This is a plain English summary of an original research article. The views expressed are those of the author(s) and reviewer(s) at the time of publication.
In utero magnetic resonance imaging (iuMRI) can provide a more accurate diagnosis when used after ultrasound in pregnancy. Adding iuMRI when a brain abnormality is suspected but unclear from ultrasound could help clinicians provide better prognostic advice and support to parents during pregnancy.
This NIHR-funded cohort study compared ultrasound and iuMRI in 570 women at more than 18 weeks' pregnancy carrying a fetus with a suspected brain abnormality. It found that iuMRI was more accurate than ultrasound at correctly identifying a brain abnormality, with an overall difference in accuracy of 25%. Both methods were similarly effective at providing prognostic information on abnormal development. However, iuMRI was better at correctly predicting that the child would have ‘normal’ or ‘at risk’ development than ultrasound.
This suggests that iuMRI would be valuable to further investigate abnormal ultrasound findings.
Why was this study needed?
Ultrasound is the main method used for antenatal screening to identify structural abnormalities during a baby’s development. However, there are limits to its accuracy and sensitivity and this can sometimes lead to incorrect diagnoses. As a result, parents may be given inaccurate advice and prognostic information about their unborn baby. This is a particular concern with brain abnormalities, which affect 3 in 1,000 pregnancies and can lead to serious long-term complications for babies who survive.
Some studies suggest that iuMRI could be a useful addition to ultrasound for detecting brain abnormalities from 18 weeks of pregnancy but the limitations of these studies make it difficult to draw firm conclusions. This clinical study aimed to assess whether iuMRI could aid the prenatal diagnosis of brain abnormalities in babies where ultrasound suggests there may be a problem.
What did this study do?
MERIDIAN was a pragmatic, prospective study involving 570 pregnant women aged over 16 years, carrying a fetus of longer than 18 weeks’ gestation with a suspected brain abnormality detected on ultrasound.
The women were referred for iuMRI at one of six UK hospitals. Radiologists used the iuMRI images to assess each abnormality detected by ultrasound and indicated their confidence in the diagnosis. They either added a diagnosis if appropriate or used ‘diagnosis excluded’ if they disagreed with the ultrasound diagnosis.
Follow-up brain imaging was carried out after birth, pregnancy termination, stillbirth or neonatal death and compared with the original diagnoses to show if the brain abnormalities had been diagnosed correctly (the reference standard against which scans were compared). Data on the subsequent development of surviving infants at age two to three years was collected using the Bayley Scales of Infant Development questionnaire.
What did it find?
- In 386 out of 570 cases (67.7%) both ultrasound and iuMRI were correct at identifying brain abnormalities. In 144 out of 570 cases (25.3%), the ultrasound was incorrect but the iuMRI report was correct. Two fetuses (0.4%) had a correct ultrasound and incorrect iuMRI, and both techniques were incorrect for 38 fetuses (6.8%).
- The overall diagnostic accuracy of ultrasound was 68% and iuMRI 93%. This equates to a difference in accuracy of 25% (95% confidence interval [CI] 21% to 29%). This difference in accuracy was more pronounced as gestational age increased: in the 18-23-week age group, the difference was 23% (95% CI 18% to 27%); in the ≥ 24 weeks group, the difference was 29% (95% CI 23% to 36%).
- Analysis of children’s development after two to three years follow-up showed that the overall difference in diagnostic accuracy was still 24.8% in favour of iuMRI (95% CI 21% to 28.5%). There was little difference in the accuracy of ultrasound and iuMRI for the prognosis of abnormal development. However, iuMRI had a higher number of correct prognoses in surviving infants who were assessed as developmentally ‘normal’ or ‘at risk’.
- Patient acceptability of iuMRI was high, with 95% to 97% of patients stating that they would agree to have iuMRI as part of their care in future. IuMRI was also broadly acceptable to clinicians, although as an adjunct rather than a replacement for ultrasound.
What does current guidance say on this issue?
Guidance from the Royal College of Obstetricians and Gynaecologists acknowledges that the use of ultrasound to screen for fetal abnormalities at 18-20 weeks results in variable detection rates, depending on the type of abnormality.
It recommends that iuMRI can be effective as an adjunct to ultrasound in diagnosing and evaluating structural abnormalities, particularly those involving the fetal central nervous system.
What are the implications?
This study shows that iuMRI is more accurate where a brain abnormality is suspected and that it could be beneficial as an adjunct to ultrasound for providing prognostic information to parents during pregnancy.
It is the first study to include outcome reference diagnosis data, allowing comparison between the initial diagnosis and the outcome of the pregnancy.
IuMRI is a new technique only available at some hospitals, and collaboration between radiologists (for example, the establishment of expert panels) may be required if it is to be offered as part of the diagnostic pathway for fetal abnormalities.
Citation and Funding
Griffiths P, Bradburn M, Campbell M et al. MRI in the diagnosis of fetal developmental brain abnormalities: the MERIDIAN diagnostic accuracy study. Health Technol Assess. 2019;23(49).
This project was funded by the NIHR Health Technology Assessment Programme (project number 09/06/01).
Bibliography
Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre
NIHR Evidence is covered by the creative commons, CC-BY licence. Written content and infographics may be freely reproduced provided that suitable acknowledgement is made. Note, this licence excludes comments and images made by third parties, audiovisual content, and linked content on other websites.
