Objective. To investigate the effect of low-flow supplemental oxygen (SupOx) on sleep architecture and cardiorespiratory events in asymptomatic preterm infants.
Methods. An overnight polysomnographic evaluation was conducted prospectively in 23 premature infants who were born at 30.0 ± 3.2 (standard deviation) weeks’ gestational age and studied at 38.1 ± 4.4 weeks’ postconceptional age. Infants were free of any adverse events, including cardiorespiratory monitor alarms in the nursery for at least 1 week before the study. Infants received room air (RA) or SupOx via nasal cannula at 0.25 L/min.
Results. Quiet sleep density was increased during SupOx (33.3 ± 10.8% vs 26.6 ± 10.0% total sleep time [TST] in RA), and a reciprocal decrease in active sleep density was observed (61.5 ± 11.1% vs 68.4 ± 9.9% TST in RA). No differences in sleep efficiency emerged (69.7 ± 10.6% SupOx vs 69.7 ± 8.8% RA). SupOx elicited significant decreases in apnea index (3.8 ± 2.4 events/h vs 11.1 ± 6.4 events/h in RA) and in the percentage of time spent in periodic breathing (1.8 ± 2.9% vs 6.7 ± 8.9% in RA). In addition, SupOx decreased the frequency of bradycardic events (0.3 ± 0.8 events vs 2.5 ± 0.03 events in RA) and improved overall oxygen saturation (98.7 ± 1.1% vs 96.4 ± 2.2%). No changes in alveolar ventilation, as derived from end-tidal CO2 measurements, was detected (38.6 ± 5.8 mm Hg in SupOx vs 38.4 ± 5.4 mm Hg in RA).
Conclusions. Asymptomatic preterm infants exhibit frequent and potentially clinically adverse cardiorespiratory events when assessed in the sleep laboratory. Administration of SupOx to these infants is associated with an increase in the overall duration and percentage TST spent in quiet sleep with reciprocal changes in active sleep. In addition, improvement in respiratory stability is observed with the use of low-flow SupOx, as evidenced by a decrease in apnea, periodic breathing, and bradycardia, without adverse effects on alveolar ventilation.
The effect of supplemental oxygen (SupOx) on cardiorespiratory events and sleep architecture in premature infants has been previously investigated, albeit not extensively. In these few studies, apnea of prematurity and periodic breathing resolve when oxygen concentration is increased to a threshold level.1,2 Rigatto and Brady3 further reported that inhalation of 100% O2 is associated with a decrease in periodic breathing and an increase in minute ventilation. In fact, a modest increase in inspired oxygen concentration decreased apnea and periodicity in preterm infants, not via an increase in alveolar ventilation but rather through a decrease in breath-to-breath variability.4 The effect of SupOx on sleep in infants with chronic neonatal lung disease (CNLD) was also studied; SupOx increased rapid eye movement (REM) sleep and decreased REM arousal in patients with CNLD.5 A later study, however, indicated that higher oxyhemoglobin saturations decreased sleep efficiency but did not modify REM sleep duration or arousal index.6 These somewhat contradictory findings prompted us to investigate how SupOx modifies sleep architecture and cardiorespiratory events during a polysomnographic study of asymptomatic preterm infants.
An overnight polysomnographic evaluation was conducted in asymptomatic premature infants who were free of any adverse events, including cardiorespiratory monitor alarms in the nursery for at least 1 week before the study. All infants were considered stable and were ready for discharge from the nursery by their medical and other caregivers. Infants were excluded from the study when they had significant medical problems, including genetic diseases, craniofacial anomalies, grade 3 and 4 intraventricular hemorrhage, chronic lung disease requiring continuous oxygen supplementation, and neuromuscular diseases. The study was approved by the Committee on Use of Human Subjects at Tulane University Health Science Center, and parental informed consent was obtained.
Both room temperature and infant clothing and bedcovers were identical to those used in the nursery. The 10-hour studies were performed by randomly dividing the study into room air or SupOx via nasal cannula at 0.25 L/min. One group was studied on room air and then followed by SupOx; the second group was initially studied on SupOx and then followed by room air. Normal randomization procedures were applied for group assignment. The SupOx flow rate was set at 0.25 L/min, based on previous studies on the effect of SupOx on sleep in chronic neonatal lung diseases.5,6 This flow delivered a fraction of inspired oxygen (Fio2) of approximately 35% to 45% in infants who weighed between 2.2 and 2.6 kg.7 Studies were considered acceptable only when both portions included at least 4 hours of valid recordings, because it has been previously shown that the reliability of apnea estimation is adequate only when study duration exceeds 3 hours.8 Although no infants received supplemental oxygen before their arrival to the sleep laboratory, room air was discontinued and SupOx instituted when infants developed significant oxygen desaturation episodes.
The polysomnographic studies were done by using Alice 3 Polygraph Recorder (Healthdyne, Marietta, GA). The standard infant montage was used, and the following variables were recorded simultaneously: body position, left and right electrooculogram, 3-channel electroencephalogram (C3A2, C4A1, and O1A2), chin electromyogram, pulse oximetry and pulse wave form, thoracic and abdominal inductance plethysmography, nasal airflow, end-tidal Pco2, and transcutaneous Po2 and Pco2. An accelerometer was also placed on 1 arm to measure gross body movements. Sleep scoring was performed using standard criteria for the scoring of states of sleep and wakefulness in newborn infants using electroencephalogram and electromyogram patterns and behavioral observation.9 The scoring was performed by 2 qualified scorers, who were blinded to the order of oxygen administration. The reliability of the scoring was accepted when there was >90% agreement between the 2 scorers, and data were recorded as the average of the findings. For periods corresponding to room air (RA) and SupOx, the time spent in active sleep (AS), quiet sleep (QS), and indeterminate sleep (IS) was calculated and expressed as percentage of total sleep time (%TST), and sleep efficiency was derived from the ratio between sleep duration as a function of recording duration. Cardiorespiratory variables were scored as apnea index (AI; number of apneic events/hour TST), periodic breathing (expressed in %TST), total number of bradycardic events, average Spo2, and average end-tidal Pco2. The average Spo2 was calculated from the actual reading of all valid Spo2, with Spo2 values during movement artifact being excluded. The average end-tidal Pco2 was obtained from the mean of the highest valid value of end-tidal Pco2 in any given 30-second epoch. Periodic breathing was defined as a series of 3 or more apneic events of at least 3 seconds’ duration, separated by <20 seconds of uninterrupted breathing.10 Apnea (central, obstructive, or mixed) that lasted >5 seconds was scored; however, hypopneic events were not scored. The distribution of different type of apnea was calculated as percentage of total apnea.
Numerical variables were expressed as the mean ± standard deviation, unless otherwise specified. All parameters were assessed and, except for the number of bradycardic events, passed the Kolmogorov and Smirnov normality test.11 Two-tailed t tests were therefore used, whereas the number of bradycardic events was compared by Wilcoxon matched-pairs signed rank test. P < .05 was considered to have achieved statistical significance.
Twenty-three of 28 premature infants had sufficient quality recordings during RA and SupOx to allow analysis. Thirteen were boys. The average gestational age at birth was 30.0 ± 3.2 weeks, and the postconceptional age at the time of polysomnographic study was 38.1 ± 4.4 weeks. Compared with RA, SupOx was associated with a significant increase in QS density (33.3 ± 10.8% vs 26.6 ± 10.0% in RA; P < .001; Fig 1) and a reciprocal decrease in AS density (61.5 ± 11.1% vs 68.4 ± 9.9% in RA; P < .001; Fig 1). No significant differences in the percentage of IS (5.2 ± 2.5% vs 4.6 ± 1.3%; not significant) and sleep efficiency (69.7 ± 10.6% vs 69.7 ± 8.8% RA; not significant) emerged.
SupOx elicited a decrease in AI (3.8 ± 2.5/h vs 11.1 ± 6.4/h in RA; P < .001; Fig 2), as well as a reduction in periodic breathing density (1.8 ± 2.9% vs 6.7 ± 8.9% RA; P < .01; Fig 2). Most of these apneas were central and mixed apnea. No significant changes in the distribution of different type of apnea was observed with the use of SupOx (central apnea: 51.7 ± 18.1% vs 48.9 ± 23.9% RA [not significant]; obstructive apnea: 10.7 ± 8.6% vs 12.0 ± 12.6% RA[not significant]; mixed apnea: 37.5 ± 18.4% vs 39.0 ± 22.8% RA[not significant]). The average Spo2 was increased by SupOx (98.7 ± 1.1% vs 96.4 ± 2.2% RA; P < .001), with no significant differences in the end-tidal CO2 (38.6 ± 5.8 torr vs 38.4 ± 5.4 torr RA; not significant; Fig 2). These findings were corroborated by increased TcPo2 with no changes in TcPco2 levels. A trend toward a decrease in the frequency of bradycardic events was also observed during the use of SupOx (0.3 ± 0.8 vs 1.3 ± 2.5 in RA; P = .055; Fig 2).
This study shows that otherwise healthy premature infants have frequent, unsuspected adverse cardiorespiratory events, including apnea and bradycardia. Administration of low-flow SupOx improved respiratory stability, as evidenced by a decrease in the frequency of respiratory and bradycardic events, a reduction in periodic breathing density, and an increase in overall oxygen saturation without adverse effects on alveolar ventilation. In addition, SupOx modified sleep architecture by increasing QS density and reciprocal decreases in AS density.
It is now well established that preterm infants with or without bronchopulmonary dysplasia may have clinically significant apnea and oxygen desaturation.12–14 In addition, preterm infants have significantly more apneic episodes, desaturation, and periodic breathing when they reach term than infants who are born at term.13 Although some infants may become symptomatic, several studies have pointed out that a large percentage of apnea and bradycardia will not be detected by the nursing staff.14,15 Furthermore, clinical observation alone not only identified significantly less true apnea and bradycardia but also substantially misclassified the type of apnea in these infants.15 The present study further substantiates these findings, particularly because infants who were enrolled in this study were reported as asymptomatic by the nursing staff, whereas the overnight polysomnographic evaluation revealed significant apnea and bradycardia. Stebbens et al16 showed that preterm infants have baseline values of oxyhemoglobin saturation in the same range as those of full-term infants and that punctual measurements may not provide reliable information. In contrast, multichannel polysomnographic recordings may identify clinically asymptomatic infants with significant apnea, although they will not predict the risk for apparent life-threatening events.17
The mechanisms underlying the effect of SupOx on apnea and periodic breathing are unclear. Several putative models of feedback respiratory control would predict increased system stability when heightened peripheral chemoreceptor afferent inputs are reduced by SupOx. Indeed, preterm infants with lower baseline oxyhemoglobin saturations are more prone to have apnea and periodic breathing.18 Furthermore, exposure to an hypoxic gas mixture may lead to prolonged or recurrent apnea in infants.19 Conversely, decreases in apnea density and periodic breathing ensued after inhalation of 100%.20,21 It is interesting that despite the silencing of peripheral chemoreceptor gain induced by SupOx, increases in minute ventilation occurred, suggesting that improved neural tissue oxygenation has a facilitatory effect on overall neural respiratory output.20,21 This latter mechanism may be mediated by increased nitric oxide release in brainstem regions subserving ventilation22 and may account for the emergence of decreased breath-to-breath variability of minute ventilation.4 Thus, our study confirms previous findings indicating that apnea of prematurity can be treated by increasing the Fio2.1 The Fio2 required to eliminate apnea, however, varies between studies. Fenner et al1 found no change in number of apneic episodes until a given “threshold” concentration was achieved, whereas Weintraub et al4 did not identify any evidence for such a threshold, such that regularization of breathing patterns and disappearance of apnea were gradual in their infants. We chose a fixed flow of oxygen and therefore cannot compare our results with other studies because the actual Fio2 was not only variable among the different infants but also varied from moment to moment in the same infant as determined from their instantaneous minute ventilation.7
The most apparent effect on cardiovascular function induced by apnea is the rapid development of bradycardia. Bradycardia may be initiated by hypoxemia or may represent a chemoreceptor-mediated vagal reflex.23 Furthermore, even short-lasting central, obstructive, or mixed apneic events may affect cerebral hemodynamics in preterm infants.24 Although the ultimate long-term adverse consequences of such bradycardic episodes and regional brain flow fluctuations are still unclear, the number of bradycardic events tended to decrease after the initiation of low-flow SupOx.
SupOx not only improved cardiorespiratory stability but also altered sleep architecture. The use of SupOx in preterm infants was associated with an increase in QS with a reciprocal decrease in AS. Our findings concur with those of Fitzgerald et al,6 who showed that elevation of inspired oxygen concentrations in infants with CNLD resulted in a trend toward higher percentage of slow-wave sleep and decreased AS. These authors further speculated that an increased percentage of slow-wave sleep may contribute to improved growth in CNLD infants who maintained higher oxygenation; the latter coincides with normalization of growth hormone urinary excretion.25 The association between improved oxygenation and resultant changes in QS density and accelerated growth would be predicted from the well-known interactions between decreased work of breathing induced by enhanced respiratory stability during SupOx breathing, as well as by the beneficial effect that SupOx would impose on growth hormone end-organ downstream signaling pathway.26,27 Furthermore, QS will enhance growth hormone pulsatile release,28 and increased growth hormone levels will further induce increased QS density.29,30 Thus, increased QS density induced by SupOx not only may ameliorate cardiorespiratory instability but also could lead to improved growth patterns in preterm infants without CNLD.
Multiple concerns have arisen regarding the use of SupOx in preterm infants. First and foremost has been the concern that SupOx will exacerbate retinopathy of prematurity (ROP), because earlier studies showed a correlation between the prolonged administration of a high percentage of oxygen and increased vision loss as a result of ROP.31 More recently, however, the use of SupOx actually has been associated with a decrease in the proliferative vasculopathy of ROP in an animal model32 and with regression of prethreshold ROP in infants.33 These observations on the beneficial effects of SupOx on visual outcome were confirmed recently by a large multicenter study in infants who had prethreshold ROP.34 Second, the use of SupOx was postulated to increase the risk of adverse pulmonary events, including pneumonia and exacerbation of chronic lung disease.34 Although evidence suggests that SupOx may superimpose an oxidant-like injury on disease lung, there is limited data on the effect of SupOx on lung injury in infants with no baseline lung disease. Third, SupOx theoretically could impose an adverse effect on alveolar ventilation by reducing peripheral chemoreceptor function and therefore induce clinically significant hypercapnia. Although such concern has been voiced for many years, our study measured CO2 homeostasis during SupOx administration, and no differences in either the end-tidal CO2 or transcutaneous Pco2 emerged.
Otherwise asymptomatic preterm infants exhibit frequent adverse cardiorespiratory events when they are studied in the sleep laboratory. The use of SupOx was associated with increased quiet sleep and with a decrease in apnea density, time spent in periodic breathing, and number of bradycardic events without any adverse consequences on alveolar ventilation. These findings support the need for a larger multicenter prospective study to examine whether more liberal use of supplemental oxygen may lead to improved cardiorespiratory stability, growth, and/or developmental outcome in these asymptomatic preterm infants.
This study was supported by the American Heart Association Southeast Affiliate (0160230B) and the Constance Kaufman Endowment Fund. Dr Gozal was supported by National Institutes of Health grant HL-65270, Department of Education grant H324E011001, and the Commonwealth of Kentucky Research Challenge Trust Fund.
- ↵Rigatto H, Brady JP. Periodic breathing and apnea in preterm infants. I. Evidence for hypoventilation possibly due to central respiratory depression. Pediatrics.1972;50 :202– 218
- ↵Weintraub Z, Alvaro R, Kwiatkowski K, Cates D, Rigatto H. Effects of inhaled oxygen (up to 40%) on periodic breathing and apnea in preterm infants. J Appl Physiol.1992;72 :116– 120
- ↵Duke JC, Stahl ML, Parrish WE, Rundell OH, Orr WC. Predicting sleep apnea rates in infants. Sleep Res.1984;13 :201
- ↵Anders T, Emde R, Parmelee A, eds. A Manual of Standardized Terminology, Techniques and Criteria for the Scoring of States of Sleep and Wakefulness in Newborn Infants. Los Angeles, CA: UCLA Brain Information Services; 1971
- ↵Kelly DH, Walker AM, Cahen L, Shannon DC. Periodic breathing in siblings of sudden infant death syndrome victims. Pediatrics.1980;66 :515– 520
- ↵Garg M, Kurzner SI, Bautista DB, Keens TG. Clinically unsuspected hypoxia during sleep and feeding in infants with bronchopulmonary dysplasia. Pediatrics.1988;81 :635– 642
- ↵Poets CF, Stebbens VA, Alexander JR, Arrowsmith WA, Salfield SA, Southall DP. Oxygen saturation and breathing patterns in infancy. 2: Preterm infants at discharge from special care. Arch Dis Child.1991;66 :574– 578
- ↵Southall DP, Levitt GA, Richards JM, et al. Undetected episodes of prolonged apnea and severe bradycardia in preterm infants. Pediatrics.1983;72 :541– 551
- ↵Stebbens VA, Poets CF, Alexander JA, Arrowsmith WA, Southall DP. Oxygen saturation and breathing patterns in infancy. 1: Full term infants in the second month of life. Arch Dis Child.1991;66 :569– 573
- ↵Haider AZ, Rehan V, Al Saedi S, et al. Effect of baseline oxygenation on the ventilatory response to inhaled 100% oxygen in preterm infants. J Appl Physiol.1995;79 :2101– 2105
- ↵Parkins KJ, Poets CF, O’Brien LM, Stebbens VA, Southall DP. Effect of exposure to 15% oxygen on breathing patterns and oxygen saturation in infants: interventional study. BMJ.1998;316 :887– 891
- ↵Rigatto H, Brady JP. Periodic breathing and apnea in preterm infants. II. Hypoxia as a primary event. Pediatrics.1972;50 :219– 228
- ↵Gozal D. Potentiation of hypoxic ventilatory response by prior O2 breathing is modulated by nNOS activity in the conscious rat. J Appl Physiol.1998;85 :129– 132
- ↵Henderson-Smart DJ, Butcher-Puech MC, Edwards DA. Incidence and mechanism of bradycardia during apnoea in preterm infants. Arch Dis Child.1986;61 :227– 232
- ↵Van Cauter E, Copinschi G. Interrelationships between growth hormone and sleep. Growth Horm IGF Res.2000;10(suppl B) :S57– S62
- ↵Obal F Jr, Fang J, Taishi P, Kacsoh B, Gardi J, Krueger JM. Deficiency of growth hormone-releasing hormone signaling is associated with sleep alterations in the dwarf rat. J Neurosci.2001;21 :2912– 2918
- ↵Tailoi CL, Gock B, Stone J. Supplemental oxygen therapy. Basis for noninvasive treatment of retinopathy of prematurity. Invest Ophthalmol Vis Sci.1995;36 :1215– 1230
- ↵Supplemental Therapeutic Oxygen for Prethreshold Retinopathy Of Prematurity (STOP-ROP), a randomized, controlled trial. I: primary outcomes. Pediatrics.2000;105 :295– 310
- Copyright © 2002 by the American Academy of Pediatrics