Factors causing acute shunt infection. Computer analysis of 1174 operations

Abstract

A series of 1174 operations performed on 802 hydrocephalic children was analyzed in an effort to find the factors causing acute postoperative infection. Culture of the cerebrospinal fluid (CSF) samples during the operation was positive in 33 cases. These cases were excluded from the series. Ninety infections were observed in the remaining 1141 operations, an overall rate of 7.9%. Most of these infections were meningitis (56 cases). Staphylococcus epidermidis was the bacterium most frequent identified (44%). Statistically significant relationships were found between shunt infection and the following factors: 1) age: infection was 2.6 times as frequent before 6 months than after 1 year of age (p = 0.03); 2) poor condition of the skin; 3) presence of intercurrent seats of infection at the time of surgery; 4) type of operation: the rate of infection was 8.4% in primary shunt insertions, 5% in shunt revisions, and 17.5% in reinsertions following shunt removal for infection (p = 0.0001); 5) end of the shunt requiring revision: an infection rate of 7.7% followed revision of the ventricular catheter alone, and 2.6% followed revision of the other end alone (p = 0.012); and 6) postoperative wound dehiscence or scalp necrosis. The surgical team involved was poorly correlated with the rate of infection (p = 0.12). No statistically significant relationships have been found between infection and 1) etiology of hydrocephalus: 2) sex; 3) recent neurosurgical operation before the shunt procedure; 4) preoperative presence of an external drainage tube or CSF fistula; 5) lumbar or ventricular taps, or ventriculography; 6) number of previous operations performed on the shunt; 7) time (month and year) of operation; or 8) sugar level and cell count in the CSF.