Cranial ultrasound and clinical studies in preterm infants

Serial ultrasound imaging of the brain was used to determine the ventricular index (VI), and the ratio (VR) of the VI to the cranial hemidiameter during the nursery course and first year post-term in preterm infants of less than 33 weeks gestation. Twenty-nine of the infant survivors with no intracranial hemorrhage or major medical complication during their nursery course composed group 1. Twenty-two survivors with intracranial hemorrhage unassociated with early ventricular dilation composed group 2. Group 3 was comprised of 10 other survivors who had neonatal intraventricular hemorrhage with early ventriculomegaly; all 10 infants had at least one major medical complication during their neonatal course. In groups 1 and 2 the VR decreased and the VI increased significantly with age post-conception. Infants in group 3, compared with those in groups 1 or 2, had decreased occipitofrontal growth during the early postnatal period and increased VR and VI during the neonatal period and first year post-term. These results suggest that the ventriculomegaly associated with neonatal intracranial hemorrhage cannot be explained by posthemorrhagic hydrocephalus alone and may also be related to cerebral atrophy or decreased brain growth or both. Neurodevelopmental assessments at 20 to 30 months of age disclosed significantly lower Bayley Motor Development scores in group 3 compared with groups 1 or 2. Four infants in group 3, but none in groups 1 or 2, had cerebral palsy. The neurodevelopmental deficits in group 3 infants may reflect the complex pathogenesis of the ventriculomegaly as well as the effects of the intraventricular hemorrhage and posthemorrhagic hydrocephalus.     

Intraventricular hemorrhage in the premature infant

Periventricular-intraventricular hemorrhage is the most important adverse neurologic event of the newborn period. It is very common and can be very severe. Such hemorrhage begins in the germinal matrix but may spread into and throughout the ventricular system. It may be accompanied by hemorrhage within the brin parenchyma. The pathogenesis of periventricular-intraventricular hemorrhage is still imperfectly understood, but relates to the anatomy and physiology of the developing cerebral vasculature and to the biophysical and biochemical environment in which that development proceeds. Periventricular-intraventricular hemorrhage may be marked by a catastrophic clinical deterioration, but is more commonly accompanied by a saltatory progression that may be difficult to detect clinically. Both concomitant neonatal disease and therapeutic intervention for such disease have been implicated in the initiation and exacerbation of periventricular-intraventricular hemorrhage. Real-time ultrasound scanning with portable instruments is now the best procedure for identifying this lesion and for assessing its sequelae. Prognosis relates principally to the severity of the lesion. Early management must be particularly directed to the maintenance of cerebral perfusion. Later management is predominantly the therapy of posthemorrhagic hydrocephalus. There is no currently available therapeutic modality that will prevent progressive posthemorrhagic hydrocephalus.     

Optimal timing for diagnostic cranial ultrasound in low-birth-weight infants: detection of intracranial hemorrhage and ventricular dilation

Intracranial hemorrhage and posthemorrhagic ventricular dilation are common problems in small preterm infants. To determine the optimal timing for ultrasound diagnosis of these abnormalities, we studied 64 preterm infants (less than 1,500 gm) by sequential cranial ultrasonography from birth until one year of age or until death. The optimal timing for ultrasound diagnosis of intracranial hemorrhage is days 4 to 7 with follow-up at day 14. The most efficient time for ultrasound examination to diagnose ventricular dilation was day 14 with follow-up at 3 months. Intracranial hemorrhage was diagnosed by ultrasound in 35 of the 64 patients (55%). In 18 of the 64 infants (28%) significant ventricular dilation was diagnosed by ultrasound during the first three months.     

Transforming growth factor-beta1: a possible signal molecule for posthemorrhagic hydrocephalus?

Posthemorrhagic hydrocephalus remains a complication of preterm birth for which we lack a clear understanding and a curative therapy. Transforming growth factor beta (TGF-beta) is a cytokine that upregulates the production by fibroblasts of extracellular matrix proteins. We hypothesized that TGF-beta might be released into cerebrospinal fluid (CSF) after intraventricular hemorrhage and play a role in posthemorrhagic hydrocephalus. Total TGF-beta1 and TGF-beta2 were measured by immunoassay in CSF samples from 12 normal preterm infants, nine preterm infants with transient posthemorrhagic ventricular dilation, and 10 infants who subsequently developed permanent hydrocephalus. Five infants received intraventricular tissue plasminogen activator, and two infants were treated by drainage irrigation and fibrinolytic therapy. Median TGF-beta1 in normal CSF was 0.495 ng/mL. In infants with transient posthemorrhagic ventricular dilation, median initial CSF TGF-beta1 was 2.1 ng/mL. Infants who subsequently had permanent hydrocephalus had median initial CSF TGF-beta1, 9.7 ng/mL (differences between groups p < 0.01). Intraventricular recombinant tissue plasminogen activator was followed by a rise in CSF TGF-beta1 (p = 0.0007). Drainage irrigation and fibrinolytic therapy was followed by a fall in CSF TGF-beta1. TGF-beta2 was detected in CSF and showed similar trends, but the CSF concentration of TGF-beta1 was more than 20 times higher. These findings support the hypothesis that TGF-beta1 is released into CSF after intraventricular hemorrhage and may play an important part in hydrocephalus. The results help to explain the failure of intraventricular fibrinolytic therapy.     

Increased hypoxanthine concentrations in cerebrospinal fluid of infants with hydrocephalus

Hypoxanthine, the end product of purine metabolism, is usually very elevated in body fluids during severe hypoxia. We measured hypoxanthine in the cerebrospinal fluid of hydrocephalic preterm infants (12 with posthemorrhagic, one with congenital hydrocephalus) to determine whether hydrocephalus is associated with anaerobic metabolism of brain tissue. Cerebrospinal fluid hypoxanthine was undetectable in normal infants. In hydrocephalic infants, the concentration of hypoxanthine ranged from 7.5 mumol/L to 28 mumol (means = 14.3 mumol/L). The hypoxanthine concentrations fell from a mean of 12.8 mumol/L to a mean of 2.0 mumol/L (P less than 0.05) with successful treatment of the ventriculomegaly by lumbar puncture or by ventriculoperitoneal shunt. Patients with acute posthemorrhagic hydrocephalus had similar concentrations of hypoxanthine (means = 14.5 mumol/L) as patients with late or with congenital hydrocephalus (means = 13.8 mumol/L), indicating that brain hypoxia is probably a consequence of the ventriculomegaly and not of the hemorrhagic insult.     

Ventricular dilation preceding rapid head growth following neonatal intracranial hemorrhage.

We describe two patients in whom ventricular dilation began within a week of neonatal intraventricular hemorrhage and preceded rapid head growth and clinical signs of increased intracranial pressure by days to weeks. The posthemorrhagic ventricular dilation is readily demonstrated in the neonatal period by computerized tomography of the cranium. These data indicate the importance of careful follow-up of survivors of intraventricular hemorrhage, early demonstration of ventricular dilation and prompt therapeutic intervention to prevent brain injury.     

Matrix metalloproteinases in infants with posthemorrhagic hydrocephalus

The cerebrospinal fluid matrix metalloproteinase (MMP) activities were measured in infants with posthemorrhagic hydrocephalus to elucidate the intrinsic mechanism for the resolution of ventricular dilation. Increased MMP-9 activities were observed in the patients who escaped a shunt operation, suggesting its potential contribution to the resolution of ventricular dilation.     

Treatment of progressive posthemorrhagic hydrocephalus with temporary external ventricular drainage. Preliminary results

Peri-intraventricular hemorrhage (PIVH) is a major problem of preterm neonates: over 40% of infants with birth weight less than 1500 g have been found to experience this disorder. The posthemorrhagic hydrocephalus (PHH) is generally believed to occur secondary to obliterative posterior fossa arachnoiditis. Its management in critically ill premature infants with multiple medical problems has thus become quite significant. This paper reports the results of the placement of a temporary external ventricular drainage in three patients in whom progressive ventricular dilatation began within two weeks after severe neonatal PIVH. In all patients the hydrocephalus and increased intracranial pressure were controlled within three weeks of treatment. One infant died from other problems associated with his prematurity. Early demonstration of progressive ventricular dilatation by ultrasound and prompt external drainage has been found to be a safe initial method to treat PHH in preterm infants.     

Posthemorrhagic hydrocephalus. Use of an intravenous-type catheter for cerebrospinal fluid drainage

STUDY OBJECTIVE: To report a 9-year experience with the treatment of posthemorrhagic hydrocephalus (PHH) with the use of an easily inserted external ventricular drain. DESIGN: A case series with a retrospective review of hospital records and cranial ultrasound results, from 1981 through 1989, in all infants with PHH. INTERVENTION: A previously defined method of identification and bedside management of PHH was applied. If infants reached 2 kg of body weight and PHH recurred, a ventriculoperitoneal shunt was inserted. RESULTS: A total of 70 procedures were performed in 24 patients, and all were associated with a decrease in head circumference and ventricular size on ultrasound scan. One infection occurred, and only 12 infants required a ventriculoperitoneal shunt. CONCLUSIONS: This technique compared favorably with other methods of intervention to avoid early placement of a ventriculoperitoneal shunt in preterm infants and offered the advantage of consistently decreasing ventricular size. A multicenter-controlled trial will be needed to compare the safety and efficacy of therapies for PHH.     

Determinants of major handicap in post-haemorrhagic hydrocephalus.

The outcome of 54 low birthweight (less than 2500 g) infants treated by ventriculoperitoneal or ventriculoatrial shunting for post-haemorrhagic hydrocephalus over six years is reviewed. Prognosis correlated with cranial ultrasound appearances in the neonatal period but not with factors relating to the aetiology or extent or management of the hydrocephalus.     

Periventricular-intraventricular hemorrhage, sonographic localization, phenobarbital, and motor abnormalities in low birth weight infants

A total of 228 low birth weight (less than 1750 g), mechanically ventilated infants with and without periventricular-intraventricular hemorrhage were examined at 18 months corrected age to assess the relationship between cranial ultrasonographic findings and specific motor abnormalities. All infants were previously enrolled in a double-blind, randomized, prospective clinical trial of phenobarbital prophylaxis against periventricular-intraventricular hemorrhage. Ultrasonographic abnormalities on the scans performed between 7 and 13 days of life were categorized as germinal matrix hemorrhage, lateral ventricular hemorrhage, parenchymal hemorrhage, ventriculomegaly, and any hemorrhage. Regardless of anatomical location, periventricular-intraventricular hemorrhage was associated with an increased risk for developing motor abnormalities. Hypertonia and hyperreflexia/ankle clonus were most common. No abnormal motor findings distinguished unilateral from bilateral germinal matrix hemorrhage and lateral ventricular hemorrhage or between phenobarbital and placebo treatment. None of the 5 infants with parenchymal hemorrhage had spastic cerebral palsy. Ventriculomegaly was associated with a fivefold increase in risk for spastic cerebral palsy and delayed walking and a threefold increase for hypertonia and hyperreflexia/clonus. The results suggest that ventriculomegaly, observed even as early as the first week of life, might be a significant antecedent of later motor abnormalities among the survivors of periventricular-intraventricular hemorrhage.     

Progressive posthämorrhagische Ventrikelerweiterung des Frühgeborenen Inzidenz, Prognose und Therapie

Background. 35% of preterm infants with intraventricular hemorrhage develop ventricular dilatation. The posthemorrhagic ventricular dilatation can persist, be transient or be progressive. Finally 1–2% of all very low birthweight (VLBW <1500 g) infants require shunt placement for the treatment of the posthemorrhagic hydrocephalus. Outcome. The neurodevelopmental outcome is extremely poor in children surviving progressive posthemorrhagic ventricular dilatation. Therapy. There are no uniform guidelines for the treatment of preterm infants with progressive posthemorrhagic ventricular dilatation. Randomised multicenter trials demonstrated, that inhibition of cerebrospinal fluid production with acetazolamide and furosemide or early cerebrospinal fluid tapping did not reduce the need for shunt placement and may worsen the outcome or can be associated with adverse side-effects. Preterm infants with posthemorrhagic ventricular dilatation should be treated by standard therapy with uniform guidelines for cerebrospinal fluid tapping and shunt placement.     

Ventricular volume in infantile hydrocephalus and its relationship to intracranial pressure and cerebrospinal fluid clearance before and after treatment. A preliminary study

The present study examines ventricular volume before and after shunting in 16 infants with hydrocephalus and mixed aetiology and relates this parameter to rates of cerebrospinal fluid clearance, intracranial pressure and outcome. Following treatment all patients showed reduction in intracranial pressure and amelioration of the clinical manifestations of hydrocephalus. Nevertheless, 4 patients showed persistence of marked ventriculomegaly, 4 had moderate ventricular enlargement, while 8 showed small or normal ventricles. Although the series was small there appeared to be a clear correlation between persistent ventriculomegaly, poor cerebrospinal fluid clearance and poor outcome despite reduction in intracranial pressure. The question is raised as to whether pressure-determined cerebrospinal fluid drainage provides optimum treatment of hydrocephalus in infants given the particular physical characteristics pertaining to the non-rigid cranium.     

A potential mechanism of pathogenesis for early posthemorrhagic hydrocephalus in the premature newborn

Ventricular dilation is common following intraventricular hemorrhage. Neuropathologic studies have demonstrated that chronic posthemorrhagic hydrocephalus most commonly is a result of an obliterative arachnoiditis in the posterior fossa or is due to obstruction of flow of CSF within the ventricular system. Recent use of ultrasound scanning has demonstrated the occurrence of ventricular dilation within days of intraventricular hemorrhage (prior to the expected time of development of arachnoiditis). In the case described, serial real-time ultrasound scans demonstrated small mobile particles within dilated ventricles seven days following intraventricular hemorrhage. There was no obstruction of CSF flow within the ventricular system. Thus, in this case, ventricular dilation may have been secondary to plugging of arachnoid villi by the small particulate matter and, as a consequence, decrease in CSF reabsorption.     

Measurement of progressive cerebral ventriculomegaly in infants after grades III and IV intraventricular hemorrhages

To develop guidelines that might help predict prospectively which infants with severe intraventricular hemorrhage (IVH) would require intervention, we obtained serial cranial sonograms to measure the rate of growth of cerebral ventricular volumes in 48 preterm infants with and without IVH. The infants were divided into three groups: (1) those with no IVH (22 infants), (2) those with IVH with acute ventricular dilation (13 infants), and (3) those with IVH with progressive ventricular dilation requiring intervention (13 infants). The decision to intervene because of progressive ventricular dilation was based on clinical criteria and the subjective assessment of increasing ventricular size on weekly cranial sonograms. The rate of cerebral ventricular volume growth in infants with IVH who needed intervention was greater (4.2 +/- 3.3 ml/day) than that in infants without IVH (0.0 +/- 0.1 ml/day; p less than 0.001) and in infants with IVH and acute ventricular dilation (0.0 +/- 0.2 ml/day; p less than 0.001). Using these data, we generated guidelines for predicting prospectively which infants with IVH and ventricular dilation will need intervention for posthemorrhagic hydrocephalus. The guidelines were then confirmed prospectively in 10 infants.     

Cisterna magna clot and subsequent post-hemorrhagic hydrocephalus

Background. Posthemorrhagic hydrocephalus (PHH) is an ominous complication of intraventricular hemorrhage (IVH) in premature infants. Previous studies have correlated lateral intraventricular clot with subsequent PHH, but there are no studies assessing the outcome of clot in the cisterna magna (CM). Objective. The purpose of this study was to determine if the identification of clot in the CM on the initial positive cranial ultrasound for intraventricular hemorrhage increased the risk of subsequent PHH. Materials and methods. A review of ultrasound records over a 4-year period identified 41 neonates with at least grade 2 IVH who had specific CM views prospectively performed. Results. Thirty-six survived more than 3 months. Overall, 22 (61 %) developed hydrocephalus. On the initial positive cranial ultrasound, CM blood was identified in 21 (58 %). Of these 18 (86 %) developed hydrocephalus [odds ratio (OR) 16.5, confidence interval (CI) 2.5–125.7, P < 0.001] and 14 (67 %) required intervention or remained dilated (OR 28.0, CI 2.8–1265.8). All patients that required permanent shunt placement had initial CM clot (P < 0.01). Initial ventriculomegaly was not a significant predictor of subsequent hydrocephalus, but a dilated third ventricle (≥ 8 mm) was (OR 9.0, CI 1.2–103.3, P < 0.04). Asymmetric intraventricular clot filling of more than 50 % was not predictive of hydrocephalus, but symmetric clot of more than 50 % was (OR 10.8, CI 1.4–61.6, P < 0.01). Nine neonates had concomitant parenchymal damage with varying outcomes (three shunted, two persistent ventriculomegaly, four resolved with normal sized ventricles). Conclusion. In this study CM clot significantly increased the risk and was a better predictor of posthemorrhagic hydrocephalus than initial hydrocephalus. Received: 3 July 2000 Accepted: 22 September 2000     

Hypoglycorrhachia in neonatal intracranial hemorrhage. Relationship to posthemorrhagic hydrocephalus

The level of cerebrospinal fluid (CSF) glucose may be lowered after subarachnoid hemorrhage. This was observed in each of 18 cases of proven posthemorrhagic hydrocephalus in infants (study group). In one of these children with a hemorrhagic spinal fluid and hypoglycorrhachia unaccompanied by clinical signs of intracranial hemorrhage or hydrocephalus, the axial tomography showed a significant although asymptomatic hydrocephalus. To further evaluate the significance of this finding (hypoglycorrhachia), we compared the incidence of hypoglycorrhachia (CSF glucose less than 40 mg) and lowered CSF glucose/blood glucose ratio (ratio less than 0.4) at three similar time intervals from the presumed time of the intracranial hemorrhage in the study group with that of a control group of 40 neonates with similar neonatal associated pathology (mainly premature infants with hyaline membrane disease) but who did not later develop posthemorrhagic hydrocephalus or cerebral palsy. There was a statistically greater frequency of these anomalies in the hydrocephalic group. Only 3 of the 40 control patients had hypoglycorrhachia and low ratio. Hypoglycorrhachia in the absence of other known causes for decreased CSF glucose is a good index of a probably significant meningeal hemorrhage with a high risk of secondary hydrocephalus which may or may not be symptomatic. Hypoglycorrhachia may be used as an indication of the frequency of clinically inapparent subarachnoid hemorrhage in these high risk newborns.     

Asymmetric growth of the lateral cerebral ventricle in infants with posthemorrhagic ventricular dilation

Lateral cerebral ventricular volume in 36 preterm infants with or without an intraventricular hemorrhage, and with or without posthemorrhagic hydrocephalus, was measured longitudinally and compared with the ventricular index measurements of the same ventricles. A poor correlation was found (r2 = 0.67). To determine a reason for this poor relationship, we analyzed the volumes of the regions of the ventricles by a segmental volume analysis. The occipital region of the lateral cerebral ventricle enlarged at a much faster rate (1.904 +/- 0.477 ml/day) than either the anterior region (0.546 +/- 0.253 ml/day; p less than 0.01) or the middle region (-0.209 +/- 0.334 ml/day; p less than 0.01) in infants with posthemorrhagic hydrocephalus. The rate of growth of the middle region of the lateral cerebral ventricles was the same for all infants. Linear indexes, such as the ventricular index and the lateral ventricular ratio, do not allow for accurate serial estimates of ventricular size in posthemorrhagic hydrocephalus because of asymmetric growth of the lateral cerebral ventricle. We conclude that sequential volume measurements are more useful than ventricular index measurements to follow ventricular size sequentially in infants with posthemorrhagic hydrocephalus.     

Intraventricular hemorrhage. Incidence and outcome in a population of very-low-birth-weight infants

The incidence, extent, and outcome of germinal matrix hemorrhage-intraventricular hemorrhage (GMH-IVH) were determined with the use of ultrasound and autopsy findings in 100 consecutive infants, with a birth weight of less than 1,500 g. Serial ultrasound examinations once or twice weekly were performed with the use of a portable real-time linear-array scanner. The overall incidence of GMH-IVH was 46%. Twenty infants had grade 1 (GMH), 24 had grade 2 (IVH +/- GMH), and two had grade 3 (IVH +/- GMH with intracerebral hemorrhage) conditions. The mortality in infants with GMH-IVH was 35%, compared with 13% in infants without GMH-IVH. Although 11 (37%) of 30 survivors with GMH-IVH had ventricular dilatation, only two infants required ventriculoperitoneal shunts for progressive hydrocephalus. The incidence of GMH-IVH was increased in outborn infants, in those delivered vaginally, and in those who required mechanical ventilation, bicarbonate therapy, or volume expansion in the first 24 hours. The long-term prognostic significance of the ultrasound findings was unknown and will be determined by follow-up studies.     

Developmental outcome following posthemorrhagic hydrocephalus in preterm infants. Comparison of twins discordant for hydrocephalus

The mental development and neurologic outcome of four sets of preterm twins, in whom one infant within each pair developed posthemorrhagic hydrocephalus during the neonatal period, were assessed. Each hydrocephalic infant presented with signs of increased intracranial pressure and required placement of a ventriculoperitoneal shunt. Three of the four hydrocephalic infants required revision or replacement of the shunt during the first year. The four nonhydrocephalic infants had normal neurologic outcomes. All four hydrocephalic infants had some neurologic abnormalities on follow-up examination. Neurologic abnormalities were transient in one infant and persistent in three others, each of whom showed evidence of mild right hemiparesis. Within each twin pair, developmental rates and mental test scores were strikingly similar throughout infancy and early childhood. Two of the twin pairs have undergone psychological assessments at 8 and 11 years of age. Relatively lower IQ scores were obtained for the hydrocephalic child in each twin pair at these ages.