Refining the use of electronic fetal monitoring

Electronic fetal monitoring (EFM) is the most commonly used instrument in obstetrics and is the perceived standard of care. However, the U.S. Preventive Services Task Force recommended against its use in low-risk women in 1996 (a “D” rating) – signifying the lack of evidence for benefit and the potential for harm – and said there was insufficient evidence to recommend for or against its use in high-risk women (a “C” rating).

Today, available data still suggest that EFM does not reduce overall perinatal mortality or the risk of cerebral palsy. Moreover, its use is associated with increased operative vaginal deliveries and cesarean deliveries.

Given the near-zero positive predictive value of EFM for stillbirth or cerebral palsy, some have called EFM “useless” and a “failure.” However, I see potential in the technology. I believe that we are beginning to see evidence emerge that – if confirmed and expanded – will enable us to quantify and interpret indeterminate EFM patterns in new ways that positively impact clinical outcomes.

Despite EFM’s routine use and our specialty’s well-ingrained clinical habits, we should critically and meaningfully examine new science and new data on category II fetal heart rate tracings as they come to light. In the meantime, there is more we can do to resolve concerning elements of these tracings – without using supplemental oxygen – or to provide reassurance of fetal well-being so that cesarean deliveries are not unnecessarily performed.

Emerging research

An abnormal or indeterminate fetal heart rate tracing is the second most common indication for primary cesarean, after labor arrest, according to a study published in 2011 of more than 32,000 live births. Given the rarity of category III tracings (“abnormal”), it is likely that category II tracings (“indeterminate”) account for most of the cesarean deliveries performed out of concern for fetal acidemia (Obstet Gynecol. 2011 Jul;118[1]:29-38).

Until recently, we’ve known very little about the patterns contained within category II of our current three-tier system for categorizing fetal heart rate patterns. The system was defined by the 2008 consensus workshop sponsored by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), the American College of Obstetricians and Gynecologists, and the Society for Maternal-Fetal Medicine (Obstet Gynecol. 2008 Sep;112[3]:661-6).

We have reasonable data to know that the vast majority of patients with category I fetal heart rate tracings will have a normal pH. We also have reasonable data showing us that patients with category III tracings have a high risk of acidemia and morbidity. However, the majority of tracings we see during labor at term fall into category II, with no clear indication of risk and characterized most often by the presence of decelerations.

As we’ve delved more deeply into the highly variable and complex category II tracings defined in 2008, we’ve begun to demonstrate that tracings can have different meanings for different patients, and that particular clinical factors can make EFM patterns more informative and predictive. In other words, EFM patterns may require different interpretations based on a priori risk and clinical factors.

One of these factors may be the presence of meconium. In a prospective cohort of more than 3,000 women with category II tracings, the presence of meconium – especially thick meconium – was associated with a higher risk of acidemia and neonatal morbidity than when meconium was absent. Interestingly, the negative predictive value was higher than the overall predictive ability in this cohort, which suggests that the absence of meconium in the setting of a category II tracing can be considered a reassuring feature (Am J Obstet Gynecol. 2014;211:644.e1-8).

We have also found through retrospective cohort studies that magnesium sulfate can impact fetal heart rate tracings, causing a transient decrease in variability (Obstet Gynecol. 2012 Jun;119[6]:1129-36 and Am J Perinatol. 2014 Nov;31[10]:869-74). In addition, intrauterine growth restricted fetuses have a higher risk of decelerations without a commensurate higher risk of morbidity (Am J Perinatol. 2015 Jul;32[9]:873-8).

Such findings need to be reproduced, expanded, and further analyzed to show us how the risk of acidemia can be better predicted. For now, just as we increasingly appreciate that tracings have a transient nature and should be considered with two lenses – one looking back in time and one looking forward – we have a growing sense that EFM should not be interpreted without consideration of clinical factors.

Research at our institution and others has shown that acidemia is more significantly associated with non-NICHD measures of fetal heart rate deceleration than with each of the main deceleration types defined by the 2008 NICHD system (i.e., repetitive variable, repetitive late, and repetitive prolonged).

For instance, Emily Hamilton, MD, and her colleagues at PeriGen, a perinatal software company, found that only prolonged decelerations, in addition to the variability within the deceleration and a depth below 60 beats per minute for more than 60 seconds, could discriminate between cases of metabolic acidosis and those with normal umbilical artery gases (J Matern Fetal Neonatal Med. 2012 Jun;25[6]:648-53).

In a 4-year retrospective cohort study of nearly 5,340 consecutive singleton, term, nonanomalous gestations, we found acidemia was most significantly associated with a calculation of the “total deceleration area” – the sum of the estimates of area within all the decelerations. This measure accounted for the frequency, depth, and duration of all decelerations in the final 30 minutes of EFM.

Each of the NICHD deceleration types was associated in our study with acidemia after adjustment for fever, obesity, prolonged first stage, and nulliparity. However, total deceleration area had superior predictive ability. After the same adjustments were made, an abnormal total deceleration area (greater than the 95th percentile) was significantly associated with an increased risk for acidemia (odds ratio, 3.79) (Am J Obstet Gynecol. 2012 Sep;207[3]:206.e1-8).

Pathophysiologically, it seems logical that the total area is most predictive, as it captures both the temporal and dose effect of decelerations. At this point, however, we can only apply this concept crudely at the bedside. There is more work to do to translate such findings into software-driven bedside tools.