Serum screening with Down's syndrome markers to predict pre-eclampsia and small for gestational age: Systematic review and meta-analysis

Table 3 Serum screening among pregnant women and number of women needed to be tested and treated with aspirin to prevent one case of SGA (birth weight < 10^th centile).

Test result	Prevalence SGA (%)	Probability of SGA after testing positive (%)	Risk of SGA after treatment*	Probability of SGA after treatment	NNTest¹	NNTreat²
No test, no treatment³	10.0	10.0	-	10.0	-	-
No test, treat all³	10.0	-	0.90	9.0	-	90
Alpha feto-protein>2.0MoM: Sensitivity 60%; Specificity 98%
Test all, treat test positives	10.0	28.3	0.90	25.4	167	35
Human chorionic gonadotrophin>2.0MoM: Sensitivity 12%; Specificity 94%
Test all, treat test positives	10.0	16.2	0.90	14.6	833	62
Unconjugated estriol<0.75MoM: Sensitvity 37%; Specifcitiy 88%
Test all, treat test positives	10.0	22.0	0.90	19.8	270	45
Pregnancy associated plasma protein A (PAPP-A)<1 ^st centile: Sensitivity 3%; Specificity 99%
Test all, treat test positives	10.0	28.0	0.90	25.2	3333	36
Inhibin A>2.0MoM: Sensitivity 11%; Specificity 98%.
Test all, treat test positives	10.0	33.1	0.90	29.8	909	30
Alpha feto-protein>2.0MoM to predict severe FGR: Sensitivity 22%, Specificity 99%
Test all, treat test positives	1.0	22.0	0.90	19.8	454	45

* RR 0.90 (95% CI 0.84–0.97) Askie et al. Antiplatelet agents for prevention of pre-eclampsia: meta-analysis of individual patient data. Lancet 2007;369:1791–98;¹¹.
¹ NNTest is number needed to test and treat with aspirin to prevent one case of SGA calculated by 1/(proportion true positives (TP) – (proportion TP * RR)).
² NNTreat is number need to treat if only treat test positives with aspirin calculated by 1/(probability after testing positive – probability after treatment).
³ Numbers are equal for all tests regardless of threshold, sensitivity and specificity.
MoM multiples of median
SGA small for gestational age

ISSN: 1471-2393