Cerebrospinal fluid (CSF) shunting and ventriculocisternostomy (ETV) in 400 pediatric patients. Shifts in understanding, diagnostics, case-mix, and surgical management during half a century

A. Henriette Paulsen, Bernt J. Due-Tønnessen, Tryggve Lundar, Karl-Fredrik Lindegaard

To characterize shifts from the 1960s to the first decade in the 21st century as to diagnostics, case-mix, and surgical management of pediatric patients undergoing permanent CSF diversion procedures. One hundred and thirty-four patients below 15 years of age were the first time treated with CSF shunt or ETV for hydrocephalus or idiopathic intracranial hypertension (IIH) in 2009-2013. This represents our current practice. Our previously reported cohorts of shunted children 1967-1970 (n = 128) and 1985-1988 (n = 138) served as backgrounds for comparison. In the 1960s, ventriculography and head circumference measurements were the main diagnostic tools; ventriculoatrial shunt was the preferred procedure (94 %), neural tube defect (NTD) was the leading etiology (33 %), and overall 2-year survival rate was 76 % (non-tumor survival 84 %). In the 1980s, computerized tomography (CT) was the preferred diagnostic imaging tool; ventriculoperitoneal shunt (VPS) had become standard (91 %), the proportion of NTD children declined to 17 %, and the 2-year survival rate was 91 % (non-tumor survival 95 %). Hydrocephalus caused by intracranial hemorrhage had, on the other hand, increased from 7 to 19 %. In the years 2009-2013, when MRI and endoscopic third ventriculocisternostomy (ETV) were matured technologies, 73 % underwent VPS, and 23 % ETV as their initial surgical procedure. The most prevalent etiology was CNS tumor (31 %). The proportion of NTD patients was yet again halved to 8 %, while intracranial hemorrhage was also reduced to 12 %. In this last period, six children were treated with VPS for Idiopathic Intracranial Hypertension (IIH) due to unsatisfactory response to medical treatment. They all had headache, papilledema, and visual disturbances and responded favorably to treatment. The 2 years of survival was 92 % (non-tumor survival 99 %). In contrast to the previous periods, there was no early shunt related mortality (2 years). Aqueductal stenosis was a small but distinctive group in all cohorts with 5, 6 and 3 % respectively. The case-mix in pediatric patients treated with permanent CSF diversion has changed over the last half-century. With the higher proportion of children with CNS tumor patients and inclusion of the IIH children, the median age at initial surgery has shifted substantially from 3.2 to 14 months. Between the 1960s and the current cohort, 2 years of all-cause mortality fell from 24 to 8 %. Prolonged asymptomatic periods, extending 15 years, were relatively common. Nevertheless, 18 patients experienced shunt failure more than 15 years after last revision, and first-time shunt failure has been observed 29 years after initial treatment. This underscores the importance of life-long follow-up.