Abstract
Purpose:
Intraventricular hemorrhage (IVH) of prematurity can lead to hydrocephalus, sometimes necessitating permanent cerebrospinal fluid (CSF) diversion. We sought to characterize the relationship between head circumference (HC) and ventricular size in IVH over time to evaluate the clinical utility of serial HC measurements as a metric in determining the need for CSF diversion.
Methods:
We included preterm infants with IVH born between January 2000 and May 2020. Three measures of ventricular size were obtained: Ventricular Index (VI) Evan’s Ratio (ER), and Frontal Occipital Head Ratio (FOHR). The Pearson correlations (r) between the initial (at birth) paired measurements of HC and ventricular size were reported. Multivariable longitudinal regression models were fit to examine the HC:ventricle size ratio, adjusting for the age of the infant, IVH grade (I/II vs. III/IV), need for CSF diversion, and sex.
Results:
639 patients with an average gestational age of 27.5 weeks were included. IVH Grade I/II & Grade III/IV patients had a positive correlation between initial HC and VI (r=0.47, p<0.001 and r=0.48, p<0.001, respectively). In our longitudinal models, patients with a low grade IVH (I/II) had an HC:VI ratio 0.52 higher than those with a high-grade IVH (p-value<0.001). Patients with low-grade IVH had an HC:ER ratio 12.94 higher than those with high-grade IVH (p-value<0.001). Patients with low-grade IVH had a HC:FOHR ratio 12.91 higher than those with high-grade IVH (p-value<0.001). Infants who did not require CSF diversion had an HC:VI ratio 0.47 higher than those who eventually did (p <0.001). Infants without CSF diversion had an HC:ER ratio 16.53 higher than those who received CSF diversion (p <0.001). Infants without CSF diversion had an HC:FOHR ratio 15.45 higher than those received CSF diversion (95%CI (11.34, 19.56), p <0.001).
Conclusions:
There is a significant difference in the ratio of HC:VI, HC:ER, and HC:FOHR size between patients with high-grade IVH and low-grade IVH. Likewise, there is a significant difference in HC:VI, HC:ER, and HC:FOHR between those who did and did not have CSF diversion. The routine assessment of both head circumference and ventricle size by ultrasound are important clinical tools in infants with IVH of prematurity.
Keywords: Prematurity, head circumference, intraventricular hemorrhage, germinal matrix hemorrhage
Introduction
Intraventricular hemorrhage (IVH) is seen in 20–50% of low birth weight, preterm infants.1,2 Also known as a germinal matrix hemorrhage (GMH), this condition arises due to immature germinal matrix vessels that normally regress by 36 weeks of gestation, but instead rupture and bleed into the cerebral ventricles in preterm infants.1 Blood products in the ventricles and resultant inflammation can block cerebrospinal fluid (CSF) flow and impair absorption, leading to progressive ventricular dilation and hydrocephalus.3
Definitive treatment for post-hemorrhagic hydrocephalus is cerebrospinal fluid (CSF) diversion via ventriculoperitoneal shunting3 or endoscopic third ventriculostomy.4 Low-weight patients may have temporary CSF diversion with lumbar puncture, external ventricular drain, ventricular access/reservoir, or a ventriculosubgaleal shunt, until they can safely undergo permanent diversion, typically when they weigh at least 2kg.3
Severity of IVH at birth is currently the best predictor of future development of hydrocephalus and requirement for permanent CSF diversion.5 IVH is graded from I-IV, with I indicating a simple GMH, II indicating IVH without ventricular dilation, III indicating IVH with acute ventricular dilation, and IV indicating hemorrhage with ventricular dilation and intraparenchymal bleeding.1,3 Incidence of permanent CSF diversion is <1% among infants with Grade I & II IVH, 9% among infants with Grade III IVH, and 12% among infants with Grade IV IVH.5
There are no standardized protocols for determining which infants will need permanent CSF diversion or the optimal timing for intervention.3,6,7 The decision to perform CSF diversion has traditionally been made with serial head circumference (HC) and serial head ultrasound (US),6 but newer studies suggest that ventricular size measurements such as the ventricular index (VI) can predict the need for early intervention and allow for improved outcomes in IVH8. Because HC can be easily assessed at the bedside and in resource-limited settings, serial HC measurement has been investigated as a possible surrogate for serial imaging. Prior studies have found that changes in HC do not correlate with changes in ventricular size as measured by Evan’s Ratio (ER).6 However, the relationship between HC and other measures of ventricular volume, including the frontal occipital head ratio (FOHR) and ventricular index (VI), has not yet been investigated. Further, the correlation of serial HC with ventricular size over time has not been previously studied.
This study aims to characterize the relationship, or lack thereof, between HC and ventricular size over time and evaluate the clinical utility of serial HC measurements as a metric in determining the need for CSF diversion.
Methods
Patient Inclusion and Data Collection
This study received approval from the Duke University Institutional Review Board (Pro00105801). The initial patient cohort was selected by searching the Duke University Health System electronic medical record for all patients born prior to 37 weeks with a diagnosis of GMH or IVH between January 1, 2000 and May 1, 2020. Patients were identified via the following International Classification of Disease (ICD) 9 & 10 codes: 765.09, 765.10, 765.19, 765.20, 772.10, 772.11, 772.12, 772.13, 772.14, P07.30, P52.0, P52.1, P52.21, P52.22, P52.3, and I61.5. These codes are detailed in Supplemental Table 1.
We conducted a retrospective chart review and collected GA at birth, race/ethnicity, birth weight, IVH grade, weekly HC, and all available ER, FOHR, and VI measurements from ultrasound, CT, or MRI. Data were collected from birth until discharge from the hospital, 40 weeks after birth, or until permanent CSF diversion was performed, whichever came first. For patients discharged prior to CSF diversion, medical records were searched until May 1, 2022 to check for later CSF diversion. After preliminary chart review, we excluded patients without a confirmed IVH or GMH diagnosis, those born at or after 37 weeks, those without measurements of head circumference before CSF diversion, and those without any radiologic imaging of the ventricles.
Ventricular Size Measurements
Ventricular size measurements were collected from head ultrasounds, CT, and MRI performed during the patients’ hospital course. Evan’s Ratio (ER) was measured as the ratio of the maximum diameter of the frontal ventricular horns to the maximum diameter of the inner skull on an axial view.9 The frontal-occipital head ratio (FOHR) was measured as the sum of the frontal and occipital ventricular horns divided by twice the maximum diameter of the inner skull on an axial view.10 The ventricular index (VI) was measured as the distance between the falx and lateral wall of the ventricle on a coronal brain view taken at the level of the Foramen of Monro.11 FOHR has shown higher inter-observer reliability compared with ER and high correlation with ventricular volumes,10 and VI has shown high reproducibility and high predictive value in early diagnosis of posthemorrhagic hydrocephalus.11
Statistical Analysis
Patient demographics were summarized with descriptive statistics. The frequency and percentage of patients who had temporary CSF diversion and permanent CSF diversion were cross-tabulated and the association was examined by the Chi-square test or Fisher exact test.
The Pearson correlations (r) between the initial paired measurements of head circumference and ventricular size measurements were reported with 95% CI. These correlations were performed on patient data stratified by IVH severity (low grade: I or II vs. high grade: III or IV). Preterm infant growth is often tracked using the standardized Fenton growth chart; thus, HC measurements were converted to Fenton z-scores using the PediTools calculator.12,13 The ratio of HC to VI, ER, and FOHR with means and 95% CI stratified by IVH grade over time were plotted.
To explore the relative change in HC and ventricle sizes over time, the ratios of HC to ventricle sizes at each postmenstrual week were estimated with means and 95% CIs stratified by IVH grade. All serial pairs of observations before CSF diversion were used after excluding patients with unspecified IVH grades. Longitudinal regression models were fit to examine the association between the ratios and the postmenstrual age of the infant after adjusting for the IVH grade (I/II vs. III/IV), need for CSF diversion, and sex using the generalized estimating equation method (GEE). GEE represents an extension of the generalized linear model to accommodate correlated data and it does not require the specification of the form of the distribution. The significance of all statistical tests were assessed at alpha = 0.05 and the unadjusted p-values were reported due to the exploratory nature of the study. Analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC) and R (version 4.1.3).
Results
Patient Cohort and Demographics
Patient demographics are listed in Table 1. Our cohort consisted of 639 patients, of whom 304 (47.6%) were female. Patients had an average birth GA of 27.5 weeks, average birth weight of 1100.4 g, and average birth HC of 25.2 cm. 340 (53.2%) patients had a Grade I IVH, 95 (14.9%) had a Grade II, 72 (11.3%) had a Grade III, and 107 (16.7%) had a Grade IV, while 25 (3.9%) had an unspecified IVH grade.
Table 1:
Patient Demographics
| Characteristic | Total (N=639) |
|---|---|
| Gestational age at Birth (Weeks) | |
| Mean (SD) | 27.5 (3.3) |
| Median | 27.0 |
| Q1, Q3 | 25.0, 30.0 |
| Range | (22.0–36.0) |
| Sex | |
| Female | 304 (47.6%) |
| Male | 335 (52.4%) |
| Ethnicity | |
| White | 229 (35.8%) |
| Black or African American | 303 (47.4%) |
| Asian | 12 (1.9%) |
| Native American or Alaska Native | 7 (1.1%) |
| Native Hawaiian or Pacific Islander | 1 (0.2%) |
| More than one race | 15 (2.3%) |
| Unknown or Not Reported | 72 (11.3%) |
| Hispanic or Latino | 62 (9.7%) |
| Not Hispanic or Latino | 524 (82.0%) |
| Birth weight (g) | |
| N | 634 |
| Mean (SD) | 1101.4 (561.5) |
| Median | 930.0 |
| Q1, Q3 | 730.0, 1310.0 |
| Range | (360.0–4840.0) |
| Head circumference at birth (cm) | |
| N | 586 |
| Mean (SD) | 25.2 (3.4) |
| Median | 24.5 |
| Q1, Q3 | 22.5, 27.5 |
| Range | (18.5–35.5) |
| Diagnosis with respect to prematurity (not mutually exclusive) | |
| None | 34 (5.3%) |
| Hydrocephalus | 73 (11.4%) |
| Intraventricular hemorrhage | 590 (92.3%) |
| Intra cranial hemorrhage | 26 (4.1%) |
| Other | 98 (15.3%) |
| Germinal Matrix Hemorrhage Grade | |
| Grade I | 340 (53.2%) |
| Grade II | 95 (14.9%) |
| Grade III | 72 (11.3%) |
| Grade IV | 107 (16.7%) |
| Not specified | 25 (3.9%) |
| Follow-Up Time from Birth (months) | |
| Mean (SD) | 3.8 (2.8) |
| Median | 3.2 |
| Q1, Q3 | 1.6, 5.5 |
| Range | (0.0–9.2) |
| Gestational age at Final Follow-Up (weeks) | |
| Mean (SD) | 44.1 (11.7) |
| Median | 41.0 |
| Q1, Q3 | 36.0, 49.0 |
| Range | (23.0–75.0) |
| Gestational Age at Permanent CSF Diversion (Weeks) | |
| N | 66 |
| Mean (SD) | 45.0 (8.3) |
| Median | 43.5 |
| Q1, Q3 | 38.0, 50.0 |
| Range | (31.0–65.0) |
A plot of HC by age for all patients is shown in Figure 1a. In general, our cohort had Fenton Z-score less than 0 which decreased over time (Figure 1b). On average, patients had a follow-up time of 3.8 months after birth and were on average 44.1 weeks GA at final follow-up.
Fig 1. Head Circumference by Postmenstrual Week of all Patients.
A) Head circumference in centimeters of all patients over time. The red trend line represents the average head circumference, with the 95% confidence interval shown as blue shading. B) Head circumference Fenton z-score of all patients over time. The red trend line represents the average head circumference, with the 95% confidence interval shown as blue shading.
Temporary vs Permanent CSF Diversion
Of 639 patients, 73 (11.4%) received temporary CSF diversion, while 66 (10.3%) received permanent CSF diversion. Permanent CSF diversion occurred at an average GA of 45.0 weeks. 55 of the 66 patients (83.3%) who received temporary CSF diversion eventually required permanent diversion. 3/340 (0.9%) Grade I, 6/95 (6.3%) Grade II, 19/72 (26.4%) Grade III, and 43/107 (40.2%) Grade IV required temporary diversion, while 2/340 (0.6%) Grade I, 3/95 (3.2%) Grade II, 21/72 (29.2%) Grade III, and 39/107 (36.5%) Grade IV patients required permanent diversion (Table 2). There was a statistically significant association between temporary CSF diversion and eventual permanent diversion (Gr I: p=0.02, Gr II: p<0.001, Gr III: p<0.001, Gr IV: p<0.001).
Table 2:
Contingency Association between Temporary and Permanent CSF Diversion by Germinal Matrix Hemorrhage Grade
| Overall | ||||
| Permanent CSF diversion/No | Permanent CSF diversion/Yes | Total | Chi-Square Test: p < 0.001 | |
| Temporary CSF diversion/No | 555 (96.9%) | 11 (16.7%) | 566 (88.6%) | |
| Temporary CSF diversion/Yes | 18 (3.1%) | 55 (83.3%) | 73 (11.4%) | |
| Total | 573 (89.7%) | 66 (10.3%) | 639 (100%) | |
| GMH Grade I | ||||
| Temporary CSF diversion/No | 336 (99.4%) | 1 (50%) | 337 (99.1%) | Fisher Exact Test: p = 0.02 |
| Temporary CSF diversion/Yes | 2 (0.6%) | 1 (50%) | 3 (0.9%) | |
| Total | 338 (99.4%) | 2 (0.6%) | 340 (100%) | |
| GMH Grade II | ||||
| Temporary CSF diversion/No | 89 (96.7%) | 0 (0%) | 89 (93.7%) | Fisher Exact Test: p < 0.001 |
| Temporary CSF diversion/Yes | 3 (3.3%) | 3 (100%) | 6 (6.3%) | |
| Total | 92 (96.8%) | 3 (3.2%) | 95 (100%) | |
| GMH Grade III | ||||
| Temporary CSF diversion/No | 46 (90.2%) | 7 (33.3%) | 53 (73.6%) | Fisher Exact Test: p < 0.001 |
| Temporary CSF diversion/Yes | 5 (9.8%) | 14 (66.7%) | 19 (26.4%) | |
| Total | 51 (70.8%) | 21 (29.2%) | 72 (100%) | |
| GMH Grade IV | ||||
| Temporary CSF diversion/No | 61 (89.7%) | 3 (7.7%) | 64 (59.8%) | Fisher Exact Test: p < 0.001 |
| Temporary CSF diversion/Yes | 7 (10.3%) | 36 (92.3%) | 43 (40.2%) | |
| Total | 68 (63.6%) | 39 (36.5%) | 107 (100%) | |
| GMH Grade Not Specified | ||||
| Temporary CSF diversion/No | 23 (95.8%) | 0 (0%) | 23 (92%) | Fisher Exact Test: p = 0.08 |
| Temporary CSF diversion/Yes | 1 (4.2%) | 1 (100%) | 2 (8%) | |
| Total | 24 (96%) | 1 (4%) | 25 (100%) | |
Correlations between Initial HC and Ventricular Size
338 patients had their first HC and ER measurements taken within one week of each other, while 336 met this criteria for HC and FOHR and 432 for HC and VI. When split into low-grade (Grade I-II) and high-grade (Grade III-IV) IVH, 311 low-grade patients had a positive correlation (r=0.47, p=02.2e-16) between HC and VI and 121 high-grade patients had a positive correlation (r=0.48, p=2.9e-8) between HC and VI (Figure 2). 217 low-grade patients had a a negative correlation (r=−0.25, p=1.5e-4) between initial HC and ER, while 121 high-grade patients had a positive correlation (r=0.24, p=0.0081) between HC and ER (Supplemental Figure 1a). 215 low-grade patients had a positive correlation (r=0.1, p=0.13) between HC and FOHR and 121 high-grade patients had a positive correlation (r=0.41, p=2.5e-6) between HC and FOHR (Supplemental Figure 1b).
Fig 2. Correlation between Initial Head Circumference and Ventricular Index.
Correlation between initial measurements of head circumference and ventricular index, in patients with low-grade (I/II) IVH (left panel) and high-grade (III/IV) IVH (right panel).
HC and Ventricular Measurements Over Time Prior to CSF Diversion
The weekly ratio of HC over either VI, ER, or FOHR were calculated and plotted over time. 432 patients (with a total of 1075 observations) with were included in the analysis of the VI:ER ratio, of which 311 (71.9%) had low-grade (I/II) IVH and 121 (28.1%) high-grade (III/IV) IVH. Figure 3 demonstrates HC:IV ratios over time by IVH grade and the need for CSF diversion.
Fig 3. Ratio of Head Circumference to Ventricular Index Over Time.
Average weekly head circumference to ventricular index ratio for infants with IVH. A) Ratio of Head circumference in cm to ventricular index stratified into low-grade (I/II) and high-grade (III/IV) IVH. B) Ratio of head circumference in cm to ventricular index by IVH grades. C) Ratio of Head circumference z-score to ventricular index stratified into low-grade (I/II) and high-grade (III/IV). D) Ratio of Head circumference in cm to ventricular index stratified by those who eventually received CSF diversion and those who did not.
In a regression analysis adjusting for the IVH grade, CSF diversion and sex (Table 3), there is no significant association between the HC:VI ratio and age (p=0.49). Patients with low grade IVH had an HC:VI ratio 0.52 higher than those with high-grade IVH (p<0.001). There is no statistically significant difference in the ratio between females and males (p=0.73). Infants who did not require CSF diversion had an HC:VI ratio 0.47 higher than those who eventually did (p <0.001).
Table 3:
Regression Analysis on the Ratio of Head Circumference to Ventricular Index
| Covariates | Estimate (95% CI) | p-value |
|---|---|---|
| Postmenstrual week (with 1-week increase) | −0.003 (−0.01, 0.005) | 0.49 |
| IVH grade I/II | 0.52 (0.43, 0.62) | <0.001 |
| IVH grade III/IV | Reference | Reference |
| CSF diversion-No | 0.47 (0.32, 0.63) | <0.001 |
| CSF diversion-Yes | Reference | Reference |
| Sex: Female | 0.01 (−0.07, 0.09) | 0.73 |
| Sex: Male | Reference | Reference |
For the ratio of HC:ER, there were 799 observations from 338 patients, of which 217 (64.2%) were low-grade and 121 (35.8%) high-grade (Supplemental Figure 2a). The HC z-score:ER ratio for Grade I/II and Grade III/IV is located in Supplemental Figure 2b. Multivariable regression of the HC:ER ratio and age showed increased by 2.30 weekly (p<0.001, Supplemental Table 2). Patients with low-grade IVH had an HC:ER ratio 12.94 higher than those with high-grade (p<0.001). There is no statistically significant difference in the ratio between females and males (p=0.92). Infants without CSF diversion had an HC:ER ratio 16.53 higher than those who received CSF diversion (p <0.001).
The ratio of HC:FOHR included 336 patients with 790 observations, with 215 low-grade patients (63.9%) and 121 high-grade patients (36.1%) (Supplemental Figure 3a). The HC z-score:FOHR are located in Supplemental Figure 3b. In the multivariable regression analysis for HC and FOHR (Supplemental Table 3), there is a significant association between the ratio of HC:FOHR and the postmenstrual weeks, increasing by 1.29 per week (p<0.001). Patients with low-grade IVH had a HC:FOHR ratio 12.91 higher than those with high-grade IVH (95% CI (9.46, 16.37), p-value<0.001). There is no statistically significant difference in the ratio between females and males (p=0.23). Infants without CSF diversion had an HC:FOHR ratio 15.45 higher than those received CSF diversion (95%CI (11.34, 19.56), p <0.001).
Discussion
We present a study analyzing associations between temporary & permanent CSF diversion & trends in HC & ventricular size measurements in patients with IVH of prematurity. Our results suggest that across all grades of IVH, temporary CSF diversion is associated with eventual permanent CSF diversion, with rates of both being higher in those with higher grades of IVH. We found that initially, HC and ventricle size are similar regardless of IVH grade. However, the ratio of HC to ventricle size varies dramatically over time. It is therefore paramount that practitioners do not rely solely on HC as a longitudinal marker for ventricle size.
Temporary vs Permanent CSF Diversion
As expected,14,15 patients who receive temporizing CSF diversion are likely to undergo permanent CSF diversion prior to discharge. Unlike prior studies that analyzed this association in high grade IVH,14,16,17 we confirmed that this holds true across all grades of IVH, even the most minor Grade I bleeds. Unfortunately, the need for both permanent and temporary CSF diversion is associated with worse neurodevelopmental outcomes and complications such as infection.16,18,19 Our limitation here lies in that we did not analyze the timing of temporary diversion and permanent CSF diversion. Multiple studies16,20 and randomized clinical trials have suggested that temporary diversion at older ages or larger ventricle sizes is predictive of the need for VP shunting and is associated with worse neurodevelopmental outcomes than early CSF diversion. The ELVIS21,22 trial demonstrated that intervention with temporary diversion at smaller ventricle sizes (early intervention) was associated with better (but statistically insignificant) odds of death or cerebral palsy and better cognitive outcomes even if the infant required VP shunt placement. Meanwhile, the DRIFT23,24 trial suggested that even more aggressive intervention with intraventricular fibrinolysis and blood drainage was associated with better long-term cognitive outcomes in IVH. Our results suggest that even infants with a low-grade IVH who require temporary diversion should be monitored carefully and families informed that a permanent diversion surgery is likely necessary, and evidence from the literature should strongly guide physicians to consider these interventions earlier if safe to do so.
Correlations between Initial HC and Ventricular Size
The Pearson correlations made between HC and ventricles show correlation between initial measurements and the need for CSF diversion in IVH. These results do not provide clinicians with a useful tool to predict the need for diversion in a neonate with IVH. If early HC and FOHR are both high, these correlations suggest that the patient would fall into the high-grade IVH group or more likely to receive CSF diversion, potentially prompting closer follow-up with serial head imaging and head circumference measurements. However, if early HC or ventricular measures are low, the severity of IVH cannot as easily be predicted and an astute clinician would continue to follow with serial head imaging and HC measurements anyway.
Ratios of HC and Ventricular Measurements Over Time
We demonstrated through longitudinal analysis of HC and ventricle size ratios that these measures taken together show a significant difference between those with low-grade IVH and those with high-grade IVH. Patients with high-grade IVH on average had an HC:VI ratio 0.52 lower, an HC:ER 12.94 lower, and an HC:FOHR 12.91 lower than low-grade IVH patients. Since HC is expected to grow over time in any infant, an infant with less severe bleeds and thus less dilated ventricles would have larger HC:ventricle measures over time while those with high-grade bleeds and larger ventricles would continue to have lower ratios.
Of the three ratios, HC:VI was the most consistent across time, with low-grade IVH infants consistently having a ratio of above 2, while those with high-grade IVH had ratios below 2. As the regression analysis shows, there is no significant week-by-week change in the HC:VI ratio, while there is a significant difference in this ratio between the IVH severity groups and between those who required CSF diversion and those who didn’t. While the regression did not show a significant week-to-week change, the plots in Figure 3 still demonstrate some changes over time, with Figures 3a, 3b, and 3d in particular highlighting a potential decrease in HC:VI between weeks 25 and 30 in high grade IVH and IVH requiring CSF diversion. HC:VI could be useful in differentiating between high and low-grade bleeds at later postmenstrual weeks, but is not as consistent at earlier ages. Further prospective study will be needed to confirm these findings and demonstrate statistical significance.
When we standardized all HC measurements to the Fenton z-score25, we observed a pattern across all 3 HC:ventricular size ratios. In both low-grade and high-grade patients, the ratio of HC z-score:ventricle size consistently decreased from birth to about 30–31 postmenstrual weeks in age (Figure 3c, Supplemental Figures 2 & 3). This pattern suggests that from birth to 31 weeks, the ventricle size increases rapidly, leading to ventriculomegaly out of proportion to HC. We additionally see in the regression models for ER and FOHR that there is a significant week-to-week increase in the HC:ventricular size ratio, whereas we would expect the change by weeks to be close to 0 if HC and ventricular size changed proportionally to each other, thus maintaining a proportional ratio. This leads us to conclude that HC is not a good longitudinal metric of ventricle size. While HC might increase over time in an infant with IVH, it underestimates the level of ventriculomegaly and might lead providers to intervene later than necessary if HC is used as the sole metric for hydrocephalus progression.
In regression analysis, it is unclear why HC:VI (Table 3) did not show a significant relationship over time whereas both HC:ER (Supplemental Table 2) and HC:FOHR (Supplemental Table 3) did. Future work will compare all three measurement methods to further assess validity. Notably, in all three ratios, week 29 appears to be an inflection point in which Grade I/II and Grade III/IV groups begin to have more of a parallel trend. This may be a reflection of the degree of parenchymal injury in higher grade IVH prior to brain myelination.
Animal models have shown that oxidative stress on white matter in the developing brain from IVH leads to ventriculomegaly.26 Oligodendrocyte precursor cells generally grow differentiate around the 23–32 week time period, meaning that the white matter is particularly sensitive to inflammatory and oxidative stress from the accumulation of blood products27,28 In preterm infants in particular, MRI studies have demonstrated that periventricular myelination in sites such as the internal capsule and corona radiata is only visible on imaging around 36 weeks, about 4 weeks after they are visible on histological slides.29 We can extrapolate that the maturation of periventricular oligodendrocytes around 31–32 weeks might account for the HC:ventricular size pattern in our data, since the myelin might then be mature enough to counter the ventricular expansion or oxidative stress caused by IVH.
Although the Papile scale already exists to grade IVH by the extent of bleeding through the ventricles and surrounding parenchyma,3 these ratios could be used to guide further management after an IVH diagnosis is established. There currently exist few standardized methods to time the neurosurgical intervention for IVH with hydrocephalus, with one major study from the Hydrocephalus Research Network agreeing that on an FOHR cutoff of 0.55 and others finding that a VI above the 97th percentile should prompt temporizing CSF diversion measures. 10,30,31
An infant with ultrasound findings of severe IVH and HC:ventricle size ratio consistent with high-grade IVH might prompt clinicians to consider earlier interventions, but these ratios only consistently differentiate low and high-grade infants after 30 weeks gestational age. The most important finding to emphasize here is that HC alone cannot be a stand-alone metric of internal ventricular size, thus necessitating some form of cranial imaging to diagnose rapid ventricular dilation and hydrocephalus.
Limitations & Future Directions
This study analyzes only head circumferences and radiographic measurements in neonates but doesn’t consider the variety of indices of growth such as weight and nutritional status that may affect HC. Further, temporary or permanent CSF diversion may be delayed by other systemic issues such as necrotizing enterocolitis, pulmonary complications, congenital defects, or infections which would increase the risk of a neurosurgical procedure.33–35 Our study is also limited by the retrospective nature of data collection. We were not able to standardize timing of measurements, thus limiting true correlation between HC and ventricular size measurements. Prospective studies enforcing specific times for measuring clinical data from patients would remove any bias introduced here. Future studies should also analyze HC and ventricle size metrics after patients undergo temporary CSF diversion measures to characterize patterns that lead to permanent CSF diversion. We should also emphasize that this study does not consider circumstances where infants present with microcephaly and hydrocephalus, which can be seen in conditions such as congenital Zika infections, encephalocele, or hypoxic-ischemic encephalopathy.36–38 Conditions in which an infant has microcephaly with hydrocephalus would be expected to have much different ratios of HC:ventricle size and would likely be managed differently from a neurosurgical perspective.
Conclusions
We analyzed relationships between head circumference and three measures of ventricular size, VI, ER, and FOHR, in premature infants with IVH. Our study shows that there is a significant difference in the ratio of head circumference to ventricular size between patients with high-grade IVH compared to low-grade IVH. “We also found significant differences in the ratio of head circumference to ventricular size in patients that eventually needed CSF diversion and those that did not. Significance persisted regardless of what method was used to measure ventricle size (VI, ER, FOHR). Notably, these HC:ventricle size ratios vary week-to-week, suggesting that HC does not proportionally change with ventricle size. Therefore, we recommend selecting one method of ventricular size measurement by ultrasound (either VI, ER, or FOHR) for routine assessment in addition to measuring head circumference measurements routinely to clinically evaluate infants with IVH of prematurity.”
Supplementary Material
Funding & Disclosures:
The authors have no conflicts of interest relevant to this article to disclose. This publication was made possible (in part) by Grant Number UL1TR002553 from the National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health (NIH), awarded to Vishal Venkatraman. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NCATS or NIH.
References
- 1.Leijser LM, de Vries LS. Preterm brain injury: Germinal matrix-intraventricular hemorrhage and post-hemorrhagic ventricular dilatation. Handb Clin Neurol. 2019;162:173–199. doi: 10.1016/b978-0-444-64029-1.00008-4 [DOI] [PubMed] [Google Scholar]
- 2.Ballabh P Intraventricular hemorrhage in premature infants: mechanism of disease. Pediatr Res. Jan 2010;67(1):1–8. doi: 10.1203/PDR.0b013e3181c1b176 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Valdez Sandoval P, Hernández Rosales P, Quiñones Hernández DG, Chavana Naranjo EA, García Navarro V. Intraventricular hemorrhage and posthemorrhagic hydrocephalus in preterm infants: diagnosis, classification, and treatment options. Childs Nerv Syst. Jun 2019;35(6):917–927. doi: 10.1007/s00381-019-04127-x [DOI] [PubMed] [Google Scholar]
- 4.Chamiraju P, Bhatia S, Sandberg DI, Ragheb J. Endoscopic third ventriculostomy and choroid plexus cauterization in posthemorrhagic hydrocephalus of prematurity. J Neurosurg Pediatr. Apr 2014;13(4):433–9. doi: 10.3171/2013.12.PEDS13219 [DOI] [PubMed] [Google Scholar]
- 5.Christian EA, Jin DL, Attenello F, et al. Trends in hospitalization of preterm infants with intraventricular hemorrhage and hydrocephalus in the United States, 2000–2010. J Neurosurg Pediatr. Mar 2016;17(3):260–9. doi: 10.3171/2015.7.PEDS15140 [DOI] [PubMed] [Google Scholar]
- 6.Ingram MC, Huguenard AL, Miller BA, Chern JJ. Poor correlation between head circumference and cranial ultrasound findings in premature infants with intraventricular hemorrhage. J Neurosurg Pediatr. Aug 2014;14(2):184–9. doi: 10.3171/2014.5.PEDS13602 [DOI] [PubMed] [Google Scholar]
- 7.Mazzola CA, Choudhri AF, Auguste KI, et al. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 2: Management of posthemorrhagic hydrocephalus in premature infants. J Neurosurg Pediatr. Nov 2014;14 Suppl 1:8–23. doi: 10.3171/2014.7.PEDS14322 [DOI] [PubMed] [Google Scholar]
- 8.Leijser LM, Miller SP, van Wezel-Meijler G, et al. Posthemorrhagic ventricular dilatation in preterm infants: When best to intervene? Neurology. Feb 20 2018;90(8):e698–e706. doi: 10.1212/wnl.0000000000004984 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Jaraj D, Rabiei K, Marlow T, Jensen C, Skoog I, Wikkelso C. Estimated ventricle size using Evans index: reference values from a population-based sample. Eur J Neurol. Mar 2017;24(3):468–474. doi: 10.1111/ene.13226 [DOI] [PubMed] [Google Scholar]
- 10.El-Dib M, Limbrick DD Jr., Inder T, et al. Management of Post-hemorrhagic Ventricular Dilatation in the Infant Born Preterm. J Pediatr. Nov 2020;226:16–27 e3. doi: 10.1016/j.jpeds.2020.07.079 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Leijser LM, Scott JN, Roychoudhury S, et al. Post-hemorrhagic ventricular dilatation: inter-observer reliability of ventricular size measurements in extremely preterm infants. Pediatr Res. Aug 2021;90(2):403–410. doi: 10.1038/s41390-020-01245-0 [DOI] [PubMed] [Google Scholar]
- 12.Fenton TR, Kim JH. A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants. BMC Pediatr. Apr 20 2013;13:59. doi: 10.1186/1471-2431-13-59 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Chou JH, Roumiantsev S, Singh R. PediTools Electronic Growth Chart Calculators: Applications in Clinical Care, Research, and Quality Improvement. J Med Internet Res. Jan 30 2020;22(1):e16204. doi: 10.2196/16204 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wellons JC 3rd, Shannon CN, Holubkov R, et al. Shunting outcomes in posthemorrhagic hydrocephalus: results of a Hydrocephalus Clinical Research Network prospective cohort study. J Neurosurg Pediatr. Jul 2017;20(1):19–29. doi: 10.3171/2017.1.PEDS16496 [DOI] [PubMed] [Google Scholar]
- 15.Lai GY, Chu-Kwan W, Westcott AB, Kulkarni AV, Drake JM, Lam SK. Timing of Temporizing Neurosurgical Treatment in Relation to Shunting and Neurodevelopmental Outcomes in Posthemorrhagic Ventricular Dilatation of Prematurity: A Meta-analysis. J Pediatr. Jul 2021;234:54–64.e20. doi: 10.1016/j.jpeds.2021.01.030 [DOI] [PubMed] [Google Scholar]
- 16.de Vries LS, Liem KD, van Dijk K, et al. Early versus late treatment of posthaemorrhagic ventricular dilatation: results of a retrospective study from five neonatal intensive care units in The Netherlands. Acta Paediatr. 2002;91(2):212–7. doi: 10.1080/080352502317285234 [DOI] [PubMed] [Google Scholar]
- 17.Nagy A, Bognar L, Pataki I, Barta Z, Novak L. Ventriculosubgaleal shunt in the treatment of posthemorrhagic and postinfectious hydrocephalus of premature infants. Childs Nerv Syst. Mar 2013;29(3):413–8. doi: 10.1007/s00381-012-1968-5 [DOI] [PubMed] [Google Scholar]
- 18.Brouwer A, Groenendaal F, van Haastert IL, Rademaker K, Hanlo P, de Vries L. Neurodevelopmental outcome of preterm infants with severe intraventricular hemorrhage and therapy for post-hemorrhagic ventricular dilatation. J Pediatr. May 2008;152(5):648–54. doi: 10.1016/j.jpeds.2007.10.005 [DOI] [PubMed] [Google Scholar]
- 19.Adams-Chapman I, Hansen NI, Stoll BJ, Higgins R, Network NR. Neurodevelopmental outcome of extremely low birth weight infants with posthemorrhagic hydrocephalus requiring shunt insertion. Pediatrics. May 2008;121(5):e1167–77. doi: 10.1542/peds.2007-0423 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ballabh P, de Vries LS. White matter injury in infants with intraventricular haemorrhage: mechanisms and therapies. Nat Rev Neurol. Apr 2021;17(4):199–214. doi: 10.1038/s41582-020-00447-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Cizmeci MN, Groenendaal F, Liem KD, et al. Randomized Controlled Early versus Late Ventricular Intervention Study in Posthemorrhagic Ventricular Dilatation: Outcome at 2 Years. J Pediatr. Nov 2020;226:28–35.e3. doi: 10.1016/j.jpeds.2020.08.014 [DOI] [PubMed] [Google Scholar]
- 22.Cizmeci MN, Khalili N, Claessens NHP, et al. Assessment of Brain Injury and Brain Volumes after Posthemorrhagic Ventricular Dilatation: A Nested Substudy of the Randomized Controlled ELVIS Trial. J Pediatr. May 2019;208:191–197.e2. doi: 10.1016/j.jpeds.2018.12.062 [DOI] [PubMed] [Google Scholar]
- 23.Luyt K, Jary S, Lea C, et al. Ten-year follow-up of a randomised trial of drainage, irrigation and fibrinolytic therapy (DRIFT) in infants with post-haemorrhagic ventricular dilatation. Health Technol Assess. Feb 2019;23(4):1–116. doi: 10.3310/hta23040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Luyt K, Jary SL, Lea CL, et al. Drainage, irrigation and fibrinolytic therapy (DRIFT) for posthaemorrhagic ventricular dilatation: 10-year follow-up of a randomised controlled trial. Arch Dis Child Fetal Neonatal Ed. Sep 2020;105(5):466–473. doi: 10.1136/archdischild-2019-318231 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Fenton TR, Dai S, Lalari V, Alshaikh B. Neonatal and Preterm Infant Growth Assessment. Clin Perinatol. Jun 2022;49(2):295–311. doi: 10.1016/j.clp.2022.02.001 [DOI] [PubMed] [Google Scholar]
- 26.Goulding DS, Vogel RC, Gensel JC, Morganti JM, Stromberg AJ, Miller BA. Acute brain inflammation, white matter oxidative stress, and myelin deficiency in a model of neonatal intraventricular hemorrhage. J Neurosurg Pediatr. Aug 28 2020;26(6):613–623. doi: 10.3171/2020.5.PEDS20124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cheng B, Ballabh P. Recovery of the brain after intraventricular hemorrhage. Semin Fetal Neonatal Med. Feb 2022;27(1):101224. doi: 10.1016/j.siny.2021.101224 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Krishna S, Cheng B, Sharma DR, et al. PPAR-gamma activation enhances myelination and neurological recovery in premature rabbits with intraventricular hemorrhage. Proc Natl Acad Sci U S A. Sep 7 2021;118(36)doi: 10.1073/pnas.2103084118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Counsell SJ, Maalouf EF, Fletcher AM, et al. MR imaging assessment of myelination in the very preterm brain. AJNR Am J Neuroradiol. May 2002;23(5):872–81. [PMC free article] [PubMed] [Google Scholar]
- 30.Nayak N, Sankhla SK. Management of Posthemorrhagic Hydrocephalus. Neurol India. Nov-Dec 2021;69(Supplement):S313–S319. doi: 10.4103/0028-3886.332257 [DOI] [PubMed] [Google Scholar]
- 31.de Vries LS, Groenendaal F, Liem KD, et al. Treatment thresholds for intervention in posthaemorrhagic ventricular dilation: a randomised controlled trial. Arch Dis Child Fetal Neonatal Ed. Jan 2019;104(1):F70–f75. doi: 10.1136/archdischild-2017-314206 [DOI] [PubMed] [Google Scholar]
- 32.Brouwer MJ, de Vries LS, Groenendaal F, et al. New reference values for the neonatal cerebral ventricles. Radiology. Jan 2012;262(1):224–33. doi: 10.1148/radiol.11110334 [DOI] [PubMed] [Google Scholar]
- 33.Su BH, Lin HY, Huang FK, Tsai ML, Huang YT. Circulatory Management Focusing on Preventing Intraventricular Hemorrhage and Pulmonary Hemorrhage in Preterm Infants. Pediatr Neonatol. Dec 2016;57(6):453–462. doi: 10.1016/j.pedneo.2016.01.001 [DOI] [PubMed] [Google Scholar]
- 34.Rich BS, Dolgin SE. Necrotizing Enterocolitis. Pediatr Rev. Dec 2017;38(12):552–559. doi: 10.1542/pir.2017-0002 [DOI] [PubMed] [Google Scholar]
- 35.Patel RM. Short- and Long-Term Outcomes for Extremely Preterm Infants. Am J Perinatol. Feb 2016;33(3):318–28. doi: 10.1055/s-0035-1571202 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ho SS, Zhou Y, Rajderkar D. Intracranial Imaging of Preterm Infants with Suspected Hypoxic Ischemic Encephalopathy: Comparing MRI and Ultrasound. Curr Pediatr Rev. 2023;19(2):179–186. doi: 10.2174/1573396318666220417233146 [DOI] [PubMed] [Google Scholar]
- 37.Faizan MI, Abdullah M, Ali S, Naqvi IH, Ahmed A, Parveen S. Zika Virus-Induced Microcephaly and Its Possible Molecular Mechanism. Intervirology. 2016;59(3):152–158. doi: 10.1159/000452950 [DOI] [PubMed] [Google Scholar]
- 38.Protzenko T, Dos Santos Gomes Junior SC, Bellas A, Salomão JFM. Hydrocephalus and occipital encephaloceles: presentation of a series and review of the literature. Childs Nerv Syst. Nov 2021;37(11):3437–3445. doi: 10.1007/s00381-021-05312-7 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.