.29, p<0.001, per unit increase in stage).
Table 3:
Multivariable Cox Model Showing Relationship Between Baseline CHD Physiologic Stage and All-cause Mortality
| HR (95% CI) | p | |
|---|---|---|
|
| ||
| CHD Physiologic Stages | ||
| Stage A | Reference | |
| Stage B | 1.18 (1.06–1.41) | 0.03 |
| Stage C | 1.33 (1.14–1.91) | <0.001 |
| Stage D | 1.72 (1.38–2.09) | <0.001 |
| Demographics | ||
| Age, per years | 1.04 (1.03–1.05) | <0.001 |
| Male sex | 1.14 (0.94–1.37) | 0.2 |
| CHD Severity | ||
| CHD anatomic groups | ||
| Simple | Reference | |
| Moderate | 1.09 (0.92–1.22) | 0.2 |
| Complex | 1.65 (0.99–2.08) | 0.08 |
| # of cardiac surgeries/pt | 1.12 (1.03–1.21) | 0.005 |
| # of EP procedures/pt | 1.42 (1.28–1.59) | 0.02 |
| Comorbidities | ||
| Hypertension | 1.45 (1.10–1.91) | 0.008 |
| Coronary artery disease | 1.38 (1.09–1.76) | 0.03 |
| Atrial fibrillation | 1.26 (1.05–1.62) | 0.02 |
| Echocardiography | ||
| S-ventricular EF, % | 0.97 (0.96–0.99) | 0.005 |
| NS-ventricular FAC, % | 0.98 (0.97–0.99) | 0.008 |
| ≥Mod NS-AVVR | 1.31 (1.21–1.43) | <0.001 |
Abbreviations: AVVR: atrioventricular valve regurgitation; CHD: congenital heart disease; EP: electrophysiology; EF: ejection fraction; FAC: fractional area change; HR: hazard ratio; NS: non-systemic; S: systemic; CI: confidence interval
Footnote: The CHD anatomic groups and physiologic stages were modeled as categorical variables using the simple CHD group and stage A as the reference groups.
Of the 737 deaths, 483 (66%) were cardiovascular deaths. The CHD physiologic stages were associated with cardiovascular mortality. Using stage A as the reference group, stages B (HR 1.22, 95% CI 1.04–1.43, p=0.02), stage C (HR 1.35, 95% CI 1.19–1.98, p<0.001), and stage D (HR 1.88, 95% CI 1.49–2.33, p<0.001) after adjustment for conventional clinical risk indices. There was a 26% increase in the risk of cardiovascular mortality for every unit increase in CHD physiologic stage (HR 1.26, 95% CI 1.19–1.32, p<0.001, per unit increase in stage).
We performed a subgroup analysis to evaluate the prognostic performance of CHD physiologic stages in patients that underwent cardiac interventions (cardiac surgery, catheter intervention, and electrophysiology procedures) versus patients without cardiac interventions during follow-up. The CHD physiologic stage was associated with all-cause mortality in patients that underwent cardiac interventions during follow-up (HR 1.17, 95 CI 1.09–1.25, p=0.007, per unit increase in stage), as well as in patients that did not undergo cardiac interventions during follow-up (HR 1.25, 95 CI 1.18–1.32, p<0.001 per unit increase in stage).
We also performed a subgroup analysis to evaluate the prognostic performance of CHD physiologic stages in patients with simple CHD (n=813) versus patients with moderate/complex CHD (n=4,505). The CHD physiologic stage was associated with all-cause mortality in patients with simple CHD (HR 1.11, 95 CI 1.01–1.29, p=0.02, per unit increase in stage), as well as in patients with moderate/complex CHD (HR 1.29, 95 CI 1.21–1.37, p<0.001 per unit increase in stage).
Temporal Change in CHD Physiologic Stage
Of 5,321 patients, 4,588 (86%) had a follow-up assessment at 29 (25–33) months from baseline. The Central Figure shows a flowchart depicting the movement of patients between different physiologic stages at baseline and follow-up assessments. Of the 4,588 patients, 3,347 (73%) were clinically stable (remained at same stage at baseline and follow-up), 855 (19%) had clinical deterioration (advanced to a worse physiologic stage at follow-up), while 386 (8%) had clinical improvement (moved to a lower physiologic stage at follow-up). Compared to the clinically stable subgroup, those with clinical deterioration had more comorbidities (hypertension, coronary artery disease, atrial fibrillation) and worse ventricular systolic function, while those with clinical improvement were less likely to have atrial fibrillation and more likely to have had cardiac interventions (cardiac surgery and catheter interventions) between baseline and follow-up assessment (Table 4). Although the patients with clinical improvement were more likely to have hemodynamically significant nonsystemic atrioventricular valve regurgitation at baseline as compared to the clinically stable subgroup (106 [28%] versus 533 [16%], p<0.001), they were also more likely to have had atrioventricular valve surgery in the interval between baseline and follow-up assessment (101 [26%] versus 412 [12%], p<0.001).
Central Figure.
A: Flowchart showing the CHD physiologic stage is at baseline and follow-up. The red arrows and boxes signify patients that had clinical deterioration, while the black arrows and boxes signify patients that had clinical improvement during follow-up.
B: Forest plot showing the relationship between CHD physiologic stage at baseline and temporal change in (Δ) CHD physiologic stages during follow-up. The hazard ratios (HR) and 95% confidence intervals (CI) were derived from the multivariable Cox regression model shown in Table 5.
Table 4:
Comparison of Clinical Characteristics Based on Clinical Course As Defined By the Temporal Change in CHD Physiologic Stage
| Baseline Indices | All pts (n=4,588) |
Stable (n=3,347;73%) |
Improved (n=386; 8%) |
Deteriorated (n=855;19%) |
p | p* |
|---|---|---|---|---|---|---|
|
| ||||||
| Age, years | 37±15 | 37±11 | 36±8 | 39±9 | 0.6 | 0.4 |
| Male sex | 2,406 (52%) | 1,755 (52%) | 198 (51%) | 453 (53%) | 0.5 | 0.5 |
| Comorbidities | ||||||
| Hypertension | 1,123 (25%) | 756 (23%) | 81 (21%) | 286 (34%) | 0.4 | <0.001 |
| Hyperlipidemia | 829 (18%) | 611 (18%) | 63 (16%) | 155 (18%) | 0.5 | 0.8 |
| Diabetes | 314 (7%) | 218 (7%) | 24 (6%) | 72 (8%) | 0.7 | 0.8 |
| Obesity | 833 (18%) | 601 (18%) | 73 (19%) | 159 (19%) | 0.2 | 0.7 |
| Coronary artery disease | 249 (5%) | 164 (5%) | 19 (5%) | 66 (8%) | 0.4 | 0.002 |
| Atrial fibrillation | 722 (16%) | 509 (15%) | 41 (11%) | 172 (20%) | 0.02 | <0.001 |
| Echocardiography | ||||||
| S-ventricular EF, % | 55±13 | 57±12 | 55±9 | 48±10 | 0.4 | 0.02 |
| ≥Moderate S-AVVR | 248 (5%) | 167 (5%) | 17 (4%) | 64 (8%) | 0.6 | 0.004 |
| NS-ventricular FAC, % | 37±12 | 36±11 | 38±7 | 29±8 | 0.5 | 0.03 |
| ≥Mod NS-AVVR | 841 (18%) | 533 (16%) | 106 (28%) | 202 (24%) | <0.001 | <0.001 |
|
| ||||||
| Interval procedures | ||||||
| Cardiac surgery | 345 (8%) | 204 (6%) | 98 (25%) | 43 (5%) | <0.001 | 0.3 |
| Catheter procedure | 220 (5%) | 133 (4%) | 59 (15%) | 28 (3%) | 0.004 | 0.5 |
| EP procedure | 101 (2%) | 68 (2%) | 12 (3%) | 21 (3%) | 0.2 | 0.2 |
| Any cardiac procedures | 632 (14%) | 387 (12%) | 163 (42%) | 81 (10%) | <0.001 | 0.08 |
Abbreviations: AVVR: atrioventricular valve regurgitation; CHD: congenital heart disease; EP: electrophysiology; EF: ejection fraction; FAC: fractional area change; NS: non-systemic; S: systemic
Footnote: p denotes comparison between the patients that were clinical stable versus the patients with clinical improvement, while p* denotes comparison between the patients that were clinical stable versus the patients with clinical deterioration.
Of the 4,588 patients, 509 (11%) died during follow-up. Both the CHD physiologic stage at baseline, and temporal change in CHD physiologic stage during follow-up were associated with all-cause mortality. Using stage A as the reference group, stages C and stage D were associated with 19%, and 35% increase in the risk of mortality, respectively, clinical deterioration was associated with more than 2-fold increase in the risk of mortality, while clinical improvement was associated with a 9% reduction in the risk of mortality (Table 5 and Central Figure). We also observed similar results, when the CHD physiologic stages were remodeled as continuous variables. In these analyses, the baseline CHD physiologic stage was associated with a 13% increase in the risk of mortality (HR 1.13, 95 CI 1.09–1.17, p=0.005, per unit increase in stage), while temporal change in CHD physiologic stage was associated with a 46% increase in the risk of mortality (HR 1.46, 95 CI 1.32–1.61, p=0.007, per unit increase in stage).
Table 5:
Multivariable Cox Model Showing Relationship Between CHD Physiologic Stage and All-cause Mortality
| HR (95% CI) | p | |
|---|---|---|
|
| ||
| CHD Physiologic Stages @ baseline | ||
| Stage A | Reference | |
| Stage B | 1.05 (0.84–1.26) | 0.4 |
| Stage C | 1.19 (1.10–1.28) | 0.02 |
| Stage D | 1.35 (1.18–1.52) | 0.007 |
| Δ CHD Physiologic Stages during follow-up | ||
| Clinically stable | Reference | |
| Clinical deterioration | 2.05 (1.49–2.53) | 0.009 |
| Clinical improvement | 0.91 (0.84–0.98) | 0.01 |
| Demographics | ||
| Age, per years | 1.04 (1.03–1.06) | <0.001 |
| Male sex | 1.09 (0.96–1.22) | 0.5 |
| CHD Severity | ||
| CHD anatomic groups | ||
| Simple | Reference | |
| Moderate | 1.04 (0.86–1.27) | 0.4 |
| Complex | 1.34 (0.93–1.69) | 0.2 |
| # of cardiac surgeries/pt | 1.43 (1.22–1.64) | 0.009 |
| Comorbidities | ||
| Atrial fibrillation | 1.19 (1.04–1.42) | 0.01 |
| Echocardiography | ||
| S-ventricular EF, % | 0.96 (0.94–0.98) | 0.001 |
| NS-ventricular FAC, % | 0.97 (0.95–0.99) | 0.02 |
Abbreviations: CHD: congenital heart disease; EF: ejection fraction; FAC: fractional area change; HR: hazard ratio; NS: non-systemic; S: systemic; CI: confidence interval
Footnote: The CHD anatomic groups and physiologic stages were modeled as categorical variables using the simple CHD group and stage A as the reference groups. Temporal change in CHD physiologic stage was modeled as time dependent covariate, and the clinically stable cohort was used as the reference group.
Of the 509 deaths, 358 (70%) were cardiovascular deaths. The baseline CHD physiologic stage was associated with a 19% increase in the risk of cardiovascular mortality (HR 1.19, 95 CI 1.12–1.25, p<0.001, per unit increase in stage), while temporal change in CHD physiologic stage was associated with a 39% increase in the risk of cardiovascular mortality (HR 1.39, 95 CI 1.21–1.53, p<0.001, per unit increase in stage).
Subgroup analyses showed that temporal change in CHD physiologic stage was associated with all-cause mortality both in patients that underwent cardiac interventions during follow-up (HR 1.39, 95 CI 1.24–1.57, p=0.02, per unit increase in stage), and in patients that did not undergo cardiac interventions during follow-up (HR 1.51, 95 CI 1.34–1.66, p=0.008 per unit increase in stage), independent of the baseline CHD physiologic stage.
Incremental Prognostic Value of CHD Physiologic Stage
In order to determine whether the CHD physiology stage provided incremental improvement in prognostic power predicted the risk of mortality above that of conventional clinical risk indices, we created a base model using conventional clinical risk indices (Table 6). The addition of baseline CHD physiologic stage to the base model resulted in an improvement in prognostic power as evidenced by an increase in c-statistic from 0.702 (0.681–0.724) to 0.753 (0.734–0.775). The addition of both baseline CHD physiologic stage and temporal change in CHD physiologic stage to the base model resulted in further improvement in prognostic power as evidence by an increase in c-statistic from 0.702 (0.681–0.724 to 0.769 (0.754–0.787), Table 7.
Table 6:
Multivariable Cox Model Showing Clinical and Hemodynamic Correlates of All-cause Mortality (Base Model)
| HR (95% CI) | p | |
|---|---|---|
|
| ||
| Demographics | ||
| Age, per years | 1.05 (1.03–1.04) | <0.001 |
| Male sex | 1.02 (0.93–1.111) | 0.4 |
| CHD Severity | ||
| CHD anatomic groups | ||
| Simple | Reference | |
| Moderate | 1.09 (1.03–1.15) | 0.006 |
| Complex | 1.97 (1.51–2.44) | <0.001 |
| # of cardiac surgeries/pt | 1.09 (1.02–1.16) | 0.02 |
| # of EP procedures/pt | 1.37 (1.14–1.59) | 0.007 |
| Comorbidities | ||
| Hypertension | 1.27 (1.10–1.41) | 0.006 |
| Coronary artery disease | 1.36 (1.02–1.74) | 0.04 |
| Atrial fibrillation | 1.48 (1.21–1.83) | <0.001 |
| End-organ function | ||
| MELD-XI, per unit | 1.19 (1.12–1.25) | 0.007 |
| Echocardiography | ||
| S-ventricular EF, % | 0.96 (0.94–0.98) | 0.004 |
| NS-ventricular FAC, % | 0.97 (0.95–0.99) | 0.04 |
| ≥Mod NS-AVVR | 1.18 (1.04–1.7) | 0.03 |
Abbreviations: AVV: atrioventricular valve; CHD: congenital heart disease; EP: electrophysiology; EF: ejection fraction; FAC: fractional area change; HR: hazard ratio; MELD-XI: Model for end-stage liver disease the excluding international normalized ratio; NS: non-systemic; S: systemic; CI: confidence interval
Footnote: The CHD anatomic groups were modeled as categorical variables using the simple CHD group as the reference group.
Table 7:
Incremental Prognostic Value of CHD Physiologic Stage
| C-statistic (95%CI) | C-statistics difference * | C-statistics difference † | |
|---|---|---|---|
|
| |||
| Model 1 | 0.702 (0.681–0.724) | ||
| Model 2 | 0.753 (0.734–0.775) | 0.049 (0.037–0.62) | |
| Model 3 | 0.769 (0.754–0.787) | 0.067 (0.053–0.071) | 0.016 (0.002–0.030) |
Abbreviations: CHD: Congenital heart disease; CI: Confidence interval
Footnote:
Model 1: Base model created using conventional clinical indices shown as shown in Table 5
Model 2: Base model + CHD physiologic stage baseline
Model 3: Base model + CHD physiologic stage baseline + temporal change in CHD physiologic stage during follow-up
The C-statistic difference and 95% CI was calculated based on 1000 bootstrap samples. The C-statistic difference
signifies the difference between Model 1 and Model 2, and between Model 2 and Model 3. The C-statistic difference
signifies the difference between Model 1 and Model 3.
DISCUSSION
The ACC/AHA recommend the use of CHD anatomic and physiologic classification for risk stratification in adults with CHD. However, the prognostic value of this classification scheme has not been empirically validated. In this study, we assessed the prognostic role of the CHD physiologic staging scheme, and the main findings are as follows: (1) CHD physiologic stage at baseline, and temporal change in physiologic stage during follow-up were associated with mortality. (2) The addition of CHD physiologic stage to conventional clinical risk indices improved prognostic performance above that of the conventional clinical risk indices alone. (3) About 8% of the cohort had clinical improvement between the baseline and follow-up assessment, and these patients were more likely to have had cardiac interventions between assessments, suggesting that temporal change in CHD physiologic stage can be used to assess response to cardiac intervention.
There has been a significant increase in the annual volume of hospital admission in adults with CHD, leading to an increase in healthcare resource utilization.2, 17–20 Additionally, hospitalizations and mortality in this population leads to loss of economic productivity since most adults with CHD are within the working age.19–21 As a result, there is a critical need for robust and novel strategies aimed at reducing cardiovascular morbidity and mortality in this population. However, data to guide risk stratification in this population are sparse.
Most of the literature on risk stratification in adults with CHD were based on studies conducted in patients with specific CHD lesions, and these studies have identified risk factors such as systemic ventricular dysfunction, non-systemic ventricular dysfunction, atrial fibrillation, impaired aerobic capacity, and prior heart failure hospitalization.13, 22–25 Other studies that reported non-lesion risk scores such as the perioperative ACHD (PEACH) score and the Aristotle score, focused on perioperative risk assessment, and hence not applicable for risk assessment in ambulatory patients.26, 27 The Seattle Heart Failure Model is a well validated risk model in patients with acquired heart disease, but its prognostic role in adults with CHD is not well understood.28 Stefanescu et al tested the prognostic performance of the Seattle Heart Failure Model in adults with CHD, and reported that the model can identify patients at an increased risk for 5-year mortality from heart failure.29 However, this study was based on a limited cohort of 153 patients with 6 selected CHD diagnoses. Hence, it is unknown how well it would apply to the general CHD population. The current study aimed to address this knowledge gap.
Another important observation from the study was that 8% of the cohort had temporal improvement in CHD physiologic stage, and these patients were more likely to have had cardiac surgery or catheter interventions. These results suggest that the physiologic classification scheme can potentially monitor response to cardiac interventions, and for prognostication after cardiac intervention since the patients that had improvement in physiologic stage after a cardiac intervention had lower risk of mortality during follow-up. Conversely, clinical deterioration should prompt an intensification in therapy because of the high risk of mortality associated with clinical deterioration (temporal change in CHD physiologic stage).
An important negative finding in this study was the lack of prognostic association between anatomic complexity (as measured by the ACC/AHA anatomic group) and mortality. This may be related to the fact that the cohort was comprised predominantly of patients with moderate and complex lesions. An alternative explanation may be that the risks conferred by anatomic complexity, is already expressed in the physiologic stage, and in other measures of disease severity use in our models such as the number of cardiac procedures.
Limitations
This was a retrospective study conducted in a single CHD referral center, and it is therefore prone to referral and ascertainment bias. The referral bias was evident in skewed distribution of the cohort with paucity of simple lesions, and predominance of moderate and complex CHD lesions. As a result, the demographic and clinical characteristics of the patients in this study may differ significantly from that of patients in other centers, thereby limiting generalizability. Additionally, the institutional expertise derived from a long cumulative experience of managing patients with complex disease could have influenced the results, and hence it may be difficult to replicate similar outcomes in smaller centers. Notwithstanding, the current study provides a risk stratification framework, and a foundation for further studies to assess is applicability to other CHD cohorts. We did not assess the effect of changes in medical therapy on CHD physiologic stage and outcomes because of difficulties in controlling for such interventions in retrospective study. Such knowledge gaps can be addressed in a prospective study.
Conclusions
The CHD physiologic stage had superior prognostic performance as compared to conventional clinical risk indices, and hence can be used for risk stratification. Furthermore, temporal change in the CHD physiologic stage could potentially be used to monitor disease progression, and response to therapy. To the best of our knowledge, this is the first study to systematically test the prognostic performance of ACC/AHA CHD classification scheme. Further studies are required to validate these results in another CHD population with different clinical characteristics, and to determine whether clinical decision making based on CHD physiologic staging would improve clinical outcomes in this population.
Supplementary Material
CLINICAL PERSPECTIVE.
What is new?
Of 4,588 patients, 73%, 8%, and 19% had clinical stability, deterioration, and improvement, respectively. Patients with clinical improvement were more likely to have undergone cardiac procedures between both assessments. Both CHD physiologic stage at baseline and temporal change in CHD physiologic stage were associated with mortality after adjustments for conventional risk indices (age/sex, functional class, comorbidities, cardiac procedures, hepatorenal dysfunction, and ventricular/valvular dysfunction).
What are the clinical implications?
Therefore, the CHD physiologic stage can potentially be used for risk stratification, and temporal change in the CHD physiologic stage could potentially be used to monitor disease progression, and response to therapy.
Funding:
Dr. Egbe is supported by National Heart, Lung, and Blood Institute (NHLBI) grants (R01 HL158517 and R01 HL160761). The MACHD Registry is supported by the Al-Bahar Research grant.
Abbreviations:
- ACC/AHA
American College of Cardiology/American Heart Association
- CHD
congenital heart disease
- HR
hazard ratio
- CI
confidence interval
Footnotes
Conflict of Interest: none
Disclosures: none
REFERENCES
- 1.Egbe AC, Miranda WR, Pellikka PA, DeSimone CV and Connolly HM. Prevalence and Prognostic Implications of Left Ventricular Systolic Dysfunction in Adults With Congenital Heart Disease. Journal of the American College of Cardiology. 2022;79:1356–1365. [DOI] [PubMed] [Google Scholar]
- 2.Zomer AC, Vaartjes I, van der Velde ET, de Jong HM, Konings TC, Wagenaar LJ, Heesen WF, Eerens F, Baur LH, Grobbee DE and Mulder BJ. Heart failure admissions in adults with congenital heart disease; risk factors and prognosis. International journal of cardiology. 2013;168:2487–93. [DOI] [PubMed] [Google Scholar]
- 3.Egbe AC, Miranda WR, Anderson JH and Borlaug BA. Hemodynamic and Clinical Implications of Impaired Pulmonary Vascular Reserve in the Fontan Circulation. Journal of the American College of Cardiology. 2020;76:2755–2763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Marelli A Trajectories of care in congenital heart disease - the long arm of disease in the womb. Journal of internal medicine. 2020;288:390–399. [DOI] [PubMed] [Google Scholar]
- 5.Egbe AC, Anderson JH, Ammash NM and Taggart NW. Left Ventricular Remodeling After Transcatheter Versus Surgical Therapy in Adults With Coarctation of Aorta. JACC Cardiovascular imaging. 2020;13:1863–1872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Egbe AC, Vojjini R, Badawy M, Jain V, Bonnichsen CR, Reddy YNV, Obokata M and Borlaug BA. Heightened Dependence of Left-Heart Filling Pressures on Right-Heart Failure in Congenital Heart Disease. The Canadian journal of cardiology. 2021;37:131–139. [DOI] [PubMed] [Google Scholar]
- 7.Diller GP, Arvanitaki A, Opotowsky AR, Jenkins K, Moons P, Kempny A, Tandon A, Redington A, Khairy P, Mital S, Gatzoulis M, Li Y and Marelli A. Lifespan Perspective on Congenital Heart Disease Research: JACC State-of-the-Art Review. Journal of the American College of Cardiology. 2021;77:2219–2235. [DOI] [PubMed] [Google Scholar]
- 8.Stout KK, Daniels CJ, Aboulhosn JA, Bozkurt B, Broberg CS, Colman JM, Crumb SR, Dearani JA, Fuller S, Gurvitz M, Khairy P, Landzberg MJ, Saidi A, Valente AM and Van Hare GF. 2018 AHA/ACC Guideline for the Management of Adults With Congenital Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Journal of the American College of Cardiology. 2019;73:e81–e192. [DOI] [PubMed] [Google Scholar]
- 9.Diller GP, Kempny A, Alonso-Gonzalez R, Swan L, Uebing A, Li W, Babu-Narayan S, Wort SJ, Dimopoulos K and Gatzoulis MA. Survival Prospects and Circumstances of Death in Contemporary Adult Congenital Heart Disease Patients Under Follow-Up at a Large Tertiary Centre. Circulation. 2015;132:2118–25. [DOI] [PubMed] [Google Scholar]
- 10.Tutarel O, Kempny A, Alonso-Gonzalez R, Jabbour R, Li W, Uebing A, Dimopoulos K, Swan L, Gatzoulis MA and Diller GP. Congenital heart disease beyond the age of 60: emergence of a new population with high resource utilization, high morbidity, and high mortality. European heart journal. 2014;35:725–32. [DOI] [PubMed] [Google Scholar]
- 11.Egbe AC, Miranda WR, Dearani J, Kamath PS and Connolly HM. Prognostic Role of Hepatorenal Function Indexes in Patients With Ebstein Anomaly. Journal of the American College of Cardiology. 2020;76:2968–2976. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Egbe AC, Miranda WR, Jain CC and Connolly HM. Right Heart Dysfunction in Adults With Coarctation of Aorta: Prevalence and Prognostic Implications. Circulation Cardiovascular imaging. 2021:CIRCIMAGING121013075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Egbe AC, Miranda WR and Connolly HM. Role of Echocardiography for Assessment of Cardiac Remodeling in Congenitally Corrected Transposition of Great Arteries. Circulation Cardiovascular imaging. 2022;15:e013477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Egbe AC, Miranda WR, Jain CC and Connolly HM. Prognostic Implications of Progressive Systemic Ventricular Dysfunction in Congenitally Corrected Transposition of Great Arteries. JACC Cardiovascular imaging. 2022;15:566–574. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L, Flachskampf FA, Foster E, Goldstein SA, Kuznetsova T, Lancellotti P, Muraru D, Picard MH, Rietzschel ER, Rudski L, Spencer KT, Tsang W and Voigt JU. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography. 2015;28:1–39 e14. [DOI] [PubMed] [Google Scholar]
- 16.Rudski LG, Lai WW, Afilalo J, Hua L, Handschumacher MD, Chandrasekaran K, Solomon SD, Louie EK and Schiller NB. Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography. 2010;23:685–713; quiz 786–8. [DOI] [PubMed] [Google Scholar]
- 17.O’Leary JM, Siddiqi OK, de Ferranti S, Landzberg MJ and Opotowsky AR. The Changing Demographics of Congenital Heart Disease Hospitalizations in the United States, 1998 Through 2010. JAMA : the journal of the American Medical Association. 2013;309:984–986. [DOI] [PubMed] [Google Scholar]
- 18.Lopez KN, Morris SA, Sexson Tejtel SK, Espaillat A and Salemi JL. US Mortality Attributable to Congenital Heart Disease Across the Lifespan From 1999 Through 2017 Exposes Persistent Racial/Ethnic Disparities. Circulation. 2020;142:1132–1147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gilboa SM, Devine OJ, Kucik JE, Oster ME, Riehle-Colarusso T, Nembhard WN, Xu P, Correa A, Jenkins K and Marelli AJ. Congenital Heart Defects in the United States Estimating the Magnitude of the Affected Population in 2010. Circulation. 2016;134:101–+. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Marelli AJ, Ionescu-Ittu R, Mackie AS, Guo L, Dendukuri N and Kaouache M. Lifetime prevalence of congenital heart disease in the general population from 2000 to 2010. Circulation. 2014;130:749–56. [DOI] [PubMed] [Google Scholar]
- 21.Moons P, Bovijn L, Budts W, Belmans A and Gewillig M. Temporal trends in survival to adulthood among patients born with congenital heart disease from 1970 to 1992 in Belgium. Circulation. 2010;122:2264–72. [DOI] [PubMed] [Google Scholar]
- 22.Egbe AC, Miranda WR, Ammash NM, Ananthaneni S, Sandhyavenu H, Farouk Abdelsamid M, Yogeswaran V, Kapa S, Fatola A, Kothapalli S and Connolly HM. Atrial Fibrillation Therapy and Heart Failure Hospitalization in Adults With Tetralogy of Fallot. JACC Clin Electrophysiol. 2019;5:618–625. [DOI] [PubMed] [Google Scholar]
- 23.Egbe AC, Miranda WR, Connolly HM and Borlaug BA. Coarctation of aorta is associated with left ventricular stiffness, left atrial dysfunction and pulmonary hypertension. American heart journal. 2021;241:50–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Egbe AC, Bonnichsen C, Reddy YNV, Anderson JH and Borlaug BA. Pathophysiologic and Prognostic Implications of Right Atrial Hypertension in Adults With Tetralogy of Fallot. J Am Heart Assoc. 2019;8:e014148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Valente AM, Gauvreau K, Assenza GE, Babu-Narayan SV, Schreier J, Gatzoulis MA, Groenink M, Inuzuka R, Kilner PJ, Koyak Z, Landzberg MJ, Mulder B, Powell AJ, Wald R and Geva T. Contemporary predictors of death and sustained ventricular tachycardia in patients with repaired tetralogy of Fallot enrolled in the INDICATOR cohort. Heart. 2014;100:247–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Constantine A, Costola G, Bianchi P, Chessa M, Giamberti A, Kempny A, Rafiq I, Babu-Narayan SV, Gatzoulis MA, Hoschtitzky A, Shore D, Aw TC, Ranucci M and Dimopoulos K. Enhanced Assessment of Perioperative Mortality Risk in Adults With Congenital Heart Disease. Journal of the American College of Cardiology. 2021;78:234–242. [DOI] [PubMed] [Google Scholar]
- 27.Lacour-Gayet F, Clarke D, Jacobs J, Comas J, Daebritz S, Daenen W, Gaynor W, Hamilton L, Jacobs M, Maruszsewski B, Pozzi M, Spray T, Stellin G, Tchervenkov C, Mavroudis and Aristotle C. The Aristotle score: a complexity-adjusted method to evaluate surgical results. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery. 2004;25:911–24. [DOI] [PubMed] [Google Scholar]
- 28.Levy WC, Mozaffarian D, Linker DT, Sutradhar SC, Anker SD, Cropp AB, Anand I, Maggioni A, Burton P, Sullivan MD, Pitt B, Poole-Wilson PA, Mann DL and Packer M. The Seattle Heart Failure Model: prediction of survival in heart failure. Circulation. 2006;113:1424–33. [DOI] [PubMed] [Google Scholar]
- 29.Stefanescu A, Macklin EA, Lin E, Dudzinski DM, Johnson J, Kennedy KF, Jacoby D, DeFaria Yeh D, Lewis GD, Yeh RW, Liberthson R, Lui G and Bhatt AB. Usefulness of the Seattle Heart Failure Model to identify adults with congenital heart disease at high risk of poor outcome. The American journal of cardiology. 2014;113:865–70. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.