Eurly Humun Devuhpmeni,

Elsevier Scientific

26 (1991)

1-l

Ireland

Ltd.

Publishers

2

EHD 01151

Patterns of oxygenation during periodic breathing in preterm infants Christian F. Poets and David P. Southall Deportment

(Received

of’ Puediuirics,

2 August

Nulionul

1990; revision

Heurr and Lung Insritute.

received

25 January

London ( U.K.)

1991; accepted

17 April

1991)

Summary

The characteristics of the arterial oxygen saturation (Sao,) signal during episodes of hypoxaemia (Saoz I 80% for 2 4 s) associated with periodic and non-periodic apnoeic pauses were studied in 16 preterm infants with cyanotic episodes (patients), and 15 asymptomatic preterm infants (controls), matched on birthweight and gestational age. The patients showed a significantly higher percentage of apnoeic pauses followed by a hypoxaemic episode (25 vs. 6X, P < O.Ol), and a two-fold increase in the slope of the desaturation curve (8.4 vs. 4.3% per s, P < 0.005) in periodic compared with non-periodic breathing. All other characteristics of oxygenation (baseline Saoz before episodes of hypoxaemia, delay between onset of apnoeic pause and onset of desaturation, lowest Saoz during episodes of hypoxaemia) were similar for periodic and non-periodic breathing patterns. Similar, but not significant, differences between isolated and periodic apnoeic pauses were also present in the controls. An analysis of episodes of bradycardia (I 100 beats per minute (bpm)) showed that out of 121 episodes in the patients 118 were accompanied by a fall in Saoz to 5 80%. and in the remaining three Sao, fell to 82X, 85% and 86% respectively. Thus all episodes of bradycardia (5 100 bpm) were associated with a fall in SaOz detected by beat-tobeat pulse oximetry. Examination of hypoxaemic episodes and their relationship with bradycardia and with apnoeic pauses, periodic and non-periodic, may help the further understanding of the control of arterial oxygenation in preterm infants with cyanotic episodes. preterm infant; hypoxaemia; bradycardia;

Currrspondenre fu: Dr. D.P. Southall,

ton Hospital.

Fulham

0378-3782/91/.$3.50 Published

and Printed

Road. 0

London

Department of Paediatrics. SW3 6HP, U.K.

1991 Elsevier in Ireland

periodic breathing

Scientific

Publishers

National

Ireland

Ltd.

Heart and Lung Institute. Bromp-

2

Introduction

Periodic breathing is a normal phenomenon in the healthy infant [33]. However, an increased proportion of periodic breathing has been associated with prematurity [8] and with airway hypoxia [3]. Moreover, relatively high proportions of periodic breathing have been been reported in infants at increased risk of sudden infant death syndrome @IDS) [l&36]. It has, however, been shown from a prospective population based study that as a group infants who suffer SIDS do not have increased levels of periodic breathing when compared with surviving controls [10,37]. There is little knowledge concerning the mechanisms underlying periodic breathing in infants born prematurely. Immaturity of the respiratory control system [31], and airway hypoxia [32] have been cited as possible causes. Although airway hypoxia may induce periodic breathing [32], the effect of periodic breathing on arterial oxygenation has not yet been systematically investigated. Decelerations in heart rate are frequently associated with apnoeic pauses in preterm infants. They have been related to the development of hypoxaemia [ 131. It has, however, not yet been systematically analysed whether such sinus-bradycardias are always related to a decrease in arterial oxygen saturation. In this paper, we investigated the relationship between periodic and non-periodic apnoeic pauses and arterial oxygen saturation in preterm infants suffering cyanotic episodes and in preterm controls. We also looked into the relationship between episodes of hypoxaemia and bradycardia in a subgroup of infants in whom ECG was recorded. Two specific questions were addressed: (i) is there a difference between isolated and periodic apnoeic pauses in the response of Saoz to the cessation in breathing; and (ii) is bradycardia a specific indicator of hypoxaemia in these patients? Patients and Methods

Sixteen preterm infants (6 male, 10 female) with a mean gestational age at birth of 30 weeks (range 25-36) and a birthweight of 1485 g (750-2600) were referred to our hospital for evaluation of recurrent cyanotic episodes of unknown cause. The cyanotic episodes were described as recurrent episodes of apnoea with cyanosis and bradycardia requiring stimulation (n = 12) or cardiopulmonary resuscitation (n = 4). In 10 infants, these episodes had predominantly occurred during or after feeding. Conditions which may cause symptomatic apnoea secondary to other disorders, such as seizures, infection, gastro-oesophageal reflux or anaemia, had been excluded by the referring hospital. A congenital heart lesion as a possible underlying cause for the cyanotic episodes had been clinically excluded in all infants. Their mean gestational age at the time of the study was 40 (36-47) weeks, their postnatal age was 10 (4-20) weeks. Eleven infants had been ventilated, and 4 had received additional inspired oxygen for more than 4 weeks, but none of them was on additional oxygen at the time of the study. Thirteen infants were bottle fed, one was breast fed, and two were fed via a nasogastric tube. Six infants were on theophylline during the study. As a control group, 16 recordings from 15 asymptomatic preterm infants, matched to the cases for birthweight, gestational age (mean 40 (35-45) weeks) and postnatal age (mean 10 (3-19) weeks), were drawn from a study of respiratory physiology in

3

preterm infants about to be discharged from three special care baby units. Details of this study have been published elsewhere [28]. In this group, 10 infants had been ventilated, and 5 had developed an O2 requirement beyond 28 days of life, but again none of them was on additional oxygen during the study. All were bottle fed. One infant received theophylline at the time of the study. For matching purposes, one infant had two recordings at different ages used for analysis. Recordings were made in the control group with the fully informed consent of parents and with the ethical approval of the Doncaster Royal Infirmary. All infants underwent 12-h overnight tape recordings (Racal Store 4) of arterial oxygen saturation (Nellcor NlOO in the beat-to-beat mode), the photoplethysmographic waveforms from the oximeter to validate the saturation signal, breathing movements (Graseby pressure capsule or Respitrace, Studley Data System) and airflow (thermistor or expired CO, sampling, Engstriim Eliza). For technical reasons the latter was not recorded in three infants with cyanotic episodes. In 6 infants with cyanotic episodes and in two controls ECG was also recorded. An activity chart including information about feeding times was kept by the nurses in all infants. Recordings were printed and analysed by two experienced workers and by one of the authors (C.F.P.) who were blind to whether they were assessing cases or controls. The duration of artefact-free oxygen saturation signal, as determined from examination of each photoplethysmographic waveform, was measured. Feeding periods were defined as the actual time of feeding plus the first 30 min after a feed. The number of apnoeic pauses, defined as a cessation in breathing movements for at least 4 s, was analysed [29] and classified according to their occurrence during non-periodic or periodic breathing. The latter was defined as a sequence of at least three apnoeic pauses, each separated by not more than 20 breaths [29]. The number of hypoxaemic episodes (Sa@ I 80% for 1 4 s) associated with an apnoeic pause was counted. The onset of desaturation was determined as the time at which the Saoz fell to 3% or more below the values in the immediately preceding stable period (> 5 s in duration). If this was short (5 5 s), the highest value between two episodes of desaturation was used. The response time to an apnoeic pause was measured as the time elapsing from the end of the last inspiration to the onset of the accompanying desaturation as defined above. The slope was calculated from this point until Sao, reached 40% or, if this was not reached, until the lowest value of that episode of desaturation (Fig. 1). The response time of the pulse oximeter, which reflects lung-toe circulation time plus the electronic response time of the oximeter, was obtained by measuring the time between the onset of inspiration following an apnoeic pause and the beginning of recovery from the associated desaturation [I]. The instantaneous heart rate signal as derived from the ECG was analysed for episodes of bradycardia (HR < 100 bpm for at least 2 s). The ECG signal was examined for artefacts during each bradycardia identified on the instantaneous heart rate signal. The lowest Sao, associated with each episode of bradycardia was measured. All variables were calculated separately with regard to their association with periodic or non-periodic breathing pattern. Except for the analysis of the relationship between bradycardia and oxygen saturation, results are presented only for those infants who showed at least one episode of hypoxaemia during each pattern. Individual means

90

D

r

30 Seconds

1

Fig. I. Typical recording of an onset of an episode of periodic breathing. Note the difference in the slope of the desaturation between the tirst apnoeic pause and the subsequent ones. The tracings are: A, arterial oxygen saturation; B, breathing movements; C, instantaneous heart rate; D, plethysmographic waveforms from oximeter; E, electrocardiogram. The arrows indicate beginning and end of a period over which the slope was measured.

TABLE

I

Comparison of the variables measured during apnoeic pauses occurring during periodic and non-periodic breathing in the infants with hypoxaemic episodes during both patterns (n = I I). Medians (and ranges) are given. Statistical comparisons were made using the Wilcoxon matched-pairs signed-ranks test. Variable

Periodic

Apnoeic pauses/h (n) Episodes Sao, 5 80% z 4 s/recording

15.5 (0.3-79.0)

(n) Desaturations associated with feeding (‘%I) Apnoeic pauses with desaturation (‘%I) Apnoea duration with episodes (s) Slope of desaturation (A’% SaOzis) Saoz before episodes (‘X) Lowest Sao, during episodes W) Delay in onset of desdturation (s) Response time of pulse oximeter (s)

breathing

9 (I-81) 67 (o--100) 25.0 (1.3--100) 7.8 (6.2-19.2) X.4 (3.7-12.6) 99.3 (95.1--100) 47.6 (36.3-64.0) 6. I (3.1-8.7) 4.8 (2.9-7.0)

Non-periodic breathing

f-value

16.4 (7.2-24.7)

ns.

IO (l-21)

n.s.

46 (o--100) 6.0 ( I .&-25.0) II.9 (7.3-21.0) 4.3 (2.7-X.1) 98.7 (92.5---100) 52.8 (38.0-68.0) 5.8 (4.1-9.1) 4.6 (3.1-5.2)

ns. < 0.01 < 0.01 c 0.005 n.s. n.s. n.s. n.s.

5

were calculated for all variables in each infant. Data are presented as the medians and ranges of these individual means. Comparisons were made using the Wilcoxon matched-pairs signed-ranks test. Results Patients with cyanotic episodes Fourteen infants showed a total of 350 hypoxaemic episodes associated with an apnoeic pause. All hypoxaemic episodes were self-resolving. Eleven infants had hypoxaemic episodes during both periodic and non-periodic apnoeic pauses. The mean duration of recording in these 11 recordings was 11.9 h (S.D. 2.3 h), the mean duration of artefact free signal was 7.5 h (SD. 1.7 h). Their results are shown in Table 1. In summary, the proportion of apnoeic pauses followed by a hypoxaemic episode was significantly higher during periodic compared with non-periodic breathing (25% vs. 6X, P < 0.01). The mean duration of apnoeic pauses associated with a hypoxaemic episode was significantly shorter during periodic breathing. The slope of desaturation was 8.4% per s during periodic apnoeic pauses and 4.3% per s during non-periodic apnoeic pauses (P < 0.005). The Sao, before the hypoxaemic episodes was similar during both breathing patterns (median 99%). The depth of desaturation during the episodes was also similar. There was no significant difference between periodic and non-periodic apnoeic pauses with regard to the time that elapsed from the end of the last inspiration to the onset of desaturation, or with regard to the time between the onset of breathing and the onset of an increase in saturation (total response time). There were also no significant differences in the proportion of desaturations associated with feeding with regard to their occurrence during periodic or non-periodic apnoea. Out of a total of 246 hypoxaemic episodes during periodic breathing, 10 occurred with the first apnoeic pause in a sequence of periodic apnoea in 7 infants (an example is given in Fig. I). The slope during these episodes was 4.1% per s (3.0-8.7) which was significantly slower than the slope in all remaining episodes observed during periodic apnoea (P < 0.02). In contrast, the duration of the apnoeic pauses at the onset of periodic apnoea was 14.4 s (9.5-25.2) which was significantly longer than the median pause duration during all remaining apnoeic episodes in this pattern (P c 0.05). A total of 121 (median 14, range 6-65) episodes of bradycardia (HR < 100 bpm) were observed in the 6 infants where ECG was recorded. Of these, 66 occurred during periodic apnoea. In 118 episodes the associated SaOz was I 80%; in the remaining three episodes Sao, was 82X, 85% and 86%. This means that bradycardia (< 100 bpm) only occurred when Sac+ fell below 90%. Controls Seven infants showed a total of 76 hypoxaemic episodes; 5 infants had episodes in both periodic and non-periodic breathing. The mean duration of recording was 10.6 h (S.D. 0.9 h), the duration of artefact free signal was 8.3 h (S.D. 0.5 h). The number of apnoeic pauses per hour was 12.3 (0.0-62.6) in periodic, and 16.3 (5.4-22.3) in non-periodic breathing. During periodic versus non-period breathing

6

the number of hypoxaemic episodes per recording was 2 (l-30) and 1 (l-28), the proportion of apnoeic pauses associated with a desaturation was 1.7% (0.7-100%) and 0.7% (0.4-45s) (P = 0.2), and the median pause duration was 5.4 s (4.G5.6) and 6.0 s (4.0-8.5) (P = 0.2). Due to the small numbers of hypoxaemic episodes in the controls, results could not be compared to those occurring in the patients above. However, an analysis of the slope of desaturation was made within the controls. This showed a value of 10.0% per s (5.0-13.80/o) in periodic, and 5.7% per s (3.7-8.5X) in non-periodic breathing (P = 0.08). One of the two infants with ECG recordings had 9 episodes of bradycardia, in 5 of them the Sao2 was I 80”/0, in the remaining 4 it was 5 90%. Discussion Characteristics of the beat-by-beat oxygen saturation signal derived from pulse oximetry during hypoxaemic episodes associated with periodic and non-periodic breathing were analysed in 16 preterm infants with cyanotic episodes and a matched control group of asymptomatic infants. Hypoxaemic episodes to or below 80%) for 4.0 s or longer were more likely to follow apnoeic pauses during periodic than during non-periodic breathing. Isolated apnoeic pauses associated with desaturations were longer in duration than those occurring during periodic breathing. Finally, there was an almost two-fold increase in the slope of the desaturation curves during pauses within periodic compared with non-periodic breathing in the patients with cyanotic episodes. Similar differences in these three variables between periodic and non-periodic breathing were also present in the control group but were not significant, possibly because of a much smaller number of hypoxaemic episodes and fewer infants showing both patterns. For these latter reasons it was not considered appropriate to compare findings in the patients with those in controls. The following discussion therefore relates only to the preterm infants with cyanotic episodes. One limitation of this study is that, predominantly, the resulting information came from a selected group of infants with cyanotic episodes. Although the patients presented with similar clinical symptoms, and despite the exclusion of patients with symptomatic apnoea secondary to other disorders, it cannot be said whether the mechanism for cyanosis was the same in all patients. Moreover, only data from 11 and 6, respectively, of the 16 patients investigated could be used for analysis. They may not be representative for the total group of infants with recurrent cyanotic episodes. However, it was not our aim to study the pathophysiology of cyanotic episodes in the preterm infant. These patients were chosen because they were likely to produce hypoxaemic episodes to a degree that permitted precise and repeatable measurements. Nevertheless, our finding that the asymptomatic preterm infants also demonstrated a comparable pattern in their desaturation slope, suggests that similar mechanisms concerning the matching of perfusion to ventilation operate in these patients. It is important not to imply that the absolute values of oxygen saturation measured from this beat-to-beat pulse oximeter during the course of short-lived hypoxaemic episodes are identical to arterial changes. Although pulse oximetry is a well validated method to determine baseline Sao, with an excellent correlation over a wide range

7

of values (35-100%) [2,5,35], it should be mentioned that validating a pulse oximeter during short-lived episodes of desaturation, as observed here, is at present technically impossible. However, the consistent relationship of these short-lived desaturations to apnoeic pauses, as observed also by others [ 12,241, and previous observations that similar but more severe desaturations occur during cyanotic episodes documented as real by arterial blood gas analysis [38] suggest that the readings of the pulse oximeter do indeed reflect true episodes of hypoxaemia. Since the variables measured were always compared within the same patient and the same overnight recording period, using the same instrument, any potential systematic error of the instrument should not have influenced the differences between isolated apnoeic pauses and those occurring during periodic breathing. A Sao, of 80% or less and a period with Saoz 5 80% of at least 4 s were determined as the inclusion criteria for the hypoxaemic episodes measured. These criteria were chosen to restrict the analysis to episodes with a relatively major change in measured oxygen saturation. This ensured precise measurements by allowing the calculation of the slope over a relatively longer period. Two of the 11 infants for whom results were presented had no airflow recorded. It is thus possible that some of the apnoeic pauses analysed in these two infants were in fact mixed pauses, that is, a pause in breathing movements followed or preceded by a pause in airflow with continued breathing movements. Since mixed pauses leading to desaturations of 4 s or longer were relatively infrequent in the infants wherein airflow was actually analysed (< 5%) [27], it was not felt that the small number of mixed pauses possibly not identified in these two infants would have seriously affected the results. In the present study two parameters were used to describe the response of oxygenation to a cessation in breathing movements: the probability with which an apnoeic pause is associated with a hypoxaemic episode, and the slope with which oxygenation changes during an apnoeic pause. These two variables are influenced by (i) the oxygen stores in the lung, (ii) the oxygen consumption, and (iii) the matching between ventilation and lung perfusion. We have not measured lung function in this study, but it has been reported that lung volume may fall during periodic breathing [25,26], and this has been explained by a decrease in tonic intercostal and diaphragmatic EMG activity occurring simultaneously with the switching from regular to periodic breathing [26]. However, a change in lung volume would have had a striking influence on the delay between the onset of apnoea and the onset of desaturation [12]. Since this response time was similar during isolated and periodic apnoeic pauses, it is unlikely that lung volumes have changed substantially during periodic breathing. The actually very fast response time observed in this study for both isolated and periodic apnoeic pauses may be explained by the low lung volume in relation to oxygen consumption that has been reported in newborn infants [4], and by the finding that lung volume generally decreases during apnoeic pauses [26], particularly, as is most likely, if they occur in active sleep when rib-cage compliance is increased [ 141. A reduced lung volume might also be partly related to the frequent apnoeic pauses occurring during or after feeding [l 11. In fact, approximately half of the desaturations investigated occurred during or shortly after feeding. However, since there were

8

no significant differences in the proportion of desaturations associated with feeding with regard to their occurrence during periodic or non-periodic apnoea, this factor has probably not significantly influenced the differences observed between periodic and non-periodic apnoea. Oxygen consumption has also not been measured, but it is unlikely that an increase in metabolic rate is responsible for the differences between periodic and non-periodic apnoea observed in this study. However, oxygen consumption is higher during active sleep [6], when periodic breathing is most likely to occur. Studies concerning the matching of perfusion to ventilation during periodic breathing in infants are not available. However, the potential for changes in the perfusion to ventilation ratio has been described in adults, in whom oscillations of this ratio were observed even during regular breathing [23]. It has also been concluded from measurements of the alveolar-arterial PO, differences at various concentrations of inspired oxygen that the intermittent right-to-left shunt which causes these oscillations is located within the lungs [22]. In infants, the existence of a right-to-left intrapulmonary shunt has been demonstrated in normal premature babies [19], in patients with recurrent cyanotic episodes [38], and as an autopsy finding in newborn infants with respiratory distress syndrome [ 151. In addition, we have previously demonstrated the presence of hypoxaemia despite continued ventilation in 6 of the patients described here [27]. A congenital heart anomaly that could serve as a source for a right-to-left shunt was excluded as part of the clinical investigations for cyanotic episodes in these infants. We cannot, however, exclude such a shunt through the foramen ovale, since Doppler-/contrast-echocardiography was not and could not be performed during these short-lived hypoxaemic episodes. As a consequence of these considerations we speculate that the higher proportion of desaturations and the increase in the desaturation slope observed during periodic apnoea most likely reflects a disturbance not only in alveolar ventilation but also in lung perfusion during periodic breathing. However, other factors such as a reduction in lung volume or an increase in oxygen consumption may have also contributed to these phenomena. A disturbance in lung perfusion may be triggered by an initial episode of airway hypoxia, induced by the first apnoeic pause of the periodic breathing sequence (Fig. 1). Particularly relevant to this suggestion was the observation that some intervals of periodic breathing were preceded by a longer apnoeic pause. There was, however, no evidence to suggest a fall in baseline Saoz before the onset of the periodic breathing. It was interesting to observe that Saoz regained normal values between each hypoxaemic episode. This suggests a normal perfusion to ventilation relationship between the apnoeic episodes within the periodic breathing. Since the intervals of breathing between pauses were normally short (< 10 s), this might indicate a rapidly reacting pulmonary vascular bed. The potential of the latter to increase its resistance in the presence of even relatively short episodes of airway hypoxia, thus inducing a right-to-left shunt of up to more than 80% [ 161, was reported 30 years ago [34], and has also been confirmed recently in animals [40]. There are also reports that pulmonary vasoconstriction is potentiated by repeated episodes of airway hypoxia [41]. The mechanisms through which airway hypoxia triggers pulmonary vasoconstric-

36

Seconbs

Fig. 2. Example showing the relationship between hypoxaemia and bradycardia during an episode of periodic breathing. There are two episodes of bradycardia (HR falling to - 90 bpm). coincidencing with two relatively prolonged

and deep desaturations.

Tracings

as in Fig. I.

tion are not completely understood. It appears that the pulmonary neuroendocrine cell system may be involved, since it has been shown that airway hypoxia can cause degranulation of these cells, which are located close to both pulmonary airways and vessels [20]. They have afferent and efferent vagal innervation [21] and may be important effecters of cardiorespiratory reflexes [7]. A reactive hyperplasia of those cells has been found in association with chronic airway hypoxia, in infants with bronchopulmonary dysplasia [17], and also in victims of SIDS [7]. The potential effects of the neuroendocrine cell system and its afferents. to the brainstem may also be one explanation for the major changes in heart rate observed during periodic breathing (Fig. 2) which occurred only in the presence of hypoxaemia. It is also possible that the carotid body chemoreceptor system may be involved in this response. Major changes in heart rate during periodic breathing have previously been reported by Gordon et al. [9]. Moreover, Henderson-Smart et al. [ 131 also demonstrated that hypoxaemia is the underlying primary initiator of bradycardia during an apnoeic pause. We confirmed this finding; bradycardia did not occur without hypoxaemia in the infants studied here. Additionally, the constant link between hypoxaemia and bradycardia gives further support to our contention that the SaOz measurements made during periodic breathing with this beat-to-beat pulse oximeter do indeed reflect true hypoxaemia. There are only a small number of studies which have analysed oxygenation during periodic breathing in healthy fullterm infants; none of them studied the slope of the

10

desaturation curve. Mok et al. [24] provided average values of arterial oxygen saturation in healthy newborns and observed that these were lowest during periodic apnoea. Similar observations to those in this paper with regard to the proportion of apnoeic pauses followed by a desaturation were made in a study of oxygenation in healthy term infants from our own group, in which the median percentage of apnoeic pauses (8-12 s in duration) followed by a desaturation to ~80% was 0% in nonperiodic, compared with 37% in periodic breathing [39]. Both studies suggest that in the healthy fullterm infant hypoxaemia is more likely to be present during periodic breathing. Whether this results from similar mechanisms to those described above requires further investigation. Our results indicate the existence of a two-way relationship between hypoxaemia and periodic breathing. Thus airway hypoxia and the resulting hypoxaemia may trigger periodic breathing, and periodic breathing itself may produce hypoxaemia. The statement that periodic breathing is a benign disorder simply because the apnoeic pauses within it are short [30] may not be appropriate. Further investigations of the relationship between periodic breathing and oxygenation are required. Acknowledgements

Out thanks to the workers who gave careful technical effort to the study: Mary Gray, Gill Crowther, Jean Tait, Barbara Neville, Pat Allison, Pauline Mills and Jackie Mills. Thanks also to the medical and nursing staff at the hospitals caring for the patients and at the Doncaster Royal Infirmary. Dr. Poets was funded by the Deutsche Forschungsgemeinschaft, Bonn, F.R.G., and Dr. Southall by the National Heart and Chest Hospitals. References Abraham, N.G., Stebbens, V.A., Samuels, M.P. and Southall, D.P. (1990): Investigation of cyanotic/apnoeic episodes and sleep-related upper airway obstruction by long-term non-invasive bedside recordings. Pediatr. Pulmonol.. 8, 259-262. Boxer, R.A., Gottesfeld. 1.. Singh. S.. LaCorte, M.A.. Parnell, V.A. and Walker, P. (1987): Noninvasive pulse oximetry in children with cyanotic congenital heart disease. Crit. Care Med., 15, 1062-1064. Brady, J.P. and McCann, E.M. (1985): Control of ventilation in subsequent siblings of victims of sudden infant death syndrome. J. Pediatr., 106, 212-217. Cook, C.D.. Cherry, R.B., O’Brien, D., Karlberg, P. and Smith, C.A. (1955): Studies of respiratory physiology in the newborn infant. 1. Observation of normal premature and full-term infants. J. Clin. Invest.. 34, 975-982. Fanconi. S., Doherty, P., Edmonds. J.F.. Barker. G.A. and Bohn. D.J. (1985): Pulse oximetry in pediatric intensive care: comparison with measured saturations and transcutaneous oxygen tension. J. Pediatr., 107, 362-366. Gaultier, C. (1990): Respiratory adaptation during sleep in infants. Lung, Suppl.. 905-91 I. Gillan, J.E., Curran, C., O’Reilly, E., Cahalane. S.F. and Unwin. A.R. (1989): Abnormal patterns of pulmonary neuroendocrine cells in victims of sudden infant death syndrome. Pediatrics, 84.828-834. Glotzbach. S.F.. Baldwin. R.B.. Lederer. B.A., Tansey. P.A. and Ariagno. R.L. (1989): Periodic breathing in preterm infants: incidence and characteristics. Pediatrics, 84, 785-792. Gordon, D., Cohen, R.J.. Kelly. D.. Akselrod, S. and Shannon, D.C. (1984): Sudden infant death syndrome: abnormalities in short term fluctuations in heart rate and respiratory activity. Pediatr. Res.. 18. 921-928.

11 10

11 12 13 14 15 16 17 18 19 20

21

22 23 24 25 26 27 28

29

30 31 32 33 34

35

Gordon, D., Southall, D.P., Kelly, D.H., Wilson. A. Akselrod, S.. Richards, J., Kenet. B.. Kenet. R., Cohen, R.J. and Shannon, D.C. (1986): Analysis of heart rate and respiratory patterns in sudden infant death syndrome victims and control infants. Pediatr. Res., 20, 68&684. Heldt, G.P. (1988): The effect of gavage feeding on the mechanics of the lung. chest wall, and diaphragm of preterm infants. Pediatr. Res., 24, 55-58. Henderson-Smart, D.J. (1980): Vulnerability to hypoxemia in the newborn. Sleep, 3. 331-342. Henderson-Smart, D.J., Butcher-Puech, M.C. and Edwards, D.A. (1986): Incidence and mechanism of bradycardia during apnoea in preterm infants. Arch. Dis. Child. 61. 227-232. Henderson-Smart, D.J. and Read, D.J.C. (1979): Reduced lung volume during behavioral active sleep in the newborn. J. Appl. Physiol.: Respirat. Environ, Exercise Physiol., 46, 1081--1085. Jaykkii, S. (1958): Capillary erection and the structural appearance of fetal and neonatal lungs. Acta Paediatr., 47, 48&500. James, L.S. and Rowe, R.D. (1957): The pattern of response of pulmonary and systemic arterial pressures in newborn and older infants to short periods of hypoxia. J. Pediatr., 51, 5-l 1, Johnson, D.E.. Lock, J.E., Elde, R.P. and Thompson. T.R. (1982): Pulmonary neuroendocrine cells in hyaline membrane disease and bronchopulmonary dysplasia. Pediatr. Res., 16. 44U54. Kelly, D.H. and Shannon, D.C. (1979): Periodic breathing in infants with near-miss sudden infant death syndrome. Pediatrics, 63, 355-360. Krauss, A.N. and Auld, P.A.M. (1969): Ventilation-perfusion abnormalities in the premature infant: triple gradient, Pediatr. Res., 3, 255-264. Lauweryns, J.M., de Bock, V., Guelinckx, P. and Decramer, M. (1983): Effects of unilateral hypoxia on neuroepithelial bodies in rabbit lungs. J. Appl. Physiol.: Respir. Environ. Exercise Physiol.. 55. 1665-1668. Lauweryns. J.M.. von Lommel. A.T. and Dom, R.J. (1985): Innervation of rabbit intrapulmonary neuroepithelial bodies: quantitative and qualitative ultrastructural study after vagotomy. J. Neurol. Sci.. 67, Lenfant, J. Appl. Lenfant,

81-89. C. (1963): Measurement of ventilation/perfusion distribution with alveolar-arterial Physiol., 18, 1090-1094. C. (1967): Time-dependent variations of pulmonary gas exchange in normal

differences. man at rest.

J. Appl. Physiol.. 22, 675-684. Mok, J.Y.Q., McLaughlin, F.J., Pintar. M., Hak, H., Amaro-Galvez. R. and Levison, H. (1986): Transcutaneous monitoring of oxygenation: what is normal? J. Pediatr., 108. 365-371. O’Brien, M.J. (1985): Respiratory EMG findings in relation to periodic breathing in infants. Early Hum. Dev., Olinsky, A., in neonates. Poets, CF.,

I I, 43-60. Bryan, M.H. and Bryan, A.C. (1974): Influence of lung inflation on respiratory control J. Appl. Physiol., 36. 426-429. Stebbens, V.A., Samuels, M.P. and Southall. D.P. (1990): Patterns of breathing during

hypoxaemic episodes in preterm infants at around term (abstract). Early Hum. Dev.. 22, 107-108. Poets, C.F., Stebbens, V.A., Alexander, J.R., Arrowsmith, W.A., Salfield. S.A.W. and Southall, D.P. (1991): Oxygen saturations and breathing patterns in infancy: II. preterm infants at discharge from special care. Arch. Dis. Child, 66, 569-573. Richards, J.M., Alexander, J.R.. Shinebourne, E.A., de Swiet, M.. Wilson, A.J. and Southall, D.P. (1984): Sequential 22-hour profiles of breathing patterns and heart rate in 1 IO full-term infants during their first 6 months of life. Pediatrics, 74, 763-777. Rigatto, H. (1982): Apnea. Pediatr. Clin. North Am., 29. 1105-l 116. Rigatto, H., Brady, J.P. (1972): Periodic breathing and apnea in preterm infants. I. Evidence for hypoventilation possibly due to central respiratory depression. Pediatrics, 50. 202-218. Rigatto, H.. Brady, J.P. (1972): Periodic breathing and apnea in preterm infants. II. Hypoxia as a primary event. Pediatrics, 50, 219-227. Shannon, D.C., Carley, D.W. and Kelly. D.H. (1988): Periodic breathing: Quantitative analysis and clinical description. Pediatr. Pulmonol.. 4, 98-102. Sinclair, J.C.. Engel, K. and Silverman, W.A. (1958): Early correction of hypoxemia and acidemia in infants of low birth weight: a controlled trial of oxygen breathing. rapid alkali infusion, and assisted ventilation. Pediatrics, 42, 565-589. Southall, D.P., Bignall, S., Stebbens,

V.A.. Alexander,

J.R., Rivers, R.P.A.

and Lissauer,

T. (1987):

12

Pulse oximeter and transcutaneous arterial sive care. Arch. Dis. Child, 62, 882-888. 36

37

38

39

40 41

oxygen measurements

in neonatal

and paediatric

inten-

Southall, D.P., Janczynski, R.E., Alexander, J.R., Taylor, V.G. and Stebbens, V.A. (1990): Cardiorespiratory patterns in infants presenting with apparent life-threatening episodes. Biol. Neonate, 57, 71-81. Southall, D.P., Richards, J.M., Stebbens. V., Wilson, A.J.. Taylor, V. and Alexander. J.R. (1986): Cardiorespiratory function in I6 full-term infants with sudden infant death syndrome. Pediatrics, 78, 787-796. Southall, D.P., Samuels, M.P. and Talbert, D.G. (1990): Recurrent cyanotic episodes with severe arterial hypoxaemia and intrapulmonary shunting: a mechanism for sudden death. Arch. Dis. Child, 65, 953-961. Stebbens, V.A., Poets, C.F., Alexander, J.A., Arrowsmith, W.A., Salfield. S.A. and Southall. D.P. (1991): Oxygen saturations and breathing patterns in infancy: 1. fullterm infants in the second month of life. Arch. Dis. Child, 66, 574-578. Taylor, B.J., Fewell, J.E. and Kearns, G.L. (1988): Pulmonary vascular response to aerosolized cromolyn sodium and repeated epochs of isocapneic alveolar hypoxia in lambs. Pediatr. Res.. 23. 513-518. Unger, M., Atkins, M., Briscoe, W.A. and King, T.K.C. (1977): Potentiation of pulmonary vasoconstrictor response with repeated intermittent hypoxia. J. Appl. Physiol.: Respir. Environ. Exercise Physiol., 43, 662-667.

Patterns of oxygenation during periodic breathing in preterm infants.

The characteristics of the arterial oxygen saturation (SaO2) signal during episodes of hypoxaemia (SaO2 less than or equal to 80% for greater than or ...
944KB Sizes 0 Downloads 0 Views