The Natural History of the Appearance of Apnea of Prematurity

0031-399819 1/2904-0372$03.00/0 PEDIATRIC RESEARCH Copyright O 1991 International Pediatric Research Foundation, Inc. Vol. 29, No. 4, 1991 Printed in U.S.A. The Natural History of the Appearance of Apnea of Prematurity KEITH BARRINGTON AND NEIL FINER Walter Mackenzie Health Sciences Center, Division of Newborn Medicine, University of Alberta Hospitals, Edmonton, Alberta, Canada T6G 2B7 [K.B.] and Division of Newborn Medicine, Royal Alexandra Hospitals, Edmonton, Alberta, Canada T5H 2V9 [N.F.] ABSTRACT. Twenty healthy preterm infants of less than 34 wk gestation were studied with continuous recordings, commencing within 8 h of birth, for up to 1 wk of age to determine the usual time course of the appearance of apnea and to classify apnea types. Airway occlusion studies were also performed on a regular basis to determine whether apneic spells were preceded or followed by a reduction in central respiratory drive. Apneic spells of greater than 15 s duration accompanied by hypoxia or bradycardia occurred in all infants before 24 h of age. The frequency of apneic spells was highest in the first 24 h after birth with a mean frequency of 0.9/h and gradually reduced thereafter, falling to 0.2/h by 5 d of age ( p < 0.01). Apneic spells were more likely to be obstructive in the first 2 d of life than thereafter ( p < 0.05). Central apnea was proportionately significantly less frequent during this time period. Reduced respiratory drive, as demonstrated by airway occlusion pressures, was associated with more frequent apnea and was evident at the first occlusion study, which frequently preceded the first significant apnea. (Pediatr Res 29: 372-375,1991) Abbreviations SpOz, pulse oximeter-derived saturation Po.,,airway pressure 0.1 s after initiation of first occluded breath ANOVA, analysis of variance It is commonly stated that idiopathic apnea of prematurity develous after the first 24 to 72 h of life and a m e a that is noted early requires vigorous investigation (1). The factual basis of this statement is uncertain. Henderson-Smart (2) previously demonstrated that apnea (more than three spells of greater than 20 s duration) developed on the 1st or 2nd postnatal d in 77% of preterm infants. He appears to have used impedance monitoring only and it is uncertain whether any recordings were made for independent confirmation of apnea (3). Carlo et al. (4), utilizing 60-min recordings, showed that apneic spells of greater than 5 s duration occurred on d 1 in eight of 10 preterm infants without lung disease. There appears to be no published information regarding the usual frequency and nature of significant apneic spells in an unselected population of healthy newborn preterm infants who have undergone prolonged recordings. Our main objective was to ascertain the usual age of the Received March 28, 1990; accepted December 5, 1990. Reprint requests: Dr. K. Barrington, 3A3 Walter Mackenzie Health Sciences Center, Division of Newborn Medicine, University of Alberta Hospitals, Edmonton G- . ...., .Alherta. . -. ..-, Canada --..- -- T.- ~- -7R7 Supported by the Northern Alberta Children's Hospital Foundation. appearance of apneic spells and the relative frequency of obstructive, mixed, and central apnea during the 1st wk of life in preterm infants without respiratory disease. Our secondary objective was to describe the development of respiratory drive, by determining the postnatal progression of occlusion pressures, in infants with and without apnea. SUBJECTS AND METHODS Twenty infants of less than 34 wk gestation with birth weights above the 5th percentile for gestational age were studied. Infants were all born in the Royal Alexandra Hospitals and were entered into the study as soon after birth as possible. We entered consecutive infants who were clinically determined to be free of acute medical problems; in particular, they were without significant respiratory disease. The criteria for this were an oxygen requirement of no more than 30%, respiratory rate less than 70/min, and no clinical suspicion of pneumonia or hyaline membrane disease. Infants requiring any supplemental oxygen had chest radiographs, which were required to be normal. Ultrasound examination of the head was performed in all infants before 4 d of age and again at approximately 2 wk of age. Any infant with intraventricular hemorrhage that distended the ventricles or evidence of intracerebral hemorrhage or penventricular leukomalacia on either examination was eliminated from the study. After the initial clinical examination and assessment to ensure that the infants were stable, each was monitored with a combination of impedance pneumogram, for the detection of chest wall movement, ECG, pulse oximetry (SpOz), and end-tidal COP monitoring; some infants were monitored with transcutaneous POPin addition. The analogue outputs of each of these monitors was digitized by a Data-translation DT2800 analogue to digital convertor board (Data Translation, Inc., Marlborough, MA) and then analyzed on a Compaq 286 personal computer (Compaq Computer Corp., Houston, TX), using programs that we wrote using the Asyst programming language (Asyst Software Industries, Rochdale, NY). At this point, parental consent was sought for the airway occlusion studies and for the continuance of noninvasive monitoring. In the event of parental refusal of the occlusions, consent for continued monitoring alone was requested. This protocol was approved by the clinical investigation committee of the Roval Alexandra Hos~itals. After obtaining consent, airway occlusions were performed as previously described ( 5 ) . A face mask was applied and the absence of leaks was confirmed. Airflow was monitored by means of a Fleisch 00 pneumotachograph and integrated electronically to give tidal volume. At end-extiration the ainvav was occluded and during the next breath the resultant airway pressures were recorded. ~h~ po,,was recorded. ~h~~~occ~usionresponses were obtained on each occasion and averaged. This process was performed as soon as possible after birtfi and then every 12 h for 373 THE APPEARANCE OF APNEA OF PREMATURITY the next 72 h. Thereafter, occlusions were performed daily until the study terminated. The study continued until the infant was 1 wk old or until therapy (theophylline) for idiopathic apnea of prematurity was prescribed by the attending physician. Thus, all of the results presented herein are from infants who were not receiving methylxanthines or any other therapy for apnea. The clinical staff had no knowledge of the results of the recordings and clinical diagnosis of apnea was according to the usual unit practice of examination of the nursing records. At the termination of the study, the recordings were analyzed by means of a semiautomated process (6). The program used will flag an epoch that is considered to contain a potential apneic spell by looking for a combination of any two of the following: loss of end-tidal CO*, fall in heart rate, (...truncated)