Respiratory Control in Neonatal Rats Exposed to Prenatal Cigarette Smoke Jonathan D. Pendlebury1, Richard J. A. Wilson2, Shehr Bano1, Kathleen J. Lumb1, Jennifer M. Schneider1, and Shabih U. Hasan1 1 Department of Pediatrics and 2Department of Physiology and Biophysics, Institute of Maternal and Child Health, Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada

Rationale: Prenatal cigarette smoke (CS) exposure, increased environmental temperature, and hypoxic episodes have been postulated as major risk factors for sudden infant death syndrome. Objectives: To test the hypothesis that maternal CS exposure disrupts eupneic breathing and depresses breathing responses of neonatal rats to thermal and hypoxic challenges. Methods: Experiments were performed on 1-week-old rat pups exposed prenatally to CS (n 5 39) or room air (sham; n 5 30). Breathing patterns were recorded by whole-body plethysmography during thermoneutral or hyperthermic states under normoxic and hypoxic conditions. Measurements and Main Results: Mean pup weight, breaths per minute, and gasping respiratory patterns were measured for both smoke- and sham-exposed groups during thermoneutral and hyperthermic states under normoxic and hypoxic conditions. Under thermoneutral conditions, hypoxia caused gasping in CS-exposed animals but not in sham-exposed animals. Furthermore, under hyperthermic conditions, whereas hypoxia induced gasping in both groups, only CSexposed animals exhibited a pronounced and longer lasting respiratory depression after the termination of hypoxia. Conclusions: We show that prenatal CS exposure increases the likelihood of gasplike respiration and provide the first experimental evidence that the combined effects of prenatal CS exposure and hyperthermia dramatically prolong the time required for neonates to return to eupneic breathing after hypoxia. These observations provide important evidence of how prenatal CS exposure, hypoxic episodes, and hyperthermia might place infants at higher risk for sudden infant death syndrome. Keywords: apnea; hyperthermia; hypoxia; nicotine; sudden infant death syndrome

Pre- and postnatal cigarette smoke (CS) exposure is currently the principal independent risk factor for the occurrence of sudden infant death syndrome (SIDS), which remains the leading cause of infant mortality after the first month of life (1, 2). On the basis of the strength and consistency of the association, CS exposure has been proposed as a causal factor in SIDS (1). However, experimental studies on the effects of prenatal CS exposure on neonatal respiratory control in both humans and animal models to date have provided divergent results (3–10).

AT A GLANCE COMMENTARY Scientific Knowledge on the Subject

Hypoxic episodes and increased body temperature have been proposed as possible triggers for sudden infant death syndrome (SIDS). How prenatal cigarette smoke exposure, the leading risk factor for SIDS, affects respiratory responses to these triggers is unknown. What This Study Adds to the Field

Increased ambient temperature and hypoxia adversely affect breathing patterns in neonates exposed prenatally to cigarette smoke. Addressing these risk factors through tobacco reduction programs and better infant care practices could potentially decrease the incidence of SIDS.

SIDS is known to occur more frequently during the winter months. Therefore, thermal stress from low or high ambient temperature and/or an increased infant body temperature due to overwrapping have been proposed as additional major risk factor for SIDS. Nonetheless, the evidence, although well analyzed and precisely documented, is largely inferential, based on interviews of the caregivers and calculation of environmental temperature or mathematical modeling (11, 12). Franco and coworkers suggested that an increased ambient temperature decreases infant arousability in REM sleep during the latter half of the night (13). Increased occurrence of SIDS in the prone sleeping position may also point toward decreased heat dissipation (higher body temperature) and a greater chance of upper airway obstruction leading to hypoxia and hypercarbia (14). However, to our knowledge, no experimental evidence exists to confirm or refute the deleterious interaction between the effects of prenatal CS exposure, increased body temperature, and hypoxia on neonatal respiratory control. The specific aim of the current study was to investigate the effects of hypoxia and/or hyperthermia on respiratory control in neonatal rats exposed to prenatal CS. Specifically, we tested the hypothesis that maternal CS exposure alters the pattern of breathing of neonatal rats in response to thermal and hypoxic challenges.

(Received in original form November 26, 2007; accepted in final form February 22, 2008) Supported by the Canadian Institutes of Health Research, the Alberta Heritage Foundation for Medical Research, and the Canadian Foundation for the Study of Infant Deaths. Correspondence and requests for reprints should be addressed to Shabih U. Hasan, M.D., Health Sciences Center, 3330 Hospital Drive NW, Calgary, AB, T2N 4N1, Canada. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Am J Respir Crit Care Med Vol 177. pp 1255–1261, 2008 Originally Published in Press as DOI: 10.1164/rccm.200711-1739OC on February 28, 2008 Internet address: www.atsjournals.org

METHODS The experimental protocols were approved by the Animal Care Committee of the University of Calgary (Calgary, AB, Canada) and all studies were performed in accordance with the Guide to the Care and Use of Experimental Animals provided by the Canadian Council on Animal Care.

Animals Experiments were performed on 69 time-dated, 1-week-old rat pups from 21 pregnant Sprague-Dawley rats (11 cigarette smoke and 10 sham

1256

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 177

exposed; Figure 1A). Pregnancy was confirmed by the presence of a vaginal mucus plug and the following day was considered as Gestational Day 1. Postnatal Day 7 was chosen because by this age, unsuccessful cardiorespiratory transition from fetal to neonatal life becomes apparent and gross congenital anomalies can be ruled out. Furthermore, feeding is well established and the hemoglobin levels begin to decline as observed in human infants (15). This age corresponds to approximately 6 months in human infants (16), a postnatal age still within the SIDS risk group.

CS Exposure CS exposure was designed to mimic that in moderate to heavy human cigarette smokers as closely as possible. Exposure to CS was initiated on Day 1 of pregnancy and continued until the day of birth (Gestational Day 22). Research cigarettes (2R1) were used throughout. Mainstream smoke was delivered via the nose-only smoke exposure system (Kentucky Tobacco Research and Development Center, University of Kentucky, Lexington, KY). The mainstream smoke from a single cigarette was diluted with room air and distributed among the eight pregnant rats at an hourly interval, 10 times per day. Ten puffs were obtained from a single cigarette over a period of 10 minutes and the puff volume was 10 ml per puff. Therefore, the total daily CS volume was 1,000 ml 5 10 ml puff volume 3 10 puffs per cigarette 3 10 cigarettes per day and was distributed equally among eight pregnant rats (125 ml per animal per day). Using this regimen, plasma nicotine concentrations (20–60 ng/ml) and maternal carboxyhemoglobin levels were similar to those observed in moderate to heavy human smokers. Further, fetal weight reductions were proportional to those of human infants born to smoking mothers (17). The sham group was treated identically to the CS group and was placed in the smoke exposure system, but received puffs of room air instead of CS. See the online supplement for additional information about methods.

Whole-Body Plethysmography Breathing patterns were recorded in unanesthetized, spontaneously breathing 1-week-old pups, using a continuous-flow, unrestrained, whole-body plethysmograph composed of two 60-ml syringes (animal and reference chambers). The ambient temperature was controlled to within 18C. Ambient temperatures of 32–338C and 37–388C were used to define thermoneutral and hyperthermic environments, respectively (15). Body temperature (Tb) was recorded during the study with a rectal thermometer and a computer-based data acquisition system (resolution, 0.018C; type T thermocouple; OMEGA Engineering, Stamford, CT). The whole-body plethysmograph was continuously flushed with air, creating a positive flow through the system (bias flow regulator; Buxco Research Systems, Wilmington, NC) and the changing of the gases from room air to hypoxia (10% O2, balance nitrogen) and vice versa was made with no interruption. A differential pressure transducer (Buxco Research Systems) was used to record the pressure difference between the two chambers, a surrogate measure of changes in lung volume. In small animals, the pressure signal is a consequence mainly of the rarefaction and compression of gas associated with moving air in and out of the thorax, whereas in larger animals the signal is dominated by humidifica-

2008

tion and warming of inspired gas (18). In the Day 7 animals, the pressure signal likely results from both of these sources, precluding the use of the signal to derive tidal volume. Whereas any tidal volume measurements would be affected by changes in the temperature of the air in the chamber, the frequency is unaffected and the waveform, although different in amplitude, is likely to maintain the same shape. Because changes in air temperature could affect the waveform, in any given experiment the air temperature in the recording chamber was maintained at a constant level throughout the baseline, hypoxic, and washout periods (Figure 1B).

Experimental Protocol The experimental protocol is given in Figure 1B. Each pup was randomly selected and weighed before the experiment. No pup was used for more than one experiment. All studies were performed between 09:00 and 15:00 hours to avoid any influence of the circadian rhythm (15, 19). The number of pups in both CS and sham groups under various experimental conditions are given in Figures 1A and 1B.

Data Analysis Pup weight was determined for both the CS-exposed and sham groups at birth and on Day 7. The frequency of breathing (number of breaths per minute) was calculated from the whole-body plethysmograph signal (Buxco Research Systems) for each minute of recording. Gasping, defined as a rapid inspiratory rise with a long expiratory phase and preceded as well as followed by cessation of breathing movements (20), was identified by the breathing signal (inspiratory and expiratory times) and visually.

Statistical Analysis An independent sample t test was performed to determine the differences in body weight between animals in the CS and sham groups. The frequency of breathing and rectal temperature under various experimental conditions was analyzed by analysis of variance for repeated measures. Further analysis of variance was done to investigate whether there were any differences between the two groups and whether those differences were dependent on the experimental condition. Paired t tests with Bonferroni correction were then performed to detail those interactions. To determine whether there was an increased incidence of gasping in a specific group, the percentage of gaspers from each group and treatment was analyzed by Fisher exact test. All data are presented as means 6 SEM and P values equal to or less than 0.05 were considered statistically significant.

RESULTS Pup Weights

Pup weight in the CS-exposed group was significantly lower at birth and on Day 7 as compared with the sham group (P , 0.001 for both age groups; Figure 2). Figure 1. (A) Number of pups in each experimental group. Of 69 pups, 39 were exposed to cigarette smoke (CS) and 30 were sham treated. The CS- and sham-exposed groups were further divided into thermoneutral and hyperthermic hypoxia groups. (B) Experimental protocol. Animals were placed in the whole-body plethysmograph for 10 minutes to allow for acclimatization (settling time). Baseline measurements were recorded over a 5-minute period (baseline). The air in the chamber was replaced with hypoxic gas (10% O2) and breathing patterns were recorded for 5 minutes (hypoxia). The hypoxic gas was replaced with room air and breathing patterns were recorded for 10 minutes (washout). Thermoneutral and hyperthermic ambient temperatures remained constant throughout the baseline, hypoxic, and washout periods.

Pendlebury, Wilson, Bano, et al.: Smoking and Infant Respiratory Control

1257

and 3B). Rectal temperature increased in correspondence with the increase in ambient temperatures (P , 0.001; Figure 3B). Eupneic Breathing Pattern

Figure 2. Pup weight at birth and Day 7. Pup weight at birth and on Day 7 was significantly lower in the cigarette smoke (CS)–exposed group (solid columns) as compared with the sham group (open columns) (*P , 0.001, CS vs. sham for both age groups).

Rectal Temperature

Rectal temperatures of CS- and sham-exposed animals under various experimental conditions are given in Figure 3. Rectal temperatures corresponded to the target ambient temperature within the chamber to 18C. No significant difference in rectal temperatures was observed between CS- and sham-exposed animals under thermoneutral or hyperthermic states during baseline, hypoxic, or washout periods (P > 0.05; Figures 3A

Figure 3. Rectal temperature during thermoneutral and hyperthermic hypoxia. (A) Rectal temperature during thermoneutral hypoxia. There was no difference in rectal temperature (8C) between the sham group (open columns) and the cigarette smoke (CS) group (solid columns) during the baseline, hypoxic, and washout periods under thermoneutral conditions (P . 0.05). (B) Rectal temperature during hyperthermic hypoxia. There was no difference in rectal temperature (8C) between the sham group (open columns) and the CS group (solid columns) during the baseline, hypoxic, and washout periods under hyperthermic conditions (P . 0.05). Rectal temperature was higher during the hyperthermic versus thermoneutral state in both sham and CS groups under baseline, hypoxic, and washout periods (*P , 0.001).

Thermoneutral hypoxia (n 5 36). Under thermoneutral conditions, the baseline breathing frequency was similar between the sham-exposed group (n 5 16) and the CS-exposed group (n 5 20). The breathing frequency showed a comparable increase in both sham and CS groups (P 5 0.003 and , 0.001, respectively) within 1–2 minutes of the onset of hypoxic challenge (Figure 4A). The breathing rate declined with continuation of hypoxia as well as during the washout period (Figure 4A). The decrease in frequency of breathing was more pronounced within the CS group (P , 0.001). However, no significant difference was observed between the two groups. Hyperthermic hypoxia (n 5 33). The breathing frequency was higher in the sham group as compared with the CS-exposed group under baseline hyperthermic conditions (P 5 0.001). The breathing frequency increased during the first 2 minutes of hypoxic exposure in the CS group only (P , 0.001), and the mean peak frequency was not different between the two groups

Figure 4. (A) Breathing responses during thermoneutral hypoxia. There was no difference in the breathing frequency between the CS group (solid squares) and the sham group (open circles) during the baseline recordings. Breathing frequency increased in both groups at the onset of hypoxia. In the cigarette smoke (CS)–exposed group, breathing frequency remained significantly lower during the washout period compared with the baseline values (*P , 0.05, experimental vs. baseline within CS group; †P , 0.05, experimental vs. baseline within sham group). (B) Breathing responses during hyperthermic hypoxia. The breathing frequency was higher in the sham group (open circles) as compared with the CS-exposed group (solid squares) under baseline conditions. Breathing frequency increased in the CS group at the onset of hypoxia (P , 0.01). The decrease in breathing frequency during hypoxic challenge was greater in the CS group as compared with the sham group. During the washout phase, recovery of breathing frequency was slower in the CS-exposed group versus the sham group. (*P , 0.05, experimental vs. baseline within CS group; †P , 0.05, experimental vs. baseline within sham group; ‡P , 0.05 CS-exposed vs. sham group). X axis: 1 5 5 minutes, 21% O2; 2 5 1–2 minutes, 10% O2; 3 5 2–3 minutes, 10% O2; 4 5 3–4 minutes, 10% O2; 5 5 4–5 minutes, 10% O2; 6 5 0–1 minute, 21% O2; 7 5 1–2 minutes, 21% O2; 8 5 2–3 minutes, 21% O2; 9 5 3–6 minutes, 21% O2; 10 5 6–10 minutes, 21% O2.

1258

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 177

2008

P , 0.0001 and 0.0006, respectively). However, during the washout period, recovery was slower and/or incomplete in the gaspers of the CS group as compared with the gaspers in the sham group (P 5 0.0001 within the group and P 5 0.01 between the groups; Figures 7A and 7B).

DISCUSSION

Figure 5. Percentage of pups exhibiting gasping. The percentage of pups exhibiting gasping during thermoneutral hypoxia as well as the total number of pups exhibiting gasping during hypoxia was significantly higher (*P 5 0.05) in the cigarette smoke–exposed group (solid columns) versus the sham group (open columns).

(Figure 4B). The breathing frequency decreased in both groups during the washout period (P < 0.03 for both groups). However, the decrease was greater in CS-exposed pups as compared with the sham group (P 5 0.009–0.001; Figure 4B). Gasping Breathing Pattern

Overall, 13% of the sham-treated animals and 36% of the CSexposed animals exhibited gasping breathing patterns (P 5 0.05; Figure 5). None of the sham-treated animals and 25% of the CS-exposed animals gasped while in the thermoneutral hypoxic state (P 5 0.05). Under hyperthermic hypoxic conditions, the gasping pattern was observed in 29 versus 47% in the sham group versus CS-exposed animals, respectively (Figure 5). The time from the start of the hypoxic challenge to the onset of gasping was not different between the two groups. However, the time spent gasping was significantly longer in the CS group (P , 0.05). The eupneic and gasping breathing patterns are illustrated in Figure 6. Under baseline hyperthermic conditions, the breathing frequency was similar between the animals that gasped (gaspers) and those that did not gasp (nongaspers) within the CS or sham group (P 5 0.2 and 0.4, respectively; Figures 7A and 7B). The breathing frequency decreased more during hypoxia in gaspers as compared with the nongaspers in both groups (CS and sham;

We provide the first direct experimental evidence that a combination of prenatal CS exposure and hyperthermia has detrimental effects on respiratory control in neonates. Specifically, we show that prenatal CS exposure significantly affects the pattern of breathing in response to hypoxia under thermoneutral conditions, increasing the occurrence of gasping during hypoxia and impairing the newborn’s ability to recover to eupneic breathing after hypoxia. Importantly, we also show that the protracted effects of prenatal CS exposure on this recovery are greatly exacerbated by hyperthermia. Our approach differed fundamentally from previous animal studies, which used nicotine to mimic the effects of CS, as we have employed a standardized animal model of CS exposure. Thus our results provide some of the most direct evidence to date suggesting that prenatal CS exposure can contribute to the destabilizing effects of hypoxia and thermal stress on neonatal breathing. Consequently, these results suggest a causal role for prenatal CS exposure in SIDS, adding weight to epidemiologic studies and further supporting the efforts to foster prenatal smoking cessation programs. It is well established that prenatal CS exposure reduces fetal growth and birth weight in a dose-dependent manner (21). The mechanisms of reduced birth weight in CS-exposed fetuses have been discussed elsewhere and evidence suggests that CS constituents other than nicotine mediate such effects (17, 22). In addition, prior work suggests that infants born to light- and heavy-smoking mothers may not achieve body weights comparable to infants born to nonsmoking mothers until 6 months of age (23). The mechanisms of the slower postnatal growth could include protracted effects of more than 4,700 toxins found in cigarette smoke (24), decreased fat deposition rates (25), and reduced maternal milk production (26). In our study, rats born to CS-exposed animals had reduced birth weights compared with the sham group. Furthermore, the CS-exposed group did not achieve body weights similar to those in the sham group during the study period. The effects of maternal CS exposure have previously been investigated in human infants whereas continuous nicotine infusion has been used as a surrogate for prenatal CS exposure

Figure 6. Breathing responses during hypoxic and hyperthermic challenges. (A) Thermoneutral hypoxia (cigarette smoke [CS] exposed). Baseline eupneic breathing turned into complex (multiphasic) gasping during hypoxia but returned toward eupneic breathing during washout. The respiratory cycle (inspiration [Insp.] and expiration [Exp.]) and time bar (2 s) applies to all panels (A–D). (B) Thermoneutral hypoxia (sham). Baseline eupneic breathing was unaffected by hypoxia and washout conditions. (C) Hyperthermic hypoxia (CS exposed). Baseline eupneic breathing turned into multiphasic gasping during hypoxia and evolved into low-amplitude gasping during washout. (D) Hyperthermic hypoxia (sham). Baseline eupneic breathing turned into gasping during hypoxia and returned to eupneic breathing during washout.

Pendlebury, Wilson, Bano, et al.: Smoking and Infant Respiratory Control Figure 7. Frequency of breathing during hyperthermia: gaspers (solid squares) versus nongaspers (solid circles). Nongaspers had similar breathing patterns during baseline, hypoxia, and washout in both sham and cigarette smoke (CS)–exposed groups. (A) Sham gaspers versus nongaspers. After a decrease during hypoxia, the breathing frequency returned toward baseline values during the washout period in sham gaspers (*P , 0.05, between gaspers and nongaspers within the group). (B) CS-exposed gaspers versus nongaspers. Gaspers in the CS-exposed group had a slower and/or incomplete recovery during the washout phase when compared with nongaspers within the CS group as well as gaspers in the sham group (*P , 0.05, between gaspers and nongaspers within the group; †P , 0.05 between CS and sham group gaspers). X axis: 1 5 baseline (21% O2); 2 5 hypoxia (10% O2); 3 5 washout (21% O2).

in a number of animal models (27). Both human data (3–5, 28, 29) and animal data (6–10) provide divergent results. Studies in humans by Sovik and coworkers, and Poole and coworkers, observed no significant difference in various ventilatory responses to hypoxia between the infants of smokers versus nonsmokers (3, 4). In contrast to these observations, Ueda and coworkers showed that infants of smoking mothers as compared with those born to nonsmoking mothers had lower respiratory drive during normoxia and blunted ventilatory responses to hypoxia (5). However, in this latter study infants were sedated with a large dose of chloral hydrate. Yet another study showed the ventilatory responses to be paradoxically higher in CS-exposed infants, and arousal thresholds to be similar between the CSexposed and control infants (28). Lewis and Bosque observed similar ventilatory responses but deficient arousal responses to hypoxia in CS-exposed infants as compared with control infants (29). The likely reasons for such divergent and often contradictory results include use of sedation, age of the infants, extent of CS exposure, and composition of gaseous mixtures for hypoxic exposures (3, 4, 28, 29). Similar to the studies in human infants, animal data have also been inconsistent (6–10). Bamford and coworkers, and Schuen and coworkers, observed no significant effect on ventilatory responses to moderate and severe hypoxia after gestational exposure to a large nicotine dose (6, 7). Fewell and Smith did not observe an effect on the time to last gasp in 5- to 6–day-old rat pups, but prenatal nicotine exposure did impair their ability to autoresuscitate (8). In an ovine model, Hafstro¨m and coworkers showed minute ventilation to be similar and inspiratory drive to be higher in nicotine-exposed animals as compared with the control group (9). Another study by the same group of investigators showed the latency for arousal to be longer and minute ventilation to be lower after moderate hypoxic challenge during quiet sleep in nicotine-exposed neonatal lambs (10). It is interesting to note that the nicotine dose and plasma concentrations in the latter two studies were modest and mimicked the concentrations observed in pregnant women, suggesting sensitivity of the ovine perinate to nicotine (9, 10). In our current study, both the

1259

CS-exposed and sham groups showed a qualitatively similar biphasic response to a hypoxic challenge under thermoneutral conditions: an initial increase followed by a decrease and protracted recovery on return to normoxia. However, recovery was significantly longer in the CS group as compared with the sham group. Furthermore, the CS group exhibited gasping respiration, indicating loss of eupneic breathing patterns. The unequivocal adverse effects of CS exposure on respiratory control in our newly standardized pregnant rat model suggest that at least in rodents, the constituents of CS are more injurious than nicotine infusion alone as observed in previous studies (6, 7). Thermal stress induced by both over- and underwrapping of infants as well as by high and low ambient temperatures may be associated with SIDS (2, 12, 30, 31). Furthermore, evidence suggests that such stress is independent of prone sleep position (30). The evidence of thermal stress in human infants is inferential as it is based on the examination of the death scene, average daily temperatures, and mathematical calculations (12, 30, 31). However, animal experimental data do suggest that high core temperature adversely affects the survival rate, time to last gasp, and ability to autoresuscitate during anoxic (97% N2 and 3% CO2) challenge (32, 33). In our studies, hyperthermia was induced by increasing ambient temperature. Even moderate hypoxia (10% O2) during hyperthermic states resulted in failure of eupneic breathing in the sham group, further corroborating the previous studies. However, the current study provides the unique observation that prenatal CS exposure has detrimental effects on eupneic breathing patterns under hypoxic thermoneutral and hyperthermic conditions. Our experimental data confirm the inferential and epidemiologic evidence that moderate hypoxia and increased ambient temperature disrupt the normal breathing patterns in the absence of CS exposure. In addition, prenatal CS exposure further enhances the failure of eupneic breathing. Despite intense research, much controversy exists regarding the neurogenesis of eupneic breathing and gasping (34, 35). Nonetheless, gasping is generally accepted as the most robust during hypoxia and a backup mechanism for restoration of regular breathing patterns and cardiorespiratory homeostasis (36, 37). In a study by Poets and coworkers, seven of nine infants manifested gasping respiration in predeath recordings even though the gasping response remained ineffective in restoring cardiorespiratory homeostasis (20). In another study, almost all infants exhibited hypoxic gasping before their death; however, SIDS infants had higher occurrence of complex gasps and decreased autoresuscitation as compared with the non-SIDS infants (38). In our current study, a significant number of CS-exposed rat pups exhibited gasping during moderate hypoxic challenge, indicating deleterious effects of prenatal CS exposure on postnatal respiratory control and the failure of normal breathing pattern. Various definitions have been used to define gasping (20, 38–40). In the current study, we defined gasping in a manner similar to that used by Poets and coworkers: ‘‘presence of rapid inspiratory rise with a retarded expiratory phase preceded and followed by a cessation of breathing movements’’ (20). Sridhar and coworkers have shown ‘‘hyperpneic’’ breaths preceding gasping with similar waveform as seen during gasping (38). However, such terminal ‘‘hyperpnea’’ does not include multiple complex gasps as illustrated in our animals. Furthermore, cessation of breathing or breathing pauses does not appear to be the consistent feature of hyperpneic breaths as observed during gasping. The respiratory waveform in our current study closely resembles that described as gasping by Sridhar and coworkers, comprising multiple and complex gasping patterns (38). Although the recording methods differ between our study and the study by Sridhar and coworkers, both methods transduce changes in lung volume and thus generate similar waveforms.

1260

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 177

According to the conventional definition, cessation of breathing movements (or primary apnea) precedes gasping. However, there is no absolute definition of apnea, especially when comparing different species with dramatically different baseline breathing rates. In human neonates at term gestation, the baseline breathing rate is approximately 30 breaths/minute and apnea is generally defined as cessation of airflow or breathing pauses greater than 20 seconds. However, in an SIDS infant, breathing pauses of approximately 10 seconds have been denoted, quite appropriately, as primary apnea (38). In neonatal rats, the breathing frequency is approximately 170 breaths/minute. Thus, it is reasonable to consider a similar phenomenon in rats lasting about 2 seconds. This is in the same order of magnitude as the interval between the events we define as gasping. The temporal relationship between terminal ‘‘hyperpnea,’’ apnea, and gasping needs to be investigated in future studies. Although disturbances of cardiorespiratory homeostasis and impaired arousal have been postulated, the precise relationship between SIDS and maternal smoking remains unclear (41–43). In an American aboriginal population, 3H-nicotinic receptor binding showed no change in SIDS infants exposed to CS, whereas a significant reduction was observed among the control infants (44). Similarly, in another study, nicotine receptor binding was not affected in SIDS infants exposed to CS (45); however, 3Hnicotinic receptor binding was upregulated in control infants exposed to CS. These observations suggest that CS exposure alters nicotinic receptor binding in SIDS infants as opposed to control infants. The observed differences in nicotinic receptor binding between the infants in the two studied control groups may reflect the demographic differences between the two study populations (44, 45). Furthermore, other neurotransmitter systems including serotonin and various isoforms of protein kinase C have been reported to be abnormal in SIDS infants and animals exposed to prenatal CS, respectively (36, 46, 47). In view of the evidence, further studies are warranted to investigate the interaction between nicotinic and serotonergic receptors (48). In conclusion, we provide the first experimental evidence that prenatal CS exposure decreases the tolerance to moderate hypoxia and that thermal stress further disrupts the eupneic breathing resulting in gasping respiration. The ‘‘Back to Sleep’’ campaign has resulted in marked reduction in SIDS rates. However, intense advocacy through public awareness for the cessation of maternal smoking during and after pregnancy, and avoidance of thermal stress, is vital to further reduce the incidence of this devastating tragedy. Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Acknowledgment: The authors thank Dr. Tak Fung, Anita Rigaux, and Linda Brigan for statistical, technical, and editorial assistance, respectively.

References 1. Blair PS, Sidebotham P, Berry PJ, Evans M, Fleming PJ. Major epidemiological changes in sudden infant death syndrome: a 20-year population-based study in the UK. Lancet 2006;367:314–319. 2. Moon RY, Horne RS, Hauck FR. Sudden infant death syndrome. Lancet 2007;370:1578–1587. 3. Sovik S, Lossius K, Eriksen M, Grogaard J, Walloe L. Development of oxygen sensitivity in infants of smoking and non-smoking mothers. Early Hum Dev 1999;56:217–232. 4. Poole KA, Hallinan H, Beardsmore CS, Thompson JR. Effect of maternal smoking on ventilatory responses to changes in inspired oxygen levels in infants. Am J Respir Crit Care Med 2000;162:801–807. 5. Ueda Y, Stick SM, Hall G, Sly PD. Control of breathing in infants born to smoking mothers. J Pediatr 1999;135:226–232. 6. Bamford OS, Schuen JN, Carroll JL. Effect of nicotine exposure on postnatal ventilatory responses to hypoxia and hypercapnia. Respir Physiol 1996;106:1–11.

2008

7. Schuen JN, Bamford OS, Carroll JL. The cardiorespiratory response to anoxia: normal development and the effect of nicotine. Respir Physiol 1997;109:231–239. 8. Fewell JE, Smith FG. Perinatal nicotine exposure impairs ability of newborn rats to autoresuscitate from apnea during hypoxia. J Appl Physiol 1998;85:2066–2074. 9. Hafstro¨m O, Milerad J, Sundell HW. Altered breathing pattern after prenatal nicotine exposure in the young lamb. Am J Respir Crit Care Med 2002;166:92–97. 10. Hafstro¨m O, Milerad J, Sundell HW. Prenatal nicotine exposure blunts the cardiorespiratory response to hypoxia in lambs. Am J Respir Crit Care Med 2002;166:1544–1549. 11. Anonymous. Prone, hot, and dead. Lancet 1990;336:1104. 12. Williams SM, Taylor BJ, Mitchell EA; National Cot Death Study Group. Sudden infant death syndrome: insulation from bedding and clothing and its effect modifiers. Int J Epidemiol 1996;25:366–375. 13. Franco P, Scaillet S, Valente F, Chabanski S, Groswasser J, Kahn A. Ambient temperature is associated with changes in infants’ arousability from sleep. Sleep 2001;24:325–329. 14. Tuffnell CS, Petersen SA, Wailoo MP. Prone sleeping infants have a reduced ability to lose heat. Early Hum Dev 1995;43:109–116. 15. Mortola JP. Respiratory physiology of newborn mammals: a comparative perspective. Baltimore, MD: Johns Hopkins University Press; 2001. 16. Quinn R. Comparing rat’s to human’s age: how old is my rat in people years? Nutrition 2005;21:775–777. 17. Farkas S, Hussein J, Ariano RE, Sitar DS, Hasan SU. Prenatal cigarette smoke exposure: pregnancy outcome and gestational changes in plasma nicotine concentration, hematocrit, and carboxyhemoglobin in a newly standardized rat model. Toxicol Appl Pharmacol 2006;214: 118–125. 18. Cummings KJ, Pendlebury JD, Sherwood NM, Wilson RJ. Sudden neonatal death in PACAP-deficient mice is associated with reduced respiratory chemoresponse and susceptibility to apnoea. J Physiol 2004;555:15–26. 19. Stephenson R, Liao KS, Hamrahi H, Horner RL. Circadian rhythms and sleep have additive effects on respiration in the rat. J Physiol 2001; 536:225–235. 20. Poets CF, Meny RG, Chobanian MR, Bonofiglo RE. Gasping and other cardiorespiratory patterns during sudden infant deaths. Pediatr Res 1999;45:350–354. 21. Salihu HM, Wilson RE. Epidemiology of prenatal smoking and perinatal outcomes. Early Hum Dev 2007;83:713–720. 22. Hussein J, Farkas S, MacKinnon Y, Ariano RE, Sitar DS, Hasan SU. Nicotine dose–concentration relationship and pregnancy outcomes in rat: biologic plausibility and implications for future research. Toxicol Appl Pharmacol 2007;218:1–10. 23. Conter V, Cortinovis I, Rogari P, Riva L. Weight growth in infants born to mothers who smoked during pregnancy. BMJ 1995;310:768–771. 24. U.S.Environmental Protection Agency, Office of Research and Development, Office of Health and Environmental Assessment. Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders pp. 44–96. Washington, DC: U.S. Government Printing Office; 1992. Publication No. EPA/600/6-90/006F. 25. Bernstein IM, Plociennik K, Stahle S, Badger GJ, Secker-Walker R. Impact of maternal cigarette smoking on fetal growth and body composition. Am J Obstet Gynecol 2000;183:883–886. 26. Hopkinson JM, Schanler RJ, Fraley JK, Garza C. Milk production by mothers of premature infants: influence of cigarette smoking. Pediatrics 1992;90:934–938. 27. Dempsey DA, Benowitz NL. Risks and benefits of nicotine to aid smoking cessation in pregnancy. Drug Saf 2001;24:277–322. 28. Campbell AJ, Galland BC, Bolton DP, Taylor BJ, Sayers RM, Williams SM. Ventilatory responses to rebreathing in infants exposed to maternal smoking. Acta Paediatr 2001;90:793–800. 29. Lewis KW, Bosque EM. Deficient hypoxia awakening response in infants of smoking mothers: possible relationship to sudden infant death syndrome. J Pediatr 1995;127:691–699. 30. Fleming PJ, Gilbert R, Azaz Y, Berry PJ, Rudd PT, Stewart A, Hall E. Interaction between bedding and sleeping position in the sudden infant death syndrome: a population based case–control study. BMJ 1990;301:85–89. 31. Ponsonby A, Dwyer T, Gibbons LE, Cochrane JA, Jones ME, McCall MJ. Thermal environment and sudden infant death syndrome: case– control study. BMJ 1992;304:277–282. 32. Kahraman L, Thach BT. Inhibitory effects of hyperthermia on mechanisms involved in autoresuscitation from hypoxic apnea in mice: a model for thermal stress causing SIDS. J Appl Physiol 2004;97:669–674.

Pendlebury, Wilson, Bano, et al.: Smoking and Infant Respiratory Control 33. Serdarevich C, Fewell JE. Influence of core temperature on autoresuscitation during repeated exposure to hypoxia in normal rat pups. J Appl Physiol 1999;87:1346–1353. 34. Ramirez JM, Garcia A 3rd. Point: medullary pacemaker neurons are essential for both eupnea and gasping in mammals vs. medullary pacemaker neurons are essential for gasping, but not eupnea, in mammals. J Appl Physiol 2007;103:717–718. 35. Paton JFR, St. John WM. Counterpoint: medullary pacemaker neurons are essential for gasping, but not eupnea, in mammals. J Appl Physiol 2007;103:718–720. 36. Thach BT. The role of respiratory control disorders in SIDS. Respir Physiol Neurobiol 2005;149:343–353. 37. Gozal D, Gozal E, Reeves SR, Lipton AJ. Gasping and autoresuscitation in the developing rat: effect of antecedent intermittent hypoxia. J Appl Physiol 2002;92:1141–1144. 38. Sridhar R, Thach BT, Kelly DH, Henslee JA. Characterization of successful and failed autoresuscitation in human infants, including those dying of SIDS. Pediatr Pulmonol 2003;36:113–122. 39. Gozal D, Torres JE, Gozal YM, Nuckton TJ. Characterization and developmental aspects of anoxia-induced gasping in the rat. Biol Neonate 1996;70:280–288. 40. St John WM. Neurogenesis, control, and functional significance of gasping. J Appl Physiol 1990;68:1305–1315. 41. Harper RM. Impaired arousals and sudden infant death syndrome: preexisting neural injury? Am J Respir Crit Care Med 2003;168:1262– 1263.

1261 42. Harper RM, Kinney HC, Fleming PJ, Thach BT. Sleep influences on homeostatic functions: implications for sudden infant death syndrome. Respir Physiol 2000;119:123–132. 43. Hunt CE, Hauck FR. Sudden infant death syndrome. CMAJ 2006;174: 1861–1869. 44. Duncan JR, Randall LL, Belliveau RA, Trachtenberg FL, Randall B, Habbe D, Mandell F, Welty TK, Iyasu S, Kinney HC. The effect of maternal smoking and drinking during pregnancy upon 3H-nicotine receptor brainstem binding in infants dying of the sudden infant death syndrome: initial observations in a high risk population. Brain Pathol 2008;18:21–31. 45. Nachmanoff DB, Panigrahy A, Filiano JJ, Mandell F, Sleeper LA, Valdes-Dapena M, Krous HF, White WF, Kinney HC. Brainstem 3Hnicotine receptor binding in the sudden infant death syndrome. J Neuropathol Exp Neurol 1998;57:1018–1025. 46. Paterson DS, Trachtenberg FL, Thompson EG, Belliveau RA, Beggs AH, Darnall R, Chadwick AE, Krous HF, Kinney HC. Multiple serotonergic brainstem abnormalities in sudden infant death syndrome. JAMA 2006;296:2124–2132. 47. Hasan SU, Simakajornboon N, MacKinnon Y, Gozal D. Prenatal cigarette smoke exposure selectively alters protein kinase C and nitric oxide synthase expression within the neonatal rat brainstem. Neurosci Lett 2001;301:135–138. 48. Rand CM, Berry-Kravis EM, Zhou L, Fan W, Weese-Mayer DE. Sudden infant death syndrome: rare mutation in the serotonin system FEV gene. Pediatr Res 2007;62:180–182.