Page 274 - 16Neonatal Jaundice_compressed

Neonatal jaundice

Bibliographic details Study type & Patient characteristics Test, Reference Standard, Results Reviewers Comments
Evidence level Threshold for a positive test
weight loss)
c-statistic 0.96 (95% 0.93 to 0.98)

Test 3 vs Test 1
p-value for difference< 0.01

Test 3 vs Test 2
p-value for difference = 0.15

Test 2 vs Test 1
p-value for difference = 0.35

Herschel M; Study Type: All consecutive babies admitted to the Objective 1: Objective 1: Unselected population but
Prospective General Care Nursery of a tertiary Diagnostic accuracy of DAT Prevalence of DAT positive results exclusion criteria not defined
Year: 2002 diagnostic study care city hospital. Test and Reference described
Test: Direct Antiglobulin Test 23/659 (3.5%) adequately
Country: USA Evidence Level: II Mean GA: 38.9 ± 1.4 weeks (DAT) done on cord blood of all Reference test a standard one
Accuracy of DAT in detecting haemolysis Blinding – Not reported
41 Mean BW: 3267 ± 480 g newborn babies. (ETCOc = 3.2 microlitre/litre) in babies of
Gender: Males = 47.6%, Reference standard: Haemolysis non-smoking mothers (n = 499)
identified by measuring ETCOc Sensitivity: 10/26 (38.5%)
Ethnicity: Specificity: 466/473 (98.5%)
black - 82.9% levels in all babies at 12 ± 6 hours PPV: 10/17 (58.8%)
white = 9.8% and 24 ± 6 hours. Significant NPV: 466/482 (96.7%)
Hispanic = 3.3% haemolysis defined as ETCOc
Asian = 2% th Accuracy of DAT in detecting haemolysis
Other = 2% levels = 95 centile in babies of (ETCOc = 2.5 microlitre/litre) in babies of
non-smoking mothers at 12 hours all mothers (n = 563)
Results given separately for babies (= 3.2 microlitre/litre), and among Sensitivity: 4/47 (8.5%)
with smoking mothers and non- all babies at 24 hours (= Specificity: 504/516 (97.6%)
smoking mothers. 2.5 microlitre/litre). PPV: 4/16 (25.0%)
NPV: 504/547 (92.1%)
Exclusion: not defined Objective 2:
Accuracy of DAT and ETCOc in Objective 2:
predicting hyperbilirubinaemia Prevalence of hyperbilirubinaemia. In
th
defined as bilirubin reading = 75 babies of non-smoking mothers
centile on the nomogram (TcB 61/499 (12.2%)
readings with BiliChek at the time
of discharge or earlier as clinically Accuracy of positive DAT test in predicting
indicated, and subsequent TSB as hyperbilirubinaemia in babies of non-
deemed necessary) smoking mothers (n = 499)
Sensitivity: 9/61 (14.7%)
Specificity: 430/438 (98.2%)
PPV: 9/17 (52.9%)
NPV: 430/482 (89.2%)
Accuracy of ETCOc
(threshold = 2.5 microlitre/litre) in

210