Abstract
Background
Bronchopulmonary dysplasia (BPD) poses a considerable burden of disease in preterm infants, with incidence rising as survival of preterm infants improves. Home oxygen plays a role in the management of infants with BPD, allowing for earlier hospital discharge while requiring supplemental oxygen.
Objectives
To describe the incidence, clinical characteristics and outcomes of infants with BPD at Charlotte Maxeke Johannesburg Academic Hospital (CMJAH), South Africa, focusing on those discharged on home oxygen. A secondary objective was to compare characteristics of infants discharged on home oxygen with those discharged on room air.
Methods
A retrospective descriptive study of infants diagnosed with BPD was conducted at CMJAH between 1 January 2015 and 31 December 2021. Clinical features, respiratory support and outcomes were described and compared between infants discharged on home oxygen and those discharged on room air.
Results
Among 866 infants, the incidence of BPD was 9.8% overall and 28.1% in very low-birthweight infants. Most cases (93.5%) were mild to moderate. Home oxygen use was low (2.1%), and systemic steroid use was high at 59.4%. Infants discharged on home oxygen had a longer hospital stay (85.1 days v. 58.3 days, p<0.001) and higher discharge weight (2 528 g v. 1 834 g, p<0.001). No significant differences in perinatal factors and comorbidities were found. By 1 year corrected age, 54.5% of those on home oxygen had been weaned off it, and 33.3% had had readmissions, mainly for respiratory tract infections.
Conclusion
The incidence of BPD is increasing locally, but home oxygen use remains low compared with high-income countries. Limited uptake may contribute to prolonged hospitalisation. Development of local guidelines and prospective evaluation of home oxygen use are needed to optimise outcomes and reduce the healthcare burden in resource-limited settings.
Study synopsis
What the study adds. Our study showed a high incidence of bronchopulmonary dysplasia (BPD), exceeding earlier local rates but remaining below those of high-income countries. Notably, home oxygen use was low. Infants discharged on home oxygen had longer hospital stays than those discharged on room air, and no clear predictors for home oxygen use were identified.
Implications of the findings. Development of standardised local protocols for home oxygen use in infants with BPD is essential. Earlier integration of home oxygen into management plans should be considered, as timely initiation may facilitate shorter hospital stay.
Keywords: Bronchopulmonary dysplasia, chronic lung disease, incidence, home oxygen, outcomes, low- and middle-income countries
Background
Bronchopulmonary dysplasia (BPD), also known as neonatal chronic lung disease (CLD), was initially described by Northway et al.[1] in 1967. With advances in neonatal care leading to lower mortality rates in preterm infants, the rate of BPD is increasing.[2] The latest definition of BPD by the National Institute of Child Health and Human Development (NICHD), revised in 2019, classifies BPD into three grades based on respiratory support at 36 weeks’ postmenstrual age (PMA). Grade 1 (mild) is nasal cannula at ≤2 L/min, grade 2 (moderate) is nasal cannula at >2 L/min or non-invasive positive airway pressure, and grade 3 (severe) is invasive mechanical ventilation.[3] Data from the Vermont Oxford Network (VON) database in 2021 showed that the incidence of grades 1 and 2 BPD was 36.9% and that of grade 3 BPD 3.7% among infants born between 22 and 29 weeks’ gestation in the USA.[4] Locally, a 2016 study at Charlotte Maxeke Johannesburg Academic Hospital (CMJAH) reported an incidence of 7.5% in infants weighing <1 500 g.[5]
Home oxygen therapy is a key consideration in BPD management, with international rates varying widely (7 - 95%) owing to differences in practice and the absence of standardised guidelines.[6,7] In high income countries, 20.4% of infants with BPD are discharged on home oxygen, with use increasing with disease severity.[4] Locally, CMJAH data published in 2016 reported a much lower rate of home oxygen use of 2.2% in infants with CLD.[5] The 2009 British Thoracic Society guidelines recommend home oxygen for BPD to reduce pulmonary hypertension, prevent intermittent desaturation, lower airway resistance and promote growth.[8] Benefits include improved weight for-length z-scores at 18 - 26 months,[9] reduced risk of nosocomial infections and iatrogenic harm,[10] and earlier discharge.[6] The 2019 American Thoracic Society guideline concluded that these benefits outweigh potential harms, costs and caregiver burden.[10]
As the incidence of BPD is on the rise, determining what patients may require home oxygen therapy becomes more relevant, and predictably the use of home oxygen will also increase.[11] There are limited data available nationally looking at the clinical characteristics of patients with BPD who require home oxygen on discharge and their outcomes. This knowledge may assist in early identification of patients who require home oxygen, thereby allowing for counselling, preparing parents, and assisting with compiling a discharge plan for these patients.[11] The present study aimed to describe the incidence, clinical characteristics and outcomes of infants with BPD at CMJAH, particularly those discharged on home oxygen therapy. The secondary objective was to examine differences in clinical characteristics between infants discharged on home oxygen and those discharged on room air.
Methods
This retrospective descriptive study included infants who were admitted to CMJAH between 1 January 2015 and 31 December 2021 and were diagnosed with BPD. Exclusions were infants with incomplete records, admission after day 3 of life, cardiac lesions (excluding patent ductus arteriosus (PDA)), major congenital anomalies, and infants transferred to another hospital who did not return within 1 week.
Data were obtained from the CMJAH neonatal Research Electronic Data Capture (REDCap) database. Infants with BPD were grouped into those discharged on room air and those on home oxygen. Follow-up data for the home oxygen group were retrieved from the pulmonology clinic files and hospital records.
Located in Johannesburg, South Africa (SA), CMJAH is a public quaternary hospital with a 95-bed neonatal unit, including 14 beds in the intensive care unit (ICU) and 35 in high care, and the remainder in low-care wards and kangaroo mother care.
Definitions
BPD was defined as oxygen dependence at day 28 of life or 36 weeks’ PMA. Severity was classified using the 2019 NICHD criteria; mild to moderate BPD was defined as requiring nasal cannula or non-invasive ventilation, and severe BPD as requiring invasive mechanical ventilation at day 28.[3] Sepsis was defined as culture proven in accordance with VON definitions, requiring isolation of a bacterial or fungal pathogen from a blood culture.[12] Resuscitation was defined as bag-valve-mask ventilation, chest compressions, adrenaline or intubation. Patients weighing 1 000 - 1 500 g were defined as very low birthweight (VLBW) and those weighing <1 000 g as extremely low birthweight (ELBW).[13] Major anomalies followed VON criteria (www.vtoxford.org).
Database
The neonatal records at CMJAH are stored on REDCap, hosted by the University of the Witwatersrand.[14,15] The following data were collected: (i) maternal data: demographics, antenatal care, place and mode of delivery, HIV status, medical conditions, and antenatal steroids; (ii) infant data: gestation, birthweight, sex, head circumference, 5-minute Apgar score, resuscitation at birth, surfactant administration, presence of bacterial or fungal sepsis, receipt of steroids for BPD, respiratory support, comorbidities, and outcomes (discharged on home oxygen or room air, length of hospital stay, discharge weight); and (iii) follow-up (oxygen group): growth, rehospitalisation (including cause), and oxygen weaning by 1 year corrected age.
Statistical analysis
Data were entered in MS Excel for Mac version 16.77.1 (Microsoft Corp., USA) and analysed in Stata version 17 SE (StataCorp LLC, USA). Categorical variables were reported as frequencies and percentages. They were compared using Fisher’s exact (n<5) or χ2 (n≥5) tests. Ordinal and continuous variables were described as medians (interquartile range (IQR)) when they were not normally distributed, and compared with the Mann-Whitney U test. When they were normally distributed, they were described as means (standard deviation (SD)) and compared with a t-test. A p-value of <0.05 was considered significant.
Ethical considerations
Approval to perform the study was obtained from the Human Research Ethics Committee at the University (Medical) (ref. no. M221129).
Results
During the study period, 10 972 infants were admitted to the CMJAH neonatal wards, of whom 1 081 were diagnosed with BPD. A total of 866 infants met the criteria for inclusion (Fig. 1).
Overall incidence of BPD
The overall incidence of BPD was 1 081/10 972 (9.8%). The incidence of BPD in ELBW infants was 319/1 033 (30.8%), the incidence in VLBW infants 570/2 023 (28.1%), and the incidence in infants weighing ≥1 500 g 192/7 916 (2.4%).
Fig. 1.
Study participants included in the analysis of infants diagnosed with BPD admitted to a Johannesburg academic hospital between 1 January 2015 and 31 December 2021. (BPD = bronchopulmonary dysplasia.)
Clinical characteristics
Table 1 displays the clinical characteristics of infants diagnosed with BPD. Among the infants with BPD, the sex distribution was balanced (females 49.1%), and most were delivered by caesarean section (54.6%). Most were inborn (80.3%) and of black ethnicity (97.2%), and they were born to mothers with a mean (SD) age of 28.8 (6.1) years. Antenatal care attendance was 81.0%, with antenatal steroid and magnesium sulphate use reported in 49.8% and 8.2% of mothers, respectively. Maternal hypertension occurred in 23.1%. HIV exposure was common (35.1%), but infection was rare (1.7%).
Resuscitation at birth was required in 73.1% of cases, although endotracheal intubation (5.1%), adrenaline administration (2.1%) and chest compressions (11.4%) were infrequent. Most infants required NCPAP during their admission (87.3%), and 38.8% required conventional mechanical ventilation (CMV). High-frequency ventilation (HFV) was rarely used (3.2%). Of those graded for severity, 93.5% had mild to moderate BPD. Steroid therapy for BPD was administered to 59.4%. Grade 3 - 4 intraventricular haemorrhage (IVH) occurred in 7.6% of cases, while 1.8% had periventricular leukomalacia (PVL). The mean (SD) length of stay was 58.9 (24.5) days. The mortality rate was 10.6%, and 2.1% were discharged on home oxygen therapy.
Table 1. Characteristics of infants with BPD admitted to Charlotte Maxeke Johannesburg Academic Hospital, 1 January 2015 - 31 December 2021.
| Variable | n (%)* |
| Neonatal characteristics | |
| Total infants enrolled, N | 866 |
| Birth weight (g) mean (SD) (n=866) | 1 163.9 (403.0) |
| Gestational age (weeks), mean (SD) (n=861) | 28.9 (2.8) |
| Apgar score at 5 minutes, median (IQR) (n=773) | 8 (8 - 10) |
| Therapeutic interventions | |
| Duration NCPAP (days), median (IQR) (n=714) | 3 (1 - 7) |
| Age at the start of CMV (days), median (IQR) (n=299) | 3 (0 - 13) |
| Duration of CMV (days), median (IQR) (n=249) | 6 (2 - 13) |
| HFV (n=835) | |
| No | 808 (96.8) |
| Yes | 27 (3.2) |
| HFNC (n=696) | |
| No | 643 (92.5) |
| Yes | 53 (7.6) |
| Surfactant therapy (n=840) | |
| No | 123 (14.6) |
| Yes | 717 (85.4) |
| Steroids for BPD (n=837) | |
| No | 340 (40.6) |
| Yes | 497 (59.4) |
| Comorbid conditions | |
| BPD grading (n=443) | |
| Mild to moderate | 414 (93.5) |
| Severe | 29 (6.5) |
| Respiratory distress syndrome (n=862) | |
| No | 66 (7.7) |
| Yes | 796 (92.3) |
| PDA (n=862) | |
| No | 666 (77.3) |
| Yes | 196 (22.7) |
| Bacterial sepsis after 3 days (n=866) | |
| No | 399 (46.1) |
| Yes | 467 (53.9) |
| Fungal sepsis (n=866) | |
| No | 765 (88.3) |
| Yes | 101 (11.7) |
| Outcomes | |
| Length of stay (days), mean (SD) (n=861) | 58.9 (24.5) |
| Outcome (n=866) | |
| Died | 92 (10.6) |
| Survived | 774 (89.4) |
| Discharge weight (g), mean (SD) (n=765) | 1 850.7 (438.4) |
BPD = bronchopulmonary dysplasia; SD = standard deviation; IQR = interquartile range; NCPAP = nasal continuous positive airway pressure; CMV = conventional mechanical ventilation; HFV = high-frequency ventilation; HFNC = high-flow nasal cannula oxygen PDA = patent ductus arteriosus.
* Except where otherwise indicated
Comparison: Home oxygen v. room air on discharge
There were no statistically significant differences between the groups in weight, gestational age, Apgar scores, mode of delivery, or most maternal and perinatal factors (Table 2). Similarly, BPD severity and requirements for NCPAP, CMV, HFV, high-flow nasal cannula and surfactant therapy were comparable (Table 3). Postnatal steroid use was higher in the home oxygen group (82.4% v. 62.6%), but the difference was not statistically significant (p=0.127). Infants discharged on home oxygen had significantly longer hospital stay (85.1 days v. 58.3 days, p<0.001) and higher discharge weight (2 528 g v. 1 834 g, p<0.001) (Table 4).
Table 2. Comparison of neonatal characteristics of patients with BPD discharged on room air with those discharged on home oxygen.
| Variable | Home oxygen, n(%)* | Room air, n(%)* | p-value |
| Sex, n | 18 | 756 | 0.340† |
| Female | 11 (61.1) | 376 (49.7) | |
| Male | 7 (38.9) | 380 (50.3) | |
| Apgar score at 5 minutes, n | 17 | 681 | 0.1030‡ |
| Median (IQR) | 8 (6 - 8) | 8 (7 - 9) | |
| Mode of delivery, n | 17 | 732 | 0.213† |
| Caesarean section | 7 (41.2) | 403 (55.1) | |
| NVD | 10 (58.8) | 329 (44.9) | |
| Birth HIV PCR, n | 9 | 303 | 1.000† |
| Negative | 9 (100) | 299 (98.7) | |
| Positive | 0 | 4 (1.3) | |
| Head circumference at birth, n | 15 | 646 | 0.1697‡ |
| Mean (SD) | 26.4 (1.8) | 27.3 (2.6) | |
| Patent ductus arteriosus, n | 18 | 753 | 0.200† |
| No | 12 (66.7) | 596 (79.2) | |
| Yes | 6 (33.3) | 157 (20.8) | |
| Bacterial sepsis after 3 days, n | 18 | 756 | 0.363† |
| No | 7 (38.9) | 376 (49.7) | |
| Yes | 11 (61.1) | 380 (50.3) | |
| Fungal sepsis, n | 18 | 756 | 0.701† |
| No | 16 (88.9) | 680 (89.9) | |
| Yes | 2 (10.1) | 76 (10.1) | |
| BPD grading, n | 15 | 382 | 1.000† |
| Mild to moderate | 15 (100) | 367 (96.1) | |
| Severe | 0 | 15 (3.9) |
BPD = bronchopulmonary dysplasia; IQR = interquartile range; SD = standard deviation; NVD = normal vaginal delivery; PCR = polymerase chain reaction.
* Except where otherwise indicated.
† χ2/Fisher’s exact test.
‡ t-test/Wilcoxon rank-sum (Mann-Whitney) test.
Table 3. Therapeutic interventions in patients with BPD discharged on room air and those discharged on home oxygen.
| Variable | Home oxygen, n(%)* | Room air, n(%)* | p-value |
| NCPAP, n | 18 | 749 | 1.000† |
| No | 2 (11.1) | 92 (12.3) | |
| Yes | 16 (88.9) | 657 (87.7) | |
| Duration of NCPAP, n | 14 | 624 | 0.1450‡ |
| Median (IQR) | 5.5 (2 - 12) | 3 (1 - 7) | |
| CMV, n | 18 | 739 | 0.385† |
| No | 9 (50.0) | 485 (65.6) | |
| Yes | 9 (50.0) | 254 (34.4) | |
| Age at start of CMV (days), n | 8 | 232 | 0.7802‡ |
| Median (IQR) | 2 (0.5 - 27.5) | 2 (0 - 9) | |
| Duration of ventilation (days), n | 7 | 242 | 0.1974‡ |
| Median (IQR) | 12 (4 - 19) | 6 (2 - 13) | |
| HFV, n | 17 | 730 | 1.000† |
| No | 17 (100) | 712 (97.5) | |
| Yes | 0 | 18 (2.5) | |
| HFNC, n | 16 | 602 | 1.000† |
| No | 15 (93.8) | 557 (92.5) | |
| Yes | 1 (6.3) | 45 (7.5) | |
| Surfactant therapy, n | 18 | 735 | 0.306† |
| No | 4 (22.2) | 103 (14.0) | |
| Yes | 14 (77.8) | 632 (86.0) | |
| Steroids for BPD, n | 17 | 733 | 0.127† |
| No | 3 (17.6) | 274 (37.4) | |
| Yes | 14 (82.4) | 459 (62.6) |
BPD = bronchopulmonary dysplasia; NCPAP = nasal continuous positive airway pressure; IQR = interquartile range; CMV = conventional mechanical ventilation; HFV = high-frequency ventilation; HFNC = high-flow nasal cannula oxygen.
* Except where otherwise indicated.
† χ2/Fisher’s exact test.
‡ t-test/Wilcoxon rank-sum (Mann-Whitney) test.
Table 4. Outcomes of infants with BPD discharged on room air and those discharged on home oxygen.
| Variable | Home oxygen (n=18), n(%)† | Room air (n=756), n(%)† | p-value |
| IVH grade 3 - 4 | 0.114‡ | ||
| No | 15 (83.3) | 707 (93.5) | |
| Yes | 3 (16.7) | 49 (6.5) | |
| Cystic PVL | 1.000‡ | ||
| No | 18 (100) | 746 (98.7) | |
| Yes | 0 | 10 (1.3) | |
| Discharge weight (g), mean (SD) | 2 528.0 (673.2) | 1 834.4 (418.5) | <0.001§* |
| Length of stay (days), mean (SD) | 85.1 (29.4) | 58.3 (21.5) | <0.001§* |
IVH = intraventricular haemorrhage; PVL = periventricular leukomalacia; SD = standard deviation.
* Statistically significant (p<0.05).
† Except where otherwise indicated.
‡ χ2/Fisher’s exact test.
§ t-test/Wilcoxon rank-sum (Mann-Whitney) test.
Follow-up of home oxygen cohort
Eleven infants (n=11/18, 61.1%) attended follow-up, with a mean (SD) corrected age at first visit of 1.81 (1.69) months and a mean weight-for height z-score of –1.8 (3.58) (Fig. 2). Thirty-three percent of infants (n=6/18) were re-hospitalised within 1 year, most commonly for lower respiratory tract infection. Three of these patients (n=3/6, 50%) were hospitalised more than once, with each admission involving an acute respiratory tract infection (Fig. 3). The median (IQR) age at first re-hospitalisation was 2.5 (0 - 4.8) months. No ICU admissions occurred during re-hospitalisations. Fifty-five percent of patients (n=6/11) were weaned from home oxygen within 1 year, at a mean (SD) corrected age of 3.2 (2.0) months.
Fig. 2.
Weight-for-height z-scores at follow-up visits.
Fig. 3.
Diagnoses associated with re-hospitalisations of patients discharged on home oxygen.
Discussion
In our study, the overall incidence of BPD was 9.8%, with the highest burden among ELBW infants, with an incidence rate of 30.8%. There has been a notable increase among VLBW infants, from 15.8% reported at CMJAH between 2013 and 2014, to 28.1%.[5] This increase could be explained by the improved survival of infants of earlier gestational age as neonatal care improves. In contrast, 715 hospitals in the USA that are part of the VON database found an incidence of 40.6% among preterm infants between 22 and 29 weeks’ gestation.[4] The greater availability of advanced respiratory support to very preterm infants in high-income countries is likely to contribute to their higher incidence of BPD, as more extremely preterm infants survive long enough to develop BPD. In our resource-limited setting, restricted capacity to provide prolonged mechanical ventilation to extremely preterm infants may partially explain our lower rates.
In our cohort, the median length of CMV was 6 days, and only 38.8% of infants diagnosed with BPD required CMV. Over the years, NCPAP has become increasingly available in our setting, with 87.3% of infants in our cohort receiving NCPAP. This increase possibly explains the lessened demand for CMV. These findings support emerging evidence that the pathophysiology of BPD is more closely associated with arrested pulmonary development than with direct ventilator-induced lung injury.[16] Among the patients for whom BPD severity could be graded in our study, the majority (93.5%) had mild to moderate disease, contrasting with findings by Regin et al.,[2] who demonstrated an increasing trend of severe BPD over time in preterm infants with a median gestational age of 26 weeks and 3 days born between 2000 and 2019.
The overall rate of home oxygen use in our study was low at 2.1%, far below the 20.4 - 65% reported in high-income countries for infants <32 weeks’ gestation with BPD.[4,6,17] The low uptake can partly be attributed to the absence of local standardised protocols for identifying eligible infants who require home oxygen, a problem highlighted in a multi-centre study by Ejiawoko et al.,[17] where home oxygen use varied from 7% to 95%. At CMJAH, domiciliary home oxygen via a concentrator is supplied by a private provider and funded by the Department of Health. Eligibility requires reliable household electricity, as well as legal identification documentation and proof of address. In our setting, these practical barriers, including challenges in obtaining legal identification for undocumented migrant families, may further limit the prescribing of home oxygen therapy.
Our study demonstrated a high rate of systemic steroid use for BPD at 59.4%. This pattern of use is comparable to what was observed in a study across 715 hospitals in the USA, with rates of 33.2 - 64.0% among infants with BPD.[4] Postnatal systemic steroids after 6 days of life have been shown to decrease the incidence of BPD and the likelihood of requiring home oxygen on discharge.[18] While high steroid use and low home oxygen use were observed together in our cohort, the retrospective design is insufficient to establish a causal relationship. Importantly, the impact of steroid use on neurodevelopmental outcomes remains a concern, as systemic steroids, while beneficial for lung disease, have been associated with risks such as growth failure and cerebral palsy.[18] Notably, we found no statistically significant differences in clinical characteristics or interventions between infants with BPD discharged on room air and those discharged on home oxygen, including no difference in BPD severity. This contrasts with findings by Lagatta et al.,[7] who reported that higher grades of BPD were associated with discharge on home oxygen, alongside other predictors such as earlier gestational age, antenatal steroid exposure, prolonged ventilation and receipt of postnatal steroids, findings not replicated in our study. Other risk factors that have been shown to increase the likelihood of discharge on home oxygen that we did not find include maternal hypertension, CMV in the first 72 hours of life, and the use of surfactant.[6,17,19]
We found no significant differences in common comorbidities associated with BPD, such as PDA, grade 3 - 4 IVH, or PVL. The small sample size of infants discharged on home oxygen in our study may have resulted in limited statistical power to detect such differences. An important finding was that infants who were discharged on home oxygen had a longer hospital stay, which contradicts the findings by Lagatta et al.[6] published in 2012, which showed that patients discharged from high home oxygen use centres had a 1.2 weeks earlier gestational age at discharge. By enabling earlier hospital discharge, the use of home oxygen decreases the risk of iatrogenic complications associated with hospitalisation, such as nosocomial sepsis.[8,10] Additionally, by shortening hospital stay, it lowers the cost of hospitalisation to our strained public healthcare system. The lack of local home oxygen use protocols and the late consideration of patients for home oxygen, in addition to the difficulty of accessibility, could explain why our patients stay in the hospital longer.
The follow-up rate in patients discharged on home oxygen in our study was 61.1%. By 1 year corrected age, 54.5% of these patients were weaned off oxygen, and growth outcomes were favourable. The readmission rate by 1 year corrected age was 33.3%, comparable to the 34% reported in a US study.[7] These findings suggest that, with appropriate selection and follow-up, infants discharged on home oxygen in our setting can achieve outcomes comparable to those in high-income countries, despite the logistical challenges.
Most readmissions were due to respiratory conditions, and every patient who experienced more than one rehospitalisation was admitted for lower respiratory tract infection. This observation underscores the heightened vulnerability of infants with BPD to severe respiratory syncytial virus (RSV) disease, which frequently necessitates readmission.[20] The increased risk of viral infection is hypothesised to result from chronic oxygen exposure during the preterm period, which disrupts and impairs the maturation and function of the pulmonary innate immune mechanisms. These alterations increase susceptibility to severe viral infections.[21] Although routine viral sampling was not performed in this setting, all respiratory admissions in our cohort clustered between February and June, coinciding with the regional RSV season. This finding is consistent with prior observations of RSV causing severe disease in infants with BPD.[20] Supporting these findings, Hong et al.[22] reported that BPD patients requiring home oxygen had a significantly higher risk of readmissions (adjusted odds ratio (OR) 3.45) and RSV infections (adjusted OR 1.95) compared with those not on home oxygen, with the greatest risk in the first year of life. This observation highlights the need for close follow-up, early viral prophylaxis and access to monoclonal antibodies against RSV, such as nirsevimab, which is efficacious against medically attended RSV-related lower respiratory tract infections.[23] Unfortunately, viral prophylaxis with monoclonal antibody drugs is not available in our public hospital setting. With the introduction of maternal RSV vaccination, future studies will be needed to assess how these interventions influence respiratory outcomes in this population.
Study limitations
This study has several limitations. Its retrospective design resulted in incomplete data capture, with important clinical variables such as retinopathy of prematurity, duration of high-flow oxygen support and the number of surfactant doses not consistently recorded in infants weighing ≥1 500 g. Grading of BPD was only possible in half of the cohort, as data on the level of oxygen support are not routinely collected for infants weighing <1 500 g. In addition, the small number of infants discharged on home oxygen limited statistical power to detect differences in clinical characteristics and outcomes. Follow-up of this subgroup was incomplete, and neurodevelopmental outcomes were not assessed. A further limitation is the lack of maternal socioeconomic data. Socioeconomic factors may influence a family’s ability to access home oxygen, which in our setting requires reliable household electricity and the provision of supporting legal documentation. These requirements may disproportionately affect families with no access to these resources and may contribute to the low home oxygen use.
Conclusion
The incidence of BPD in our cohort was 9.8%, with a significantly higher burden among VLBW and ELBW infants. Compared with earlier local data, the rise in incidence is likely to reflect improved survival in increasingly premature infants. While most affected infants had mild to moderate disease and relatively short exposure to CMV, systemic steroid use was high, contrasting with the practice pattern in high-income settings. Infants discharged on home oxygen represented a small subgroup, but their longer hospital stays highlight the gaps in standardised protocols and barriers to accessibility. Despite these challenges, the post-discharge outcomes, including successful weaning from oxygen and comparable readmission rates, demonstrate the feasibility of home oxygen therapy in our resource-limited setting. Our findings emphasise the need for locally adapted guidelines to optimise the identification of infants requiring home oxygen and reduce unnecessary prolonged hospitalisation.
Footnotes
The datasets generated and analysed during the present study are available from the corresponding author (LAM) on reasonable request. Any restrictions or additional information regarding data access can be discussed with the corresponding author.
The research for this study was done in partial fulfilment of the requirements for LAM’s MMed (Paed) degree at the University of the Witwatersrand.
LAM and RTS conceptualised the project. LAM analysed the data and wrote the manuscript. RTS and DAW supported data interpretation, reviewed and revised the manuscript, and approved the final version.
References
- 1.Northway WH, Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline-membrane disease – bronchopulmonary dysplasia. N Engl J Med. 1967;276(7):357–368. doi: 10.1056/NEJM196702162760701. [DOI] [PubMed] [Google Scholar]
- 2.Regin Y, Gie A, Eerdekens A, Toelen J, Debeer A. Ventilation and respiratory outcome in extremely preterm infants: Trends in the new millennium. Eur J Pediatr. 2022;181:1899–1907. doi: 10.1007/s00431-022-04378-y. [DOI] [PubMed] [Google Scholar]
- 3.Jansen EA, Dysart K, Gantz MG, et al. The diagnosis of bronchopulmonary dysplasia in very preterm infants: An evidence-based approach. Am J Respir Crit Care Med. 2019;200(6):751–759. doi: 10.1164/RCCM.201812-2348OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Jensen EA, Edwards EM, Greenberg LT, Soll RF, Ehret DEY, Horbar JD. Severity of bronchopulmonary dysplasia among very preterm infants in the United States. Pediatrics. 2021;148(1):e2020030007. doi: 10.1542/peds.2020-030007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mphaphuli AV, Ballot DE. A review of chronic lung disease in neonates at Charlotte Maxeke Johannesburg Academic Hospital from 1 January 2013 to 31 December 2014. S Afr J Child Health. 2016;10(3):161–165. doi: 10.7196/SAJCH.2016.v10i3.1060. [DOI] [Google Scholar]
- 6.Lagatta J, Clark R, Spitzer A. Clinical predictors and institutional variation in home oxygen use among preterm infants. J Pediatr. 2012;160(2):232–238. doi: 10.1016/j.jpeds.2011.08.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lagatta J, Murthy K, Zaniletti I, et al. Home oxygen use and 1-year readmission among infants born preterm with bronchopulmonary dysplasia discharged from children’s hospital neonatal intensive care units. J Pediatr. 2020;220:40–48. doi: 10.1016/j.jpeds.2020.01.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Balfour-Lynn IM, Field DJ, Gringras P, et al. BTS guidelines for home oxygen in children. Thorax. 2009;64(Suppl 2):ii1–26. doi: 10.1136/thx.2009.116020. [DOI] [PubMed] [Google Scholar]
- 9.DeMauro SB, Jensen EA, Bann CM, et al. Home oxygen and 2-year outcomes of preterm infants with bronchopulmonary dysplasia. Pediatrics. 2019;143(5):220182956. doi: 10.1542/peds.2018-2956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hayes D, Wilson KC, Krivchenia K, et al. Home oxygen therapy for children: An official American Thoracic Society clinical practice guideline. Am J Respir Crit Care Med. 2019;199(3):e5–e23. doi: 10.1164/rccm.201812-2276ST. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Teoh S, Bhat R, Greenough A, Dassios T. Predicting the duration of supplemental home oxygen in prematurely-born infants at discharge from neonatal care. Early Hum Dev. 2021;157:105353. doi: 10.1016/j.earlhumdev.2021.105353. [DOI] [PubMed] [Google Scholar]
- 12.Manual of Operations: Part 2. Data definitions & infant data booklets. Release 26.2. Vermont Oxford Network [Internet] 2022. [10 December 2025]. https://vtoxford.my.salesforce.com/sfc/p/#1N000002CGJf/a/PX000009LWuA/_hPBVzSKkWD_ZpPn89ytoeVMRyqXLHdambazKJBO4fc
- 13.World Health Organization. WHO; 2025. [9 December 2025)]. ICD-11 for Mortality and Morbidity Statistics.https://icd.who.int/browse/2025-01/mms/en [Google Scholar]
- 14.Harris PA, Taylor R, Minor BL, et al. The REDCap consortium: Building an international community of software platform partners. J Biomed Inform. 2019;95:103208. doi: 10.1016/j.jbi.2019.103208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap) – a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–381. doi: 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Higgins RD, Jobe AH, Koso-Thomas M, et al. Bronchopulmonary dysplasia: Executive summary of a workshop. J Pediatr. 2018;197:300–308. doi: 10.1016/J.JPEDS.2018.01.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ejiawoko A, Lee HC, Lu T, Lagatta J. Home oxygen use for preterm infants with bronchopulmonary dysplasia in California. J Pediatr. 2019;210:55–62. doi: 10.1016/j.jpeds.2019.03.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Doyle LW, Cheong JL, Hay S, Manley BJ, Halliday HL. Late (≥ 7 days) systemic postnatal corticosteroids for prevention of bronchopulmonary dysplasia in preterm infants. Cochrane Database Syst Rev. 2021;(11) doi: 10.1002/14651858.CD001145.PUB5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lagatta JM, Clark RH, Brousseau DC, Hoffmann RG, Spitzer AR. Varying patterns of home oxygen use in infants at 23-43 weeks’ gestation discharged from United States neonatal intensive care units. J Pediat. 2013;163(4):976–982. doi: 10.1016/j.jpeds.2013.04.067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Paes B, Fauroux B, Figueras-Aloy J, et al. Defining the risk and associated morbidity and mortality of severe respiratory syncytial virus infection among infants with chronic lung disease. Infect Dis Ther. 2016;5(4):453–471. doi: 10.1007/s40121-016-0137-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Domm W, Misra RS, O’Reilly MA. Affect of early life oxygen exposure on proper lung development and response to respiratory viral infections. Front Med (Lausanne) 2015;2:55. doi: 10.3389/fmed.2015.00055. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Hong T, Bolisetty S, Bajuk B, et al. A population study of respiratory rehospitalisation in very preterm infants in the first 3 years of life. J Paediatr Child Health. 2016;52(7):715–721. doi: 10.1111/jpc.13205. [DOI] [PubMed] [Google Scholar]
- 23.Simões EAF, Madhi SA, Muller WJ, et al. Efficacy of nirsevimab against respiratory syncytial virus lower respiratory tract infections in preterm and term infants, and pharmacokinetic extrapolation to infants with congenital heart disease and chronic lung disease: A pooled analysis of randomised controlled trials. Lancet Child Adolesc Health. 2023;7(3):180–189. doi: 10.1016/S2352-4642(22)00321-2. [DOI] [PMC free article] [PubMed] [Google Scholar]