Predictive Values of Lumbar Infusion Testing in Idiopathic Normal Pressure Hydrocephalus Diagnosis.

Abstract

Background and objectives: Idiopathic normal-pressure hydrocephalus (iNPH) remains a diagnostic challenge because of the lack of a single reliable diagnostic test. The aim of this study was to evaluate the performance of the lumbar infusion test (LIT) in the decision-making process for shunting in patients with suspected iNPH.

Methods: A total of 201 patients with suspected iNPH underwent complete clinical, radiological, and hydrodynamic evaluation, including LIT, between January 2016 and January 2024 at Lille University Hospital. Patients were categorized into unlikely, possible, or probable iNPH groups before and after LIT. The decision to shunt was made in a multidisciplinary setting. The sensitivity, specificity, and predictive values of the resistance coefficient (Rout) thresholds were analyzed. Clinical outcomes after shunting were assessed using the Larsson categorization, walking test, and Mini-Mental State Examination.

Results: Among 165 patients who completed an interpretable LIT for iNPH diagnosis suspicion, the mean age was 79 years. Gait disturbances were present in 98.8%, cognitive impairment in 86.6% and urinary disturbances in 58.8% of cases. The mean Rout values varied significantly across the diagnostic groups (P < .0001). The optimal threshold for Rout was determined to be 12 mm Hg/mL/min, with a sensitivity of 78.1% and specificity of 74.3% (area under the curve = 0.8441, P < .0001). Among 41 patients in the possible iNPH group who underwent shunting based on LIT findings, 75.6% demonstrated clinical improvement. In the probable iNPH group, 87.1% of shunted patients showed improvement. None of the unlikely patients with iNPH with negative LIT results underwent surgery.

Conclusion: Although LIT does not provide a definitive diagnosis of iNPH, it serves as a safe and valuable adjunct in cases of diagnostic uncertainty, aiding in early decision making and potentially improving postshunting outcomes. Its reproducibility, ease of implementation, and low morbidity make it a useful tool in the diagnosis of iNPH when combined with clinical and radiological assessments.