A multicentre randomised controlled trial of low versus high threshold treatment in preterm infants with progressive posthaemorrhagic ventricular dilatation.

Posthaemorrhagic ventricular dilatation (PHVD) is the most serious direct complication of intraventricular haemorrhage. If progressive ventricular enlargement exceeds 4 mm over the 97th centile for gestational age, PHVD carries a poor prognosis with about 50 – 60 % being shunt dependent, over 60 % disabled and about 20 % not surviving the neonatal period. However, whether lower threshold treatment for PHVD decreases the need for shunting and improves long term neurodevelopmental outcome is still under debate.
In a recent retrospective study in 5 Dutch neonatal intensive care units 95 surviving infants with a gestational age equal to or below 34 wk, diagnosed as having a grade III haemorrhage according to Volpe who developed PHVD (ventricular enlargement above the 97th centile for gestational age) were included. Intervention was not deemed necessary in 22 infants, because of lack of progression of ventricular dilatation. Low threshold intervention (progressive PHVD exceeding the 97th centile) was associated with a strongly reduced risk of ventriculoperitoneal shunting (odds ratio = 0.22, 95% confidence interval: 0.08-0.62) and a lower number with a moderate or severe handicap (5/31; 16%) compared to high threshold intervention (PHVD exceeding 4 mm over the 97th centile) (11/42; 26%).
A randomised prospective intervention study is needed to prove the beneficial role of low threshold intervention on the risk of ventriculoperitoneal shunting and neurodevelopmental outcome.

We hypothesize that in preterm infants with a gestational age below 34 weeks a low threshold intervention (progressive PHVD with a ventricular enlargement above the 97th centile for gestational age according to Levene and a diagonal width enlargement of the frontal horn above 6 mm according to Davies) will decrease the need for a ventriculoperitoneal shunt as compared to high threshold intervention (PHVD exceeding 4 mm over the 97th centile according to Levene and an increase in diagonal width of the frontal horn above 10 mm according to Davies) and will improve neurodevelopmental outcome at two years of age.

Comparison: low threshold versus high threshold intervention.
Low threshold: intervention when an
increase in ventricular width according to Levene above the 97th centile towards the P97+4 but without crossing the > P97+4 and an increase
in diagonal width according to Davies above 6 mm > towards 10 mm, but not
above 10 mm.
High threshold: intervention after an increase in ventricular width according to Levene above the P97+4 and an increase in diagonal width according to Davies above 10 mm. Intervention:
Lumbar punctures (LP; 10 ml/kg) on 2 days. Cranial ultrasound is repeated
daily. If on the third day a LP is still required, a subcutaneous
reservoir will be inserted. Daily 10 cc/kg will be drained in 2 taps a
day. Punctures from the reservoir will be continued over the next days
or weeks. The amount of CSF drained will be increased or decreased in
order to reach and keep the ventricular Index according to Levene < P97
and diagonal anterior horn width < 6 mm. If punctures are still necessary
exceeding 28 days after the first LP, a ventriculoperitoneal shunt is
inserted. If the bodyweight of the infant is less than 2,5 kg, the
insertion of the shunt will be postponed until the bodyweight is over 2,5
kg, if CSF drainage is still needed then.