Abstract
Rationale
To determine if ventilatory pattern instability, manifested as periodic breathing (PB) during physiologic challenge testing, affects postmenstrual age (PMA) at discharge.
Methods
Eighty infants underwent challenge testing at 36 weeks PMA. Infants breathing supplemental O2 received a room air challenge (RAC, N = 51); those breathing ambient air underwent a hypoxic challenge test (HCT, N = 29). Infants were assigned one of four ventilatory control phenotypes based on the presence or absence of PB during their test, and if they passed or failed because of hypoxemia during the challenge test.
Results
There were no clinical or demographic differences between groups. Infants who passed their challenge testing were, on average, discharged 1.6 weeks sooner than those who failed. The groups of ventilatory control phenotypes differed in PMA at discharge (p = 0.0020), but those with PB were younger by PMA at discharge.
Conclusions
Ventilatory pattern instability did not prolong time to discharge. Passing either challenge was associated with earlier discharge, suggesting these tests might identify infants who can have nasal cannula support removed and be safely discharged sooner. Most of the infants who failed their challenge tests with PB were receiving nasal cannula support. Nasal cannula support may be not only treating hypoxemia due to bronchopulmonary dysplasia (BPD), but also mitigating their ventilatory pattern instability.
Introduction
The criteria for the diagnosis of chronic lung disease (CLD) in preterm infants, originally described as bronchopulmonary dysplasia (BPD), have evolved over the past 45 years [1, 2]. It was recently defined for infants born before 32 weeks gestation as the need for supplemental oxygen at 28 days of life. Those receiving supplemental O2 at 28 days of life but breathing room air by 36 weeks postmenstrual age (PMA) are labeled mild BPD, those requiring <0.30 fraction of inspired oxygen concentration (FIO2), as moderate BPD, and those requiring >0.30 FIO2, as severe BPD [3]. This common characterization of CLD assumes that need for supplemental oxygen at 28 days of life or at 36 weeks PMA is due to lung disease, and does not take into account that supplemental oxygen may be needed for other reasons, such as apnea or periodic breathing (PB).
Ventilatory pattern instability is often manifested by PB [4–7]. In recovering premature infants, PB reflects immaturity of central respiratory control (e.g., low apnea threshold), low lung volume, and the normal increase in peripheral chemoreceptor function expected in the weeks after birth [8, 9]. Supplemental oxygen or increasing resting lung volume (e.g., with CPAP) can stabilize ventilatory pattern in these infants and reduce or eliminate PB [10, 11]. We have shown that as many as 43.2% of premature infants receiving supplemental oxygen at 36 weeks PMA developed PB and hypoxemia when transition to breathing room air was attempted during a physiologic challenge test (Room air challenge test, RAC) [12]. Given the impact similar episodes of hypoxemia while breathing room air can have on discharge readiness, we hypothesized that infants with ventilatory pattern instability, defined by PB, are discharged at a later PMA than those without instability.
Subjects and types of support
From August 12, 2011 to November 30, 2013, one hundred thirty-seven (137) neonates born between 24 and 28 weeks gestation were enrolled prospectively at Saint Louis Children’s Hospital, one of the seven centers participating in the National Institutes of Health-supported Prematurity and Respiratory Outcome Program (PROP) (NHLBI, No. UO1HL1017, Clinicaltrials.gov NCT01607216). At 36 weeks PMA, the age at which the diagnosis of CLD is typically assigned, subjects underwent physiologic respiratory testing according to the PROP protocol. Subjects were categorized into one of four ventilatory control phenotypes based on the presence or absence of PB during respiratory testing, and whether or not they maintained passing oxygen saturations (SpO2%) during the testing. Infants requiring mechanical ventilation, noninvasive positive pressure support, high-humidity nasal cannula flow >3 L min−1, or those deemed unstable by the clinical team were excluded. The clinical team, according to general NICU guidelines, chose the type and amount of respiratory support provided for care, without a standard protocol from PROP for adjusting support.
Informed consent was obtained from the parents of each participant. The Washington University Human Research Protection Office and the PROP Observational Study Monitoring Board approved the study.
Methods for assigning ventilatory control phenotypes
At 36 weeks PMA, infants receiving supplemental oxygen with FIO2 > 0.21 or airflow support via nasal cannula with FIO2 ≥ 0.21 underwent a physiologic challenge to room air (RAC). Infants off respiratory support and supplemental O2 at 36 weeks PMA received the hypoxic challenge test (HCT). HCT was done as part of PROP to assess alveolar O2 reserve among infants who had ostensibly recovered from lung disease of prematurity. Both types of challenge tests were done according to the PROP Manual of Procedures [12].
Physiologic challenge to room air (RAC)
Infants were studied in the supine position. The recording was started and data selected for analysis when infants were in behaviorally-defined quiet sleep (QS) [13]. QS was identified when infants had their eyes closed, were breathing regularly, and had at most brief spontaneous movements. Sleep state was continuously assessed during challenge testing; data recorded while awake or during active sleep were not analyzed. During a 5-min baseline period, the infant received the baseline FIO2 and nasal cannula airflow rate prescribed by the clinicians supervising care. SpO2% was recorded continuously using a NONINR pulse oximeter (NONIN Medical, Inc., Minneapolis, MN). In response to a change in arterial oxyhemoglobin saturation the oximeter has an effective averaging time in newborns of 1.5–3.0 s, depending on the pulse rate [14].
Rib cage and abdominal excursions showing ventilatory patterns during spontaneous breathing were recorded using respiratory inductance plethysmography (RIP) (Biocapture, Cleveland Medical Devices, Cleveland, OH). Elastic bands for RIP were placed around the infant at the nipple line and the umbilicus. RIP was recorded during the baseline period, when the infant received prescribed FIO2 and flow, and continued during the challenge periods at our center. Data describing SpO2%, percent of quiet sleep time with %PB, and minute ventilation during quiet sleep from the sum signal of the QDC-calibrated RIP, were compared before and during the challenge, and are reported elsewhere [12, 15].
For the RAC, after the 5-min baseline recording, FIO2 was reduced to 0.21 in steps of 0.20 or less, followed by stepwise decrements of 0.5 L min−1 in airflow provided by nasal cannula. For each stepwise decrement in FIO2 the patient was observed for 5 min, and for each stepwise decrement in flow, for 10 min. Failure (see below) at any step in the weaning protocol resulted in termination of the challenge with resumption of the previously prescribed support. If the infants tolerated all decrements of support, recordings were continued for up to 60 min with the infant breathing only ambient air and with the small nasal cannula undisturbed. RAC failure was defined as SpO2 < 90% for 5 consecutive minutes within any step of the challenge including the 60-min observation period breathing ambient air, or SpO2 < 80% for 15 consecutive seconds. If the infant did not meet criteria for failure during the 60-min observation, he or she was deemed to have passed the physiologic challenge to room air. In some instances we refer to the challenge as a “physiologic” challenge, rather than strictly a wean to ambient room air, because many infants failed before breathing room air alone.
Hypoxic challenge test (HCT)
Before the HCT, infants were placed supine with their heads in a high-flow hood delivering FIO2 = 0.21. Once the infants achieved QS, a 15-min baseline recording was made. FIO2 measured in the hood was then reduced over 30 s to 0.15 ±0.01. If the infant aroused briefly, she or he was allowed to return to QS, and QS status was documented every 5 min. The infants continued to breathe air with FIO2 = 0.15 ± 0.01 for up to 20 min, and infants were considered to have passed the HCT if they did not meet the failure criteria during that time. HCT failure was defined as SpO2% < 85% for 60 consecutive seconds, or SpO2% < 80% for 15 consecutive seconds [12].
Assignment of phenotypes based on challenge testing
PB was defined as at least three central apneas lasting at least 3 s separated by fewer than 20 s of regular breathing. The percent of time with PB during the 5 min baseline period was compared with the 1 min before challenge test failure or completion. The infant was said to have PB if the percent of time with PB in the final minute was greater than the PB percent in the 5 min baseline period. Apneas longer than, 10 s were very rare, and were not linked to challenge failure.
We characterized each infant in terms of ability to maintain acceptable SpO2%, a primary criterion for discharge from our NICU. In addition, we sought to determine how ventilatory pattern instability, manifested as PB, might destabilize SpO2%. To accomplish these descriptions, infants undergoing either the physiologic challenge test or the HCT were grouped into one of four phenotypes. These phenotypes were based on whether or not they passed their respective test, and whether or not they developed an unstable breathing pattern defined by PB during their respective test [12]. Group A passed their challenge with a stable breathing pattern without PB; group B failed their challenge with a stable breathing pattern without PB. Group C passed their challenge despite having an unstable breathing pattern that included PB; group D failed their challenge, with a decrease in oxygen saturations below failure thresholds during a time when they had an unstable breathing pattern with PB. These phenotypes were designed to allow us to examine the association, if any, among SpO2%, PB, and PMA at discharge.
Statistical methods
Descriptive statistics for normally distributed, continuous variables are mean ± standard deviation (SD).
For normally distributed results, means for two groups of continuous variables were compared by two sample t test (Tables 1–3), and continuous variables from the four phenotypes (Table 2) were compared by one-way ANOVA. Post hoc comparison among groups was made by the Bonferroni correction for any ANOVA p < 0.05. Comparisons of the Apgar scores were done using the Mann–Whitney U Test. Chi-squared test and Fisher’s Exact Test were used for categorical comparisons. Statistical significance was defined as p < 0.05. We used R-statistics (R core team, Vienna, Austria) software for post hoc comparisons, Mann–Whitney U, and Fisher’s Exact tests; all other statistics were calculated using Microsoft Excel 2016 (Microsoft, Redmond, Washington).
Table 1.
Passed physiologic test or hypoxic challenge test vs. failed
| Test outcome | Pass (n = 19) | Fail (n = 61) | p value |
|---|---|---|---|
| N for RAC/N for HCT | 11/8 | 40/21 | |
| Birth weight, means, g | 979 ± 163 | 906 ± 219 | 0.126 |
| Gestational age, mean ± SD, weeks | 27 ± 1.1 | 26.7 ± 1.4 | 0.284 |
| Male sex, n (%) | 6 (32) | 23 (38) | 0.628 |
| Race (AA/White/Other) | |||
| African-American, n (%) | 14 (74) | 32 (52) | 0.0264 |
| Caucasian, n (%) | 4 (21) | 29 (48) | |
| Other, n (%) | 1 (5) | 0 (0) | |
| PMA on testing date, mean ± SD, weeks | 36.1 ± 0.6 | 36.2 ± 0.7 | 0.404 |
| Weight on testing date, mean ± SD, g | 2261 ± 0.310 | 2260 ± 338 | 0.993 |
| PMA at discharge, mean ± SD, weeks | 38.1 ± 1.9 | 39.7 ± 2.2 | 0.0046 |
Table 3.
RAC, failed with PB vs. no PB, home O2 requirement
| Physiologic test outcome | Failed with PB (n = 18) | Failed without PB (n = 21) | p value |
|---|---|---|---|
| PMA at discharge, mean ± SD, weeks | 39.7 ± 1.9 | 41.0 ± 2.2 | 0.054 |
| Home on supplemental O2, n (%) | 16 (89) | 18 (86) | 0.723 |
| PMA when off O2, mean ± SD, weeks | 45.0 ± 4.5 | 48.5 ± 9.0 | 0.157 |
Table 2.
Phenotype comparison
| Phenotype | A passed w/o PB (n = 13) | B failed w/o PB (n = 36) | C passed with PB (n = 6) | D failed with PB (n = 25) | p-value |
|---|---|---|---|---|---|
| N for RAC/N for HCT | 10/3 | 22/14 | 1/5 | 18/7 | |
| Birth weight, mean ± SD, g | 921 ± 142 | 894 ± 202 | 1105 ± 138 | 924 ± 245 | 0.152 |
| Gestational age, mean ± SD, weeks | 26.7 ± 1.16 | 26.8 ± 1.3 | 27.6 ± 0.6 | 26.4 ± 1.4 | 0.265 |
| Male sex, n (%) | 5 (38) | 15 (42) | 1 (17) | 8 (36) | 0.642 |
| Race (AA/White/Other) | |||||
| African-American, n (%) | 9 (69) | 19 (53) | 5 (83) | 13 (52) | 0.219 |
| Caucasian, n (%) | 3 (23) | 17 (47) | 1 (17) | 12 (48) | |
| Other, n (%) | 1 (8) | 0 (0) | 0 (0) | 0 (0) | |
| PMA on testing date (weeks) | 36.2 ± 0.6 | 36.3 ± 0.8 | 35.9 ± 0.6 | 36.1 ± 0.7 | 0.490 |
| Weight on testing date, mean ± SD, g | 2222 ± 340 | 2221 ± 330 | 2345 ± 252 | 2317 ± 348 | 0.618 |
| PMA at discharge, mean ± SD, weeks | 38.9 ± 1.9 | 40.1 ± 2.4 | 36.5 ± 0.5 | 39.2 ± 2.0 | 0.0020 |
Results
Of the 137 enrolled subjects, 32 were not eligible for testing at 36 weeks PMA because they were intubated and mechanically ventilated [N = 6], were receiving noninvasive ventilation [N = 8], were receiving high-humidity nasal cannula >3 L [N = 2], had transferred to another NICU [N = 3], or had died [N = 13]. Nineteen infants were eligible but did not undergo testing because of clinical team or parental refusal, or because of other scheduled procedures (e.g., surgery or eye examinations). Six subjects were tested but reliable RIP recordings were not obtained during the challenges. A total of 80 subjects underwent challenge testing, with reliable RIP recordings obtained, at 36.2 ± 0.7 weeks PMA; 51 physiologic RAC, 29 HCT. One patient was transferred to an outside hospital after undergoing challenge test, and was not included in the analysis of length of stay but his data were used for the pass vs. fail descriptive comparisons. There were no significant differences in gestational age, birth weight, gender, ethnicity, or PMA at discharge, between the eligible but not tested group and the group tested.
Irrespective of breathing patterns and type of challenge, 19 infants passed, and 61 failed (Table 1). Those who failed either challenge had an older PMA at discharge, 39.7 ± 2.3 weeks, compared with those who passed, 38.1 ±1.9 (p = 0.0046).
Infants are grouped according to their assigned phenotype in Table 2. The four phenotypes did not differ by birth weight or EGA at birth, or by PMA or weight on date of testing. Significant differences among phenotypes were found in PMA at time of discharge (Table 2, ANOVA, p = 0.0020). Post hoc analysis with the Bonferroni test showed the PMA at discharge was significantly younger for the subgroup that passed despite having PB (Phenotype C) compared with each “failed” subgroup, but did not differ from the other “passed” phenotype (group A).
Table 3 shows more detailed results from the 39 infants failing the RAC physiologic test (18 with PB, and 21 without PB). The majority (87.2%) of infants on nasal cannula support at 36 weeks PMA who underwent a RAC were discharged on supplemental O2. Infants who failed with PB were not more likely to be discharged on supplemental O2. There was a suggestion, however, that infants who failed the physiologic challenge without PB had an older PMA at discharge than those who failed with PB (Table 3, p = 0.054).
Infants breathing room air alone at 36 weeks given the HCT (N = 29) were younger at discharge than those receiving respiratory support and given the RAC (PMA 38.1 ± 1.8 vs. 40.1 ± 2.2 weeks, p < 0.001). Failing the HCT with PB was not associated with an older PMA at discharge (38.0 ± 1.6 weeks).
Discussion
We have shown in this study that infants born between 24 and 28 weeks PMA who passed either a room air physiologic challenge or a hypoxic challenge at 36 weeks PMA were discharged by the clinical team at a younger PMA. This finding is not based on overrepresentation of infants breathing ambient air at 36 weeks, as only 8 of 29 infants receiving an HCT passed.
Several of our results contradicted our hypothesis that respiratory pattern instability, manifested as PB, would be associated with longer hospitalization and older PMA at discharge. Phenotype C infants, who passed despite increased PB during the challenge study, in fact had the youngest PMA at discharge. Furthermore, infants who failed their RAC (Table 3) with PB had younger PMA at discharge than those failing without PB, though the difference did not reach statistical significance. We speculate that infants failing their RAC during PB may have healthier lungs that allow for earlier discharge.
There are likely two principal groups of infants who require supplemental O2 at 36 weeks PMA: those primarily with less healthy lungs, and those primarily with more ventilatory pattern instability. This possible distinction may not be appreciated because supplemental O2 both shortens the duration of PB episodes causing hypoxemia and reduces their frequency [10, 11], as well as relieving alveolar hypoxia in infants without PB. RAC testing as part of larger prospective studies using specific discharge criteria might identify many infants who, if they had their course of caffeine extended, would not require supplemental O2 at discharge [16].
Small numbers makes statistical analysis by phenotype subgroups open to type II error (e.g., Table 2). It is also possible that the group passing and the group failing were not completely similar, and that a larger study would show the passing group have greater birth weights and longer gestations. However, PMAs and weights at the time of challenge at 36 weeks were similar for infants in all groups (Tables 1and 2). Furthermore, the statistically significant differences in PMA at discharge for the passing groups (A and C) vs. fail groups (B and D Table 1) is consistent with the significant differences in PMA by ANOVA for the four smaller sample phenotype subgroups.
When evaluating discharge PMA, our decision to combine data from infants on room air alone (HCT group) and RAC infants deserves more discussion. It is likely that the respiratory system—lungs per se and respiratory control—of most infants on room air was healthier than those who were prescribed support. However, 13 of 51 infants receiving support and thus undergoing RAC, maintained their SpO2% in the passing range when on room air for 60 consecutive minutes. Infants passing the RAC may have had respiratory system health that was closer to those prescribed room air alone (HCT group) than to those failing the RAC. Nearly all infants receiving the RAC who developed PB (17 of 18, 94.4%) failed the challenge, suggesting marked inability to tolerate even brief apneas without hypoxemia. A smaller percentage of those receiving HCT had PB, but overall, 12 of 28 developed PB, and of these 7 of 12 (58.3%) failed. Taken together, for the physiologic parameters we focused on for the challenge tests—maintaining SpO2% in the face of respiratory pattern instability—the findings just described suggest that there was considerable overlap among the HCT and RAC groups.
There are other limitations to this study. Discharge readiness is certainly multifactorial, and evaluation of the usefulness of challenge tests before discharge will require a larger number of infants to measure the effects of other expected factors, such as feeding difficulties, readiness of home caregivers, etc. These associated factors may have been important among our infants receiving the HCT who were still hospitalized but breathing ambient room air.
It has been suggested that the diagnosis of CLD based on the need for noninvasive support be postponed to a later PMA than 36 weeks, with the expectation that respiratory pattern instability would be less common, e.g., nearer to 40 weeks PMA. Our results support this idea, in general, but do not allow us to be more specific about the PMA at which supplemental O2 would no longer be needed to mitigate PB and treat its associated hypoxemia. Our results suggest that infants still on supplemental oxygen at 36 weeks, with respiratory pattern instability, are discharged at a younger age (p = 0.054) (Table 3). Evaluating the need for noninvasive support beyond 36 weeks PMA, perhaps nearer to 40 weeks PMA, as confirmed or refuted by a RAC test, may provide a more useful and rigorous definition for CLD that better reflects the status of the pulmonary parenchyma. Furthermore, larger, prospective studies might show that infants passing either challenge, particularly the RAC, might be weaned from support and prepared for discharge more aggressively.
Funding
NHLBI, No. UO1HL1017, Clinicaltrials.gov NCT01607216.
Footnotes
Conflict of interest The authors declare that they have no conflict of interest.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Maitre NL, Ballard RA, Ellenberg JH, Davis SD, Greenberg JM, Hamvas A, et al. Respiratory consequences of prematurity: evolution of a diagnosis and development of a comprehensive approach. J Perinatol 2015;35:313–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Poindexter BB, Feng R, Schmidt B, Aschner JL, Ballard RA, Hamvas A, et al. Comparisons and limitations of current definitions of bronchopulmonary dysplasia for the prematurity and respiratory outcomes program. Ann Am Thorac Soc 2015;12:1822–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kinsella JP, Greenough A, Abman SH. Bronchopulmonary dysplasia. Lancet. 2006;367:1421–31. [DOI] [PubMed] [Google Scholar]
- 4.Hoppenbrouwers T, Hodgman JE, Harper RM, Hofmann E, Sterman MB, McGinty DJ. Polygraphic studies of normal infants during the first six months of life: III. Incidence of apnea and periodic breathing. Pediatrics. 1977;60:418–25. [PubMed] [Google Scholar]
- 5.Khoo MC, Gottschalk M, Pack AI. Sleep-induced periodic breathing and apnea: a theoretical study. J Appl Physiol. 1985;70:2014–24. 1991 May [DOI] [PubMed] [Google Scholar]
- 6.Al-Matary A, Kutbi I, Qurashi M, Khalil M, Alvaro R, Kwiatkowski K, et al. Increased peripheral chemoreceptor activity may be critical in destabilizing breathing in neonates. Semin Perinatol. 2004;28:264–72. [DOI] [PubMed] [Google Scholar]
- 7.Patel M, Mohr M, Lake D, Delos J, Moorman JR, Sinkin RA, et al. Clinical associations with immature breathing in preterm infants: part 2—periodic breathing. Pediatr Res 2016;80:28–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Edwards BA, Sands SA, Berger PJ. Postnatal maturation of breathing stability and loop gain: the role of carotid chemoreceptor development. Respir Physiol Neurobiol. 2013;185:144–55. [DOI] [PubMed] [Google Scholar]
- 9.Fleming PJ, Goncalves AL, Levine MR, Woollard S. The development of stability of respiration in human infants: changes in ventilatory responses to spontaneous sighs. J Physiol 1984;347:1–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.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 1985;72:116–20. 1992 Jan [DOI] [PubMed] [Google Scholar]
- 11.Wilkinson MH, Sia KL, Skuza EM, Brodecky V, Berger PJ. Impact of changes in inspired oxygen and carbon dioxide on respiratory instability in the lamb. J Appl Physiol 1985;98:437–46. 2005 Feb [DOI] [PubMed] [Google Scholar]
- 12.Coste F, Ferkol T, Hamvas A, Cleveland C, Linneman L, Hoffman J, et al. Ventilatory control and supplemental oxygen in premature infants with apparent chronic lung disease. Arch Dis Child Fetal Neonatal Ed 2015;100:F233–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Grigg-Damberger M, Gozal D, Marcus CL, Quan SF, Rosen CL, Chervin RD, et al. The visual scoring of sleep and arousal in infants and children. J Clin Sleep Med 2007;3:201–40. [PubMed] [Google Scholar]
- 14.Nonin pureSAT SpO2 averaging. http://www.nonin.com/wp-content/uploads/2018/10/NoninMedical_APSS_AveragingwithFAQa.pdf
- 15.Adams JA, Zabaleta IA, Stroh D, Johnson P, Sackner MA. Tidal volume measurements in newborns using respiratory inductive plethysmography. Am Rev Respir Dis 1993;148:585–88. [DOI] [PubMed] [Google Scholar]
- 16.Dobson NR, Rhein LM, Darnall RA, Corwin MJ, Heeren TC, Eichenwald E, et al. Caffeine decreases intermittent hypoxia in preterm infants nearing term-equivalent age. J Perinatol 2017;37:1135–40. [DOI] [PubMed] [Google Scholar]