BACKGROUND: Switzerland introduced newborn screening (NBS) for CF in 2011, using an IRT/DNA/IRT protocol. This paper describes the results of the first year and compares two versions of the protocol with different IRT cut-offs, particularly effects on recall rate, sensitivity and specificity.
METHODS: IRT cut-offs were >45 ng/ml (99.0th percentile) in period 1 (months 1-4) and >50 ng/ml (99.2nd percentile) in period 2 (months 5-12). In period 2 we abstained from recalls when none of the 7 most common CF mutations were detected and IRT was <60 ng/ml.
RESULTS: In periods 1 and 2, 26,535 and 56,663 tests were performed. Recall rates were 0.94% and 0.48%, respectively (p<0.001), PPV increased from 23% to 47% (p=0.024) and sensitivity was 90% and 100%.
CONCLUSIONS: Raising initial IRT cut-off from the 99.0th to the 99.2nd percentile and abstaining from recalls for children with an IRT<60 ng/ml and carrying no major CFTR mutation significantly reduced the recall rate without affecting sensitivity.