Downloaded

Improvement of Short- and Long-Term Outcomes for Very Low Birth Weight Infants: Edmonton NIDCAP Trial Kathrine Leigh Peters, Rhonda Jean Rosychuk, Leonora Hendson, Judith Jean Coté, Catherine McPherson and Juzer Mohamed Tyebkhan Pediatrics 2009;124;1009-1020; originally published online Sep 28, 2009; DOI: 10.1542/peds.2008-3808

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://www.pediatrics.org/cgi/content/full/124/4/1009

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication, it has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2009 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from www.pediatrics.org. Provided by Royal Alexandra Hospital on September 28, 2009

ARTICLES

Improvement of Short- and Long-Term Outcomes for Very Low Birth Weight Infants: Edmonton NIDCAP Trial AUTHORS: Kathrine Leigh Peters, PhD, RN,a,b Rhonda Jean Rosychuk, PhD, PStat,c,d Leonora Hendson, MBBCh, MSc, FRCPC,c,d,e,f Judith Jean Cote´, MN, RN,a,e Catherine McPherson, BScN, RN,e and Juzer Mohamed Tyebkhan, MBBS, MRCP, FRCPCc,d,e Faculty of Nursing, bPerinatal Clinical Research Centre, Department of Pediatrics, and dWomen and Children’s Health Research Institute, University of Alberta, Edmonton, Canada; eNorthern Alberta Neonatal Intensive Care Program, Stollery Children’s Hospital, Edmonton, Canada; and fNeonatal and Infant Follow-up Clinic, Glenrose Rehabilitation Hospital, Edmonton, Canada a c

KEY WORDS developmental care, Newborn Individualized Developmental Care and Assessment Program, very low birth weight infant, premature infant, neonatology, neurodevelopmental outcomes ABBREVIATIONS CI— confidence interval CLD— chronic lung disease CP— cerebral palsy CPAP— continuous positive airway pressure HR— hazard ratio LOS—length of stay MDI—Mental Developmental Index NIDCAP—Newborn Individualized Developmental Care and Assessment Program OR— odds ratio RCT—randomized, controlled trial VLBW—very low birth weight This trial has been registered at www.clinicaltrials.gov (identifier NCT00552383). www.pediatrics.org/cgi/doi/10.1542/peds.2008-3808 doi:10.1542/peds.2008-3808 Accepted for publication Apr 9, 2009 Address correspondence to Juzer M. Tyebkhan, MBBS, MRCP, FRCPC, Northern Alberta Neonatal Intensive Care Program, Royal Alexandra Hospital, Room 5027 DTC, 10240 Kingsway Ave, Edmonton, Alberta, T5H 3V9, Canada. E-mail: Juzer.Tyebkhan@ albertahealthservices.ca PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275). Copyright © 2009 by the American Academy of Pediatrics FINANCIAL DISCLOSURE: Dr Peters, Dr Hendson, Ms Cote´, and Dr Tyebkhan have received honoraria and travel expenses for presentation of results at conferences.

WHAT’S KNOWN ON THIS SUBJECT: NIDCAP-based care, with caregiving guided by the preterm infant’s behavior, reduces hospital stays and time of respiratory support. However, longterm developmental outcomes with NIDCAP have not been reported in any large study. WHAT THIS STUDY ADDS: NIDCAP-based care in this large RCT reduced hospital stays and incidence of CLD and also reduced the number of children with cognitive delays at 18 months.

abstract OBJECTIVE: Our objective was to determine the impact of Newborn Individualized Developmental Care and Assessment Program (NIDCAP)based care on length of stay of very low birth weight (VLBW) infants. Secondary outcome measures were days of ventilation, incidence of chronic lung disease, and 18-month neurodevelopmental outcomes. METHODS: This cluster-randomized, controlled trial took place in a large NICU in Canada, with follow-up evaluation at 18 months of age, from September 1999 to September 2004. One hundred VLBW singleton infants and 10 VLBW twin sets were assigned randomly to NIDCAPbased or control care, and 90% participated in follow-up assessments. The intervention was NIDCAP-based care (N ! 56), that is, care by NIDCAP-educated staff members and behavioral observations. The control group (N ! 55) received standard NICU care. Statistical analyses were adjusted for cluster randomization. Although the intervention was not blinded, the pediatricians making the decisions to discharge the infants were not involved in the study, and the follow-up staff members were blinded with respect to group. RESULTS: NIDCAP group infants had reduced length of stay (median: NIDCAP: 74 days; control: 84 days; P ! .003) and incidence of chronic lung disease (NIDCAP: 29%; control: 49%; odds ratio: 0.42 [95% confidence interval: 0.18 – 0.95]; P ! .035). At 18 months of adjusted age, NIDCAP group infants had less disability, specifically mental delay (NIDCAP: 10%; control: 30%; odds ratio: 0.25 [95% confidence interval: 0.08 – 0.82]; P ! .017). CONCLUSION: NIDCAP-based care for VLBW infants improved shortand long-term outcomes significantly. Pediatrics 2009;124:1009–1020

PEDIATRICS Volume 124, Number 4, October 2009


1009

Infant experiences during the NICU stay may contribute to the disturbingly high disability rates for surviving very low birth weight (VLBW) infants.1–5 The uterine environment promotes modulated, rhythmic, and predictable sensory inputs to the fetal brain, to support optimal development and maturation.5,6 The loss of these modulated inputs and separation from the mother as the behavioral regulator7 lead to haphazard sensory input. Together with the physiologic stress of NICU care practices,8 this sensory input may affect brain structure and function, leading to developmental disabilities.8–10 Newborn care that minimizes the impact of the NICU environment and invasive care practices and encourages parental participation may promote more-appropriate brain development. This approach is termed developmental care. The Newborn Individualized Developmental Care and Assessment Program (NIDCAP) is one developmental care framework.7,11 Infant behavior is conceptualized as the expression of brain function. When parents and staff members synchronize caregiving to the infant’s behavior and therefore brain function, ongoing brain development should be supported.9,12,13 Infant behavior is described by the synactive theory of development of Als7 as a continuous, reciprocal interaction among infant behavioral subsystems (autonomic, motor, state, and attention/interaction), environment, family, and caregivers. Peters14 showed a significant interaction between autonomic and motor levels of behavioral function in the clinical setting. Care that supports !1 of these subsystems (eg, environmental or motor support) promotes overall stability and should improve neurobehavioral outcomes. Previous studies showed that NIDCAP for VLBW infants reduced hospital length of stay (LOS),12,15–17 duration of 1010

PETERS et al

ventilation,15–18 duration of continuous positive airway pressure (CPAP) treatment,18,19 duration of supplemental oxygen treatment,15,16,19 and NICU hospital costs.12,15,16,18,20 Criticisms of these studies21–25 include small sample size, performance before the routine use of surfactant and prenatal corticosteroid therapy,15,16 and nonrandomized design.15,17,26,27 The 5 published randomized, controlled trials (RCTs) are summarized in Table 1. The latest Cochrane review28 found that NIDCAP group infants had significantly fewer days of ventilation, fewer days of CPAP therapy, and reduced rates of chronic lung disease (CLD). Although the meta-analysis found no difference for LOS, 1 of the 3 studies in the meta-analysis included more-mature infants, for whom reduction of LOS would not be expected.9 Positive effects of NIDCAP use include higher Bayley scores at 9 months9,16 and 12 months.19,29,30 Only 2 studies, 1 of which was not a RCT, reported developmental outcomes after 18 months, with marginal positive effects on behavior.31,32 LOS was selected as the primary outcome measure, because it reflects optimal behavioral functioning in preterm infants. Criteria determining LOS are the infant’s ability to regulate temperature, freedom from apnea and bradycardia (ie, autonomic stability), ability to demand feeding and to gain weight (ie, motor and state stability), and integration into the family (ie, attention/interaction). If NIDCAP promotes better functioning in these behavioral subsystems, then infants should reach discharge criteria sooner.

METHODS Hypotheses The primary hypothesis was that NIDCAP would result in shorter LOS. Secondary hypotheses were that NIDCAP would result in fewer days of ventila-

tion, CPAP therapy, and supplemental oxygen treatment, would reduce the incidence of CLD, and would improve neurodevelopmental outcomes at adjusted age of 18 months. Study Design This cluster RCT was conducted at the Royal Alexandra Hospital (Edmonton, Canada). The University of Alberta health research ethics board approved the study protocol. Part I (May 1998 to September 1999) involved the education of 52 volunteer NICU staff members (of "250 staff members) in providing NIDCAP-based care, according to the requirements of the NIDCAP training program.33 This volunteer group included senior and junior neonatal nurses, 1 neonatal nurse practitioner, and 2 physicians. Two staff nurses, Dr Hendson, Ms Cote´, and Dr Tyebkhan were certified to reliability in NIDCAP behavioral observation and care plan development. Part II (September 1999 to September 2004) consisted of the RCT and neurodevelopmental follow-up monitoring. Participants Inclusion criteria were birth weight of 500 to 1250 g, gestational age of "32 weeks, birth weight between the 3rd and 97th percentiles for gestational age,34 and age of 2 to 7 days at the time of study entry. Exclusion criteria were chromosomal or major congenital anomalies, known maternal alcohol or drug abuse, known congenital infection, and/or decision to withdraw life support before 48 hours of life. Twins were enrolled as 1 cluster if both infants met all criteria. Randomization Parents of eligible infants were approached for written informed consent. Randomization was performed with computer-generated random numbers, which were held in sequentially numbered, sealed, opaque enve-




1992–1994

1994–1997

1990–1992

2000–2002

Fleischer et al18

Westrup et al19

Als et al12

Als et al9

16/14

45/47

12/13

17/18

20/18

No. of Infants, NIDCAP/ Control

1648/1730

800/850 (approximate values; 3-center RCT with data reported for each center separately)

1083/840

893/815

872/862

Birth Weight, NIDCAP/ Control, Mean, g

31.2/31.8

26/26 (approximate values; 3-center RCT with data reported for each center separately)

27.6/26.1

26.5/26.1

27.1/26.5

Gestational Age, NIDCAP/ Control, Mean, wk

None; not designed to demonstrate differences in short-term outcomes (low-risk preterm infants)

Reduced LOS, days of ventilation, days of parenteral nutrition, days of gavage feeding, incidence of NEC, and hospital costs; improved growth

No significant differences between groups

Reduced LOS, days of oxygen therapy, days of gavage feeding, incidence of IVH, incidence of CLD, and hospital costs Reduced days of CPAP therapy

Statistically Significant Results (Short Term)a

IVH indicates intraventricular hemorrhage; APIB, Assessment of Preterm Infants’ Behavior; EEG, electroencephalography. a Short- and long-term results for the NIDCAP group.

Not stated

Date of RCT

Als et al15

Authors

TABLE 1 RCTs of NIDCAP

2 wk and 9 mo after due date

2 wk after due date

69.9 mo; NIDCAP: n ! 11; control: n ! 1532

2 wk after due date

9 mo after due date

Age at Postdischarge Follow-up Assessment

More-mature EEG and MRI findings, improved APIB and Prechtl scores at 2 wk; improved Bayley scale II scores at 9 mo

Improved APIB scores; reduced parental stress

Improved odds for survival with normal behavior

Improved APIB scores

Improved APIB and Bayley scale scores

Statistically Significant Results (Long Term)a

6 of 17 NIDCAP group infants and 5 of 18 control infants received prenatal steroid treatment 9 of 12 NIDCAP group infants and 8 of 13 control infants received prenatal steroid treatment; study halted 20 of 45 NIDCAP group infants and 24 of 47 control infants received prenatal steroid treatment; 27 of 45 NIDCAP group infants and 35 of 47 control infants received surfactant; differences persisted after adjustment for site 11 of 16 NIDCAP group infants and 13 of 14 control infants received prenatal steroid treatment; only 5 NIDCAPtreated infants and 4 control infants required surfactant

Only 4 infants (NIDCAP: n ! 3; control: n ! 1) received prenatal steroid treatment; no surfactant

Comments

ARTICLES

1011

lopes kept in a locked office. Stratification and block randomization were not used. Study Intervention NIDCAP-based care is guided by the infant and requires caregivers to respond to individual behaviors, to be more flexible in caregiving practices, to modify the environment, and to encourage parental involvement. Detailed descriptions of how NIDCAPbased care differs from traditional NICU care were provided by Als et al9,15 and Buehler et al.13 Behavioral observations facilitate formulation of descriptive care plans to adapt caregiving to the individual infant.9 These observations are performed before, during, and after a planned caregiving procedure, by scoring autonomic, motor, state, and attention/interaction behaviors at 2-minute intervals. Behaviors are conceptualized as either disorganized (eg, flaccidity, agitated movements, respiratory pauses, finger splaying, or gaze aversion) or regulatory (eg, hand-to-mouth movements, hand clasping, grasping, efforts to suck, or tucking). This process allows staff members and parents to pace caregiving in synchrony with the infant’s behavioral readiness.9 NIDCAP reaffirms the parents as their child’s consistent caregivers by actively including them in caregiving. NIDCAP cannot be a blinded intervention, because the environment is modified to reduce inappropriate stimulation, the infant’s selfregulatory efforts are supported to help the infant gain or regain balance, and appropriate social stimulation is promoted. This study intervention required infant care by NIDCAP-educated nurses for !50% of the time, plus behavioral observations and care plans performed by NIDCAP-certified staff members approximately every 2 weeks while at the study site. Control infants were to re1012

PETERS et al

ceive no care by the NIDCAP-educated nurses and no behavioral observations/care plans. The 2 groups of infants were cared for in the same nursery, sometimes side by side. Incubator covers, positioning devices, and kangaroo care were available for all infants but were used only at nurses’ discretion for control infants. Such basic developmental care may have little effect on LOS.35 Attending neonatologists directed routine care for all infants in both groups, with assistance from neonatal nurse practitioners and neonatal fellows. During part II of the study, there were 26 neonatologists, neonatal nurse practitioners, and fellows working in this NICU. The 5 NIDCAP-certified staff members made suggestions for modification of care according to the NIDCAP behavioral observations only for NIDCAP group infants. Infants were transferred to intermediate-care nurseries when they no longer required nasal CPAP therapy and parenteral nutrition, if the study site nursery was full, or if it was more convenient for the family. After transfer, NIDCAP group infants no longer received the intervention. All infants were tracked prospectively by telephone until discharge. Infants who stayed at the study site until discharge were transferred to pediatricians for routine care for the last part of their NICU stay. Forty-five pediatricians were practicing in this region (study site and intermediate-care nurseries) in part II of the study. They received no NIDCAP education, were not involved in any part of the RCT, and were not aware of the outcome measures. They made the final decisions to discharge infants according to policies at all nurseries at that time. Discharge criteria were that infants no longer required incubator care, had no symptomatic apnea, and were being fully orally fed and gaining weight.

The follow-up visit at adjusted age of 18 months included a neurologic examination, visual assessment, and auditory assessment by certified staff members and a neurodevelopmental assessment by a certified developmental psychologist, with the Bayley Scales of Infant Development II.36 Follow-up evaluations were performed prospectively, as described previously,37 and examiners were blinded to group assignment. Operational Definitions LOS was defined as the number of calendar days spent in the hospital. Days of ventilation indicated the number of days on which mechanical ventilation was required. Days of CPAP therapy indicated the number of days on which nasal CPAP and/or high-flow nasal cannula oxygen therapy38 was required, not including days on which both ventilation and CPAP therapy were required. Days of supplemental oxygen indicated the number of days on which supplemental oxygen was required, not including days on which ventilation and/or CPAP therapy also were required. CLD indicated the need for supplemental oxygen to maintain oxygen saturation levels of 92% to 96%, at gestational age of 36 weeks. Disability at adjusted age of 18 months was recorded when a child had !1 of cerebral palsy (CP) of any type or severity,39 visual impairment (corrected visual acuity in the better eye of #20/60), binaural/bilateral sensorineural hearing loss of $40 dB at any frequency from 250 to 4000 Hz, or mental delay. Moderate mental delay was defined as Mental Developmental Index (MDI) score of #70 and severe mental delay as MDI score of #55 on the Bayley Scales of Infant Development II.36 Statistical Analyses The data were analyzed on an intention-to-treat basis. Data are described as mean % SD, median and


ARTICLES

range, or frequency and proportion. Two-sample t tests (or Wilcoxon ranksum tests) and #2 tests (or Fisher’s exact tests) were used to compare groups for cluster-level continuous and categorical variables, respectively. Subject-level variables were compared with cluster-adjusted t tests,40 Wilcoxon rank-sum tests,41 or #2 tests.40 Kaplan-Meier curves displayed time-to-event variables. Cox proportional-hazard regression techniques were used to assess group differences in time-to-event variables, with a robust variance estimator.40 Deaths and withdrawals were treated as censored events. Multivariate proportional-hazard models were developed for LOS, to assess the effect of group with adjustment for baseline variables. Gestational age, rather than birth weight, was entered because of strong correlation. Models were assessed for adequacy of the proportional-hazard assumption42 and fit (eg, deviance and score residuals). Hazard ratios (HRs) and odds ratios (ORs) were provided with 95% confidence intervals (CIs). S-PLUS 8.0 for Windows (TIBCO Software Inc, Palo Alto, CA) was used for analyses, and 2-sided P values of #.05 were considered statistically significant. Sample Size Fifty-five infants per group, on the basis of a clinically significant 15% reduction in LOS, assuming $ of .05 and power of 0.80, were required. Given that the median LOS for infants admitted the year before part I was 90 days, this translated to a clinically meaningful reduction of 13 days. Sixty infants per group were recruited.

RESULTS During part II, 328 infants with birth weights of #1250 g were admitted to the study site, 201 infants were eligible, and 120 infants enrolled, with 60 infants (50 singletons and 5 sets of PEDIATRICS Volume 124, Number 4, October 2009

FIGURE 1

Study flow. *Parents of the control infant who was initially withdrawn agreed to participate in the 18-month evaluation.

twins) in each group (Fig 1). While at the study site, NIDCAP group infants received a median of 83% (range: 49%– 95%) of total nursing care hours from NIDCAP-educated nurses, whereas all except 1 control infant received 0% (range: 0%–27%) NIDCAP-educated nursing care hours. The 5 NIDCAP-certified staff members completed 1 to 7 behavioral observations per NIDCAP group infant (median: 3 observations) and actively encouraged the parents to participate in ward rounds and ongoing care. No control infant was assessed by using behavioral observations, and participation in care for control group parents was at staff members’ discretion.

No evidence of a statistically significant difference between groups was found for any maternal or infant demographic characteristic (Table 2). Four infants in each group died (1 from each group as a result of intraventricular hemorrhage, meningitis, and CLD; 1 control infant as a result of necrotizing enterocolitis; and 1 NIDCAP group infant as a result of aspiration pneumonia), and 1 infant in the control group was withdrawn at parental request on day 8, leaving 55 control and 56 NIDCAP group infants with complete data sets. No evidence of a statistically significant difference was found for any clinical course variable (Table 3) except for the use of inotropes, which


1013

TABLE 2 Parent and Patient Characteristics at Baseline (Randomization) Cluster-level information Maternal age, y Mean % SD Median (range) Gravidity Mean Median (range) Parity Mean Median (range) Blishen scoreb Mean Median (range) Prenatal steroid use, n (%) Inborn, n (%) Gestational age, wk Mean % SD Median (range) Age at randomization, d Mean % SD Median (range) Subject-level information Cesarean section, n (%) Birth weight, g Mean % SD Median (range) Male, n (%) Apgar score at 1 min Mean % SD Median (range) Apgar score at 5 min Mean % SD Median (range) SNAPPE-II score Mean % SD Median (range) SNAPPE-II–predicted mortality rate, % Mean Median (range) Ventilator support at randomization, n/N (%) Time ventilated at randomization, d Mean Median (range)

NIDCAP

Control

N ! 55

N ! 55

P

27.4 % 5.6 27 (16–42)

26.6 % 6.3 25 (14–40)

.49

2.8 2 (1–10)

2.3 2 (1–6)

.32a

0.6 0 (0–4) N ! 45 42.1 41 (21–101) 45 (82) 51 (93)

0.5 0 (0–4) N ! 49 42.7 39 (21–101) 49 (89) 54 (98)

27.5 % 1.4 28 (25–30)

27.0 % 2.3 27 (23–32)

.18

4.1 % 1.3 4 (2–7) N ! 60 33 (55)

4.1 % 1.3 4 (2–7) N ! 60 30 (50)

.93

988.2 % 183.7 1005 (570–1250) 31 (52)

927.1 % 204.0 960 (550–1240) 29 (48)

.11

5.2 % 2.0 5 (1–9)

5.0 % 2.1 5 (1–9)

.73

7.3 % 1.2 8 (3–10)

7.3 % 1.5 8 (2–10)

.90

22.7 19 (0–68)

26.4 25 (0–77)

.41a

.84a .42 .36c

.61

.72

.34a

7.5 2.2 (0.5–32.3) 23/60 (38)

9.0 4.9 (0.5–43.8) 31/60 (52)

.45a

1.6 1 (0–5)

2.1 2 (0–6)

.16a

.16

SNAPPE-II indicates Score for Neonatal Acute Physiology Perinatal Extension II (measure of newborn illness severity and mortality risk).55 a Highly skewed data; the Wilcoxon rank-sum test or clustered Wilcoxon rank-sum test was used. b Blishen scores are Canadian scores for socioeconomic status.56 c Fisher’s exact test.

were used for more control infants (NIDCAP: n ! 18; control: n ! 28; P ! .046). The median LOS was significantly reduced for NIDCAP group infants (NIDCAP: 74.0 days; control: 84.0 days; P ! .003) (Table 4 and Fig 2). The estimated HR was 1.83 (95% CI: 1.24 –2.72). When other important factors such as gestational age and days of ventilation 1014

PETERS et al

before randomization were considered (Table 5), LOS was still significantly reduced for NIDCAP group infants (HR: 1.65 [95% CI: 1.05–2.59]). Similar numbers of patients were discharged while receiving methylxanthines (NIDCAP: n ! 35; control: n ! 37) and/or oxygen therapy (NIDCAP: n ! 3; control: n ! 4). Similar numbers of patients were discharged from

the study site and from other nurseries (from the study site: NIDCAP: n ! 19; control: n ! 15; from other nurseries: NIDCAP: n ! 37; control: n ! 40). For the 54 infants who were undergoing ventilation at the time of randomization, the median time of ventilation after randomization for NIDCAP-treated infants was 16 days and that for control infants was 27 days (Table 3 and Fig 3). However, there was no evidence of a statistically significant difference (HR: 1.68 [95% CI: 0.87–3.27]; P ! .12) because of 1 NIDCAP group infant who underwent ventilation for 81 days before death. An additional 6 infants (NIDCAP: n ! 5; control: n ! 1) were not undergoing ventilation before randomization but required ventilation because of apnea after randomization. The control infant required 7 days of ventilation and the NIDCAP group infants required a median of 4 days (range: 1–5 days). No evidence of a statistically significant difference between the 2 groups was found for days of CPAP therapy or supplemental oxygen therapy (Table 3). Fewer surviving NIDCAP group infants had CLD, compared with surviving control infants (NIDCAP: 29%; control: 49%; OR: 0.42 [95% CI: 0.18 – 0.95]; P ! .035). In addition, fewer ventilated, surviving, NIDCAP group infants had CLD (NIDCAP: 38%; control: 66%; OR: 0.31 [95% CI: 0.12– 0.77]; P ! .011). Fifty-one NIDCAP group infants and 50 control infants (91%) were available for follow-up evaluations at 18 months of adjusted age. One control infant died after discharge, as a result of complications related to short-bowel syndrome, and 5 infants in each group were lost to follow-up monitoring. Parents of the control infant who was initially withdrawn agreed to participate for the 18-month evaluation. The incidence of overall disability, specifically moderate mental delay, was lower in the NIDCAP group (NIDCAP: 10%; con-


ARTICLES

TABLE 3 Neonatal Clinical Course n/N (%) All infants Ventilator support Surfactant use for ventilated infants Inotrope use Survivors only IVH with no ventriculomegaly IVH with ventriculomegaly and/or IPED Sepsisa Severe ROP Methylxanthine use Dexamethasone use

P

OR (95% CI)

NIDCAP

Control

47/60 (78) 36/47 (77) 18/60 (30)

43/60 (72) 35/43 (81) 29/60 (48)

.42 .60 .05

1.43 (0.60–3.39) 0.75 (0.25–2.20) 0.46 (0.21–0.99)

7/56 (13) 2/56 (4) 20/56 (36) 6/56 (11) 56/56 (100) 4/56 (7)

7/55 (13) 1/55 (2) 23/55 (42) 12/54 (22) 54/55 (98) 7/55 (13)

.97 .57 .52 .11 .35 .33

0.98 (0.30–3.17) 2.00 (0.18–22.72) 0.77 (0.36–1.68) 0.42 (0.14–1.25) 0.53 (0.15–1.92)

IVH indicates intraventricular hemorrhage; IPED, intraparenchymal echodensity; ROP, retinopathy of prematurity (stage !3). a Sepsis was defined through positive blood culture results.

TABLE 4 Short-Term Outcomes

LOS, d All, median Survivors onlya Mean Median (range) Censored onlyb Censored times, d Ventilation for subjects ventilated at randomization, d All, median Survivors only Mean Median (range) Censored only Censored times, d CPAP treatment, d All, median Survivors only Mean Median (range) Censored only Censored times, d Supplemental oxygen therapy, d All, median Survivors only Mean Median (range) Censored only Censored times, d CLD, n/N (%) All living infants Ventilated living infants

NIDCAP (N ! 60)

Control (N ! 60)

P

74.0 N ! 56 75.5 71.5 (40–126) n!4 15, 22, 23, 82 N ! 23

84.0 N ! 55 90.2 84.0 (32–169) n!5 8,c 12, 19, 34, 37 N ! 31

.003

16 N ! 19 14.7 12 (0–47) n!4 5, 13, 21, 81

27 N ! 26 31.0 24 (0–103) n!5 0,c 9, 14, 15, 32

.12

30.4 N ! 56 29.0 30.0 (0–69) n!4 0, 0, 0.3, 15.0

27.9 N ! 55 28.1 27.9 (0–70) n!5 0,c 0, 0, 0, 13.8

.960

7.0 N ! 56 13.5 6.5 (0–98) n!4 0, 0, 0, 1

10.0 N ! 55 15.2 10.0 (0–72) n!6 0,c 0, 0, 0, 0, 0

.590

16/56 (29) 16/43 (38)

27/55 (49) 25/38 (66)

.04 .01

Survivors with complete data. Censored for death, withdrawal, or missing data. c For the withdrawn infant. a

b

trol: 30%; OR: 0.25 [95% CI: 0.08 – 0.82]; P ! .017) (Table 6). The MDI score of the NIDCAP group (mean: 85.1; SD: 15.3) was higher than that of the control PEDIATRICS Volume 124, Number 4, October 2009

group (mean: 79.5; SD: 18.3; difference: 5.60 [95% CI: &1.49 to 12.7]; P ! .120), but the difference did not reach statistical significance. No evidence of a sta-

tistically significant difference in Psychomotor Developmental Index scores was found (mean % SD: NIDCAP: 85.1 % 15.4; control: 81.1 % 15.3; difference: 3.95 [95% CI: &2.78 to 10.69]; P ! .246). The rate of severe disability (!1 of MDI score of #55, CP, visual impairment, or sensorineural hearing loss) was lower in the NIDCAP group (NIDCAP: 6%; control: 20%; OR: 0.25 [95% CI: 0.06 – 0.97]; P ! .034).

DISCUSSION This RCT shows significant reductions in LOS, incidence of CLD, and incidence of developmental disability for infants who receive NIDCAP-based care. Results from earlier studies12,15,16,18,19 are replicated with a sample reflective of current practice; 85% of the study infants received prenatal steroid therapy, and 76% of those undergoing ventilation received surfactant. The effect of NIDCAP remained significant for LOS after adjustment for predictor variables. Gestational age, Score for Neonatal Acute Physiology Perinatal Extension II values, days of ventilation before randomization, and male gender were significant predictors for LOS, as expected. Further multivariate analysis for the 18-month outcomes was not performed, because the sample size was not powered for it. At the time of randomization, 54 infants were still undergoing ventilation. With this reduced subset, no evidence of a statistically significant reduction in days of ventilation for NIDCAP group infants was found. One NIDCAP group infant, with severe cystic bronchopulmonary dysplasia that was unresponsive to maximal treatment, underwent prolonged ventilation until death at 81 days. The parents were not able to accept the NICU team’s recommendation, made some weeks before death, to withdraw life support. When this infant was removed from the analysis, there was a statistically significant difference


1015

TABLE 6 Neurodevelopmental Outcomes at 18 Months n (%)

Any disability MDI score of #70 CP Sensorineural hearing loss Visual impairment Severe disability

P

NIDCAP (N ! 51)

Control (N ! 50)

5 (10) 5 (10) 0 (0) 0 (0)

15 (30) 15 (30) 3 (6) 2 (4)

.02 .02 .11 .19

0 (0) 3 (6)

0 (0) 10 (20)

NA .03

Severe disability indicates !1 of MDI score of #55, CP, visual impairment, or sensorineural hearing loss; NA, not applicable.

FIGURE 2

LOS according to group.

TABLE 5 Multivariate Cox Proportional-Hazard Model for LOS Variable

Coefficient, Estimate % SE

P

HR (95% CI)

NIDCAP Gestational age Log (SNAPPE-II–predicted mortality rate) Ventilator supported at randomization Time ventilated at randomization Male

0.501 % 0.229 0.478 % 0.082 &0.294 % 0.100 &0.976 % 0.311 &0.221 % 0.103 &0.471 % 0.228

.03 #.001 .00 .002 .03 .04

1.65 (1.05–2.59) 1.61 (1.37–1.89) 0.75 (0.61–0.91) 0.38 (0.21–0.69) 0.80 (0.66–0.98) 0.62 (0.40–0.98)

SNAPPE-II indicates Score for Neonatal Acute Physiology Perinatal Extension II (measure of newborn illness severity and mortality risk).55

FIGURE 3

Days of ventilation according to group, for infants who were undergoing ventilation at the time of randomization.

favoring NIDCAP (unadjusted HR: 2.41 [95% CI: 1.32– 4.38]; P ! .004). The trend of fewer days of ventilation for the NIDCAP group is similar to the results of previous studies.15–18 1016

PETERS et al

No evidence of a statistically significant difference between the 2 groups was found for days of CPAP therapy and supplemental oxygen treatment, which indicates that the trend of re-

duced duration of ventilation was likely not offset by increased duration of other forms of respiratory support. The incidence of CLD also was significantly reduced. It is postulated that the enhanced behavioral stability with NIDCAP helped NIDCAP group infants regulate autonomic behavior (ie, breathing), leading to these outcomes. NIDCAP group infants demonstrated improved neurodevelopmental outcomes at 18 months of adjusted age (Table 6). CP, hearing loss, and visual loss were not seen among NIDCAP group infants. The incidence of moderate mental delay (MDI score of #70), the most common developmental problem for VLBW infants,1,2 was significantly reduced. The difference between the 2 groups in mean MDI scores at 18 months was not statistically significant, but this RCT was not powered to detect such a difference. Although some children with MDI scores of #70 shift into the normal range with time, this often is only into the low-average range (MDI scores of 70 – 84), and the children remain at significant academic disadvantage.43,44 MDI scores of #70 at 18 months are clinically significant; in this region, such children often qualify for government-funded early education. Until now, prenatal corticosteroid treatment has been the only intervention with a documented, long-term,


ARTICLES

neurodevelopmental advantage for VLBW infants.2 Surfactant treatment for respiratory distress syndrome has led to increased survival rates but not better neurodevelopmental outcomes.45 Studies of other neonatal interventions designed to demonstrate improved neurodevelopmental outcomes, such as prophylactic indomethacin treatment46 and low-dose, postnatal, corticosteroid treatment,47 found no differences. NIDCAP, however, by enabling caregivers (including parents) to work “in synchrony” with the infants, led to improved neurodevelopmental outcomes at 18 months. In NIDCAP, caregivers (including parents) learn to recognize states of alertness and sleep, and they time caregiving to when the infant is ready for interaction.9 The synactive model suggests that this should lead to better brain organization, and NIDCAP has been shown to optimize behavioral functioning of preterm infants.9,12,13,15,16,32 It is inferred that this improved behavioral outcome has a structural equivalent. Early experiences, whether pleasant or noxious, influence synaptic formation and cortical development.6,8 These experiences are both physical and emotional. The absence of parental emotional support for VLBW infants in the traditional NICU, together with the physical NICU environment, may contribute to differences in brain structure and function of preterm infants.10,48 Parental involvement supports positive physical and emotional experiences. Teaching parents how to interpret their infants’ behavior has long been known to enhance parenting skills49 and should improve long-term outcomes. Als et al9 provided evidence, by using electroencephalographic cortical spec-


tral coherence imaging and MRI, of enhanced brain structure and function for infants who received NIDCAP. Those findings, together with the results of this RCT, provide support for the synactive model and the continuous, reciprocal interaction between the caregivers, the environment, and behavioral subsystems. Although a prospective RCT design was used, the limitations of this study are the unblinded intervention and possible volunteer bias. Interventions such as high-frequency oscillatory ventilation50–52 and cesarean section because of breech presentation53 are other examples of treatments that cannot be blinded but have been well studied in RCTs. A meta-epidemiological study showed that there is little evidence of exaggerated treatment effect for unblinded intervention studies with adequate allocation concealment and objectively determined outcomes.54 Concealed, random allocation was used, and objective outcomes were determined by clinicians blinded to the study group (ie, LOS by pediatricians not involved in the RCT and 18-month outcomes by the follow-up team). There might have been immeasurable differences between the volunteers and their peers, leading to subtle modifications of caregiving by the NIDCAP (volunteer) nurses, for whom the NIDCAP approach might have been intuitive. However, this study intervention involved changes in caregiving that were based on behavioral observations, which were performed only for NIDCAP group infants, and the ability to understand and to respond to preterm infant behavior, which was provided by NIDCAP education. It is unlikely that these results were attributable simply to inherent differences in the nurses.

NIDCAP is an intervention that includes caregivers, education, and caregiving practices in the NICU, and the effects of the components cannot be separated. A multicenter RCT in which entire NICUs are assigned randomly to NIDCAP or control conditions could be conducted to overcome these limitations but would have major financial and logistic requirements. Even so, volunteer bias would not be eliminated, because education and implementation would require voluntary staff member participation at each participating site.

CONCLUSION In this RCT, NIDCAP-based care for VLBW infants improved LOS, CLD incidence, and neurodevelopmental outcomes at 18 months of adjusted age significantly. This study provides additional evidence to support NIDCAPbased care.

ACKNOWLEDGMENTS Funding was provided by the Alberta Heritage Foundation of Medical Research; Canadian Lung Association; Canadian Nurses Respiratory Society; Alberta Lung Association; Perinatal Clinical Research Centre, University of Alberta; and Neonatal Research Trust Fund. Dr Rosychuk is salary-supported as a health scholar from the Alberta Heritage Foundation for Medical Research (Edmonton, Canada). We thank all of the staff members of the Royal Alexandra Hospital NICU who participated in this study; the infants and their families who participated in this study; Jean Gardner Cole, Boston Medical Center (Boston, MA), our NIDCAP trainer; and the Perinatal Clinical Research Centre, University of Alberta, for assistance and support during this study.


1017

REFERENCES 1. Marlow N, Wolke D, Bracewell MA, Samara M; EPICure Study Group. Neurologic and developmental disability at six years of age after extremely preterm birth. N Engl J Med. 2005;352(1):9 –19 2. Vohr BR, Wright LL, Poole WK, McDonald SA. Neurodevelopmental outcomes of extremely low birth weight infants #32 weeks’ gestation between 1993 and 1998. Pediatrics. 2005;116(3):635– 643 3. Anderson PJ, Doyle LW; Victorian Infant Collaborative Study Group. Executive functioning in schoolaged children who were born very preterm or with extremely low birth weight in the 1990s. Pediatrics. 2004;114(1):50 –57 4. American Academy of Pediatrics, Committee on Environmental Health. Noise: a hazard for the fetus and newborn. Pediatrics. 1997;100(4):724 –727 5. White RD, ed. The Sensory Environment of the NICU: Scientific and Design-Related Aspects. Philadelphia, PA: Saunders; 2004 6. Eisenberg L. Experience, brain, and behavior: the importance of a head start. Pediatrics. 1999; 103(5):1031–1035 7. Als H. Reading the premature infant. In: Goldson E, ed. Nurturing the Premature Infant. Oxford, England: Oxford University Press; 1999:18 – 85 8. Anand KJ. Pain, plasticity, and premature birth: a prescription for permanent suffering? Nat Med. 2000;6(9):971–973 9. Als H, Duffy FH, McAnulty GB, et al. Early experience alters brain function and structure. Pediatrics. 2004;113(4):846 – 857 10. Inder TE, Warfield SK, Wang H, Huppi PS, Volpe JJ. Abnormal cerebral structure is present at term in premature infants. Pediatrics. 2005;115(2):286 –294 11. Als H. Toward a synactive theory of development: promise for the assessment and support of infant individuality. Infant Ment Health J. 1982;3(4):229 –243 12. Als H, Gilkerson L, Duffy FH, et al. A three-center, randomized, controlled trial of individualized developmental care for very low birth weight preterm infants: medical, neurodevelopmental, parenting, and caregiving effects. J Dev Behav Pediatr. 2003;24(6):399 – 408 13. Buehler DM, Als H, Duffy FH, McAnulty GB, Liederman, J. Effectiveness of individualized developmental care for low-risk preterm infants: behavioral and electrophysiologic evidence. Pediatrics. 1995;96(5):923–932 14. Peters K. Association between autonomic and motoric systems in the preterm infant. Clin Nurs Res. 2001;10(1):82–90 15. Als H, Lawhon G, Brown E, et al. Individualized behavioral and environmental care for the very low birth weight preterm infant at high risk for bronchopulmonary dysplasia: neonatal intensive care unit and developmental outcome. Pediatrics. 1986;78(6):1123–1132 16. Als H, Lawhon G, Duffy FH, McAnulty GB, Gibes-Grossman R, Blickman JG. Individualized developmental care for the very low-birth-weight preterm infant: medical and neurofunctional effects. JAMA. 1994;272(11):853– 858 17. Brown LD, Heermann JA. The effect of developmental care on preterm infant outcome. Appl Nurs Res. 1997;10(4):190 –197 18. Fleisher BE, VandenBerg K, Constantinou J, et al. Individualized developmental care for very-lowbirth-weight premature infants. Clin Pediatr (Phila). 1995;34(10):523–529 19. Westrup B, Kleberg A, von Eichwald K, Stjernqvist K, Lagercrantz H. A randomized, controlled trial to evaluate the effects of the Newborn Individualized Developmental Care and Assessment Program in a Swedish setting. Pediatrics. 2000;105(1):66 –72 20. Petryshen P, Stevens B, Hawkins J, Stewart M. Comparing nursing costs for preterm infants receiving conventional vs. developmental care. Nurs Econ. 1997;15(3):138 –145, 150 21. Garland JS. Developmental care for very low-birth-weight infants [letter]. JAMA. 1995;273(20): 1575 22. Lacy JB. Developmental care for very low-birth-weight infants [letter]. JAMA. 1995;273(20): 1575–1576 23. Ohlsson A. Developmental care for very low-birth-weight infants [letter]. JAMA. 1995;273(20):1576 24. Saigal S, Streiner D. Developmental care for very low-birth-weight infants [letter]. JAMA. 1995; 273(20):1576 –1577 25. Sepkowitz S. Developmental care for very low-birth-weight infants [letter]. JAMA. 1995;273(20): 1577 26. Stevens B, Petryshen P, Hawkins J, Smith B, Taylor P. Developmental versus conventional care: a comparison of clinical outcomes for very low birth weight infants. Can J Nurs Res. 1996;28(4): 97–113 27. Westrup B, Kleberg A, Wallin L, Lagercrantz H, Wikblad K, Stjernqvist K. Evaluation of the Newborn

1018

PETERS et al


ARTICLES

28. 29.

30.

31.

32.

33. 34. 35.

36. 37. 38.

39. 40. 41. 42. 43.

44. 45. 46. 47.

48. 49. 50.

51. 52.

Individualized Developmental Care and Assessment Program (NIDCAP) in a Swedish setting. Prenat Neonat Med. 1997;2:366 –375 Symington A, Pinelli J. Developmental care for promoting development and preventing morbidity in preterm infants. Cochrane Database Syst Rev. 2006;(2):CD001814 Ariagno RL, Thoman EB, Boeddiker MA, et al. Developmental care does not alter sleep and development of premature infants. Pediatrics. 1997;100(6). Available at: www.pediatrics.org/cgi/ content/full/100/6/e9 Kleberg A, Westrup B, Stjernqvist K, Lagercrantz H. Indications of improved cognitive development at one year of age among infants born very prematurely who received care based on the Newborn Individualized Developmental Care and Assessment Program (NIDCAP). Early Hum Dev. 2002;68(2): 83–91 Kleberg A, Westrup B, Stjernqvist K. Developmental outcome, child behaviour and mother-child interaction at 3 years of age following Newborn Individualized Developmental Care and Intervention Program (NIDCAP) intervention. Early Hum Dev. 2000;60(2):123–135 Westrup B, Bohm B, Lagercrantz H, Stjernqvist K. Preschool outcome in children born very prematurely and cared for according to the Newborn Individualized Developmental Care and Assessment Program (NIDCAP). Acta Paediatr. 2004;93(4):498 –507 Als H. Program Guide: Newborn Individualized Developmental Care and Assessment Program (NIDCAP). Boston, MA: NIDCAP Federation International; 1997 Arbuckle TE, Wilkins R, Sherman GJ. Birth weight percentiles by gestational age in Canada. Obstet Gynecol. 1993;81(1):39 – 48 Maguire CM, Veen S, Sprij AJ, et al. Effects of basic developmental care on neonatal morbidity, neuromotor development, and growth at term age of infants who were born at #32 weeks. Pediatrics. 2008;121(2). Available at: www.pediatrics.org/cgi/content/full/121/2/e239 Bayley N. Bayley Scales of Infant Development II. San Antonio, TX: Psychological Corp; 1993 Robertson C, Sauve RS, Christianson HE. Province-based study of neurologic disability among survivors weighing 500 through 1249 grams at birth. Pediatrics. 1994;93(4):636 – 640 Sreenan C, Lemke RP, Hudson-Mason A, Osiovich H. High-flow nasal cannulae in the management of apnea of prematurity: a comparison with conventional nasal continuous positive airway pressure. Pediatrics. 2001;107(5):1081–1083 Bax MCO. Terminology and classification of cerebral palsy. Dev Med Child Neurol. 1964;6(3): 295–297 Donner A, Klar N. Design and Analysis of Cluster Randomization Trials in Health Research. London, England: Arnold; 2000 Rosner B, Glynn RJ, Loe MT. Incorporation of clustering effects for the Wilcoxon rank sum test: a large sample approach. Biometrics. 2003;59(4):1089 –1098 Grambsch PM, Therneau TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika. 1994;81(3):515–526 Hack M, Taylor HG, Drotar D, et al. Poor predictive validity of the Bayley Scales of Infant Development for cognitive function of extremely low birth weight children at school age. Pediatrics. 2005;116(2):333–341 Aylward GP. Cognitive function in preterm infants: no simple answers. JAMA. 2003;289(6):752–753 Soll RF, Morley CJ. Prophylactic versus selective use of surfactant in preventing morbidity and mortality in preterm infants. Cochrane Database Syst Rev. 2009;(3):CD000510 Schmidt B, Davis P, Moddemann D, et al. Long-term effects of indomethacin prophylaxis in extremely-low-birth-weight infants. N Engl J Med. 2001;344(26):1966 –1972 Doyle LW, Davis PG, Morley CJ, McPhee A, Carlin JB; DART Study Investigators. Outcome at 2 years of age of infants from the DART Study: a multicenter, international, randomized, controlled trial of low-dose dexamethasone. Pediatrics. 2007;119(4):716 –721 Mewes AU, Huppi PS, Als H, et al. Regional brain development in serial magnetic resonance imaging of low-risk preterm infants. Pediatrics. 2006;118(1):23–33 Das Eiden R, Reifman A. Effects of Brazelton demonstrations on later parenting: a meta-analysis. J Pediatr Psychol. 1996;21(6):857– 868 Truffert P, Paris-Llado J, Escande B, et al. Neuromotor outcome at 2 years of very preterm infants who were treated with high-frequency oscillatory ventilation or conventional ventilation for neonatal respiratory distress syndrome. Pediatrics. 2007;119(4). Available at: www.pediatrics.org/ cgi/content/full/119/4/e860 Johnson AH, Peacock JL, Greenough A, et al. High-frequency oscillatory ventilation for the prevention of chronic lung disease of prematurity. N Engl J Med. 2002;347(9):633– 642 Courtney SE, Durand DJ, Asselin JM, et al. High-frequency oscillatory ventilation versus conventional mechanical ventilation for very-low-birth-weight infants. N Engl J Med. 2002;347(9):643– 652



1019

53. Hannah ME, Hannah WJ, Hewson SA, Hodnett ED, Saigal S, Willan AR. Planned caesarean section versus planned vaginal birth for breech presentation at term: a randomized multicentre trial. Lancet. 2000;356(9239):1375–1383 54. Wood L, Egger M, Gluud LL, et al. Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: meta-epidemiological study. BMJ. 2008;336(7644):601– 605 55. Richardson DK, Corcoran JD, Escobar GJ, Lee SK. SNAP-II and SNAPPE-II: Simplified newborn illness severity and mortality risk scores. J Pediatr. 2001;138(1):92–100 56. Blishen BR, Carroll WK, Moor C. The 1981 socioeconomic index for occupation in Canada. Can Rev Social Anthropol. 1987;24:465– 488

1020

PETERS et al


Improvement of Short- and Long-Term Outcomes for Very Low Birth Weight Infants: Edmonton NIDCAP Trial Kathrine Leigh Peters, Rhonda Jean Rosychuk, Leonora Hendson, Judith Jean Coté, Catherine McPherson and Juzer Mohamed Tyebkhan Pediatrics 2009;124;1009-1020; originally published online Sep 28, 2009; DOI: 10.1542/peds.2008-3808 Updated Information & Services

including high-resolution figures, can be found at: http://www.pediatrics.org/cgi/content/full/124/4/1009

References

This article cites 46 articles, 30 of which you can access for free at: http://www.pediatrics.org/cgi/content/full/124/4/1009#BIBL

Subspecialty Collections

This article, along with others on similar topics, appears in the following collection(s): Developmental/Behavior http://www.pediatrics.org/cgi/collection/developmental:behavior

Permissions & Licensing

Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://www.pediatrics.org/misc/Permissions.shtml

Reprints

Information about ordering reprints can be found online: http://www.pediatrics.org/misc/reprints.shtml
