Key Points
Question
What is the significance of nonhemorrhagic ventriculomegaly detected on cranial ultrasonography prior to 36 weeks’ postmenstrual age in extremely preterm neonates?
Findings
This observational cohort study of 4193 neonates born at less than 27 weeks’ gestational age found that those with nonhemorrhagic ventriculomegaly had higher odds of neurodevelopmental impairment, poor cognitive outcome, moderate to severe cerebral palsy, and death or neurodevelopmental impairment at 18 to 22 months’ corrected age compared with neonates with normal cranial ultrasonograms. Behavioral outcomes did not differ.
Meaning
Nonhemorrhagic ventriculomegaly is a marker of brain injury and might be associated with early neurodevelopmental risk among extremely preterm neonates.
This longitudinal study characterizes neurodevelopmental outcomes of extremely preterm neonates after nonhemorrhagic ventriculomegaly is detected via cranial ultrasonography prior to 36 weeks’ postmenstrual age.
Abstract
Importance
Studies of cranial ultrasonography and early childhood outcomes among cohorts of extremely preterm neonates have linked periventricular-intraventricular hemorrhage and cystic periventricular leukomalacia with adverse neurodevelopmental outcomes. However, the association between nonhemorrhagic ventriculomegaly and neurodevelopmental and behavioral outcomes is not fully understood.
Objective
To characterize the outcomes of extremely preterm neonates younger than 27 weeks’ gestational age who experienced nonhemorrhagic ventriculomegaly that was detected prior to 36 weeks’ postmenstrual age.
Design, Setting, and Participants
This longitudinal observational study was conducted at 16 centers of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Infants born prior to 27 weeks’ gestational age in any network facility between July 1, 2006, and June 30, 2011, were included if they had a cranial ultrasonogram performed prior to 36 weeks’ postmenstrual age. Comparisons were made between those with ventriculomegaly and those with normal cranial sonograms. Data analysis was completed from August 2013 to August 2017.
Main Outcomes and Measures
The main outcome was neurodevelopmental impairment, defined as a Bayley Scales of Infant and Toddler Development III cognitive score less than 70, moderate/severe cerebral palsy, a Gross Motor Function Classification System score of level 2 or more, vision impairment, or hearing impairment. Secondary outcomes included Bayley Scales of Infant and Toddler Development III subscores, components of neurodevelopmental impairment, behavioral outcomes, and death/neurodevelopmental impairment. Logistic regression was used to evaluate the association of ventriculomegaly with adverse outcomes while controlling for potentially confounding variables and center differences as a random effect. Linear regression was used similarly for continuous outcomes.
Results
Of 4193 neonates with ultrasonography data, 300 had nonhemorrhagic ventriculomegaly (7%); 3045 had normal cranial ultrasonograms (73%), 775 had periventricular-intraventricular hemorrhage (18.5%), and 73 had cystic periventricular leukomalacia (1.7%). Outcomes were available for 3008 of 3345 neonates with ventriculomegaly or normal scans (90%). Compared with normal cranial ultrasonograms, ventriculomegaly was associated with lower gestational age, male sex, and bronchopulmonary dysplasia, late-onset sepsis, meningitis, necrotizing enterocolitis, and stage 3 retinopathy of prematurity. After adjustment, neonates with ventriculomegaly had higher odds of neurodevelopmental impairment (odds ratio [OR], 3.07; 95% CI, 2.13-4.43), cognitive impairment (OR, 3.23; 95% CI, 2.09-4.99), moderate/severe cerebral palsy (OR, 3.68; 95% CI, 2.08-6.51), death/neurodevelopmental impairment (OR, 2.17; 95% CI, 1.62-2.91), but not death alone (OR, 1.09; 95% CI, 0.76-1.57). Behavioral outcomes did not differ.
Conclusions and Relevance
Nonhemorrhagic ventriculomegaly is associated with increased odds of neurodevelopmental impairment among extremely preterm neonates.
Introduction
Magnetic resonance imaging studies have reported an association between ventriculomegaly accompanied by white matter loss and neurocognitive and developmental outcomes among preterm neonates. The association between cranial ultrasonography measures of cerebral ventriculomegaly and long-term outcomes among extremely preterm (EP) neonates without prior or concurrent hemorrhage remains unclear, despite an established role for cranial ultrasonography to detect important pathologic lesions, such as periventricular-intraventricular hemorrhage (PIVH) and cystic periventricular leukomalacia. One small single-center study of 44 infants born prior to 30 weeks’ gestational age reported an association between ventriculomegaly at age 1 month and delayed motor and language development at 2 years. Another prospective single-center study of very low-gestational-age neonates reported poor neurodevelopmental outcomes among neonates with ventriculomegaly and other brain lesions, but not among those with ventriculomegaly alone.
Compared with the poor interrater agreement for echodense lesions suggestive of white matter injury, there is an overall better agreement in assessment of ventriculomegaly. In the United States, guidance for the timing and definitions for the classification of preterm cranial ultrasonographic findings, including ventriculomegaly, have been established by the quality standards subcommittee of the American Academy of Neurology and the practice committee of the Child Neurology Society since 2002. Similar to prenatal ultrasonographic guidelines, these guidelines define significant ventricular enlargement as a ventricular measurement of 1 cm or larger. The atrial diameter of the fetal lateral ventricles is constant from 14 to 38 weeks’ gestation, with a mean (SD) of 7.6 (0.6) mm.
We sought to explore the association between nonhemorrhagic ventriculomegaly in neonates and neurodevelopmental and behavioral outcomes at age 18 to 22 months. The objectives of this observational cohort study were to (1) characterize the frequency of nonhemorrhagic ventriculomegaly on cranial ultrasonograms performed during the neonatal intensive care unit stay of neonates born before 27 weeks’ gestational age, (2) explore short-term morbidities, especially systemic inflammation, associated with ventriculomegaly and neonatal brain injury, and (3) characterize the neurodevelopmental and behavioral outcomes of neonates with nonhemorrhagic ventriculomegaly compared with neonates with normal cranial ultrasonograms in a contemporary multicenter cohort within the 16 centers participating in the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network (NICHD NRN). We hypothesized that EP neonates with evidence of nonhemorrhagic ventriculomegaly would be at increased odds of neurodevelopmental impairment (NDI), cerebral palsy, gross motor functional limitations, and poor behavioral outcomes at 18 to 22 months’ corrected age compared with EP neonates who had had normal cranial ultrasonographic findings. We also hypothesized that a greater proportion of EP neonates would have ventriculomegaly reported on late cranial ultrasonographic findings (closest to 36 weeks’ postmenstrual age) compared with cranial ultrasonograms performed within 28 days of birth.
Methods
This study was a retrospective analysis of prospectively collected data from the Eunice Kennedy Shriver NICHD NRN’s generic database and follow-up studies. Institutional review board approval was obtained at each participating NRN center, and informed consent was obtained from the parent(s) or legal guardian as specified by each participating center.
As part of the generic database registry, the Eunice Kennedy Shriver NICHD NRN prospectively collects maternal and neonatal data from birth until patient death, hospital discharge, transfer, or 120 days postdelivery. In-hospital outcomes include mortality (ie, death before 12 hours or death before hospital discharge), early-onset sepsis, late-onset sepsis, bronchopulmonary dysplasia, PIVH, ventriculomegaly, cystic periventricular leukomalacia, necrotizing enterocolitis (NEC), retinopathy of prematurity at stage 3 or greater, and inflammation-initiating illnesses, among others. Inflammation-initiating illness is defined as any 1 or more of the following: mechanical ventilation at the neonatal age of 2 weeks, culture-proven sepsis or meningitis, and surgical NEC or intestinal perforation.
Our research team identified infants who were born between July 1, 2006, and June 30, 2011, at any of the 16 centers participating in the Eunice Kennedy Shriver NICHD NRN, at a gestational age of 26 weeks and 6 days or less, with a birth weight greater than 401 g, and with ultrasonographic results and outcome data available. Neonates who died prior to 12 hours of age were excluded, because many of these neonates were not resuscitated and ultrasonographic data were often missing.
Cranial Ultrasonographic Data
Routine cranial ultrasonograms were performed on infants per existing standards of care; ultrasonography was not specified by the study protocol. Ultrasonographical guidelines established by the quality standards subcommittee of the American Academy of Neurology and the practice committee of the Child Neurology Society serve as minimum requirements at NRN centers.
Cranial imaging findings were reported by local site radiologists for sonograms performed within 28 days of birth and 36 weeks’ postmenstrual age (or closest to 36 weeks’ postmenstrual age). If multiple scans were performed, the scan closest to 7 days within the 36 weeks’ postmenstrual age date was used. The date of the cranial ultrasonogram with the most severe findings, along with the specific findings, location, and laterality of the ultrasonogram, were included in the data set of the present study. For neonates with multiple scans prior to 28 days, the most severe findings were identified, and the earliest scan that showed these findings was used.
The following lesions were defined: blood or echodensity in the germinal matrix or subependymal area, blood or echodensity in the ventricle, ventricular size enlarged with concurrent or prior blood in the ventricles, ventricular size enlarged without concurrent or prior blood in the ventricles, blood echodensity in the parenchyma, cerebellar hemorrhage, and cystic areas in the parenchyma (cystic periventricular leukomalacia and porencephalic cyst). Nonhemorrhagic ventriculomegaly was defined as enlarged ventricular size without concurrent or prior blood in the ventricles detected on cranial sonogram. Periventricular leukomalacia was defined as cystic areas in the white matter around the ventricles (most commonly dorsal and lateral to the external angle of the lateral ventricle). These echolucencies could be single or multiple, unilateral or bilateral, diffuse or focal, and could vary in size.
Follow-up at Corrected Age 18 to 22 Months
Neonatal data collected during the initial hospitalization of EP infants were linked to follow-up data from the Eunice Kennedy Shriver NICHD NRN follow-up study at 18 to 22 months (corrected age). In this study, surviving infants underwent formal neurodevelopmental testing and behavioral screenings performed by certified examiners who were blinded to neonatal clinical variables (including ultrasonographic findings); all investigators were certified and trained for reliability. Psychometric testing was performed using the Bayley Scales of Infant and Toddler Development III (Bayley III), which includes cognitive, language, and motor subscales and is standardized to a mean (SD) of 100 (15), and thus a score of less than 70 represents 2 SD or more below the mean. Children who are so developmentally delayed that they cannot be assessed are assigned scores of 54 for severe cognitive delay and 46 for severe language or motor delay. Behavioral screening was performed using the Brief Infant-Toddler Social and Emotional Assessment administered to the primary caregiver in the form of a structured interview. The Brief Infant-Toddler Social and Emotional Assessment is a nationally standardized, normative value–referenced behavioral screening instrument that can be used to determine whether diagnostic assessment or referral for socioemotional and behavioral problems may be indicated. Total problem scores can be compared with specific percentile rankings of normative populations; lower percentile rankings (25% or less) are associated with higher problem scores. Specific items address externalizing, internalizing, and dysregulation problems, as well as behaviors or deficits often seen in autism spectrum disorders.
Cerebral palsy is defined as a nonprogressive central nervous system disorder with abnormal muscle tone in at least 1 extremity and abnormal control of movement and posture that interferes with age-appropriate activities. Disabling cerebral palsy is classified based on a modified Gross Motor Function Classification System score of level 2 or greater. Hearing impairment is defined as permanent hearing loss that does not permit the child to understand or communicate despite amplification. Visual impairment is defined as corrected visual acuity less than 20/200 in both eyes. Neurodevelopmental impairment is defined by 1 or more of the following: a Bayley III cognitive score less than 70, disabling cerebral palsy, a Gross Motor Function Classification System score of level 2 or greater, blindness, or hearing impairment. The primary outcome was composite NDI. Secondary outcome measures included cerebral palsy; gross motor functional limitation; Bayley III cognitive, language, and motor subscores; behavioral scores according to the Brief Infant-Toddler Social and Emotional Assessment; risk of death or NDI combined as a single unit of statistical analysis; and death.
Statistical Analysis
Data analysis was completed from August 2013 to August 2017. For the statistical analysis, continuous variables were described using mean (SD) or median (interquartile range). Categorical variables were described using frequency and percentage. Unadjusted comparisons of maternal demographics, neonatal baseline characteristics, neurodevelopmental outcomes, and behavioral outcomes between the ventriculomegaly group and the normal cranial ultrasonogram group were made using χ2 tests or Fisher exact tests for categorical data and Wilcoxon 2-sample tests (as an approximation for the t test) for continuous data. Logistic regression modeling was used to evaluate the association between nonhemorrhagic ventriculomegaly and adverse outcomes while controlling for center differences (defined as differences in health care populations, facilities, and clinical practices in the centers at which data were collected) as a random effect and for other potentially confounding variables (estimated gestational age, sex, antenatal steroids, maternal educational attainment of less than a high school diploma, late-onset sepsis, physiologic BPD, NEC, and retinopathy of prematurity) as fixed effects. Linear regression was used in a similar manner for continuous outcomes. The NRN center was included in all models as a random effect to control for potential differences in clinical management and interpretation of cranial ultrasonography at different health care centers.
Results
Among the Eunice Kennedy Shriver NICHD NRN centers, 4193 EP neonates younger than 27 weeks’ gestational age and more than 401 g birth weight survived longer than 12 hours and had a cranial ultrasonogram available within or after 28 days of life (Figure). Of these, 3045 had normal cranial ultrasonographic findings (72.6%); 775 neonates had PIVH identified within 28 days of life (18.5%), 73 had findings of isolated cystic periventricular leukomalacia without ventriculomegaly (1.74%), and 300 neonates had nonhemorrhagic ventriculomegaly (7.15%). Nineteen participants (n = 19 of 300; 6.3%) with ventriculomegaly were lost to follow-up. Among neonates with nonhemorrhagic ventriculomegaly who were followed up to 18 to 22 months (corrected age), most had bilateral ventriculomegaly (n = 209 of 281; 74%) and half developed ventricular enlargement after 28 days of life (n = 147 of 281; 52%) (Table 1). Thirty-nine participants had both nonhemorrhagic ventriculomegaly and cystic periventricular leukomalacia. Analyzing the data with and without these participants did not alter the results (eTables 1 and 2 in the Supplement). None of the neonates with nonhemorrhagic ventriculomegaly required ventricular drainage.
Figure. Flow Diagram of Study Participants.
Included and excluded infants. Three additional infants were excluded owing to missing data, including 1 missing data for blood in ventricle within 28 days and 2 missing data for periventricular leukomalacia.
Table 1. Timing and Laterality of Ultrasonographic Ventriculomegaly Findings.
| Nonhemorrhagic Ventriculomegaly | Neonates, No. (%) | ||
|---|---|---|---|
| Within 28 d (n = 80) |
After 28 d (Through 36 wk) (n = 147) |
Both Periods (n = 54) |
|
| Unilateral | 28 (35) | 29 (20) | 13 (24) |
| Bilateral | 52 (65) | 118 (80) | 39 (72) |
| Mixed | NA | NA | 2 (4) |
Abbreviation: NA, not applicable.
Outcome data were available for 3008 of the 3345 neonates with either normal cranial sonograms or ventriculomegaly (90%). Maternal and neonatal characteristics among those who were successfully followed up were compared with those who were lost to follow-up (Table 2). Neonates lost to follow-up were more likely than neonates assessed at follow-up to be singleton, be in a higher gestational age and birthweight strata, have higher umbilical cord pH, and have been exposed to chorioamnionitis. Other baseline characteristics did not differ.
Table 2. Maternal and Neonatal Characteristics of Patients Followed up vs Patients Lost to Follow-up.
| Characteristic | Patients Followed up (n = 3008) |
Patients Lost to Follow-up (n = 337) |
P Valuea |
|---|---|---|---|
| Maternal | |||
| Mother’s age, y | .11 | ||
| Mean (SD) | 27.4 (6.4) | 26.8 (6.4) | |
| Median (IQR) | 27 (22-32) | 26 (22-31) | |
| Maternal race/ethnicity | .10 | ||
| Black | 1227 (42) | 119 (36) | |
| White | 1559 (53) | 188 (57) | |
| Other | 168 (6) | 24 (7) | |
| Hispanic ethnicity | 500 (17) | 61 (18) | .65 |
| High school graduate | 2053 (77) | 235 (77) | >.99 |
| ≥1 prenatal visit | 2852 (95) | 312 (93) | .12 |
| Diabetes (insulin-dependent) | 147 (5) | 13 (4) | .50 |
| Parity | .47 | ||
| 0 or 1 | 1198 (40) | 146 (43) | |
| 2 or 3 | 1377 (46) | 146 (43) | |
| >3 | 433 (14) | 45 (13) | |
| Hypertension | .51 | ||
| Chronic (preexisting) | 319 (11) | 29 (9) | |
| Pregnancy-induced | 347 (12) | 38 (11) | |
| None | 2317 (78) | 268 (80) | |
| Prepartum hemorrhage | 614 (20) | 64 (19) | .62 |
| PPROM >18 h | 857 (29) | 89 (27) | .44 |
| Duration, h | |||
| Mean (SD) | 100.2 (179.8) | 76.0 (129.4) | .53 |
| Median (IQR) | 24.0 (2.8-121.7) | 16.6 (2.4-111.7) | .36 |
| Chorioamnionitis | 567 (19) | 81 (24) | .03 |
| Antenatal antibiotics during inpatient stay | 2190 (73) | 248 (74) | .95 |
| Antenatal steroids | 2682 (89) | 307 (91) | .40 |
| Multiple birth | 686 (23) | 60 (18) | .04 |
| Cesarean delivery | 1927 (64) | 215 (64) | .95 |
| Insurance | .42 | ||
| Medicaid/public | 1538 (52) | 171 (51) | |
| Private | 1214 (41) | 132 (39) | |
| Self/other/uninsured | 233 (8) | 33 (10) | |
| Neonatal | |||
| Gestational age, wk | <.001 | ||
| Mean (SD) | 24.9 (1.0) | 25.1 (1.0) | |
| Median (IQR) | 25 (24-26) | 25 (24-26) | |
| Birth weight, g | .04 | ||
| 401-500 | 168 (6) | 11 (3) | |
| 501-750 | 1473 (49) | 148 (44) | |
| 751-1000 | 1171 (39) | 153 (45) | |
| >1000 | 196 (7) | 25 (7) | |
| Male | 1531 (51) | 154 (46) | .07 |
| SGA at birth | 212 (7) | 17 (5) | .21 |
| Delivery room resuscitation | |||
| Intubation | 2455 (82) | 269 (80) | .42 |
| Epinephrine | 119 (4) | 8 (2) | .18 |
| Chest compression | 252 (8) | 20 (6) | .14 |
| Apgar score <3 | |||
| At 1 min | 1023 (34) | 98 (29) | .07 |
| At 5 min | 227 (8) | 19 (6) | .23 |
| Cord pH | .004 | ||
| Mean (SD) | 7.3 (0.1) | 7.3 (0.1) | |
| Median (IQR) | 7.3 (7.2-7.4) | 7.3 (7.3-7.4) | |
| Base deficit, mmol/L | .98 | ||
| Mean (SD) | 4.3 (4.4) | 4.5 (5.3) | |
| Median (IQR) | 3 (1-6) | 4 (1-6) | |
Abbreviations: IQR, interquartile range; PPROM, prolonged premature rupture of membranes; SGA, small for gestational age.
P values are from Fisher exact test for categorical variables and nonparametric Wilcoxon 2-sample tests (t approximation) for continuous variables.
Table 3 presents maternal and neonatal demographic data for the study participants. Compared with participants with normal cranial ultrasonograms, those with ventriculomegaly had lower gestational age, were male, died prior to discharge, and had neonatal morbidities commonly associated with systemic inflammation: respiratory distress syndrome treated with the administration of surfactant, late-onset sepsis, meningitis, BPD, NEC, stage 3 retinopathy of prematurity, and inflammation-initiating illnesses (P < .05; all data supporting comparisons are detailed in Table 3).
Table 3. Participant Demographics.
| Characteristic | Normal Cranial Ultrasonogram (n = 2727) |
Nonhemorrhagic Ventriculomegaly (n = 281) |
P Value |
|---|---|---|---|
| Maternal | |||
| Mother’s age, y | .57 | ||
| Mean (SD) | 27.3 (6.4) | 27.5 (6.4) | |
| Median (IQR) | 27 (22-32) | 28 (22-32) | |
| Maternal race/ethnicity | .28 | ||
| Black | 1122 (42) | 105 (39) | |
| White | 1406 (52) | 153 (57) | |
| Other | 157 (6) | 11 (4) | |
| Hispanic ethnicity | 451 (17) | 49 (19) | .55 |
| High school graduate | 1874 (77) | 179 (74) | .23 |
| ≥1 Prenatal visit | 2582 (95) | 270 (96) | .40 |
| Diabetes | 127 (5) | 20 (7) | .08 |
| Hypertension | .89 | ||
| Chronic (preexisting) | 287 (11) | 32 (11) | |
| Pregnancy-induced | 315 (12) | 32 (11) | |
| None | 2102 (78) | 215 (77) | |
| PPROM >18 h | 783 (29) | 74 (26) | .37 |
| Duration of PPROM, h | .26 | ||
| Mean (SD) | 99.8 (174.6) | 103.7 (222.8) | |
| Median (IQR) | 24.8 (3.0-122.7) | 13.5 (1.3-115.6) | |
| Chorioamnionitis | |||
| Histologicala | 1229 (54) | 125 (55) | .89 |
| Clinicala | 524 (19) | 43 (15) | .11 |
| Antenatal antibiotics | 1990 (73) | 200 (71) | .48 |
| Antenatal steroids | 2436 (90) | 246 (88) | .26 |
| Cesarean delivery | 1754 (64) | 173 (62) | .36 |
| Neonatal characteristics | |||
| Gestational age, wk | <.001 | ||
| Mean (SD) | 24.9 (1.0) | 24.7 (1.0) | |
| Median (IQR) | 25 (24-26) | 25 (24-26) | |
| Birth weight, g | .47 | ||
| 401-500 | 153 (6) | 15 (5) | |
| 501-750 | 1323 (49) | 150 (53) | |
| 751-1000 | 1073 (39) | 98 (35) | |
| >1000 | 178 (7) | 18 (6) | |
| Male | 1367 (50) | 164 (58) | .01 |
| SGA at birth | 195 (7) | 17 (6) | .62 |
| Neonatal | |||
| Delivery room resuscitation | |||
| Intubation | 2224 (82) | 231 (82) | .87 |
| Epinephrine | 110 (4) | 9 (3) | .63 |
| Chest compression | 233 (9) | 19 (7) | .37 |
| Apgar score <3 | |||
| At 1 min | 922 (34) | 101 (36) | .47 |
| At 5 min | 202 (7) | 25 (9) | .35 |
| Cord pH | .32 | ||
| Mean (SD) | 7.3 (0.1) | 7.3 (0.1) | |
| Median (IQR) | 7.3 (7.3-7.4) | 7.3 (7.2-7.4) | |
| Base deficit, mmol/L | .12 | ||
| Mean (SD) | 4.3 (4.4) | 4.7 (4.1) | |
| Median (IQR) | 3 (1-6) | 4 (2-6) | |
| Death prior to discharge | 572 (21) | 78 (28) | .01 |
| RDS treated with surfactant | 2457 (90) | 264 (94) | .04 |
| Postnatal steroids | 471 (17) | 50 (18) | .80 |
| Physiologic BPD | 1144 (53) | 142 (66) | <.001 |
| Early-onset sepsis | 55 (2) | 7 (2) | .51 |
| Late-onset sepsis | 928 (35) | 127 (46) | <.001 |
| Early-onset meningitisb | 6 (0.2) | 0 (0) | >.99 |
| Meningitis (any) | 60 (2) | 15 (5) | .004 |
| NEC | 342 (13) | 56 (20) | .001 |
| ROP ≥ stage 3 | 470 (21) | 66 (29) | .01 |
| IFIc | 1137 (45) | 140 (53) | .01 |
| Chorioamnionitis plus IFI | |||
| Histological | 490 (20) | 63 (26) | .04 |
| Clinical | 223 (8) | 30 (11) | .17 |
Abbreviations: BPD, bronchopulmonary dysplasia; IFI, inflammation initiating illness; IQR, interquartile range; NEC, necrotizing enterocolitis; PPROM, prolonged premature rupture of membranes; RDS, respiratory distress syndrome; ROP, retinopathy of prematurity; SGA, small for gestational age.
Clinical chorioamnionitis was based on the mother’s medical record; histologic chorioamnionitis was only available when placental pathology was performed.
Early-onset meningitis includes meningitis within 3 days of birth; since we only collect dates (not times), only episodes occurring on the day of birth or the subsequent 2 days were included.
Inflammation-initiating illness was defined as any 1 or more of the following: mechanical ventilation at 2 weeks, culture-proven sepsis or meningitis, and surgical NEC or intestinal perforation.
Outcomes at 18 to 22 Months’ Corrected Age
Outcomes of neonates with normal cranial ultrasonographic findings were compared with outcomes of those with ventriculomegaly (Table 4) at the corrected age of 18 to 22 months. Models were adjusted for the following covariates: nonhemorrhagic ventriculomegaly, gestational age, sex, use of antenatal steroids, maternal educational attainment less than high school diploma, late-onset sepsis, physiologic BPD, NEC, and retinopathy of prematurity, as well as by Eunice Kennedy Shriver NICHD NRN center (as a random effect). Neonates with ventriculomegaly had higher odds of NDI (odds ratio [OR], 3.07; 95% CI, 2.13-4.43), cognitive impairment (OR, 3.23; 95% CI, 2.09-4.99), moderate to severe cerebral palsy (OR, 3.68; 95% CI, 2.08-6.51), and death or NDI (OR, 2.17; 95% CI, 1.62-2.91) compared with premature neonates with normal cranial ultrasonograms; however, they did not have higher odds of death alone (OR, 1.09; 95% CI, 0.76-1.57). Median cognitive composite scores, language scores, and motor scores were 6 to 7 points lower (SD, −0.40 to 0.47) among neonates with nonhemorrhagic ventriculomegaly compared with neonates with normal cranial ultrasonographic findings; this is a clinically meaningful result. Behavioral outcomes assessed with the Brief Infant-Toddler Social and Emotional Assessment were not associated with ventriculomegaly, although rates of behavioral problems were higher than those typically reported for normally developing children.
Table 4. Neurodevelopmental Outcomes by Cranial Ultrasonographic Findingsa.
| Outcomes | Normal Cranial Ultrasonogram | Nonhemorrhagic Ventriculomegaly | OR or β (95% CI)b |
|---|---|---|---|
| 18-22 mo outcomes, No. (%) | 2727 | 281 | |
| Neurodevelopmental impairment, No. (%) | 249 (12) | 63 (32) | 3.07 (2.13 to 4.43) |
| Cognitive scorec | |||
| Cognitive score <85, No. (%) | 503 (24) | 93 (47) | 2.74 (1.96 to 3.82) |
| Cognitive score <70, No. (%) | 130 (6) | 41 (21) | 3.23 (2.09 to 4.99) |
| Cognitive score <55, No. (%) | 27 (1) | 14 (7) | 4.41 (2.10 to 9.26) |
| Median (IQR) | 90 (85-100) | 85 (70-95) | −6.57 (−8.60 to −4.55) |
| Mean (SD) | 90.7 (14.1) | 82.9 (17.6) | |
| Language scorec | |||
| Language score <85, No. (%) | 975 (47) | 127 (66) | 2.05 (1.46 to 2.89) |
| Language score <70, No. (%) | 311 (15) | 58 (30) | 2.08 (1.44 to 2.99) |
| Language score <55, No. (%)d | 32 (2) | 14 (7) | 4.07 (2.01 to 8.24) |
| Median (IQR) | 86 (74-97) | 77 (68-89) | −6.00 (−8.38 to −3.62) |
| Mean (SD) | 85.7 (16.5) | 77.8 (17.2) | |
| Cognitive and language scores <85, No. (%)c | 418 (20) | 79 (41) | 2.56 (1.82 to 3.60) |
| Motor scorec | |||
| Motor score <85, No. (%) | 389 (27) | 65 (47) | 2.00 (1.36 to 2.94) |
| Motor score <70, No. (%) | 119 (8) | 37 (27) | 3.23 (2.01 to 5.19) |
| Motor score <55, No. (%) | 32 (2) | 10 (7) | 1.73 (0.70 to 4.32) |
| Median (IQR) | 91 (82-100) | 85 (64-97) | −6.92 (−9.51 to −4.34) |
| Mean (SD) | 90.0 (14.9) | 81.0 (19.2) | |
| Cerebral palsy, moderate to severe, No. (%) | 61 (3) | 22 (11) | 3.68 (2.08 to 6.51) |
| Behavioral problemse | |||
| Rating <26th percentile, No. (%)f | 803 (39) | 66 (36) | 0.92 (0.66 to 1.27) |
| Parental worry ≥3, No. (%)g | 175 (8) | 18 (9) | 1.20 (0.71 to 2.03) |
| Hearing impairment, No. (%) | 54 (3) | 10 (5) | 1.57 (0.75 to 3.30) |
| Vision impairment, No. (%) | 13 (1) | 7 (4) | 4.22 (1.55 to 11.5) |
| Death prior to 18 mo, No. (%)h | 608 (22) | 85 (30) | 1.09 (0.76 to 1.57) |
| Death or NDI, No. (%)h | 856 (31) | 148 (53) | 2.17 (1.62 to 2.91) |
Abbreviations: BPD, bronchopulmonary dysplasia; IQR, interquartile range; NDI, neurodevelopmental impairment; OR, odds ratio.
Models adjust for the following covariates: nonhemorrhagic ventriculomegaly, gestational age, sex, antenatal steroids, maternal education less than high school, late-onset sepsis, physiologic BPD, necrotizing enterocolitis, retinopathy of prematurity, and the health care center at which the data was collected(as a random effect).
For categorical outcomes, the adjusted OR (95% CI) from logistic models is presented. For continuous outcomes, the regression coefficient for cranial ultrasonographic findings (β, 95% CI) from linear regression models reflects the adjusted difference in score with exposure to ventriculomegaly.
Scores from Bayley Scales of Infant and Toddler Development-III.
The statistical model for language score <55 was not adjusted for maternal education or physiologic BPD.
Scores from the Brief Infant Toddler Socioemotional Adjustment Scale; a problem cut score at or lower than 25% indicates a possible behavioral problem.
The statistical model for rating <26th percentile was not adjusted for maternal education.
Parental worry ≥3 means the parent is either “worried” or “very worried” about the child’s behaviors, emotions, and relationships.
Models for NDI or Death and Death are not adjusted for physiologic BPD or retinopathy of prematurity; those variables are largely missing for infants who died.
Discussion
Nonhemorrhagic ventriculomegaly is an important but underrecognized neuroimaging finding that predicts early childhood neurodevelopmental risk. Despite an association in magnetic resonant imaging studies with poor neurodevelopmental outcomes data on nonhemorrhagic ventriculomegaly remain sparse, to our knowledge. Although identification is potentially easy, these findings are not routinely collected by clinicians and researchers in many quality-improvement collaboratives focused on very low-birth-weight infants.
Several of our findings are worthy of comment. First, nonhemorrhagic ventriculomegaly was observed in 7% of participants, typically bilateral, and often identified after multiple cranial ultrasonograms had been completed. Second, compared with neonates with normal cranial ultrasonographic findings, neonates with nonhemorrhagic ventriculomegaly had a 2-fold to 3-fold higher odds of NDI, cognitive impairment, moderate to severe cerebral palsy, and death or NDI, but not death alone. Finally, there was no association between nonhemorrhagic ventriculomegaly and parent-reported behavioral outcomes, although behavioral problems were more common in children with ventriculomegaly than in term population norms per parental reports.
Most studies on the association between ventriculomegaly detected by cranial ultrasonography and outcomes among preterm neonates include lesions with prior or concurrent PIVH. Ment and colleagues were among the first to study the cause and outcome of cerebral ventriculomegaly in very low-birth-weight neonates. This pioneering work found a strong association between ventriculomegaly at term-equivalent age and 2 short-term neonatal morbidities: high-grade PIVH and BPD. Long-term outcomes were assessed prospectively at the corrected age of 4.5 years; ventriculomegaly was associated with both motor and cognitive impairment and was the most important predictor of IQ less than 70.
Further work by the ELGAN group also investigated the association between unspecified ventriculomegaly (with and without hemorrhage) and short-term and long-term morbidities among neonates born prior to 28 weeks’ gestation. These investigators showed that 12% of neonates had moderate or severe ventriculomegaly on assessment by cranial ultrasonography at up to 3 time points: between days 1 and 4, days 5 to 14, and days 15 through the 40th postconceptional week. Moderate to severe ventriculomegaly was identified with the use of templates for evaluating ventricular size. Ultrasonograms were reviewed by at least 2 study ultrasonographic technicians, who were blinded to clinical data; a third, tie-breaking technician addressed discrepancies between the 2 independent assessments.
Ventriculomegaly was associated with an increased risk of microcephaly and cerebral palsy at age 2 years, with cerebral palsy affecting 44% of children with ventriculomegaly. Compared with children without cranial ultrasonographic abnormalities, those with ventriculomegaly were 17 times more likely to have quadriparesis or hemiparesis and 11 times more likely to have significant gross motor delay (ie, Gross Motor Function Classification System scores of 2 or more). Ventriculomegaly was also associated with an increased risk of developmental delay, including a 4-fold increase in the risk of psychomotor impairment and a 3-fold increase in the risk of mental impairment as measured by the Bayley Scales of Infant Development II.
Other investigators have reported on the natural history of posthemorrhagic ventricular dilation; the severity of PIVH has been found to be a major predictor of adverse outcomes. Neonates with ventriculomegaly or posthemorrhagic hydrocephalus had worse neurodevelopmental outcomes at 18 to 22 months’ corrected age than those who had hemorrhage without persistent ventricular enlargement.
Similar to prior work, the present study reports poor neurodevelopmental outcomes among EP neonates with ventriculomegaly. However, our focus was on ventriculomegaly without concomitant or preceding hemorrhage. Our data suggest that the accumulation of inflammatory diagnoses might affect the likelihood of ventriculomegaly progression in an infant younger than 28 days to 36 weeks old and thus impact the brain development and subsequent neurodevelopmental trajectory of that infant. In the future, we hope to investigate the pathogenesis and timing of progressive ventriculomegaly in greater detail with serial ventricular measurements. The recent publication of new ultrasonographic reference values might facilitate more consistent identification of ventriculomegaly and facilitate further study.
Important strengths of our work included the large sample size, which permitted assessment of a variety of neurodevelopmental outcomes of interest; the assimilation of a multicenter cohort (which might have increased the generalizability of the results); the prospective data collection; and the carefully designed and detailed follow-up assessments. Follow-up assessments were performed by blinded examiners who were certified for reliability on an annual basis.
Limitations
The early end point of neurodevelopmental follow-up is a limitation deserving mention. Given the progressive trajectories of many neurocognitive and behavioral skills, longer follow-up of development as the child grows is essential for accurate appraisal of neurobehavioral functioning.
Second, we admit that an obvious limitation of our work is the potential use of differing strategies by individual site radiologists in the interpretation of cranial ultrasonograms. Center differences in cranial ultrasonography technique might result in failures to identify milder ventriculomegaly or mild hemorrhage. Thus, the exclusion of mild PIVH prior to the development of ventriculomegaly might not be possible.
Third, specific ventricular measurements were not recorded in the NRN generic database. Given the fairly constant size of the fetal lateral ventricles from 14 to 38 weeks’ gestation, however, a clinical diagnosis of nonhemorrhagic ventriculomegaly remains an important and readily identifiable neuroimaging finding.
As our intent was to capture significant brain injury, the approach of the present study to ultrasonography may predict severe adverse outcomes without overidentifying brain injuries that are amenable to adaptive mechanisms of the developing brain. As a result, the generalizability of our results may also be broader for the practicing neonatologist because this approach uses clinically available data.
An additional limitation was that the precise timing of onset of ventriculomegaly was not reported. Cranial ultrasonogram findings were recorded for sonograms performed in 2 time periods: within 28 days of birth and at 36 weeks’ postmenstrual age (or the date closest to that point).
Finally, a disproportionate number of participants with exposure to maternal chorioamnionitis were lost to follow-up. Chorioamnionitis is associated with subsequent development of ventriculomegaly and poor outcomes, so we might have underestimated the effect of ventriculomegaly on our participants.
Conclusions
Though less sensitive than magnetic resonance imaging in detecting white and gray matter loss, the bedside capability for cranial ultrasonography to identify clinically significant brain lesions in a timely and cost-effective manner make it an important neuroimaging technique for screening large numbers of critically ill neonates. Nonhemorrhagic ventriculomegaly is associated with early childhood developmental risk. Despite the greater emphasis on long-term outcomes of neonates with PIVH in the neonatal literature, nonhemorrhagic ventriculomegaly on sonogram is associated with worse outcomes. In a contemporaneous cohort of EP neonates born at NRN centers, severe PIVH was associated with an approximately 1.7-fold higher odds of NDI; the present study shows nearly 3-fold higher odds of NDI among neonates with ventriculomegaly. This work provides important data to consider when evaluating the need for advanced neuroimaging, parental counseling, and referral to early intervention and support services. Further study is needed to assess more stringent definitions of ventriculomegaly, the specific brain areas that are most compromised (eg, white matter, association tracts, basal ganglia, or cortex), the progression and timing of ventriculomegaly, the intervening factors, and prospective neuroprotective strategies that might ameliorate systemic inflammation and ventriculomegaly.
eTable 1. Participant Demographics (excluding neonates with PVL)
eTable 2. Neurodevelopmental Outcomes by Cranial Ultrasound Findings (excluding neonates with PVL)
References
- 1.Woodward LJ, Clark CA, Bora S, Inder TE. Neonatal white matter abnormalities an important predictor of neurocognitive outcome for very preterm children. PLoS One. 2012;7(12):e51879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Miller SP, Ferriero DM, Leonard C, et al. Early brain injury in premature newborns detected with magnetic resonance imaging is associated with adverse early neurodevelopmental outcome. J Pediatr. 2005;147(5):609-616. [DOI] [PubMed] [Google Scholar]
- 3.Inder TE, Wells SJ, Mogridge NB, Spencer C, Volpe JJ. Defining the nature of the cerebral abnormalities in the premature infant: a qualitative magnetic resonance imaging study. J Pediatr. 2003;143(2):171-179. [DOI] [PubMed] [Google Scholar]
- 4.Payne AH, Hintz SR, Hibbs AM, et al. ; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network . Neurodevelopmental outcomes of extremely low-gestational-age neonates with low-grade periventricular-intraventricular hemorrhage. JAMA Pediatr. 2013;167(5):451-459. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.O’Shea TM, Allred EN, Kuban KC, et al. ; ELGAN Study Investigators . Intraventricular hemorrhage and developmental outcomes at 24 months of age in extremely preterm infants. J Child Neurol. 2012;27(1):22-29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Stoll BJ, Hansen NI, Bell EF, et al. ; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network . Neonatal outcomes of extremely preterm infants from the NICHD Neonatal Research Network. Pediatrics. 2010;126(3):443-456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Shankaran S, Langer JC, Kazzi SN, Laptook AR, Walsh M; National Institute of Child Health and Human Development Neonatal Research Network . Cumulative index of exposure to hypocarbia and hyperoxia as risk factors for periventricular leukomalacia in low birth weight infants. Pediatrics. 2006;118(4):1654-1659. [DOI] [PubMed] [Google Scholar]
- 8.Hamrick SE, Miller SP, Leonard C, et al. Trends in severe brain injury and neurodevelopmental outcome in premature newborn infants: the role of cystic periventricular leukomalacia. J Pediatr. 2004;145(5):593-599. [DOI] [PubMed] [Google Scholar]
- 9.Fox LM, Choo P, Rogerson SR, et al. The relationship between ventricular size at 1 month and outcome at 2 years in infants less than 30 weeks’ gestation. Arch Dis Child Fetal Neonatal Ed. 2014;99(3):F209-F214. [DOI] [PubMed] [Google Scholar]
- 10.Maunu J, Lehtonen L, Lapinleimu H, et al. ; PIPARI Study Group . Ventricular dilatation in relation to outcome at 2 years of age in very preterm infants: a prospective Finnish cohort study. Dev Med Child Neurol. 2011;53(1):48-54. [DOI] [PubMed] [Google Scholar]
- 11.O’Shea TM, Allred EN, Dammann O, et al. ; ELGAN study Investigators . The ELGAN study of the brain and related disorders in extremely low gestational age newborns. Early Hum Dev. 2009;85(11):719-725. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Brouwer MJ, de Vries LS, Groenendaal F, et al. New reference values for the neonatal cerebral ventricles. Radiology. 2012;262(1):224-233. [DOI] [PubMed] [Google Scholar]
- 13.Hintz SR, Slovis T, Bulas D, et al. ; NICHD Neonatal Research Network . Interobserver reliability and accuracy of cranial ultrasound scanning interpretation in premature infants. J Pediatr. 2007;150(6):592-596, 596.e1-596.e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ment LR, Bada HS, Barnes P, et al. Practice parameter: neuroimaging of the neonate: report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society. Neurology. 2002;58(12):1726-1738. [DOI] [PubMed] [Google Scholar]
- 15.Wyldes M, Watkinson M. Isolated mild fetal ventriculomegaly. Arch Dis Child Fetal Neonatal Ed. 2004;89(1):F9-F13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cardoza JD, Goldstein RB, Filly RA. Exclusion of fetal ventriculomegaly with a single measurement: the width of the lateral ventricular atrium. Radiology. 1988;169(3):711-714. [DOI] [PubMed] [Google Scholar]
- 17.Stoll BJ, Hansen NI, Bell EF, et al. ; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network . Trends in care practices, morbidity, and mortality of extremely preterm neonates, 1993-2012. JAMA. 2015;314(10):1039-1051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Newman JE, Bann CM, Vohr BR, Dusick AM, Higgins RD; Follow-Up Study Group of Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network . Improving the Neonatal Research Network annual certification for neurologic examination of the 18-22 month child. J Pediatr. 2012;161(6):1041-1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.O’Shea TM, Shah B, Allred EN, et al. ; ELGAN Study Investigators . Inflammation-initiating illnesses, inflammation-related proteins, and cognitive impairment in extremely preterm infants. Brain Behav Immun. 2013;29:104-112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Vohr BR, Stephens BE, Higgins RD, et al. ; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network . Are outcomes of extremely preterm infants improving? Impact of Bayley assessment on outcomes. J Pediatr. 2012;161(2):222-8.e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Bayley N. Bayley Scales of Infant Development. 3rd ed San Antonio, TX: The Psychological Corporation; 2006. [Google Scholar]
- 22.Briggs-Gowan MJCA. Brief Infant-Toddler Social and Emotional Assessment (BITSEA) Manual, version 2. New Haven, CT: Yale University; 2002. [Google Scholar]
- 23.Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997;39(4):214-223. [DOI] [PubMed] [Google Scholar]
- 24.Spittle AJ, Ferretti C, Anderson PJ, et al. Improving the outcome of infants born at <30 weeks’ gestation—a randomized controlled trial of preventative care at home. BMC Pediatr. 2009;9(73). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Pocock SJ, Smith M, Baghurst P. Environmental lead and children’s intelligence: a systematic review of the epidemiological evidence. BMJ. 1994;309(6963):1189-1197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.McLendon D, Check J, Carteaux P, et al. Implementation of potentially better practices for the prevention of brain hemorrhage and ischemic brain injury in very low birth weight infants. Pediatrics. 2003;111(4 Pt 2):e497-e503. [PubMed] [Google Scholar]
- 27.Ment LR, Vohr B, Allan W, et al. The etiology and outcome of cerebral ventriculomegaly at term in very low birth weight preterm infants. Pediatrics. 1999;104(2 Pt 1):243-248. [DOI] [PubMed] [Google Scholar]
- 28.Krishnamoorthy KS, Kuban KC, O’Shea TM, Westra SJ, Allred EN, Leviton A; ELGAN Study Co-investigators . Early cranial ultrasound lesions predict microcephaly at age 2 years in preterm infants. J Child Neurol. 2011;26(2):188-194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kuban KC, Allred EN, O’Shea TM, et al. ; ELGAN study investigators . Cranial ultrasound lesions in the NICU predict cerebral palsy at age 2 years in children born at extremely low gestational age. J Child Neurol. 2009;24(1):63-72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.O’Shea TM, Kuban KC, Allred EN, et al. ; Extremely Low Gestational Age Newborns Study Investigators . Neonatal cranial ultrasound lesions and developmental delays at 2 years of age among extremely low gestational age children. Pediatrics. 2008;122(3):e662-e669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Murphy BP, Inder TE, Rooks V, et al. Posthaemorrhagic ventricular dilatation in the premature infant: natural history and predictors of outcome. Arch Dis Child Fetal Neonatal Ed. 2002;87(1):F37-F41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Adams-Chapman I, Hansen NI, Stoll BJ, Higgins R; NICHD Research Network . Neurodevelopmental outcome of extremely low birth weight infants with posthemorrhagic hydrocephalus requiring shunt insertion. Pediatrics. 2008;121(5):e1167-e1177. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Brouwer AJ, Groenendaal F, Han KS, de Vries LS. Treatment of neonatal progressive ventricular dilatation: a single-centre experience. J Matern Fetal Neonatal Med. 2015;28(suppl 1):2273-2279. [DOI] [PubMed] [Google Scholar]
- 34.Pappas A, Kendrick DE, Shankaran S, et al. ; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network . Chorioamnionitis and early childhood outcomes among extremely low-gestational-age neonates. JAMA Pediatr. 2014;168(2):137-147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Hintz SR, Barnes PD, Bulas D, et al. ; SUPPORT Study Group of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network . Neuroimaging and neurodevelopmental outcome in extremely preterm infants. Pediatrics. 2015;135(1):e32-e42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Inder TE, Anderson NJ, Spencer C, Wells S, Volpe JJ. White matter injury in the premature infant: a comparison between serial cranial sonographic and MR findings at term. AJNR Am J Neuroradiol. 2003;24(5):805-809. [PMC free article] [PubMed] [Google Scholar]
- 37.Maalouf EF, Duggan PJ, Counsell SJ, et al. Comparison of findings on cranial ultrasound and magnetic resonance imaging in preterm infants. Pediatrics. 2001;107(4):719-727. [DOI] [PubMed] [Google Scholar]
- 38.Miller SP, Cozzio CC, Goldstein RB, et al. Comparing the diagnosis of white matter injury in premature newborns with serial MR imaging and transfontanel ultrasonography findings. AJNR Am J Neuroradiol. 2003;24(8):1661-1669. [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Participant Demographics (excluding neonates with PVL)
eTable 2. Neurodevelopmental Outcomes by Cranial Ultrasound Findings (excluding neonates with PVL)
