The Berlin questionnaire, a common tool for identifying obstructive sleep apnea, does not accurately identify pregnant women whose breathing is intermittently interrupted or stopped (a condition called apnea) during sleep, said researchers from Baylor College of Medicine in a report that appears online in the American Journal of Obstetrics and Gynecology.
"We wanted to know if the questionnaire was a valid tool for screening for obstructive sleep apnea in pregnancy compared to the gold standard, which is polysomnography (a monitor that uses sensors to measure various parameters such as brain electrical activity, movements of eyes, jaw and leg muscles, airflow, breathing effort, heart rhythm and oxygen levels in the blood)," said Dr. Kjersti Aagaard, assistant professor of obstetrics and gynecology at BCM and senior author of the report. "Other investigators had reported that a diagnosis of this kind of apnea was associated with infants born small for their gestational age and mothers at risk for a complication of pregnancy called preeclampsia.
Validating gold standard
"However, because these studies relied on the Berlin questionnaire to diagnose obstructive sleep apnea, we felt it was important to first test the Berlin questionnaire for diagnostic accuracy in pregnant women. Because some of the questions on Berlin involved common complaints in pregnancy—snoring, sleepiness, and obesity—it was possible that the observed complication rate may not have been directly attributable to the sleep apnea per se. It was important to first validate the Berlin questionnaire in pregnant subjects."
In the study, she and her colleagues enrolled 100 women in the last one-third of pregnancy. The women took the Berlin questionnaire, and then underwent continual sleep apnea and fetal heart rate monitoring for at least three hours.
Compared to the results of polysomnography, the Berlin questionnaire was not an accurate tool to diagnose obstructive sleep apnea. It had a sensitivity of 35 percent, which means that it identified just a bit more than one-third of those who actually had sleep apnea. Its specificity was 63.8 percent – which means that it correctly identified just under two-thirds of those who did not have apnea.
"Clearly, this was disappointing as it raises questions regarding data linking obstructive sleep anea [as diagnosed by the Berlin questionnaire] to adverse pregnancy outcomes," said Aagaard. "We were reassured by our findings that regardless of the accuracy of the Berlin questionnaire in diagnosing obstructive sleep apnea in pregnancy, there was no concerning changes in the fetal heart rate pattern in women who did experience intermittent apnea or hypoxia during sleep."
Of interest, the researchers did find that snoring did seem to correlate with apnea episodes in these women.
"Our data supports further investigations into the use of a modified Berlin questionnaire which emphasizes snoring symptoms," they wrote, noting that questionnaire must be validated in a larger group of pregnant women. They also advised more research to identify how sleep apnea might lead to problems in pregnancy and resulting offspring.
Others from BCM who took part in this work include Drs. Sofia A. Olivarez, Bani Maheshwari, Meghan McCarthy, Nikolaos Zacharias, Ignatia van den Veyver and Haleh Sangi-Haghpeykar and Lata Casturi, a sleep technologist.
Funding for this work came from the National Institutes of Health New Innovator Award.
An abstract of the article is available.