Abstract
This study compares neurodevelopment at 5 years of age in children who were extremely preterm infants and who underwent early systematic echocardiographic screening for patent ductus arteriosus (PDA) vs those who did not undergo screening.
In extremely preterm infants, patent ductus arteriosus (PDA) has been associated with pulmonary hemorrhage, intraventricular hemorrhage, necrotizing enterocolitis, and mortality.1 However, all treatments aiming at closing the PDA have failed to demonstrate significant clinical benefits,2 resulting in various management strategies from no treatment to early pharmacological treatment.3
In a national population-based study, early systematic echocardiographic screening for PDA (intended to diagnose and treat a large PDA at a preclinical stage) was associated with lower in-hospital mortality (14.2% vs 18.5% for no-screening group; odds ratio [OR], 0.73 [95% CI, 0.54-0.98]).4 The influence of a PDA and its management on the neurodevelopmental outcome of children is debated. Safety data are scarce in premature infants exposed to early echocardiographic screening and targeted treatment of a PDA with ibuprofen, and data are absent beyond 2 years of age. We evaluated the cohort at the age when entering school.
Methods
This French prospective study (Etude Epidémiologique sur Petits Ages Gestationnels [EPIPAGE-2]), initiated in 2011, was approved by the Committee for Protection of People, and parents provided written informed consent. Eligible children were born at 24 to 28 weeks’ gestation, were hospitalized at 68 neonatal intensive care units, and were alive on the third postnatal day (day 3). Infants were evaluated according to whether systematic echocardiographic screening was performed before the end of day 3 (exposed group) or not performed (nonexposed group).4 Assessment at 5.5 years of age occurred from September 2016 to December 2017. Neurodevelopmental tests included screening for cerebral palsy and assessment of hearing and visual acuity. The French version of the Wechsler Preschool and Primary Scale of Intelligence was used to measure the full-scale intelligence quotient (FSIQ) and its domains. A contemporary national cohort of children5 was used to estimate the mean FSIQ of 100 (eMethods 1-3 in the Supplement).
The primary outcome was moderate to severe neurodevelopmental disabilities, defined by at least 1 of the following: (1) cerebral palsy level of 2 or greater on the Gross Motor Function Classification System; (2) binocular visual acuity less than 3.2/10; (3) unilateral or bilateral hearing loss greater than 40 dB; or (4) FSIQ less than 79 (SD, <−2). The secondary outcomes included mortality and survival without neurodevelopmental disabilities. Statistical analysis compared matched groups using logistic regression fit with generalized estimating equations to account for paired data. Infants were matched on gestational age, sex, and propensity score when estimating an infant’s probability of undergoing early echocardiographic screening.4 Participants and nonparticipants of follow-up in both groups were compared using the χ2 or t test. Missing variables were handled with multiple imputation, including predictors of nonresponse and outcomes. A 2-sided P < .05 was considered significant. Analyses were performed using SAS version 9.4 (SAS Institute Inc).
Results
Neurocognitive assessment was performed in 373 of 516 (72.3%) exposed children and 341 of 487 (70.0%) nonexposed children. The reasons for absence of assessment were parental refusal of follow-up (58 exposed and 48 nonexposed) and nonparticipation (85 and 98, respectively). Mothers of children lost to follow-up were younger, were born less often in France, and had lower levels of education (Table 1).
Table 1. Baseline Characteristics of Surviving Children at 5.5 Years of Age.
| No. of events/total children (%) at 5.5 y of age among those exposed to systematic echocardiographic screening for PDA (n = 516)a | P valueb | No. of events/total children (%) at 5.5 y of age among those not exposed to systematic echocardiographic screening for PDA (n = 487)a | P valueb | P value for exposed group vs not exposed group who participated in follow-upb | |||
|---|---|---|---|---|---|---|---|
| Participated in follow-up | Lost to follow-up | Participated in follow-up | Lost to follow-up | ||||
| Maternal age at birth, y | |||||||
| <25 | 64/373 (17.2) | 43/143 (30.1) | .005 | 48/341 (14.1) | 40/146 (27.4) | .002 | .31 |
| 25-34 | 230/373 (61.7) | 72/143 (50.3) | 207/341 (60.7) | 74/146 (50.7) | |||
| ≥35 | 79/373 (21.2) | 28/143 (19.6) | 86/341 (25.2) | 32/146 (21.9) | |||
| Maternal birth in France | 296/373 (79.4) | 89/139 (64.0) | <.001 | 262/337 (77.7) | 99/141 (70.2) | <.001 | .60 |
| Parents’ socioeconomic statusc | |||||||
| Professional | 76/361 (21.1) | 19/131 (14.5) | .001 | 74/314 (23.6) | 16/134 (11.9) | .08 | .40 |
| Intermediate | 81/361 (22.4) | 14/131 (10.7) | 72/314 (22.9) | 20/134 (14.9) | |||
| Administrative, public service, self-employed, student | 100/361 (27.7) | 40/131 (30.5) | 77/314 (24.5) | 32/134 (23.9) | |||
| Shop assistant, service worker | 55/361 (15.2) | 30/131 (22.9) | 50/314 (15.9) | 24/134 (17.9) | |||
| Manual worker | 38/361 (10.5) | 16/131 (12.2) | 38/314 (12.1) | 26/134 (19.4) | |||
| Unemployed | 11/361 (3.0) | 12/131 (9.2) | 3/314 (1.0) | 16/134 (11.9) | |||
| Maternal level of education | |||||||
| ≤Secondary school | 103/363 (28.4) | 46/109 (42.2) | <.001 | 85/323 (26.3) | 60/126 (47.6) | <.001 | .29 |
| High school | 66/363 (18.2) | 34/109 (31.2) | 78/323 (24.1) | 24/126 (19.0) | |||
| College degree | 84/363 (23.1) | 14/109 (12.8) | 72/323 (22.3) | 17/126 (13.5) | |||
| Postgraduate or higher | 110/363 (30.3) | 15/109 (13.8) | 88/323 (27.2) | 25/126 (19.8) | |||
| Antenatal corticosteroid use | 313/368 (85.1) | 113/140 (80.7) | .24 | 284/333 (85.3) | 118/145 (81.4) | .28 | .93 |
| Cesarean delivery | 227/372 (61.0) | 88/143 (61.5) | .91 | 216/340 (63.5) | 74/145 (51.0) | .01 | .49 |
| Multiple birth | 133/373 (35.7) | 44/143 (30.8) | .30 | 107/341 (31.4) | 51/146 (34.9) | .44 | .23 |
| Male sex | 198/373 (53.1) | 80/143 (55.9) | .56 | 177/341 (51.9) | 78/146 (53.4) | .76 | .75 |
| Gestational age, mean (SD), wk | 26.6 (1.2) | 26.7 (1.2) | .43 | 26.8 (1.2) | 26.6 (1.2) | .12 | .43 |
| Birthweight z score, mean (SD) | 0 (1.0) | −0.1 (1.0) | .74 | 0 (0.9) | 0.1 (0.7) | .15 | .56 |
| 5-min Apgar score <7 | 83/342 (24.3) | 38/128 (29.7) | .23 | 77/321 (24.0) | 23/128 (18.0) | .17 | .93 |
| Resuscitation in delivery room | 36/367 (9.8) | 13/137 (9.5) | .91 | 30/329 (9.1) | 19/141 (13.5) | .16 | .76 |
| Intubation in delivery room | 309/369 (83.7) | 111/140 (79.3) | .24 | 272/336 (81.0) | 120/143 (83.9) | .44 | .33 |
| Surfactant, dose | |||||||
| None | 35/365 (9.6) | 16/137 (11.7) | .22 | 31/336 (9.2) | 17/140 (12.1) | .42 | .95 |
| 1 | 241/365 (66.0) | 79/137 (57.7) | 220/336 (65.5) | 94/140 (67.1) | |||
| ≥2 | 89/365 (24.4) | 42/137 (30.7) | 85/336 (25.3) | 29/140 (20.7) | |||
| Respiratory distress syndrome | 287/368 (78.0) | 114/141 (80.9) | .48 | 267/337 (79.2) | 111/145 (76.6) | .51 | .69 |
| Antibiotic treatment within first 3 d of life | 272/371 (73.3) | 109/142 (76.8) | .43 | 254/339 (74.9) | 112/144 (77.8) | .50 | .62 |
| Inborn status | 323/373 (86.6) | 118/143 (82.5) | .24 | 299/341 (87.7) | 119/146 (81.5) | .07 | .66 |
| Infant volume in neonatal unitd | |||||||
| ≤15 | 18/373 (4.8) | 4/143 (2.8) | .11 | 19/341 (5.6) | 8/146 (5.5) | .59 | .84 |
| 16-30 | 216/373 (57.9) | 92/143 (64.3) | 188/341 (55.1) | 87/146 (59.6) | |||
| 31-45 | 71/373 (19.0) | 32/143 (22.4) | 72/341 (21.1) | 23/146 (15.8) | |||
| >45 | 68/373 (18.2) | 15/143 (10.5) | 62/341 (18.2) | 28/146 (19.2) | |||
Abbreviation: PDA, patent ductus arteriosus.
Denominators vary according to the number of missing data for each variable. Matched on gestational age in weeks, sex, and propensity score. Some of the data are expressed as mean (SD) as indicated.
Calculated using the χ2 test for categorical variables and the t test for continuous variables.
Defined as the highest occupational status between occupations of the mother and the father or only the mother if living alone.
Defined by the number of preterm infants born at 28 weeks’ gestation or earlier who were included in the EPIPAGE-2 study.
After multiple imputation, rates of moderate to severe neurodevelopmental disabilities were 23.8% (123/516) in exposed children and 26.7% (130/487) in nonexposed children (OR, 0.85 [95% CI, 0.61-1.20]; P = .36). The exposed group scored higher on 2 domains of intelligence and had a lower mortality rate at 5.5 years of age vs the nonexposed group (14.7% [89/605] vs 19.5% [118/605], respectively; OR, 0.71 [95% CI, 0.53-0.95]; P = .02) (3/89 vs 6/118 deaths occurred after discharge). Survival without moderate to severe neurodevelopmental disabilities was 65% in the exposed group and 58.9% in the nonexposed group (OR, 1.29 [95% CI, 1.00-1.68]; P = .05) (Table 2).
Table 2. Neurodevelopmental Outcome by Early Echocardiographic Screening Status.
| No. of events/total children (%)a | Effect (95% CI) | P value | ||
|---|---|---|---|---|
| Exposed to systematic echocardiographic screening for PDAb | Not exposed to systematic echocardiographic screening for PDAc | |||
| Neurodevelopmental disabilities | ||||
| Severed | 55/516 (10.7) | 62/487 (12.7) | OR, 0.75 (0.45 to 1.24) | .43 |
| Moderatee | 68/516 (13.2) | 68/487 (14.0) | OR, 0.85 (0.54 to 1.36) | |
| Mildf | 196/516 (38.0) | 188/487 (38.6) | OR, 0.89 (0.65 to 1.23) | |
| Noneg | 197/516 (38.2) | 169/487 (34.7) | OR, 1 [Reference] | |
| WPPSI scores, mean (SD) | ||||
| Verbal comprehension | 95.2 (17.5) | 94.4 (18.2) | MD, 0.8 (−1.9 to 3.5) | .57 |
| Visual-perceptual reasoning | 94.7 (14.9) | 91.9 (15.7) | MD, 2.9 (0.4 to 5.3) | .02 |
| Fluid reasoning | 95.6 (15.1) | 93.9 (15.8) | MD, 1.7 (−0.8 to 4.2) | .17 |
| Working memory | 94.7 (14.0) | 92.1 (13.9) | MD, 2.7 (0.4 to 4.9) | .02 |
| Processing speed | 93.0 (15.2) | 91.6 (16.5) | MD, 1.4 (−0.9 to 3.7) | .24 |
| Full-scale intelligence quotient | 92.7 (15.1) | 90.7 (15.9) | MD, 2.0 (−0.4 to 4.4) | .10 |
| Survival at 5.5 y of age | 516/605 (85.3) | 487/605 (80.5) | OR, 1.40 (1.05 to 1.88) | .02 |
| Survival without moderate to severe neurodevelopmental disabilities at 5.5 y of age | 393/605 (65.0) | 357/605 (59.0) | OR, 1.29 (1.00 to 1.68) | .05 |
Abbreviations: MD, mean difference; OR, odds ratio; PDA, patent ductus arteriosus; WPPSI, Wechsler Preschool and Primary Scale of Intelligence (4th edition).
Unless otherwise indicated.
Assessed in 373 of 516 children (72%). Matched on gestational age in weeks, sex, and propensity score. Results were obtained after multiple imputation.
Assessed in 341 of 487 children (70%). Matched on gestational age in weeks, sex, and propensity score. Results were obtained after multiple imputation.
May have met more than 1 of the following criteria: cerebral palsy Gross Motor Function Classification System (GMFCS) level of IV or V, bilateral binocular visual acuity less than 1/10, unilateral or bilateral hearing loss greater than 70 dB, or full-scale intelligence quotient SD less than −3.
May have met more than 1 of the following criteria: cerebral palsy GMFCS level of II or III, bilateral binocular visual acuity of 1/10 or greater to less than 3.2/10, unilateral or bilateral hearing loss greater than 40 dB to 70 dB, full-scale intelligence quotient SD between −3 and less than −2.
May have met more than 1 of the following criteria: cerebral palsy GMFCS level of I, binocular visual acuity of 3.2/10 or greater to less than 5/10, unilateral or bilateral hearing loss of 40 dB or less, full-scale intelligence quotient SD between −2 and less than −1, movement assessment battery for children (2nd edition) score less than fifth percentile, behavioral difficulties according to a Strengths and Difficulties Questionnaire score greater than 90th percentile.
Reference group born at term.
Discussion
This study found that echocardiographic screening for PDA before day 3 among extremely preterm infants was not associated with adverse neurodevelopmental outcomes at 5.5 years of age. Higher survival at this age mirrored the higher survival during the initial hospitalization. Differences in survival rates may generate a competing risk for neurodevelopmental outcomes, which is partly addressed by assessing survival without neurodevelopmental disabilities. This outcome favored the exposed group, but the difference was not statistically significant. However, the lower bound of the 95% CI was 1.00. Study limitations include selection bias during follow-up and residual bias arising from missing data even though multiple imputation was used. These results for school-aged children favored early echocardiographic screening for PDA management, but the results must be confirmed in randomized clinical trials.
Section Editors: Jody W. Zylke, MD, Deputy Editor; Kristin Walter, MD, Associate Editor.
eMethods 1. Neurodevelopmental evaluation
eMethods 2. Classification of neurodevelopmental disabilities
eMethods 3. Reference group
eReferences
References
- 1.Hamrick SEG, Sallmon H, Rose AT, et al. Patent ductus arteriosus of the preterm infant. Pediatrics. 2020;146(5):e20201209. doi: 10.1542/peds.2020-1209 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Benitz WE, Watterberg KL, Aucott S, et al. Patent ductus arteriosus in preterm infants. Pediatrics. 2016;137(1):e20153730. doi: 10.1542/peds.2015-3730 [DOI] [PubMed] [Google Scholar]
- 3.Clyman RI, Liebowitz M. Treatment and nontreatment of the patent ductus arteriosus. J Pediatr. 2017;189:13-17. doi: 10.1016/j.jpeds.2017.06.054 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Rozé JC, Cambonie G, Marchand-Martin L, et al. Association between early screening for patent ductus arteriosus and in-hospital mortality among extremely preterm infants. JAMA. 2015;313(24):2441-2448. doi: 10.1001/jama.2015.6734 [DOI] [PubMed] [Google Scholar]
- 5.Pierrat V, Marchand-Martin L, Marret S, et al. Neurodevelopmental outcomes at age 5 among children born preterm. BMJ. 2021;373(741):n741. doi: 10.1136/bmj.n741 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods 1. Neurodevelopmental evaluation
eMethods 2. Classification of neurodevelopmental disabilities
eMethods 3. Reference group
eReferences