Communicating hydrocephalus in adults. Diagnostic tests and results of treatment with medium pressure shunts.

In a study of 60 patients with communicating hydrocephalus, sources of error included lumbar cerebrospinal fluid (CSF) leakage, epidural infusion, and cerebrovascular reactions during the CSF infusion test; a high frequency of failing shunts; and initially undetected tumors in the group whose hydrocephalus was idiopathic. Though no sharp distinction could be made between positive and negative diagnositc tests, each variable indicated a certain probability of benefit, ranging from high to low, from shunting. CSF opening pressure in the lateral and sitting positions, the CSF infusion test, and cisternography were helpful in selecting patients for shunting. If clinical improvement was to take place, it did so within 3 months of shunting. Only patients with an opening pressure at or within the upper half of supposedly normal CSF pressure improved from a medium-pressure shunt. Improvement occurred in all patients having pressures of 580 mm H2O or more during an infusion test at a rate of 1.5 ml per minute. Neither preoperative signs and symptoms nor a single electroencephalogram had predictive value.