To investigate the feasibility of Region 4 Stork (R4S) project used for newborn screening by tandem mass spectrometry in China.

This retrospective study was performed among 362 822 neonates screened by tandem mass spectrometry from May 2015 to April 2016 in Zhejiang newborn screening center. Infants were grouped by screening result category: 83 true positive cases, 360 554 true negative cases and 2 185 false positive cases. Raw data was uploaded into R4S website to perform postanalytical interpretive tools, then results were analyzed with interpretation rules. The comparisons of normal population percentiles were done at five selected percentiles between Zhejiang newborn screening center and R4S project with min-max normalization.

Compared with cutoff system by using R4S project with interpretation rules, the positive predictive value increased from 3.7% to 8.3%, the specificity increased from 99.40% to 99.75%, and the false positive rate declined from 0.6% to 0.2%. The two cases of true positive hyperprolinemia were reported negative, and one case of β-ketothiolase deficiency was misdiagnosis. Totally 311 638 cases in true negative group were resolved by postanalytical interpretive tools, and the remaining 48 916 cases were excluded with interpretation rules. False positive cases were reduced to 897 cases. Results of percentiles comparison showed that levels of some markers were significantly different between zhejiang newborn screening center and R4S project.

R4S project effectively improved the newborn screening performance, whereas leaded to a small number of misdiagnosis and missed diagnosis. Besides, many true negative cases should be excluded with interpretation rules. Optimization should be achieved based on local normal population.(Chin J Lab Med, 2018, 41: 300-304)

Metabolism, inborn errors; Tandem mass spectrometry; Neonatal screening; False positive reactions