Background: The false positive rate for critical congenital heart disease (CCHD) oxygen saturation (SpO2) screening is 0.14% but is likely higher at high altitude. Physiologic models suggest there is ∼35 mmHg decrease in PO2 at 6,000 feet elevation with associated decrease in SpO2. We evaluated the prevalence of false positive CCHD screen at a hospital at >6,000 feet elevation with standard SpO2 cut-off of <95% vs. lowered SpO2 threshold <93%. Methods: We conducted a retrospective cohort of newborns that underwent CCHD screening at two extreme altitudes (near sea level vs >6,000 feet). We estimated 144 newborns in each group would provide power 0.9 with alpha 0.01 to detect an increase in initial screen pass rate from 90% to 99.9%. We randomly selected newborns from each site evenly distributed over three years. Well child checks at least 6 weeks of age were reviewed to identify potential false negative screens. Continuous data was presented as medians with interquartile ranges (IQR) and compared between the groups. Categorical data was presented with frequencies and were compared with chi square or Fisher exact. Results: 764 newborns (N = 331 sea level, N = 433 high-altitude) were reviewed. C-Section was more common in the sea level group, and neonatal respiratory conditions were more common in the high-altitude group. Other maternal and birth characteristics did not differ between the groups (Table 1). The median SpO2 was lower in high-altitude newborns vs sea level, 96% vs 99% (p<0.001) respectively for both pre and postductal SpO2 (Table 2). The hospital near sea level uses the Kemper algorithm (any limb SpO2 94% with difference <4% = pass, both limb SpO2 90-94% or difference >3% = repeat). 330/331 (99.7%) sea level newborns passed the screen on first attempt (1 required a repeat measurement and then passed). When applying the Kemper algorithm to the high-altitude newborns, only 387/433 (89.4%) passed on first attempt; 2 failed and 44 (10%) required repeat measurement (p < 0.001, Table 2). The adjusted high-altitude screen used a SpO2 threshold <93% instead of <95%. With this adjusted algorithm, 97% of high-altitude infants passed the first screen, 3% (N=12) required repeat measurements, and two failed the screen. Both of the newborns that failed the adjusted high-altitude screen had normal echocardiograms. Conclusions: Lowering SpO2 CCHD threshold by just 2% from 6,000 feet, significantly reduced false positive screens, increased pass rate on first screen and does not appear to increase false negative screens. This adjustment may reduce nursing time associated with unnecessary repeat measurements, and avoid unnecessary transfers to tertiary hospitals for echocardiograms. However, larger samples are needed to confirm these findings.

Table 1

Summary Demographic Data of Newborns Screened for CCHD at Two Extreme Altitudes

*Neonatal respiratory conditions included pulmonary hypertension, meconium aspiration, lung malformation, sepsis, respiratory distress syndrome, pneumothorax, and transient tachypnea of the newborn

Table 1

Summary Demographic Data of Newborns Screened for CCHD at Two Extreme Altitudes

*Neonatal respiratory conditions included pulmonary hypertension, meconium aspiration, lung malformation, sepsis, respiratory distress syndrome, pneumothorax, and transient tachypnea of the newborn

Close modal
Table 2

Critical Congenital Heart Disease Screening Results for Newborns Screened at Two Extreme Altitudes

*Kemper algorithm Pass = either limb SpO2 > 95% and difference < 4%, Fail = either limb SpO2 < 90%, and Repeat = both limb SpO2 90-94% or difference > 3 **N = 1 ***N = 12, CCHD = Critical Congenital Heart Disease, SpO2 = oxygen saturation

Table 2

Critical Congenital Heart Disease Screening Results for Newborns Screened at Two Extreme Altitudes

*Kemper algorithm Pass = either limb SpO2 > 95% and difference < 4%, Fail = either limb SpO2 < 90%, and Repeat = both limb SpO2 90-94% or difference > 3 **N = 1 ***N = 12, CCHD = Critical Congenital Heart Disease, SpO2 = oxygen saturation

Close modal