New tool developed to predict severe risks in mothers with pregnancy complications

Overview

Researchers have developed a new triage tool that could save mothers’ and babies’ lives by predicting whether pregnant women diagnosed with early onset pre-eclampsia are safe to prolong pregnancy.

The study provides the most robust evidence yet to help clinicians decide on management of mothers-to-be with early onset pre-eclampsia – a condition with raised blood pressure and protein in the urine diagnosed before 34 weeks of pregnancy. It provides advice to determine women’s place of delivery, the need for hospital admission, or outpatient monitoring, using risk estimates.

The PREP study, funded by the National Institute for Health Research (NIHR) is published in BMC Medicine. It was led by Queen Mary University (QMUL) and led to the development of two prediction models, which were able to accurately predict the risk of complications in up to 84 per cent of mothers.

The PREP-S model predicted individualised risk to mothers at various time points after pre-eclampsia diagnosis, while PREP-L measured the overall risk of experiencing complications by postnatal discharge.

Around one per cent of pregnant women are diagnosed with early onset pre-eclampsia, which occurs before 34 weeks, which is much more severe than pre-eclampsia occurring later in pregnancy.

Those diagnosed before 34 weeks are at risk of potentially fatal complications including seizures, liver and kidney failure, bleeding disorders and heart and lung problems. The complications necessitate intensive care in a third of these women who are at risk of unexpected clinical deterioration.

The only known cure is delivery of the baby. In early onset pre-eclampsia babies are often delivered extremely prematurely due to perceived risk of complications in the mother. Most require immediate neonatal care and early birth can lead to serious complications such as respiratory problems. An estimated 1,000 babies a year die from complications caused by pre-eclampsia.

Shakila Thangaratinam

Project lead Professor Shakila Thangaratinam, pictured, professor in maternal and perinatal health at QMUL, said: “Often babies are delivered prematurely to ensure the mother is safe from complications. But this decision depends on the individual clinician, with no robust test to guide whether early birth is needed or not.

“Given the scarcity of neonatal intensive care beds and high-dependency hospital wards for mothers and the high costs of these facilities, an accurate estimation of the health risks at various time points after diagnosis of early onset pre-eclampsia is needed to prioritise and plan care.”

The models used information routinely gathered in the NHS to predict the risk of complications. Data used included, mother’s age, gestation at which pre-eclampsia was diagnosed, blood pressure, urine protein level, liver and kidney function, oxygen levels in blood and the need for treatment to control blood pressure and prevent seizures.

Of the 946 women with early onset pre-eclampsia recruited from 53 NHS Trusts for the study, 169 (18 per cent) suffered complications within 48 hours of being diagnosed. Overall 633 (67 per cent) mothers experienced a complication by the time they were discharged from hospital after their babies were delivered.

The PREP-S accurately predicted risk of complications in 84 per cent of mothers by 48 hours of diagnosis, while PREP-L accurately predicted in 82 per cent of mothers by discharge after delivery.

When the models were used in different groups of mothers outside the study, the PREP-S model’s predictions of level of risk were similar to what was actually observed, particularly in women who were grouped to be low and intermediate risk of complications. The PREP-L’s predictions of risk of complications by discharge for individual mothers were very similar to what was observed.

Professor Thangaratinam added: “The PREP models could play a role in helping clinicians decide whether mothers need to be transferred to hospital for intensive maternal and neonatal care.

“Women categorised to be low risk could be followed-up in an outpatient setting, with high- and very high-risk women monitored as inpatients with regular intensive monitoring.

“The next stage is to evaluate the impact of using PREP models in clinical practice for doctors to use – this requires well-planned, robust clinical trials.

“The provision of personalised risk information will allow parents to have the opportunity to discuss the expected outcomes.

“At present the model is available as an Excel spreadsheet and we are developing it as a mobile app for use on tablet computers and phones.”

Professor Hywel Williams, director of the NIHR Health Technology Assessment (HTA) Programme, said: “The NIHR is proud to have supported this independent research which should make a difference to the health of mothers and babies in the NHS.”

Yvonne Muwalo, senior research midwife at St Barts Health NHS Trust, London, who was involved in the study, said: “The tool will help provide women with a better understanding of the risks of their condition. It will aid clinicians in planning management of mothers’ care and to plan appropriate follow-ups and transfer to specialist units if required.”

Marcus Green

Marcus Green, chief executive of the Action on pre-eclampsia (APEC) charity, pictured, which worked with the study team, said: “We welcome the results of this research. It has been a sound study and there is no doubt that triaging care to mums to be who need it most is really important, especially when predicting risk in early onset pre-eclampsia.

“This devastating condition frightens patients, comes on quickly, is unpredictable and can kill. Knowing when to intervene and when to deliver is crucial and this work is very helpful in identifying the women who really need careful medical attention and to ensure they get the care they need.”

The journal article is available to read at https://bmcmedicine.biomedcentral.com/articles/10.1186/s12916-017-0827-3

For more information on the study visit the NIHR Journals Library