http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, 2015; 28(3): 262–266 ! 2014 Informa UK Ltd. DOI: 10.3109/14767058.2014.913131

ORIGINAL ARTICLE

Obstructive sleep apnea is associated with alterations in markers of fetoplacental wellbeing Ghada Bourjeily1,2, Kristen Butterfield3, Patrizia Curran2, and Geralyn Lambert-Messerlian1,4 1

Warren Alpert Medical School of Brown University, Providence, RI, USA, 2The Miriam Hospital, Providence, RI, USA, 3Rhode Island Hospital, Providence, RI, USA, and 4Women and Infants Hospital of Rhode Island, Providence, RI, USA Abstract

Keywords

Background: Obstructive sleep apnea (OSA) has been associated with adverse fetal outcomes in some studies. Second trimester Down syndrome screening markers reflect fetal and fetoplacental wellbeing. We aimed to compare markers of fetal and feto-placental wellbeing in women with OSA and low risk controls. Methods: A retrospective case-control study of pregnant women with OSA and available second trimester markers was performed. Controls were screened for sleep disordered breathing (SDB) at the time of delivery using a questionnaire. Women at low risk for OSA were selected. Marker levels were adjusted for gestational age and race and reported as multiples of median and later adjusted for body mass index (BMI). Results: Twenty-four OSA cases and 166 controls were identified. Women with OSA had a higher mean BMI when compared to controls (37.1 ± 12.7 versus 24.1 ± 5.1, p ¼ 0.03). Estriol (uE3) multiples of the median (MoM) levels were lower in women with OSA compared to controls, even after adjusting for BMI, 0.74 (interquartile range (IQR) 0.45) versus 1.06 (IQR 0.38), respectively, p ¼ 0.026. Once adjusted for BMI, alpha feto-protein (AFP) MoM levels were no longer significantly different in women with OSA compared to controls. Conclusion: OSA is associated with reduced serum uE3 levels, independently of BMI, possibly indicating fetal distress.

Fetal markers, fetal wellbeing, obstructive sleep apnea, pregnancy, sleep disordered breathing

Introduction The incidence of obstructive sleep apnea (OSA) is likely on the rise due to the obesity epidemic, and women of reproductive age are no exception. In addition, pregnant women are at an increased risk for sleep disordered breathing (SDB) due to the physiologic changes of pregnancy, evidenced by a high prevalence of snoring in gravidas. Interestingly, SDB including snoring and OSA have both been associated with adverse pregnancy outcomes such as gestational hypertension [1–3] and gestational diabetes [2,4]. SDB is also a concern because of potential adverse fetal outcomes such as preterm birth and growth restriction. Studies evaluating both snoring and OSA have shown an increased risk of preterm birth [2,3,5]. However, in many

Address for correspondence: Ghada Bourjeily, MD, Associate Professor of Medicine, Department of Medicine, Warren Alpert Medical School of Brown University, 146 West River Street, 11C, Providence, RI 02904, USA. Tel: (401) 793-7410. Fax: (401) 793-7801. E-mail: [email protected]

History Received 4 February 2014 Accepted 5 April 2014 Published online 14 May 2014

studies this risk was most significant for induced delivery and appeared to be mediated by preeclampsia [2,5]. Snoring has also been associated with growth restriction in some studies [6,7] but not others [2,8,9]. Population differences and methodological differences in the assessment of growth restriction likely explain these discrepancies. Preliminary data suggest alterations in markers of fetal wellbeing such as interleukin-6 and nucleated red blood cells in cord blood of newborns of pregnant women who snore habitually compared to controls [10]. Alpha feto-protein (AFP) and estriol (uE3) are second trimester markers of fetoplacental unit wellbeing, and are used to assess the risk of Down syndrome in the offspring. AFP is also recognized as a biomarker of perinatal distress and has been used to monitor fetal wellbeing in pregnancy [11]. Estriol is also a marker of fetoplacental unit wellbeing and can be altered in conditions affecting certain fetal organs or the placenta. As OSA is associated with potential tissue ischemia, hypoperfusion and inflammation related to sympathetic stimulation, recurrent arousals and intermittent hypoxia, we hypothesized that women with OSA may have alterations in markers of fetoplacental wellbeing. The goal of this study was to evaluate levels of AFP and uE3 in women with OSA compared to pregnant controls at low risk for OSA.

DOI: 10.3109/14767058.2014.913131

Methods Participants A retrospective case-control study was performed. OSA cases Cases were identified by a review of ICD-9 codes for OSA, associated with an ICD-9 code for delivery from 2003 to 2009 at a large tertiary care obstetric hospital. Medical records were reviewed to confirm a history of OSA in the index pregnancy. Women with available polysomnography in the record or those with documentation of CPAP therapy having been initiated for sleep apnea at any time were included.

Obstructive sleep apnea and alteration in fetal markers

263

Fisher’s exact test, and reported the results as counts and percents. We analyzed continuous variables between cases and controls using Student’s t-test and reported results as mean and standard deviation. We used the Wilcoxon rank sum test to report significant differences between marker levels and OSA status, and used a multivariable linear regression model to examine this association controlling for BMI. We used a multivariable linear regression model to examine the association between markers and oxygen saturation and AHI, controlling for body mass index (BMI). We used commercial software (SAS version 9.1; SAS Institute, Inc, Cary, NC) for all analyses.

Results Controls A secondary analysis of a convenient sample of control subjects was identified from a database of English-speaking subjects screened for SDB by a questionnaire at the time of delivery as part of a study evaluating SDB and pregnancy outcomes. Detailed methods of this survey-based study have been published elsewhere [2,12,13]. Briefly, questions regarding snoring, gasping and witnessed apneas were used as in the multivariable apnea prediction index [14] and were answered by subjects in the presence of trained research personnel. Answers were scored as 0 (never), 1 (rarely), 2 (sometimes), 3 (often) and 4 (always). One thousand patients were recruited into the study. A group of subjects who scored zero for snoring, gasping and witnessed apneas (n ¼ 321) was then identified. Of those, subjects who had available screening markers were therefore selected as a control group (n ¼ 166). Polysomnography Polysomnographic reports were obtained from medical records and were performed at American Academy of Sleep Medicine (AASM) accredited laboratories, using AASM recommended technology, scoring methods and interpretation. Apnea hypopnea index (AHI) is defined as the number of apneas and hypopneas per hour of sleep. AHI 5 events per hour is considered abnormal. Disease severity is defined as mild (AHI 5–14), moderate (15–30) and severe (430). Laboratory testing Records of eligible patients were reviewed for second trimester screening markers. Levels of alpha fetoprotein (AFP) and estriol (uE3) sampled in the second trimester were measured at the time of sample receipt using automated chemiluminescent immunoassays on the DxI instrument (Beckman Coutler Inc., Chaska, MN). Data are expressed as multiples of the median (MoM) after correction for gestational age and maternal race. Although MoM usually correct for weight as well, we opted to adjust for BMI later in the analysis instead, as BMI reflects better on body habitus than weight. Inter- and intra-assay coefficients of variation were below 15%. Twin data were adjusted using an appropriate correction factor for each marker. Statistical analysis To examine patient characteristics between cases and controls, we analyzed categorical variables using two-tailed

Maternal characteristics A total of 40 pregnant women with OSA were identified by ICD-9 codes. Record review was used to confirm that OSA was diagnosed during or prior to the index pregnancy. Of those 40 women, 24 had second trimester markers obtained during the index pregnancy. As mentioned above, a total of 166 controls that scored ‘‘0’’ on snoring, gasping and witnessed apneas had available second trimester markers. Mean age was 28.9 ± 5.8 in OSA cases compared to 28.1 ± 6.2 in controls, p ¼ 0.5. Women with OSA had a higher mean BMI when compared to controls (37.1 ± 12.7 versus 24.1 ± 5.1, p ¼ 0.03), Table 1. Detailed polysomnography data were available at our institution in a total of 12 out of 24 OSA cases. Mean AHI was 20.2 ± 25.1, with the majority of women having mild OSA (AHI 5–15 events per hour). To evaluate for a potential selection bias, subjects with OSA who underwent second trimester screening were compared to subjects with OSA who did not undergo such screening for baseline characteristics. Both groups had similar demographics including age and a similar prevalence of comorbidities. Fetal characteristics There were no significant differences in fetal characteristics between cases and controls, even after adjusting for confounders. The median APGAR score at one minute and five minutes were similar for cases and controls (8.0 ± 2.1 versus 8.0 ± 1.3, p ¼ 0.9183) and (9.0 ± 1.7 versus 9.0 ± 0.5, p ¼ 0.0988), respectively. The median gestational age at delivery (in weeks) was not significantly different for cases Table 1. Maternal and fetal characteristics in cases and controls.

Maternal characteristics Mean and SD Age (years) BMI (kg/m2) AHI (events/h) Fetal characteristics Median and SD APGAR 1 minute APGAR 5 minutes Gestational age at birth Birth weight (g)

Cases

Controls

p value

28.9 ± 5.8 37.1 ± 12.7 20.2 ± 25.1

28.1 ± 6.2 24.1 ± 5.1 –

0.5 0.03

8.0 + 2.1 9.0 + 1.7 38.2 + 4.2 3448 + 986

8.0 + 1.3 9.0 + 0.5 39.0 + 2.3 3365 + 583

0.9 0.1 0.3 0.7

264

G. Bourjeily et al.

compared to controls (38.2 ± 4.2 versus 39.0 ± 2.3, p ¼ 0.3273). The median birth weight (in grams) was not significantly different for cases compared to controls (3448 ± 986 versus 3365 ± 583, p ¼ 0.6906). There were five preterm births in women with OSA and seven in the control group. None of the newborns in the OSA group were diagnosed with Down syndrome, trisomy 18 or neural tube defects. Markers There was a significant reduction in median AFP MoM levels in cases compared to controls (0.75, IQR 0.4 versus 1.0, IQR 0.52; p ¼ 0.03) (Figure 1). However, this difference was no longer significant when adjusted for BMI. There was a significant reduction in uE3 MoM levels in cases compared to controls (0.74, IQR 0.45; 0.51, 0.96 versus 1.06, IQR 0.38; 0.8, 1.25; p ¼ 0.0003) that persisted after adjusting for BMI (p ¼ 0.02) (Figure 2). These data are equivalent to 0.74/1.06 MoM (0.69 MoM) for uE3.

Discussion To our knowledge, this is the first study evaluating serum screening markers of feto-placental wellbeing in women with OSA. This study has found a significant reduction in both AFP and uE3 levels in women with OSA compared to controls; only the reduction in uE3 levels persisted even after correction for maternal BMI. Our findings may be explained by several mechanisms. It is possible that through intermittent hypoxia and oxidative stress, hypoperfusion or inflammation OSA affects the wellbeing of the fetus, the placenta or both. Estriol is secreted in large amounts during late pregnancy and functions as a marker of fetoplacental health, in that the biosynthetic pathway for the production of estriol requires three different organs to be fully functioning: fetal adrenals, fetal liver and the placenta. Disruptions in this pathway such as those Figure 1. Alpha feto protein levels in cases and controls. Data are shown in multiples of median. Lower end of the box indicates first quartile; upper end indicates third quartile; line inside the box indicates median; whiskers indicate data distribution; asterisk indicates outliers.

J Matern Fetal Neonatal Med, 2015; 28(3): 262–266

occurring in fetal anencephaly, placental sulfatase deficiency, fetal death, chromosomal abnormalities or molar pregnancies will lead to low levels of maternal serum uE3. In our study, we have observed one case of stillbirth in the OSA group, which was thought to be related to severe early preeclampsia. However, estriol levels were measured prior to the occurrence of death in this case. In order to identify women who had the diagnosis of OSA during pregnancy, we only included women with an ICD-9 code of OSA and either vaginal or cesarean delivery. Therefore, this study design excluded women with early losses. In addition, women with fetal demise at any point in pregnancy were excluded in the control group. More so, uE3 levels in this study were performed in the second trimester, where they are much less predictive of loss given the significantly lower incidence of fetal loss compared to the first trimester. Hence, the association of low uE3 levels with fetal death cannot be evaluated in this cohort. It is possible, however, that OSA may, through the mechanisms discussed above, have an effect on one or more of the organs involved in the estriol production pathway, leading to lower circulating levels of this protein in pregnant women with OSA. This study has also shown no significant reduction in AFP in OSA cases compared to controls once levels were adjusted for BMI. AFP is a glycoprotein that is initially produced by the fetal yolk sac and then by the fetal liver, as the fetal yolk sac degenerates. In maternal serum, AFP is first detectable around the 10th week of gestation and peaks around 25 weeks of gestation. Developmental abnormalities such as neural tube defects and gastrointestinal anomalies have been associated with elevated AFP [15,16] levels, while chromosomal anomalies such as Down syndrome and trisomy 18 are associated with low serum AFP. AFP levels have shown only a weak association with negative fetal outcomes; hence, they are not used clinically in the diagnosis of fetal death or preterm labor [17]. In our study, there was no significant association in AFP levels with preterm birth in either cases or controls, though the study was not powered or designed to

DOI: 10.3109/14767058.2014.913131

Obstructive sleep apnea and alteration in fetal markers

265

Figure 2. Estriol (uE3) levels in cases and controls. Data are shown in multiples of median. Lower end of the box indicates first quartile; upper end indicates third quartile; line inside the box indicates median; whiskers indicate data distribution; asterisk indicates outliers.

evaluate this association. Studies evaluating a link between AFP and hypoxia are scarce. In a study by Harrison and Peat [18] published in 1989, AFP levels were found to be significantly elevated in neonates who suffered from recurrent apneas of the newborn. The risk of apnea persisted until AFP levels reached the reference range. This study was never reproduced and no further studies linking AFP to apnea are available in the literature, to our knowledge. However, AFP has been found to be increased in cord blood of neonates exposed to smoking in-utero compared to neonates born to non-smokers and these levels appear to correlate with the number of cigarettes smoked [19]. Data linking maternal sleep disturbances and subtle indicators of fetal wellbeing are scarce. Tauman et al. [10] have found that newborns of mothers with habitual snoring had elevated levels of nucleated red cells in cord blood, erythropoietin and interleukin 6 when compared to nonsnorers. The increase in nucleated blood cells is a marker of chronic fetal hypoxia that is mediated by erythropoietin and interleukin-6. Another study by Fung et al. [20] has shown that pregnant women with OSA had lower levels of insulin growth factor (IGF)-1 with a corresponding increase in IGF binding protein (IGFBP)-1 and IGFBP-2 compared to nonOSA pregnant controls. The findings in this study did not reach statistical significance. Due to the lack of documentation of a sleep study in early pregnancy or prior to pregnancy in many cases in this study, we could not ascertain whether OSA had occurred de novo in the index pregnancy or whether it had predated pregnancy. It

is possible that the time frame in which OSA is present during pregnancy may play a key role in determining the effect on the fetal environment. Similar to other types of exposure (such as drugs or radiation), it is possible that there is a critical period during fetal development where the occurrence of maternal OSA would be more detrimental to the fetus or that having OSA de novo may not allow maternal adaptive mechanisms, which may be crucial for fetal well being, to be functional [10]. Future longitudinal studies starting in early pregnancy can help answer these questions. Our study certainly has strengths including the identification of a control group at low risk for OSA as well as the expertise in marker testing. This study is limited by its retrospective design and the identification of markers in only 60% of subjects. Currently, correction for body weight using the multiple of medians model only occurs to a given body weight (270–300 lbs in most labs) and marker levels may be altered in women with weights exceeding 300 lbs, possibly affecting our data. Thus, by adjusting for BMI in the regression analysis rather than using the multiple of medians model we have been able to curtail this limitation. Given the increasing prevalence of obesity in women of reproductive age, new weight correction equations may need to be established. As predictive power of screening tools for OSA in pregnancy have not been validated [21], it is also possible that some controls may still be affected had they been tested, potentially affecting the associations. However, when we performed a comparison of cases with OSA with only a subgroup of controls with BMI 525 (an even lower risk for

266

G. Bourjeily et al.

OSA), results were similar (data not shown). As objective CPAP adherence could not be confirmed in OSA cases, the effect of CPAP as a potential confounder of our results is difficult to assess and it is possible that the use of CPAP may have contributed to an underestimation of our findings. In summary, the present results suggest, for the first time, that maternal serum uE3 levels in women with OSA are reduced, independently of BMI. This marker alteration may be indicative of fetal distress, perhaps resulting from OSA-associated hypoxemia, and warrants further intensive investigation.

Declaration of interest The authors report no declarations of interest.

References 1. O’Brien LM, Bullough AS, Owusu JT, et al. Pregnancy-onset habitual snoring, gestational hypertension, and preeclampsia: prospective cohort study. Am J Obstet Gynecol 2012;207: 487.e481–9. 2. Bourjeily G, Raker CA, Chalhoub M, Miller MA. Pregnancy and fetal outcomes of symptoms of sleep-disordered breathing. Eur Respir J 2010;36:849–55. 3. Chen YH, Kang JH, Lin CC, et al. Obstructive sleep apnea and the risk of adverse pregnancy outcomes. Am J Obstet Gynecol 2012; 206:136 e131–5. 4. Qiu C, Enquobahrie D, Frederick IO, et al. Glucose intolerance and gestational diabetes risk in relation to sleep duration and snoring during pregnancy: a pilot study. BMC Womens Health 2010;10:17. doi: 10.1186/1472-6874-10-17. 5. Louis JM, Auckley D, Sokol RJ, Mercer BM. Maternal and neonatal morbidities associated with obstructive sleep apnea complicating pregnancy. Am J Obstet Gynecol 2010;202:261 e261–5. 6. Franklin KA, Holmgren PA, Jonsson F, et al. Snoring, pregnancyinduced hypertension, and growth retardation of the fetus. Chest 2000;117:137–41.

J Matern Fetal Neonatal Med, 2015; 28(3): 262–266

7. Micheli K, Komninos I, Bagkeris E, et al. Sleep patterns in late pregnancy and risk of preterm birth and fetal growth restriction. Epidemiology 2011;22:738–44. 8. Loube DI, Poceta JS, Morales MC, et al. Self-reported snoring in pregnancy. Association with fetal outcome. Chest 1996;109: 885–9. 9. Tauman R, Sivan Y, Katsav S, et al. Maternal snoring during pregnancy is not associated with fetal growth restriction. J Matern Fetal Neonatal Med 2012;25:1283–6. 10. Tauman R, Many A, Deutsch V, et al. Maternal snoring during pregnancy is associated with enhanced fetal erythropoiesis – a preliminary study. Sleep Med 2011;12:518–22. 11. Mizejewski GJ. Physiology of alpha-fetoprotein as a biomarker for perinatal distress: relevance to adverse pregnancy outcome. Exp Biol Med (Maywood) 2007;232:993–1004. 12. Bourjeily G, Raker C, Chalhoub M, Miller M. Excessive daytime sleepiness in late pregnancy may not always be normal: results from a cross-sectional study. Sleep Breath 2013;17:735–40. 13. Bourjeily G, Raker C, Paglia MJ, et al. Patient and provider perceptions of sleep disordered breathing assessment during prenatal care: a survey-based observational study. Ther Adv Respir Dis 2012;6:211–19. 14. Maislin G, Pack AI, Kribbs NB, et al. A survey screen for prediction of apnea. Sleep 1995;18:158–66. 15. Muller F. Prenatal biochemical screening for neural tube defects. Childs Nerv Syst 2003;19:433–5. 16. Drugan A, Weissman A, Evans MI. Screening for neural tube defects. Clin Perinatol 2001;28:279–87, vii. 17. Bartha JL, Comino-Delgado R, Arce F, et al. Relationship between alpha-fetoprotein and fetal erythropoiesis. J Reprod Med 1999;44: 689–97. 18. Harrison VC, Peat GM. Alpha-fetoprotein: a marker for recurrent apnoea of prematurity? Early Human Develop 1989;20: 175–82. 19. Beratis NG, Varvarigou A, Christophidou M, et al. Cord blood alpha-fetoprotein concentrations in term newborns of smoking mothers. Eur J Pediatr 1999;158:583–8. 20. Fung AM, Wilson DL, Lappas M, et al. Effects of maternal obstructive sleep apnoea on fetal growth: a prospective cohort study. PLoS One 2013;8:e68057. 21. Olivarez SA, Ferres M, Antony K, et al. Obstructive sleep apnea screening in pregnancy, perinatal outcomes, and impact of maternal obesity. Am J Perinatol 2011;28:651–8.

Copyright of Journal of Maternal-Fetal & Neonatal Medicine is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Obstructive sleep apnea is associated with alterations in markers of fetoplacental wellbeing.

Obstructive sleep apnea (OSA) has been associated with adverse fetal outcomes in some studies. Second trimester Down syndrome screening markers reflec...
244KB Sizes 0 Downloads 4 Views