Randomized trial of occlusive wrap for heat loss prevention in preterm infants: neurodevelopmental outcome

Abstract

Background

Heat loss in the delivery room in very preterm infants is associated with increased mortality, hypothermia and other adverse outcomes. The Heat Loss Prevention (HeLP) trial resulted in a significant reduction of hypothermia but not a reduction in mortality. The HeLP follow-up trial was designed to determine whether application of occlusive wrap applied immediately after birth reduces adverse long-term neurodevelopmental outcomes at 18–24 months corrected gestational age in very preterm infants.

Method

The original HeLP trial was a randomized controlled trial in infants < 28 weeks gestation placed at birth in occlusive wrap (wrap group) or not placed in occlusive wrap (control group). The HeLP follow-up trial was the later assessment of these infants enrolled in the HeLP trial to determine the composite outcome of death or neurodevelopmental disability at 18–24 month corrected gestational age.

Results

This report constitutes a description of outcome for a subgroup of infants enrolled at centers with perinatal follow-up programs. 356 infants were eligible for follow-up from the HeLP trial with a loss to follow-up of 116/356 (32.6%). The primary outcome occurred in 82/113 (72.6%) in the wrap group and 80/127 (63.0%) in the control group (p = 0.115). An unadjusted odds ratio was 1.55 (95% confidence interval = 0.90–2.69) and because of a large loss to follow-up, odds ratios adjusted for potential predictors of the outcome were determined and showed no differences in the primary outcome by group assignment. Secondary outcomes in the wrap versus control group were: cerebral palsy 9/70 (12.9%) versus 5/87 (5.7%) (p = 0.20), Mental Developmental Index < 85 (Bayley Scales) 30/53 (56.6%) versus 33/67 (49.3%) (p = 0.54), blindness at 0/68 (0.0%) versus 1/86 (1.2%) (p > 0.99) and deafness 5/70 (7.1%) versus 6/87 (6.9%) (p > 0.99). No difference was noted in prescription glasses, 7/70 (10.0%) in the wrap group versus 2/87 (2.2%) in the control group (p = 0.09).

Conclusions

Use of polyethylene occlusive wrap after birth in very preterm infants was associated with similar death or long-term neurodevelopmental disability rate.

Trials Registion

This trial was registered with Clinicaltrials.gov NCT00607464 on 22 January 2008. Ethics approval was provided by each site and held Mount Sinai Hospital Research Ethics Board (15–0243-C; 7 October 2015).

Background

Heat loss is common shortly after birth in very preterm (VP) infants and is due mainly to evaporative losses [1] and to a lesser extent convective and conductive heat losses. The preterm infant is prone to hypothermia when exposed to the relatively cool environment in the delivery room. [2] These concerns were known even in the nineteenth century and recommendations to use incubators to impede heat loss were made [3]. The initial randomized trial by Vohra [4] showed that those wrapped in polyethylene occlusive wrap maintained body temperature closer to the normal range than those infants not wrapped in polyethylene occlusive wrap, particularly in infants < 28 weeks gestational age and this was supported by additional randomized trials [5–12].Despite reduced hypothermia in the VP infant, other outcomes, such as death prior to hospital discharge, major brain injury, duration of oxygen therapy, necrotizing entercolitis, late onset sepsis and duration of hospital stay were not different although pulmonary hemorrhage was slightly lower in the wrap group [13, 14]. A network meta-analysis showed that plastic wrap combined with a plastic head cap lowered mortality and a plastic wrap combined with heated humidified gases lowered brain injury [15]. The Neonatal Resuscitation Program advocates the use of polyethylene occlusive wrap to maintain body temperature in the delivery room [16].

The Heat Loss Prevention (HeLP) trial [17] is the most recent and largest trial to date of the use of polyethylene occlusive wrap in VP infants in the delivery room compared to not wrapping VP infants in the delivery room. The current study aims to compare death or adverse neurodevelopmental outcome by 18–24 months corrected gestational age (CGA), using maternal due date instead of baby’s birth date, in infants being placed in a polyethylene occlusive wrap (wrap group) or to not being placed in a polyethylene occlusive wrap (control group) shortly after birth while still in the delivery room and thus extending the findings of the HeLP trail and the parallel exploratory trial of infant < 24 weeks gestation [18] to long-term neurodevelopmental outcome [19].

Methods

Study design

The HeLP trial was a multicentered study assisted by the Vermont Oxford Network (VON) in seeking willing NICU’s to participate. The trial included 39 centers in which VP infants with gestational age < 28 weeks were randomly assigned in a 1:1 ratio either to be placed in a polyethylene occlusive wrap (wrap group) or to not being placed in a polyethylene occlusive wrap (control group) shortly after birth while still in the delivery room [19]. The trial used permuted blocks of 4 or 6 using random number generating software. The infants were stratified by center and by three gestational age groups: 1) 26 0/7 to 27 6/7 weeks; 2) 24 0/7 to 25 6/7 weeks; and 3) < 24 weeks. Treatment allocation was contained in a sealed, opaque, and sequentially numbered randomization envelopes which was opened by the team before entering the delivery room.

Sample and setting

Local ethics approval at each center was provided and the trial protocol was registered at Clinicaltrials.gov on 22 January 2008 (NCT00607464). Each participating study center decided to use one of three formats for obtaining consent: 1) a signed antenatal informed consent from a parent or legal guardian to allow for their child to be included in the trial; 2) a waiver of consent in which the parent or legal guardian was approached after birth and informed that their child was enrolled in the study and they gave verbal consent for study participation (i.e. for use of their child’s data); or 3) a delay of consent in which the parent or legal guardian was approached after birth and informed that their child was enrolled in the study and they provided a signed consent for study participation (i.e. for use of their child’s data) [20]. The study team had the approval from Steering Committee site at Sunnybrook Health Sciences Centre, Toronto, (Research Ethics Board #15–0243-C) where the data was stored. Each participating institution submitted a copy of their institution’s Institutional Review Board approval and subsequent annual renewals to the VON HeLP Trial Coordinating Center. This approval process included the enrollment of their child in their local follow-up program (where such a program existed) to evaluate the child between 18–24 months CGA. This work has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans. While group assignment was done in a blinded fashion, because of the nature of the intervention, blinding to each group was not possible after group assignment.

Sample size

The initial hypothesis of the HeLP trial of infants 24–28 weeks [17] and the parallel exploratory trial of infants < 24 weeks in the wrap group versus the control group was a reduction: 1) in all cause mortality by 6 months of age; or 2) alive and still in NICU at 6 months. The sample size was determined by this outcome and has been published previously [17]. The original study enrolled infants from November 2004 until June 2010 when a futility analysis determined that the study could not reach a 25% reduction in mortality even if all 1604 infants calculated in the sample size requirements had been enrolled. When the study was stopped, 801 infants had been enrolled in the HeLP trial [17] and 28 had been enrolled in the parallel exploratory trial [18].

Study population

Among the intended secondary outcomes proposed in the initial HeLP protocol were the evaluation of long-term neurodevelopmental outcomes of infants [19]. Of the initial 39 centers participating in the VON HeLP trial, 11 had formal perinatal follow-up programs allowing them to evaluate the long-term neurodevelopmental outcome of these infants. One of the stratification features was by center, therefore, infants in the original trial born in a center with a follow-up program constitute a valid subgroup analysis with a random allocation of patients to either the wrap group or the control group. The neurodevelopmental outcome measured at 18–24 months CGA included a complete physical examination (including a neurologic evaluation), developmental testing with either of the Bayley Scales of Infant Development (second edition) [21] or the Bayley Scales of Infant and Toddler Development (third edition), [22] hearing assessment and visual assessment.

Outcome measures

The long-term outcome was a major secondary outcome specified for the HeLP trial [19]. The composite of death or neurodevelopmental disability as defined by any one of: a) cerebral palsy of any degree, or b) mental developmental index (MDI) < 85 on the Bayley (version two) or calculated MDI < 85 on the Bayley (version three) [calculated MDI = 88.8—(61.6 × (language score/100)^{− 1}) + (0.67 × cognitive score)], [23] c) blindness (visual acuity < 20/200 in the best eye) or d) hearing impairment (requiring amplification in the best ear).

Statistical analysis

Chi square tests, odds ratios with 95% confidence intervals (CI) and risk difference with 95% CI were used to compare outcomes of categorical variables. Logistic regression analysis adjusting for gestational age, waiver of consent, sex of the infant, birth weight, intrauterine growth restriction, antenatal corticosteroids, chorioamnionitis, neonatal sepsis, necrotizing enterocolitis, bronchopulmonary dysplasia treated with steroids or maternal education, factors potentially predictive of the primary outcome (death or disability in the follow-up of HeLP), was used.

Results

This report constitutes a description of outcome for a subgroup of infants enrolled at centers with perinatal follow-up programs. From these 11 centers, 356 were enrolled and randomized in the trial. The primary outcome of death or disability included babies from these centers who were known to have died at any time prior to 18–24 months CGA and the number with any disability detected by 18–24 months CGA. Infants lost to follow-up were not categorized. 183 were randomly assigned to the wrap group and 173 randomly assigned to the control group (Fig. 1). All infants were analyzed according to intention-to-treat. Initial state variables for those infants with known outcomes are noted in Table 1.

Fig. 1.

Infants in the HeLP trial enrolled in follow-up programs

Table 1.

Initial state variables

	wrap group	control group
Lost to follow-up‡	70/183 (38.3%)	46/173 (26.6%)
Gestational age in weeks (median & range)	25.4 (22–27) n = 113	25.9 (22–27) n = 127
Waiver of consent	32/113 (28.3%)	30/126* (23.8%)
Infant female	52/112* (46.4%)	59/125* (47.2%)
Birth weight in grams (median & range)	720 (340–1307) n = 111*	790 (370–1595) n = 125*
Intrauterine growth restriction	7/111* (6.3%)	4/125* (3.2%)
Antenatal corticosteroid use	105/110* (95.5%)	113/125* (90.4%)
Chorioamnionitis	32/110* (29.1%)	24/122* (19.7%)
Sepsis in baby	49/105* (46.7%)	59/115* (51.3%)
Necrotizing enterocolitis (≥ stage 2)	11/105* (10.5%)	9/109* (7.6%)
Bronchopulmonary dysplasia treated with steroids	105/110* (95.5%)	113/125* (90.4%)
Mother’s education (high school diploma or less)	60/112* (53.6%)	72/125* (57.6%)
Maternal temperature in labour degrees C (median & range)	36.9 (35.6–39.7) n = 101*	36.8 (35.6–38.9) n = 101*
Newborn temperature after admission to NICU in degrees C (median & range)†	36.3 (33.6–38.4) n = 104*	35.9 (32.1–37.7) n = 117*
Age (minutes) left delivery room (median & range)	15 (4–130) n = 71*	15 (5–45) n = 65*
Age in days to final extubation (median & range)†	42 (1–109) n = 109*	24 (1–143) n = 88*
Intraventricular hemorrhage grades III or IV	14/98* (14.3%)	18/115* (15.7%)
Cystic periventricular leukomalacia	3/97* (3.1%)	6/115* (5.2%)
Pulmonary hemorrhage	9/109* (8.3%)	17/124* (13.7%)

^‡loss to follow-up is significantly higher in the wrap group p < 0.025 (Fisher’s exact test)

^*do not add up to 113 for the wrap group and 127 for the control group due to missing data

^†p < 0.05 (Fisher exact test)

Table 1 also indicates Neonatal Intensive Care Unit (NICU) outcomes by a group assignment. As with the larger trial [17] the admitting temperature to the NICU was higher in the wrap group. Age in days to final extubation was noted to be older in the wrap group compared to the control group (Table 1). Pulmonary hemorrhage was significantly lower in the wrap group in the original HeLP trial [17] but was not noted to be lower in the infants in the follow-up component of the HeLP trial. No other differences in neonatal outcomes were identified.

The loss to follow-up rate was 32.6% (116/356) with 3 of 11 centers achieving at least 85% follow-up rate. There was a significantly higher loss rate in the wrap group, 70/183 (38.3%), compared to the control group, 46/173 (26.6%) (p = 0.025). Median birth weight was 818 g (interquartile range 710 to 920) in those lost to follow-up compared to 751 g (interquartile range 641 to 910) to those not lost (p = 0.008) and median gestational age was 26 weeks (interquartile range 25 to 27) in those lost compared to 25 weeks (interquartile range 24 to 26) in those not lost (p < 0.001). There was no significant difference in the primary outcome of death or disability in the wrap group compared to the control group for all infants enrolled in the follow-up study or for the subgroup of infants < 26 weeks enrolled in the follow-up study (Table 2). No differences were found in survivors for overall neurodevelopmental disability or any of the components of neurodevelopmental disability including, cerebral palsy, low MDI, visual impairment or auditory impairment (Table 3). The use of prescription glasses was similar in between the groups.

Table 2.

Primary outcome at 18–24 months corrected gestational age for all infants (with known outcomes) in HeLP follow-up study

	wrap group	control group	Odds ratio (95% CI)	Risk difference (95% CI)	p-value (Fisher’s exact)
Primary outcome (Disabled or death) for entire group enrolled in long-term outcome	82/113	80/127	1.55 (0.90–2.69)	+ 10.9% (− 2.4 to + 24.2%)	0.115
Primary outcome (Disabled or death) for subsample of infants < 26 weeks enrolled in long-term outcome	51/67	55/65	0.58 (0.24–1.39)	− 8.5% (− 5.0 to + 22.0%)	0.157

Table 3.

Neurodevelopmental disabilities at 18–24 months corrected gestational age in surviving infants (with known outcomes) in HeLP follow-up study

	wrap group	control group	Odds ratio (95% CI)	Risk difference (95% CI)	p-value (Fisher’s exact)
Any disability	39/70	40/87	1.42 (0.79–2.78)	+ 9.8% (− 5.9 to + 25.4%)	0.293
Cerebral Palsy	9/70	5/87	2.42 (0.77–7.58)	+ 7.1% (− 2.1 to + 16.4%)	0.204
MDI (< 85)	30/53	33/67	1.34 (0.65–2.77)	+ 7.4% (− 10.6 to + 25.3%)	0.538
Blind	0/68	1/86	cannot calculate	− 1.2% (− 3.4 to + 1.1%)	> 0.999
Deaf	5/70	6/87	1.04 (0.95–3.56)	+ 0.3% (− 7.7 to + 8.3%)	> 0.999
Prescription glasses	7/70	2/87	4.72 (0.95–23.50)	+ 7.7% (+ 0.0 to + 15.4%)	0.085

No additional data from community services, child health services or family doctors regarding infants lost to follow-up could be acquired because it was outside the scope of funding for this trial.

Because of the disproportionate loss to follow up affecting the wrap group compared to those in the control group, logistic regression analyses of the primary outcome by randomized group assignment while forcing in each of the first 11 variables from Table 1 (gestational age, waiver of consent, infant sex, birth weight, intrauterine growth restriction, antenatal corticosteroids, chorioamnionitis, sepsis, necrotizing enterocolitis, bronchopulmonary dysplasia treated with steroids or maternal education) might assist with the robustness of observing no difference in the primary outcome in relation to group assignment. Gestational age and birthweight were converted to a categorical variable as noted in Table 4. The unadjusted odds ratio for death or disability in the wrap group compared to the control group was 1.55 (95% CI = 0.90–2.69). When the odds ratios were adjusted for each variables from Table 1 added to logistic regression models, the odds ratios were virtually unchanged.

Table 4.

Logistic regression of primary outcome by group assignment with stated variable forced in the model

	odds ratio	95% CI	p-value
group assignment alone	1.55	0.90–2.69	0.12
with gestational age group*	1.52	0.85–2.69	0.16
with waiver of consent	1.53	0.87–2.68	0.14
with sex of baby	1.57	0.91–2.73	0.11
with birth weight group†	1.32	0.74–2.34	0.35
with intrauterine growth restriction	1.53	0.88–2.65	0.13
with antenatal steroids	1.67	0.95–2.92	0.07
with chorioamnionitis	1.64	0.93–2.87	0.09
with sepsis	1.47	0.84–2.60	0.18
with necrotizing enterocolitis	1.46	0.83–2.55	0.19
with bronchopulmonary dysplasia treated with steroids	1.52	0.87–2.66	0.14
with maternal education‡	1.29	0.72–2.30	0.39

^*gestational age groups (< 24 weeks, 24–25 weeks & ≥ 26 weeks)

^†birth weight groups (< 500 g, 500–749 g, 750–999 g & ≥ 1000 g)

^‡high school or less vs more than high school education

No harms occurred resulting from the follow-up of these infants. As reported in the original HeLP trial, [17] there was a slight increase in hyperthermia in the wrap group.

Discussion

This trial is the first and only report of long-term outcome of delivery room management of VP infants placed in polyethylene occlusive wrap to avoid hypothermia. This randomized controlled trial showed no differences in adverse neurodevelopmental outcome in either the wrap group or the control group.

The strength of this study is that it was a large multicenter randomized controlled trial with 11 of 39 centers reporting on neurodevelopmental outcomes although with only 356 enrolled in the follow-up component of the trial, the power to detect a 50% reduction in death or disability at the 0.05 significance level is only 40%. Unfortunately, a weakness is that the loss to follow-up rate was 32.6% with a significantly higher loss rate noted in the wrap group compared to the control group. Logistic regression analyses of primary outcome by randomized group assignment were performed adjusting for each of the first 11 variables in Table 1. Table 4 shows these regression analyses and indicates that unadjusted odds and adjusted odds for the primary outcome by study group assignment remained virtually unchanged making it impossible to reject the null hypothesis. Individual neurodevelopmental outcomes (Table 3) evaluated in surviving infants enrolled in the follow-up component of HeLP also failed to show any differences. Other outcomes such as growth and general health problems could not be addressed because of the small sample size and missing data points.

Despite the data shown being 15–20 years old, polyethylene occlusive wrap is still a recommended treatment for VP infants in the delivery room [14]. The failure to decrease mortality and long-term disability in this population may be because other supporting techniques to improve survival occur at the same time as occlusive wrapping. If other techniques had been used to impede heat loss, such as use of plastic caps, which improves mortality, or use of heated humidified respiratory gases, which decreases brain injury [15], then the HeLP trial may have shown improved mortality or long-term neurodevelopmental outcomes. The possibility of repeating HeLP is unlikely given the failure to document and improvement in survival [17, 18] or 18–24 month outcomes as noted in this study.

This data is most applicable to patients of the type enrolled in the HeLP trial, namely those from high income countries. In low and middle income countries of the world with less abundance of resources, the neonatal mortality rate can be as much as 40 times higher than that in the countries with the lowest neonatal mortality rate [24]. There may be some opportunity to improve survival and possibly long-term disability where more than half of newborns are hypothermic shortly after birth by including polyethylene occlusive wrap as part of a bundle of care in the delivery room [25].

Conclusions

The original HeLP trial resulted in a significant reduction in hypothermia in very preterm infants; however, the HeLP follow-up trial failed to identify differences in the long-term adverse outcomes in these infants. Nevertheless, occlusive wrap is a simple and inexpensive way to reduce neonatal hypothermia while at the same time resulting in similar long-term outcomes.

Abbreviations

VP

Very preterm

HeLP

Heat Loss Prevention

MDI

Mental Developmental Index (of the Bayley Scales of Infant Development)

NICU

Neonatal Intensive Care Unit

VON

Vermont Oxford Network

CGA

Corrected gestational age

CI

Confidence interval

Appendix

Members of the VON HeLP Trial Study Group include: Co-Principal Investigators: Maureen Reilly, RRT (Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada); Sunita Vohra, MD (Stollery Children’s Hospital, Edmonton, Alberta, Canada). HeLP Trial Steering Committee Members: Michael Dunn, MD, FRCPC (Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Ontario, Canada); Karla Ferrelli, BA (VON, Burlington, VT); Alex Kiss, PhD (Department of Research Design and Biostatistics Scientist, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada); Valeria E. Rac, MD, PhD (The Toronto Health Economics and Technology Assessment, Collaborative, Institute of Health Policy, Management and Evaluation, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada); Roger F. Soll, MD (President and Director of Clinical Trials and Followup, VON; H. Wallace Professor of Neonatology, University of Vermont, Burlington, VT); Michael Vincer, MD, FRCPC (IWK Health Centre, Associate Professor, Dalhousie University, Halifax, Nova Scotia, Canada); John Wimmer, MD (Associate Professor of Pediatrics, University School of Medicine, Greensboro, NC); Denise Zayack, RN, BA (Director, Neonatal Intensive Care Services; Operations Co-Director, Perinatal & Gynaecology Program, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada). Participating centers: Albert Einstein Medical Center, Philadelphia, PA (David L. Schutzman, MD); Aultman Hospital, Canton, OH (Diana L. Hannay, NNP-BC); Avera McKennan Hospital, Sioux Falls, SD (Brenda H. Lewis, MD); Blank Children’s Hospital, Des Moines, IA (Samir Alabsi, MD); Brookdale Hospital Medical Center, Brooklyn, NY (Roger Kim, MD); Caritas St Elizabeth’s Medical Center, Boston, MA (Ronald Pye, MD, Gopal Gupta, MD); Children’s at Cooper University Medical Center, Camden, NJ (Linda Wicker, RN, MSN, CCRN); Children’s Hospital of Wisconsin, Milwaukee, WI (Michael R. Uhing, MD, Laura Lane, RN); CHOC Children’s Hospital, Orange, CA (Sudeep Kukreja, MD, Vijay Dhar, MD); Connecticut Children’s NICU at UConn Health Center, Farmington, CT (Mariann Pappagallo, MD, Lisa Dion, RN, MSN); Elliot Hospital, Manchester, NH (Andrea LaRose, NNP); Goryeb Children’s Hospital, Morristown, NJ (Catherine Sawtell, MSN, CRNP, Melissa Mohn, CCRN); Hennepin County Medical Center, Minneapolis, MN; Henry Ford Hospital, Detroit, MI (Christine Newman, CNNP, Shelley Campbell, RN); Hospital Clinic (sede Maternitat), Barcelona, Spain (Francesc Botet, MD, Josep Figueras-Aloy, MD); Hospital de S Joao, Porto, Portugal (Gustavo Rocha, MD, Hercilia Guimaraes, MD); Hospital de Santa Maria, Lisbon, Portugal (Joana Saldanha, MD, Lincoln Silva, MD); Hospital Universitari Joan XXIII de Tarragona, Tarragona, Spain (Ricardo Closa, MD, Mar Albujar, MD); Jackson Memorial Hospital, Miami, FL (Shahnaz Duara, MD, Karina Lifschitz, MA, MPH); K.K. Women’s & Children’s Hospital, Singapore, Singapore (Mary Fong, CNS, Pratibha K. Agarwal, MD); Lafayette General Medical Center, Layfayette, LA (Jaye Harris, NNP); The Medical Center at Columbus Regional, Columbus, GA (Dana Cason, MD, David H. Levine, MD); Mercy Health Center, Oklahoma City, OK (Lori McDonald, NNP); Mt Carmel Health Systems, Columbus, OH (Peggy O’Connor, RN, Lynda Gage, RN); National Maternity Hospital, Dublin, Ireland (Brid O’Brien, CMS, Geraldine Duffy, RGN, RM, Bsc); Nationwide Children’s Hospital, Columbus, OH (John Seguin, MD, RayeLynn Leukart, NNP); Ochsner Clinic Foundation, New Orleans, LA (Marie McGettigan, MD, Kathy Bell, NNP); Rapides Women’s and Children’s Hospital, Alexandria, LA (Patricia Hope Monk, NP-BC, Heidi Peavy, RN); Royal Maternity Service, Belfast, Ireland (David Sweet, MD, Joanne Hegarty, MD); St Elizabeth Regional Medical Center, Lincoln, NE (BJ Wilson, Jr, MD, Kris Schwarzkopf, RNC, BSN); St John Hospital and Medical Center, Detroit, MI (John T. Adams, MD, Laurie Aman, RN, NNP, MSN); St Luke’s Hospital, Kansas City, MO (Katherine Claflin, MD, Janet Owen, RN); Stamford Hospital, Stamford, CT (Laura Swaan Lasley, MD, Gerald B. Rakos, MD); Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada (Maureen C. Reilly, RRT, Michael Dunn, MD); Temple University Hospital, Philadelphia, PA (Bernice Duesler, MD, Patricia McMahon-Hextron, PA-C); University Kebangsaan Malaysia, Kuala Lumpur, Malaysia (Nem Yun Boo, MD, Jaafar Rohana, MD); University of Iowa Children’s Hospital, Iowa City, IA (Edward F. Bell, MD, Karen J. Johnson, RN, Gretchen A. Cress, RN, BSN); Waukesha Memorial Hospital, Waukesha, WI (Sharon Nelson, NP, Dorothy Koepsell, RN); and Wesley Medical Center, Wichita, KS (Paula Delmore, RNC, MSN, Kathryn Filby, APRN).

Authors’ contributions

MV produced the first draft of this manuscript, oversaw the conduct of this study, analysed the data and is the guarantor of the study. All authors reviewed and approved the final version of the manuscript. MR and SV participated in the conception, design, data acquisition, interpretation and drafting of all manuscripts and manual of operations. MR oversaw study implementation, conduct and monitoring across all participating centers. VER contributed to study implementation and conduct and data acquisition and interpretation. DZ participated in the study design. JW helped in the conception and conduct of the study. KF participated in the study design and data collection. RS contributed to the study design and conduct and reviewed data and analysis plans. MD participated in the study design and conduct and reviewed data and analysis plans. AK was the study statistician.

Funding

During the time of the study, S.V. received salary support as an Alberta Innovates-Health Solutions Health Scholar (G118160495). R.S. received salary support as the director of trials and follow-up at VON. No honorarium, grant or other form of payment was given to produce the manuscript.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study team had the approval from Steering Committee site at Sunnybrook Health Sciences Centre, Toronto, (Research Ethics Board #15–0243-C; 7 October 2015) where the data was stored. Each participating institution submitted a copy of their institution’s Institutional Review Board approval and the need for a consent form or waiver of consent and subsequent annual renewals to the VON HeLP Trial Coordinating Center. This approval process included the enrollment of their child in their local follow-up program (where such a program existed) to evaluate the child between 18–24 months CGA. This work has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.