In the Journals

Screening algorithm accurately predicts OSA in pregnant women

A new algorithm may help screen pregnant women, particularly African American women, for obstructive sleep apnea, according to a recent study.

“Predicting OSA risk in the rst trimester may allow health care providers to prevent the development of OSA or decrease its impact on associated adverse outcomes,” the researchers wrote. “These findings indicate that, when the BMI, age, tongue enlargement (BATE) algorithm screens an individual as OSA negative, the likelihood of having OSA is very low, which is a desirable characteristic in a screening tool.”

Researchers conducted a secondary analysis of 121 pregnant women during their first trimester and 87 women during their third trimester, all of whom were 14 weeks’ gestation or less. Of the first trimester participants, 91 were African American, 24 were white and six were of other races or ethnicities. Of third trimester participants, 63 were African American, 21 were white and three were of other races or ethnicities. Exclusion criteria included inability to communicate, behavioral or cognitive impairments that had the potential to interfere with informed consent, lack of a telephone, self-reported alcoholism or illicit drug use, serious preexisting chronic medical conditions, use of sedative or hypnotic medication at least three times a week and current treatment for sleep-disordered breathing.

Participants submitted the Multivariable Apnea Prediction Questionnaire and underwent full polysomnography, with apnea defined as complete cessation of airflow for 10 seconds or longer and hypopnea defined as a decrease in airflow greater than 50% or a lesser airflow reduction with more than 3% oxygen arousal or desaturation.

OSA prevalence was 10.7% and 24.1% during the first and third trimesters, respectively.

The researchers used logistic regression analysis and area under the curve to identify characteristics that could be added to a model to help predict risk for OSA. The final screening algorithm predicted OSA risk by examining participant age, BMI and presence of tongue enlargement.

The AUC for the algorithm was 0.86 (95% CI, 0.76-0.96) during the first trimester and 0.87 (95% CI, 0.77-0.96) during the third trimester. Additionally, the AUC was 0.87 (95% CI, 0.77-0.97) when first trimester data were used to predict OSA risk in the third trimester. The algorithm also had high sensitivity and specificity when used during early and late pregnancy.

Because the study population was predominantly African American, these findings may be particularly applicable to this patient population, according to the researchers.

“Our model can be easily and quickly administered in busy clinical practices without depending on patients’ awareness of experiencing apnea symptoms,” they wrote. “Future studies are needed to further validate the scoring algorithms.” – by Eamon Dreisbach

Disclosure: One of the study authors reports receiving a research grant from BluTech Inc. to study the role of blue-blocking lenses in the secretion of melatonin in healthy adults.

A new algorithm may help screen pregnant women, particularly African American women, for obstructive sleep apnea, according to a recent study.

“Predicting OSA risk in the rst trimester may allow health care providers to prevent the development of OSA or decrease its impact on associated adverse outcomes,” the researchers wrote. “These findings indicate that, when the BMI, age, tongue enlargement (BATE) algorithm screens an individual as OSA negative, the likelihood of having OSA is very low, which is a desirable characteristic in a screening tool.”

Researchers conducted a secondary analysis of 121 pregnant women during their first trimester and 87 women during their third trimester, all of whom were 14 weeks’ gestation or less. Of the first trimester participants, 91 were African American, 24 were white and six were of other races or ethnicities. Of third trimester participants, 63 were African American, 21 were white and three were of other races or ethnicities. Exclusion criteria included inability to communicate, behavioral or cognitive impairments that had the potential to interfere with informed consent, lack of a telephone, self-reported alcoholism or illicit drug use, serious preexisting chronic medical conditions, use of sedative or hypnotic medication at least three times a week and current treatment for sleep-disordered breathing.

Participants submitted the Multivariable Apnea Prediction Questionnaire and underwent full polysomnography, with apnea defined as complete cessation of airflow for 10 seconds or longer and hypopnea defined as a decrease in airflow greater than 50% or a lesser airflow reduction with more than 3% oxygen arousal or desaturation.

OSA prevalence was 10.7% and 24.1% during the first and third trimesters, respectively.

The researchers used logistic regression analysis and area under the curve to identify characteristics that could be added to a model to help predict risk for OSA. The final screening algorithm predicted OSA risk by examining participant age, BMI and presence of tongue enlargement.

The AUC for the algorithm was 0.86 (95% CI, 0.76-0.96) during the first trimester and 0.87 (95% CI, 0.77-0.96) during the third trimester. Additionally, the AUC was 0.87 (95% CI, 0.77-0.97) when first trimester data were used to predict OSA risk in the third trimester. The algorithm also had high sensitivity and specificity when used during early and late pregnancy.

Because the study population was predominantly African American, these findings may be particularly applicable to this patient population, according to the researchers.

PAGE BREAK

“Our model can be easily and quickly administered in busy clinical practices without depending on patients’ awareness of experiencing apnea symptoms,” they wrote. “Future studies are needed to further validate the scoring algorithms.” – by Eamon Dreisbach

Disclosure: One of the study authors reports receiving a research grant from BluTech Inc. to study the role of blue-blocking lenses in the secretion of melatonin in healthy adults.