Ventriculoperitoneal Shunts in Low Birth Weight Infants with Intracranial Hemorrhage: Neurodevelopmental Outcome

Abstract

Fifty preterm infants (mean birth weight, 1266 +/- 303 g; mean gestational age, 30 +/- 2 weeks) who required a ventriculoperitoneal (VP) shunt for posthemorrhagic hydrocephalus (92% with Grade III or IV hemorrhage) were followed for neurodevelopmental problems. VP shunts were placed at a median age of 29 days (range, 18 to 87 days) after serial lumbar punctures failed to control progressive and symptomatic ventriculomegaly. A total of 34 infants (68%) required one shunt revision or more, and the overall infection rate per patient was 50%. Seven infants died, 2 from shunt infections. The infants were evaluated with audiological, ophthalmological, and neurodevelopmental examinations. Of the survivors, 11 (28%) have severe visual loss and 10 (24%) have hearing impairment. Of the infants, 21 (49%) have severe motor handicaps and 19 (38%) have seizure disorders. Developmental and motor scores were obtained using the Bayley or Knobloch-Gesell scales. Seven infants (18%) have normal developmental outcomes; 26 (60%) have multiple handicaps. Grade IV hemorrhage or the occurrence of seizures was a predictor of poor neurodevelopmental outcome. We conclude that progressive posthemorrhagic hydrocephalus in low birth weight infants is associated with multiple handicaps despite early VP shunt placement.

Early Treatment of Posthemorrhagic Hydrocephalus in Low-Birth-Weight Infants with Valveless Ventriculoperitoneal Shunts

“…However, the quantities of fluid to be absorbed are low, as demonstrated by Cornips et al [7]; none of our patients developed ascites or a hydrothorax, and hydroceles were not particularly a problem. Early shunting of LBWI with PHH using very-low-resistance shunts or even valveless shunts has been proposed previously [4, 9]. The risk of shunt obstruction is minimal with no-resistance shunts, but the risk of overdrainage must be taken into account.…”

“…The incidence of shunt infection in LBWI with PHH differs markedly from one study to another, ranging from 20–22% when only the first shunt insertion is considered [8, 15, 22], to as high as 50% when infections after shunt revision are included [4]. Other authors consider that ‘shunt infections are not a concern in the low birth weight infant’ [9]; in a recent study, the infection rate was as low as 3% [19].…”

“…The risk of obstruction decreases as the ventricular clots dissolve, but delaying surgery may result in an unacceptable cost in terms of the effects on psychomotor development [9]. The total number of shunt revisions in LBWI shunted for PHH may be as high as 4–6.6 per patient in some series [4, 19], but we do not think that this number indicates a poorer prognosis. In this situation, shunt revision is the solution, not the problem, and all efforts to maintain shunt patency are warranted.…”

“…However, cerebrospinal fluid (CSF) diversion in LBWI is associated with a high rate of failure and infection, which can pose daunting problems. These complications can be life-threatening in a number of cases [4]and contribute significantly to long-term morbidity in survivors [5]. Lumbar taps are generally performed initially, as a temporary measure, during the time needed for the blood clots to dissolve, before a ventriculoperitoneal (VP) shunt can be inserted [6].…”

“…External drainage has the advantage of good control of the amount of fluid drained [7], but is associated with a high risk of infection in most series, especially when used for long periods, and an internal shunt is often preferred. To lessen the risk of shunt clogging by bloody CSF, some authors have stressed the need for low-opening-pressure, low-resistance valves [4, 8], or even valveless shunts [9], but these bear a high incidence of slit-ventricle syndrome in the long term. In the short term, the ability of the premature child’s peritoneum to absorb bloody CSF has been questioned as well [10].…”