NIHR Signal Computerised interpretation of fetal heart rate during labour does not improve outcomes

Published on 27 June 2017

Using a computerised decision-support system to interpret the outputs of continuous electronic fetal monitoring during labour does not show any advantages over clinicians interpreting the outputs themselves.  This is the first trial to assess decision support of this kind.

In this large NIHR-funded trial, rates of poor neonatal outcomes, caesarean sections and assisted deliveries were not affected by whether clinicians were alerted to potential problems by the decision-support system or by their own interpretation of the data. Developmental outcomes at two years of age were not affected either.

The system used in the trial only looked at fetal heart rate patterns, and didn’t use other data about the labour (which may affect how a clinician manages the labour). This suggests that maternity units should focus on other ways of improving recognition of problems and decision making when they are detected rather than investing in such decision-support systems at this time.

Share your views on the research.

Why was this study needed?

There are about 700,000 live births each year in England and Wales. Continuous electronic monitoring during labour is not routinely used in the UK for all women. But it is used widely when changes to a baby’s heart rate are detected from manual checks, or when women have a higher risk of complications. This selective use aims to detect any abnormalities early, so that clinicians can intervene to speed up the birth or undertake caesarean section to improve outcomes. However, there are sometimes difficulties in interpreting the electronic outputs, and it has been suggested that decision-support software could help.

Programs such as INFANT analyse fetal heart signals and assess overall patterns, generating a colour-coded alert if necessary. INFANT does not make recommendations for action following an alert – this is left to the clinicians, not the software. Up until now, there have not been any studies assessing the use of such decision-support systems and whether they improve neonatal outcomes.

What did this study do?

This was a randomised controlled trial of 46,042 women, carried out in 24 maternity units in the UK and Ireland between January 2010 and August 2013. Each unit dealt with women who were in the intervention group and control group and were aware of the allocation. All the maternity units used a software system called Guardian to display the outputs from the cardiotocograph. 

In the trial, all the women were connected to the Guardian system, and half were also connected to the INFANT decision-support system, which was developed to run on the Guardian system. Neonatal outcomes were recorded. A subset of the surviving children also had a developmental assessment at two years.

The trial had some limitations, including the potential for clinicians working with women in both arms of the trial to learn from the decision-support software, which may have affected how they responded to abnormal cardiotocographs.

What did it find?

  • The incidence of poor neonatal outcome did not differ between the groups: 172 out of 23,263 babies (0.7%) had a poor outcome in the decision-support group, compared with 171 out of 23,351 babies (0.7%) in the no-decision-support group (adjusted risk ratio 1.01, 95% confidence interval [CI] 0.82 to 1.25). The trial sample size was large so that it had 90% power to detect meaningful changes in this primary outcome. Poor neonatal outcome was defined as stillbirth and neonatal death up to 28 days after birth, moderate or severe neonatal encephalopathy, admission to neonatal unit within 48 hours for more than 48 hours because of feeding difficulty, respiratory symptoms, or encephalopathy and evidence of compromise at birth.
  • Just over 50% of all births were spontaneous vaginal births and the same in each group. 50.8% in the decision-support group and 51.2% in the no-decision-support group (adjusted risk ratio 0.99, 95% CI 0.97 to 1.01).
  • The rates of caesarean section (about 24%) and instrumental delivery (about 25%) also showed no differences between the groups.
  • The number of thumbprint entries on the Guardian system was used as a measure for how often a clinician was with the woman during labour. This did not differ significantly between the groups about 4.2 visits per hour. About one visit every 15 minutes.
  • Families were contacted again when their surviving children reached two years of age, and parents were asked to complete the Parent Report of Children’s Abilities – Revised (PARCA-R) questionnaire. From the 6,707 returned questionnaires, no differences between the groups were seen. The score ranges from 0 to 158 with higher scores indicating better ability: the mean composite score in the decision-support group was 98.0, and in the no-decision-support group was 97.2 (mean difference 0.63, 95% CI -0.98 to 2.25).

What does current guidance say on this issue?

The NICE guideline on intrapartum care for healthy women and babies provides information to help define and interpret cardiotocograph traces and to guide the management of labour for women who are at high risk and may be having continuous cardiotocography.

It recommends that a clinician stays with a woman, to continue providing one-to-one support. It does not make any recommendations about the use of computerised decision-support systems.

What are the implications?

This study found that using a decision-support system to help interpret cardiotocographs did not improve clinical outcomes for mothers or babies. The decision-support software used in the trial alerts clinicians to abnormal fetal heart rates, but doesn’t use other information about the labour, such as duration, rate of progress, or meconium (the baby’s first faeces). This other information may affect how a clinician interprets the fetal heart rate and then manages the labour.

This large trial suggests that investing in a computerised decision-support system that only looks at fetal heart rate does not appear to be practically helpful and other things will need looking at to further improve infant mortality rates.

Citation and Funding

INFANT Collaborative Group. Computerised interpretation of fetal heart rate during labour (INFANT): a randomised controlled trial. Lancet. 2017;389(10080):1719-29.

This project was funded by the National Institute for Health Research Health Technology Assessment programme (project number 06/38/01).

Bibliography

NICE. Intrapartum care for healthy women and babies. CG190. London: National Institute for Health and Care Excellence; 2017.

NHS Choices. What happens during labour and birth? London: Department of Health; 2017.

Why was this study needed?

There are about 700,000 live births each year in England and Wales. Continuous electronic monitoring during labour is not routinely used in the UK for all women. But it is used widely when changes to a baby’s heart rate are detected from manual checks, or when women have a higher risk of complications. This selective use aims to detect any abnormalities early, so that clinicians can intervene to speed up the birth or undertake caesarean section to improve outcomes. However, there are sometimes difficulties in interpreting the electronic outputs, and it has been suggested that decision-support software could help.

Programs such as INFANT analyse fetal heart signals and assess overall patterns, generating a colour-coded alert if necessary. INFANT does not make recommendations for action following an alert – this is left to the clinicians, not the software. Up until now, there have not been any studies assessing the use of such decision-support systems and whether they improve neonatal outcomes.

What did this study do?

This was a randomised controlled trial of 46,042 women, carried out in 24 maternity units in the UK and Ireland between January 2010 and August 2013. Each unit dealt with women who were in the intervention group and control group and were aware of the allocation. All the maternity units used a software system called Guardian to display the outputs from the cardiotocograph. 

In the trial, all the women were connected to the Guardian system, and half were also connected to the INFANT decision-support system, which was developed to run on the Guardian system. Neonatal outcomes were recorded. A subset of the surviving children also had a developmental assessment at two years.

The trial had some limitations, including the potential for clinicians working with women in both arms of the trial to learn from the decision-support software, which may have affected how they responded to abnormal cardiotocographs.

What did it find?

  • The incidence of poor neonatal outcome did not differ between the groups: 172 out of 23,263 babies (0.7%) had a poor outcome in the decision-support group, compared with 171 out of 23,351 babies (0.7%) in the no-decision-support group (adjusted risk ratio 1.01, 95% confidence interval [CI] 0.82 to 1.25). The trial sample size was large so that it had 90% power to detect meaningful changes in this primary outcome. Poor neonatal outcome was defined as stillbirth and neonatal death up to 28 days after birth, moderate or severe neonatal encephalopathy, admission to neonatal unit within 48 hours for more than 48 hours because of feeding difficulty, respiratory symptoms, or encephalopathy and evidence of compromise at birth.
  • Just over 50% of all births were spontaneous vaginal births and the same in each group. 50.8% in the decision-support group and 51.2% in the no-decision-support group (adjusted risk ratio 0.99, 95% CI 0.97 to 1.01).
  • The rates of caesarean section (about 24%) and instrumental delivery (about 25%) also showed no differences between the groups.
  • The number of thumbprint entries on the Guardian system was used as a measure for how often a clinician was with the woman during labour. This did not differ significantly between the groups about 4.2 visits per hour. About one visit every 15 minutes.
  • Families were contacted again when their surviving children reached two years of age, and parents were asked to complete the Parent Report of Children’s Abilities – Revised (PARCA-R) questionnaire. From the 6,707 returned questionnaires, no differences between the groups were seen. The score ranges from 0 to 158 with higher scores indicating better ability: the mean composite score in the decision-support group was 98.0, and in the no-decision-support group was 97.2 (mean difference 0.63, 95% CI -0.98 to 2.25).

What does current guidance say on this issue?

The NICE guideline on intrapartum care for healthy women and babies provides information to help define and interpret cardiotocograph traces and to guide the management of labour for women who are at high risk and may be having continuous cardiotocography.

It recommends that a clinician stays with a woman, to continue providing one-to-one support. It does not make any recommendations about the use of computerised decision-support systems.

What are the implications?

This study found that using a decision-support system to help interpret cardiotocographs did not improve clinical outcomes for mothers or babies. The decision-support software used in the trial alerts clinicians to abnormal fetal heart rates, but doesn’t use other information about the labour, such as duration, rate of progress, or meconium (the baby’s first faeces). This other information may affect how a clinician interprets the fetal heart rate and then manages the labour.

This large trial suggests that investing in a computerised decision-support system that only looks at fetal heart rate does not appear to be practically helpful and other things will need looking at to further improve infant mortality rates.

Citation and Funding

INFANT Collaborative Group. Computerised interpretation of fetal heart rate during labour (INFANT): a randomised controlled trial. Lancet. 2017;389(10080):1719-29.

This project was funded by the National Institute for Health Research Health Technology Assessment programme (project number 06/38/01).

Bibliography

NICE. Intrapartum care for healthy women and babies. CG190. London: National Institute for Health and Care Excellence; 2017.

NHS Choices. What happens during labour and birth? London: Department of Health; 2017.

Computerised interpretation of fetal heart rate during labour (INFANT): a randomised controlled trial

Published on 21 March 2017

P Brocklehurst, D Field, K Greene, E Juszczak, R Keith, S Kenyon, L Linsell, C Mabey, M Newburn, R Plachcinski, M Quigley, E Schroeder, P Steer

The Lancet , 2017

Background Continuous electronic fetal heart-rate monitoring is widely used during labour, and computerised interpretation could increase its usefulness. We aimed to establish whether the addition of decision-support software to assist in the interpretation of cardiotocographs affected the number of poor neonatal outcomes. Methods In this unmasked randomised controlled trial, we recruited women in labour aged 16 years or older having continuous electronic fetal monitoring, with a singleton or twin pregnancy, and at 35 weeks' gestation or more at 24 maternity units in the UK and Ireland. They were randomly assigned (1:1) to decision support with the INFANT system or no decision support via a computer-generated stratified block randomisation schedule. The primary outcomes were poor neonatal outcome (intrapartum stillbirth or early neonatal death excluding lethal congenital anomalies, or neonatal encephalopathy, admission to the neonatal unit within 24 h for ≥48 h with evidence of feeding difficulties, respiratory illness, or encephalopathy with evidence of compromise at birth), and developmental assessment at age 2 years in a subset of surviving children. Analyses were done by intention to treat. This trial is completed and is registered with the ISRCTN Registry, number 98680152. Findings Between Jan 6, 2010, and Aug 31, 2013, 47 062 women were randomly assigned (23 515 in the decision-support group and 23 547 in the no-decision-support group) and 46 042 were analysed (22 987 in the decision-support group and 23 055 in the no-decision-support group). We noted no difference in the incidence of poor neonatal outcome between the groups—172 (0·7%) babies in the decision-support group compared with 171 (0·7%) babies in the no-decision-support group (adjusted risk ratio 1·01, 95% CI 0·82–1·25). At 2 years, no significant differences were noted in terms of developmental assessment. Interpretation Use of computerised interpretation of cardiotocographs in women who have continuous electronic fetal monitoring in labour does not improve clinical outcomes for mothers or babies. Funding National Institute for Health Research.

Expert commentary

Continuous cardiotocographic fetal monitoring in labour aims to reduce adverse neonatal outcomes due to hypoxia. It does not always achieve this effect. Perhaps because caregivers do not realise there is an abnormality, or because they do not respond appropriately.

The INFANT study showed that provision of decision support software which alerts caregivers to abnormal fetal monitoring in labour had no effect on short term neonatal outcomes, or the baby’s development aged two.

The study has been criticised because clinical management after the alert was left to the clinicians’ discretion, and because of the risk of “learning” by staff exposed to both arms. Nevertheless, it suggests cardiotocographic decision support, as used in many hospitals around the UK, is unlikely to be of significant benefit in improving patient outcomes.

Jane E Norman, Professor of Maternal and Fetal Health, Director of the Tommy's Centre for Maternal and Fetal Health, Vice Principal, People and Culture, University of Edinburgh MRC Centre for Reproductive Health, Queen's Medical Research Institute