Trends In and Barriers to the prenatal detection of major Congenital Heart Disease in Alberta

Abstract

Introduction Congenital heart disease (CHD) affects approximately 8-13 per 1000 live births globally, and it is the leading cause of mortality and morbidity among neonates with birth defects. Most CHD can be diagnosed prenatally, which, particularly for more severe disease, is associated with improved outcomes. Despite the ability to diagnose almost all major CHD prenatally, current rates of detection, even in high-income countries, range between 30% to 60%. Whether recent modifications of obstetrical ultrasound (OB US) guidelines have led to improvements in the prenatal detection of certain forms of CHD in North America and internationally is unclear. As well, while the ease of detection of CHD subtypes at OB US is likely a significant predictor of prenatal diagnosis, other factors, including socioeconomic status (SES) and remoteness of residence (RoR) from tertiary care OB US screening and fetal echocardiography services, may also play a relevant role. Objectives

To examine trends in prenatal diagnosis of major CHD in Alberta from 2008 through 2018 To examine the impact of SES and geographic RoR on prenatal detection rates and timing in Alberta from 2008 to 2018 Methods Using provincial databases, we retrospectively identified all fetuses and infants diagnosed between January 2008 to December 2018 with major CHD requiring surgical intervention within the first postnatal year. Objective 1: We evaluated individual lesions and categorized CHD subtypes based on the OB US fetal cardiac views required for detection: Group 1 - 4 chamber view (e.g. hypoplastic left heart syndrome, Ebstein’s anomaly, single ventricle), Group 2 - outflow tract view (e.g. tetralogy of Fallot, double-outlet right ventricle, truncus arteriosus), Group 3 - 3 vessel view/3 vessel tracheal view (3VV-3VT) or other non-standard cardiac views (e.g. coarctation, anomalous pulmonary veins), and 4 - isolated ventricular septal defects (VSDs) using any view. Objective 2: Using maternal residence postal code and geocoding, SES quintiles and geographic distance from fetal tertiary care, both continuous and categorical, were calculated. Outcome measures included the presence of a prenatal diagnosis and the gestational age at prenatal diagnosis when it occurred. Results From 2008-2018, 1405 patients (fetuses and infants) with major CHD were encountered pre and/or postnatally in Alberta, of whom 814 (58%) were diagnosed prenatally. Live births occurred in 1202 (84%), intrauterine fetal death (IUFD) in 47 (3.2%), elective termination of pregnancy (TOP) in 118 (8.3%) and missing data of 38 patient (4.5%). Objective 1: Over the study period, the proportion of prenatal diagnosis of major CHD significantly improved overall from 277/560 [49% 95% CI (confidence interval) 45, 54] in 2008-2012, 255/417 [61%, 95% CI 56, 65] in 2013-2015, to 282/428, [66%, 95% CI 61, 70] in 2016-2018 (p-value <0.001). Linear regression of the proportion of fetal diagnoses by year predicted an increase of 2.3% per year (r2 = 0.73, p=0.0008). By US OB view, Groups 1 and 2 demonstrated a significant increase in prenatal detection, from 75% to 88% (p=0.008) and from 56% to 79% (p=0.0002), respectively. While rates of prenatal detection increased for Group 3 and 4, from 27% to 43% (p=0.007) and 13% to 30% (p=0.04), even in the more recent era, rates remain low compared to Groups 1 and 2, with less than half being detected. Objective 2: Prenatal detection rates of CHD were not associated with SES quintiles; however, the lowest quintile was associated with an estimated 24-28% higher risk of a late diagnosis of CHD. For RoR, maternal residence of >100 km from the tertiary care center was associated with 16% greater risk of a postnatal diagnosis and 47% higher chance of a prenatal diagnosis after 22 weeks of gestation. Conclusions Prenatal detection rates of major CHD have significantly increased in Alberta from 2008-2018. With respect to OB US fetal cardiac views, prenatal detection of CHDs associated with four-chamber and outflow tract view abnormalities have increased. Although CHD associated with abnormalities of the 3VV-3VT and nonstandard views and isolated VSDs have also observed improved prenatal detection, rates for these subgroups remain suboptimal. While SES does not appear to impact rates of prenatal detection in our province, lower SES is associated with later gestational age when a prenatal diagnosis is made. In contrast, greater RoR from tertiary OB US and fetal echocardiography services in Alberta is significantly associated with both reduced rates of prenatal diagnosis and later prenatal diagnosis. Further work is needed to enhance prenatal screening through optimized OB US assessments, and to determine factors responsible for inequity in prenatal detection of CHD particularly for remote pregnancies in Alberta.