Prognostic Value of the Anatomic-Physiologic Classification in Adults With Congenital Heart Disease

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.

Methods: Retrospective study of adults with CHD at the Mayo Clinic (2003-2019). The CHD physiologic stage was assessed at baseline and 36 (24-48) months, and patients were classified into stages A to 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 5321 patients, 1649 (31%), 1968 (37%), 1224 (23%), and 480 (9%) were in stages A, B, C, and D at baseline. Of 5321 patients, 4588 (86%) also had assessments at 36 months, and of these patients, 3347 (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, 1.13 [95% CI, 1.09-1.17]; P<0.001, per unit increase in stage) and change in CHD physiologic stage (hazard ratio, 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: adult; heart defects, congenital; humans; mortality; prognosis.

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.