|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Pulmonary Medicine (Drs. Maasilta, Bachour, and Laitinen) and Obstetrics (Dr. Teramo), Helsinki University Hospital, Helsinki; and Department of Physiology (Dr. Polo), University of Turku, Turku, Finland.
Correspondence to: Paula Maasilta, MD, FCCP, Department of Pulmonary Medicine, PO Box 340, FIN 00029 HUS, Finland; e-mail: paula.maasilta{at}helsinki.fi
| Abstract |
|---|
|
|
|---|
Design: A case-control study to compare sleep-related breathing in obese pregnant women (mean prepregnancy body mass index [BMI] > 30 kg/m2) with pregnant women of normal weight (mean BMI, 20 to 25 kg/m2).
Setting: University teaching hospital with a sleep laboratory.
Participants: We recruited 11 obese women (BMI, 34 kg/m2; mean age 31 years) and 11 control women (BMI, 23 kg/m2; mean age 32 years).
Interventions: Overnight polysomnography was performed during early (after 12 weeks) and late (after 30 weeks) pregnancy.
Measurements and results: During pregnancy, obese mothers gained 13 kg and control women gained 16 kg. Sleep characteristics did not differ between the groups. During late pregnancy, the women in both groups slept more poorly and slept in supine position less. During early pregnancy, their apnea-hypopnea indexes (1.7 events per hour vs 0.2 events per hour; p < 0.05), 4% oxygen desaturations (5.3 events per hour vs 0.3 events per hour; p < 0.005), and snoring times (32% vs 1%, p < 0.001) differed significantly. These differences between the groups persisted in the second polysomnography, with snoring time further increasing in the obese. Preeclampsia and mild obstructive sleep apnea were diagnosed in one obese mother. One obese mother delivered a baby showing growth retardation (weight - 3 SD).
Conclusions: We have shown significantly more sleep-related disordered breathing occurring in obese mothers than in subjects of normal weight, despite similar sleeping characteristics.
Key Words: obesity pregnancy sleep-disordered breathing
| Introduction |
|---|
|
|
|---|
Obesity, as such, predisposes to sleep-related breathing disorders.10 Weight gain and increased nasal obstruction during pregnancy are considered potentially detrimental, leading to sleep breathing disorders.1 A decrease in arterial PO2 has been noted during pregnancy when women change from a sitting to the supine position while awake, a decrease more important in obese patients.11 It may be that a combination of the supine position and sleep leads to worsened hypoxemia during pregnancy.
One of the potentially detrimental changes during pregnancy is the enlarging uterus, which alters diaphragmatic function.12 This results in reduced functional residual capacity, and in some patients a closing capacity greater than functional residual capacity, potentially causing shunting and hypoxemia; this decreased functional residual capacity reduces lung oxygen stores, which may lead to earlier occurrence of hypoxemia during hypoventilation in sleep.13 Nocturnal breathing and oxygenation have, according to one study,14 been well maintained in multiple pregnancies in nonobese mothers; however, that study included one mother with a slightly elevated body mass index (BMI), and she had significant partial upper-airway obstruction.
Two reports4 5 suggest that pregnancy may precipitate or worsen sleep apnea. One study15 looking at eight obese snoring pregnant women with a clinical diagnosis of sleep apnea found intrauterine growth retardation in all cases, but others3 16 have reported good fetal outcome. However, Crawford17 and Joel-Cohen and Schoenfeld18 noticed that even a small degree of hypoxemia during pregnancy may have profound effects on fetal development, resulting in intrauterine growth retardation.
This study was designed to evaluate sleep-related breathing in obese women during early (after 12 weeks) and late (after 30 weeks) pregnancy in comparison to that of normal-weight control subjects. Furthermore, we assessed obstetric and delivery data and data on the newborns.
| Materials and Methods |
|---|
|
|
|---|
All subjects were referred for the first polysomnography after their first routine obstetric ultrasound examination at 12 weeks of pregnancy. The second polysomnography was performed after 30 weeks. A prestudy decision was made to propose nasal continuous positive airway pressure to all mothers with an apnea-hypopnea index (AHI) > 10 events per hour. A follow-up polysomnography was planned 6 months postpartum for those with major pathologic findings. In connection with both polysomnographies, venous blood was drawn for estradiol and progesterone measurement. Furthermore, routine obstetric and delivery data were noted, as well as data on the newborn.
Overnight Sleep Studies
Prior to each sleep study, the subjects were asked to keep a
sleep diary for 2 weeks. A sleep questionnaire based on the Basic
Nordic Sleep Questionnaire19
was filled in by all before
the first sleep study. A shorter follow-up questionnaire was used prior
to the second study. In connection with both overnight recordings,
daytime sleepiness was evaluated with the Epworth Sleepiness
Scale.20
The overnight hospital recordings were performed with a computerized 24-channel polygraph (Alice 3; Healthdyne Technologies; Marietta, GA). This included a four-channel EEG (C3/A2, C4/A1, O2/A1, O1/A2), electro-oculogram, and submental and leg (two separate) electromyograms. Heart rate was monitored through standard leads. Airflow was detected by monitoring with a nasal and oral thermistor. Thoracic and abdominal belts (Healthdyne effort sensor; Healthdyne Technologies) were used for respiratory movement detection. Pulse oximetry (BCI Oximetry 3100; BCI International; Waukesha, WI) was included in every recording. A body position sensor was included to record body position: supine vs other. A skin microphone was attached to the throat for snoring detection. Furthermore, the study included an overnight video recording. A nurse trained in sleep medicine attended the recording.
Sleep stage was scored manually in 30-s epochs and was based on the
criteria of Rechtschaffen and Kales.21
Respiratory and
nonrespiratory events were both scored visually. An apneic event was
defined as absence of nasal or buccal flow for
10 s. Hypopnea was
scored as the diminution of flow amplitude of > 50% and for > 10 s
associated either with an arousal or an oxygen desaturation
3%.
The AHI was defined as the number of apneas and hypopneas per hour of
sleep. An oxygen desaturation index of 4 (ODI4)
was automatically registered by the polygraph when there was a
4%
drop in oxygen saturation during sleep. An arousal was defined as an
EEG frequency shift to the
range for at least 3
s.22
A respiratory event was scored when diminution of
flow amplitude was observed for > 10 s associated with paradoxical
chest and abdominal movement sometimes with a crescendo pattern of
snoring. A respiratory arousal (RA) was scored when an arousal was
preceded by an apnea, a hypopnea, or a respiratory event. The RA index
was calculated by dividing the total number of RAs by total sleep time.
Snoring was detected with a microphone attached to the subjects throat, and the analog signal was transferred to the monitor screen. Another microphone was attached to the ceiling, 2 m from the patients head, to record sounds on a videotape. Subjects were asked to snore as loud as they could while lying supine during the calibration process before the start of the recording. The maximal snoring signal during calibration was given a value on an arbitrary scale from 0 to 100. With a snoring signal of < 50, no snoring was heard on the videotape. A snoring event was scored visually if the signal was at least 50% of the calibration signal. A snoring episode included at least one snoring event and terminated when no snoring event was detected for two breathing cycles. The time spent in snoring episodes was divided by total sleeping time to give the figure for snoring time. Breathing irregularity was scored visually in epochs of 5 min. Irregular breathing was noted when either the respiratory frequency or the respiratory amplitude was changed for > 50% of the epoch. Periodic leg movement scoring was based on American Sleep Disorders Association Task Force recommendations.23 All respiratory parameters were calculated separately for REM and non-REM sleep as well as for the supine position.
Statistical Analysis
The data are expressed as mean ± SEM. The Mann-Whitney
U test, z-adjusted for ties, and Yates corrected
2 test were used to compare findings between
the obese and control groups. Wilcoxons matched-pairs test was used
to compare the findings within the groups at two separate assessment
points. All computations were performed with a commercial statistical
package (Statistica v.5; StatSoft; Tulsa, OK); p values
0.05 were
considered to indicate statistical significance
Ethical Considerations
The study was approved by the Ethics Committee of the Helsinki
University Central Hospital. Written consent was obtained from each
subject prior to the study.
| Results |
|---|
|
|
|---|
|
|
- and
ß-blocker medication, her BP returned to normal. Gestational diabetes
developed significantly (p < 0.05) more often in the obese subjects
(8 of 11 women) than in the control group (1 of 11 women). However, no
medication was needed for diabetes control in any of these subjects.
Sleepiness and Sleep Characteristics
Sleepiness was evaluated on the Epworth Sleepiness Scales. If we
consider values > 8 of 24 to indicate abnormal
sleepiness,22
two subjects in the obese group and one
subject in the control group were exceptionally sleepy at the time of
the first polysomnography. Only the same three subjects were sleepy at
the time of the second recording.
The main sleep characteristics are shown in Table 2 . Three obese and two control subjects refused to undergo a second polysomnography. When the early- and late-pregnancy polysomnographies were performed during 16.3 ± 1.2 weeks and 32.7 ± 0.8 weeks in the obese group and 15.3 ± 0.7 weeks and 34.7 ± 0.7 weeks of pregnancy in the control group, respectively, no significant differences appeared between the groups at these time points.
|
Sleep-Related Breathing Parameters
Mean awake oxygen saturations were 95.8 ± 0.6% and
95.8 ± 0.6% in the obese subjects and 96.9 ± 0.5% and
96.3 ± 0.4% in the nonobese subjects during the two
polysomnographies, respectively. In the latter recording, one obese
subject slept 6% of the time (corresponding AHI, 0.2 events per hour),
and three obese subjects and one control subject slept 1% of the time
at an oxygen saturation < 90%; all other values were > 90%.
The obese woman with preeclampsia was found to have mild obstructive sleep apnea (AHI, 12 events per hour) during late pregnancy. Within 48 h, she was admitted to the sleep laboratory for initiation of treatment with nasal continuous positive airway pressure; however, on admission to the sleep laboratory, she was already in labor.
The main sleep-related breathing characteristics with arousal data are shown in Table 3 . The difference in AHI (1.7 events per hour vs 0.2 events per hour, respectively) between the obese and nonobese subjects was significant (p < 0.05), although all indexes were < 10 events per hour. Even more highly significant (p < 0.005) was the difference recorded in ODI4 (5.3 events per hour vs 0.3 events per hour). Obese subjects snored for one third of their sleeping time during the first recording, whereas control subjects snored only 1% of the time (p < 0.001). The RA indexes differed significantly (p < 0.001), with an index of 7.4 for the obese subjects and 0.8 for control subjects. When irregular breathing was scored as a percentage of REM sleep, non-REM sleep, or total sleep time, at all stages this finding was significantly (p < 0.05) more common in the obese women.
|
The sleep-related breathing parameters and arousal data within the groups at the two assessment points show several significant changes. With advancing pregnancy, the number of arousals only in the control group increased significantly (p < 0.05), but increase in snoring time was statistically significant (p < 0.05) only in the obese group. Increase in RA index in the obese group was also significant (p < 0.05). In the control group, irregular breathing as a percentage of sleeping time increased significantly in REM sleep (p < 0.01), non-REM sleep (p < 0.02), and also when calculated as a percentage of total sleeping time (p < 0.05).
Hormonal Data
The estradiol and progesterone levels are shown in Figure 2
. At neither level did obese or control subjects show significant
difference at any assessment point. The greatest difference was in
progesterone level during late pregnancy, 298 ± 39 nmol/L in the
obese subjects and 437 ± 57 nmol/L in the control subjects; as
expected, within both groups, both hormone levels increased
significantly (p < 0.05) between early and late pregnancy.
|
Six of the babies were girls in the obese group, and five were girls in the control group. The babies weights were 3,506 ± 223 g and 3,622 ± 128 g in the respective groups. Their corresponding lengths were 49.2 ± 0.8 cm and 50.5 ± 0.7 cm. The mother with mild obstructive sleep apnea gave birth to a normal child with a relative body weight of - 0.6 SD.24 The mother who spent 6% of her sleep time with an oxygen saturation < 90% in late pregnancy had a baby weighing 3 SD below the mean.
Follow-up Polysomnography
A follow-up polysomnography was performed for the two mothers with
abnormal findings in the second recording. At 6 months
postpartum, sleep-related breathing had normalized in both mothers,
although the mother with mild obstructive sleep apnea during pregnancy
still snored 59% of her sleep time.
| Discussion |
|---|
|
|
|---|
The sleep parameters in the pregnant women of the present study are in accordance with previous findings.2 6 7 8 Both groups of women slept more when monitored during early pregnancy, although this was not associated with marked sleepiness as evaluated by the Epworth Sleepiness Scale. As expected, during the last trimester all women spent less time in REM and SWS sleep. Their poorer sleep quality was further manifested in more arousals and more awake time during late pregnancy. However, periodic leg movement was recorded in neither group. As documented previously,6 both groups slept significantly less often in supine position during the third trimester.
Sleep-related breathing parameters differed significantly between our obese and control groups. The obese subjects had greater AHI and ODI4 values even during early pregnancy, and these values increased, although not significantly, during late pregnancy. Although the means and the great majority of individual AHI and ODI4 values even in our obese subjects fell within current reference values,26 no findings and no change occurred in the control group. This pilot study was not planned to look at the effect of other possible risk factors, eg, craniofacial characteristics, on sleep-related breathing and these issues need to be resolved in future, larger studies. Although our obese mothers were clearly overweight, they were not morbidly obese,25 and findings may be more prominent in cases of extreme obesity. Furthermore, our methodology with thermal sensors may have underestimated the number of hypopneas,27 which probably also explains the difference between AHI and ODI4 values.
Differences in sleep-related parameters between these obese and normal weight pregnant women are reflected even more with regard to snoring. The control group did not snore in either recording period, whereas the obese women spent one third and increasing to one half of their sleeping time snoring. In this study, the polygraph recorded snoring. Unfortunately, to our knowledge no data exist on the correlation between recorded and self-reported snoring. Studies looking at self-reported snoring have found it to be common in pregnancy (23%27 to 14%28 ), and to be either a risk for growth retardation of the fetus28 or no risk for the newborn.27 Thus, the issue of snoring and any links with adverse effects on pregnancy await further studies.
The optimal weight or BMI for women who wish to become pregnant is unknown.29 That weight gain during pregnancy is poorly correlated with prepregnancy BMI30 was also reflected in our study. The normal weight group gained more, meaning that weight gain cannot explain the difference observed in sleep-related breathing parameters. However, the obese mothers experienced more complications. Only one obese subject had preeclampsia. Gestational diabetes developed in eight women in the obese group and one woman in the control group. Because gestational diabetes is not known to affect respiration,31 it is unlikely that the difference in its incidence had an effect on the differences observed in sleep-related breathing parameters. However, a larger study will be needed to clarify conclusively the role of these complications on sleep-related breathing during pregnancy.
The high circulating level of progesterone during pregnancy is suggested to prevent sleep-related disordered breathing by increasing ventilatory drive.9 The mechanism leading to snoring and sleep apnea in obese pregnant women is still unclear. Whether the smaller rise in serum progesterone levels in our obese women in late pregnancy favored their increased snoring needs further study. However, significant differences in sleep-related breathing parameters were observed at both assessment points.
The sleep-related disordered breathing in these obese mothers offers one possible explanation for the adverse outcomes of pregnancy reported for this group. Further, larger studies are needed to confirm our preliminary findings and to characterize potential additional risk factors.
| Footnotes |
|---|
This study has been funded by the Helsinki University Hospital Special Funds.
Received for publication February 21, 2001. Accepted for publication May 16, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B. Izci, M. Vennelle, W. A. Liston, K. C. Dundas, A. A. Calder, and N. J. Douglas Sleep-disordered breathing and upper airway size in pregnancy and post-partum Eur. Respir. J., February 1, 2006; 27(2): 321 - 327. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Virkkula, A. Bachour, M. Hytonen, T. Salmi, H. Malmberg, K. Hurmerinta, and P. Maasilta Snoring Is Not Relieved by Nasal Surgery Despite Improvement in Nasal Resistance Chest, January 1, 2006; 129(1): 81 - 87. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Virkkula, A. Bachour, M. Hytonen, H. Malmberg, T. Salmi, and P. Maasilta Patient- and Bed Partner-Reported Symptoms, Smoking, and Nasal Resistance in Sleep-Disordered Breathing Chest, October 1, 2005; 128(4): 2176 - 2182. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Bachour and P. Maasilta Mouth Breathing Compromises Adherence to Nasal Continuous Positive Airway Pressure Therapy Chest, October 1, 2004; 126(4): 1248 - 1254. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Saaresranta and O. Polo Sleep-disordered breathing and hormones Eur. Respir. J., July 1, 2003; 22(1): 161 - 172. [Abstract] [Full Text] [PDF] |
||||
![]() |
N Edwards, P G Middleton, D M Blyton, and C E Sullivan Sleep disordered breathing and pregnancy Thorax, June 1, 2002; 57(6): 555 - 558. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |