Arousal and ventilatory responses to hypoxia in sleeping infants: effects of maternal smoking

Respiratory Physiology & Neurobiology 140 (2004) 77–87

Arousal and ventilatory responses to hypoxia in sleeping infants: effects of maternal smoking Peter M. Parsiow a , Susan M. Cranage a , T. Michael Adamson a , Richard Harding b,1 , Rosemary S.C. Horne a,∗,1 a b

Department of Paediatrics, Ritchie Centre for Baby Health Research, Monash University, Level 5, Monash Medical Centre, 246 Clayton Road, Clayton, Vic. 3168, Australia Department of Physiology, Monash University, Wellington Rd, Clayton, Vic. 3168, Australia Accepted 26 January 2004

Abstract Our aim was to determine whether maternal cigarette smoking affects arousal and ventilatory responses to hypoxia in infants. Infants born to non-smoking (NS, n = 15) and smoking mothers (SM, n = 9) were studied at 2–5 weeks, 2–3 and 5–6 months. Ventilatory responses to 15% O2 were determined preceding arousal. At each age and in both groups, infants aroused more frequently and earlier to hypoxia in active sleep (AS) than quiet sleep (QS). Arousal latency was longer in SM infants (in QS) at 5–6 months (P < 0.05). Baseline respiratory parameters were not different between groups, except that, at 2–3 months, SM infants had higher SpO2 during AS than NS infants. Maternal smoking did not affect ventilatory responses preceding hypoxia-induced arousal in either sleep-state at any age. We conclude that mild hypoxia stimulates ventilation and arousal in infants up to 6 months and that arousability is depressed in SM infants at 5–6 months; however, ventilatory responses preceding arousal are not adversely affected by smoking. © 2004 Elsevier B.V. All rights reserved. Keywords: Development, infants; Hypoxia, arousal; Mammals, humans; Sleep, arousal, cigarette smoke; Smoking, maternal

1. Introduction Maternal cigarette smoking during pregnancy is a major risk factor for the Sudden Infant Death Syndrome (SIDS). More than forty epidemiological studies have shown that smoking increases the risk of SIDS ∗ Corresponding author. Tel.: +61-3-9594-5100; fax: +61-3-9594-5700. E-mail address: [email protected] (R.S.C. Horne). 1 Joint senior authors.

by a ratio of between 0.7 and 4.85 (Anderson and Cook, 1997; Sullivan and Barlow, 2001). Although the physiological pathway by which maternal smoking could increase the risk of SIDS remains unknown, nicotine may alter cardio-respiratory function via nicotinic cholinergic receptors (nAChRs) involved in cardiovascular control, sleep and arousal (Grenhoff and Svensson, 1989; Lambers and Clark, 1996). Arousal serves as an important protective mechanism against cardio-respiratory failure during sleep and if maternal tobacco smoking impaired an infant’s ability to arouse, it could contribute to the occurrence SIDS.

1569-9048/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.resp.2004.01.004

78

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

The effects of nicotine or tobacco smoke exposure on arousal and cardio-respiratory function during sleep in early life have been investigated using animals (Bamford et al., 1996; Cohen et al., 2002; Fewell and Smith, 1998; Schuen et al., 1997) and human infants (Browne et al., 2000; Campbell et al., 2001; Chang et al., 2003; Franco et al., 1999, 2000; Galland et al., 2000; Horne et al., 2002; Kahn et al., 1994; Lewis and Bosque, 1995; Sovik et al., 2001). However, only two of these studies provide data on cardio-respiratory responses to hypoxia in human infants (Lewis and Bosque, 1995; Sovik et al., 2001) and few have been conducted in both active sleep (AS) and quiet sleep (QS) (Franco et al., 2000; Galland et al., 2000; Kahn et al., 1994). When investigating ventilatory responses of sleeping infants to hypoxia, it is particularly important to account for arousal, as infants are less arousable to exogenous stimuli and have fewer spontaneous arousals in QS than in AS (Horne et al., 2000, 2001, 2002; McNamara et al., 2002). In addition, arousal from sleep produces a transient increase in cardiac and respiratory activation (Horner, 1996). Owing to the paucity of knowledge on the effects of smoking on infant arousal from sleep the aim of this study was to determine whether maternal tobacco smoking affects neonatal arousal and ventilatory responses to hypoxia.

2. Methods Ethical approval for this project was obtained from the Monash Medical Centre Human Ethics Committee. Written informed parental consent was obtained prior to each investigation. Participation in the study was entirely voluntary with no monetary incentive provided. 2.1. Subjects The non-smoking (NS) group consisted of 15 (9 females/6 males) healthy term infants born at 38–41 weeks (39.7 ± 0.3 weeks, mean ± S.E.M.) with birth weights between 2890 and 4725 g (3549±132 g); Apgar scores were 3–9 (median 9) at 1 min and 6–10 (median 9) at 5 min. Some data for this group have previously been published (Parslow et al., 2003). The infants’ urinary cotinine levels at 2–3 months of age were all less than 10 ng ml−1 (2.8 ± 0.3 ng ml−1 ).

The smoking group (SM) consisted of 9 healthy term infants (4 females/5 males) born at 39–41 weeks (39.8 ± 0.3 weeks) with birth weights of 2700–3870 g (3538 ± 124 g) and Apgar scores of 3–9 (median 9) at 1 min and 9–10 (median 9) at 5 min. Urinary cotinine levels of these infants were all greater than 10 ng ml−1 (181.7 ± 70.3 ng ml−1 ). The mothers of the SM infants reported smoking 5–20 (12 ± 2) cigarettes per day during pregnancy and were still smoking a similar number (14 ± 2) at the time of the infant’s first sleep study. Polysomnographic studies were performed on each infant at 2–5 weeks, 2–3 and 5–6 months after birth. Two of the NS infants were unavailable for investigation at 2–3 months and one was unavailable at 5–6 months. Two of the SM infants were siblings studied 18 months apart and one SM infant was unavailable for the 2–5-week study. There were no significant differences between the NS and SM infants in gestational age at birth, birth-weight, Apgar scores at 1 and 5 min after birth, or in age or weight at any of the three studies. 2.2. Arousal criteria Sub-cortical arousal was determined according to four criteria (Horne et al., 2000, 2001), with the simultaneous occurrence of at least three of the criteria being required to designate an arousal. The four criteria were: a change in both amplitude and frequency of the ventilatory pattern for more than 2 breaths; an observed behavioural response; a heart-rate (HR) acceleration of greater than 10% above baseline; or increase in submentalis electromyogram (EMG) activity. The 10 s of recording immediately preceding the hypoxia-induced arousals were used for baseline measures to assess the degree of change in each physiological variable. 2.3. Respiratory measurements Nasal airflow was measured using a miniaturised pneumotachograph attached to a nose-mask, as described previously (Parslow et al., 2003). A continuous bias flow of medical-grade gas was passed across the pneumotachograph at 5 l min−1 , causing a pressure of 0.25 cm H2 O within the mask. A differential pressure transducer (Model PT5A, Grass Instrument

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

Co., Quincy, MASS, USA) was used to measure airflow via two ports positioned 5 mm either side of a fine wire-mesh screen within the pneumotachograph. An additional port was used to monitor expired CO2 . A custom-drilled 3-way tap (Swagelock, OH, USA) was used to change the bias flow gas supply from air (21% O2 ) to 15% O2 (balance N2 ). 2.4. Polysomnography The electroencephalogram (EEG), electro-oculogram (EOG), submentalis electromyogram (EMG), electrocardiogram (ECG) and instantaneous heart-rate were recorded using a polygraph (Model 78A, Grass Instrument Co., Quincy, MASS, USA). Thoracic and abdominal breathing movements were recorded using Resp-ez Piezo-electric sensors (EPM Systems, Midlothian, VA, USA), while oxygen saturation (SpO2 ) was recorded in a fast (3 s) response mode via a pulse oximeter attached to the ear (Biox 3700e, Ohmeda, Louisville, CO, USA). We also recorded abdominal skin temperature (YSI 400, Yellow Springs Instruments, USA) and transcutaneous CO2 (tcpCO2 ) (Model E5280, Radiometer, Copenhagen, Denmark). Data were simultaneously stored using digital data acquisition systems for respiratory analysis (MacLab/8e, ADInstruments, Sydney, NSW, Australia) and manual scoring of sleep-stages (S-series Sleep-System V5.2, Compumedics, Melbourne, Vic., Australia). 2.5. Study protocol Polysomnography was performed between 10:00 and 16:00 h. Electrodes were attached during a morning feed, following which all infants slept supine in a bassinette. Both morning and afternoon sleep periods were obtained following the infant’s regular feeding schedule. Environmental temperature remained constant (21–24 ◦ C), and noise and light levels were kept to a minimum. The hypoxic challenges were randomly presented in either the morning or afternoon sleep period to eliminate any “time of day” effect. Sleep-state was determined as AS, QS or indeterminate sleep according to standard criteria (Anders et al., 1971); indeterminate sleep was excluded from the analyses. Parents completed infant sleep diaries for the two days preceding each study to ensure that

79

the sleep obtained in the laboratory was similar to typical sleep patterns in the home. A 2 min control period (with mask/pneumotachograph attached) at the commencement of each sleep period provided baseline values of cardio-respiratory variables and respiratory gases for each sleep-state. A further one-minute control period was then obtained prior to each hypoxia test, following which the inspired gas was changed from air to 15% O2 (balance N2 ). Tests were terminated at either arousal, at 5 min in the absence of arousal (failure to arouse, FTA) or if SpO2 fell below 85%. Replicate tests (n = 3) were obtained in each sleep-state following recovery periods of at least 2 min during which the infant breathed room air. Those tests in which periodic breathing was induced were continued until either of the three termination criteria was met but were not included in the analyses. 2.6. Data analysis Hypoxia tests that were terminated due to SpO2 < 85% were excluded from arousal latency analyses. Arousal latency in those tests in which the infants did not arouse was denoted as 300 s. Sleep-period length (AS and QS duration) was determined for complete AS and QS periods that were naturally terminated, while sleep-cycle length (combined duration of a complete AS–QS cycle) was calculated as the time between the onset of successive AS periods. All data were assessed for normality using the Shapiro–Wilks statistic with significance taken at P < 0.05 and equality of variances assessed using Levene’s test. Paired-sample Student’s t-tests or Wilcoxon Signed Rank tests were used to compare the effect of sleep-state at each study age on: (a) sleep period duration; (b) baseline respiratory parameters; and, (c) arousal latency to hypoxia. Student’s t-tests were also used to compare total sleep times (TST) at home and in the laboratory. Chi square or Fisher exact analysis was used to compare the effects of sleep-state and maternal smoking on probability of arousal, calculated as the number of arousals divided by the total number of hypoxic stimuli presented and expressed as a percentage. Tests failing to induce arousal were excluded from respiratory analyses. Tests providing respiratory data for at least 15 breaths preceding arousal were analysed

80

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

breath-by-breath (b–b) for measurement of respiratory rate (f), tidal volume (VT (ml kg−1 )), inspiratory minute ventilation (VI (ml min kg−1 )), SpO2 and expired CO2 . Data from tests repeated within infants were averaged in each sleep-state. Data were then grouped for the 5, 10 and 15 breath time periods immediately preceding arousal. Initially, absolute values of respiratory variables were compared between ventilation immediately preceding hypoxic arousal and values obtained during the 1 min control period prior to each hypoxic challenge. To account for sleep-state related differences in baseline values, the ventilatory responses were then expressed as percentage changes from baseline levels for f, VT (kg−1 ), VI (kg−1 ) and SpO2 . Independent samples Student’s t-tests were used to compare data from NS and SM infants. Results are presented as mean±S.E.M., significance taken at P < 0.05.

3. Results 3.1. Demographic data There were no differences between the two groups in weight or gestational age at birth, nor in weight or age at any of the three studies.

nificantly lower (P < 0.01) in AS than in QS at 2–5 weeks and 2–3 months of age. VI (kg−1 ) was significantly higher in AS than QS at each age (P < 0.05). In addition, a significant difference in the duration of AS and QS periods was only present at 5–6 months, with QS periods being significantly longer than AS periods (P < 0.05). Sleep cycle duration did not change with age. Laboratory TST (matched for time of day) was not significantly different from TST at home, except at 5–6 months of age, when infants slept longer in the laboratory (91 ± 6 min) compared with at home (66 ± 8 min, P < 0.05). In SM infants there were no significant sleep-state related differences in SpO2 , f, VT (kg−1 ) or VI (kg−1 ). Expired CO2 levels were significantly lower in AS than in QS only at 2–3 months of age (P < 0.01). In addition, a significant difference in the duration of AS and QS periods was only demonstrated at 5–6 months, with QS peréods being significantly longer than AS periods (P < 0.05). Laboratory TST was not different to home TST and did not change with postnatal age. Between groups there were no significant differences in any of the baseline respiratory parameters at 2–5 weeks, 2–3 or 5–6 months of age, with the exception of baseline SpO2 during AS at 2–3 months. At this age, SM infants had higher baseline SpO2 levels during AS than NS infants (P < 0.01). The groups did not differ in AS or QS period duration and sleep cycle duration at any age.

3.2. Test data 3.4. Arousal latency and FTA analyses In NS infants, a total of 254 hypoxic challenges were performed, 115 in AS and 139 in QS. Of all tests, 63 in AS (55% of all performed tests) and 65 in QS (47%) provided ventilatory data for the 15 breaths prior to arousal. In SM infants, a total of 156 hypoxic challenges were performed, 72 in AS and 84 in QS. Of all tests, 39 in AS (46% of all performed tests) and 41 in QS (57%) provided ventilatory data for analysis of the 15 breaths prior to arousal. 3.3. Baseline respiratory and sleep data Baseline respiratory data obtained before each hypoxic challenge are provided in Table 1. In NS infants there were no significant differences in SpO2 or VT (kg−1 ) between sleep-states. Respiratory rate was significantly higher (P < 0.05) and CO2 levels sig-

Arousal latency (FTA tests included as 300 s) was calculated both for the entire set of data from all infants and for the subgroup of infants that provided suitable data for the ventilatory analyses. In NS infants using all available data, arousal latency to hypoxia was significantly shorter in AS than in OS at 2–5 weeks (n = 13, P = 0.001), 2–3 months (n = 13, P < 0.001) and 5–6 months (n = 10, P < 0.001) (Fig. 1). In the sub-group of infants used in the ventilatory analyses, arousal latency was also shorter in AS than in QS. In SM infants using all available data, arousal latency to hypoxia was significantly shorter in AS than in QS at 2–5 weeks (n = 8, P = 0.01), 2–3 months (n = 9, P < 0.001) and 5–6 months (n = 9, P < 0.001) (Fig. 1). When we analysed only data from the sub-group of infants used in the ventilatory analyses,

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

81

Table 1 Comparison of baseline respiratory values obtained during 1 min periods preceding hypoxic challenges and laboratory sleep data (min) in infants from smoking (SM) and non-smoking (NS) mothers 2–5 weeks AS

2–3 months QS

AS

5–6 months QS

AS

QS 94.7 ± 0.7 41.2 ± 1.7 30.6 ± 3.1 5.5 ± 0.2 164.6 ± 15.9∗ 22 ± 2 (14)∗ (10)

infantsa

NS SpO2 (%) CO2 (mmHg) f (bpm) VT (ml kg−1 ) VI (ml min−1 kg−1 ) Sleep period Sleep cycle

95.2 ± 0.6 95.9 ± 0.4 37.8 ± 1.3 40.7 ± 1.2∗∗ 50.8 ± 4.2 44.6 ± 4.4∗ 5.5 ± 0.3 5.2 ± 0.4 263.9 ± 19.3 214.7 ± l1.8∗∗ 24 ± 2 (15) 20 ± 1 (15) 43 ± 2 (10)

93.7 ± 0.5 94.6 ± 0.4 40.6 ± 1.9 43.2 ± 1.6∗∗ 36.7 ± 1.6 31.1 ± 2.6∗ 6.1 ± 0.2 6.0 ± 0.4 218.9 ± 2.7 182.1 ± 3.6∗∗ 18 ± 2 (11) 21 ± 1 (11) 39 ± 3 (10)

94.2 ± 0.7 41.9 ± 0.9 29.5 ± 2.7 7.0 ± 0.8 196.4 ± 17.3 15 ± 2 (14) 37 ± 2

SM infantsa SpO2 (%) CO2 (mmHg) f (bpm) VT (ml kg−1 ) VI (ml min−1 kg−1 ) Sleep period Sleep cycle

95.3 ± 0.5 95.9 ± 0.4 38.3 ± 0.8 41.8 ± 1.9 45.7 ± 2.6 35.5 ± 0.4 5.1 ± 0.3 5.4 ± 0.4 220.6 ± 4.2 188.7 ± 12.7 20 ± 3 (8) 18 ± 2 (8) 40 ± 5 (7)

95.9 ± 0 5† 95.8 ± 0.5 39.9 ± 1.3 41.6 ± 1.3∗∗ 35.8 ± 0.9 32.1 ± 3.6 5.6 ± 0.8 6.0 ± 0.3 194.3 ± 29.1 190.4 ± 25.1 14 ± 2 (7) 19 ± 2 (7) 41 ± 3 (5)

94.1 ± 0.5 94.3 ± 0.6 41.0 ± 2.1 41.4 ± 1.9 30.9 ± 0.9 31.0 ± 2.7 5.5 ± 0.6 4.6 ± 0.6 170.7 ± 20.7 142.8 ± 25.6 13 ± 1 (9) 20 ± 2 (9)∗ 35 ± 2 (9)

Results are expressed as mean ± S.E.M. Asterisks indicate sleep-state related difference within each study. Value in parentheses are number of infants. a Indicates difference between NS and SM groups. ∗ P < 0.05. ∗∗ P < 0.01. † P < 0.01.

arousal latency tended to be shorter in AS than in QS at all ages, with significant sleep-state related differences at 2–3 and 5–6 months. When arousal latencies were compared between groups at each postnatal age, although latencies were greater in QS in SM infants at both 2–5 weeks and 5–6 months, the difference only reached significance at the latter age (P < 0.05) (Fig. 1). The probability of arousal to hypoxia (% tests) was significantly higher in AS compared with QS at all three ages for both groups of infants. There was no significant difference in the probability of arousal between groups in either sleep-state at any age; however this just failed to reach significance at 2–5 weeks, with 80% of NS infants arousing compared to 44% of SM infants (P = 0.058) (Fig. 2). 3.5. Ventilatory responses preceding arousal 3.5.1. 2–5 weeks 3.5.1.1. Normoxia versus hypoxia data. In NS infants, VI (kg−1 ) and CO2 levels during hypoxia

were not significantly different from normoxic values in either sleep-state for 5, 10 and 15 breaths preceding arousal. Significant O2 desaturation was evident in both sleep-states for all three-time periods (Table 2). Different findings were obtained for VT (kg−1 ) and f in the different time periods (Table 2). In SM infants (Table 3), VI (kg−1 ) and CO2 levels during hypoxia were not significantly different from normoxic levels in either sleep-state for 5, 10 and 15 breaths preceding arousal. Although not significantly different, mean SpO2 levels were lower at the point of arousal than during the normoxic control periods. Different findings were obtained for VT (kg−1 ) and f in the different time periods. VT (kg−1 ) increased prior to arousal only in QS over 5 and 15 breaths (P = 0.07), while f was decreased only in AS at 10 and 15. When data were expressed relative to values in normoxia, there were no significant differences between groups in VT (kg−1 ), SpO2 , CO2 , VT (kg−1 ) or f in either of the three-time periods.

82

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

Fig. 1. Comparison of the effects of sleep-state on hypoxic arousal latency in NS (black bars) and SM (grey bars) infants, in AS (solid bars) and QS (hatched bars), at 2–5 weeks, 2–3 and 5–6 months (all available data). ∗∗ P < 0.01, ∗∗∗ P < 0.001. # P < 0.05 NS vs. SM.

Fig. 2. Comparison of the effects of sleep-state on the probability of arousal to hypoxia (% tests) in NS (black bars) and SM (grey bars) infants, in AS (solid bars) and QS (hatched bars), at 2–5 weeks, 2–3 and 5–6 months. ∗∗ P < 0.01, ∗∗∗ P < 0.001.

Normoxia

5 breaths hypoxia

10 breaths hypoxia

15 breaths hypoxia

AS

QS

AS

QS

AS

QS

AS

QS

2–5 weeks (10) VI (ml min−1 kg−1 ) SpO2 (%) CO2 (mmHg) VT (ml kg−1 ) f (bpm)

264 ± 19 95 ± 0.6 38 ± 1 5.5 ± 0.3 51 ± 4

215 ± 12 96 ± 0.4 41 ± 1 5.2 ± 0.4 45 ± 4

260 ± 20 91 ± 1.1∗ 38 ± 0.9 5.8 ± 0.4 49 ± 5

232 ± 13 91 ± 0.7∗ 41 ± 1.1 6.4 ± 0.5∗∗ 38 ± 4

269 ± 22 91 ± 1.0 38 ± 0.9 5.9 ± 0.5∗∗ 49 ± 5

227 ± 12 92 ± 0.7 41 ± 1.1 6.2 ± 0.4∗∗ 39 ± 4∗∗

262 ± 19 91 ± 0.9∗∗ 37 ± 1.1 5.7 ± 0.5∗∗ 50 ± 4∗

229 ± 13 92 ± 0.7∗∗∗ 41 ± 1.2 6.2 ± 0.4∗∗ 40 ± 4∗

2–3 months (6) VI (ml min−1 kg−1 ) SpO2 (%) CO2 (mmHg) VT (ml kg−1 ) f (bpm)

219 ± 3 94 ± 0.5 41 ± 2 6.1 ± 0.2 37 ± 2

182 ± 4 95 ± 0.4 43 ± 2 6.0 ± 0.4 31 ± 3

220 ± 13 89 ± 0.6∗∗∗ 40 ± 1.9∗∗ 6.3 ± 0.4 36 ± 2

186 ± 4 90 ± 0.5∗∗∗ 42 ± 1.9∗∗ 7.2 ± 0.5 27 ± 2

230 ± 15 89 ± 0.7∗∗∗ 40 ± 1.9∗∗ 6.2 ± 0.4 38 ± 2

185 ± 5 90 ± 0.5∗∗∗ 42 ± 1.8∗∗ 6.8 ± 0.4 29 ± 3∗

236 ± 17 89 ± 0.7∗∗∗ 39 ± 2.1∗∗ 6.2 ± 0.4 39 ± 1

187 ± 5 90 ± 0.5∗∗∗ 42 ± 1.8∗∗ 6.8 ± 0.4 29 ± 3∗

5–6 months (6) VI (ml min−1 kg−1 ) SpO2 (%) CO2 (mmHg) VT (ml kg−1 ) f (bpm)

196 ± 17 94 ± 0.7 42 ± 0.9 7.0 ± 0.8 29 ± 3

165 ± 16 95 ± 0.7 41 ± 1.7 5.5 ± 0.2 31 ± 3

223 ± 16 91 ± 0.7∗∗∗ 37 ± 2.8∗ 7.1 ± 0.7 34 ± 2

165 ± 20 91 ± 0.7∗∗∗ 41 ± 1.0 6.0 ± 0.5 28 ± 2

229 ± 16∗ 91 ± 0.6∗∗∗ 40 ± 1.0∗∗ 7.4 ± 0.8 33 ± 2

169 ± 19 91 ± 0.7∗∗∗ 40 ± 1.4∗∗ 6.0 ± 0.4 28 ± 2

229 ± 18∗∗ 91 ± 0.6∗∗∗ 40 ± 1.0∗∗ 7.3 ± 1.0 34 ± 3

170 ± 19 91 ± 0.8∗∗∗ 40 ± 1.7∗∗ 6.0 ± 0.4 28 ± 2

Results presented as mean ± S.E.M. Value in parentheses are number of infants. ∗ P < 0.05 normoxia vs. hypoxia. ∗∗ P < 0.01 normoxia vs. hypoxia. ∗∗∗ P < 0.001 normoxia vs. hypoxia.

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

Table 2 Comparison of ventilatory indices during normoxia and hypoxia in NS infants (5, 10 and 15 breaths preceding arousal) using absolute values

83

84

Normoxia

5 breaths hypoxia

10 breaths hypoxia

15 breaths hypoxia

AS

QS

AS

QS

AS

QS

AS

QS

2–5 weeks (4) VI (ml min−1 kg−1 ) SpO2 (%) CO2 (mmHg) VT (ml kg−1 ) f (bpm)

221 ± 4 95 ± 0.5 38 ± 0.8 5.1 ± 0.3 46 ± 3

189 ± 13 96 ± 0.4 42 ± 1.9 5.4 ± 0.4 36 ± 0.4

217 ± 28 92 ± 1.0 37 ± 1.3 5.2 ± 0.8 45 ± 4

200 ± 8 92 ± 0.8 40 ± 1.0 6.4 ± 0.7 32 ± 2

223 ± 21 92 ± 0.8 37 ± 1.0 5.5 ± 0.7 43 ± 2

202 ± 5 92 ± 0.9 41 ± 1.0 6.3 ± 0.7 33 ± 3

221 ± 20 92 ± 0.8 38 ± 0.9 5.5 ± 0.7 43 ± 2

203 ± 5 92 ± 0.8 41 ± 1.0 6.4 ± 0.7 33 ± 3

2–3 months (5) VI (ml min−1 kg−1 ) SPO2 (%) CO2 (mmHg) VT (ml kg−1 ) f (bpm)

194 ± 29 96 ± 0.5 40 ± 1.3 5.6 ± 0.8 36 ± 1

190 ± 25 96 ± 0.5 42 ± 1.3 6.0 ± 0.3 32 ± 4

219 + 47 92 ± 1.0∗∗ 39 ± 1.8 6.0 ± 1.0∗∗ 36 ± 2

206 ± 31 91 ± 0.7∗ 39 ± 1.5 7.4 ± 0.4∗∗ 29 ± 4

225 ± 42 92 ± 0.8∗∗ 39 ± 2 6.2 ± 0.9∗∗ 36 ± 2

203 ± 26 91 ± 0.7∗ 40 ± 1 7.2 ± 0.3∗∗ 29 ± 4

227 ± 36∗ 92 ± 0.8∗∗ 39 ± 2 6.2 ± 0.7∗ 37 ± 3

204 ± 28 91 ± 0.6∗ 40 ± 2 7.1 ± 0.3∗ 30 ± 4

5–6 months (5) VI (ml min−1 kg−1 ) SpO2 (%) CO2 (mmHg) VT (ml kg−1 ) f (bpm)

171 ± 21 94 ± 0.5 41 ± 2.1 5.5 ± 0.6 31 ± 1

143 ± 26 94 ± 0.6 41 ± 1.9 4.6 ± 0.6 31 ± 3

152 ± 23 90 ± 0.9∗∗∗ 40 ± 2.9 5.3 ± 0.8 29 ± 1

134 ± 25 91 ± 0.6∗ 40 ± 2.3 4.5 ± 0.9 31 ± 2

160 ± 25 90 ± 1.1∗∗ 40 ± 2.4 5.4 ± 0.6 30 ± 1

144 ± 27 90 ± 0.8∗ 40 ± 2.4 4.7 ± 0.8 31 ± 2

163 ± 25 90 ± 11∗∗ 40 ± 2 5.4 ± 0.6 30 ± 1

144 ± 28 90 ± 1.1∗∗ 40 ± 2 4.6 ± 0.7 31 ± 2

Results presented as mean ± S.E.M. Asterisk highlights difference between normoxic and hypoxic values within each study. ∗ P < 0.05. ∗∗ P < 0.01. ∗∗∗ P < 0.001.

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

Table 3 Comparison of ventilatory indices during normoxia and hypoxia in SM infants (5, 10 and 15 breaths preceding arousal) using absolute values

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

3.5.2. 2–3 months 3.5.2.1. Normoxia versus hypoxia data. In NS infants (Table 2), VI (kg−1 ) during hypoxia was not significantly different from normoxic values in either sleep-state in all three-time periods; however, significant O2 desaturation was present (P < 0.01) and CO2 levels significantly decreased in both sleep-states in all time periods (P < 0.01). VT (kg−1 ) remained unchanged in each time period, except for the 5 breaths preceding arousal in QS, when VT (kg−1 ) significantly increased prior to arousal (P < 0.05). Although f did not change prior to arousal in AS, it significantly decreased preceding hypoxic arousal over 5, 10 and 15 breaths in QS (P < 0.05). In SM infants (Table 3), f and CO2 levels during hypoxia remained unchanged in each time period. Significant O2 desaturation was present (P < 0.05) and VT (kg−1 ) significantly increased in both sleep-states in all time periods (P < 0.05). VT (kg−1 ) significantly increased immediately preceding arousal only in AS when averaged over 15 breaths (P < 0.05). When data were expressed relative to values during normoxia, there were no significant differences between groups in VT (kg−1 ), SpO2 CO2 , VT (kg−1 ) or f in either of the three-time periods. 3.5.3. 5–6 months 3.5.3.1. Normoxia versus hypoxia data. In NS infants (Table 2), VT (kg−1 ) and f during hypoxia were not significantly different from values in normoxia in either sleep-state for all three-time periods. Significant O2 desaturation was evident in both sleep-states for all time periods (P < 0.001). CO2 levels during hypoxia significantly decreased in both sleep-states in each time period (P < 0.05), except in QS at 5 breaths. VI (kg−1 ) significantly increased from normoxic levels only in AS in the 10 and 15 breaths preceding arousal (P < 0.05), remaining unchanged in QS in all time periods. In SM infants (Table 3), there was significant O2 desaturation in both sleep-states for all time periods (P < 0.05); however, VI (kg−1 ), CO2 , VT (kg−1 ) and f during hypoxia were not significantly different from normoxic levels in either sleep-state for all three-time periods. When data were expressed relative to values in normoxia, there were no significant differences between

85

NS and SM infants in VI (kg−1 ), SpO2 , CO2 , VT (kg−1 ) or f in either of the three-time periods.

4. Discussion Our study has shown that moderate levels of maternal cigarette smoking can modify arousal responses of infants to hypoxia in QS but do not affect ventilatory responses immediately preceding this arousal. Our study is the first to examine the effects of maternal smoking on responses to hypoxia in both AS and QS, as well as examining responses of the same infants during the first 6 months of life when the risk of SIDS is greatest. Our finding that maternal smoking depresses infant arousability in QS is consistent with previous studies showing that maternal smoking decreases arousability in QS in response to hypoxia (Lewis and Bosque, 1995), nasal air-jet stimulation (Horne et al., 2002) and auditory stimulation (Chang et al., 2003). However, other studies have failed to show an effect of smoking on arousal responses to tilting (Galland et al., 2000) and hypercapnic-hypoxia (Campbell et al., 2001). It has been postulated that altered arousal responses may be due to alterations in autonomic function (Horne et al., 2002; Franco et al., 1998; Galland et al., 1998). In support of this, several studies have now identified altered autonomic function in infants whose mothers smoked (Franco et al., 2000; Browne et al., 2000; Sovik et al., 2001). Our study is the first to examine the effects of maternal smoking on ventilatory responses prior to arousal. Our findings are consistent with those of Lewis and Bosque (1995), who also showed no effect of maternal smoking on ventilatory responses to hypoxia, although arousal responses were depressed in QS. Other studies, using different stimuli, have found an increased ventilatory response to hypercapnic-hypoxia in infants exposed to maternal smoking (Campbell et al., 2001). It was found that the degree of O2 desaturation was significantly greater in infants of smoking mothers than control infants at 1 and 3 months of age, despite an increased ventilatory sensitivity and a greater increase in VT in the smoking group (Campbell et al., 2001). The authors proposed that this may have been caused by inherent ventilation/perfusion mismatching due to increased airway wall thickness, as observed

86

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

in SIDS infants of smoking mothers (Elliot et al., 1998). Differences between our findings and those of previous studies may be due to the level of exposure to cigarette smoke during pregnancy. In our study, mothers were included as smokers if they reported smoking ≥5 cigarettes per day while pregnant (mean 12 ± 2 cigarettes per day) and/or had urinary cotinine levels >10 ng ml−1 . Other investigators have defined smokers as those who smoked 1–9 cigarettes per day (Kahn et al., 1994), ≥5 cigarettes per day (Campbell et al., 2001; Galland et al., 2000), ≥10 cigarettes per day (Franco et al., 2000; Kahn et al., 1994), 2–22 cigarettes per day (Sovik et al., 2001) or 15.6 ± 3.6 cigarettes per day (Browne et al., 2000). Only three other studies have verified maternal smoking with cotinine analysis (Lewis and Bosque, 1995; Browne et al., 2000; Campbell et al., 2001). In our study, birth-weight was not different between the two groups of infants indicating that the level of maternal smoking was not severe enough to induce overt perinatal compromise. Other studies have also demonstrated this (Horne et al., 2002; Franco et al., 1999, 2000). In contrast, some studies have shown that birth weights were significantly reduced in the smoking groups (Lewis and Bosque, 1995; Galland et al., 2000; Campbell et al., 2001; Browne et al., 2000). As in studies of human infants, some investigations in the rat have failed to demonstrate an effect of nicotine exposure on cardio-respiratory control and ventilatory responses to hypoxia (Bamford et al., 1996) or anoxia (Schuen et al., 1997). Differences in the level of smoke exposure between studies may explain the conflicting results. The effects of maternal smoking during pregnancy on cardio-respiratory development and function before and after birth are poorly understood. Nicotine, the major constituent of cigarette smoke, readily crosses the placenta and has been found in fetal cord serum in concentrations generally 15% higher than those in maternal serum (Lambers and Clark, 1996). It has been suggested that maternal cigarette smoking may alter neural control of cardio-respiratory, autonomic and arousal responses via a complex interplay of fetal hypoxia, neurotoxicity and metabolic abnormalities. Potentially, maternal smoking induces fetal hypoxemia by two mechanisms. Firstly, smoking leads to reductions in uterine blood flow (30%–40%), depending on

the dose of nicotine and this, in turn, reduces the supply of oxygen and nutrients to the fetus (Lambers and Clark, 1996). Secondly, infants of smoking mothers have been reported to have elevated levels of carboxyhaemoglobin (Bureau et al., 1983). The combination of decreased oxygen delivery and decreased oxygen carrying capacity of the fetal haemoglobin may expose the fetus to hypoxia during uterine life (Grenhoff and Svensson, 1989; Milerad and Sundell, 1993; Peacock et al., 1991). In support of this, several studies in animal models have demonstrated that nicotine exerts powerful inhibitory influences on cardio-respiratory control during the neonatal period, particularly in response to hypoxia (Fewell and Smith, 1998; Slotkin et al., 1997; Cohen et al., 2002). Our study has confirmed previous findings by ourselves and others that sleep-state has a marked influence on arousability, with both the frequency of arousal and latency to arousal being delayed in QS compared with AS (Horne et al., 2000, 2001, 2002; Parslow et al., 2003; Galland et al., 2000). This highlights the importance of examining arousal responses of infants in both sleep-states. In summary, our findings support and extend those of Lewis and Bosque (1995) that maternal cigarette smoking depresses the arousability of infants in QS while not affecting their ventilatory responses to mild hypoxic challenges during sleep. Any depression of arousability may make an infant more vulnerable to respiratory challenges during sleep and thus increase the risk of SIDS.

Acknowledgements The authors thank the parents and infants who participated in this study and the staff of the maternity wards and Jessie MacPherson Private Hospital, Monash Medical Centre. We also wish to thank Frank Meacco and Paul Nadalin (Biomedical Engineers, Monash Medical Centre) and Vojta Brodecky (Ritchie Centre for Baby Health Research) for their assistance in the development of our gas delivery system. We would also like to acknowledge the invaluable assistance of our research assistants Ms. Jessica Vitkovic and Mrs. Catherine Sangster, and the statistical advice provided by Mr. David Caddy and Dr. Kais Hamza, Monash University. This work was supported by the

P.M. Parsiow et al. / Respiratory Physiology & Neurobiology 140 (2004) 77–87

National Health and Medical Research Council of Australia, the Sudden Infant Death Research Foundation (South Australia) and SIDassist.

References Anders, T., Emde, R., Parmelee, A., 1971. A manual of standardized terminology, techniques and criteria for scoring of states of sleep and wakefulness in newborn infants. BRI Publications, Los Angeles. Anderson, H.R., Cook, D.G., 1997. Health effects of passive smoking. 2. Passive smoking and sudden infant death syndrome: review of the epidemiological evidence. Thorax 52, 1003–1009. Bamford, O.S., Schuen, J.N., Carroll, J.L., 1996. Effect of nicotine exposure on postnatal ventilatory responses to hypoxia and hypercapnia. Respir. Physiol. 106, 1–11. Browne, C.A., Colditz, P.B., Dunster, K.R., 2000. Infant autonomic function is altered by maternal smoking during pregnancy. Early Hum. Dev. 59, 209–218. Bureau, M.A., Shapcott, D., Berthiaume, Y., Monette, J., Bloum, D., Blanchard, P., Begin, R., 1983. Maternal cigarette smoking and fetal oxygen transport: a study of P50, 2,3diphosphoglycerate, total hemoglobin, hematocrit and type F hemoglobin in fetal blood. Pediatrics 72, 22–26. Campbell, A.J., Galland, B.C., Bolton, D.P.G., Taylor, B.J., Sayers, R.M., Williams, S.M., 2001. Ventilatory responses to rebreathing in infants exposed to maternal smoking. Acta Pediatr. 90, 793–800. Chang, A.B., Wilson, S.J., Masters, I.B., Yuill, M., Williams, J., Williams, G., Hubbard, M., 2003. Altered arousal response in infants exposed to cigarette smoke. Arch. Dis. Child 88, 30–33. Cohen, G., Han, Z., Grailhe, R., Gallego, J., Gaultier, C., Changeux, J., Lagercrantz, H., 2002. ␤2 Nicotinic acetylcholine receptor subunit modulates protective responses to stress: a receptor basis for sleep-disordered breathing after nicotine exposure. PNAS 99, 13272–13277. Elliot, J., Vullermin, P., Robinson, P., 1998. Maternal cigarette smoking is associated with increased inner airway wall thickness in children who die from sudden infant death syndrome. Am. J. Respir. Crit. Care Med. 158, 802–806. Fewell, J.E., Smith, F.G., 1998. Perinatal nicotine exposure impairs ability of newborn rats to autoresuscitate from apnea during hypoxia. J. Appl. Physiol. 85, 2066–2074. Franco, P., Pardou, A., Hassid, S., Lurquin, P., Kahn, A., 1998. Auditory arousal thresholds are higher when infants sleep in the prone position. J. Pediatr. 132, 240–243. Franco, P., Groswasser, J., Hassid, S., Lanquart, J.P., Scaillet, S., Kahn, A., 1999. Prenatal exposure to cigarette smoking associated with a decrease in arousal in infants. J. Pediatr. 135, 34–38. Franco, P., Chabanski, S., Szliwowski, H., Dramaix, M., Kahn, A., 2000. Influence of maternal smoking on autonomic nervous system in healthy infants. Pediatr. Res. 47, 215–220.

87

Galland, B.C., Reeves, G., Taylor, B.J., Bolton, D.P.G., 1998. Sleep position, autonomic function and arousal. Arch. Dis. Child Fetal Neonatal Ed. 78, F189–F194. Galland, B.C., Hayman, R.M., Taylor, B.J., Bolton, D.P.G., Sayers, R.M., Williams, S.M., 2000. Factors affecting heart rate variability and heart rate responses to tilting in infants aged 1 and 3 months. Pediatr. Res. 48, 360–368. Grenhoff, J., Svensson, T.H., 1989. Pharmacology of nicotine. Br. J. Addict. 84, 477–492. Horne, R.S.C., Sly, D.J., Cranage, S.M., Chau, B., Adamson, T.M., 2000. Effects of prematurity on arousal from sleep in the newborn infant. Pediatr. Res. 47, 468–474. Horne, R.S.C., Ferens, D., Watts, A., Vitkovic, J., Lacey, B., Andrew, S., Cranage, S.M., Chau, B., Adamson, T.M., 2001. The prone sleeping position impairs arousability in term infant. J. Pediatr. 138, 811–816. Horne, R.S.C., Ferens, D., Watts, A.M., Vitkovic, J., Lacey, B., Andrew, S., Cranage, S.M., Chau, B., Greaves, R., Adamson, T.M., 2002. Effects of maternal tobacco smoking, sleeping position, and sleep state on arousal in healthy term infants. Arch. Dis. Child. Fetal Neonatal Ed. 87, F100–F105. Horner, R.L., 1996. Autonomic consequences of arousal from sleep: mechanisms and implications. Sleep 19, S193–S195. Kahn, A., Groswasser, J., Sottiaux, M., Kelmanson, I., Rebuffat, E., Franco, P., Dramaix, M., Wayenberg, J.L., 1994. Prenatal exposure to cigarettes in infants with obstructive sleep apneas. Pediatrics 93, 778–783. Lambers, D.S., Clark, K.E., 1996. The maternal and fetal physiologic effects of nicotine. Semin. Perinatol. 20, 115–126. Lewis, K.W., Bosque, E.M., 1995. Deficient hypoxia awakening response in infants of smoking mothers: possible relationship to sudden infant death syndrome. J. Pediatr. 127, 691–699. McNamara, F., Lijowska, A.S., Thach, B.T., 2002. Spontaneous arousal activity in infants during NREM and REM sleep. J. Physiol., London 538, 263–269. Milerad, J., Sundell, H., 1993. Nicotine exposure and risk of SIDS. Acta Paediatr. Suppl. 389, 70–72. Parslow, P.M., Harding, R., Cranage, S.M., Adamson, T.M., Horne, R.S.C., 2003. Ventilatory responses preceding hypoxia-induced arousal in infants: effects of sleep-state. Respir. Physiol. Neurobiol. 136, 235–247. Peacock, J.L., Bland, J.M., Anderson, H.R., et al., 1991. Cigarette smoking and birth weight: type of cigarette smoked and a possible threshold effect. Int. Epidemiol. 20, 405–412. Schuen, J.N., Bamford, O.S., Carroll, J.L., 1997. The cardiorespiratory response to anoxia: normal development and the effect of nicotine. Respir. Physiol. 109, 231–239. Slotkin, T.A., Saleh, J.L., McCook, E.C., Seidler, F.J., 1997. Impaired cardiac function during postnatal hypoxia in rats exposed to nicotine prenatally: implications for perinatal morbidity and mortality, and for sudden infant death syndrome. Teratology 55, 177–184. Sovik, S., Lossius, K., Walbe, L., 2001. Heart rate response to transient chemoreceptor stimulation in term infants is modified by exposure to maternal smoking. Pediatr. Res. 49, 558–565. Sullivan, F.M., Barlow, S.M., 2001. Review of risk factors for SIDS. Paediatr. Perinatal Epidemiol. 15, 144–200.