I spoke recently at both the Normal Labour and Birth Research Conference in Aarhus, Denmark, and Birth Festival in Drammen, Norway. At both, I summarised what is known about the risks and benefits of fetal heart rate monitoring in labour from research. I ended with a call for more and better research. Speaking with Professor Jenny Gamble afterwards, she challenged me to come up with a list of what I see as the gaps in our knowledge that could be addressed by research.
This post is that list.
Who benefits and who doesn’t?
So far, the evidence from randomised controlled trials has showed a small reduction in neonatal seizures when CTG monitoring rather than IA was used (15 fewer babies out of every 10,000). This difference was seen in the “low-risk” trials but didn’t reach statistical significance in the “high-risk” trials. No difference in perinatal mortality has been found, regardless of the risk categorisation of the population studied.
Most of the high-risk trials piled women in together, regardless of the specific risk factor, not separating outcome data by the individual risk factor. It might be that (for example) infants born to women with pre-eclampsia do worse when CTG monitoring is used, but those born to women who had oxytocin used during the labour do better, and so the effects disappear when both groups of women are included in the same trial. Only three trials have compared CTG monitoring with IA in women with one specific risk factor. These were:
- The Seattle based trial (Luthy, et al., 1987; Shy, et al., 1990) where the risk factor was preterm labour (26 to 32 weeks of gestation). 386 women were enrolled. No differences in perinatal mortality or neonatal seizures were found, but the cerebral palsy rate was higher in the CTG monitoring arm of the trial.
- The unpublished 1989 Pakistan based trial, summarised in the Alfirevic et al., 2017 Cochrane review. 200 women were enrolled. Here the risk factor was meconium staining of the liquor. No differences in perinatal mortality were found and no data were provided about neurological outcomes.
- The 2006 trial conducted in India (Madaan & Trivedi, 2006), enrolling 100 women who had one previous lower segment caesarean section. No deaths occurred and no data regarding neonatal seizures were provided.
The small size and lack of some of the important outcome measures in these studies makes it challenging to make use of them to advise women as they make decisions about fetal heart rate monitoring.
This leaves us in a situation where there is a HUGE list of unanswered questions, of the “if I have <insert risk factor here>, will CTG monitoring help, harm, or make no difference compared to using IA during labour?” variety. We don’t know whether CTG monitoring is better when women:
- Have a multiple pregnancy
- Are in labour before 37 or after 42 weeks
- Have a growth restricted fetus
- Are induced, or their labour is augmented
- Have a breech presentation
- Have pre-eclampsia, diabetes, or some other health condition that might mean the placenta doesn’t work as well as it might
We don’t even know whether changing to CTG monitoring when an abnormal heart rate pattern is heard with IA is a good idea or not. For every situation in which CTG monitoring is currently recommended, we genuinely have no evidence that it helps for that particular issue. It would be useful to have well conducted studies to give us know whether our current practice of CTG use for women with risk factors is doing something useful, or not. That would be a HUGE undertaking however, and I’m not convinced that we will ever see such research being done.
Better understandings of physiology
One of the main shortcomings of CTG monitoring is that it is based on a lot of assumptions and some not very good, really old research that attempted to understand what the various wiggles on a CTG recording meant. As better research has been done, it has become clearer that most of the changes in the heart rate that are seen when oxygen levels fall are signs of compensation. That is, they are adjustments the fetus makes to prevent damage occurring due to low oxygen levels. As a result, there aren’t good indicators that help people reading the CTG to tell the difference between a fetus who is coping with labour and one that is in real trouble. Therefore, we have higher rates of caesarean and instrumental birth and yet continue to have babies born in poor condition, when CTG monitoring is used.
If we were better able to tell the heart rate patterns that meant “I’m ok!” and the ones that meant “get me out of here, now!”, appropriate use of intervention could be applied with the expectation of improved outcomes. This knowledge would be useful not only for women using CTG monitoring but also to understand heart rate patterns identified by IA. Fortunately, there is research (like this) being done in this area and I look forward to seeing this begin to be translated into guidelines.
Is central fetal monitoring safe?
Central fetal monitoring seems like a good idea. Let’s digitise the CTG and send that data to a screen in the staff office in the birth suite where lots of people can see it. In theory, more people looking at the CTG might mean that missing an abnormality happens less often, and perhaps inappropriate over reactions might also be avoided. Whether central fetal monitoring improves safety, has no impact, or even makes things less safe has never been assessed in a randomised controlled trial. Three studies have reported what happened at hospitals where central fetal monitoring had been in use, but broke and wasn’t replaced for some time (Brown et al., 2016; Weiss et al., 1997; Withiam-Leitch et al., 2006). None showed a difference in perinatal outcomes, and some showed higher rates of caesarean section and that doctors and midwives spent less time in birth rooms when central monitoring was in use. Our own research has raised concerns that changes in behaviours that happened with the introduction of central fetal monitoring might undermine safety (Small, et al., 2022).
Before health services continue to introduce these expensive systems, it seems reasonable to expect quality research about central fetal monitoring. If central fetal monitoring systems don’t help, or even make things worse, there are better ways to invest the money they cost into other aspects of maternity care. If they do help, then they should be financially supported by health departments.
Alternatives to CTG monitoring
During the 60 years we have had CTG monitoring, we have never been able to convincingly prove that it works. Rather than keep fiddling about trying to get CTG monitoring to work we could find something else that does the job. I wonder about fetal oximetry. It makes logical sense to me that if you want to know whether the fetus has adequate levels of oxygen, then you should test the levels of oxygen rather than look at the heart rate and try to guess what that means for oxygen levels.
We do have some research on fetal oximetry. Several randomised trials have been brought together in a Cochrane review (East, et al., 2014). The Cochrane team compared CTG use on its own, with CTG use plus oximetry at the same time, looking to see if the addition of oximetry might reduce the caesarean section rate. It didn’t. We have never had big enough trials about oximetry to see if it improved outcomes for the fetus / baby. And we have never run a trial assessing fetal oximetry as an alternative to the CTG rather than an addition to it.
Technology that makes it possible to measure fetal oxygen from a sensor worn on the woman’s abdomen seems to be just across the horizon (here is more information). This is a less invasive option than the previous monitors that had to be inserted through the vagina into the uterus to sit on the cheek of the fetus. Other new approaches to assessing fetal health during labour are also being developed (like this or this). Accepting that CTG monitoring isn’t working would help to make assessing these a research priority.
In summary – there are big holes in our knowledge about how to support the best outcomes for the fetus as it makes its journey during labour. What would you like to see being done in research about fetal monitoring in labour? (And yes, if you are looking for a research supervisor, I can help. Or, if you are looking for a doctorally qualified obstetric researcher to be part of a project that answers these questions, contact me.)
References
Alfirevic, Z., Devane, D., Gyte, G. M. L., & Cuthbert, A. (2017, Feb 03). Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour. Cochrane Database of Systematic Reviews, 2(CD006066), 1-137. https://doi.org/10.1002/14651858.CD006066.pub3
Brown, J., McIntyre, A., Gasparotto, R., & McGee, T. M. (2016, May 31). Birth outcomes, intervention frequency, and the disappearing midwife-potential hazards of central fetal monitoring: a single center review. Birth, 43(2), 100-107. https://doi.org/10.1111/birt.12222
East, C. E., Begg, L., Colditz, P. B., & Lau, R. (2014). Fetal pulse oximetry for fetal assessment in labour. Cochrane Database of Systematic Reviews, 10, CD004075. https://doi.org/10.1002/14651858.CD004075.pub4
Luthy, D. A., Shy, K. K., van Belle, G., Larson, E. B., Hughes, J. P., Benedetti, T., Brown, Z. A., Effer, S., King, J. F., & Stenchever, M. A. (1987, May 01). A randomized trial of electronic fetal monitoring in preterm labor. Obstetrics & Gynecology, 69(5), 687-695. https://www.ncbi.nlm.nih.gov/pubmed/3554055
Madaan, M., & Trivedi, S. S. (2006, Aug). Intrapartum electronic fetal monitoring vs. intermittent auscultation in postcesarean pregnancies. International Journal of Gynecology & Obstetrics, 94(2), 123-125. https://doi.org/10.1016/j.ijgo.2006.03.026
Shy, K. K., Luthy, D. A., Bennett, F. C., Whitfield, M., Larson, E. B., van Belle, G., Hughes, J. P., Wilson, J. A., & Stenchever, M. A. (1990, Mar 01). Effects of electronic fetal-heart-rate monitoring, as compared with periodic auscultation, on the neurologic development of premature infants. New England Journal of Medicine, 322(9), 588-593. https://doi.org/10.1056/NEJM199003013220904
Small, K. A., Sidebotham, M., Fenwick, J., & Gamble, J. (2022, Mar). “I’m not doing what I should be doing as a midwife”: An ethnographic exploration of central fetal monitoring and perceptions of clinical safety. Women & Birth, 35(2), 193-200. https://doi.org/10.1016/j.wombi.2021.05.006
Weiss, P. M., Balducci, J., Reed, J., Klasko, S. K., & Rust, O. A. (1997, Jan). Does centralized monitoring affect perinatal outcome? Journal of Maternal-Fetal and Neonatal Medicine, 6(6), 317-319. https://doi.org/10.3109/14767059709162013
Withiam-Leitch, M., Shelton, J., & Fleming, E. (2006, Dec). Central fetal monitoring: effect on perinatal outcomes and cesarean section rate. Birth, 33(4), 284-288. https://doi.org/10.1111/j.1523-536X.2006.00120.x
Categories: CTG, EFM, IA, Reflections
Tags: central fetal monitoring, Evidence, oximetry, Physiology, research
Birth Small Talk 