A comparison of the incidence of undiagnosed congenital heart disease in hospital born and home born children.

Objective: To evaluate the incidence of otherwise undiagnosed congenital heart disease (CHD) in a population of children born in a hospital with routine pulse oximetry (RPO) screening compared to children born at home.

Methods: We reviewed 15 years of births at 2 hospitals for incidence of undiagnosed CHD with RPO. The Health Department reviewed the same data for out of hospital births.

Results: A total of 50,545 hospital births were screened and 1,274 children were born outside the hospital. There were 28 hospital-born babies diagnosed with cyanotic CHD prior to nursery discharge. Only one of these babies would not have been diagnosed without RPO. Three children were missed and there were 3 false positives. Sensitivity and positive predictive value of RPO was 25%, specificity and negative predictive value of RPO exceed 99%. The incidence of CHD requiring RPO diagnosis was roughly one birth per 50,000. Two children born at home with undiagnosed CHD were missed. One of these children presented with neonatal demise.

Conclusion: RPO screening is still valuable in diagnosing CHD only diagnosable with RPO. However, the incidence of CHD requiring RPO to diagnose is similar to other congenital diseases which are not mandated national screening tests. In our limited experience a patient is roughly 25 times more likely to have undiagnosed CHD if they are born outside of a hospital.