Utility of Radiologic Variables to Predict the Result of Lumbar Infusion Test in the Diagnosis of Idiopathic Normal Pressure Hydrocephalus.

ABSTRACT

BACKGROUND: Diagnosis of idiopathic normal pressure hydrocephalus is based in clinical data, radiologic variables, and invasive cerebrospinal fluid (CSF) testing, such as the lumbar infusion test. Several neuroimaging findings are inconclusively related to improvement after CSF shunt surgery. CSF tests are invasive and have complications. The aim of this study was to select radiologic variables related to a positive lumbar infusion test so as to avoid this test in patients. METHODS: Patients with possible idiopathic normal pressure hydrocephalus were reviewed. The collected radiologic data were cingulate sulcus sign, disproportionately enlarged subarachnoid space, callosal angle, and width of temporal horns. Two groups were established group 1, comprising patients with resistance to CSF outflow <12 mm Hg/mL/minute, and group 2, comprising patients with resistance to CSF outflow >12 mm Hg/mL/minute. Negative and positive predictive values were determined. RESULTS: The study included 43 patients in group 1 and 64 patients in group 2. Group 2 significantly showed more acute callosal angle with higher accuracy cutoff value of 90.6°, lower width of temporal horns with higher accuracy cutoff value of 8 mm, and higher percentage of cingulate sulcus sign and disproportionately enlarged subarachnoid space. Matching the radiologic variables, positive predictive values were >80%; however, negative predictive values were low. CONCLUSIONS: Owing to high positive predictive values of matched radiological variables, the lumbar infusion test could be avoided in the diagnosis of idiopathic normal pressure hydrocephalus. However, when 1 or 2 of the variables are negative, this invasive test should be performed.