Published ahead of print on December 18, 2003, doi:10.1164/rccm.200305-692OC
© 2004 American Thoracic Society
The Effect of Maternal Smoking on Respiratory and Arousal Patterns in Preterm Infants during SleepConstance S. Kaufman Pediatric Pulmonary Research Laboratory, Department of Pediatrics, Tulane University School of Medicine; and Comprehensive Sleep Medicine Center, Tulane University Hospital and Clinics, New Orleans, Louisiana Correspondence and requests for reprints should be addressed to Narong Simakajornboon, M.D., Department of Pediatrics, SL-37, Tulane University School of Medicine, New Orleans, LA 70112. E-mail: nsimaka{at}tulane.edu
Prenatal exposure to cigarette smoke is associated with an increased risk of sudden infant death syndrome. The effect of maternal smoking on apnea and arousal patterns in preterm infants is currently unknown. Multichannel polysomnographic studies were performed in preterm infants. Thirty infants were enrolled into the study: 16 exposed prenatally to cigarette smoke (S) and 14 control infants (C). There was no difference in the gestational and postconceptional ages at the time of study. Maternal smoking was associated with a significant increase in the apneic index in these infants (28.6 ± 6.4/hour [S] vs. 13.2 ± 3.9 [C]; p < 0.05), and the difference was noted for obstructive events and only during active sleep. The arousal index was significantly decreased in the maternal smoking group (34.5 ± 2.3/hour [S] vs. 46.3 ± 5.6/hour [C]; p < 0.05), with a specific decrease in percentage of arousal after respiratory events (10.7 ± 2.1% [S] vs. 29.4 ± 5.4% [C]; p < 0.05). In conclusion, preterm infants exposed prenatally to cigarette smoke have increased respiratory events during active sleep, predominantly due to obstructive apnea, and possibly a higher arousal threshold during apneic events. These alterations in respiratory and arousal patterns in preterm infants born to smoking mothers may lead to significant vulnerability in this population.
Key Words: maternal smoking • apnea of prematurity • preterm infant • infant apnea • infant arousal Sudden infant death syndrome (SIDS) is the leading cause of death during the first year of life. Maternal cigarette smoking is an important, independent risk factor for SIDS (1–5), and the relationship is dose dependant (4, 6). Both active and passive prenatal exposures to cigarette smoke are associated with increased risk of SIDS (2, 7). The mechanism underlying the correlation between maternal cigarette smoking and SIDS is currently unknown. SIDS has been attributed to a variety of pathophysiologic mechanisms, including respiratory control abnormalities. Previous studies have demonstrated the effect of prenatal exposure to cigarette smoke on apnea and arousal response in term infants. Kahn and colleagues demonstrated a dose-dependent increase in frequency and duration of obstructive apnea in infants born to smoking mothers (8). Tirosh and colleagues examined the effect of maternal smoking on arousal response and found that infants whose mothers smoked during pregnancy had a significant decrease in arousal response after obstructive events (9). A later study has confirmed an increase in arousal threshold in response to auditory stimuli and a decrease in overall behavioral awakenings in the newborn of smokers (10). The studies of the effect of maternal smoking on ventilatory drive revealed conflicting results. Ueda and colleagues demonstrated a reduction in respiratory drive and attenuation of hypoxic ventilatory response in newborns exposed prenatally to cigarette smoke (11); however, Campbell and colleagues revealed no significant alteration in ventilatory drive in these infants (12). There is limited information on the effect of prenatal exposure to cigarette smoke on apnea and arousal pattern in preterm infants. Prematurity is one of the risk factors for SIDS (13–15), and prenatal exposure to cigarette smoke may have profound effects on the respiratory pattern during sleep in this vulnerable population. This study examined the effect of maternal smoking during pregnancy on the sleep architecture, apnea, and arousal patterns in preterm infants. Some of the results of these studies have been reported previously in the form of an abstract (16).
Overnight polysomnographic evaluations were prospectively conducted in asymptomatic preterm infants. All infants were clinically stable for discharge from the nursery. The medical records were reviewed with special attention to history of maternal smoking during pregnancy. A short questionnaire was used to acquire additional information to determine the presence of significant prenatal complications and to inquire into the possibility of multiple substances abuse. The study was approved by the Committee on Use of Human Subjects at Tulane University Health Science Center, and parental informed consent was obtained. Additional details on the method are provided in an online supplement. Infants included in the control group were less than 34 weeks' gestational age, were not on any respiratory stimulants, and did not have any charted cardiorespiratory events for at least 1 week before the sleep study. An additional criterion for the smoking group was a history of maternal smoking of 10 or more cigarettes per day. Infants who had central nervous system malformations, inborn errors of metabolism, endocrinopathies, neuromuscular diseases, congenital heart disease, craniofacial malformation, oxygen-dependent bronchopulmonary dysplasia, Grade II–IV intracranial hemorrhage, and history of maternal drug abuse were excluded from the study. The polysomnographic studies were performed between 8 P.M. and 6 A.M. by using Alice3 Polygraph Recorders (Healthdyne Technologies, Marietta, GA). Studies were considered acceptable only when the sleep duration was at least 4 hours. It has been previously shown that the reliability of sleep and apnea estimation is adequate only when the study duration exceeds 3 hours (17). The standard infant montage was used and the following variables were recorded simultaneously: body position, electrooculogram (ROC/A1, LOC/A2), 3-channel EEG (O1A2, C4A1, C3A2), chin EMG, ECG, pulse oximetry and pulse wave form (Nellcore N200; Nellcore, Harvey, CA), thoracic and abdominal inductance plethysmography, nasal thermistor, end-tidal PCO2 monitoring (Pryon SC-300; Pryon Corporation, Menomonee Falls, WI), transcutaneous PO2 and PCO2 (Tina; Radiometer, Copenhagen, Denmark). An accelerometer was placed on one arm to measure gross body movements. The scoring was performed by two qualified scorers, each individually blinded to the group the infants belonged to. The reliability of the scoring was accepted when there was more than 90% agreement between scorers, and results were recorded as an average of their findings. Sleep scoring was performed using standard criteria for infants (18). The scoring was analyzed in 30-second epochs and assigned as awake, active, indeterminate, or quiet sleep. Sleep efficiency was defined by the percentage of total sleep time divided by the time in bed.
Apnea was defined as cessation of airflow for at least two respiratory cycles. Hypopnea was not scored. Apnea index, expressed as the number of apneic events per hour, was calculated in each sleep state. Arousal was defined by modified criteria applicable to infants. The criteria included a shift in the EEG pattern to frequencies of
Analysis
Thirty infants met the criteria for enrollment into the study. There were 16 infants in the maternal smoking (S) group and 14 infants in the control (C) group. The general characteristics of the population studied are shown in Table E1 (see online supplement). There was no significant difference between the two groups in sex, gestational age (29.6 ± 0.9 weeks [S] vs. 28.2 ± 0.8 weeks [C]; p = not significant [NS]) and postconceptional age (37.9 ± 1.4 weeks [S] vs. 36.4 ± 0.6 weeks [C]; p = NS) at the time of sleep study, and birth weight. The amount of cigarette exposure during pregnancy in the study group ranged from 10 to 40 cigarettes/day (19.4 ± 2.2). None of the mothers in the control group smoked during pregnancy. Infants in each group spent a similar period of time in the neonatal intensive care unit before the sleep study (8.3 ± 1.4 weeks [S] vs. 8.1 ± 1.0 weeks [C]; p = NS). The characteristics of sleep state distribution in the infants of the smoking mothers and control group are illustrated in Table E2 (see online supplement). Both groups spent comparable amount of time in active sleep (62.2 ± 2.3% [S] vs. 65.1 ± 2.6% [C]; p = NS) and quiet sleep (31.5 ± 2.4% [S] vs. 30.4 ± 2.9% [C]; p = NS). There was no significant difference in sleep efficiency or state shifts. Prenatal exposure to cigarette smoke was associated with a significant increase in the overall apnea index (28.6 ± 6.4/hour [S] vs. 13.2 ± 3.9 [C]; p < 0.05). This difference was noted only during active sleep (38.1 ± 9.1/hour [S] vs. 15.2 ± 4.4/hour [C]; p < 0.05) (Figure 1) . Although, the apneic index during quiet sleep was also higher in infants born to smoking mothers, the difference was not statistically significant (14.5 ± 2.8/hour [S] vs. 7.9 ± 2.4/hour [C]; p = 0.08). Further analysis of the apnea type revealed that preterm infants born to smoking mothers had an increased percentage of obstructive apnea compared with control infants (25.7 ± 5.4% [S] vs. 9.9 ± 2.1% [C]; p < 0.05). No significant difference in percentage of central (61.2 ± 5.3% [S] vs. 72.5 ± 4.3% [C]; p = NS) and mixed apnea (13.1 ± 2.5% [S] vs. 17.5 ± 3.1% [C]; p = NS) was noted between the two groups (Figure 2) . In addition, there was a tendency toward longer duration of obstructive apnea in preterm infants born to smoking mothers; however, the difference did not reach statistical significance (6.8 ± 0.8 seconds [S] vs. 4.9 ± 0.6 [C]; p = 0.06]. Analysis of other cardiorespiratory parameters showed no significant difference in the average SaO2 (96.5 ± 0.5% [S] vs. 97.8 ± 0.1% [C]; p = NS) or end-tidal PCO2 (36.4 ± 1.6 mm Hg [S] vs. 38.8 ± 1.2 mm Hg [C]; p = NS). The total number of cardiac decelerations was similar between the two groups (7.3 ± 4.5 [S] vs. 8.7 ± 7.3 [C]; p = NS) (see Table E3 in the online supplement).
A significant decrease in the total arousal index was observed in preterm infants born to smoking mothers (34.5 ± 2.3/hour [S] vs. 46.3 ± 5.6/hour [C]; p < 0.05) (Figure 3) . The subanalysis of various types of arousals revealed that prenatal exposure to cigarette smoke was associated with a specific decrease in the percentage of arousals after apneic events (i.e., respiratory arousals) (10.7 ± 2.1% [S] vs. 29.4 ± 5.4% [C]; p < 0.05). The reduction in respiratory arousals was noted only during active sleep (12.9 ± 2.8% [S] vs. 37.7 ± 8.4% [C]; p < 0.05) (Figure 4) . No significant difference in respiratory arousals was noted during quiet sleep (7.5 ± 2.5% [S] vs. 12.6 ± 4.6% [C]; p = NS). The spontaneous arousals during active sleep were higher than during quiet sleep in both preterm infants born to smoking mothers (36.5 ± 2.9/hour [active] vs. 18.6 ± 2.4/hour [quiet]; p < 0.05) and in those in the control group (45.2 ± 5.5/hour [active] vs. 28.6 ± 9.7/hour [quiet]; p < 0.05). However, there was no significant difference between the two groups in the overall spontaneous arousal index (30.8 ± 2.3/hour [S] vs. 40.0 ± 5.2/hour [C]; p = NS) (Figure 3) or spontaneous arousal index during active and quiet sleep (Figure 4).
The main finding of the current study is that preterm infants exposed prenatally to cigarette smoke have a significant increase in respiratory events, especially obstructive apnea. In addition, a specific decrease in respiratory arousals is noted in these infants, suggesting an increased arousal threshold during apneic events. Both an increase in apneas and a decrease in respiratory arousals are observed only during active sleep. Before the implications of our study are addressed, certain methodologic limitations deserve comment. First, the evaluation of prenatal exposure to cigarette smoke was based on parental questionnaire and medical records; no objective method was used. Presenting a questionnaire to a mother who smokes may provoke some anxiety and may lead to underreporting of smoking during pregnancy. Although some studies have shown that urinary cotinine is not well correlated with maternal reports of smoking (23), others have indicated a good correlation between self-reports of parental smoking and urinary cotinine levels (24). Second, exposure to cigarette smoke was not evaluated quantitatively. Maternal smoking of 10 or more cigarettes per day during pregnancy was used as the cutoff point. The amount of cigarette exposure was comparable with previous studies in term infants (8, 9). The nicotine levels may vary between each mother and infant depending on the nicotine content in each cigarette, the smoking habits, and the nicotine metabolism (8, 25, 26). Third, there was a possibility that some infants may have been exposed to caffeine and illicit drugs, which may not be revealed by the mothers in the questionnaire. Maternal drug abuse and caffeine consumption during pregnancy have been linked to an increase in apnea in infants (27). Fourth, potential confounding factors associated with maternal smoking, such as maternal hematocrit and nutrition, were not assessed (6, 28, 29). Finally, the sleep study was performed for only one night, and the results may vary due to night-to-night variability. This is unlikely in infants because the polysomnographic evaluation on a single night study has been shown to be adequate to assess the sleep architecture and apnea (30). The present study has shown that asymptomatic preterm infants exhibit frequent respiratory events. Several studies have revealed that preterm infants with or without bronchopulmonary dysplasia may have significant apnea and oxygen desaturation (31–35). In addition, some studies have pointed out that a large percentage of apneas will not be detected by nursing staff (34–36). Our study further corroborated these findings. All infants who were enrolled in our study were reportedly asymptomatic for at least 7 days before the sleep studies, although overnight polysomnographic studies indicated the presence of frequent apnea. More importantly, these apneic events in preterm infants were significantly accentuated by prenatal exposure to cigarette smoke. The effect of maternal cigarette smoking on apnea has been previously reported in term infants. Infants born to smoking mothers have been shown to have an increase in obstructive events, even though the overall apneic index may be similar (8, 9). The present study is the first to evaluate the effect of prenatal exposure to cigarette smoke on apnea and arousal patterns in preterm infants. The findings indicated that preterm infants exposed prenatally to cigarette smoke had not only an increased percentage of obstructive apnea but also an elevation in the overall apneic index. No significant difference in other cardiorespiratory parameters including average SaO2, average end-tidal PCO2, or the number of cardiac decelerations was noted. The increased obstructive events were particularly noted during active sleep, which was the predominant sleep state in premature infants. The percentages of central and mixed apnea were comparable in both groups. The specific increase in obstructive events may have significant clinical implication. A recent study has demonstrated that infants who died of SIDS had experienced more frequent episodes of obstructive and mixed apnea (37). One of the important protective mechanisms during apneic and hypoxic events is arousal. Failure of the arousal mechanism was speculated as playing an important role in the pathophysiology of SIDS (38, 39). The role of arousal in the termination of respiratory events in infants is controversial. Although McNamara and colleagues suggested that arousal was not an important mechanism in the termination of respiratory events (40), other studies indicated that most obstructive apneas in infants and children were terminated by arousal (20, 41). Our study revealed that about one third (37%) of the respiratory events were followed by arousal in the control group. The discrepancy in the results may originate from the methodologic difference in the scoring criteria of arousals. Several recent studies have investigated the effect of maternal cigarette smoking on arousal mechanism in term infants. Franco and colleagues have demonstrated that infants born to smoking mothers have a higher arousal threshold to auditory stimuli during REM sleep (10). Subsequent studies have substantiated this finding and have showed that maternal tobacco smoking leads to impairment of both stimulus-induced arousal and spontaneous arousal (23, 42). Furthermore, one study has specifically indicated that the percentage of arousals after obstructive events is reduced in term infants born to smoking mothers (9). Our study suggested that the diminished arousal after respiratory events was also observed in preterm infants born to smoking mothers, and this change was noted only during active sleep. The arousal threshold in the premature infant is modulated by sleep state, gestational age, and postnatal age (43). Infants from both groups in the present study had similar gestational age and postconceptional age. The current study also revealed that spontaneous arousal during active sleep was higher than during quiet sleep in both preterm infants born to smoking mothers and in control infants. Therefore, it confirmed previous studies that the probability of spontaneous arousal from sleep was greater in active sleep than in quiet sleep (21, 22, 43, 44). In contrast, no significant difference in spontaneous arousal was noted between infants of smoking mothers and those of the control group. This data suggested the disparity between intact spontaneous arousals and defective respiratory arousals in premature infants born to smoking mothers. Prematurity and prenatal exposure to cigarette smoke are important risk factors for SIDS (1–5, 13–15). The relationship between SIDS and prenatal smoke exposure is dose dependent (4, 6), whereas that of prematurity is inversely related to gestational age (14, 15). The presence of both risk factors may lead to significant vulnerability to SIDS as suggested by simultaneous increased respiratory events and decreased respiratory arousals in the present study.
The mechanism underlying the alteration in respiratory and arousal patterns in infants born to smoking mothers is currently unknown. Several animal models have been used to examine to the effect of prenatal nicotine exposure on ventilatory and arousal response in developing animals. Milerad and colleagues examined the ventilatory response to hypoxia in the developing lamb after nicotine infusion. Nicotine attenuated ventilatory response to hypoxia and augmented the response to hyperoxia, suggesting that nicotine may alter peripheral chemoreceptor oxygen sensitivity (45). Bamford and colleagues demonstrated no change in the hypoxic and hypercapnic ventilatory response in nicotine-exposed developing rats (46). However, a subsequent study by the same laboratory revealed that prenatal nicotine exposure resulted in abnormal ventilatory response to hyperoxia, suggesting an alteration in dynamic ventilatory response after withdrawal of baseline peripheral chemoreceptor drive (47). A recent study has confirmed that prenatal nicotine exposure attenuates ventilatory response to hypoxia and hyperoxia during sleep in the developing lamb (48). Furthermore, nicotine-exposed neonatal rats had significantly lower All infants in our study stayed in the nursery for an average of 2 months before sleep studies; therefore, recent postnatal exposure to cigarette smoke was very unlikely. Although we could not exclude nicotine exposure from expressed breast milk from the mother, a previous study has revealed no significant correlation between breastfeeding in smoking mothers and frequency of obstructive apnea in term infants (8). The lack of postnatal exposure implies that prenatal exposure to cigarette smoke has a long-lasting effect on respiratory and arousal patterns in these preterm infants. Even though premature infants born to smoking mothers had greater respiratory events, the sleep state distribution, including the percentage of active and quiet sleep and sleep efficiency, was similar in the two groups. This could be explained by the insufficient arousal response to these respiratory events. In conclusion, our study has shown that preterm infants born to smoking mothers have increased respiratory events, especially obstructive apnea, and a decrease in arousal associated with respiratory events during active sleep. The concomitant increase in apneic events and elevated arousal threshold after apnea may result in significant vulnerability in these infants, and could underlie the pathophysiologic mechanism of increased risk of SIDS in preterm infants born to smoking mothers.
Supported by the American Heart Association 016023B, the Constance Kaufman Fund, the General Clinical Research Center, 5M01 RR05096-10 at Tulane University Health Sciences Center, and Louisiana SIDS Reduction Program. H.S. received an American Thoracic Society trainee travel award for this research project. Part of the study was presented as an abstract at the International Conference of the American Thoracic Society, 2002. This article has an online supplement, which is accessible from the issue's table of contents online at www.atsjournals.org Conflict of Interest Statement: H.S. has no declared conflict of interest; T.J. has no declared conflict of interest; T.M. has no declared conflict of interest; R.B. has no declared conflict of interest; N.S. has no declared conflict of interest. Received in original form May 24, 2003; accepted in final form December 13, 2003
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
(8–13 Hz) or above 16 Hz for a minimum of 1 second and that arousal in active sleep must be accompanied by a concurrent increase in submental EMG amplitude (
E in room air, indicating that prenatal nicotine exposure depressed eucapnic ventilation (
