Skip to main content
. 2019 Dec 6;16(12):e1002988. doi: 10.1371/journal.pmed.1002988

Table 2. Subgroup analyses based on indication for use from randomised controlled trials: Comparison 1—Magnesium sulphate versus placebo or no treatment.

Outcome and subgroup Studies Participants Method (I2) RR (95% CI) χ2, P value, I2
1.1 Perinatal death
1.1.1 Tocolysis 2 257 F (NA) 7.99 (1.00, 63.49) 4.07, 0.13, 50.8%
1.1.2 Pre-eclampsia 2 9,259 F (0%) 1.01 (0.91, 1.13)
1.1.3 Fetal neuroprotection 4 4,138 F (0%) 0.96 (0.80, 1.15)
1.2 Stillbirth
1.2.1 Tocolysis 2 257 F (NA) 5.70 (0.28, 116.87) 1.45, 0.49, 0%
1.2.2 Pre-eclampsia 3 9,961 F (8%) 0.99 (0.87, 1.12)
1.2.3 Fetal neuroprotection 4 2,122 F (0%) 0.85 (0.40, 1.80)
1.3 Neonatal death
1.3.1 Tocolysis 4 445 R (61%) 0.78 (0.11, 5.67) 0.48, 0.79, 0%
1.3.2 Pre-eclampsia 2 9,259 R (35%) 1.03 (0.64, 1.65)
1.3.3 Fetal neuroprotection 5 3,283 R (0%) 0.86 (0.68, 1.08)
1.4 Death > 28 days, before discharge
1.4.1 Tocolysis 3 412 F (0%) 0.76 (0.19, 3.09) 0.37, 0.83, 0%
1.4.2 Pre-eclampsia 1 9,024 F (NA) 1.13 (0.55, 2.31)
1.4.3 Fetal neuroprotection 1 1,255 F (NA) 0.88 (0.44, 1.74)

Statistically significant effect estimates in bold. Test for heterogeneity represented by I2 statistic; where I2 > 30%, summary estimates were calculated using random-effects meta-analysis. Result of test subgroup differences represented by χ2 statistic, P value, and I2 statistic.

CI, confidence interval; F, fixed-effects; NA, not applicable; R, random-effects; RR, risk ratio.