Objectives The diagnosis of shunt malfunction is often not straightforward. We have explored, in symptomatic shunted patients with hydrocephalus or pseudotumour cerebri syndrome (PTCS), the accuracy of CSF infusion tests in differentiating a functioning shunt from one with possible problems, and the health economic consequences. Methods Participants: hydrocephalus/PTCS patients with infusion tests performed from January 2013 until December 2015. We followed patients up after 6 and 12 months from the test to determine whether they had improved, had persisting symptoms or had required urgent revision. We calculated the total cost savings of revision versus infusion tests and standard protocol of revision and ICP monitoring versus infusion tests. Results Three hundred sixty-five shunt infusion tests had been performed where a shunt prechamber/reservoir was present. For hydrocephalus patients, more than half of the tests (~55%, 155 out of 280) showed no shunt malfunction versus 125 with possible malfunction (ages 4 months to 90 years old). For PTCS patients aged 10 to 77 years old, 47 had possible problems and 38 no indication for shunt malfunction. Overall, > 290 unnecessary revisions were avoided over 3 years' time. Two hundred fiftyeight (> 85%) of those non-surgically managed, remained well, did not deteriorate and did not require surgery. No infections were associated with infusion studies. For Cambridge, the overall savings from avoiding revisions was £945,415 annually. Conclusions Our results provide evidence of the importance of shunt testing in vivo to confirm shunt malfunction. Avoiding unnecessary shunt revisions carries a strong health benefit for patients that also translates to a significant financial benefit for the National Health Service and potentially for other healthcare systems worldwide.