|| Checking for direct PDF access through Ovid
Early diagnosis of critical congenital heart defects (CHDs) in newborn babies can improve the likelihood of a good outcome. A number of studies have reported that pulse oximetry is a useful screening method for detection of these defects in asymptomatic newborn babies. However, the value of this test as a newborn screening strategy has been questioned because of concern over false-positive rates and test accuracy. The low prevalence of CHDs also limited the accuracy of previous studies and systematic reviews.The aim of this systematic review and meta-analysis was to assess the performance of pulse oximetry as a screening method in asymptomatic newborn babies for the detection of critical CHDs. A electronic search was performed of MEDLINE (1951–2011), EMBASE (1974–2011), Cochrane Library (2011), and Scisearch (1974–2011) for relevant citations that met predefined selection criteria. Summary estimates of sensitivity and specificity were calculated using a random-effects model; true-positive and false-positive rates at various test thresholds were plotted on a receiver operating characteristic curve. The effect of test timing (<24 vs ≥24 hours after birth) on test accuracy was evaluated.Among the 552 studies screened, 13 eligible studies were identified with data for 229,421 newborn babies. Overall, pulse oximetry showed high specificity (99.9%; 95% confidence interval [CI], 99.7–99.9) and moderately high sensitivity (76.5%; 95% CI, 67.7–83.5) for detection of critical CHDs. The false-positive rate was low (0.014%; 95% CI, 0.06–0.33). When pulse oximetry was done after 24 hours, the false-positive rate for detection of defects was substantially lower than when done before 24 hours (0.05% vs 0.50; P = 0.0017).These findings show that pulse oximetry is a highly specific test with moderate sensitivity for detection of critical CHDs in asymptomatic newborn babies. The data provide strong-evidence use of this test as a screening method in clinical practice.