Longitudinal, 15-Year Follow-up of Children Born at Less Than 29 Weeks’ Gestation After Introduction of Surfactant Therapy Into a Region: Neurologic, Cognitive, and Educational Outcomes
Objective. To measure the primary and secondary school-age neurologic, cognitive, and educational outcomes in a cohort of extremely premature infants born after the introduction of exogenous surfactant therapy in a circumscribed region.
Methods. Two hundred thirteen infants born at <29 weeks’ gestation were cared for at a regional referral center during 1985–1987. At primary school age, neurologic and cognitive outcomes, educational achievement, school placement, health status, and socioeconomic status were determined by follow-up visit. At secondary school age, school placement and health status were evaluated by telephone interview.
Results. One hundred thirty-two infants survived to school age, of whom 127 (96%) were evaluated in 1992–1995 and 126 (95%) were evaluated in 2000. Mean ages were 7.0 years at first follow-up and 14.1 years at second follow-up. At primary-school age follow-up, 19 children (15%) had cerebral palsy, 24 (19%) had a general cognitive index <70, and 41 (32%) were placed in a self-contained, special classroom. Thirty-nine children (31%) had no physical or educational impairment, whereas 27 (21%) had at least 1 severe disability. At secondary school age, cerebral palsy incidence remained unchanged, whereas 36 children (29%) were placed in a special classroom. Fifty-one children (41%) had no physical or educational impairment, whereas 24 (19%) had at least 1 severe disability. Neonatal intraventricular hemorrhage and low socioeconomic status were the strongest predictors of adverse outcomes.
Conclusions. Premature infants born in the surfactant era remain at high risk of neurodevelopmental compromise. Although many of these children do well, a significant minority will require intensive special educational services through secondary school age.
- premature infants
- very low birth weight infants
- follow-up studies
- pulmonary surfactants
- developmental disabilities
The introduction of surfactant has dramatically decreased mortality among very premature infants.1 Although regular use of exogenous surfactant therapy did not begin nationally until US Food and Drug Administration licensure of synthetic surfactants in the early 1990s, several intensive care nurseries, including our own, began using surfactant regularly as part of research protocols as early as 1983. The school-age outcome of those premature infants who survive has implications for the health and educational care of this group of children. Only a limited number of school-age follow-up studies have been reported since surfactant use became routine, and the majority of these have reported outcomes only through the early primary school years.2–5
Before the use of surfactant, multiple studies reported elevated rates of adverse neurologic outcomes including cerebral palsy and visual impairment in very low birth weight (<1500 g) infants.6,7 In addition, many of these children also had poor brain growth, limited cognitive abilities, poor academic achievement, behavior problems, and an increased need for special education and educational services.6–16
After the introduction of surfactant, short-term reports suggested that although survival of extremely premature infants improved, there were no differences in neurologic outcomes between infants treated with or without surfactant.17 A study from our group reported no differences at school age (5–7 years of age) in cognitive abilities between infants treated with surfactant or saline.2 However, that study also found that nearly half of school-age survivors had General Cognitive Index (GCI) scores <85 on the Bayley Scales of Infant Development,18 and that one third of the children were receiving special education services.2 We have since reported that school-age (4- to 8-year-old) children from a trial comparing prophylactic administration of surfactant (before the first breath) with administration only in the presence of established respiratory distress syndrome (RDS) showed no differences between the 2 groups in neurologic, cognitive, or educational outcomes.3 No studies to date have followed surfactant-treated children beyond the early school years.
We longitudinally examined the performance of a cohort of premature infants cared for in a single regional neonatal intensive care unit (NICU) to determine their long-term outcomes after the introduction of surfactant usage in that NICU. Children were examined during the primary school years and contacted again at 12 to 15 years of age. We hypothesized that premature infants would remain at risk for poor educational achievement through the middle and high school years.
All infants <29 completed weeks’ gestational age born between January 1, 1985, and December 31, 1987, and cared for in the NICU of the Strong Memorial Hospital (SMH) were eligible. Both inborn and transported infants were included. The University of Rochester Research Subjects Review Board approved this study. Informed consent was obtained at the time of follow-up evaluations.
The NICU at SMH is the Level IV Regional NICU for the 12-county Finger Lakes Region of New York State. An active regional maternal transport program was in place, and all infants in the region born at <29 weeks’ gestation were transferred to this unit. Gestational age was determined at birth by the neonatologist’s synthesis of the obstetric history and Ballard examination.19 Infants <24 weeks’ gestation were generally considered previable and were seldom offered aggressive resuscitation or ventilatory support during the study period.
During the first year of the period under study, calf lung surfactant extract (Infasurf, ONY, Buffalo, NY) was available under a study protocol comparing a single dose of intratracheal surfactant (as a 3 mL, 90 mg, preventilatory, bolus dose) with saline at birth in infants at risk for RDS.20 During the second 2 years of the period, surfactant was available to infants <29 weeks’ gestation under a study protocol comparing prophylactic, preventilatory surfactant (Infasurf) administered in the delivery room with its administration in established RDS. Preventilatory doses were delivered as detailed above. Rescue doses were delivered as four 0.75-mL aliquots. Both the prophylactic and rescue groups in the second study were eligible for up to 3 additional doses of surfactant (given in 4 aliquots) at intervals of at least 12 hours if signs of respiratory distress continued. Surfactant (Infasurf) was also available during the second 2 years under a nonrandomized, compassionate use protocol for infants with established RDS.
Prenatal corticosteroids were used infrequently and postnatal corticosteroids were used intermittently during the years studied. However, specific information about corticosteroid use was not collected in this data set. Brain imaging was performed on most infants within the first 2 weeks after birth. Intraventricular hemorrhage (IVH) was graded 0 through IV using the method described by Papile and colleagues.21 Routine brain neuroimaging for periventricular leukomalacia was not being performed during this time period. Routine audiologic screening using brainstem auditory-evoked responses was performed on all premature infants in the NICU during 1985–1987.
Four- to 10-Year Evaluation
During the years 1992–1995, the children were recalled for evaluation, their hospital records were reviewed, and their physicians and teachers were contacted (Table 1). A neurologist and neuropsychologist (blinded to the neonatal medical diagnoses and current school placement) evaluated the children individually. The neurologic assessment consisted of an evaluation of mental status, activity, cranial nerves, motor system (gait, tone, strength, deep tendon reflexes, and coordination), cerebellar system, and sensation. A single pediatric neurologist who made the final determination about normality of the examination reviewed all examination reports. Examination results were classified as normal or abnormal (any abnormality), with a subset of children with abnormal examinations having cerebral palsy. Cerebral palsy was defined as a fixed motor deficit diagnosed by the pediatric neurologist. The neuropsychologist determined cognitive abilities using the McCarthy Scale of Children’s Abilities,22 the Children’s Auditory Verbal Learning Test-2, the Peabody Picture Vocabulary Test-Revised,23 and/or the Developmental Test of Visual-Motor Integration,24 depending on the age of the child. School-administered IQ testing was used if the child was uncooperative or unable to be tested during the clinic visit, and the study psychologist reviewed the results. School testing included the Wechsler Intelligence Scale for Children Version III,25 the Wechsler Intelligence Scale for Children-Revised,26 Wechsler Preschool and Primary Scale of Intelligence,27 Vineland Adaptive Behavior Scale, the Peabody Picture Vocabulary Test-Revised, and the Bayley Scales of Infant Development.18 After the child was tested, and without knowledge of the child’s school placement, the evaluators independently recommended services they deemed to be indicated.
Presence of bronchopulmonary dysplasia (defined as need for supplemental oxygen at 36 weeks’ postmenstrual age), total length of hospital stay (until final discharge home), the presence of severe visual impairment (corrected vision worse than 20/200 in the better eye and unable to navigate using visual cues), and hearing impairment (need for hearing aids) were determined from hospital records. Primary care physicians completed a standardized questionnaire to determine medical status and limitations. Parents were interviewed to determine the child’s physical and behavioral status and to gather socioeconomic variables. Socioeconomic status was determined at the time of the first follow-up evaluation using the Hollingshead Scale.28
The subjects’ teachers completed a questionnaire rating each child’s school performance and placement. Academic performance was rated on a scale of 1 to 5, as follows: 1) far below average, 2) below average, 3) average, 4) above average, and 5) far above average.3 Class placement was rated in the following 7 categories: 1) regular class/no services; 2) consultation: children in regular class who have an itinerant teacher or therapist work with them one-on-one at regular intervals; 3) resource room: child leaves regular class for scheduled period of time to go to a resource room and work with a resource teacher; 4) Option I: a special classroom with a 15:1 pupil:teacher ratio; 5) Option II: a special classroom with a 12:1 pupil:teacher ratio with 1 additional paraprofessional; 6) Option III: a classroom with an 8:1 pupil:teacher ratio and 1 additional paraprofessional; 7) Option IV: a classroom with a 12:1 pupil:teacher ratio and 1 additional adult for every 3 children in class. Children “mainstreamed” or “blended” into regular classroom settings while still receiving extensive services appropriate for a separate classroom were considered to require the equivalent option, as defined above. Teachers also identified special education classification, if applicable, and type of additional services the child was receiving. Additional services were defined as occupational therapy, physical therapy, speech and language, adaptive physical education, tutoring, counseling, remedial reading, or remedial mathematics.
Twelve- to 15-Year Evaluation
During the year 2000, subjects’ parents were contacted again by telephone and interviewed using a structured, 50-item questionnaire covering school placement, sensory difficulties, physical and medical limitations, and behavior (Table 1). If parents answered “yes” to questions regarding mental retardation, cerebral palsy, or Option III or IV educational placement, the Children’s Functional Independence Measure (WeeFIM), a standardized evaluation of self care, sphincter control, mobility, locomotion, communication, and social cognition skills, was also administered.29
Summaries of continuous results are reported as mean ± standard deviation, unless otherwise specified. The associations of multiple risk factors with continuous outcomes were modeled using forward-selection, multiple linear regression. Similar associations with dichotomous outcomes were modeled using forward-selection, multiple logistic regression. Both regression models were prospectively designed to include birth weight, gestational age at birth, sex, race, age at evaluation, surfactant therapy strategy, presence of BPD, hospital of birth (inborn vs outborn), IVH (as an ordinal variable, by grade), and socioeconomic status (Hollingshead score) as independent variables. Variables with P (entry) <.20 were retained in the model. Because we wished to analyze outcomes by exposure to surfactant rather than by the study protocol under which surfactant was administered, infants were classified into 1 of 4 surfactant strategy groups: 1) surfactant available under research protocol but not needed (referent group), 2) surfactant prophylaxis in the delivery room, 3) surfactant rescue therapy either within a trial or through compassionate use, or 4) no surfactant available under protocol or compassionate use. Single sets of dichotomous variables were compared statistically only when these variables were not included in the prospectively designed regression models. These comparisons were made using χ2 or Fisher exact analysis as appropriate. Two-sided P values <.05 were considered significant.
Two hundred thirteen neonates with a gestational age <29 weeks were cared for at SMH during the study period. Seventy-six (36%) died in the NICU. Five (2%) died after discharge. Of those who died after discharge, 3 had a gestational age of 26 weeks and 2 of 28 weeks. Four of the 5 received surfactant. Causes of death included sudden infant death, sepsis, and respiratory failure. One hundred thirty-two children survived to school age. All survivors were contacted. The demographic characteristics of the 127 (96%) who agreed to participate are shown in Table 2. One hundred nine infants were inborn and 18 were transported. Birth weights ranged from 588 to 1550 g. The median gestational age was 27 weeks. No child born at <24 weeks’ gestation survived. There were 69 (54%) males, and 75 (59%) of the subjects were white. Ninety-one (72%) received exogenous surfactant therapy. Of these, 60 received surfactant as part of the study that compared prophylactic with rescue surfactant,30 whereas 9 received it as part of the study comparing surfactant to saline20 and 22 received it under the compassionate use protocol. When considered by surfactant strategy, 29 infants had surfactant available but did not require it, 35 children received prophylactic surfactant, 56 children received rescue surfactant, and 7 children did not have surfactant available to them. Sixty-seven children enrolled in the prophylactic surfactant trial had undergone neuropsychological testing at early school age.3 Those results are included in this report.
One hundred nineteen children had cerebral ultrasounds. Of these, 79 (66%) were normal, 14 (12%) had Grade I IVH, 7 (6%) had Grade II IVH, 6 (5%) had Grade III IVH, and 13 (11%) had Grade IV IVH.
Three families declined to participate at the first evaluation. Two of these families had twins of 27 weeks’ gestation. The birth weights of children whose families refused participation ranged from 860 to 1102 g. The family of a former 26-week multihandicapped child refused participation at the second evaluation. Age at the first evaluation ranged from 4 to 10 years (7.0 ± 1.2 years), whereas age at the second evaluation ranged from 12 years, 10 months to 15 years, 9 months (14.1 ± 0.8 years).
Severe visual impairment (corrected vision worse than 20/200 in the better eye and unable to navigate using visual cues) was infrequent (1 child born at 24 weeks and 2 children born at 26 weeks at the first evaluation). One of these children’s families reported an improvement in vision at the second evaluation, but another 27-week child had developed severe visual impairment. Four children had hearing impairment requiring hearing aids, but in none was it so severe as to preclude speech as a primary means of communication. Eight children’s families reported asthma requiring medication at the first evaluation, whereas 22 children (17%) required asthma medication at the second evaluation. Four children (3%) at the first evaluation and 10 (8%) at the second evaluation required medication for seizures (Table 3).
Neurologic examination (Table 3) was performed at the first evaluation. One hundred seven (84%) of the neurologic assessments were conducted by a pediatric neurologist, 14 (11%) were performed by a pediatrician, and 6 (5%) were performed by a psychologist. Eighty children (63%) had a normal neurologic examination. Children with grade IV IVH were more likely to have abnormal neurologic examination (92% with grade IV IVH vs 32% with no IVH at the first evaluation) or cerebral palsy (77% with grade IV IVH vs 9% with no IVH at the first evaluation, and 83% with grade IV IVH vs 8% with no IVH at the second evaluation). Nineteen children (15%) were diagnosed with cerebral palsy at the first evaluation, and 19 (15%) carried this diagnosis at the second evaluation. In a multiple logistic regression model, increasing grade of IVH and administration of prophylactic surfactant were independently associated with an increased risk of cerebral palsy at both the first and second evaluations, while BPD was associated with an increased risk at the second evaluation (Table 4).
At the first evaluation, the families of 6 children (5%) reported that the child received medication for attention deficit disorder and/or hyperactivity. The families of 17 children (13%) did so at the second evaluation.
A study neuropsychologist evaluated 99 (78%) of the children at 4 to 10 years of age (Table 3). Cognitive examination results of 25 children (20%) were obtained from school records, and 3 (2%) from other psychologists. Sixty-five children (51%) had a GCI ≥85, and 103 (81%) had a GCI ≥70. Nine of the 127 children had GCI scores >115 and all were ≥26 weeks’ gestation at birth. The children with a GCI of <70 included 7 (54%) of those children with Grade IV IVH and 12 (15%) of those children with no IVH. Twenty children (44%) with an abnormal neurologic examination and 4 (5%) with a normal neurologic examination had a GCI of <70. In a multiple linear regression model, lower gestational age, lower socioeconomic status, increasing grade of IVH, having received rescue surfactant, and non-Caucasian race were independently associated with poorer cognitive outcome (Table 5).
At the first evaluation, educational outcome was categorized by class placement and teacher-rated academic performance (Table 6). Class placement was reassessed at the second evaluation. Eighty-six children (68%) were placed in regular classes at the first evaluation, and 90 (71%) were in regular classrooms at the second evaluation. Forty-six children (36%) required no additional services in school at the first evaluation; this was true for 63 children (50%) at the second evaluation. At the second evaluation, 8 children who were not requiring additional services had repeated at least 1 grade in school. Risk factors for needing any special services were modeled by multiple logistic regression (Table 4). At both times of evaluation, lower socioeconomic status and increasing grade of IVH were independently associated with the need for school services. At the first evaluation, outborn status and presence of BPD were also associated with the need for school services.
The median teacher rating of children’s school performance at the first evaluation was 2 (below average), with a range of 1 to 5 (Table 6). Nineteen children (16%) were performing above average or far above average. School performance was examined in relationship to the family’s socioeconomic group, as shown in Table 7. Twenty-nine percent of the highest socioeconomic group had below average school performance ratings compared with 71% of the lowest socioeconomic group. In a multiple linear regression model, higher grade of IVH, lower socioeconomic status, non-Caucasian race, and administration of rescue surfactant were all independently associated with lower teacher rating of school performance (Table 5).
At the first evaluation, thirty-eight children (30%) were not receiving some of the services in school that the study evaluators thought were indicated. Twelve of those children were receiving no special services. Twenty-six children who were receiving school services were considered by the study evaluators to need additional or different services.
At the second evaluation, the 29 children with mental retardation (n = 11), cerebral palsy (n = 19), and/or Option III or IV educational placement (n = 16) were evaluated further with the Children’s Functional Independence Measure, an interview-based measure of capacity for independent self-care and social function. The highest achievable score (126) corresponds to the normal functional independence of a 7-year-old child. The median score on the Children’s Functional Independent Measure (WeeFIM) among the children evaluated was 108 (range: 35–126), indicating functional independence equivalent to that of a normal 5-year-old child.
As shown in Table 7, children in higher socioeconomic groupings had higher cognitive and academic functioning, but neuromotor outcome was not associated with socioeconomic status. At the first evaluation, 2 (7%) of children in the lowest socioeconomic group needed no school services, whereas 11 (52%) of children in the highest socioeconomic group needed no school services. In addition, children in the higher socioeconomic groups were more likely to be getting services considered indicated by the study evaluators. Despite poorer functioning, 12 children (43%) in the lowest socioeconomic group had unmet service needs, whereas only 3 (14%) of the highest group were not receiving indicated services (P = .03).
Correlation Between First and Second Evaluations
As expected, the outcome of the first evaluation was highly predictive of the outcome at the second evaluation. The finding of cerebral palsy at the first evaluation conferred a 48-fold (95% confidence interval [CI]: 12, 192) increase in risk for the same finding at the second evaluation (P < .0001). Similarly, the need for school services at the first evaluation conferred a 5.3-fold (95% CI: 3.0, 9.3) increase in risk for service need at the second evaluation (P < .0001).
When functioning in all areas was considered, 39 children (31%) had no impairment of any type at the first evaluation (Table 8), whereas the families of 51 children (41%) reported no impairment at the second evaluation (Table 9). Twenty-seven children (21%) had at least 1 severe disability (cerebral palsy, corrected vision less than 20/200 in better eye, or need for Option III or IV special education) at first evaluation, and 24 (19%) had at least 1 severe disability at second evaluation.
We report a 15-year, near-complete follow-up of a cohort of infants born at <29 weeks’ gestation after the introduction of surfactant therapy into the tertiary care nursery for a single perinatal/neonatal referral region. We found that 2 factors—IVH and socioeconomic status—most strongly influenced the neurologic, cognitive, and educational status of survivors through secondary school age. There were few changes in diagnosis between 4 to 10 years and 12 to 15 years, but school functioning improved between the primary and secondary school age evaluations. Most of the improvement in school function occurred in a group of children who, although in regular classes, were receiving additional school services at 4 to 10 years. Many of these children were functioning without extra services by the secondary school years. Although this could reflect decreases in the overall level of special services provided by school districts, it may also indicate an increasing ability to function in the school setting as these children grew and matured.
Before the introduction of surfactant, school age outcome studies found prematurity was associated with cerebral palsy, poor developmental outcome, and visual impairment.6,7 Cognitive abilities were better in children with higher birth weights,6 and as birth weights decreased, the need for educational services increased.7 However, most school age studies followed children only through school entry or into the primary grades.
Investigators in the Netherlands, Sweden, Australia, and the United States have followed regional cohorts of children born in the presurfactant era at <32 weeks’ gestation and/or <1500 g birth weight for 10 to 14 years. These children tended to score more poorly on standardized tests of intelligence than full-term children12,31 and were more likely to have behavior problems.12 Six percent to 10% of children had cerebral palsy.13,32 Up to one third of these children required special or separate classroom placement at school.9,11,32 Among children born at <750 g, 56% required special classroom placement.32 Up to 38% of children were placed in classes below grade level in school.9,12,31 In an Australian cohort of children born at <1000 g and evaluated at 14 years of age, 6% had bilateral blindness, 5% had deafness requiring hearing aids, and 10% were severely disabled.13 Children born at <1000 g were also more likely than their peers to have other health problems, including seizures and respiratory illnesses, although these conditions decreased by 12 to 16 years of age.14 In a Dutch cohort born at <1500 g, 10% of children were classified as having severe disability at 14 years old. However, the burden of milder developmental, behavioral, and learning abnormalities was such that the authors estimated that up to 40% of children would not become fully independent adults.33
In a recent report of outcomes of young adults (20 years of age) born in 1977–1979, those born at <1500 g were less likely than full-term controls to have completed high school (74% vs 83%), were more likely to have neurosensory impairments (10% vs <1%), and were more likely to have at least 1 chronic medical condition (33% vs 21%).15 Former very low birth weight adults also scored more poorly on standardized tests of intelligence and academic achievement. However, these same young adults were less likely than their peers to have had contact with the police, to abuse drugs or alcohol, or to have been pregnant.
Attempts to compare neurodevelopmental outcomes between the presurfactant and surfactant eras have primarily been limited by the short duration of follow-up. A recent evaluation of 446 premature children born at a single center over a 12-year period spanning presurfactant, transitional, and surfactant eras showed that the rate of neurologic/neurosensory abnormalities remained constant at ∼11%.1 The rate of cognitive abnormalities varied from 16% in the presurfactant era to 10% in the surfactant era. However, none of the children from the surfactant era had reached school age at the time of evaluation. A similar study measured cerebral palsy and functional outcomes at a mean age of 5 years in 425 infants born at ≤1500 g during the period of transition to surfactant use.4 Neonatal mortality improved with the introduction of surfactant. The cerebral palsy rate remained 12.6% before and after surfactant introduction. Although measures of self-care and mobility remained unchanged, social function was slightly decreased in survivors born after the general availability of surfactant. Doyle5 recently evaluated 225 Australian infants born in 1991–1992 at 23 to 27 weeks’ gestation who survived to 5 years of age. Forty-three percent of these infants had received exogenous surfactant. Among infants who survived to 5 years of age, 80% were without major disability (blindness, requiring hearing aids, cerebral palsy, intelligence quotient 2 standard deviations below control mean). However, school performance data were not available for this cohort. In a recent review of the world experience with outcome of extremely low birth weight (<1000 g) children born in the 1990’s, Hack and Fanaroff34 reported that children born at 24 weeks’ gestation had rates of severe neurodevelopmental disability (subnormal cognitive function, cerebral palsy, blindness, and/or deafness) of 22% to 45%. Those born at 25 weeks suffered severe neurodevelopmental disability at rates of 12% to 35%.
Without a full-term control group, we cannot directly compare the outcomes in our cohort with full-term children, although the rates of disability in the cohort exceed those generally reported in the full-term population.15 Our data also did not include a concurrent nonsurfactant-available comparison group to whom direct comparisons could be made. Thus, any comparisons to the presurfactant era are hampered by differing settings, times, and evaluation methods. Contemporaneous comparisons at school age of surfactant-treated and nonsurfactant-treated cohorts are few. At 5- to 7-year follow-up of 39 children born at 25 to 29 weeks’ gestation, including several who were reevaluated for this report, Wagner and colleagues2 found no differences in neurodevelopmental outcome between those assigned randomly at birth to be treated with intratracheal surfactant or saline. However, they found that, overall, 47% of children had GCI scores below 85 and 23% had cerebral palsy. We have previously reported school age follow-up of 148 children born at <30 weeks’ gestation who participated in a study comparing prophylactic and rescue surfactant, including a subset of children reevaluated for this report.3 At 4 to 8 years of age, 19% of those subjects had a GCI <70, 31% had an abnormal neurologic examination, 47% had academic performance below average, and 18% required a separate special education class.3 No differences were found between infants who received surfactant by the prophylactic and rescue strategies.
The current report details rates of impairment similar to those measured for elementary and middle school-aged former premature infants from both the presurfactant and surfactant eras. Of interest, 2 respiratory variables—BPD and surfactant administration strategy—seemed to be associated with outcome in some areas of cognitive, neurologic, and educational functioning. Several other studies have associated the presence of bronchopulmonary dysplasia with neurodevelopmental abnormalities, including poorer cognitive, motor, and functional outcomes, at follow-up at ages ranging from 3 to over 10 years.4,35–41 This association persisted when other factors were controlled.35–37 Early data suggest that it holds true in the surfactant era.38,39 Although the effect of BPD in our group of subjects did not seem to be as strong or consistent as that of IVH or socioeconomic status, severe respiratory illness in the newborn period did predispose to neurodevelopmental difficulties in later life.
Previous reports of school age follow-up of 2 randomized, clinical trials containing some of the subjects assessed in this study did not find an association of surfactant therapy with alterations in neurodevelopmental outcome.2,3 The current evaluation yields conflicting results regarding the relationship between surfactant therapy and neurodevelopmental outcome. When compared with children who did not need any surfactant therapy, children who received prophylactic surfactant therapy seemed to be more likely to suffer from cerebral palsy, whereas children who had received rescue surfactant seemed to be more likely to have cognitive and educational impairments. Because the comparison in each case was to a nonrandom group of subjects without respiratory disease, it is difficult to reconcile the results with those from the follow-up of randomized trials. Small numbers of subjects in some of the groups (eg, only 7 subjects did not have surfactant available to them) also hamper the analysis. Despite the rising survival rates of extremely premature infants after the advent of surfactant therapy,1,4 it does not seem that increasing percentages of survivors are suffering neurodevelopmental problems at school age. The preponderance of evidence, including our own from previous controlled, clinical trials, diminishes the concern that surfactant therapy salvages only children so immature that they would be destined for later developmental difficulties.1–5,17 Nonetheless, continued, long-term neurodevelopmental follow-up studies of children receiving surfactant therapy are warranted.
Our findings are also consistent with those of many others in revealing a strong and almost overwhelming relationship between neurodevelopmental outcome and presence of neonatal neuroimaging abnormalities, even when other known predictors such as gestational age and birth weight are taken into account. In a study controlled for socioeconomic, perinatal, and neonatal variables, Whitaker and colleagues42 reported an odds ratio of 4.6 for mental retardation in low birth weight children with germinal matrix hemorrhage or IVH, and an odds ratio of 65.8 for mental retardation in those with parenchymal lesions or ventricular enlargement. In our population, presence of IVH in the newborn period was a consistent predictor of abnormal neurologic outcome, cognitive difficulties, and school problems through 12 to 15 years of age.
Our finding of poorer cognitive and school outcomes in children of lower socioeconomic status is disturbing. The import of this finding is heightened by the direct relationship between higher socioeconomic status and a child receiving appropriate school services at age 4 to 10 years. Like their full-term counterparts, preterm children in indigent populations have higher rates of developmental delay.42,43 They often receive little attention in neonatal follow-up studies because of their high rates of attrition.43 In addition, former preterm infants who are lost to epidemiologic follow-up are more likely than are those who return to have neurodevelopmental problems.44 The very high rate of follow-up in our study (95% through 12–15 years) minimizes the chance of this sort of ascertainment bias in our data. It will be particularly important in both research and clinical settings to make extra efforts to follow the most difficult to locate former premature infants.
Extremely premature children born after the introduction of surfactant therapy into a region and evaluated serially through secondary school age have outcomes similar to children of comparable gestation born in the presurfactant era. Although significant numbers of children have major developmental problems, by secondary school half of the youngsters are able to function in regular classes without any additional help. Despite concerns that the introduction of surfactant could have resulted in the survival of increasing proportions of children with significant neurodevelopmental disabilities, this does not seem to be the case. The additional risks of poor developmental outcome imposed by IVH and poverty provide clear indications of the physiologic and social interventions that may most benefit the smallest neonatal survivors.
This study was funded in part by National Institutes of Health SCOR grant HL-36543 and General Clinical Research Center grant RR00044–34, and by the New York State Education Department.
We thank the Neonatal Continuing Care Clinic staff and Linda Reubens for her help with data analysis. We also thank the subjects and their parents.
- Received November 6, 2002.
- Accepted June 19, 2002.
- Address correspondence to Carl T. D’Angio, MD, Box 651, Neonatology, Golisano Children’s Hospital at Strong, University of Rochester, 601 Elmwood Ave, Rochester, NY 14642. E-mail:
- ↵Piecuch RE, Leonard CH, Cooper BA, Sehring SA. Outcome of extremely low birth weight infants (500 to 999 grams) over a 12-year period. Pediatrics.1997;100 :633– 639
- ↵Sinkin RA, Kramer BM, Merzbach JL, et al. School age followup of prophylactic versus rescue surfactant trial: pulmonary, neurodevelopmental, and educational outcomes. Pediatrics.1998;101(5) . Available at: http://www.pediatrics.org/cgi/content/full/101/5/e11
- ↵Doyle LW. Outcome at 5 years of age of children 23 to 27 weeks’ gestation: refining the prognosis. Pediatrics.2001;108 :134– 141
- ↵Hille ET, den Ouden AL, Bauer L, van den Oudenriin C, Brand R, Verloove-Vanhorick SP. School performance at nine years of age in very premature and very low birth weight infants: perinatal risk factors and predictors at five years of age. Collaborative Project on Preterm and Small for Gestational Age (POPS) Infants in The Netherlands. J Pediatr.1994;125 :426– 434
- Klebanov PK, Brooks-Gunn J, McCormick MC. Classroom behavior of very low birth weight elementary school children. Pediatrics.1994;94 :700– 708
- ↵Doyle LW, Casalaz D. Outcome at 14 years of extremely low birthweight infants: a regional study. Arch Dis Child Fetal Neonatal Ed.2001;85 :F159– F164
- ↵Saigal S, Stoskopf BL, Streiner DL, Burrows E. Physical growth and current health status of infants who were of extremely low birth weight and controls at adolescence. Pediatrics.2001;108 :407– 415
- ↵Saigal S, Lambert M, Russ C, Hoult L. Self-esteem of adolescents who were born prematurely. Pediatrics.2002;109 :429– 433
- ↵Bayley N. Bayley Scales of Infant Development. San Antonio, TX: The Psychological Corporation; 1969
- ↵Kendig JW, Notter RH, Cox C, et al. Surfactant replacement therapy at birth: final analysis of a clinical trial and comparisons with similar trials. Pediatrics.1988;82 :756– 762
- ↵McCarthy D. McCarthy Scales of Children’s Abilities. San Antonio, TX: The Psychological Corporation; 1972
- ↵Dunn L. Peabody Picture Vocabulary Test-Revised. Circle Pines, MN: American Guidance Service; 1981
- ↵Berry KE. Developmental Test of Visual Motor Integration, Third Revision. Severna Park, MD: Modern Curriculum Press; 1989
- ↵Wechsler D. Wechsler Scale of Intelligence for Children-III. San Antonio, TX: The Psychological Corporation; 1991
- ↵Wechsler D. Wechsler Intelligence Scale for Children-Revised. San Antonio, TX: The Psychological Corporation; 1974
- ↵Wechsler D. Wechsler Preschool and Primary Scale of Intelligence-Revised. San Antonio, TX: The Psychological Corporation; 1989
- ↵Hollingshead AB. Four Factor Index of Social Status. New Haven, CT: Yale University; 1975
- ↵Hack M, Taylor HG, Klein N, Mercuri-Minich N. Functional limitations and special health care needs of 10- to 14-year-old children weighing less than 750 grams at birth. Pediatrics.2000;106 :554– 560
- Robertson CM, Etches PC, Goldson E, Kyle JM. Eight-year school performance, neurodevelopmental, and growth outcome of neonates with bronchopulmonary dysplasia: a comparative study. Pediatrics.1992;89 :365– 372
- ↵Singer L, Yamashita T, Lilien L, Collin M, Baley J. A longitudinal study of developmental outcome of infants with bronchopulmonary dysplasia and very low birth weight. Pediatrics.1997;100 :987– 993
- Schaap AH, Wolf H, Bruinse HW, et al. Influence of obstetric management on outcome of extremely preterm growth retarded infants. Arch Dis Child Fetal Neonatal Ed.1997;77 :F95– F99
- ↵Whitaker AH, Feldman JF, Van Rossem R, et al. Neonatal cranial ultrasound abnormalities in low birth weight infants: relation to cognitive outcomes at six years of age. Pediatrics.1996;98 :719– 729
- ↵Tin W, Fritz S, Wariyar U, Hey E. Outcome of very preterm birth: children reviewed with ease at 2 years differ from those followed up with difficulty. Arch Dis Child Fet Neonatal Ed.1998;79 :F83– F87
- Copyright © 2002 by the American Academy of Pediatrics