Intracranial pressure changes during intermittent CSF drainage.

Premature very-low-birth-weight infants with posthemorrhagic hydrocephalus are often managed with intermittent cerebrospinal fluid drainage from a ventricular reservoir. There are little data regarding intracranial pressure changes during intermittent drainage to determine the amount and frequency of cerebrospinal fluid removal or to determine the correct resistance of future programmable shunts. The objective of this study was to determine the feasibility of using a commercially available intracranial pressure transducer to measure changes in pressure associated with this procedure. Continuous intracranial pressure was measured in three infants with a transducer placed at the time of ventricular reservoir insertion. Daily reservoir taps began 48 hours after placement and intracranial pressure was monitored for 7 days. Intracranial pressure before the initial tap was comparable to levels previously reported as normal. The daily removal of 10 cc/kg of cerebrospinal fluid was sufficient to lower intracranial pressure below baseline, however it was associated with wide swings in pressure and, in one patient, sustained negative pressure. The use of direct intracranial pressure monitoring may be useful in determining the optimal amount and frequency of cerebrospinal drainage from infants with posthemorrhagic hydrocephalus managed with a ventricular reservoir, as well as determining resistance settings of subsequent programmable shunts.

Indices of hemodynamic and respiratory functions in premature infants at risk for the development of cerebral white matter injury.

OBJECTIVE: To examine potential differences in clinical risk factors, including indices of hemodynamic and respiratory functions, of premature infants developing periventricular hemorrhagic infarction (PHI) or periventricular leukomalacia (PVL). STUDY DESIGN: Indices of hemodynamic stability and respiratory function were measured prospectively during the first week of life in a cohort of 100 premature infants with respiratory distress. Maternal history was retrospectively reviewed. These data were correlated with cranial ultrasonography using one-way ANOVA, Bonferroni multiple comparisons, and Wilcoxon rank sum tests. Longitudinal analysis was performed using Generalized Estimating Equations. RESULTS: Fifty-two infants with normal cranial ultrasound studies were compared to 12 with PHI and 9 with PVL. Infants developing PHI had significantly lower birth weights, lower Apgar scores, were more often male and multiple gestations, and required more vasopressor support than infants with normal ultrasound studies. Infants with PHI had significantly worse indices of respiratory function than either normal infants or those with PVL. PVL was significantly associated with maternal chorioamnionitis, whereas PHI was not. CONCLUSION: These data suggest that there are important differences in the pathogenesis of PHI and PVL. A clear understanding of these differences is required before future preventive strategies can be formulated.

Proinflammatory cytokine-receptor interaction model improves the predictability of cerebral white matter injury in preterm infants.

Proinflammatory cytokines have been variably linked to development of cerebral white matter injury (WMI) in preterm infants. Because soluble receptors tightly control cytokine bioactivity, we modeled cytokine-receptor interaction as a predictor of WMI. Plasma from 100 preterm infants was assayed for cytokines (tumor necrosis factor alpha, interleukin (IL-1beta, IL-6) and their soluble receptors (sTNF-RI), sTNF-RII, sIL-1RA, and sIL-6R). Cranial ultrasound (US) results were correlated with cytokine and receptor concentrations individually and with cytokine-receptor interaction models (PROC LOGISTIC; SAS Software). Receiver operating characteristic curves were constructed to determine the predictability of WMI. Fifty-two infants with normal US exams were compared with 21 infants with evidence of WMI. There was no association between individual cytokine or receptor concentrations and the development of WMI. However, modeling cytokines with their soluble receptors significantly improved the predictability of WMI. We concluded that consideration of cytokine-receptor interaction may be more important than individual cytokine concentrations alone in determining the role of inflammation in the pathogenesis of WMI in preterm infants.