Background and purpose
Idiopathic normal pressure hydrocephalus (iNPH) is a potentially treatable disorder, but prognostic tests or biomarkers are lacking. The aim was to study the predictive power of clinical, neuroimaging and lumbar infusion test parameters (resistance to outflow Rout, cardiac-related pulse amplitude PA and the PA to intracranial pressure ICP ratio).

Methods
In all, 127 patients diagnosed with iNPH who had a lumbar infusion test, a subsequent ventriculo-peritoneal shunt operation and at least 2 months of postoperative follow-up were retrospectively included. Preoperative magnetic resonance images were visually scored for NPH features using the iNPH Radscale. Preoperative and postoperative assessment was performed using cognitive testing, as well as gait and incontinence scales.

Results
At follow-up (7.4 months, range 2–20 months), an overall positive response was seen in 82% of the patients. Gait was more severely impaired at baseline in responders compared to non-responders. The iNPH Radscale score was borderline significantly higher in responders compared with non-responders, whereas no significant differences in infusion test parameters were seen between responders and non-responders. Infusion test parameters performed modestly with high positive (75%–92%) but low negative (17%–23%) predictive values. Although not significant, PA and PA/ICP seemed to perform better than Rout, and the odds ratio for shunt response seemed to increase in patients with higher PA/ICP, especially in patients with lower iNPH Radscale scores.

Conclusion
Although only indicative, lumbar infusion test results increased the likelihood of a positive shunt outcome. Pulse amplitude measures showed promising results that should be further explored in prospective studies.