- Email: [email protected]
Clinical correlates of periodic breathing in neonatal polysomnography
Clinical Neurophysiology 115 (2004) 2247–2251 www.elsevier.com/locate/clinph
Clinical correlates of periodic breathing in neonatal polysomnography Andrea J. Oliveirab, Magda L. Nunesa,b,c,d,*, Alaides Fojo-Olmosb, Fernando M. Reise, Jaderson C. da Costaa,b,c a
Division of Neurology, Sa˜o Lucas University Hospital, Pontifı´cia Universidade Cato´lica do Rio Grande do Sul, 90610-000, Porto Alegre, Brazil b Clinical Neurophysiology Laboratory, Sa˜o Lucas University Hospital, Pontifı´cia Universidade Cato´lica do Rio Grande do Sul, 90610-000, Porto Alegre, Brazil c Department of Internal Medicine, Sa˜o Lucas University Hospital, Pontifı´cia Universidade Cato´lica do Rio Grande do Sul, 90610-000, Porto Alegre, Brazil d Department of Pediatrics, Sa˜o Lucas University Hospital, Pontifı´cia Universidade Cato´lica do Rio Grande do Sul, 90610-000, Porto Alegre, Brazil e Department of Obstetrics and Gynecology, UFMG, Belo Horizonte, Brazil Accepted 4 May 2004 Available online 8 June 2004
Abstract Objective: Periodic breathing is a respiratory pattern typical of preterm infants, but its clinical significance has not been clarified yet. The present study was designed to investigate whether the presence of periodic breathing is specifically associated to low post-conceptional ages, preterm birth, or common clinical disorders related to preterm birth. Methods: The study included 271 consecutive infants submitted to neonatal polysomnography, of whom 138 were born before 37 complete gestational weeks (preterm) and 133 were full-term. The main outcome measure was periodic breathing. A multivariate analysis was performed to test the specific impact of preterm birth, respiratory distress syndrome and hypoxic-ischemic encephalopathy on the occurrence of periodic breathing, with adjustment for potential confounding factors such as the post-conceptional age by the time of the polysomnography. Results: Periodic breathing was twice more frequent in infants born before term (83/138 [60%]) than in full-term babies (41/133 [31%], x2 ¼ 22:3; P , 0:0001). The presence of periodic breathing was not significantly associated to either hypoxic-ischemic encephalopathy or respiratory distress syndrome. After a multivariate analysis, only preterm birth remained in the regression model as a specific risk factor for periodic breathing (adjusted odds ratio ¼ 5.62, P , 0:0001). Conclusions: Periodic breathing is a respiratory pattern independently associated with preterm birth, and this association cannot be imputed to the coexistence of either respiratory distress syndrome or hypoxic-ischemic encephalopathy among preterm infants. Significance: The finding of periodic breathing in neonatal polysomnography correlates with preterm birth but not with its associated clinical disorders. q 2004 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. Keywords: Periodic breathing; Neonatal EEG; Polysomnography; Respiratory distress syndrome; Hypoxia
1. Introduction The development of respiratory patterns parallels the ontogeny of sleep stages in the human neonate (Gaultier, 1991). In very young infants, respiratory pauses can be used as a criterion of differentiation between sleep stages as they become more numerous during active (REM) sleep (Curzi-Dascalova and Christova-Gue´orguie´va, 1983). The respiratory pattern of the preterm newborn during sleep is * Corresponding author. Tel./fax: þ 55 51 33 39 4936. E-mail address: [email protected] (M.L. Nunes).
often unstable and characterized by frequent brief apneic episodes and periodic breathing. Periodic breathing is generally defined as groups of respiratory movements interrupted regularly by intervals of apnea, whereas the minimal frequency and duration of apneas found necessary to characterize periodic breathing varies according to different authors (Avery and Fletcher, 1974; Booth et al., 1983; Kelly and Shannon, 1979; Rigatto and Brady, 1972). Periodic breathing is frequent in premature newborns and at 30 weeks of gestational age, it can account for 25% of the respiration time. In full-term newborns, periodic breathing is more common in active sleep until 3 weeks of age, then it starts to decrease.
1388-2457/$30.00 q 2004 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2004.05.007
2248
A.J. Oliveira et al. / Clinical Neurophysiology 115 (2004) 2247–2251
At 2 months, it occurs in less than 3% of total sleep time and at 5 months in less than 1% (Parmelee et al., 1972). While there is general agreement that periodic breathing is more common in preterm than in full-term infants (Albani et al., 1985; Booth et al., 1983; Curzi-Dascalova and Christova-Gue´orguie´va, 1983; Parmelee et al., 1972), the clinical significance of this respiratory pattern remains uncertain, mainly because previous studies used small samples and did not adjust their results for the confounding effect of clinical complications of prematurity, which might affect the maturation of the respiratory system more than preterm birth itself. Only one study has evaluated the effect of respiratory distress syndrome on the occurrence of periodic breathing in preterm and full-term infants (Bentele et al., 1985), but the small numbers precluded any statistical support to the estimated prevalence of periodic breathing in the two groups. The aim of the present study was to investigate whether the presence of periodic breathing is specifically associated to low post-conceptional ages, preterm birth, or common clinical disorders related to preterm birth such as respiratory distress syndrome and hypoxic-ischemic encephalopathy.
2. Patients and methods 2.1. Patient population and study design The study included 271 consecutive infants submitted to neonatal polysomnography at the EEG Laboratory of our University Hospital. All infants were born at or admitted to the nursery or to the Neonatal Intensive Care Unit (NICU) of our hospital and were examined after parental consent. Clinical data obtained on the newborns from review of hospital charts included: any neonatal disorder or complication identified during hospital admission, gestational age, conceptional age at the moment of the polysomnographic study, Apgar scores on the 1st and 5th minutes, neuroradiological data whenever available and data of general physical and neurological examinations. Some clinical characteristics of the population studied are summarized in Table 1. The study protocol was approved by the Institutional Review Board. Gestational age refers to the time between onset of the mother’s last menstrual period and the infant’s birth. Conceptional age refers to the gestational age plus the age in weeks and days from birth. Preterm birth was defined as delivery before 37 complete weeks of gestation. Respiratory distress syndrome was defined as an acute illness, usually of preterm infants, characterized clinically by a respiratory rate over 60/min, dyspnoea, with a predominantly diaphragmatic pattern and chest X-ray with a reticulogranular appearance as a result of widespread atelectasis (Greenough and Roberton, 1999). Hypoxic-ischemic encephalopathy was defined by the presence of clearcut neurological signs of neonatal encephalopathy with documented birth asphyxia
Table 1 Birth characteristics of the study population ðn ¼ 271Þ
Gestational age (weeks) Conceptional age (weeks) Apgar score at 1 min Apgar score at 5 min Sex (proportion of males)
Full-term ðn ¼ 133Þ
Preterm ðn ¼ 138Þ
Significancea
39 (38–40) 40 (39–42) 8 (5–9) 9 (8–9) 53%
34 (32–35) 37 (35–39) 7 (5–8) 8 (8–9) 47%
P , 0:001 P , 0:001 P , 0:05 P , 0:05 P ¼ 0:372
Data are expressed as medians and interquartile ranges. Mann –Whitney U test for ages and Apgar scores and x 2 test for proportion of males/females. a
(depressed Apgar scores, cord blood acidosis) and/or neuroradiological evidence of hypoxic-ischemic lesions. 2.2. Polysomnographic procedures The neonatal polysomnographies were recorded in the neurophysiology laboratory, in warmed cribs, usually between noon and 3 p.m., in the interval between two meals. The duration of the recording was of at least 60 min—in those cases showing undetermined sleep only, or lacking a sleep organization—or until a complete sleep cycle was recorded, which happened in most cases (Albani et al., 1985; Bentele et al., 1985). Ten scalp electrodes were placed according to a modified International 10 –20 system, with double electrode distance, in the positions F1-2, C3-4, T3-4, O1-2, Fz, and Cz (Anders et al., 1971). A 13-channel Toshiba, or a 16-channel Berger machine were used to record at least 8 channels of EEG in a standard bipolar montage, and 5 extracerebral monitoring channels, including submental electromyogram, eye movement, electrocardiogram, nasal thermistor, and thoracic respiratory movements recorded by a strain gauge, with a paper speed of 15 mm/s. All polysomnographies were reviewed blindly, with no knowledge of the clinical data of the patient. These assessments were performed independently by two examiners, who scored the presence of periodic breathing with a 98% interrater agreement (k ¼ 0:949; P , 0:001). The sleep stages (REM and NREM) were identified by electroencephalographic, respiratory, eye movement and behavioral parameters (Nunes et al., 1997). 2.3. Definition of periodic breathing Periodic breathing was defined by the presence of at least 3 consecutive apneas of at least 3 s each, separated by intervals not longer than 20 s (Kelly and Shannon, 1979). Recorded data included the total duration of the tracing, duration of each sleep stage, duration of each periodic breathing episode, the sleep stage in which they occurred, and the percentages of the whole tracing and of the specific sleep stages during which periodic breathing was present.
A.J. Oliveira et al. / Clinical Neurophysiology 115 (2004) 2247–2251
2249
Fig. 1. Representative example of neonatal polysomnography showing periodic breathing in REM sleep (speed 15 mm/s).
An illustrative example of periodic breathing is presented in Fig. 1. 2.4. Statistical analysis As a preliminary analysis, the possible association between periodic breathing and preterm birth or complications like respiratory distress syndrome or hypoxicischemic encephalopathy was tested by contingence tables and x 2 tests with continuity correction. After calculating individual odds ratios with 95% confidence intervals, a multivariate analysis was performed to test the specific impact of preterm birth, respiratory distress syndrome and hypoxic-ischemic encephalopathy on the occurrence of periodic breathing. The method used was stepwise forward logistic regression. Adequacy of sample size was checked using the package Epi-Info 6.04b (CDC Atlanta) with power set at 0.8 and significance level at P , 0:05:
(x 2 ¼ 2:80; p ¼ 0:09). Conversely, the prevalence of periodic breathing was not increased in the newborns complicated by hypoxic-ischemic encephalopathy (21/36 [58%], Table 2). On an average, the newborns spent 53% of their sleep time on REM sleep (mean ^ SD ¼ 47 ^ 19 min), 40% on NREM sleep (36 ^ 16 min) and 7% on REM – NREM transition. Overall, the proportion of the sleep time spent on periodic breathing was 35%. The proportions of REM sleep and NREM sleep time spent on periodic breathing did not differ significantly in full-term and preterm infants and were not influenced by the presence of either respiratory distress syndrome or hypoxic-ischemic encephalopathy (Table 2). Table 3 summarizes the results of univariate and multivariate analyses performed to assess whether preterm Table 2 Prevalence of periodic breathing in the study population and proportion of each sleep phase spent on periodic breathing Number of infants having periodic breathing
3. Results The overall prevalence of periodic breathing in our study population was 46%. While the time of complete disappearance of this respiratory pattern has not been documented by the present study due to the cross-sectional design, periodic breathing was still present in 2/8 (25%) of the children examined at 48 weeks of conceptional age. As shown in Table 2, periodic breathing was twice more frequent in infants born before term (83/138 [60%]) than in full-term babies (41/133 [31%], x 2 ¼ 22:3; P , 0:0001). Among the infants born preterm, periodic breathing was particularly more prevalent in those complicated by respiratory distress syndrome (21/29 [72%]) than in preterm counterparts without perinatal complications (21/40 [53%]) but this difference was not statistically significant
All infants ðn ¼ 271Þ Born full-term ðn ¼ 133Þ Born preterm ðn ¼ 138Þ Preterm, uncomplicated ðn ¼ 40Þc Preterm with RDS ðn ¼ 29Þ Preterm with HIE ðn ¼ 36Þ
124 (46%) 41 (31%) 83 (60%) 21 (53%) 21 (72%) 21 (58%)
Proportion of each sleep phase spent on periodic breathinga,b REM
NREM
33 30 36 51
(17–60) (13–60) (23–62) (27–70)
13 (0–50) 3 (0–27) 21 (0–62) 29 (3–79)
32 (13–47) 27 (13–56)
20 (4–55) 16 (3–62)
RDS, respiratory distress syndrome; HIE, hypoxic-ischemic encephalopathy. a Excluding the infants who did not present any periodic breathing. b Data are expressed as percentages (median and interquartile range). c Without any kind of perinatal complication except prematurity.
2250
A.J. Oliveira et al. / Clinical Neurophysiology 115 (2004) 2247–2251
Table 3 Potential risk factors for periodic breathing Risk factor
Odds ratio
95% CI
Adjusted ORa
P value
Preterm birth Conceptional age (per week) Respiratory distress syndrome Hypoxic-ischemic encephalopathy
3.39 –
2.05–5.59 –
5.62 1.11
,0.0001 0.028
4.31
1.85–10.01
Not included
.0.05
0.75
0.45–1.22
Not included
.0.05
Data are expressed as odds ratios (OR) and 95% confidence intervals (CI) for univariate analysis, or as adjusted odds ratios for multivariate analysis. The method used was a stepwise forward logistic regression. a Adjusted for conceptional age, sex, Apgar scores, and for the potential risk factors listed in the table.
birth, respiratory distress syndrome and hypoxic-ischemic encephalopathy are specific risk factors for periodic breathing. The putative risk factors emerging from univariate analysis were preterm birth (odds ratio ¼ 3.39) and respiratory distress syndrome (odds ratio ¼ 4.31). However, after a multivariate analysis with adjustment for potentially confounding variables, including conceptional age, only preterm birth remained in the regression model as a specific risk factor for periodic breathing (adjusted odds ratio ¼ 5.62, P , 0:0001).
4. Discussion The present study was designed to identify the clinical correlates of periodic breathing in neonatal polysomnography. Two clinical findings, namely preterm birth and respiratory distress syndrome, were grossly associated to periodic breathing. However, only preterm birth proved to be specifically associated to periodic breathing after adjustment for potential confounding variables. Thus, the apparent relationship between respiratory distress syndrome and periodic breathing seems to be only coincidental, since this syndrome is more frequent in premature babies and prematurity itself appears to be the true risk factor for periodic breathing. As far as we know, only one study by Bentele et al. (1985) has previously assessed the relationship between respiratory distress syndrome and periodic breathing. Their findings suggested that the preterm infants affected by the syndrome are less prone to have periodic breathing than their healthy, preterm counterparts, but this was not statistically confirmed. The fact that periodic breathing was significantly more prevalent among infants born before term even being recorded at a mean conceptional age of 37 weeks suggests that this respiratory pattern may persist at term as a sequel of preterm birth and/or some condition related to it. Our results show that the prevalence of periodic breathing is not affected by neonatal hypoxic-ischemic encephalopathy,
which is a frequent complication of prematurity. Thus, doubt remains on which mechanism explains the effect of prematurity per se on the persistence of periodic breathing at later conceptional ages. The mechanisms involved in the generation of periodic breathing include an increased latency for activation of peripheral chemoreceptors, leading to short periods of apnea and mild hypoxemia alternated with a strong respiratory drive to restore pulmonary ventilation (Khoo et al., 1982). In the event that preterm birth is the primary reason by which preterm infants persist having periodic breathing at advanced conceptional ages, one possible interpretation to this finding is that preterm infants have a slower maturation of respiration control due to their precocious exposure to the extrauterine environment. This hypothesis, however, remains to be tested. The characteristics of respiratory patterns in preterm infants have been evaluated in the past, but some conflicting data are difficult to resolve due to the variety of methods used to define and quantify respiratory pauses (Albani et al., 1985; Booth et al., 1983; Curzi-Dascalova and ChristovaGue´orguie´va, 1983). In agreement, these studies have suggested that periodic breathing is a common finding in non-complicated preterm infants, not necessarily of pathological significance. Supporting this view, in the present study, we have identified episodes of periodic breathing in over 50% of the subgroup of preterm infants without clinical complications. We have also observed that, whichever be the gestational age at birth or the health status, the children who had periodic breathing spent nearly the same percentage of sleep time on it. This finding suggests that, at least for the clinical correlates, quantifying this phenomenon may not add relevant information to what is obtained by the simple identification of periodic breathing using the standard criteria applied here. Despite its high prevalence in preterm infants without neonatal complications, periodic breathing has been suspected to be a risk factor for sudden infant death syndrome (SIDS). Kelly and Shannon (1979) recorded respiration and heart rate in term infants with near-miss SIDS and concluded that periodic breathing is present in excessive amounts during sleep in near-miss infants, compared to healthy controls. However, this finding has not been confirmed by Guilleminaut et al. (1979), who monitored full-term infants with near-miss SIDS for 24 h and found only an increased number of mixed and obstructive apneas, whereas the number of periodic breathing events was not altered, compared to healthy controls. A definite conclusion about the relationship between periodic breathing and the risk of SIDS is hampered by the rarity of this syndrome in some countries, which limits the power of cohort studies, combined with the lack of information on the neonatal respiration patterns in case-control studies. In conclusion, periodic breathing is a respiration pattern independently associated with preterm birth, and this association cannot be imputed to the coexistence of either respiratory distress syndrome or hypoxic-ischemic
A.J. Oliveira et al. / Clinical Neurophysiology 115 (2004) 2247–2251
encephalopathy among preterm infants. Thus, the finding of periodic breathing in neonatal polysomnography correlates with preterm birth but not with its associated clinical disorders.
References Albani M, Bentele KHP, Budde C, Schulte FJ. Infant sleep apnea profile: preterm vs. term infants. Eur J Pediatr 1985;143:261–8. Anders T, Emde K, Parmelee AH. A manual of standardized terminology, techniques and criteria for scoring of states of sleep and wakefulness in newborns infants. Los Angeles: UCLA Brain Information Service; 1971. Avery ME, Fletcher BD. The lung and its disorders in the newborn infant. Philadelphia: Saunders; 1974. Bentele KHP, Albani M, Budde C, Schulte FJ. Sleep apnea profile in preterm infants recovering from respiratory distress syndrome. Arch Dis Child 1985;60:547 –54. Booth CL, Morin VN, Weite SP, Thoman EB. Periodic and non-periodic sleep apnea in premature and full-term infants. Dev Med Child Neurol 1983;25:283 –96. Curzi-Dascalova L, Christova-Gue´orguie´va E. Respiratory pauses in normal prematurely born infants. Biol Neonate 1983;44:325– 32.
2251
Gaultier C. Respiratory adaptation during sleep in infants and children: risk factors. In: Peter JH, Podszus T, von Wichert P, editors. Sleep and healthy risk. Berlin: Springer; 1991. p. 399–407. Greenough A, Roberton NRC. Acute respiratory disesase in the newborn (part 2). In: Rennie JM, Roberton NRC, editors. Textbook of neonatology, 3rd ed. Edinburgh: Churchill Livingstone; 1999. p. 481– 607. Guilleminaut C, Ariagno R, Korobkin R, Nagel L, Baldwin R, Coons S, Owen M. Mixed and obstructive sleep apnea and near miss for sudden infant death syndrome. 2. Comparison of near miss and normal control infants by age. Pediatrics 1979;64:882–91. Kelly DH, Shannon DC. Periodic breathing in infants with near miss sudden infant death syndrome. Pediatrics 1979;63:355 –60. Khoo MC, Kronauer RC, Strohl KP, Slutsky AS. Factors inducing periodic breathing in humans: a general model. J Appl Physiol 1982; 53:644–59. Nunes ML, Da Costa JC, Moura-Ribeiro MV. Polysomnographic quantification of bioelectrical maturation in preterm and full-term newborns at matched conceptional ages. Electroencephalogr Clin Neurophysiol 1997;102:186 –91. Parmelee AH, Stern E, Harris MA. Maturation of respiration in prematures and young infants. Neuropaediatrics 1972;3:294–304. Rigatto H, Brady JP. Periodic breathing and apnea in preterm infants. I. Evidence for hypoventilation possibly due to central respiratory depression. Pediatrics 1972;50:202–17.
Clinical correlates of periodic breathing in neonatal polysomnography Andrea J. Oliveirab, Magda L. Nunesa,b,c,d,*, Alaides Fojo-Olmosb, Fernando M. Reise, Jaderson C. da Costaa,b,c a
Division of Neurology, Sa˜o Lucas University Hospital, Pontifı´cia Universidade Cato´lica do Rio Grande do Sul, 90610-000, Porto Alegre, Brazil b Clinical Neurophysiology Laboratory, Sa˜o Lucas University Hospital, Pontifı´cia Universidade Cato´lica do Rio Grande do Sul, 90610-000, Porto Alegre, Brazil c Department of Internal Medicine, Sa˜o Lucas University Hospital, Pontifı´cia Universidade Cato´lica do Rio Grande do Sul, 90610-000, Porto Alegre, Brazil d Department of Pediatrics, Sa˜o Lucas University Hospital, Pontifı´cia Universidade Cato´lica do Rio Grande do Sul, 90610-000, Porto Alegre, Brazil e Department of Obstetrics and Gynecology, UFMG, Belo Horizonte, Brazil Accepted 4 May 2004 Available online 8 June 2004
Abstract Objective: Periodic breathing is a respiratory pattern typical of preterm infants, but its clinical significance has not been clarified yet. The present study was designed to investigate whether the presence of periodic breathing is specifically associated to low post-conceptional ages, preterm birth, or common clinical disorders related to preterm birth. Methods: The study included 271 consecutive infants submitted to neonatal polysomnography, of whom 138 were born before 37 complete gestational weeks (preterm) and 133 were full-term. The main outcome measure was periodic breathing. A multivariate analysis was performed to test the specific impact of preterm birth, respiratory distress syndrome and hypoxic-ischemic encephalopathy on the occurrence of periodic breathing, with adjustment for potential confounding factors such as the post-conceptional age by the time of the polysomnography. Results: Periodic breathing was twice more frequent in infants born before term (83/138 [60%]) than in full-term babies (41/133 [31%], x2 ¼ 22:3; P , 0:0001). The presence of periodic breathing was not significantly associated to either hypoxic-ischemic encephalopathy or respiratory distress syndrome. After a multivariate analysis, only preterm birth remained in the regression model as a specific risk factor for periodic breathing (adjusted odds ratio ¼ 5.62, P , 0:0001). Conclusions: Periodic breathing is a respiratory pattern independently associated with preterm birth, and this association cannot be imputed to the coexistence of either respiratory distress syndrome or hypoxic-ischemic encephalopathy among preterm infants. Significance: The finding of periodic breathing in neonatal polysomnography correlates with preterm birth but not with its associated clinical disorders. q 2004 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. Keywords: Periodic breathing; Neonatal EEG; Polysomnography; Respiratory distress syndrome; Hypoxia
1. Introduction The development of respiratory patterns parallels the ontogeny of sleep stages in the human neonate (Gaultier, 1991). In very young infants, respiratory pauses can be used as a criterion of differentiation between sleep stages as they become more numerous during active (REM) sleep (Curzi-Dascalova and Christova-Gue´orguie´va, 1983). The respiratory pattern of the preterm newborn during sleep is * Corresponding author. Tel./fax: þ 55 51 33 39 4936. E-mail address: [email protected] (M.L. Nunes).
often unstable and characterized by frequent brief apneic episodes and periodic breathing. Periodic breathing is generally defined as groups of respiratory movements interrupted regularly by intervals of apnea, whereas the minimal frequency and duration of apneas found necessary to characterize periodic breathing varies according to different authors (Avery and Fletcher, 1974; Booth et al., 1983; Kelly and Shannon, 1979; Rigatto and Brady, 1972). Periodic breathing is frequent in premature newborns and at 30 weeks of gestational age, it can account for 25% of the respiration time. In full-term newborns, periodic breathing is more common in active sleep until 3 weeks of age, then it starts to decrease.
1388-2457/$30.00 q 2004 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2004.05.007
2248
A.J. Oliveira et al. / Clinical Neurophysiology 115 (2004) 2247–2251
At 2 months, it occurs in less than 3% of total sleep time and at 5 months in less than 1% (Parmelee et al., 1972). While there is general agreement that periodic breathing is more common in preterm than in full-term infants (Albani et al., 1985; Booth et al., 1983; Curzi-Dascalova and Christova-Gue´orguie´va, 1983; Parmelee et al., 1972), the clinical significance of this respiratory pattern remains uncertain, mainly because previous studies used small samples and did not adjust their results for the confounding effect of clinical complications of prematurity, which might affect the maturation of the respiratory system more than preterm birth itself. Only one study has evaluated the effect of respiratory distress syndrome on the occurrence of periodic breathing in preterm and full-term infants (Bentele et al., 1985), but the small numbers precluded any statistical support to the estimated prevalence of periodic breathing in the two groups. The aim of the present study was to investigate whether the presence of periodic breathing is specifically associated to low post-conceptional ages, preterm birth, or common clinical disorders related to preterm birth such as respiratory distress syndrome and hypoxic-ischemic encephalopathy.
2. Patients and methods 2.1. Patient population and study design The study included 271 consecutive infants submitted to neonatal polysomnography at the EEG Laboratory of our University Hospital. All infants were born at or admitted to the nursery or to the Neonatal Intensive Care Unit (NICU) of our hospital and were examined after parental consent. Clinical data obtained on the newborns from review of hospital charts included: any neonatal disorder or complication identified during hospital admission, gestational age, conceptional age at the moment of the polysomnographic study, Apgar scores on the 1st and 5th minutes, neuroradiological data whenever available and data of general physical and neurological examinations. Some clinical characteristics of the population studied are summarized in Table 1. The study protocol was approved by the Institutional Review Board. Gestational age refers to the time between onset of the mother’s last menstrual period and the infant’s birth. Conceptional age refers to the gestational age plus the age in weeks and days from birth. Preterm birth was defined as delivery before 37 complete weeks of gestation. Respiratory distress syndrome was defined as an acute illness, usually of preterm infants, characterized clinically by a respiratory rate over 60/min, dyspnoea, with a predominantly diaphragmatic pattern and chest X-ray with a reticulogranular appearance as a result of widespread atelectasis (Greenough and Roberton, 1999). Hypoxic-ischemic encephalopathy was defined by the presence of clearcut neurological signs of neonatal encephalopathy with documented birth asphyxia
Table 1 Birth characteristics of the study population ðn ¼ 271Þ
Gestational age (weeks) Conceptional age (weeks) Apgar score at 1 min Apgar score at 5 min Sex (proportion of males)
Full-term ðn ¼ 133Þ
Preterm ðn ¼ 138Þ
Significancea
39 (38–40) 40 (39–42) 8 (5–9) 9 (8–9) 53%
34 (32–35) 37 (35–39) 7 (5–8) 8 (8–9) 47%
P , 0:001 P , 0:001 P , 0:05 P , 0:05 P ¼ 0:372
Data are expressed as medians and interquartile ranges. Mann –Whitney U test for ages and Apgar scores and x 2 test for proportion of males/females. a
(depressed Apgar scores, cord blood acidosis) and/or neuroradiological evidence of hypoxic-ischemic lesions. 2.2. Polysomnographic procedures The neonatal polysomnographies were recorded in the neurophysiology laboratory, in warmed cribs, usually between noon and 3 p.m., in the interval between two meals. The duration of the recording was of at least 60 min—in those cases showing undetermined sleep only, or lacking a sleep organization—or until a complete sleep cycle was recorded, which happened in most cases (Albani et al., 1985; Bentele et al., 1985). Ten scalp electrodes were placed according to a modified International 10 –20 system, with double electrode distance, in the positions F1-2, C3-4, T3-4, O1-2, Fz, and Cz (Anders et al., 1971). A 13-channel Toshiba, or a 16-channel Berger machine were used to record at least 8 channels of EEG in a standard bipolar montage, and 5 extracerebral monitoring channels, including submental electromyogram, eye movement, electrocardiogram, nasal thermistor, and thoracic respiratory movements recorded by a strain gauge, with a paper speed of 15 mm/s. All polysomnographies were reviewed blindly, with no knowledge of the clinical data of the patient. These assessments were performed independently by two examiners, who scored the presence of periodic breathing with a 98% interrater agreement (k ¼ 0:949; P , 0:001). The sleep stages (REM and NREM) were identified by electroencephalographic, respiratory, eye movement and behavioral parameters (Nunes et al., 1997). 2.3. Definition of periodic breathing Periodic breathing was defined by the presence of at least 3 consecutive apneas of at least 3 s each, separated by intervals not longer than 20 s (Kelly and Shannon, 1979). Recorded data included the total duration of the tracing, duration of each sleep stage, duration of each periodic breathing episode, the sleep stage in which they occurred, and the percentages of the whole tracing and of the specific sleep stages during which periodic breathing was present.
A.J. Oliveira et al. / Clinical Neurophysiology 115 (2004) 2247–2251
2249
Fig. 1. Representative example of neonatal polysomnography showing periodic breathing in REM sleep (speed 15 mm/s).
An illustrative example of periodic breathing is presented in Fig. 1. 2.4. Statistical analysis As a preliminary analysis, the possible association between periodic breathing and preterm birth or complications like respiratory distress syndrome or hypoxicischemic encephalopathy was tested by contingence tables and x 2 tests with continuity correction. After calculating individual odds ratios with 95% confidence intervals, a multivariate analysis was performed to test the specific impact of preterm birth, respiratory distress syndrome and hypoxic-ischemic encephalopathy on the occurrence of periodic breathing. The method used was stepwise forward logistic regression. Adequacy of sample size was checked using the package Epi-Info 6.04b (CDC Atlanta) with power set at 0.8 and significance level at P , 0:05:
(x 2 ¼ 2:80; p ¼ 0:09). Conversely, the prevalence of periodic breathing was not increased in the newborns complicated by hypoxic-ischemic encephalopathy (21/36 [58%], Table 2). On an average, the newborns spent 53% of their sleep time on REM sleep (mean ^ SD ¼ 47 ^ 19 min), 40% on NREM sleep (36 ^ 16 min) and 7% on REM – NREM transition. Overall, the proportion of the sleep time spent on periodic breathing was 35%. The proportions of REM sleep and NREM sleep time spent on periodic breathing did not differ significantly in full-term and preterm infants and were not influenced by the presence of either respiratory distress syndrome or hypoxic-ischemic encephalopathy (Table 2). Table 3 summarizes the results of univariate and multivariate analyses performed to assess whether preterm Table 2 Prevalence of periodic breathing in the study population and proportion of each sleep phase spent on periodic breathing Number of infants having periodic breathing
3. Results The overall prevalence of periodic breathing in our study population was 46%. While the time of complete disappearance of this respiratory pattern has not been documented by the present study due to the cross-sectional design, periodic breathing was still present in 2/8 (25%) of the children examined at 48 weeks of conceptional age. As shown in Table 2, periodic breathing was twice more frequent in infants born before term (83/138 [60%]) than in full-term babies (41/133 [31%], x 2 ¼ 22:3; P , 0:0001). Among the infants born preterm, periodic breathing was particularly more prevalent in those complicated by respiratory distress syndrome (21/29 [72%]) than in preterm counterparts without perinatal complications (21/40 [53%]) but this difference was not statistically significant
All infants ðn ¼ 271Þ Born full-term ðn ¼ 133Þ Born preterm ðn ¼ 138Þ Preterm, uncomplicated ðn ¼ 40Þc Preterm with RDS ðn ¼ 29Þ Preterm with HIE ðn ¼ 36Þ
124 (46%) 41 (31%) 83 (60%) 21 (53%) 21 (72%) 21 (58%)
Proportion of each sleep phase spent on periodic breathinga,b REM
NREM
33 30 36 51
(17–60) (13–60) (23–62) (27–70)
13 (0–50) 3 (0–27) 21 (0–62) 29 (3–79)
32 (13–47) 27 (13–56)
20 (4–55) 16 (3–62)
RDS, respiratory distress syndrome; HIE, hypoxic-ischemic encephalopathy. a Excluding the infants who did not present any periodic breathing. b Data are expressed as percentages (median and interquartile range). c Without any kind of perinatal complication except prematurity.
2250
A.J. Oliveira et al. / Clinical Neurophysiology 115 (2004) 2247–2251
Table 3 Potential risk factors for periodic breathing Risk factor
Odds ratio
95% CI
Adjusted ORa
P value
Preterm birth Conceptional age (per week) Respiratory distress syndrome Hypoxic-ischemic encephalopathy
3.39 –
2.05–5.59 –
5.62 1.11
,0.0001 0.028
4.31
1.85–10.01
Not included
.0.05
0.75
0.45–1.22
Not included
.0.05
Data are expressed as odds ratios (OR) and 95% confidence intervals (CI) for univariate analysis, or as adjusted odds ratios for multivariate analysis. The method used was a stepwise forward logistic regression. a Adjusted for conceptional age, sex, Apgar scores, and for the potential risk factors listed in the table.
birth, respiratory distress syndrome and hypoxic-ischemic encephalopathy are specific risk factors for periodic breathing. The putative risk factors emerging from univariate analysis were preterm birth (odds ratio ¼ 3.39) and respiratory distress syndrome (odds ratio ¼ 4.31). However, after a multivariate analysis with adjustment for potentially confounding variables, including conceptional age, only preterm birth remained in the regression model as a specific risk factor for periodic breathing (adjusted odds ratio ¼ 5.62, P , 0:0001).
4. Discussion The present study was designed to identify the clinical correlates of periodic breathing in neonatal polysomnography. Two clinical findings, namely preterm birth and respiratory distress syndrome, were grossly associated to periodic breathing. However, only preterm birth proved to be specifically associated to periodic breathing after adjustment for potential confounding variables. Thus, the apparent relationship between respiratory distress syndrome and periodic breathing seems to be only coincidental, since this syndrome is more frequent in premature babies and prematurity itself appears to be the true risk factor for periodic breathing. As far as we know, only one study by Bentele et al. (1985) has previously assessed the relationship between respiratory distress syndrome and periodic breathing. Their findings suggested that the preterm infants affected by the syndrome are less prone to have periodic breathing than their healthy, preterm counterparts, but this was not statistically confirmed. The fact that periodic breathing was significantly more prevalent among infants born before term even being recorded at a mean conceptional age of 37 weeks suggests that this respiratory pattern may persist at term as a sequel of preterm birth and/or some condition related to it. Our results show that the prevalence of periodic breathing is not affected by neonatal hypoxic-ischemic encephalopathy,
which is a frequent complication of prematurity. Thus, doubt remains on which mechanism explains the effect of prematurity per se on the persistence of periodic breathing at later conceptional ages. The mechanisms involved in the generation of periodic breathing include an increased latency for activation of peripheral chemoreceptors, leading to short periods of apnea and mild hypoxemia alternated with a strong respiratory drive to restore pulmonary ventilation (Khoo et al., 1982). In the event that preterm birth is the primary reason by which preterm infants persist having periodic breathing at advanced conceptional ages, one possible interpretation to this finding is that preterm infants have a slower maturation of respiration control due to their precocious exposure to the extrauterine environment. This hypothesis, however, remains to be tested. The characteristics of respiratory patterns in preterm infants have been evaluated in the past, but some conflicting data are difficult to resolve due to the variety of methods used to define and quantify respiratory pauses (Albani et al., 1985; Booth et al., 1983; Curzi-Dascalova and ChristovaGue´orguie´va, 1983). In agreement, these studies have suggested that periodic breathing is a common finding in non-complicated preterm infants, not necessarily of pathological significance. Supporting this view, in the present study, we have identified episodes of periodic breathing in over 50% of the subgroup of preterm infants without clinical complications. We have also observed that, whichever be the gestational age at birth or the health status, the children who had periodic breathing spent nearly the same percentage of sleep time on it. This finding suggests that, at least for the clinical correlates, quantifying this phenomenon may not add relevant information to what is obtained by the simple identification of periodic breathing using the standard criteria applied here. Despite its high prevalence in preterm infants without neonatal complications, periodic breathing has been suspected to be a risk factor for sudden infant death syndrome (SIDS). Kelly and Shannon (1979) recorded respiration and heart rate in term infants with near-miss SIDS and concluded that periodic breathing is present in excessive amounts during sleep in near-miss infants, compared to healthy controls. However, this finding has not been confirmed by Guilleminaut et al. (1979), who monitored full-term infants with near-miss SIDS for 24 h and found only an increased number of mixed and obstructive apneas, whereas the number of periodic breathing events was not altered, compared to healthy controls. A definite conclusion about the relationship between periodic breathing and the risk of SIDS is hampered by the rarity of this syndrome in some countries, which limits the power of cohort studies, combined with the lack of information on the neonatal respiration patterns in case-control studies. In conclusion, periodic breathing is a respiration pattern independently associated with preterm birth, and this association cannot be imputed to the coexistence of either respiratory distress syndrome or hypoxic-ischemic
A.J. Oliveira et al. / Clinical Neurophysiology 115 (2004) 2247–2251
encephalopathy among preterm infants. Thus, the finding of periodic breathing in neonatal polysomnography correlates with preterm birth but not with its associated clinical disorders.
References Albani M, Bentele KHP, Budde C, Schulte FJ. Infant sleep apnea profile: preterm vs. term infants. Eur J Pediatr 1985;143:261–8. Anders T, Emde K, Parmelee AH. A manual of standardized terminology, techniques and criteria for scoring of states of sleep and wakefulness in newborns infants. Los Angeles: UCLA Brain Information Service; 1971. Avery ME, Fletcher BD. The lung and its disorders in the newborn infant. Philadelphia: Saunders; 1974. Bentele KHP, Albani M, Budde C, Schulte FJ. Sleep apnea profile in preterm infants recovering from respiratory distress syndrome. Arch Dis Child 1985;60:547 –54. Booth CL, Morin VN, Weite SP, Thoman EB. Periodic and non-periodic sleep apnea in premature and full-term infants. Dev Med Child Neurol 1983;25:283 –96. Curzi-Dascalova L, Christova-Gue´orguie´va E. Respiratory pauses in normal prematurely born infants. Biol Neonate 1983;44:325– 32.
2251
Gaultier C. Respiratory adaptation during sleep in infants and children: risk factors. In: Peter JH, Podszus T, von Wichert P, editors. Sleep and healthy risk. Berlin: Springer; 1991. p. 399–407. Greenough A, Roberton NRC. Acute respiratory disesase in the newborn (part 2). In: Rennie JM, Roberton NRC, editors. Textbook of neonatology, 3rd ed. Edinburgh: Churchill Livingstone; 1999. p. 481– 607. Guilleminaut C, Ariagno R, Korobkin R, Nagel L, Baldwin R, Coons S, Owen M. Mixed and obstructive sleep apnea and near miss for sudden infant death syndrome. 2. Comparison of near miss and normal control infants by age. Pediatrics 1979;64:882–91. Kelly DH, Shannon DC. Periodic breathing in infants with near miss sudden infant death syndrome. Pediatrics 1979;63:355 –60. Khoo MC, Kronauer RC, Strohl KP, Slutsky AS. Factors inducing periodic breathing in humans: a general model. J Appl Physiol 1982; 53:644–59. Nunes ML, Da Costa JC, Moura-Ribeiro MV. Polysomnographic quantification of bioelectrical maturation in preterm and full-term newborns at matched conceptional ages. Electroencephalogr Clin Neurophysiol 1997;102:186 –91. Parmelee AH, Stern E, Harris MA. Maturation of respiration in prematures and young infants. Neuropaediatrics 1972;3:294–304. Rigatto H, Brady JP. Periodic breathing and apnea in preterm infants. I. Evidence for hypoventilation possibly due to central respiratory depression. Pediatrics 1972;50:202–17.