In 2017, Cochrane researchers combined the results of 12 randomized, controlled trials including more than 37,000 participants (Alfirevic et al. 2017). Most of the studies were of poor quality and took place in the 1970s and 1980s. Both technology and clinical practice have changed over the years and we do not know what effect this would have on the results if these studies were re-done today. In all of these studies, people were randomly assigned to receive either continuous EFM or hands-on listening during labor.
The researchers found no differences between the continuous EFM group and the hands-on listening group in Apgar scores or cord blood gases, rates of low-oxygen brain damage, admission to the neonatal intensive care unit, or perinatal death. They also found no difference between groups in the percentage of people using medication for pain relief during labor. The overall findings were consistent for people with both low-risk and high-risk pregnancies. When the researchers removed the poorer quality trials from the analysis it did not change the overall findings.
Importantly, there was a 50% lower risk of newborn seizures in the continuous electronic fetal monitoring group; however, overall, seizure events were rare (about one in 500 births, or 0.2%). The risk of a newborn seizure was 0.15% for people with continuous EFM vs. 0.3% for people with hands-on listening during labor. Unfortunately, there have not been any long-term follow-up studies looking at the harms of newborn seizures, so we don’t know how important preventing a seizure is to the long-term health of a child (Alfirevic et al. 2017).
One concerning finding was that people in the continuous electronic fetal monitoring group were 63% more likely to have a Cesarean and 15% more likely to experience the use of vacuum or forceps when compared to those in the hands-on listening group. The overall Cesarean rate varied widely between the different studies—it ranged from a low of 2.3% to a high of 35%. However, two-thirds of the data on Cesarean rates in the meta-analysis came from the study with an extremely low Cesarean rate of 2.3%. In the meta-analysis, the absolute risk of Cesarean was 3.6% for people receiving hands-on listening and 5.9% for those receiving continuous EFM. Of course, the overall Cesarean rate is much higher in most settings today.
The Cochrane researchers estimated that if the Cesarean rate were increased to a more modern rate of 24% among people receiving continuous EFM, then the Cesarean rate would be 15% among those receiving hands-on listening. This means that there would be one additional Cesarean for every 11 women monitored by continuous EFM. In addition, as many as 667 women would have to be monitored with continuous EFM to prevent one newborn seizure. This amounts to an estimated 61 unnecessary Cesareans from continuous EFM in order to prevent one newborn seizure event. As you can see, the risk-benefit debate focuses on preventing Cesareans vs. preventing rare newborn seizures, and not everyone agrees on which is more important to prevent.
It’s possible that the effect of continuous EFM on Cesareans may be even greater in real-world hospital settings today. In an earlier version of the Cochrane review, the researchers grouped the studies into those with a Cesarean rate of less than 10% and those with a Cesarean rate of greater than 10% (Alfirevic et al. 2006). They found that in hospitals where there are higher Cesarean rates, continuous EFM leads to an even higher risk of Cesarean.
The Cochrane review did not find a difference in the rate of cerebral palsy between the continuous EFM group and the group who received hands-on listening. Other researchers have found that continuous electronic fetal monitoring is a very poor test for detecting potential cerebral palsy. False positive rates for predicting cerebral palsy are as high as 99.8%, even in the presence of “dangerous” signs such as multiple late decelerations (slowing of the heart rate after contractions) or decreased variability between heart beats (Nelson et al. 1996). In their 2009 practice bulletin, ACOG concludes that there is Level A evidence (based on good and consistent scientific evidence) that the false-positive rate of EFM for predicting cerebral palsy is greater than 99%. Put another way, most positive test results will be wrong. The false positive rate is so high that for every 1,000 fetuses with an abnormal heart rate pattern that indicates cerebral palsy is at risk of occurring, only one or two will go on to develop cerebral palsy (ACOG 2009).
In a commentary in the Journal of Obstetrics & Gynecology, Grimes and Peipert (2010) make the case that that EFM was adopted as a public health screening program without meeting important criteria for what make a good screening test. In order to be considered a “good” screening test, the test must be able to tell the difference between people with a real problem and those without. A poor screening test for a rare outcome could be considered unethical, since it can cause healthy people to think they are sick, lead to unnecessary medical tests and procedures with harmful side-effects, and waste money and other resources. This is some researchers’ concern with the regular use of continuous EFM to screen laboring people for fetal complications.
The rate of cerebral palsy has stayed the same over time, despite the widespread adoption of using EFM during labor. About one out of 500 children have been diagnosed with cerebral palsy since 1985 (Van Naarden Braun et al. 2016). ACOG suggests that the main reason that the rate of cerebral palsy hasn’t changed is because 70% of cerebral palsy cases occur before labor begins. A review published in 2013 looked at 23 studies to determine how often birth asphyxia, or a lack of oxygen during birth, is linked to cerebral palsy (Ellenberg and Nelson 2013). They found that only a minority of cerebral palsy cases are linked to birth asphyxia. In other words, most cerebral palsy cases are due to prenatal factors before labor begins, and cannot be prevented by EFM.
Some researchers think that another basic assumption of EFM may also be faulty (Lear et al. 2016). It’s possible, they say, that the majority of decelerations, or rapid falls in fetal heart rate, are normal and not dangerous. A healthy fetus may be able to adapt to brief but repeated periods of low oxygen during contractions by triggering something called the peripheral chemoreflex. This theory would help to explain why many babies are born healthy despite repeated brief decelerations during labor. If this theory is correct, it means that what qualifies as normal fetal heart rate patterns during labor is broader than previously thought.
Rates of intrapartum death (stillbirth) were already falling when continuous electronic monitoring was introduced in the 1970s (Hornbuckle et al. 2000). This makes it difficult to interpret the evidence from observational studies. A review by Hornbuckle et al. (2000) found that 12 out of 13 observational studies showed a decrease in stillbirths during labor, after the introduction of continuous EFM. In addition, nine out of nine observational studies comparing labors monitored with continuous EFM vs. those monitored with hands-on listening showed lower rates of stillbirths during labor with continuous EFM. On average, the stillbirth rate in low-risk monitoring groups was lower by about 0.5 stillbirths per 1,000 births, or 1 stillbirth per 2,000 births.
Studies of this type provide lower quality evidence than randomized trials, because the studies could be showing a decrease in stillbirths during labor over time (called a secular trend), not caused by the introduction of continuous EFM. There may also be publication bias, where researchers are more likely to publish studies which show falling death rates. On the other hand, there could be a true relationship between continuous EFM and lower stillbirth rates.
As we discussed earlier, the meta-analysis of randomized trials shows that continuous EFM does not have an effect on stillbirth or newborn death (Alfirevic et al. 2017). The limitation with randomized trials, however, is that a rare outcome like stillbirth requires a very large sample size to detect a difference between groups. The Cochrane reviewers estimate that more than 50,000 women would have to be randomly assigned to continuous EFM or hands-on listening in order to detect a difference in one death out of 1,000 births. Since the Cochrane analysis only included around 37,000 participants, there is a chance that continuous EFM has an effect on stillbirth that was not detected.
If continuous EFM leads to a decrease in stillbirths during labor, it does not necessarily mean that continuous EFM should be used all the time for all laboring people. Any decrease in the risk of stillbirth during labor would be very small, especially among low-risk births, while the known increase in Cesarean rates with continuous EFM is very large (Hornbuckle et al. 2000). It is for this reason that practice guidelines around the world discourage the routine use of continuous EFM during labor (Society of Obstetrics and Gynecologists of Canada; National Institute for Health and Care Excellence in the U.K.; American College of Nurse Midwives and the California Maternal Quality Care Collaborative in the U.S.).
People have asked us if there is any evidence that “putting someone on the monitor” for a set time (e.g., 20 minutes of every hour) is any better than continuous EFM. There have only been two randomized trials on this topic:
In one study, researchers in Sweden randomly assigned more than 4,000 low-risk participants to receive either continuous EFM or intermittent EFM (Herbst and Ingemarsson 1994). They defined intermittent EFM as being on the monitor for 10 to 30 minutes every two to two-and-a-half hours during the active first stage of labor plus the use of hands-on listening every 15-30 minutes in between EFM periods. So, in other words, the intermittent EFM group also had hands-on listening. In the second stage of labor, all of the participants were monitored continuously with EFM. The researchers found no differences in any outcomes.
There has only been one randomized, controlled trial that compared intermittent EFM alone with hands-on listening alone (Mahomed et al. 1994). In this study, 1,255 low-risk participants giving birth at a hospital in Zimbabwe were randomly assigned to either intermittent EFM or one of three different methods of hands-on listening—Doppler ultrasound, Pinard fetal stethoscope used by a research midwife, or Pinard fetal stethoscope used by the attending midwife (as was routine in that hospital). Intermittent EFM was defined as wearing the sensors for the last 10 continuous minutes of every 30 minutes if the results were normal, or the last 10 continuous minutes of every 20 minutes if the results were abnormal.
They found that intermittent EFM detected more abnormal fetal heart rates (54%) than any of the methods of hands-on listening. As a result, more people who used intermittent EFM received Cesareans (28%) than in any other group. Compared to the two Pinard fetal stethoscope groups, the Doppler ultrasound detected more abnormal fetal heart rates (32%) and led to more Cesareans (24%). However, the Doppler ultrasound group had the best newborn health outcomes overall. The research midwives in the study used Huntleigh pocket Doppler ultrasound monitors to listen to the fetal heart rate during the last 10 minutes of every half hour, especially before and immediately after a contraction.
The authors concluded that the use of a handheld Doppler device is a more reliable test for abnormal fetal heart rates than intermittent EFM or the use of a Pinard fetal stethoscope. They also note that handheld Dopplers are simple, affordable, and probably cause less discomfort than Pinard fetal stethoscopes. In contrast, it appears that intermittent EFM alone (when not combined with other monitoring methods) is not based on research evidence. So, some researchers have concluded that it should not be recommended (Martis et al. 2017).
There is very little research on wireless or mobile continuous electronic fetal monitors. Two small pilot studies (in Uganda and the U.S.) have looked at full-term pregnant women to see how well mobile monitors function and how acceptable they are to women and hospital staff (Mugyenyi et al. 2017); (Boatin et al. 2015). Birthing people in these studies reported that they like the mobility that they had with the wireless monitors. Monitoring was successfully recorded and stored in 92% and 88% of the pregnant participants in the Mugyenyi et al. (2017) and Boatin et al. (2015) studies. Both studies experienced some data loss and delays from wireless connection problems.
As we mentioned, there are a variety of devices that can be used for hands-on listening during labor. Cochrane researchers conducted a review and meta-analysis to find out which types of listening tools and timing protocols are most effective (Martis et al. 2017). They were only able to find two randomized, controlled trials to contribute data to the meta-analysis. The studies were conducted in Zimbabwe and Uganda and included a total of 3,242 participants.
When the two studies were combined, they found that a handheld Doppler (battery and wind-up) is linked to more Cesareans for abnormal fetal heart rate compared to a Pinard fetal stethoscope, but without a clear difference in newborn health outcomes (low Apgar scores, newborn seizures, or perinatal death). However, the quality of the evidence is low and other important newborn health outcomes were not assessed. There is not enough evidence at this time to recommend a Doppler ultrasound or a type of fetal stethoscope as the preferred listening device.
Some women are told they can labor “off the monitor” with intermittent EFM and/or hands-on listening as long as they have EFM as a baseline measure for a short period, usually about 20 minutes, upon admission to the hospital. Researchers have looked into the evidence for this practice. They found four studies (from the U.K. and Ireland) to include in a review and meta-analysis (Devane et al. 2017). Altogether, the studies included more than 13,000 low-risk participants.
The researchers found a tendency towards more Cesareans among the people randomly assigned to EFM on admission compared to those assigned to hands-on listening on admission, but the finding was not statistically significant. This means that more data is needed before we can detect if there is a real impact of admission EFM on Cesareans. People assigned to EFM on admission were more likely to end up being put on continuous EFM for the rest of their labor. There were no differences in newborn health outcomes between the groups, including newborn seizures. The authors concluded that there is no evidence of benefit for using EFM on admission in labor among low-risk women, and that hands-on listening is the preferred method.
For more information Fetal Monitoring Belts, OEM Maternity Belts, Pelvic Support Band Pregnancy, please get in touch with us!