記住我
Introduction: Our objective was to evaluate the temporal trend of systemic postnatal steroid (PNS) receipt in infants of 24–28 weeks’ gestational age, identify characteristics associated with PNS receipt, and correlate PNS receipt with the incidence of bronchopulmonary dysplasia (BPD) and BPD/death from an international cohort included in the iNeo network. Methods: We conducted a retrospective study using data from 2010 to 2018 from seven international networks participating in iNeo (Canada, Finland, Israel, Japan, Spain, Sweden, and Switzerland). Neonates of 24 and 28 weeks’ gestational age who survived 7 days and who received PNS were included. We assessed temporal trend of rates of systemic PNS receipt and BPD/death. Results: A total of 47,401 neonates were included. The mean (SD) gestational age was 26.4 (1.3) weeks and birth weight was 915 (238) g. The PNS receipt rate was 21% (12–28% across networks) and increased over the years (18% in 2010 to 26% in 2018; p < 0.01). The BPD rate was 39% (28–44% across networks) and remained unchanged over the years (35.2% in 2010 to 35.0% in 2018). Lower gestation, male sex, small for gestational age status, and presence of persistent ductus arteriosus (PDA) were associated with higher rates of PNS receipt, BPD, and BPD/death. Conclusion: The use of PNS in extremely preterm neonates increased, but there was no correlation between increased use and the BPD rate. Research is needed to determine the optimal timing, dose, and indication for PNS use in preterm neonates.
© 2023 The Author(s). Published by S. Karger AG, Basel
IntroductionThe past few decades have seen significant improvement in the survival of extremely low gestational age (GA) infants, but morbidities such as bronchopulmonary dysplasia (BPD) have not decreased. Incidence of BPD has either remained static or increased in most countries [1]. Infants born before 29 weeks’ GA are at a higher risk of developing BPD compared to those born after 29 weeks [2, 3]. Antenatal steroids and postnatal surfactant replacement therapy have improved survival by accelerating fetal lung maturation and preventing or treating respiratory distress syndrome, but their overall impact in reducing BPD has been less impressive [4, 5].
The pathogenesis of BPD is believed to be due to an imbalance between pulmonary inflammation and repair [6, 7]. To counter inflammatory effects, systemic and topical postnatal corticosteroids (PNS) are utilized. The incidence of BPD varied from 10.2% to 24.8% in 10 European regions, and PNS use varied from 3% to 50% among neonates from 19 regions in 11 European countries [8, 9]. The guidelines for the use of systemic postnatal steroids for BPD differs among academic bodies and networks [10–12]. Most of these bodies and studies warn against the early use of systemic corticosteroids in the first week after birth [13–15]. Currently, suggestions are to use low-dose steroid after the first week of birth who remain ventilated and have increasing oxygen requirements and worsening lung disease [16, 17]. However, the use of systemic hydrocortisone on postnatal days 14–28 was not associated with higher survival without moderate or severe BPD [18]. The NEUROSIS trial reported that early use of inhaled corticosteroids significantly decreased BPD, but increased mortality at 36 weeks postmenstrual age [19, 20]. With these conflicting reports, the practice of PNS use differs among neonatal intensive care units (NICU) across the world, as well as within a single NICU.
The International Network for Evaluation of Outcomes (iNeo) of neonates is a collaborative effort involving neonatal networks from 11 high-income countries/regions to identify care practices that improve neonatal outcomes [21]. We reported that while mortality decreased between years 2007 and 2015 in most countries, BPD increased for neonates born very preterm [22]. The use of systemic PNS use has swung from extremely restricted use in 2010 to a slowly increasing trend in the current era, accompanying more active perinatal care of infants born at 22–24 weeks of GA [23]. However, this changing trend in increasing use of PNS has not been clearly documented and various practices across different countries provide a good platform to evaluate contemporary practices [24]. Our aim was to evaluate and compare the temporal trend of systemic PNS use among infants 24–28 weeks’ GA, identify characteristics of patients for whom PNS are used, and correlate PNS use with the incidences of BPD or composite of BPD/death.
Material and MethodsDesign and SettingIn this retrospective cohort study, we included neonates admitted to the participating NICU in 7 national neonatal networks included in iNeo. Neonates born between January 1, 2010, and December 31, 2018 with GA between 240 and 286/7 weeks were included. Neonates born ≥29 weeks’ GA with major congenital malformations and those who received palliative care in the delivery room were excluded. Neonates who died in the first 7 days were excluded, as it was less likely that they would be exposed to PNS for the purpose of preventing or treating evolving BPD.
The iNeo dataset contains anonymized individual data for neonates admitted to 10 independent network or registries. Data from 7 of these registries were used for this study, including 33 units from the Canadian Neonatal Network unit (CNN), 28 units from the Finnish Medical Birth register units (FinMBR), 27 units from the Israel Neonatal network units (INN), 159 units from the Neonatal Research Network Japan units (NRNJ), 50 units from the Spanish Neonatal Network units (SEN1500), 37 units from the Swedish Neonatal Quality Register units (SNQ), and 10 units from the Swiss Neonatal Network units (SwissNeoNet). The characteristics of included units have been reported previously [25]. The study was approved by the Research Ethics Board at Mount Sinai Hospital and by iNeo Directors. Informed consent was waived due to retrospective nature of the study.
Data Source and DefinitionsData harmonization and data collection at each participating unit were described previously [21]. GA was determined based on information available in each country, typically using either early ultrasound or last menstrual period. Birth weight z-scores were calculated using a country-specific birth weight standard [26]. BPD was defined as a receipt of supplemental oxygen or positive pressure ventilation (including mechanical ventilation, continuous positive airway pressure, and high-flow nasal cannula) at 36 weeks postmenstrual age or at the time of discharge or transfer from the NICU if it occurred earlier [7]. Mortality was defined as death during the NICU stay. A composite parameter of death or BPD was also investigated due to competing nature of these outcomes.
Statistical AnalysisAnnual data for PNS use and the incidence of BPD across all networks were obtained from iNeo dataset. Baseline characteristics were reported as frequency (percentage) or mean (standard deviation) for continuous and categorical variables, respectively. The rates of PNS use, BPD, and BPD/death were calculated for each network stratified by characteristics of patients (GA, sex, small for GA, receipt of antenatal steroid, and presence/absence of patent ductus arteriosus). Rates with 95% confidence interval (CI) were reported. Rates of BPD and BPD/death were calculated for each network according to receipt of PNS, and the risk ratio with 95% CI was estimated. Furthermore, duration of respiratory support was presented for each network according to receipt of PNS, and median (IQR) was reported. The trend of PNS use and incidence of BPD were plotted graphically by birth year and assessed by the Cochrane-Armitage trend test. All analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA) with a two-sided significance level of 0.05.
ResultsA total of 47, 401 infants were eligible for this study. The baseline characteristics of neonates were similar across networks (Table 1). The mean (SD) GA and birth weight were 26.4 (1.3) weeks and 915 (238) g, respectively. Antenatal steroids were received by 81%, birth via cesarean section occurred in 70%, surfactant was received by 74% of neonates, and PDA was diagnosed or treated in 53% of neonates (Table 1).
Table 1.Characteristics of infants included in the study from various networks
CharacteristicCNNFinMBRINNNRNJSNQSNNSEN1500TotalNumber of infants12,5851,2224,39915,9253,3032,0447,91647,394Antenatal steroids, n (%)11,115 (90.2)1,165 (96.1)3,625 (82.4)10,047 (64.4)2,832 (88.8)1,902 (93.5)6,557 (90.1)37,243 (80.9)GA, weeks mean (SD)26.4 (1.3)26.5 (1.4)26.6 (1.3)26.3 (1.4)26.4 (1.4)26.5 (1.3)26.6 (1.3)26.4 (1.3)Birth weight, g mean (SD)950 (242)947 (247)936 (215)865 (228)935 (249)901 (224)937 (242)915 (238)Birth weight, Z score (SD)0.04 (0.92)−0.02 (0.92)0.02 (0.88)−0.17 (1.00)−0.04 (0.93)−0.14 (0.84)0.04 (1.05)−0.05 (0.97)SGA, n (%)981 (7.8)120 (9.8)353 (8.0)2,208 (13.9)336 (10.2)210 (10.3)745 (9.4)4,953 (10.4)Male n (%)6,826 (54.3)637 (52.1)2,396 (54.5)8,478 (53.3)1,785 (54.0)1,089 (53.3)4,190 (52.9)25,401 (53.6)PDA, n (%)6,975 (55.7)467 (38.2)2,353 (51.5)8,263 (52.4)1,583 (47.9)945 (46.1)4,213 (54.3)24,799 (52.7)Other characteristics Cesarean birth, n (%)7,546 (60.1)792 (65.8)3,140 (71.4)12,678 (80.1)2,247 (68.6)1,640 (80.3)5,119 (64.7)33,162 (70.2) Apgar <7 at 5 min4,814 (38.7)588 (49.1)651 (15.2)6,137 (38.9)1,113 (34.5)889 (43.6)1,434 (18.2)15,626 (33.4) Surfactant, n (%)8,765 (69.6)1,039 (89.0)3,374 (76.7)12,644 (80.6)2,208 (66.9)1,516 (74.0)5,378 (68.4)34,924 (73.7)Overall, the PNS receipt rate was 21.2% (varying between 12% and 28% across different networks), which increased over years (18% in 2010–26% in 2018; p < 0.01). The rate of receipt of PNS varied from 30% to 76% at 24 weeks and from 3% to 11% at 28 weeks’ gestation between networks (Table 2). The rate of BPD among networks was 35.3% (varying between 24% and 41% across different networks), which remained unchanged over years (35.2% in 2010 and 35.0% in 2018). The rate of BPD/death was 38.7% (varying between 28 and 41% across different networks) (Table 2). The rates of PNS receipt, BPD, and BPD/deaths are reported by infant characteristics in Table 3. There was a significantly higher rate of BPD and BPD/death among those who received steroid compared to those who did not receive steroids (Table 4). The rates of PNS receipt, BPD, and BPD/death were higher in males compared to females; in SGA neonates compared to non-SGA neonates; and in neonates treated for PDA compared to those not treated for PDA. There was no difference in rates of PNS receipt, BPD, and BPD/death among those who received antenatal steroid and those did not receive them (Table 3).
Table 2.Rates of steroid use, BPD, and BPD/death by GA in the iNeo network
GAOutcomeCNNFinMBRINNNRNJSNQSNNSEN 1500Total, n (%), 95% CI24 weeks,n (%)Rates of steroid use, BPD, and BPD/death according to patient characteristics
CharacteristicOutcomeCNNFinMBRINNNRNJSNQSNNSEN 1500Total, n (%) 95% CIMale, n (%)Rate of BPD and BPD/death according to receipt of postnatal steroid
CharacteristicCNNFinMBRINNNRNJSEN 1500SNNSNQTotalBPD Received steroid, n/N (%)1,680/2,451 (68.5)189/311 (60.8)622/946 (65.8)2,885/3,999 (72.1)777/1,196 (65.0)158/228 (69.3)368/517 (71.2)6,679/9,648 (69.2) Not received steroid, n/N (%)3,228/9,484 (34.0)140/870 (16.1)392/3,077 (12.7)3,244/11,584 (28.0)1,094/5,969 (18.3)323/1,741 (18.6)774/2,642 (29.3)9,195/35,367 (26.0) Ratio (95% CI) of BPD among those who received steroid versus not received steroid2.0 (1.9–2.1)3.8 (3.2–4.5)5.2 (4.7–5.7)2.5 (2.5–2.7)3.5 (3.3–3.8)3.7 (3.3–4.3)2.4 (2.2–2.6)2.7 (2.6–2.7)BPD/death Received steroid, n/N (%)1,780/2,458 (69.9)217/338 (64.2)680/1,001 (67.9)2,932/4,044 (72.5)880/1,295 (68.0)177/247 (71.7)411/559 (73.5)7,077/10,032 (70.5) Not received steroid, n/N (%)3,771/10,024 (37.6)155/884 (17.5)717/3,398 (21.1)3,577/11,881 (30.1)1,753/6,622 (26.5)402/1,816 (22.1)882/2,744 (32.1)11,257/37,369 (30.1) Ratio (95% CI) of BPD/death among those who received steroid versus not received steroid1.9 (1.9–2.0)3.7 (3.1–4.7)3.2 (3.0–3.5)2.4 (2.3–2.5)2.6 (2.4–2.7)3.2 (2.9–3.6)2.5 (2.3–2.6)2.3 (2.3–2.4)When evaluated as a longitudinal trend over years, the overall rate of BPD for all networks did not show a significant change from 2010 to 2018, but the PNS receipt rate increased over the years (18% in 2010 to 26% in 2018, p < 0.01). With respect to the trends within individual networks, Canada and Spain showed a decreasing trend of BPD and an increasing trend of PNS use. The trend of steroid use and BPD both increased in Japan, while Israel, Finland, Sweden, and Switzerland did not show a significant change in the trends of either BPD or PNS over the study period (Fig. 1). Correspondingly, the durations of receipt of mechanical ventilation (median 32 days vs. 5 days) and supplemental oxygen (median 81 days vs. 34 days) were higher among neonates who received PNS compared to those who did not receive PNS confirming that at-risk neonates received PNS at higher rate (Table 5).
Fig. 1.Trend of postnatal steroid use in various networks over study years. CNN, Canadian Neonatal Network; FinMBR, Finnish Medical Birth Register; INN, Israel Neonatal Network; NRNJ, Neonatal Research Network Japan; SNQ, Swedish Neonatal Quality Register; SNN, Swiss Neonatal Network; SEN1500, Sociedad Espanola de Neonatologia.
Duration of respiratory support among those who received steroids versus those who did not receive steroid
Respiratory supportGroupCNNFinMBRINNNRNJSEN1500SNNSNQTotalDuration of mechanical ventilation, median (IQR)Received steroid31 (15, 46)19 (11, 34)25 (10, 44)41 (27, 57)25 (13, 39)18 (10, 27)21 (13, 32)32 (18, 49)Not received steroid3 (0, 14)4 (1, 10)3 (0, 10)16 (4, 36)4 (0, 13)2 (0, 6)2 (0, 8)5 (1, 20)Duration of oxygen supplementation, median (IQR)Received steroid78 (43, 111)NA72 (50, 101)88 (61, 125)74 (43, 102)71 (52, 85)84 (56, 108)81 (53, 114)Not received steroid22 (4, 51)NA24 (8, 45)50 (25, 72)25 (6, 50)31 (7, 53)40 (15, 63)34 (9, 60)DiscussionIn this large international cohort from 7 countries, we identified variations in outcomes of BPD or BPD/death among preterm neonates of <29 weeks’ GA between countries. Approximately 1 in 5 neonates of 240–286 received PNS, and 1 in 3 neonates was diagnosed with BPD. The overall BPD rate for the majority of networks did not show a change between years 2010 and 2018; however, the receipt of PNS increased during the study period. Neonates of lower GA; male neonates compared to female neonates; SGA compared to non-SGA, those diagnosed with PDA compared to those not diagnosed as having PDA received PNS at a higher rate and had higher rates of BPD and BPD/death.
The strengths of our study include the large, multinational cohort of high-risk neonates for whom data were collected using a standardized reporting system within each country. However, our study has limitations. First, we excluded neonates of <24 weeks’ GA who may be at very high risk of both exposure to PNS and BPD. Neonates of 22–23 weeks’ GA may drive the change in practice over time as resuscitation at 22–23 weeks’ GA has become more common. However, in order to keep the cohort homogeneous and not affected by variations in resuscitation practices and active care of neonates at lower GA, we excluded those neonates. It is possible that practices of 22 and 23 weeks’ GA neonates may have influenced practice of 24–29 weeks’ GA neonates with regard to PNS; however, we do not have data to prove this speculation. Second, we did not include inhaled steroids as the data were not available for most of the networks. Third, we do not have data on indication or threshold each unit or practitioner used or reasons for initiation of PNS. Such t
留言 (0)