INTRODUCTION — Sleep-disordered breathing is a broad term that represents the continuum from snoring at the mild end of the spectrum to overt obstructive sleep apnea (OSA) at the severe end of the spectrum. The prevalence of obesity in reproductive-aged women has increased in recent years. As a result, the prevalence of obesity-related comorbid conditions complicating pregnancy is also increasing. One of these conditions is OSA, which refers to apneas and hypopneas (absent or severely reduced airflow) during sleep despite respiratory effort. OSA is characterized by repetitive episodes of upper airway obstruction during sleep. The obstruction results in a reduction of airflow, hypoxemia, sympathetic discharge, and recurrent arousals from sleep. Central sleep apnea, which will not be reviewed here, is a different disorder that is defined by episodes of cessation of airflow due to absent breathing effort. (See "Polysomnography in the evaluation of sleep-disordered breathing in adults".)

While the diagnosis and management of sleep-disordered breathing and OSA are similar in pregnant and nonpregnant women, some aspects are unique to pregnancy. This topic will focus on OSA in pregnancy. Despite the increase in studies addressing this topic, there remains a scarcity of studies that address treatment and long-term effects of OSA in pregnant women.

Detailed discussions of sleep related breathing disorders (eg, OSA, central sleep apnea) in nonpregnant individuals are available separately.

●(See "Clinical presentation and diagnosis of obstructive sleep apnea in adults".)

●(See "Polysomnography in the evaluation of sleep-disordered breathing in adults".)

●(See "Clinical presentation and diagnosis of obstructive sleep apnea in adults".)

●(See "Management of obstructive sleep apnea in adults".)

PREVALENCE AND RISK FACTORS — The prevalence of OSA in pregnant women is not well defined, and depends on the method of ascertainment.

●In a small prospective cohort study of 105 pregnant women who underwent first- and third-trimester in-laboratory polysomnography, the prevalence of OSA (apnea-hypopnea index [AHI] ≥5 events/hour of sleep) in the first and third trimesters was 10.5 and 26.7 percent, respectively [1]. Of note, the subjects were not representative of the general obstetric population, but were selected to provide a similar number of women in each of four body mass index (BMI) groups (normal <25.0 kg/m², overweight 25.0 to <30.0, class I obesity 30.0 to <35.0, class II to III obesity ≥35.0). Using data on the BMI distribution in their local clinical population, the authors estimated overall OSA prevalence to be 8.4 percent in the first trimester and 19.7 percent in the third trimester. Among the 28 patients with OSA in the third trimester, the disorder was mild (AHI 5 to 14) in 23, moderate (AHI 15 to 29) in 4, and severe (AHI ≥30) in 1. The prevalence of OSA in this study is higher than that reported in the nonpregnant population of reproductive age women, where the prevalence of OSA has been estimated to be 0.7 to 6.5 percent by various diagnostic criteria [2-5].

●The single largest prospective study included a subset of participants in the Nulliparous Pregnancy Outcomes Study Monitoring Mothers-to-be (NuMoM2b) cohort who underwent level 3 home sleep apnea tests in early pregnancy (between 6 and 15 weeks of gestation) and in midpregnancy (between 22 and 31 weeks of gestation) [6]. Among these 3306 women, who were recruited from eight clinical sites across the United States and included substantive proportions of non-Hispanic Black and Hispanic women, the prevalence of gestational OSA was 3.6 percent in early pregnancy and 8.3 percent in midpregnancy.

Increasing age and obesity are independent risk factors for OSA in both pregnant and nonpregnant women [1,7,8].

Physiologic changes of pregnancy predisposing to OSA — The numerous hormonal and physiologic changes occurring in pregnancy may play a role in the prevalence and severity of OSA among pregnant women:

●The oropharyngeal diameter appears to narrow and the Mallampati grade increases as pregnancy progresses [9,10] (see "Approach to the anatomically difficult airway in adults outside the operating room", section on 'M: Mallampati score'). Although the role of airway narrowing in OSA has not been studied in pregnancy, there is a significant link between OSA and the internal size of the airway in the general population.

●Nasal patency is reduced during pregnancy secondary to hyperemia and edema of the nasal mucosa. Changes in the nasal mucosa may be due to increased blood flow and variations in circulating levels of estrogen and progesterone. The increased nasal congestion may lead to increased upper airway resistance and more negative intra-pharyngeal pressure with inspiration, in turn predisposing women in late pregnancy to airway narrowing, and possibly snoring and obstructed breathing in sleep. Pregnant women are more likely to report snoring compared with their nonpregnant counterparts, and the prevalence of frequent snoring (≥3 nights/week) increases during pregnancy from the first to the third trimester [9,11], as estrogen and progesterone levels increase.

●Changes in upper airway patency from other etiologies may also predispose pregnant women to snoring and possibly sleep-disordered breathing. In healthy nonpregnant individuals, studies have shown that fluid displacement from the legs into the neck reduces upper airway size and increases upper airway collapsibility [12,13]. Fluid shifts induced by prolonged sitting and recumbency have been strongly related to the degree of sleep-disordered breathing [14], although other data have found that such rostral fluid shifts do not increase the frequency of obstructed breathing events [15]. Given that maternal blood volume increases, on average, 40 to 45 percent during pregnancy above nonpregnant levels [16], recumbency may adversely affect upper airway function and possibly increase the likelihood of sleep-disordered breathing.

●The increase in progesterone levels during pregnancy leads to increased tidal volume and results in increased minute ventilation [17], either due to direct effects on respiratory drive or enhanced sensitivity of the brain's respiratory center to carbon dioxide [18]. Increased ventilatory drive during pregnancy may enhance loop gain and this may predispose to OSA. (Loop gain is the ratio of a corrective ventilatory response to a disturbance. If the response exceeds the disturbance [ie, loop gain >1], then self-sustaining periodic breathing with central apneas may result.) (See "Treatment-emergent central sleep apnea", section on 'Pathogenesis'.)

CLINICAL MANIFESTATIONS AND DIAGNOSIS — The general clinical manifestations and diagnosis of OSA are discussed in detail separately. (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults".)

A gender disparity in the recognition and treatment of OSA has been recognized. While initial reports of OSA suggested eight or nine cases of sleep apnea in men for each case of sleep apnea in women, epidemiologic studies report a ratio of two to four affected men for each affected woman [19]. Studies consistently show that women are more likely to be misdiagnosed or diagnosed with OSA at later stages than are men. Although women can have the same classic symptoms described in men (witnessed apnea, choking, and restless sleep), women appear to be less likely to self-report these symptoms, leading to misdiagnosis or delayed diagnosis [19]. In addition, women are more likely to report other symptoms as noted in a community-based sample of 551 men and 388 women, in which women were more likely than men to report daytime fatigue (60.8 versus 49.1 percent), morning headaches (12.5 versus 3.3 percent), and symptoms of depression (6.6 versus 2.7 percent), regardless of the severity of OSA [19].

The difficulty in identifying abnormal sleep symptoms in pregnant women in particular is compounded by the accepted belief that poor sleep is to be expected during pregnancy. Sleepiness and increased sleep need are widely reported by pregnant women, especially during the first trimester, and may occur due to increasing progesterone levels. These physiologic changes make it difficult to know what degree of increased sleepiness may be normal in pregnancy. Nevertheless, in one cross-sectional survey of patients and obstetric providers, 32 percent of women reported snoring but only 5 percent were asked about snoring by their provider [20].

Screening — Screening questionnaires, including the STOP-Bang and Berlin questionnaires, which are inexpensive and easy to administer, are poorly predictive in women, and their weaknesses may be even more prominent in pregnancy [21-25]. In the largest cohort of pregnant women who completed Berlin questionnaires and underwent overnight polysomnography, the sensitivity and specificity of the Berlin questionnaire for identifying OSA were only 35 and 63 percent, respectively [26]. (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults", section on 'Screening questionnaires'.)

In our patient population, we do not use screening questionnaires because of their poor performance. In our study of sleep-disordered breathing during pregnancy in nulliparous women, logistic regression modeling found that three variables, age, body mass index, and frequent snoring (self-reported snoring ≥3 days per week), achieved good prediction of prevalent and incident sleep-disordered breathing [27]. Sensitivity ranged from 46 to 61 percent at 90 percent specificity. Another model utilizing body mass index, age, and presence of tongue enlargement had sensitivity of 76 to 79 percent at 82 percent specificity and was particularly accurate in African Americans [28]. These models, when externally validated, can help with screening for sleep-disordered breathing and OSA in the clinical setting, referral for objective testing, and future clinical treatment trials.

Individuals with obesity-related comorbid conditions such as diabetes mellitus and hypertension are also considered high risk. If they report snoring, witnessed apneas and/or drowsy driving, they are referred for evaluation by a sleep medicine specialist.

Objective testing — There are no studies that specifically address the best method for objective OSA diagnosis in pregnant women. The gold standard for OSA diagnosis is the full-night, attended, in-laboratory polysomnogram. While unattended, home sleep apnea testing (HSAT) is a reasonable alternative for patients in whom there is a high likelihood of moderate or severe OSA, the American Academy of Sleep Medicine (AASM) guideline for the use of HSAT appears to exclude pregnant women from that category [29]. This guideline recommended that HSAT be used only in populations with substantive data on sensitivity and specificity; such data are not available for pregnancy [30]. The guideline also excluded individuals with comorbid medical or sleep conditions as candidates for HSAT, which may include pregnancy. Nevertheless, increased risk for adverse outcomes among women with gestational OSA was observed in the nuMoM2b study using HSAT to identify mothers with OSA, suggesting that this methodology may be sufficiently sensitive to identify clinically significant cases of OSA in pregnant women.

In our practice, due to insurance restrictions, we use HSAT to evaluate pregnant women for OSA when polysomnography is not initially approved. If home testing does not demonstrate OSA but clinical suspicion remains high, we generally pursue laboratory polysomnography. Polysomnography is more sensitive than HSAT because sleep disordered breathing events associated with arousals can be detected and because HSAT tends to overestimate total sleep time, thus underestimating apnea-hypopnea index (AHI). All pregnant individuals with suspected sleep abnormalities should be referred to a sleep medicine specialist for evaluation, testing, and discussion of OSA management, if indicated.

Scoring severity — OSA severity is scored as mild, moderate, or severe, corresponding to an AHI (average number of apneas plus hypopneas per hour slept) of 5 to 15, 15 to 30, and >30 events per hour, respectively (see "Clinical presentation and diagnosis of obstructive sleep apnea in adults"). The scoring of individual hypopneic and apneic events is dependent on the scoring criteria used in individual laboratories. The 2017 AASM-recommended criteria for identifying a hypopnea event are ≥30 percent reduction in nasal pressure signal excursion from baseline lasting ≥10 seconds, with an associated ≥3 percent oxygen desaturation from pre-event baseline or an associated arousal [31]. The AASM alternative scoring criteria for a hypopnea event are ≥30 percent reduction in nasal pressure signal excursion or alternative hypopnea sensor lasting ≥10 seconds with an associated ≥4 percent oxygen desaturation. (See "Polysomnography in the evaluation of sleep-disordered breathing in adults", section on 'Hypopneas'.)

There are no studies describing normative data in the obstetric population. Some of the physiologic changes in pregnancy may contribute to the same symptoms as OSA, such as frequent arousals and daytime sleepiness. Pregnancy-related causes of maternal awakening and arousal may be due to fetal movement, urination urge, nocturnal dyspepsia, or leg cramps. All of these symptoms are common in late pregnancy and contribute to sleep fragmentation [2,19]. (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults".)

SIGNIFICANCE OF SNORING AND OSA — Snoring is an acoustic symptom that may be associated with increased upper airway resistance, and/or obstructive apneas and hypopneas in some, but not all, patients.

Overview — Data from studies in the general population indicate an increased risk of adverse clinical outcomes associated with untreated OSA:

●Systemic hypertension, mild pulmonary hypertension, coronary artery disease, cerebrovascular disease, cardiac arrhythmias, and ischemic stroke are associated with OSA. A number of studies have demonstrated that while risk factors for OSA and many of these conditions overlap, OSA persists as an independent risk factor for many adverse cardiometabolic outcomes after adjusting for the effects of comorbid disease.

●Excessive daytime sleepiness, inattention, and fatigue are also associated with OSA. These outcomes may be caused by OSA; however, sleepiness due to OSA cannot be distinguished behaviorally from sleepiness associated with sleep deprivation.

●Perioperative complications due to intubation difficulty or impaired arousal from sedatives. (See "Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea".)

●Diabetes or insulin resistance, although this may be due, in part, to risk factors common to both conditions.

●All-cause mortality, if OSA is moderate or severe. Patients with untreated mild OSA may not be at increased risk.

These conditions have been mostly described in middle aged men; the significance of OSA in reproductive age women is less certain [2]. In the meantime, the significance of OSA to maternal and fetal health during the short course of pregnancy is becoming clearer. (See "Obstructive sleep apnea and cardiovascular disease in adults" and "Clinical presentation and diagnosis of obstructive sleep apnea in adults".)

Effect of snoring on pregnancy and infant outcomes — Among pregnant women, the prevalence of self-reported snoring has been estimated to be between 14 and 46 percent, with snoring prevalence increasing as gestation advances [32,33]. Sleep-disordered breathing has been proposed as a risk factor for adverse maternal-fetal outcomes, including preeclampsia and small-for-gestational age births.

In most individuals, relatively small absolute increases in the number of apneas and hypopneas occur over the course of pregnancy [34,35]. This raises the issue of whether other manifestations of sleep-disordered breathing, such as snoring and flow limitation, are important.

Several observational studies on snoring in pregnant women have reached differing conclusions [32,33,36-39]. Higher rates of gestational hypertension, preeclampsia [32,37], and delivery of small-for-gestational age infants [32,39] have been reported among habitual snorers compared with nonsnorers. One study reported new-onset snoring in pregnancy, but not chronic snoring, was a risk factor for preeclampsia and gestational hypertension [38]. In a prospective cohort study including over 15,700 women, women who reported habitual snoring were more likely than nonsnorers to develop gestational diabetes (odds ratio [OR] 2.50, 95% CI 1.34-4.67) [40]. In contrast, studies focusing primarily on infant outcomes (birth weight, Apgar scores, stillbirth) have failed to find significant differences between infants born to snorers and nonsnorers [33,36,41].

However, there is no consensus about snoring terminology, such as what constitutes simple-primary-benign versus mild-to-moderate snoring [42]. The distinction between different types of snoring may be important in outcome studies.

Effect of OSA on pregnancy

Maternal effects — Studies of sleep apnea in pregnancy tend to include women with any of a number of sleep-related breathing disorders (ie, abnormal respiratory pattern or abnormal oxyhemoglobin desaturation), and thus may include a small number of women with central sleep apnea syndrome and hypoventilation syndromes, as well as OSA. The broader classification was used, in part, because most studies have relied on questionnaires and patient-reported symptoms, without a polysomnogram, so it was not possible to differentiate among the different forms of sleep-related breathing disorders. A significant limitation of these studies is that most did not adjust for potential maternal confounders.

Severe maternal morbidity — A study based on discharge codes from the United States Nationwide Inpatient Sample database reported that among more than 55 million women, OSA was associated with an increased odds of severe morbidity including: eclampsia (OR 5.4, 95% CI 3.3-8.9), cardiomyopathy (OR 9.0, 95% CI 7.5-10.9), pulmonary embolism (OR 4.5, 95% CI 2.3-8.9) and in-hospital mortality (OR 5.3, 95% CI 2.4-11.5) [43]. The findings were confirmed in another, smaller sample of 1.5 million gravida with a discharge diagnosis of OSA [44].

Preeclampsia and gestational hypertension — OSA has been associated with development of preeclampsia or gestational hypertension, as well as eclampsia (see above [43]). A 2013 systematic review of retrospective and small cohort studies reported a twofold increase in preeclampsia among women with sleep-disordered breathing (adjusted OR 2.34, 95% CI 1.60-3.09; five studies) [45]. The presence of sleep-disordered breathing was based on data obtained from questionnaires (three studies) or polysomnography (two studies). Results from a large prospective cohort study including 3306 nulliparous women who underwent home sleep apnea testing in early pregnancy and midpregnancy demonstrated increased risk for preeclampsia among women found to have sleep apnea in either early pregnancy (OR 1.94, 95% CI 1.07-3.51) or midpregnancy (OR 1.95, 95% CI 1.18-3.23) [6]. The largest cohort study of women with polysomnogram-confirmed OSA (n = 791) found that, compared with women without the OSA diagnosis, women with existing OSA diagnosed prior to pregnancy had an increased risk of preeclampsia (OR 1.6, 95% CI 2.16-11.26). The population-based dataset lacked body mass index (BMI) data, resulting in a reported obesity rate of only 1.6 percent; therefore, whether risk conferred by OSA was independent of obesity could not be ascertained [46].

The association between OSA and hypertension has been hypothesized to be a physiologic response to intermittent nocturnal hypoxia, which activates the sympathetic nervous system. The recurrent hypoxia and reoxygenation cycles also result in oxidative stress, endothelial dysfunction, and increased oxidative vascular injury [47]. This is believed to be part of the mechanism underlying the development of preeclampsia, as well [2], providing a plausible common pathway for the development of early preeclampsia among women with OSA.

Data on the clinical course of OSA in the general population also lend support. In two large cohort studies, individuals with untreated OSA were more likely to develop hypertension [48]. This association persisted after controlling for confounding variables such as body mass index, ethnicity, and socioeconomic status. However, no studies have associated either presence or severity of sleep-related hypoxemia with any pregnancy outcome.

The use of nasal continuous positive airway pressure (CPAP), the first-line therapy for OSA, for treating the hypertensive conditions of pregnancy was examined in a small, uncontrolled study. When auto-titrating nasal CPAP (APAP) was administered to 11 women with severe preeclampsia, mean nocturnal blood pressure was reduced [49]. Nevertheless, given the lack of a control group and other study limitations, the results should be interpreted cautiously.

Gestational diabetes — The relationship between OSA and insulin resistance in the general population is well established [50-52]. Independent of other risk factors, individuals with OSA have an increased risk of developing type II diabetes, hyperinsulinemia, and metabolic syndrome. While a causal link has not been proven, improved glucose control has been observed after initiating treatment of OSA with CPAP [53,54].

An increased risk of gestational diabetes mellitus (GDM) in OSA patients is also plausible since increasing insulin resistance occurs in both pregnancy and OSA. Although there is no evidence of the value of early screening in this population, we screen pregnant women with OSA for diabetes early in pregnancy and repeat screening at 24 to 28 weeks in those with an initially negative screen [55]. (See "Gestational diabetes mellitus: Screening, diagnosis, and prevention".)

A 2013 systematic review of observational studies noted that pregnant women with sleep-disordered breathing were at increased risk of developing GDM (adjusted OR 1.86, 95% CI 1.30-2.42; five studies) [56-58]. The presence of sleep-disordered breathing was based on data obtained from questionnaires (four studies) or polysomnography (one study). Conclusions drawn from such data must take into consideration the limited sensitivity and specificity of standard questionnaires to predict OSA in pregnant women [26]. However, prospective data from the large sleep substudy cohort of NuMoM2b confirmed this finding [6]. After controlling for confounding variables, GDM was associated with OSA in early pregnancy (OR 3.47, 95% CI 1.95-6.19) or midpregnancy (OR 2.79, 95% CI 1.63-4.77).

In addition, severity of sleep disordered breathing may be directly associated with higher glucose levels. In one cross-sectional study including 65 pregnant patients with GDM between 24 and 34 weeks of gestation, patients with an apnea-hypopnea index (AHI) ≥30 events/hour of sleep had higher glucose levels during nocturnal and early morning periods compared with those who had an AHI <10 events/hour, even after adjustment for BMI and use of insulin/metformin [59]. Specifically, with every 10-unit increase in AHI, nocturnal and morning glucose levels rose by 0.20 mmol/L (3.6 mg/dL; 95% CI 0.04-0.40) and 0.26 mmol/L (4.7 mg/dL; 95% CI 0.08-0.40), respectively.

Preterm delivery — Population-based studies have described an increased frequency of preterm birth in women with sleep apnea, and the preterm births were primarily medically or obstetrically indicated (iatrogenic) rather than spontaneous [46,60-62]. In a large study of women with OSA (n = 791 women with OSA and 3955 randomly selected women without OSA), OSA doubled the chances of preterm birth (adjusted OR 2.31, 95% CI 1.77-3.01) [46]. A systematic review of sleep-disordered breathing in pregnancy reported that it was associated with a modest increase in preterm birth (OR 1.86, 95% CI 1.50-2.31) [8], but the findings were limited by inclusion of studies with different forms of sleep-related breathing disorders, lack of rigorous diagnostic criteria, and lack of adjustment for potential maternal confounders, particularly maternal age and body mass index.

Fetal effects — Case reports and small cohort studies have attempted to determine whether maternal OSA affects the fetus. To date, existing data, although sparse and observational, suggest that it does not. In a population-based study in which over 1,400,000 maternal records were linked to live newborn records, OSA was associated with an increased risk for congenital anomalies (adjusted OR 1.26, 1.11-1.43), with the highest risk for musculoskeletal anomalies (adjusted OR 1.89, 1.16-3.07) after adjusting for comorbidities and potential teratogens [63]. Newborns of mothers with OSA were also more likely to be admitted to the intensive care unit (25.3 versus 8.1 percent). Reports of adverse fetal effects are likely related to factors such as lack of adjustment for confounders (eg, obesity, preeclampsia), inadequate case ascertainment, and publication bias [64-68].

Fetal heart rate — Although case reports have described an association between OSA and fetal heart decelerations [64,65], the best available data come from a cohort study including 100 pregnant women who underwent simultaneous fetal monitoring and in hospital polysomnography [26]. In this study, none of the apnea episodes were associated with any fetal tracing abnormality.

Birth weight — Although case reports and retrospective studies have described an association between OSA and impaired fetal growth, confounders such as preeclampsia were generally present [46,66-69]. In a prospective cohort study of women who underwent home polysomnography, sleep-disordered breathing was associated with having a small-for-gestational age infant (OR 2.65, 95% CI 1.15-6.10). Larger studies that take into account the multiple factors that affect birth weight negatively (eg, preeclampsia, poor gestational weight gain) and positively (eg, obesity, diabetes, excessive gestational weight gain) are needed. A systematic review of sleep-disordered breathing in pregnancy reported that it was associated with a modest increase in low birth weight (<2500 g, OR 1.67, 95% CI 1.00-2.78) [8]. As discussed above, the findings were limited by inclusion of studies with different forms of sleep-related breathing disorders, lack of rigorous diagnostic criteria, and lack of adjustment for potential maternal confounders, particularly maternal age and body mass index.

Intrauterine fetal death — There are no high-quality data on how gestational OSA affects the risk of intrauterine fetal demise. The systematic review discussed above reported a modestly increased risk of stillbirth or perinatal death (OR 2.02, 95% CI 1.25-3.28) [8].

Effect of pregnancy on OSA — There are no studies investigating the difference in clinical outcomes between women with preexisting OSA and those who develop the condition during pregnancy. Evidence from small cohort studies indicates that pregnancy may exacerbate underlying sleep-related breathing disorders. In one longitudinal study that included 10 women who presented with a sleep-related breathing disorder in the third trimester, sleep apnea severity improved after delivery [70]. However, 70 percent of the women still had the disorder up to three months postpartum. This study did not account for possible body position effects on the degree of sleep-disordered breathing, postpartum weight loss, and changes in body fat composition and distribution, all of which possibly contributed to these findings. No larger scale longitudinal studies have assessed sleep-related breathing disorder in pregnancy using objective measures. (See "Gestational diabetes mellitus: Screening, diagnosis, and prevention".)

MANAGEMENT

Options — As with other aspects of OSA, the approach to treatment of OSA in pregnancy is based on extrapolation from data in the general population. There are no pregnancy-specific guidelines for OSA treatment, no data on which to base fetal or maternal parameters for treatment, and no evidence that treatment in the short term impacts maternal or neonatal morbidity. In the absence of scientific evidence, clinicians have used their own protocols [71] or followed the published recommendations for treatment in the general population [72-74]. In our practice, we treat all pregnant women with moderate or severe OSA by apnea-hypopnea index (AHI) criteria. In the general population, therapies for OSA include continuous positive airway pressure (CPAP), oral appliances, upper airway surgery, and behavior modification. The indications for treatment and these treatments are reviewed separately. (See "Management of obstructive sleep apnea in adults".)

Pharmacologic therapy — No pharmacologic agent prevents or overcomes upper airway obstruction well enough to justify pharmacologic therapy as a primary therapy in the routine management of patients with OSA. Modafinil has been used to treat the sleepiness (not the apneic events) associated with OSA. It is generally prescribed only for patients with OSA who experience problematic residual hypersomnolence despite adequate usual total sleep time and effectively treated OSA (eg, who are adherent to CPAP therapy). Women with OSA who become pregnant while being treated with modafinil or similar stimulants should be counseled about the paucity of safety data in pregnancy. The Nuvigil and Provigil Pregnancy Registry reported that the frequency of major congenital anomalies (17 percent) and cardiac anomalies (4 percent) associated with the exposure to modafinil and/or armodafinil in utero was above the frequency observed in the general population (3 and 1 percent, respectively), although some women took additional drugs. Based on these findings and those from post-marketing cases, in June 2019, Teva Pharmaceuticals Ireland, in agreement with the European Medicines Agency and Health Products Regulatory Authority, issued a direct health care professional communication stating that the use of modafinil during pregnancy is suspected to cause congenital malformations and should not be used during pregnancy [75]. Shortly afterwards, Health Canada issued a warning that modafinil is contraindicated in women who are pregnant or may become pregnant [76]. While the US Food and Drug Administration has not issued a similar warning, these alerts suggest modafinil should not be used by women who are pregnant or who may become pregnant. Modafinil may be present in breast milk, but data are limited [77].

In light of available data, we recommend stopping modafinil during pregnancy and lactation. When discontinued, patients should be warned about increased sleepiness and should take appropriate precautions, such as when driving. (See "Evaluation and management of residual excessive sleepiness in adults with obstructive sleep apnea".)

CPAP — CPAP is generally regarded as first-line therapy for pregnant women with OSA, similar to the general population. We recommend in-laboratory polysomnography for the initial titration of CPAP, if the study (eg, a single-night split baseline and CPAP titration study, or a titration study performed within a few days of a full-night baseline polysomnography) and subsequent ordering of CPAP therapy can be done expeditiously. In-laboratory CPAP titration allows patients’ initial experience with CPAP to occur in a monitored setting so that questions and problems (such as CPAP mask leaks) can be addressed in real-time. Following an in-laboratory CPAP titration study, fixed-pressure CPAP therapy is an option; providers should recognize, however, that the therapeutic CPAP pressure may need to be increased over the course of pregnancy according to patient symptoms (eg, snoring while using CPAP) [78]. A prior study examining this issue observed that an increase in CPAP pressure of 1 to 2 cm H₂O was required among some women with OSA over the course of pregnancy [78]. For this reason, we prefer prescribing auto-titrating CPAP (APAP) for treatment of OSA during pregnancy, using the results of the CPAP titration study as a guide to the range of pressures to be prescribed. When in-laboratory CPAP titration cannot be obtained quickly, auto-titrating CPAP therapy can be initiated empirically without an additional sleep study. In either case, follow-up should occur within one to two weeks after initiation of CPAP therapy and regularly thereafter (eg, every one to three months during pregnancy) so that adherence to and efficacy of CPAP treatment can be assessed and adjusted as needed.

The safety of CPAP therapy in pregnancy has not been systematically studied, but case reports have not described adverse effects [19,79]. In the general population, the most common side effects of CPAP use include rhinitis, skin abrasions, mouth dryness and aerophagia [80]. These complaints are echoed by pregnant patients in our clinics. One theoretical concern of nasal CPAP is the potential for decreased cardiac output due to the applied positive intrathoracic pressure. A study of 24 women with severe preeclampsia observed that their cardiac output decreased during sleep (from 7.7 L/min to 5.7 L/min), but remained at baseline when the women were treated with CPAP [81]. These data may not apply to pregnant women with OSA who are not preeclamptic.

It is estimated that only 50 percent of individuals are adherent to CPAP therapy, but rates of compliance in pregnancy have not been reported. In our pregnant population, we have found adherence rates that are similar to the general population [82]. Pregnant patients in our clinics often cite nasal congestion and lack of symptomatic improvement as reasons for nonadherence. (See "Titration of positive airway pressure therapy for adults with obstructive sleep apnea".)

Devices — Customized oral mandibular repositioning devices, which are noninvasive and generally seek to keep the airway open by pulling the lower jaw forward, can be an effective treatment for mild to moderate OSA for individuals with good dentition [83]. However, only 35 to 40 percent of patients achieve a reduction in the degree of sleep-disordered breathing to normal levels (ie, AHI <5 events/hour), and oxyhemoglobin saturation levels often do not return to normal [84]. This latter point is an important consideration, given the potential for fetal hypoxemia among pregnant women with OSA. Furthermore, the multiple fittings that can be needed to initiate oral appliance use, the time it takes to fabricate the device, and the potential need for frequent refitting due to the potentially rapid weight gain of pregnancy are likely to make this treatment impractical for pregnant women. No mandibular repositioning device data exist in pregnant women with OSA. (See "Oral appliances in the treatment of obstructive sleep apnea in adults".)

Surgery — Among the general population, surgical treatments for mild to moderate OSA, such as uvulopalatopharyngoplasty, generally achieve a reduction in AHI to the normal range in less than half of patients [85]. Given the relative lack of efficacy and the potential for adverse consequences associated with surgery and anesthesia (which are likely to be magnified during pregnancy), uvulopalatopharyngoplasty and other surgeries for OSA are generally not suitable approaches.

Gestational weight gain — Weight loss, which has been demonstrated to improve comorbid conditions and decrease the overall AHI, is not recommended in pregnancy. However, limiting weight gain may benefit an individual's global risk for pregnancy related morbidity. (See "Gestational weight gain", section on 'Recommendations for gestational weight gain'.)

Other — General measures that can be helpful in OSA include position adjustment to favor the lateral recumbent or head elevated position and avoidance of respiratory destabilizing drugs.

Antepartum management

Preexisting OSA — Women with known OSA who become pregnant should be evaluated and followed by a sleep medicine provider. This is particularly important if the patient has not been evaluated in the previous six months or is not being treated. Their provider can assess symptoms and evaluate the need for a CPAP titration or, in some cases, repeat polysomnography. These individuals may be at risk for developing pregnancy related hypertensive disease, so blood pressure and urine protein should be monitored closely. Given the high risk of underlying diabetes in individuals with OSA, we suggest early testing for gestational diabetes. In our practice, we perform an early glucose-tolerance test. If it is normal, it is repeated at 24 to 28 weeks gestational age. (See "Gestational diabetes mellitus: Screening, diagnosis, and prevention", section on 'Screening for undiagnosed type 2 diabetes in early pregnancy'.)

Newly diagnosed OSA — Women with excessive sleepiness, apneic symptoms or who are otherwise suspected of having OSA should be referred to a sleep medicine specialist for evaluation. This provider can evaluate the patient and participate in her care throughout the pregnancy and after delivery. In the absence of pregnancy-specific data to direct treatment, we suggest treatment for all women with moderate or severe OSA. The goal of treatment is to normalize the AHI and to alleviate any nocturnal hypoxemia. Among pregnant women with mild OSA, usual indications for treatment of OSA (such as excessive daytime somnolence) can first be considered when deciding whether to recommend treatment. In addition, we believe that pregnant women with mild OSA should be treated if they have recurrent oxyhemoglobin desaturations to less than 90 percent, given the potential for adverse effects of hypoxemia on the fetus. Among otherwise asymptomatic pregnant women with mild OSA who develop adverse outcomes that have been associated with OSA (eg, the hypertensive diseases of pregnancy or gestational diabetes), a trial of CPAP therapy may be considered, but is not clearly indicated [86].

Intrapartum management — Pregnancies affected by OSA should be considered "high risk" pregnancies. Maternal risks include hypoxemia and morbidity from the comorbid conditions that often track with OSA. Many intrapartum issues related to OSA overlap with the risks of obesity (see "Obesity in pregnancy: Complications and maternal management"). Continuous pulse oximetry during labor and the postpartum period will help identify hypoxemia and help tailor further management.

Ideally, the anesthesia service is consulted prenatally, but at least early in the patient's labor. Early placement of regional anesthesia may prevent the need for general anesthesia if emergency cesarean delivery becomes necessary. General anesthesia has been associated with an increase in maternal morbidity and mortality, which is more profound in individuals with OSA [87]. Regional anesthesia also avoids the potential need for opioids to manage labor pain. (See "Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea".)

Women who have been using CPAP, another positive pressure modality, or an oral appliance prior to delivery for OSA should be instructed to bring their devices when they present for delivery. Patients with OSA should have their oxygenation continuously monitored during labor with a pulse oximeter. If hypoxemia is observed, a comprehensive clinical evaluation should be performed to determine the etiology. The clinical evaluation should focus on respiratory causes of hypoxemia and, in particular, any potentially life-threatening causes such as pulmonary embolus. Pregnant women who are suspected of having OSA, or who are found to have OSA and are not on effective therapy, should be referred to a physician with expertise and experience in sleep medicine for possible intrapartum management and postpartum follow-up.

Postpartum management — The use of opioids in individuals with sleep apnea is an area of significant concern. Opioids can suppress the respiratory drive, blunt the arousal response, and cause nocturnal hypoxemia. There have been reports of sudden death with the use of narcotics in subjects with sleep apnea [88]. The following general guidelines apply to management of postpartum patients with OSA after vaginal delivery and after cesarean delivery. These issues are discussed in more detail separately. (See "Postoperative management of adults with obstructive sleep apnea".)

●Patients with OSA who were receiving therapy (eg, positive airway pressure, oral appliances, etc) prior to delivery should resume the therapy as soon as is feasible after delivery [74]. A physician with experience and expertise in sleep medicine should be consulted to participate in management if such a physician has not already been involved in the patient's care.

●Use of an analgesic strategy that minimizes the need for systemic opioids, such as use of nonsteroidal anti-inflammatory drugs. Neuraxial analgesia is also acceptable for women who have undergone cesarean delivery, but it must be employed with caution. Use of any drugs that suppress CNS/respiratory activity should be minimized.

●If hypoxemia is observed, provide supplemental oxygen after delivery to maintain oxygen saturation by pulse oximetry ≥96 percent while sleeping and undisturbed. This should be continued until the patient is able to maintain her baseline oxyhemoglobin saturation while breathing room air. This management is recommended for all women with diagnosed or suspected OSA. The use of CPAP in women with OSA in the postpartum period may be sufficient to treat the hypoxemia; however, the CPAP settings should be adjusted to assure normoxic values [71].

●The supine position worsens OSA in some patients. Thus, it seems reasonable to care for patients who have OSA in the lateral or upright position, if possible, especially when they are taking opioids.

●A scoring system developed by the American Society of Anesthesiologists (ASA) to assess the risk of perioperative complications in patients at risk for OSA may be helpful in identifying patients at increased risk for postdelivery complications (table 1); patients with a score of four or more should be monitored with continuous pulse oximetry. An enhanced but more complicated version of this system has also been created (table 2).

●Patients with OSA should be counseled about the risks related to breastfeeding and childcare if they have symptoms of excessive sleepiness.

Follow-up — All women diagnosed with, or suspected of having, OSA during pregnancy should be followed postpartum by a physician or other provider with specific training and experience in sleep medicine to allow for re-assessment of OSA severity and overall management/treatment strategy (eg, re-evaluation of CPAP prescription requirements, oral appliance prescription, weight management, etc). Women diagnosed with OSA during pregnancy should have repeat testing, particularly after losing their pregnancy weight. At our institutions, we generally recommend that this evaluation take place 8 to 12 weeks after delivery. However, it is important to keep in mind that although up to 80 percent of gestational weight gain is usually lost within six weeks of delivery, weight loss can continue for several months postpartum [89]. Thus, postpartum management of sleep apnea needs to be individualized. In obese women, we recommend weight management strategies which, in addition to providing benefit with regard to OSA, will help in the management of their comorbid conditions and improve overall health before a subsequent pregnancy.

There are no data regarding the effect of an interval pregnancy on the natural history of OSA, or regarding whether the development of incident OSA during one pregnancy increases the risk for OSA in subsequent pregnancies. Similarly, whether a prior history of gestational OSA affects risk for adverse maternal-fetal outcomes in subsequent pregnancies (when OSA has not recurred) has not been reported.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sleep-related breathing disorders in adults".)

SUMMARY AND RECOMMENDATIONS

●Physiologic changes related to pregnancy may worsen preexisting obstructive sleep apnea (OSA) or lead to gestational OSA. (See 'Physiologic changes of pregnancy predisposing to OSA' above and 'Effect of pregnancy on OSA' above.)

●Clinicians and patients may think that poor sleep and excessive daytime sleepiness are normal symptoms of pregnancy, and thus fail to evaluate the patient for OSA. (See 'Clinical manifestations and diagnosis' above.)

●Two potential effects of OSA on pregnancy include increased risks of developing preeclampsia and gestational diabetes. There are no clear adverse fetal effects, apart from sequelae from these pregnancy complications. (See 'Effect of OSA on pregnancy' above.)

●There are no pregnancy-specific guidelines for OSA treatment, no data on which to base fetal or maternal parameters for treatment, and no evidence that treatment in the short term impacts pregnancy outcome. In the general population, therapies for OSA include continuous positive airway pressure (CPAP), oral appliances, and behavior modification. The indications for treatment and these treatments are reviewed separately. Women with pregestational OSA should have an evaluation by a sleep provider for CPAP management during pregnancy. (See "Management of obstructive sleep apnea in adults" and 'Options' above.)

●Pregnant women with OSA should be managed with a sleep medicine specialist. Intrapartum care should include an early consultation with the anesthesia service for placement of regional anesthesia and continuous pulse oximetry. Patients receiving opiates should be observed closely for respiratory suppression. (See 'Intrapartum management' above.)

●Postpartum, an analgesic strategy that minimizes the need for systemic opioids and other drugs that suppress CNS/respiratory function should be used. (See 'Postpartum management' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Dr. Robert Basner, who contributed to an earlier version of this topic review.