Poster Presentation Society of Obstetric Medicine of Australia and New Zealand and Australasian Diabetes in Pregnancy Society Joint Scientific Meeting 2017

Validation of clinical practice guidelines for predicting pre-eclampsia and tailoring aspirin prophylaxis: individual participant data meta-analysis   (#118)

Ziad TA Al-Rubaie 1 , Lisa M Askie 2 , H Malcolm Hudson 2 3 , Joel G Ray 4 , Gregory Jenkins 5 , Sarah J Lord 1 2
  1. School of Medicine, The University of Notre Dame Australia, ORAN PARK, NEW SOUTH WALES, Australia
  2. NHMRC Clinical Trial Centre, University of Sydney, Sydney, NSW, Australia
  3. Department of Statistics , Macquarie University, Sydney, NSW, Australia
  4. Departments of Medicine, Health Policy Management and Evaluation, and Obstetrics and Gynecology, St. Michael's Hospital, University of Toronto, Toronto, Ontario, Canada
  5. Department of Obstetrics, Westmead Hospital, Westmead, NSW, Australia

Introduction
The Society of Obstetric Medicine of Australia and New Zealand (SOMANZ) considers low-dose aspirin for preventing pre-eclampsia in pregnant women at elevated risk. Doppler ultrasound and serum biomarkers have demonstrated very good accuracy for identifying high-risk women, but are not widely available.  The National Institute for Health and Care Excellence (NICE) and United States Preventive Services Task Force (USPSTF) provide guidelines to identify high-risk women using routinely collected maternal risk factors, but these require clinical validation.

Objective
To assess the accuracy of NICE and USPSTF guidelines for predicting pre-eclampsia.

Methods
We conducted an individual-participant data meta-analysis using the Perinatal Antiplatelet Review of International Studies (PARIS) dataset. Randomised controlled trials were eligible if they enrolled pregnant women before 28 weeks’ gestation, reported risk factors, and assessed pre-eclampsia. Women assigned to no antiplatelet therapy were included. Both guidelines recommend aspirin if ≥1 high-risk factors or ≥2 moderate-risk factors. Two moderate-risk factors (body mass index and pregnancy interval) were unavailable. For each guideline, we assessed sensitivity, specificity, positive (PPV) and negative (NPV) predictive value for predicting pre-eclampsia, and the number-needed-to-screen (NNS) to prevent one pre-eclampsia event.

Results
Eleven trials (n=10,659 women) had sufficient data to assess the guidelines. The pooled prevalence of pre-eclampsia was 8·3%. NICE sensitivity was 57·1% (95%CI 53·8–60·3%), specificity 72·4% (71·5–73·2%), PPV 15·7% (14·5–17·0%), NPV 94·9% (94·4–95·4%), NNS 211. USPSTF sensitivity was 58·7% (55·4–61·9%), specificity 69·8% (68·8–70·7%). For nulliparous women, NICE sensitivity was 25·4% (21·3–30·0%), specificity 91·4%(90·6–92·1%), PPV 16·7% (13·9–19·9%) and NPV 94·7% (94·1–95·3%), NNS 616.

Conclusion
NICE and USPSTF guidelines offer a simple and specific approach for identifying women at high-risk of pre-eclampsia where more advanced screening methods are not available. These findings can inform the development of Australian and international guidelines, and patient counselling for aspirin prophylaxis.