Abstract
Objectives:
To determine the incidence, timing, progression, and risk factors for intracranial hemorrhage (ICH) in infants 24–0/7 to 27–6/7 weeks’ gestational age (GA), and to characterize the association between ICH and death or neurodevelopmental impairment (NDI) at 2 years corrected age (CA).
Study design:
Infants enrolled in the Preterm Erythropoietin Neuroprotection Trial had serial cranial ultrasounds performed on day 1, day 7–9, and 36-weeks postmenstrual age (PMA) to evaluate ICH. Potential risk factors for development of ICH were examined. Outcomes included death or severe NDI as well as Bayley Scales of Infant and Toddler Developmental, 3rd Edition (BSID-III) at 2-years CA.
Results:
ICH was identified in 38% (n=339) of 883 enrolled infants. Multiple gestation and cesarean delivery reduced the risk of any ICH on day 1. Risk factors for development of bilateral Grade 2, or Grade 3 or 4 ICH at day 7–9 included any ICH at day 1, and two or more doses of prenatal steroids decreased risk. Bilateral Grade 2, Grade 3, or Grade 4 ICH at 36-weeks were associated with prior ICH at day 7–9. Bilateral Grade 2, any Grade 3, and any Grade 4 ICH at 7–9 days or 36-weeks PMA were associated with increased risk of death or severe NDI and lower BSID-III scores.
Conclusion:
Risk factors for ICH varied by timing of bleed. Bilateral and increasing grade of ICH were associated with death or NDI in infants born extremely preterm.
Advances in the care of infants born before 28 weeks of gestation have been associated with a substantial improvement in survival.(1) However, the risk of neurodevelopmental impairment (NDI) for extremely preterm (EP, <28 weeks’ gestation) and extremely low birth weight (<1000 grams) infants remains high. More than half of surviving infants develop cognitive, social, behavioral, learning and/or motor deficits.(2, 3)
Germinal matrix intraventricular hemorrhage is the most common form of intracranial hemorrhage (ICH) and the most recognized cerebral lesion on cranial ultrasound (CUS) in premature infants. The incidence of hemorrhage is as high as 45% in infants born <1000 grams and has not decreased despite advances in obstetric and neonatal intensive care.(4) A review of 34,636 infants born 1993 to 2012 at 22 to 28 weeks’ gestation with birthweights 401 to 1500g reported reduction in severe ICH for more mature infants, but no change in the incidence of severe ICH in children born <26 weeks.(1)
The Papile classification system for ICH severity diagnosed by CUS remains a widely used clinical decision-making and parental counseling tool.(5) Grades 3 and 4 ICH have been shown to be important contributors to morbidity and mortality in preterm infants, whereas the consequences of Grade 1 and 2 hemorrhages are less certain.(6) In some studies, low-grade ICH has not been associated with adverse outcomes beyond the risks associated with prematurity alone, and other studies have challenged this notion.(7–10)
Multiple studies have examined the timing of, as well as pre- and postnatal risk factors for, ICH in preterm infants; however, most of the available data are from older studies and are mostly inconclusive.(11–18) Prediction of infants who are most at risk for subsequent neurodevelopmental disability in this population remains a major challenge to the clinician.
Using the available data from infants enrolled in the Preterm Erythropoietin (Epo) Neuroprotection (PENUT) Trial, our objectives were to determine the incidence, timing, and progression of ICH in this contemporary cohort of EPs, to evaluate potential risk factors for ICH linked to worse outcomes, and to examine the association between ICH and neurodevelopmental outcomes at 2 years corrected age (CA). We hypothesized that ICH would be more frequent at lower gestational ages, and that higher ICH grades would have a greater impact on 2-year CA outcomes.
METHODS
This study is a secondary analysis of pre-existing data from the PENUT Trial (NCT #01378273), a randomized placebo-controlled trial designed to study the effects of Epo on the primary outcome of death or severe NDI in infants born between 24 and 28 weeks of gestation. Details of the protocol, IRB approvals, consent, enrollment, methods, and primary outcome have been previously described.(19, 20)
Study Population:
All infants enrolled in the PENUT Trial who underwent a CUS on day of life 1 were eligible for this study. Infants with known life-threatening anomalies, chromosomal anomalies, disseminated intravascular coagulopathy, hydrops fetalis, twin-to-twin transfusion that resulted in hydrops, a hematocrit level above 65%, or known congenital infection were excluded from PENUT enrollment.
Definitions:
Imaging and ICH grading.
The PENUT Trial manual of operations included a standardized protocol for obtaining CUS images across all sites. By protocol, enrolled infants received a CUS within the first 24 hours after birth prior to study drug administration and then a repeat CUS at 7–9 days of age. At 36-weeks PMA, infants received a CUS, magnetic resonance imaging (MRI), or both, depending on the study site. MRI results and outcomes will be evaluated in a separate study as term equivalent age MRI is not a current standard of care, and thus inclusion of this data would make our results less generalizable to the majority of infants.
CUS examinations were performed in the NICU at each individual site. A central reader, who was blinded to the patient’s identification and study arm, reviewed the images and filled out Redcap form online with details of the study. The classification of cranial ultrasound findings published in the 2002 Practice parameter were used,(21) and are briefly outlined in Table I (available at www.jpeds.com). When present, ICH was scored according to the Papile criteria as Grades 1 to 4 with the highest grade bleed on each side of the brain documented.(5) When bilateral bleeds were present, they were given an overall score based on the most severe grade present. Presence of periventricular leukomalacia (PVL), ventriculomegaly, and/or cerebellar abnormalities were also noted.
Table 1: Classification and scoring system for CUS.
All CUS images were centrally read, with intracranial hemorrhage, white matter injury, and ventriculomegaly scored according to previously-published criteria.
| Classification | Findings | |
|---|---|---|
| Intracranial hemorrhage1 | Grade 1 | Germinal matrix hemorrhage |
| Grade 2 | Blood within the ventricular system but not distending it | |
| Grade 3 | Intracranial hemorrhage with ventricular dilatation | |
| Grade 4 | Parenchymal involvement | |
| Preterm white matter injury2, 3 | Cystic lesions | Periventricular |
| Ventriculomegaly4, 5 | Mild | 0.5–1.0 cm§ |
| Moderate | 1.0–1.5 cm§ | |
| Severe | >1.5 cm§ |
Measurements at the midbody of the lateral ventricle on sagittal scan.
Neurodevelopmental outcomes:
Subjects were evaluated by centrally certified examiners who were unaware of the patient’s treatment arm, medical history, and imaging results at 2 years corrected age (22–26 months) using Bayley Scales of Infant and Toddler Development, 3rd Edition (BSID-III) and standardized neurologic examination. Severe NDI was defined as severe cerebral palsy (CP) and/or BSID-III Composite Motor score or Composite Cognitive score <70. CP was classified by location (diplegia, hemiplegia, or quadriplegia) and severity as determined by the Gross Motor Function Classification System (GMFCS), with a level of 2 considered moderate and 3 or greater considered to be severe.(22, 23)
Potential risk factors for ICH:
Detailed data were collected prospectively by centrally trained research coordinators and included maternal and infant baseline characteristics. Maternal characteristics included maternal education, obesity, drug or substance abuse, hypertensive disorders of pregnancy, receipt of prenatal care, multiple births, preterm labor, prolonged rupture of membranes (PROM), acute chorioamnionitis, antibiotic exposure, prenatal steroid and magnesium administration, and mode of delivery. Infant characteristics included treatment group, gestational age (GA), sex, birth weight, delayed cord clamping, Apgar score at 5 minutes, resuscitation in the delivery room, presence of an ICH bundle at the delivering hospital, medications received, and complications of prematurity such as sepsis, severe necrotizing enterocolitis (NEC), surgical closure of the patent ductus arteriosus (PDA), severe bronchopulmonary dysplasia (BPD), and spontaneous intestinal perforation (SIP).
Gestational age was determined by the best obstetrical estimate. PROM was defined as rupture of membranes ≥18 hours prior to delivery. Maternal obesity was defined as BMI ≥ 30. Severe sepsis was defined as culture-proven bacterial or fungal sepsis resulting in blood-pressure support or substantive new respiratory support. Severe NEC was defined as Stage 2b or 3 according to the Modified Bell’s Staging Criteria. Severe BPD was defined as supplemental oxygen requirement at 36 weeks PMA. The use of an indomethacin protocol or ICH bundle was determined independently by each site, and specific data regarding which patients received which interventions was not collected.
Statistical Analyses
Simple descriptive statistics and graphical illustrations were used to initially describe the timing, severity, and progression of ICH. For ultimate statistical inference, generalized estimating equations (GEE) with robust standard errors were used to appropriately account for potential correlation of outcomes for same-birth siblings.(24) Using GEE logistic regression models adjusting for GA and treatment, we estimated odds ratios (OR) with 95% confidence intervals (CI) to characterize the association between the primary combined outcome measure of death or severe NDI and the severity of ICH separately at each measurement time (conditional upon surviving to be imaged). Results were then stratified separately for death and for each of the BSID-III subscales, where GEE models were also used to compare BSID-III scores by ICH status at each scan time point. For infants who received all three scans a “worst” ICH grade was calculated and then associated with the dichotomous primary NDI outcome and BSID-III subscales in survivors.
Kaplan-Meier survival curves were stratified by day 7–9 ICH status, where a ‘landmark’ approach was utilized and included participants conditional upon survival to day 7–9 and having a CUS.(25) Cox proportional hazard ratio (HR) models were used to compare mortality risk over time by ICH status.
Exploratory analyses were performed to determine the effect of bilateral ICH after adjusting for overall ICH grade. As a result of the imbalance in unilateral and bilateral bleeds in Grades 1, 3, and 4 ICH, the effect of bilateral ICH could only meaningfully be explored in Grade 2 ICH at day 7–9 and at 36 weeks PMA.
For each imaging time point we considered a modified grading criterion that split Grade 2 into unilateral and bilateral groups. We then empirically defined “NDI-associated ICH” based on the modified severity grades that were associated with death or severe NDI and decreased BSID-III scores. Ultimately, our NDI-associated ICH groupings were: any ICH at day 1; bilateral Grade 2, any Grade 3, and any Grade 4 ICH for both day 7–9 and week 36. Maternal and infant variables associated with an increased risk of NDI-associated ICH at each time point were then investigated. Each candidate risk factor was assessed individually after adjusting for GA, treatment group, and grade of ICH on the previous scan (for day 7–9 and week 36 models), with Bonferroni correction of p-values. Those candidate risk factors whose unadjusted p-values were <0.05 were subsequently entered into the final model.
All analyses were conducted using the R statistical package (Version 3.6.1, Foundation for Statistical Computing, Vienna, Austria).(26)
RESULTS
Of the 936 infants enrolled in the PENUT study, 94% (n=883) of infants had a CUS within the first 24 hours after birth and were included in the analyses, as shown in Table 2 and Figure 1 (available at www.jpeds.com). Ninety-three percent (n=827) of included infants had a follow-up CUS at day 7–9, and 66% (n=584) had a CUS at 36 weeks PMA (Table 3). There were no significant differences in the number or severity of bleeds between placebo and Epo groups at any imaging time point.
Table 2: Demographics of CUS cohort at baseline.
Individual maternal and infant factors in the subset of PENUT infants for whom a CUS was available on Day 1.
| n (%) | |
|---|---|
| Total Day 1 CUS | 883 |
| Maternal Factors | |
| Maternal Education | |
| High school or less | 288 (32.6) |
| Some college | 269 (30.5) |
| College degree or greater | 228 (25.8) |
| Unknown or not reported | 98 (11.1) |
| Maternal Obesity | 90 (10.2) |
| Drugs, Alcohol, Smoking | 141 (16.0) |
| Maternal Hypertension | 179 (20.3) |
| Prenatal Care | 840 (95.1) |
| Multiple Gestation Pregnancy | 234 (26.5) |
| Preterm Labor | 545 (61.7) |
| Prolonged Rupture of Membranes | 232 (26.3) |
| Chorioamnionitis | 115 (13.0) |
| Prenatal Antibiotics | 319 (36.1) |
| Prenatal Steroids | |
| 0 Doses | 90 (10.2) |
| 1 Dose | 167 (18.9) |
| 2 Doses | 546 (61.8) |
| 3+ Doses | 72 (8.2) |
| Unknown | 8 (0.9) |
| Prenatal Magnesium | 706 (80.0) |
| C-section | 612 (69.3) |
| Infant Factors | |
| Gestational Age | |
| 24 weeks | 219 (24.8) |
| 25 weeks | 232 (26.3) |
| 26 weeks | 208 (23.6) |
| 27 weeks | 224 (25.4) |
| Female Sex | 426 (48.2) |
| Weight <10th Percentile | 137 (15.5) |
| Delayed Cord Clamping | 304 (34.4) |
| Intubation/Chest Compressions | 721 (81.7) |
| Apgar <5 at 5 mins | 178 (20.2) |
| Postnatal Factors | |
| NEC | 61 (6.9) |
| Severe Sepsis | 72 (8.2) |
| Severe BPD | 307 (34.8) |
Figure 1: CONSORT Diagram.
Flow and number of participants at each time point of imaging.
Table 3: Death/Severe NDI Outcomes by ICH Grade at 3 Time Points.
Prevalence of unilateral and bilateral bleeds by overall ICH grade at each CUS time point (upper table). Incidence and odds ratio (OR) with 95% confidence interval (CI) of death or severe NDI by grade of ICH at each time point (lower table).
| ICH Grade | |||||
|---|---|---|---|---|---|
| Day 1 CUS | 1 | 2 | 3 | 4 | |
| Unilateral n (%) | 41 (77%) | 85 (97%) | 24 (59%) | 3 (50%) | |
| Bilateral n (%) | 12 (23%) | 3 (3%) | 17 (41%) | 3 (50%) | |
| Day 7–9 CUS | |||||
| Unilateral n (%) | 41 (82%) | 63 (44%) | 5 (7%) | 4 (24%) | |
| Bilateral n (%) | 9 (18%) | 79 (56%) | 65 (93%) | 13 (76%) | |
| 36 Weeks PMA CUS | |||||
| Unilateral n (%) | 9 (69%) | 14 (33%) | 1 (5%) | 2 (19%) | |
| Bilateral n (%) | 4 (31%) | 29 (67%) | 21 (95%) | 5 (71%) | |
| Total (N=) | Death or severe NDI | ||||
| Day 1 CUS | 883 | n/N (%) * | OR (95%CI) | p-value | |
| No ICH | 695 | 132/606 (22%) | Reference | - | |
| Grade 1 | 53 | 13/46 (28%) | 1.34 (0.66–2.73) | 0.42 | |
| Grade 2 | 88 | 22/78 (28%) | 1.35 (0.79–2.33) | 0.27 | |
| Grade 3 | 41 | 14/38 (37%) | 1.94 (0.97–3.87) | 0.06 | |
| Grade 4 | 6 | 2/6 (33%) | 2.54 (0.47–13.8) | 0.28 | |
| Day 7–9 CUS | 827 | ||||
| No ICH | 548 | 66/474 (14%) | Reference | - | |
| Grade 1 | 50 | 9/45 (20%) | 1.48 (0.70–3.13) | 0.31 | |
| Grade 2 Unilateral | 63 | 11/53 (21%) | 1.45 (0.69–3.06) | 0.33 | |
| Grade 2 Bilateral | 79 | 27/73 (37%) | 3.19 (1.89–5.39) | <0.0001 | |
| Grade 3 | 70 | 22/60 (37%) | 2.99 (1.61–5.56) | 0.0005 | |
| Grade 4 | 17 | 12/17 (71%) | 13.5 (5.10–35.5) | <0.0001 | |
| 36 Weeks PMA CUS | 584 | ||||
| No ICH | 499 | 57/433 (13%) | Reference | - | |
| Grade 1 | 13 | 2/10 (20%) | 1.61 (0.34–7.62) | 0.55 | |
| Grade 2 Unilateral | 14 | 1/12 (8%) | 0.51 (0.06–4.09) | 0.53 | |
| Grade 2 Bilateral | 29 | 8/26 (31%) | 2.87 (1.16–7.10) | 0.02 | |
| Grade 3 | 22 | 10/19 (53%) | 6.26 (2.66–14.7) | <0.0001 | |
| Grade 4 | 7 | 3/5 (60%) | 11.5 (1.85–71.0) | 0.0004 | |
| Worst ICH (Of 3 scans) | 574 | ||||
| No ICH | 350 | 30/302 (10%) | Reference | - | |
| Grade 1 | 48 | 8/43 (19%) | 1.85 (0.79–4.37) | 0.16 | |
| Grade 2 Unilateral | 49 | 6/39 (15%) | 1.45 (0.54–3.87) | 0.46 | |
| Grade 2 Bilateral | 59 | 14/53 (26%) | 2.77 (1.35–5.66) | 0.005 | |
| Grade 3 | 52 | 14/44 (32%) | 3.38 (1.66–6.90) | 0.0008 | |
| Grade 4 | 16 | 8/14 (57%) | 12.6 (4.42–36.1) | <0.0001 | |
Percentage of infants from whom primary outcome data were available
Prevalence and Progression of ICH
ICH was identified in 38% (n=339) of the infants evaluated. Many hemorrhages progressed or resolved in severity over time, and new hemorrhages were noted at every time point (Figure 2). Overall, 56% of ICH occurred by day 1, 38% by day 7–9, and the remaining 6% by 36-weeks PMA.
Figure 2. Progression of ICH Grades following day 1 and day 7–9 CUS ICH status.
(A) status of infants on day 7–9 by day 1 ICH status. (B) status of infants at week 36 by day 7–9 ICH status.
Day 1 CUS:
Within the first 24 hours of birth, 21% (n=188) of infants had ICH documented with similar incidence across GAs (Table 4; available at www.jpeds.com). Nineteen percent (n=35) of day 1 bleeds were bilateral.
Table 4: ICH Grade by GA over time.
Prevalence of ICH by grade and gestational age at each CUS imaging time point.
| Gestational Age (weeks) | |||||
|---|---|---|---|---|---|
| 24 | 25 | 26 | 27 | All Infants | |
| Day 1 | n=219 | n=232 | n=208 | n=205 | n=883 |
| No ICH | 167 (76%) | 177 (76%) | 165 (79%) | 186 (83%) | 695 (79%) |
| Grade 1 ICH | 16 (7%) | 10 (4%) | 18 (9%) | 9 (4%) | 53 (6%) |
| Grade 2 ICH | 22 (10%) | 31 (13%) | 19 (9%) | 16 (8%) | 88 (10%) |
| Grade 3 ICH | 14 (6%) | 12 (5%) | 6 (2.9%) | 9 (4%) | 41 (5%) |
| Grade 4 ICH | 0 (0%) | 2 (1%) | 0 (0%) | 4 (2%) | 6 (1%) |
| Day 7–9 | n=199 | n=213 | n=199 | n=216 | n=827 |
| No ICH | 112 (56%) | 134 (63%) | 138 (69%) | 164 (76%) | 548 (66%) |
| Grade 1 ICH | 16 (8%) | 8 (4%) | 16 (8%) | 10 (5%) | 50 (6%) |
| Grade 2 ICH | 40 (20%) | 46 (22%) | 27 (14%) | 29 (13%) | 142 (17%) |
| Grade 3 ICH | 24 (12%) | 21 (10%) | 15 (8%) | 10 (5%) | 70 (8%) |
| Grade 4 ICH | 7 (4%) | 4 (2%) | 3 (2%) | 3 (1%) | 17 (2%) |
| Week 36 | n=136 | n=153 | n=141 | n=154 | n=584 |
| No ICH | 110 (80.9%) | 126 (82.4%) | 123 (87.2%) | 140 (90.9%) | 499 (85.4%) |
| Grade 1 ICH | 2 (1%) | 4 (3%) | 4 (3%) | 3 (2%) | 13 (2%) |
| Grade 2 ICH | 15 (11%) | 15 (10%) | 6 (4%) | 7 (5%) | 43 (7%) |
| Grade 3 ICH | 8 (6%) | 7 (5%) | 5 (3%) | 2 (1%) | 22 (4%) |
| Grade 4 ICH | 1 (1%) | 1 (1%) | 3 (2%) | 2 (1%) | 7 (1%) |
Percentages are column percentages for each age of scan.
Day 7–9 CUS:
The prevalence of ICH increased to 34% (n=279) by day 7–9 of life, 46% (n=129) of which were newly diagnosed. The prevalence of infants with ICH was highest in those born at 24 weeks GA (44%) and progressively decreased with increasing GA (Table 4). Fifty-nine percent (n=166) of all bleeds documented by 7–9 days were bilateral (Table 3). Of the newly diagnosed ICH cases, 19% were Grade 1, 57% were Grade 2, 18% were Grade 3, and 5% were Grade 4. Resolution of ICH between day 1 and day 7–9 was documented on 28 scans. Of these, 39% (n=11) were Grade 1, 50% (n=14) were Grade 2, 7% (n=2) were Grade 3, and 4% (n=1) were Grade 4.
36 weeks PMA CUS:
By 36 weeks PMA, the overall prevalence of ICH was 15% (n=85). Infants of 24 weeks’ GA again had the highest prevalence of ICH relative to infants of older GAs (Table 4), and 69% (n=59) of these were bilateral (Table 3). Of the 22 newly-identified ICHs, 32% (n=7) were Grade 1, 45% (n=10) were Grade 2, 14% (n=3) were Grade 3, and 9% (n=2) were Grade 4 bleeds. Of the 125 bleeds that resolved, 24% (n=30) were Grade 1, 52% (n=65) were Grade 2, 20% (n=25) were Grade 3, and 4% (n=5) were Grade 4.
Derived Maximum ICH grade Across All 3 Time Points:
Grade 1 ICH - 48 infants had a Grade 1 ICH as their worst ICH. Of these, 54% (n=26) were present on day 1, 69% (n=33) on day 7–9, and 25% (n=12) on the 36-week CUS. Grade 2 ICH - 108 infants had a Grade 2 ICH as their worst ICH. Of these, 38% (n=41) were present on day 1, 82% (n=89) on day 7–9, and 32% (n=35) on the 36-week CUS. Grade 3 ICH - 52 infants had a Grade 3 ICH as their worst ICH. Of these, 40% (n=21) were present on day 1, 88% (n=46) on day 7–9, and 35% (n=18) on the 36-week CUS. Grade 4 ICH - 16 infants had a Grade 4 ICH as their worst ICH. Of these, 31% (n=5) were present on day 1, 63% (n=10) on day 7–9, and 44% (n=7) on the 36-week CUS. Of note, the number of each grade of “worst” ICH at each time point does not necessarily sum to the total number of infants who had that grade as their worst ICH, as many infants had their worst grade of ICH present at multiple time points.
Secondary Intracranial Findings
PVL was seen in 3% (n=27) of infants at day 1, 8% (n=63) by day 7–9, and 3% (n=20) at 36-weeks PMA. Ventriculomegaly was documented in 4% (n=33) of infants at day 1, 10% (n=91) at day 7–9, and 12% (n=69) of infants at 36 weeks PMA. Cerebellar hemorrhage was identified in 1% (n=8) of infants on day 1, 2% (n=18) at day 7–9, and 1% (n=6) at 36 weeks PMA.
Risk Factors for ICH Associated with Death or Severe NDI
Uncorrected assessment of individual risk factors for ICH at day 1, days 7–9, and week 36 are presented in Tables 5–7 (available at www.jpeds.com). Following Bonferroni correction, multiple gestation and cesarean delivery were associated with a decreased risk of ICH on day 1 in the final model (Figure 3, A). Similarly, after adjustment for multiple comparisons, the following infant characteristics were associated with a statistically increased OR for Grade 2–4 ICH at 7–9 days: 25 weeks’ GA or prior ICH of any grade (Figure 3, B). Prenatal steroids decreased the odds in a dose-dependent manner. Increasing grade of prior day 7–9 ICH significantly increased the risk of 36 weeks PMA ICH (Figure 3, C).
Table 5: Maternal and infant risk factors for No ICH vs Any ICH on Day 1 CUS.
Individual maternal and infant risk factors for any ICH on day 1, adjusted for gestational age and treatment group.
| n= (%) | n= (%) | p-value# | |
|---|---|---|---|
| Total Day 1 CUS | 695 (78.7) | 188 (21.3) | |
| Maternal Factors | |||
| Maternal Education | |||
| High school or less | 226 (78.5) | 62 (21.5) | Reference |
| Some college | 209 (77.7) | 60 (22.3) | 0.8 |
| College degree or greater | 181 (79.4) | 47 (20.6) | 0.9 |
| Unknown or not reported | 79 (80.6) | 19 (19.4) | 0.6 |
| Maternal Obesity | |||
| No | 622 (78.4) | 171 (21.6) | 0.56 |
| Yes | 73 (81.1) | 17 (18.9) | |
| Drugs, Alcohol, Smoking | |||
| No | 541 (79.6) | 139 (20.4) | 0.78 |
| Yes | 111 (78.7) | 30 (21.3) | |
| NA | 43 (69.4) | 19 (30.6) | |
| Maternal Hypertension | |||
| No | 550 (78.1) | 154 (21.9) | 0.51 |
| Yes | 145 (81.0) | 34 (19.0) | |
| Prenatal Care | |||
| No | 22 (78.6) | 6 (21.4) | 0.98 |
| Yes | 662 (78.8) | 178 (21.2) | |
| Unknown | 11 (73.3) | 4 (26.7) | |
| Multiple Gestation Pregnancy | |||
| No | 493 (76.0) | 156 (24.0) | 0.003 |
| Yes | 202 (86.3) | 32 (13.7) | |
| Preterm Labor | |||
| No | 276 (81.7) | 62 (18.3) | 0.12 |
| Yes | 419 (76.9) | 126 (23.1) | |
| Prolonged Rupture of Membranes | |||
| No | 512 (78.6) | 139 (21.4) | 0.98 |
| Yes | 183 (78.9) | 49 (21.1) | |
| Chorioamnionitis | |||
| No | 606 (78.9) | 162 (21.1) | 0.77 |
| Yes | 89 (77.4) | 26 (22.6) | |
| Prenatal Antibiotics | |||
| No | 442 (78.4) | 122 (21.6) | 0.66 |
| Yes | 253 (79.3) | 66 (20.7) | |
| Prenatal Steroids | |||
| 0 Doses | 68 (75.6) | 22 (24.4) | |
| 1 Dose | 132 (79.0) | 35 (21.0) | 0.51 |
| 2 Doses | 427 (78.2) | 119 (21.8) | |
| 3+ Doses | 62 (86.1) | 10 (13.9) | |
| Unknown | 6 (75.0) | 2 (25.0) | |
| Prenatal Magnesium | |||
| No | 103 (72.5) | 39 (27.5) | 0.09 |
| Yes | 560 (79.3) | 146 (20.7) | |
| Unknown | 31 (91.2) | 3 (8.8) | |
| Caesarean Delivery | |||
| No | 195 (72.0) | 76 (28.0) | 0.002 |
| Yes | 500 (81.7) | 112 (18.3) | |
| Infant Factors | |||
| Gestational Age* | |||
| 24 weeks | 167 (76.3) | 52 (23.7) | |
| 25 weeks | 177 (76.3) | 55 (23.7) | 0.25 |
| 26 weeks | 165 (79.3) | 43 (20.7) | |
| 27 weeks | 186 (83.0) | 38 (17.0) | |
| Sex | |||
| Female | 346 (81.2) | 80 (18.8) | 0.08 |
| Male | 349 (76.4) | 108 (23.6) | |
| Weight <10th Percentile | |||
| No | 577 (77.7) | 166 (22.3) | 0.11 |
| Yes | 115 (83.9) | 22 (16.1) | |
| Delayed Cord Clamping | |||
| No | 258 (77.2) | 76 (22.8) | 0.31 |
| Yes | 245 (80.6) | 59 (19.4) | |
| Unknown | 192 (78.4) | 53 (21.6) | |
| Intubation/Chest Compressions | |||
| No | 136 (84.0) | 26 (16.0) | 0.18 |
| Yes | 559 (77.5) | 162 (22.5) | |
| Apgar <5 at 5 min | |||
| No | 559 (80.7) | 134 (19.3) | 0.4 |
| Yes | 134 (75.3) | 44 (24.7) | |
| General Appearance | |||
| Well | 390 (78.6) | 106 (21.4) | 0.85 |
| Sick | 303 (79.1) | 80 (20.9) |
All models adjusted for treatment group and gestational age. P-values are derived from comparison to a baseline model only including group and gestational age. In models with multiple levels, P-values are either for levels as factors (with individual P-values for each level) or for a linear trend across levels (single P-value).
Gestational age model adjusted for treatment group only
Bonferroni correction for 21 comparisons, adjusted significant p-value=0.05/20=0.0025
Table 7: Risk factors for ICH on 36 weeks CUS.
Individual maternal and infant risk factors for bilateral grade 2, grade 3 or grade 4 ICH on week 36 CUS, adjusted for day 7–9 ICH status, gestational age, and treatment group.
| No ICH or Grade 1 or Unilateral Grade 2 ICH n= (%) | Bilateral Grade 2, Grade 3 or 4 ICH n= (%) | p-value# | |
|---|---|---|---|
| Total Week 36 CUS | 526 (90.1) | 58 (9.9) | |
| Gestational Age* | |||
| 24 weeks | 117 (86.0) | 19 (14.0) | 0.9 |
| 25 weeks | 136 (88.9) | 17 (11.1) | |
| 26 weeks | 129 (91.5) | 12 (8.5) | |
| 27 weeks | 144 (93.5) | 10 (6.5) | |
| Infant Sex | |||
| Female | 257 (90.2) | 28 (9.8) | 0.93 |
| Male | 269 (90.0) | 30 (10.0) | |
| Prenatal Steroids | |||
| 0 Doses | 45 (84.9) | 8 (15.1) | 0.94 |
| 1 Dose | 103 (90.4) | 11 (9.6) | |
| 2 Doses | 325 (90.3) | 35 (9.7) | |
| 3+ Doses | 48 (96.0) | 2 (4.0) | |
| Weight <10th Percentile | |||
| No | 451 (90.2) | 49 (9.8) | 0.49 |
| Yes | 73 (90.1) | 8 (9.9) | |
| Treatment Group^ | |||
| Placebo | 256 (88.6) | 33 (11.4) | 0.14 |
| Epo | 270 (91.5) | 25 (8.5) | |
| Day 7–9 ICH$ | |||
| No ICH | 377 (97.7) | 9 (2.3) | <0.0001 |
| Grade 1 | 37 (97.4) | 1 (2.6) | |
| Unilateral Grade 2 | 37 (90.2) | 4 (9.8) | |
| Bilateral Grade 2 | 35 (68.6) | 16 (31.4) | |
| Grade 3 | 27 (56.3) | 21 (43.8) | |
| Grade 4 | 5 (50.0) | 5 (50.0) | |
| Serious Adverse Events | |||
| Severe Sepsis | |||
| No | 488 (92.8) | 38 (7.2) | 0.82 |
| Yes | 54 (93.1) | 4 (6.9) | |
| NEC | |||
| No | 499 (89.7) | 57 (10.3) | 0.63 |
| Yes | 54 (93.1) | 4 (6.9) | |
| PDA requiring surgery | |||
| No | 462 (90.9) | 46 (9.1) | 0.29 |
| Yes | 64 (84.2) | 12 (15.8) | |
| Severe BPD | |||
| No | 319 (89.9) | 36 (10.1) | 0.9 |
| Yes | 207 (904) | 22 (9.6) |
All models adjusted for treatment group, gestational age, and grade of ICH on day 7–9. P-values are derived from comparison to a baseline model only including group, gestational age, and day 7–9 ICH. In models with multiple levels, P-values are either for a linear trend across levels.
Gestational age model adjusted for treatment group and Day 7–9 ICH grade
Treatment group model adjusted for gestational age and Day 7–9 ICH
Day 7–9 ICH grade model adjusted for treatment group and gestational age
Bonferroni correction for 12 comparisons, adjusted significant p-value=0.05/10=0.005
Figure 3: Risk Factors for ICH.
Composite models for any ICH at day 1 (A), and bilateral grade 2, grade 3, or grade 4 ICH at day 7–9 (B) and week 36 (C) showing odds ratio OR) and 95% confidence interval (CI). Models included factors that were individually associated with ICH at each time point before adjustment for multiple comparisons.
Association of ICH with Death and Neurodevelopment
A total of 108 infants died; 30% (n=33) prior to the day 7–9 CUS, 44% (n=47) between the day 7–9 and 36-weeks CUSs, and 26% (n=28) after the 36 weeks CUS but prior to the follow-up examination at 2 years CA. Primary outcome data were available on >85% of included infants (Table 3). Approximately one third of infants with any ICH died or had severe NDI. Infants with a higher ICH grade were more likely to have a bilateral bleed and were more likely to die or have severe NDI. No individual grade of ICH on day 1 was significantly associated with the primary NDI outcome (Figure 4; available at www.jpeds.com), but the presence of any ICH on day 1 was associated with increased odds of death or severe NDI (OR 1.50; 95%CI 1.02–2.22). By comparison, ICH at day 7–9, 36 weeks PMA, and for the worst ICH grade identified across all three images, bilateral Grade 2, any Grade 3, and any Grade 4 ICH all individually increased the odds of death or severe NDI (Table 3 and Figure 4).
Figure 4. Neurodevelopmental and survival outcomes by ICH status on day 1, day 7–9, 36 weeks and by worst CUS.
Odds ratio and 95% confidence interval of death or severe disability in infants by grade of ICH at day 1 (A), day 7–9 (B), week 36 (C), or worst ICH grade across all three timepoints (C). Proportion with primary outcome is separated out as survival without severe neurodevelopmental impairment (NDI) in light blue, or severe NDI or death in darker shades of blue.
The hazard ratio (HR) of death among infants with bilateral Grade 2, any Grade 3, or any Grade 4 ICH at day 7–9 was increased compared with those with no ICH, Grade 1 ICH, or unilateral Grade 2 ICH (HR 1.90, 95% CI 1.13–3.20; Figure 5 [available at www.jpeds.com]). The increase in mortality risk was robust to sensitivity analyses, with a significantly increased mortality risk still present when those with any ICH on day 7–9 were compared with those with no ICH (HR 1.76, 95% CI 1.08–2.88). Among the 666 infants who survived and were evaluated at 2 years CA, 11% (n=75) were diagnosed with severe NDI.
Figure 5: Kaplan Meier survival curves by Day 7–9 CUS ICH status.
In infants who survived to day 7–9 CUS, those who had no ICH or a grade 1 or unilateral grade 2 ICH had a significantly increased chance of survival compared to those with a bilateral grade 2, grade 3, or grade 4 ICH.
Bayley Subscales at 2 years
The median (IQR) BSID-III Cognitive, Motor, and Language scores in infants who had all three CUS scans with no evidence of ICH were 90 (85–100), 94 (85–100), and 89 (78.5–100), respectively. On day 1 CUS, Grade 3 and 4 ICH were associated with a decreased BSID-III Cognitive score, and Grade 4 ICH was also associated with a decreased Motor score (Table 8 and Figure 6, A [available at www.jpeds.com]). A bilateral Grade 2 ICH on day 7–9 was associated with a significant reduction in Motor score, Grade 3 ICH was associated with a significant reduction in both Cognitive and Motor scores, and Grade 4 ICH was associated with a significant reduction in all three subscale scores (Table 8 and Figure 6, B [available at www.jpeds.com]). At 36 weeks PMA, bilateral Grade 2 ICH was associated with significantly lower Cognitive and Language scores, Grade 3 ICH was associated with significant reduction in all subscales, and Grade 4 ICH was associated with lower Motor and Language scores (Table 8 and Figure 6, C [available at www.jpeds.com]). Of note, Grade 1 ICH was associated with a significantly lower Cognitive score, however this evaluation was limited by a small number of infants in this group. In those who received all three scans, having a worst ICH of bilateral Grade 2 was associated with significantly reduced Cognitive score, Grade 3 with a lower Cognitive and Motor Score, and Grade 4 with lower scores on all three subscales (Table 8 and Figure 6, D [available at www.jpeds.com]).
Table 8: BSID-III Subscale scores at 2 years by ICH grade.
Estimated effect of each ICH Grade at day 1, day 7–9, and week 36 on BSID-III Cognitive, Motor, and Language subscales at 2 years of age. Worst ICH grade was determined in infants who had CUS imaging at all three time points.
| BSID-III Subscale | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Cognitive | Motor | Language | |||||||
| N= | Coef (95% CI)* | p-value | N= | Coef (95% CI)* | p-value | N= | Coef (95% CI)* | p-value | |
| Day 1 CUS | 654 | 644 | 640 | ||||||
| No ICH | 515 | Reference | 507 | Reference | 502 | Reference | |||
| Grade 1 | 38 | 2.04 (−2.83, 6.90) | 0.4 | 38 | 0.60 (−6.04, 7.24) | 0.9 | 38 | 3.08 (−3.33, 9.49) | 0.3 |
| Grade 2 | 68 | −1.16 (−5.24, 2.92) | 0.6 | 68 | −2.81 (−7.53, 1.91) | 0.2 | 68 | −0.16 (−4.70, 4.37) | 0.9 |
| Grade 3 | 28 | −5.90 (−11.1, −0.67) | 0.03 | 27 | −2.31 (−8.67, 4.04) | 0.5 | 28 | −4.27 (−9.61, 1.06) | 0.1 |
| Grade 4 | 5 | −15.0 (−20.2, −9.74) | <0.0001 | 4 | −19.8 (−25.8, −13.8) | <0.0001 | 4 | −5.51 (−23.9, 1.06) | 0.6 |
| Day 7–9 CUS | 637 | 628 | 624 | ||||||
| No ICH | 430 | Reference | 423 | Reference | 419 | Reference | |||
| Grade 1 | 39 | −0.81 (−5.54, 3.92) | 0.7 | 39 | 1.30 (−4.27, 6.87) | 0.6 | 39 | −0.07 (−5.44, 5.30) | 1.0 |
| Grade 2 Unilateral | 48 | −2.09 (−7.14, 2.97) | 0.4 | 48 | −1.44 (−6.68, 3.81) | 0.6 | 48 | −0.07 (−5.51, 5.37) | 1.0 |
| Grade 2 Bilateral | 57 | −3.44 (−7.48, 0.61) | 0.1 | 57 | −4.92 (−9.59, −0.25) | 0.04 | 56 | −1.30 (−5.88, 3.27) | 0.6 |
| Grade 3 | 52 | −9.29 (−13.5, −5.04) | <0.0001 | 51 | −9.52 (−14.8, −4.21) | <0.0001 | 52 | −4.46 (−9.37, 0.46) | 0.08 |
| Grade 4 | 11 | −14.8 (−23.4, −6.34) | 0.0006 | 10 | −24.8 (−35.2, −14.3) | <0.0001 | 10 | −17.6 (−26.6, −8.57) | 0.0001 |
| Week 36 CUS | 472 | 464 | 463 | ||||||
| No ICH | 407 | Reference | 399 | Reference | 398 | Reference | |||
| Grade 1 | 9 | −8.39 (−16.1, −0.67) | 0.03 a | 9 | −6.35 (−16.4, 3.74) | 0.2 | 9 | −0.76 (−9.04, 7.52) | 0.9 |
| Grade 2 Unilateral | 12 | 5.47 (−1.16, 12.1) | 0.1 | 12 | −0.42 (−8.07, 7.23) | 0.9 | 12 | 2.85 (−5.03, 10.7) | 0.5 |
| Grade 2 Bilateral | 24 | −6.98 (−12.8, −1.12) | 0.02 | 24 | −6.38 (−12.9, 0.157) | 0.06 | 24 | −6.96 (−13.2, −0.69) | 0.03 |
| Grade 3 | 16 | −11.8 (−17.2, −6.37) | <0.0001 | 16 | −20.8 (−27.2, −14.4) | <0.0001 | 16 | −9.69 (−17.2, −2.20) | 0.01 |
| Grade 4 | 4 | −1.94 (−5.92, 2.04) | 0.3 | 4 | −16.9 (−33.4, −0.36) | 0.04 | 4 | −11.8 (−16.7, −6.84) | <0.0001 |
| Worst ICH | 462 | 455 | 453 | ||||||
| No ICH | 288 | Reference | 283 | Reference | 280 | Reference | |||
| Grade 1 | 37 | −0.74 (−5.37, 3.79) | 0.7 | 37 | 1.26 (−4.56, 7.08) | 0.7 | 37 | 2.98 (−3.04, 9.00) | 0.3 |
| Grade 2 Unilateral | 38 | −0.27 (−5.11,4.57) | 0.9 | 38 | −0.79 (−6.35, 4.77) | 0.8 | 38 | 0.15 (−5.62, 5.92) | 1.0 |
| Grade 2 Bilateral | 48 | −6.35 (−10.3, −2.45) | 0.001 | 48 | −6.36 (−11.1, −1.6) | 0.009 | 48 | −4.47 (−9.25, 0.31) | 0.07 |
| Grade 3 | 39 | −8.82 (−13.3, −4.32) | 0.0001 | 38 | −9.15 (−15.1, −3.23) | 0.002 | 39 | −5.23 (−10.7, 0.26) | 0.06 |
| Grade 4 | 12 | −13.8 (−20.7, −6.92) | <0.0001 | 11 | −25.9 (−34.9, −16.9) | <0.0001 | 11 | −15.7 (−25.4, −6.06) | 0.001 |
Coefficient is the number of points gained/lost on the BSID-III subscale associated with that grade of ICH.
Analysis limited by small sample size
Figure 6: Bayley-III scores by ICH Grade on Day 1, Day 7–9, 36 weeks and Worst ICH.
BSID-III Cognitive (left), Motor (middle), and Language (right) scores at 2 years of age by day 1 (A), day 7–9 (B), week 36 (C), and worst across all three timepoints (D) ICH status.
Cerebral Palsy
Documented CP status was available in 75% (n=662) of included infants, with 8% of those (n=51) having mild CP, and 5% (n=31) having moderate-to-severe CP. Similarly, in the n=469 infants who had all three CUS scans and documented CP status, mild CP was diagnosed in 7% (n=35), and moderate-to-severe CP in 5% (n=23). In infants who never had a bleed documented across the three scans (n=292), mild CP was diagnosed in 6% (n=17), and moderate-to-severe CP was diagnosed in 2% (n=6). For the worst grade of ICH identified across all three time points, mild CP was diagnosed in 3% (n=1) of Grade 1, 3% (n=1) of unilateral Grade 2, 12% (n=6) of bilateral Grade 2, 18% (n=7) of Grade 3, and 21% (n=3) of Grade 4 ICH. Moderate-to-severe CP was diagnosed in 3% (n=1), 5% (n=2), 4% (n=2), 15% (n=6) and, 43% (n=6) of those with a worst ICH of Grade 1, unilateral Grade 2, bilateral Grade 2, Grade 3, or Grade 4, respectively.
Secondary Intracranial Findings
Of the 108 infants who died, 13% (n=14) had PVL, 19% (n=20) had ventriculomegaly, and 6% (n=7) had cerebellar hemorrhage. In the 51 infants diagnosed with mild CP, 16% (n=8) had PVL, 31% (n=16) had ventriculomegaly, and <2% (n=1) had cerebellar hemorrhage. In the 31 infants diagnosed with moderate-to-severe CP, 45% (n=14) had PVL, 61% (n=19) had ventriculomegaly, and 16% (n=5) had cerebellar hemorrhage. In the 75 children who survived with severe NDI, 28% (n=21) had PVL, 49% (n=37) had ventriculomegaly, and 11% (n=8) had cerebellar hemorrhage.
ICH Progression
After adjusting for day 7–9 ICH grade, infants who had a new or worsening ICH on day 7–9 relative to day 1 (n=162) had a higher of risk of death or severe NDI compared with infants who had no ICH or an ICH that had been stable or resolved since day 1 (OR 1.87; 95% CI 1.04–3.38). At 36 weeks PMA, after adjusting for ICH grade, infants who had a new or worsening ICH relative to their day 7–9 ICH also had a similar odds of death or severe NDI compared with infants who had no ICH or an ICH that had been stable or resolved since day 7–9 (OR 1.41; 95% CI 0.42–4.67).
ICH Resolution
We considered the subgroup of infants who had no ICH at 36 weeks and then classified them according to whether they had a history of ICH on day 7–9. Of the 499 infants who had no ICH identified at 36-week PMA, 25% (n=125) had documented ICH on day 7–9 (n=30 Grade 1, n=65 Grade 2, n=25 Grade 3, n=5 Grade 4). Although all of these infants had no identified ICH on week 36, those with a history of earlier ICH remained at increased risk of death or severe NDI. Compared with infants who had no ICH at either time point, the risk for death or severe NDI in infants who had no ICH at 36 weeks PMA after documented ICH at day 7–9, was only significant for those who had bilateral Grade 2 and Grade 4 ICH at day 7–9 ICH. The ORs were 1.64 (95% CI 0.60–4.49), 1.66 (95% CI 0.50–5.54), 2.85 (95% CI 1.17–6.95), 1.63 (95% CI 0.54–4.97) and 15.6 (95% CI 2.69–89.9) for Grade 1, unilateral Grade 2, bilateral Grade 2, Grade 3, and Grade 4 ICH, respectively.
DISCUSSION
We describe the impact of the timing, severity, and progression of ICH on neurodevelopmental outcomes at 2 years CA. One-third of infants were diagnosed with an ICH during their NICU stay, with the overall rate of ICH in 24 weeks’ GA infants twice that seen in 27 week’s GA infants. Although this is consistent with Stoll et al who demonstrated that less mature infants remain at the highest risk of bleeding, it is notable that in our cohort GA was not a risk factor for bleeds present on the day 1 HUS, but was a risk factor for later bleeds.(1) This suggests that the physiology and risk factors for early ICH in the immediate postnatal period after birth differ from those of ICH seen later in life. We did observe a bimodal distribution of ICH with half of ICH occurring between 0–6 hours of life which is consistent with prior literature.(27)
A wide range of risk factors from maternal and infant health to socio-economic status and even genetic predisposition have been implicated in development of ICH.(28–35) Perinatal factors previously demonstrated to modify risk of bleeding include GA and prenatal steroids, while the benefits of magnesium sulfate, mode of delivery, and delayed cord clamping remain controversial.(36–38) Contrary to previous reports, maternal education as a proxy of socioeconomic status was not associated with increased risk of ICH in our cohort.(33, 39) Conflicting data exist regarding preventive interventions proposed to reduce the risk of ICH after birth, including prophylactic indomethacin therapy or implementation of an ICH Bundle (a set of guidelines for EP care during the first 72 hours of life including midline head positioning and decreased handling and phlebotomy).(40–47) Although limited by the variability in dosing and practice between sites, our prospective data found that infants who received indomethacin or were cared for at an institution with an ICH bundle showed no overall reduction in prevalence of ICH.
We found that the risks associated with development of ICH changed over time. Both multiple gestation and cesarean delivery were associated with reduced odds of ICH on day 1. Administration of 2 or more doses of prenatal steroids reduced the risk of ICH at day 7–9 CUS, whereas lower GA and prior ICH increased the risk of bleeding. We also found an association between prior ICH and risk of ICH at 36 weeks. We did not find any association between treatment group (Epo and placebo) and development of ICH, and although adjusted for in our outcome models, treatment group was never found to be a significant variable.
One unclear aspect of ICH is the implication of unilateral versus bilateral bleeds on neurodevelopmental outcomes. More than half of the ICHs in our cohort were bilateral, with the presence of bilateral bleeds increasing over time and with increasing ICH severity. Although the risk of death or NDI in infants with unilateral or bilateral Grade 3 or 4 ICH has been described in the literature, we also found a significantly increased risk in infants with bilateral Grade 2 ICH.(48–52) The impact of Grade 2 ICH has been controversial: although Bolisetty et al and Patra et al both reported increased risk of NDI in infants with Grade 1 and Grade 2 ICH when compared with infants with no ICH, Payne et al and Reubsaet et al found no significant differences.(7, 8, 10, 53) These studies combined infants with Grade 1 with those who had Grade 2 ICH and did not evaluate unilateral compared with bilateral hemorrhages, thereby potentially masking the neurodevelopmental implications of bilateral Grade 2 ICH. When evaluated separately, we found that bilateral Grade 2 ICH had significantly different effects on death, NDI, and BSID-III scores than any Grade 1 or unilateral Grade 2 ICH and should be assessed independently. This may help explain the conflicting results regarding neurodevelopmental outcomes of infants with Grade 1 or 2 ICH found in the literature.
Our findings suggest each grade of ICH severity should be considered separately rather than considering both Grades 1 and 2 “mild” and Grades 3 and 4 as “severe,” particularly when predicting outcomes or counseling parents. We found substantial similarity in the rate and risk of death, severe NDI, and BSID-III scores between infants with bilateral Grade 2 ICH and those with any Grade 3 ICH. We also found outcomes between Grade 3 and Grade 4 ICH to be quite different, with the risk of death or severe NDI for Grade 4 ICH up to 7 times greater than the risk for Grade 3 ICH. BSID-III scores at 2 years were also significantly worse in infants with Grade 4 ICH across all three subscales relative to all other ICH Grades.
Infants with NDI-associated ICH (bilateral Grade 2, any Grade 3, or any Grade 4 ICH) on the day 7–9 CUS had a 2-fold risk of death compared with infants with no, Grade 1, or unilateral Grade 2 ICH. Importantly, the presence of any grade ICH on day 1 of life did not carry similar predictive value, so using this early information for parental counseling should be avoided. In infants who survived, our results are consistent with prior studies demonstrating that while significant ICH may predict poor outcome, the absence of ICH does not ensure normal development in EP infants.(18) For those infants with NDI-associated ICH, median BSID-III scores, although significantly lower, did not meet criteria for severe NDI.
It is unknown how changes in ICH grade over time affect an infant’s risk of death or neurodevelopment. As only 28 bleeds on day 1 resolved by day 7–9, we were unable to determine if ICH resolution early in the clinical course was associated with improved neurodevelopmental outcomes. However, infants with new or worsening ICH on the day 7–9 or 36-week CUS had 2 and 1.5 times the risk of death or severe NDI, respectively. Interestingly, for bleeds diagnosed on day 7–9, the risk remained elevated despite resolution by 36 weeks PMA. This finding suggests that ICH identified at either time point increases the risk of poorer outcomes.
Based on our findings, the value of cranial imaging at each time point (or a combination of all three) changes over time. Imaging at day 1 provides limited insight into long-term neurodevelopmental outcomes, but it does allow identification of infants at risk for ICH progression during the first week of life. Day 7–9 imaging has the best predictive power for the composite outcome of death and severe NDI by ICH grade as well as for death independently. The utility of CUS at 36 weeks PMA is the ability to detect new or missed bleeds that are associated with increased risk of death or severe NDI. Finally, although limited in its ability to predict death as infants had to survive to 36 weeks PMA, the best predictor of neurodevelopmental outcome is identification of the worst ICH grade across all three time points.
There is no current consensus regarding the timing and frequency of neurosonography for EP infants. In 2002, the American Academy of Neurology published guidelines for CUS at 7–14 days with a follow up between 36- and 40-weeks’ gestation; however these recommendations were retracted in 2010.(54) Ibrahim et. al. recommends increased frequency of serial CUSs at 3–5 days, 10–14 days, 28–30 days, monthly if indicated, and at term-equivalent age to enhance hemorrhage detection.(55) In our study, we found three results regarding CUS timing noteworthy. First, although ICH on day 1 is predictive of future ICH, knowing the Grade of ICH on day 1 does not predict any of the outcomes evaluated. Second, we show that bilateral Grade 2, any Grade 3, and any Grade 4 ICH on day 7–9 is associated with increased risk of death or severe NDI, lower BSID-III scores at 2 years CA, and is independently predictive of death. Third, as is consistent with prior studies, 6% of the infants in our study had a new ICH diagnosis at 36 weeks PMA, and there is a strong link between term-adjusted findings and neurodevelopmental outcomes.(56) Based on these findings, we recommend CUS imaging on day 7–9 and at 36 weeks PMA for clinical decision-making, prognostication, parental counseling, and appropriate follow-up referrals upon discharge.
The limitations of this study primarily fall into two categories: those due to the imaging modality and those due to the number of infants available for analysis. Infant cranial ultrasound imaging limitations include poor image clarity with patient movement, variations in ultrasonographer technique, and a spectrum of available window sizes due to small birth weights. These limitations may explain why some higher-grade bleeds were documented on day 1 but not seen on day 7–9. However, all scans were centrally read by the same individual which provides a certain level of confidence in the results. If a bleed was present previously but not seen on a later scan, we feel it is reasonable to label it as resolved in the absence of other evidence. There are also limitations of the Papile grading system for subependymal and intraventricular hemorrhages, which was first described in 1978.(5) This grading system primarily focuses on the presence of blood in and around the ventricles and associated ventriculomegaly. Although it is simple and widely-used, the grading system is not linear, and the assignment of grades may include some subjectivity. The Papile system also does not take into account the laterality of the bleeds or any other associated cerebral abnormalities such as white matter damage seen as echogenic or echolucent lesions, which are the strongest predictors of death, NDI, and cerebral palsy.(57, 58) However, we feel that the limitations of the system were largely mitigated by the fact that all ultrasounds were read centrally, and that statistical analyses considered each grade of ICH as a separate factor rather than looking for trends across grades. Finally, although the posterior fossa was imaged using the mastoid fontanel, the gold standard for diagnosing cerebellar hemorrhages is by MRI. We might therefore have underdiagnosed cerebellar pathology, which has been independently associated with impaired neurodevelopmental outcomes.(59–61)
The low incidence of secondary intracranial findings outcomes also warrants discussion. For example, we were not able to assess the association between SIP and Grade 2–4 ICH as it did not reach 5% prevalence within our cohort between day 7–9 and 36 weeks PMA CUS. Although we were able to evaluate the association between worst grade of ICH and development of CP, we were not able to evaluate whether timing of bleed was an important variable due to the low number of infants at each time point who developed CP. Similarly, the low incidence of PVL or hydrocephalus limited our ability to determine whether these were predictive of death, severe NDI, CP, or BSID-III scores at 2 years CA. We were also unable to confirm the Neuroimaging and Neurodevelopmental Outcomes (NEURO) study findings of increased risk of death and NDI in infants with high-grade ICH and/or cerebellar anomalies identified by CUS or MRI at 35–42 weeks PMA as only 20 new cases were identified at 36 weeks PMA.(56) Finally, there was no statistical adjustment for clinical site. Adjusting for 19 separate sites would introduce the same number of independent variables into the models, which would particularly reduce our ability to detect differences in relatively rare outcomes such as death or higher grades of ICH.
Our findings may be valuable in clinical decision making and counselling families, as well as in identifying risk and protective factors for ICH at different time points that may warrant inclusion into protocols to reduce the incidence of ICH in this fragile population.
Supplementary Material
Table 6: Maternal and infant risk factors for day 7–9 ICH.
Individual maternal and infant risk factors for bilateral grade 2, grade 3 or grade 4 ICH on day 7–9 CUS, adjusted for day 1 ICH status, gestational age, and treatment group.
| No ICH or Grade 1 or Unilateral Grade 2 ICH n= (%) | Bilateral Grade 2, Grade 3 or 4 ICH n= (%) | p-value# | |
|---|---|---|---|
| Total Day 7–9 CUS | 661 (79.9) | 166 (20.1) | |
| Maternal Factors | |||
| Hypertension | |||
| No | 523 (79.1) | 138 (20.9) | 0.28 |
| Yes | 138 (83.1) | 28 (16.9) | |
| Prenatal Care | |||
| No | 16 (61.5) | 10 (38.5) | 0.017 |
| Yes | 636 (80.8) | 151 (19.2) | |
| Unknown | 9 (64.3) | 5 (35.7) | |
| Prolonged Rupture of Membranes | |||
| No | 480 (78.8) | 129 (21.2) | 0.25 |
| Yes | 181 (83.0) | 37 (17.0) | |
| Multiple Gestation | |||
| No | 490 (79.9) | 123 (20.1) | 0.15 |
| Yes | 171 (79.9) | 43 (20.1) | |
| Preterm Labor | |||
| No | 257 (82.1) | 56 (17.9) | 0.39 |
| Yes | 404 (78.6) | 110 (21.4) | |
| Chorioamnionitis | |||
| No | 577 (79.9) | 145 (20.1) | 0.6 |
| Yes | 84 (80.0) | 21 (20.0) | |
| Prenatal Antibiotics | |||
| No | 420 (80.0) | 105 (20.0) | 0.82 |
| Yes | 241 (79.8) | 61 (20.2) | |
| Prenatal Steroids | |||
| 0 Doses | 56 (66.7) | 28 (33.3) | |
| 1 Dose | 120 (75.0) | 40 (25.0) | <0.0001 |
| 2 Doses | 418 (89.4) | 89 (17.6) | |
| 3+ Doses | 62 (91.2) | 6 (8.8) | |
| Prenatal Magnesium | |||
| No | 102 (76.1) | 32 (23.9) | 0.34 |
| Yes | 535 (80.9) | 126 (19.1) | |
| Unknown | 23 (82.1) | 5 (17.9) | |
| Infant Factors | |||
| Gestational Age* | |||
| 24 weeks | 148 (74.4) | 51 (25.6) | |
| 25 weeks | 159 (74.6) | 54 (25.4) | 0.007 |
| 26 weeks | 166 (83.4) | 33 (16.6) | |
| 27 weeks | 188 (87.0) | 28 (13.0) | |
| Sex | |||
| Female | 321 (80.7) | 77 (19.3) | 0.79 |
| Male | 340 (79.3) | 89 (20.7) | |
| Weight <10th Percentile | |||
| No | 553 (78.8) | 149 (21.2) | 0.19 |
| Yes | 105 (86.1) | 17 (13.9) | |
| Delayed Cord Clamping | |||
| No | 246 (78.1) | 69 (21.9) | 0.39 |
| Yes | 238 (82.6) | 50 (17.4) | |
| NA | 177 (79.0) | 47 (21.0) | |
| Intubation and/or Chest Compressions in Delivery Room | |||
| No | 144 (89.4) | 17 (10.6) | 0.029 |
| Yes | 517 (77.6) | 149 (22.4) | |
| Apgar <5 at 5 min | |||
| No | 549 (82.3) | 118 (17.7) | 0.075 |
| Yes | 109 (69.4) | 48 (30.6) | |
| General Appearance | |||
| Well | 385 (81.4) | 88 (18.6) | 0.29 |
| Sick | 273 (78.0) | 77 (22.0) | |
| Postnatal Factors | |||
| Treatment Group^ | |||
| Placebo | 325 (80.0) | 81 (20.0) | 0.6 |
| Epo | 336 (79.8) | 85 (20.2) | |
| ICH | |||
| No | 398 (81.6) | 90 (18.4) | 0.36 |
| Yes | 263 (77.8) | 75 (22.2) | |
| Day 1 ICH$ | |||
| No ICH | 579 (89.2) | 70 (10.8) | <0.0001 |
| Grade 1 | 38 (74.5) | 13 (25.5) | |
| Grade 2 | 40 (47.6) | 44 (52.4) | |
| Grade 3 | 3 (8.1) | 34 (91.9) | |
| Grade 4 | 1 (16.7) | 5 (83.3) | |
| Indomethacin | |||
| No | 258 (80.1) | 64 (19.9) | 0.71 |
| Yes | 403 (79.8) | 102 (20.2) | |
| Antibiotics | |||
| No | 185 (84.5) | 34 (15.5) | 0.23 |
| Yes | 476 (78.3) | 132 (21.7) | |
| Benzodiazepines | |||
| No | 486 (83.1) | 99 (16.9) | 0.012 |
| Yes | 175 (72.3) | 67 (27.7) | |
| Opioids | |||
| No | 294 (83.3) | 59 (16.7) | 0.074 |
| Yes | 367 (77.4) | 107 (22.6) | |
| Steroids | |||
| No | 604 (81.4) | 138 (18.6) | 0.005 |
| Yes | 57 (67.1) | 28 (32.9) | |
| Vasopressors | |||
| No | 561 (82.9) | 116 (17.1) | 0.0065 |
| Yes | 100 (66.7) | 50 (33.3) | |
| Spontaneous Intestinal Perforation | |||
| No | 645 (80.4) | 157 (19.6) | 0.31 |
| Yes | 16 (64.0) | 9 (36.0) | |
| Culture Positive Sepsis | |||
| No | 624 (80.2) | 154 (19.8) | 0.97 |
| Yes | 37 (75.5) | 12 (24.5) | |
All models adjusted for treatment group, gestational age, and grade of ICH on day 1. P-values are derived from comparison to a baseline model only including group, gestational age, and day 1 ICH. In models with multiple levels, P-values are either for levels as factors (with individual P-values for each level) or for a linear trend across levels (single P-value).
Gestational age model adjusted for treatment group and day 1 ICH grade
Day 1 ICH grade model adjusted for treatment group and gestational age
Treatment group model adjusted for gestational age and Day 7–9 ICH
Bonferroni correction for 32 comparisons, adjusted significant p-value=0.05/27=0.0019
ACKNOWLEDGEMENTS
We thank all study participants and their parents, who made the PENUT study possible. A list of additional members of the PENUT consortium is available at www.ipeds.com (Appendix).
Supported by the National Institute of Neurological Disorders and Stroke U01NS077953 and U01NS077955. The authors declare no conflicts of interest.
Abbreviations:
- BSID-III
Bayley Scales of Infant and Toddler Development, 3rd Edition
- CA
corrected age
- CI
confidence interval
- CP
cerebral palsy
- CUS
cranial ultrasound
- ELGAN
extremely low gestational age neonate
- Epo
erythropoietin
- GA
gestational age
- GEE
generalized estimating equations
- GMFCS
Gross Motor Function Classification System
- HR
hazard ratio
- ICH
intracranial hemorrhage
- MRI
magnetic resonance imaging
- NDI
neurodevelopmental impairment
- NEC
necrotizing enterocolitis
- OR
odds ratio
- PENUT
Preterm Erythropoietin Neuroprotection
- PMA
postmenstrual age
- PROM
prolonged rupture of membranes
- PVL
periventricular leukomalacia
Footnotes
Data Sharing: Preterm Erythropoietin Neuroprotection (PENUT) Trial; NCT #01378273; De-identified individual participant data will be made available through the NINDS Data Archive: https://www.ninds.nih.gov/Current-Research/Research-Funded-NINDS/Clinical-Research/Archived-Clinical-Research-Datasets. The data will be de-identified, and a limited access data set will be available by the end of May 2021 through a request form on that page. Data dictionaries, in addition to study protocol, the statistical analysis plan, and the informed consent form will be included. The data will be made available upon publication of all PENUT Trial related manuscripts to researchers who provide a methodologically sound proposal for use in achieving the goals of the approved proposal.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Stoll BJ, Hansen NI, Bell EF, Walsh MC, Carlo WA, Shankaran S, et al. Trends in Care Practices, Morbidity, and Mortality of Extremely Preterm Neonates, 1993–2012. JAMA. 2015;314(10):1039–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Volpe JJ. Brain injury in premature infants: a complex amalgam of destructive and developmental disturbances. Lancet Neurol. 2009;8(1):110–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.McAdams RM, McPherson RJ, Mayock DE, Juul SE. Outcomes of extremely low birth weight infants given early high-dose erythropoietin. J Perinatol. 2013;33(3):226–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mukerji A, Shah V, Shah PS. Periventricular/Intraventricular Hemorrhage and Neurodevelopmental Outcomes: A Meta-analysis. Pediatrics. 2015; 136(6):1132–43. [DOI] [PubMed] [Google Scholar]
- 5.Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr. 1978;92(4):529–34. [DOI] [PubMed] [Google Scholar]
- 6.Hintz SR, Vohr BR, Bann CM, Taylor HG, Das A, Gustafson KE, et al. Preterm Neuroimaging and School-Age Cognitive Outcomes. Pediatrics. 2018;142(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Payne AH, Hintz SR, Hibbs AM, Walsh MC, Vohr BR, Bann CM, et al. Neurodevelopmental outcomes of extremely low-gestational-age neonates with low-grade periventricular-intraventricular hemorrhage. JAMA Pediatr. 2013;167(5):451–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bolisetty S, Dhawan A, Abdel-Latif M, Bajuk B, Stack J, Lui K, et al. Intraventricular hemorrhage and neurodevelopmental outcomes in extreme preterm infants. Pediatrics. 2014;133(1):55–62. [DOI] [PubMed] [Google Scholar]
- 9.O’Shea TM, Allred EN, Kuban KC, Hirtz D, Specter B, Durfee S, et al. Intraventricular hemorrhage and developmental outcomes at 24 months of age in extremely preterm infants. J Child Neurol. 2012;27(1):22–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Patra K, Wilson-Costello D, Taylor HG, Mercuri-Minich N, Hack M. Grades I-II intraventricular hemorrhage in extremely low birth weight infants: effects on neurodevelopment. J Pediatr. 2006;149(2):169–73. [DOI] [PubMed] [Google Scholar]
- 11.Perlman JM, Volpe JJ. Intraventricular hemorrhage in extremely small premature infants. Am J Dis Child. 1986;140(11):1122–4. [DOI] [PubMed] [Google Scholar]
- 12.Funato M, Tamai H, Noma K, Kurita T, Kajimoto Y, Yoshioka Y, et al. Clinical events in association with timing of intraventricular hemorrhage in preterm infants. J Pediatr. 1992;121(4):614–9. [DOI] [PubMed] [Google Scholar]
- 13.Ment LR, Oh W, Ehrenkranz RA, Allan WC. Antenatal exposure to steroids and the additional prophylaxis effect of indomethacin to prevent intraventricular hemorrhage. Pediatrics. 1996;97(4):601. [PubMed] [Google Scholar]
- 14.Paul DA, Kepler J, Leef KH, Siscione A, Palmer C, Stefano JL. Effect of preeclampsia on mortality, intraventricular hemorrhage, and need for mechanical ventilation in very low-birth-weight infants. Am J Perinatol. 1998;15(6):381–6. [DOI] [PubMed] [Google Scholar]
- 15.Synnes AR, Chien LY, Peliowski A, Baboolal R, Lee SK, Network CN. Variations in intraventricular hemorrhage incidence rates among Canadian neonatal intensive care units. J Pediatr. 2001;138(4):525–31. [DOI] [PubMed] [Google Scholar]
- 16.Heuchan AM, Evans N, Henderson Smart DJ, Simpson JM. Perinatal risk factors for major intraventricular haemorrhage in the Australian and New Zealand Neonatal Network, 1995–97. Arch Dis Child Fetal Neonatal Ed. 2002;86(2):F86–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ment LR, Oh W, Philip AG, Ehrenkranz RA, Duncan CC, Allan W, et al. Risk factors for early intraventricular hemorrhage in low birth weight infants. J Pediatr. 1992;121(5 Pt 1):776–83. [DOI] [PubMed] [Google Scholar]
- 18.van Wezel-Meijler G, De Bruine FT, Steggerda SJ, Van den Berg-Huysmans A, Zeilemaker S, Leijser LM, et al. Ultrasound detection of white matter injury in very preterm neonates: practical implications. Dev Med Child Neurol. 2011;53 Suppl 4:29–34. [DOI] [PubMed] [Google Scholar]
- 19.Juul SE, Mayock DE, Comstock BA, Heagerty PJ. Neuroprotective potential of erythropoietin in neonates; design of a randomized trial. Maternal Health, Neonatology and Perinatology 2015;1:27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Juul SE, Comstock BA, Wadhawan R, Mayock DE, Courtney SE, Robinson T, et al. A Randomized Trial of Erythropoietin for Neuroprotection in Preterm Infants. N Engl J Med. 2020;382 233–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Dammann O, Leviton A. Duration of transient hyperechoic images of white matter in very low birth weight infants: a proposed classification. Dev Med Child Neurol. 1997;39:2–5. [PubMed] [Google Scholar]
- 22.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. Developmental medicine and child neurology. 1997;39(4):214–23. [DOI] [PubMed] [Google Scholar]
- 23.Palisano RJ, Rosenbaum P, Bartlett D, Livingston MH. Content validity of the expanded and revised Gross Motor Function Classification System. Dev Med Child Neurol. 2008;50(10):744–50. [DOI] [PubMed] [Google Scholar]
- 24.Zeger SL, Liang KY, Albert PS. Models for longitudinal data: a generalized estimating equation approach. Biometrics. 1988;44(4):1049–60. [PubMed] [Google Scholar]
- 25.Dafni U Landmark analysis at the 25-year landmark point. Circ Cardiovasc Qual Outcomes. 2011;4(3):363–71. [DOI] [PubMed] [Google Scholar]
- 26.R Development Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical computing; 2019. [Google Scholar]
- 27.Al-Abdi SY, Al-Aamri MA. A Systematic Review and Meta-analysis of the Timing of Early Intraventricular Hemorrhage in Preterm Neonates: Clinical and Research Implications. J Clin Neonatol. 2014;3(2):76–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Sarkar S, Bhagat I, Dechert R, Schumacher RE, Donn SM. Severe intraventricular hemorrhage in preterm infants: comparison of risk factors and short-term neonatal morbidities between grade 3 and grade 4 intraventricular hemorrhage. Am J Perinatol. 2009;26(6):419–24. [DOI] [PubMed] [Google Scholar]
- 29.Kaiser JR, Gauss CH, Pont MM, Williams DK. Hypercapnia during the first 3 days of life is associated with severe intraventricular hemorrhage in very low birth weight infants. J Perinatol. 2006;26(5):279–85. [DOI] [PubMed] [Google Scholar]
- 30.Mohamed MA, Aly H. Male gender is associated with intraventricular hemorrhage. Pediatrics. 2010;125(2):e333–9. [DOI] [PubMed] [Google Scholar]
- 31.Ment LR, Ådén U, Bauer CR, Bada HS, Carlo WA, Kaiser JR, et al. Genes and environment in neonatal intraventricular hemorrhage. Semin Perinatol. 2015;39(8):592–603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Bhandari V, Bizzarro MJ, Shetty A, Zhong X, Page GP, Zhang H, et al. Familial and genetic susceptibility to major neonatal morbidities in preterm twins. Pediatrics. 2006;117(6):1901–6. [DOI] [PubMed] [Google Scholar]
- 33.Shankaran S, Lin A, Maller-Kesselman J, Zhang H, O’Shea TM, Bada HS, et al. Maternal race, demography, and health care disparities impact risk for intraventricular hemorrhage in preterm neonates. J Pediatr. 2014;164(5):1005–11.e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Pathogenesis Ballabh P. and prevention of intraventricular hemorrhage. Clin Perinatol. 2014;41(1):47–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Kenet G, Kuperman AA, Strauss T, Brenner B. Neonatal IVH--mechanisms and management . Thromb Res. 2011; 127 Suppl 3:S120–2. [DOI] [PubMed] [Google Scholar]
- 36.Handley SC, Passarella M, Lee HC, Lorch SA. Incidence Trends and Risk Factor Variation in Severe Intraventricular Hemorrhage across a Population Based Cohort. J Pediatr. 2018;200:24–9.e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Garofalo M, Abenhaim HA. Early versus delayed cord clamping in term and preterm births: a review. J Obstet Gynaecol Can. 2012;34(6):525–31. [DOI] [PubMed] [Google Scholar]
- 38.Weintraub Z, Solovechick M, Reichman B, Rotschild A, Waisman D, Davkin O, et al. Effect of maternal tocolysis on the incidence of severe periventricular/intraventricular haemorrhage in very low birthweight infants. Arch Dis Child Fetal Neonatal Ed. 2001;85(1):F13–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Villamor-Martinez E, Fumagalli M, Mohammed Rahim O, Passera S, Cavallaro G, Degraeuwe P, et al. Chorioamnionitis Is a Risk Factor for Intraventricular Hemorrhage in Preterm Infants: A Systematic Review and Meta-Analysis. Front Physiol. 2018;9:1253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Fuchs H German experience in the management of ELGAN infants. Acta Biomed. 2015;86 Suppl 1:16–20. [PubMed] [Google Scholar]
- 41.Mehler K, Grimme J, Abele J, Huenseler C, Roth B, Kribs A. Outcome of extremely low gestational age newborns after introduction of a revised protocol to assist preterm infants in their transition to extrauterine life. Acta Paediatr. 2012;101(12):1232–9. [DOI] [PubMed] [Google Scholar]
- 42.de Bijl-Marcus K, Brouwer AJ, De Vries LS, Groenendaal F, Wezel-Meijler GV. Neonatal care bundles are associated with a reduction in the incidence of intraventricular haemorrhage in preterm infants: a multicentre cohort study. Arch Dis Child Fetal Neonatal Ed. 2019. [DOI] [PubMed] [Google Scholar]
- 43.Chiriboga N, Cortez J, Pena-Ariet A, Makker K, Smotherman C, Gautam S, et al. Successful implementation of an intracranial hemorrhage (ICH) bundle in reducing severe ICH: a quality improvement project. J Perinatol. 2019;39(1):143–51. [DOI] [PubMed] [Google Scholar]
- 44.Ment LR, Oh W, Ehrenkranz RA, Philip AG, Vohr B, Allan W, et al. Low-dose indomethacin and prevention of intraventricular hemorrhage: a multicenter randomized trial. Pediatrics. 1994;93(4):543–50. [PubMed] [Google Scholar]
- 45.Kalani M, Shariat M, Khalesi N, Farahani Z, Ahmadi S. A Comparison of Early Ibuprofen and Indomethacin Administration to Prevent Intraventricular Hemorrhage Among Preterm Infants. Acta Med Iran. 2016;54(12):788–92. [PubMed] [Google Scholar]
- 46.Gillam-Krakauer M, Slaughter JC, Cotton RB, Robinson BE, Reese J, Maitre NL. Outcomes in infants <29 weeks of gestation following single-dose prophylactic indomethacin. J Perinatol. 2020:1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Kumar P, Carroll KF, Prazad P, Raghavan A, Waruingi W, Wang H. Elevated supine midline head position for prevention of intraventricular hemorrhage in VLBW and ELBW infants: a retrospective multicenter study. J Perinatol. 2020. [DOI] [PubMed] [Google Scholar]
- 48.Davis AS, Hintz SR, Goldstein RF, Ambalavanan N, Bann CM, Stoll BJ, et al. Outcomes of extremely preterm infants following severe intracranial hemorrhage. J Perinatol. 2014;34(3):203–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Merhar SL, Tabangin ME, Meinzen-Derr J, Schibler KR. Grade and laterality of intraventricular haemorrhage to predict 18–22 month neurodevelopmental outcomes in extremely low birthweight infants. Acta Paediatr. 2012;101(4):414–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Calisici E, Eras Z, Oncel MY, Oguz SS, Gokce i K, Dilmen U. Neurodevelopmental outcomes of premature infants with severe intraventricular hemorrhage. J Matern Fetal Neonatal Med. 2015;28(17):2115–20. [DOI] [PubMed] [Google Scholar]
- 51.Cizmeci MN, de Vries LS, Ly LG, van Haastert IC, Groenendaal F, Kelly EN, et al. Periventricular Hemorrhagic Infarction in Very Preterm Infants: Characteristic Sonographic Findings and Association with Neurodevelopmental Outcome at Age 2 Years. J Pediatr. 2020;217:79–85.e1. [DOI] [PubMed] [Google Scholar]
- 52.McCauley KE, Carey EC, Weaver AL, Mara KC, Clark RH, Carey WA, et al. Survival of Ventilated Extremely Premature Neonates With Severe Intraventricular Hemorrhage. Pediatrics. 2021;147(4). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Reubsaet P, Brouwer AJ, van Haastert IC, Brouwer MJ, Koopman C, Groenendaal F, et al. The Impact of Low-Grade Germinal Matrix-Intraventricular Hemorrhage on Neurodevelopmental Outcome of Very Preterm Infants. Neonatology. 2017;112(3):203–10. [DOI] [PubMed] [Google Scholar]
- 54.Ment LR, Bada HS, Barnes P, Grant PE, Hirtz D, Papile LA, 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):1726–38. [DOI] [PubMed] [Google Scholar]
- 55.Ibrahim J, Mir I, Chalak L. Brain imaging in preterm infants <32 weeks gestation: a clinical review and algorithm for the use of cranial ultrasound and qualitative brain MRI. Pediatr Res. 2018;84(6):799–806. [DOI] [PubMed] [Google Scholar]
- 56.Hintz SR, Barnes PD, Bulas D, Slovis TL, Finer NN, Wrage LA, et al. Neuroimaging and neurodevelopmental outcome in extremely preterm infants. Pediatrics. 2015;135(1):e32–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Kuban KC, Allred EN, O’Shea TM, Paneth N, Pagano M, Dammann O, et al. 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]
- 58.Al-Mouqdad M, Al-Abdi S, Scott JN, Hurley A, Tang S, Creighton D, et al. A New IVH Scoring System Based on Laterality Enhances Prediction of Neurodevelopmental Outcomes at 3 Years Age in Premature Infants. Am J Perinatol. 2017;34(1):44–50. [DOI] [PubMed] [Google Scholar]
- 59.Boswinkel V, Steggerda SJ, Fumagalli M, Parodi A, Ramenghi LA, Groenendaal F, et al. The CHOPIn Study: a Multicenter Study on Cerebellar Hemorrhage and Outcome in Preterm Infants. Cerebellum. 2019;18(6):989–98. [DOI] [PubMed] [Google Scholar]
- 60.Limperopoulos C, Bassan H, Gauvreau K, Robertson RL, Sullivan NR, Benson CB, et al. Does cerebellar injury in premature infants contribute to the high prevalence of long-term cognitive, learning, and behavioral disability in survivors? Pediatrics. 2007;120(3):584–93. [DOI] [PubMed] [Google Scholar]
- 61.Zayek MM, Benjamin JT, Maertens P, Trimm RF, Lal CV, Eyal FG. Cerebellar hemorrhage: a major morbidity in extremely preterm infants. J Perinatol. 2012;32(9):699–704. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.