A fetal heart monitor that provides an analysis of the fetal ECG during labor has won Food and Drug Administration approval almost 4 years after it was originally rejected for use in the United States.
The American College of Obstetricians and Gynecologists is standing firm, however, in its refusal to endorse the STAN S31 fetal heart monitoring system “until it's proved efficacious” in everyday clinical use.
The device, which is used in 22 countries, is now labeled for use in the United States as an adjunct to standard electronic fetal monitoring (EFM) for determining “whether obstetric intervention is warranted when there is increased risk of developing metabolic acidosis.” Physicians who use the device, the labeling says, must be certified and credentialed in its use.
The new system monitors the fetal ECG and heart rate via a scalp electrode, automatically identifying and analyzing T-wave and ST-segment changes, which reflect myocardial ability to respond to hypoxia. When the STAN system was first reviewed in 2002, members of the FDA's Obstetrics and Gynecology Devices Panel agreed that findings from a randomized Swedish trial of almost 5,000 women in labor demonstrated its safety and effectiveness.
Among women who were monitored with both the STAN monitor and conventional monitoring, as opposed to conventional monitoring alone, metabolic acidosis was reduced by 54%, and operative deliveries for nonreassuring fetal heart rate were reduced by 19%. Moderate and severe neonatal encephalopathy were also significantly reduced.
The panel recommended nonapproval, however, citing concern about differences between Sweden and the United States in labor management and medical terminology. The FDA asked the manufacturer to conduct bridging studies to show that U.S. clinicians could learn the STAN system.
Last June, satisfied with the results of the two bridging studies, the panel unanimously recommended approval. Panelists voiced hope that the device could decrease its false-positive rate and thus reduce the rate of unnecessary cesarean sections. But at the same time they expressed concern that it could do the opposite and increase the cesarean section rate.
Dr. Gary D.V. Hankins, who chairs ACOG's Committee on Obstetric Practice, said he and others on the committee share this concern. For ACOG to endorse use of the device, it has to know “if it does indeed prevent injury without inordinately increasing operative delivery” once it is used in U.S. institutions. “Other technologies that were going to 'get rid of the neurologically impaired baby' haven't delivered on their promises,” said Dr. Hankins of the University of Texas, Galveston. “It's a responsible position to be conservative and not endorse any technology until it's proved efficacious—that is, it provides something that's worth the cost.”
Dr. Julia Carey-Corrado, the FDA's clinical reviewer of the device application, said the FDA is requiring the manufacturer to submit annual reports that address a specified list of adverse events, including rates of perinatal death, neonatal encephalopathy, acidemia, and acidosis; the rate of device malfunction; the number of monitors sold; the number of institutions using the monitors; the proportion of patients monitored with STAN as opposed to standard EFM; and the number of physicians credentialed.
“The reports [will provide for] a more systematic, intense review” than normally occurs through the FDA's standard adverse event reporting requirements, Dr. Carey-Corrado said.
“The approval is unique in that we have been very explicit [in our conditions],” she said. “And having a denominator [on the extent of the device's use] will allow us to interpret the significance of outcomes.”
The FDA will not, however, require the manufacturer to submit data on operative delivery rates, which is something its advisory panel recommended in June.
According to Colin Pollard, chief of the FDA's ob.gyn. devices branch, the agency decided not to require collection of these data, largely because it felt the issue of operative delivery rates had been addressed in the pivotal study.
In the pivotal Swedish randomized trial, the rate of operative delivery decreased significantly with the use of the STAN system. The rate of cesarean section for fetal distress was not significantly lower when all enrollees were included (the intent-to-treat analysis), but it was significantly lower when only those with adequate recordings were included.
Under the FDA's requirement for training, clinicians must be certified based on a written test and credentialed based on an oral exam that is administered after successful completion of at least five “practice cases,” according to Dr. Carey-Corrado.
Physician certification is something the FDA's advisory panel called for in June, and the FDA deliberately structured its training requirement to resemble the training that was required of clinicians in the U.S. bridging studies, she said.