I suspect that when asked to discuss the evidence about CTG monitoring, many maternity clinicians would struggle to name a specific paper, and those who did would probably mention the “Dublin trial” (McDonald, et al., 1985). There have been twelve randomised controlled trials which have compared CTG monitoring with intermittent auscultation (IA). The first ever trial was published by the Denver, USA based team of Haverkamp, Thompson, McFee, and Cetrulo, in 1976. When I first read it a few years ago, I found it fascinating. I’d like to draw it to people’s attention as I think it raises an interesting question which has been essentially ignored in research since then.
In 1976, CTG monitoring had been in use for more than a decade and it was widely assumed to have played an important role in the fall in perinatal mortality that had been documented over that decade. As Haverkamp et al. pointed out, leading obstetric authorities of the time were calling for the routine use of CTG monitoring, yet there was no real evidence to support this. Another research team had previously attempted to obtain ethical approval to conduct a randomised controlled trial, but this had been denied, on the basis that it was unethical to withhold CTG monitoring (Martin, 1998). (The irony of which is now apparent in retrospect!)
483 women considered to be at high risk for poor perinatal outcome were recruited for the trial. Focussing on a high risk population makes sense, because if CTGs work as proposed, then the greatest benefit would be expected when poor outcomes occur most often. It also reduces the number of women needed to show a difference in outcome. A power calculation to guide the sample size was not used (this wasn’t a typical approach to research at the time). A numerical scoring system for risk was used (but the details of this were not given), and in addition women with meconium stained liquor, who had abnormal heart sounds on auscultation at admission, or when an oxytocin infusion was in use were also approached.
Women were approached in early labour or if they were being induced and those who agreed to take part in the trial had a fetal scalp electrode and an intrauterine pressure catheter inserted. Once this was done, women then were randomised to either IA or CTG monitoring. When randomised to IA, the CTG monitor was moving out of the birth room, covered, and the CTG trace was not used to guide management during labour. This group of women had the fetal heart rate monitored by IA by a “study nurse” every 15 minutes in the first stage and every 5 minutes in the second stage of labour. Women randomised to CTG monitoring had the monitor in the room, did not have IA, and the CTG trace was used to guide clinical management.
It is this particular design that makes this study so interesting, as it meant that a CTG trace was available retrospectively for all women enrolled in the trial. No subsequent trial has used the same approach. Doing the study this way reduced the difference between the “intervention” and the “control” groups as the only difference was the addition of IA in the “control” group. In other trials only women randomised to CTG monitoring would have the interventions associated with internal monitoring, and differences in outcome (like infection) might have been due to this.
So what did they find? There were no intrapartum deaths, and three neonatal deaths: two of which occurred in the group monitored by CTG. This didn’t reach statistical significance due to the small sample size. There were no significant differences in cord blood gas results, low Apgar scores at 1 or 5 minutes, nor any measure of neonatal morbidity, including seizures. There was a statistically significant increase in caesarean section (both for any reason, and specifically for fetal concerns), a higher rate of general anaesthesia, and a higher rate of postnatal infection for women when CTG monitoring was used, which remained the case when corrected for the higher rate of caesarean section. The authors were surprised by these findings.
The apparent lack of improvement in perinatal outcome in high-risk pregnancies by the use of electronic fetal monitoring in this study in comparison to auscultation is unexpected, when considering the current opinion expressed in the world literature.Haverkamp, et al,., 1976, p. 316
The really interesting finding, buried in the midst of the paper, was that there were more “ominous” CTG abnormalities in “early” labour (prior to five cm of cervical dilatation) when IA was not in use. The authors had two theories to explain this difference.
The patients who were auscultated had individualized nursing care with one of the project nurses at the bedside almost continually. Very close physical contact with the patient was necessary for the nurse to auscultate the fetal heart tones adequately. This was not true to the same degree with the monitored group. Nursing attention to the gravida with respect to maternal comfort, emotional support, and ‘laying on of hands’ could have a significant impact on the fetus. Also, despite the fact that a monitor was attached to the patient, the machinery was not at the bedside but in the hall. Patients who have a monitor adjacent to their bed are often bothered and stressed by flashing light, the sound produced by each fetal heart beat, and hearing or seeing decelerations even when they are benign. The authors have the impression that the reassuring psychological atmosphere created by personal nurse interaction and the absence of the recording machine in auscultated patients contributed to the excellent infant outcome in auscultated patients.Haverkamp, et al,., 1976, p. 316
Which leads me to the question that remains unanswered in all subsequent research – could it be that heightened anxiety in the birthing women because of the presence of the CTG monitor and the relative absence of personalized midwifery / nursing support created the very problem that the CTG monitor was designed to address? There are some clues which back this up as a possibility. Shalev, et al. (1985) found significant increases in the levels of several stress hormones during and for 30 minutes after antenatal CTG monitoring was performed. More recently, Mancuso, et al. (2008) recorded higher levels of anxiety after antenatal CTG monitoring. Higher levels of stress hormones may cause redistribution of blood away from the placenta, decreasing fetal oxygen levels and therefore initiating compensatory fetal heart pattern changes (Buckley, 2015).
What fascinates me the most about this unanswered, yet really important question is that in the 44 years since the possibility that the use of CTG monitoring might cause fetal heart rate pattern abnormalities, no research group has followed up on this. I learned through doing my PhD that there are certain areas of knowledge that Obstetrics focusses on and other that are simply ignored as they don’t support the grand narrative. This seems to be one of those.
Buckley, S. J. (2015). Hormonal physiology of childbearing: Evidence and implications for women, babies, and maternity care. https://www.nationalpartnership.org/our-work/resources/health-care/maternity/hormonal-physiology-of-childbearing.pdf
Haverkamp, A. D., Thompson, H. E., McFee, J. G., & Cetrulo, C. (1976). The evaluation of continuous fetal heart rate monitoring in high-risk pregnancy. American Journal of Obstetrics and Gynecology, 125(3), 310-320.
Mancuso, A., Vivo, A., Fanara, G., Denaro, A., Laganà, D., Accardo, F. (2008). Effects of antepartum electronic fetal monitoring on maternal emotional state. Acta Obstetricia et Gynecologica Scandinavica, 87(2), 184 – 189. https://dx.doi.org/10.1080/00016340701823892
Martin, C. (1998). Electronic fetal monitoring: a brief summary of its development, problems and prospects. European Journal of Obstetrics, Gynecology, and Reproductive Biology, 78(2), 133 – 140.
MacDonald, D., Grant, A., Sheridan-Pereira, M., Boylan, P., Chalmers, I. (1985). The Dublin randomized controlled trial of intrapartum fetal heart rate monitoring. American Journal of Obstetrics and Gynecology, 152(5), 524-539. https://dx.doi.org/10.1016/0002-9378(85)90619-2
Shalev, E., Eran, A., Harpaz-kerpel, S., Zuckerman, H. (1985). Psychogenic Stress in Women During Fetal Monitoring (Hormonal Profile). Acta Obstetricia et Gynecologica Scandinavica, 64(5), 417 – 420. https://dx.doi.org/10.3109/00016348509155159
Thanks Kirsten, this is a fascinating potential explanation as is so often the case in ‘screening that doesn’t work’. People just cant get their heads around the idea that there are always unintended harms of our well-intentioned interventions, and often those harms overwhelm what was only ever purported benefit. Once you can explain a mechanism for the harm, it becomes psychologically easier to give up a false belief (for example, maybe sticking needles into pre-cancers like DIC f & then irradiating the overdiagnosed breast cancers might be killing as many women as we lie to ourselves we ‘save’ – see https://www.hardingcenter.de/en/transfer-and-impact/fact-boxes/early-detection-of-cancer/early-detection-of-breast-cancer-by-mammography-screening)
It is fascinating isn’t it. I’m just not sure how to get it back into people’s awareness, given that it has been hiding in plain sight for almost 50 years yet remains ignored. You and I might need to come up with a visual representation of the data, like the Harding Centre one for mammography.
As a risk analyst and new doula, I am comforted by the accidental finding that a continuous support person (in the nurse) had a positive effect on foetal outcomes. Shame the sample size wasn’t larger but I thought this study was amazing and as you say very much hiding in plain sight!
I think even though an environment that provides ‘psychological safety’ doesn’t seem like a risk control seeing as it’s so individual and variable (thus can’t be easily controlled) it should be studied as affecting the baseline risk of common poor outcomes that result in cesarean.
CTG monitoring is a great example of applying a control over the majority in order to ‘catch’ and marginally improve risk outcomes in a minority, with the side effect of also increasing the baseline risk in the majority without the benefit for that larger group.
I’d love to see a study where traditional CTG monitoring was compared similarly with a trace monitor that was not visible in the room and where that group received true doula care rather than IA as monitoring even in a positive sense in such frequency is still such a distraction to the normal physiological process.
New to your blog but thoroughly enjoying it thus far!
Thank you – glad you are finding it useful