We would like to congratulate the National Institute of Child Health and Human Development, Society for Maternal-Fetal Medicine, and the American College of Obstetricians and Gynecologists (ACOG) for their recent efforts in addressing the nomenclature and role of intrapartum electronic fetal heart rate monitoring (EFM) in current obstetrical practice. Certainly there are many areas that need further research including management recommendations for the new 3-category classification of EFM patterns and improved interobserver and intraobserver agreement in EFM pattern interpretation.

ACOG Practice Bulletin 106 suggests that EFM has no value because it does not decrease the incidence of cerebral palsy (CP) and that it causes excessive interventions by cesarean section and instrumented vaginal delivery. Yet their final recommendation is to electronically monitor all high-risk labors and to use amnioinfusion for the relief of variable decelerations. We would like to examine the basis for the nihilistic representation of the cost/benefit ratio of EFM, which, if accepted, might more logically lead to a recommendation against the use of the technique.

First, we should examine the accuracy of intrapartum EFM with respect to the purpose of the technique. When epidemiologists say that the EFM technique has a 99% false-positive rate, such a calculation is based upon the assumption that the technique being evaluated is intended to predict a specific outcome, ie, CP. Clinicians use EFM to detect significant hypoxia and allow intervention to prevent fetal injury, including central nervous system damage, rather than to accurately predict the specific adverse outcome of CP. As such, the intervention strives to prevent damage or complications associated with significant hypoxia and thus if effective would remove the opportunity for any correlation with the outcome (CP). An electrocardiogram has a very high false-positive rate if one suggests that the purpose of the technique is to identify those individuals who will die from cardiac disease. It is fully understood by obstetrical providers that the vast majority of abnormal EFM tracings are not associated with critical hypoxia/acidosis resulting in measurable fetal injury. The role of clinical judgment, treatment, and timely intervention must be considered to play a part in this lack of correlation rather than simply claiming a 99% false-positive rate with abnormal EFM. In fact, if EFM were perfect there would be a 100% false-positive rate because all CP resulting from acute intrapartum hypoxia would be prevented and no CP would be predicted by the technique. Clearly, damaging acute intrapartum hypoxia in a previously normal fetus will always be detected by EFM but CP caused by nonhypoxic etiologies and hypoxic etiologies not occurring during intrapartum observation are not remedied by intervention.

The vast majority of diagnostic tools, especially monitoring devices, have never been subjected to randomized trials where the endpoint was improved outcome. EFM is a diagnostic tool that was never proposed as a predictor of outcomes. One can question whether this should even be done. Imagine how difficult it would be to prove that the use of a thermometer would improve outcome with any infection.

While it is clear that the risk of CP has not declined in association with the introduction of EFM, there has been an increased survival of neonates at all gestational ages and the major portion of CP is attributable to prematurity. So if overall neonatal survival has increased it is reasonable to opine that infants who once died now may survive with damage and that some of those who were damaged now survive intact. Thus we think the statement on page 3 that says EFM did not reduce the risk of CP needs to consider the moving window from death to damage and from damage to intact survival.

Estimates by epidemiological scientists of the low percent of CP due to pure acute intrapartum hypoxia is misleading. One case-controlled study concluded that only 5% of neonatal encephalopathy was due solely to intrapartum hypoxia because if there were any antepartum factors such as diabetes, preeclampsia, or intrauterine growth retardation, it was assumed that the damaging process began before labor and thus disqualified such patients as having only intrapartum hypoxia even if the EFM pattern on admission to labor was category I. Such analysis certainly risks understating the true frequency of damaging intrapartum asphyxia, which could be reduced with appropriate interpretation and management of abnormal EFM.

While there are no randomized trials comparing EFM to no monitoring, it is clear that if one uses historical data from institutions that introduced EFM where auscultation had not been intensive (as in the randomized trials), there was a >3-fold reduction in term stillbirth rates when EFM was introduced. Intrapartum stillbirth rates prior to the introduction of EFM were consistently in the range of 3/1000, and currently with EFM or intensive auscultation intrapartum stillbirths are indeed a rare event. Thus we take issue with the statement on page 3 stating that EFM did not reduce perinatal mortality, which should have read that EFM and intensive auscultation have both been shown to decrease the rate of stillbirth when compared to nonintensive auscultation.

It is certainly not our motive to criticize the excellent ACOG Practice Bulletin 106 but rather to provide some evidence to support their recommendations with respect to the use of intrapartum EFM. We really do not believe that the technique of EFM would continue to be so widely accepted if practicing clinicians were of the mind that it has no value.