S0(1 month) = 0.87377
Outcomes
The primary outcome for the PEACOCK study was clinically indicated need for delivery for pre-eclampsia (or delivery for related conditions such as eclampsia or HELLP syndrome) within seven days of assessment. Secondary outcomes included clinically indicated need for delivery for pre-eclampsia within 48 hours of assessment and within 14 days of assessment, perinatal deaths and neonatal unit admission. On analysis of the main trial,(2) it became clear that neonatal unit admissions did not directly reflect neonatal morbidity (as intended), but rather clinician behaviour. In the PHOENIX trial,(2) the proportions of infants with confirmed morbidity diagnoses were similar, but admission for the indication of prematurity was higher in the planned delivery group, and it was not possible to mask trial allocation to the attending clinicians. It was therefore decided that neonatal unit admission could not be used in this cohort as a surrogate marker of neonatal morbidity and further analysis of this secondary outcome was not undertaken. There were no perinatal deaths in the trial, and further analysis of this secondary outcome did not proceed.
Sample Size
It has been recommended that external validation of a prognostic model should ideally involve a minimum of 100 informative events.(8) We estimated that the primary outcome (delivery within seven days due to clinical indication) would occur in around 40% of women receiving expectant management, based on our previous work and other literature. The sample size for estimation of the sensitivity (within 7%) and specificity (within 7%), assuming a sensitivity of 0.90, specificity 0.70, and 95% confidence intervals (2-tailed) required 120 women with the primary outcome (and 180 without) in the expectant management arm, giving a minimum of 10 events per candidate variable. We estimated that two-thirds of the 500 women recruited to PEACOCK would receive expectant management (the group on which the model will be validated). We therefore expected 134 primary outcome events (500 x 68% x 40%).
Statistical Analysis
The validation sample for the primary analysis (for delivery within 7 days), and secondary analysis evaluating clinically indicated need for delivery for pre-eclampsia within 14 days of assessment was restricted to women in PEACOCK who underwent expectant management: that is, women recruited to PHOENIX (and also enrolled in PEACOCK) who were randomised to the expectant management arm and women who declined the PHOENIX trial and who were recruited to the PEACOCK study only who underwent the usual care strategy of expectant management. An additional analysis was conducted for evaluating clinically indicated need for delivery for pre-eclampsia within 48 hours of assessment which additionally included the PEACOCK women randomised to the planned delivery arm in the PHOENIX trial.
The stages of analysis were as follows: external validation of the PREP-S model, limited updating of the PREP-S model by recalibration, assessment of the model performance of the updated PREP-S model, assessment of the predictive performance of PIGF (Quidel test) and sFlt-1:PlGF (Roche test), comparison of PlGF, sFlt-1:PlGF and PREP-S, assessment of the addition of PlGF and sFlt-1:PlGF to the PREP-S model. The performance of the models was assessed by calibration and discrimination. Model discrimination was assessed primarily using ROC areas, and calibration was assessed and reported graphically using calibration plots and estimated calibration slopes. The recalibrations were additionally reported graphically, with actual event rates compared to predicted rates for specified risk groups. Assessment of the model performance in relation to the primary and secondary outcomes were determined using ROC areas. Test performance of PlGF and sFlt-1:PlGF was evaluated with sensitivity, specificity, positive and negative predictive values, and positive and negative likelihood ratios. When PlGF was assessed as a single predictor, we used a PlGF cut-off of <100 pg/ml. This was based on the evidence that in those presenting <35 weeks’ gestation, PlGF <100 pg/ml has a high diagnostic accuracy (0.96; 95% confidence interval, 0.89–0.99) and negative predictive value (0.98; 0.93–0.995) of determining preeclampsia requiring delivery in 14 days.(9) We have previously reported that a PlGF threshold of <100 pg/mL predicted preeclampsia requiring delivery within 14 days or before 37 weeks’ gestation (whichever was sooner) with sensitivity and negative predictive values similar to diagnostic accuracy estimates obtained by using a <5th centile cut-off.(9) When sFlt-1:PlGF ratio was reported, the threshold was >38. Kaplan-Meier survival curves of the time from test to delivery were determined, stratified by four categories of risk determined by the PREP-S model. Assessment of PREP-S, PlGF and sFlt-1:PlGF and the combined model was conducted on the primary outcome and all secondary maternal outcomes.
Missing Data
In line with the approach used in the original PREP study, missing variables were handled as follows: where alanine aminotransferase was not available, aspartate aminotransferase was used instead (like for like); oxygen saturation was assumed to be normal if not recorded in clinical care; no women had exaggerated tendon reflexes, which was imputed as no, as such women were ineligible for the PHOENIX trial; urinary protein:creatinine ratio was derived from 24 hr urinary protein excretion where there was sufficient data to derive a conversion factor; missing values for serum urea were replaced by a value derived from serum creatinine by linear regression (serum urea = 0.053883*serum creatinine + 0.7874831; for numbers derived from linear regression, as described, in those with sufficient data, the correlation between the measurements was 0.5434 for 264 observations).
Funding
This study was supported by the National Institute for Health Research (NIHR) HTA Monitoring Add on Studies Programme (project reference 15/59/06), and National Institute for Health Research Professorship (Chappell RP-2014-05-019).