How NPH Shunt Surgery Works and What to Expect

An NPH shunt is a surgically implanted device that drains excess cerebrospinal fluid (CSF) from the brain’s ventricles to another body cavity, most often the abdominal cavity, to relieve the symptoms of normal pressure hydrocephalus. NPH is a condition in which the brain’s fluid-filled chambers enlarge and press on surrounding tissue even though CSF pressure readings look roughly normal on a standard lumbar puncture. The shunt remains the primary treatment, and studies consistently show it improves the hallmark symptoms of walking difficulty, cognitive decline, and urinary incontinence in a majority of patients. But deciding who will benefit, choosing the right hardware, and managing the shunt over years involves more nuance than the basic description suggests.

What Normal Pressure Hydrocephalus Actually Looks Like

NPH is sometimes called a treatable cause of dementia, which is what makes it worth recognizing. The classic presentation is a triad of symptoms: a distinctive shuffling gait, cognitive deterioration, and urinary incontinence. In typical cases, walking problems appear first, followed by mental slowing and then bladder control issues, all alongside enlarged ventricles visible on a CT or MRI scan.¹ The condition mostly affects older adults, and because its symptoms overlap heavily with Alzheimer’s disease, Parkinson’s disease, and ordinary aging, it gets missed frequently.

Not every patient walks in with the full triad. In one study of 200 patients referred for evaluation, about 84% had all three symptoms, but only about half of those turned out to actually have NPH rather than brain atrophy or another condition. Gait disturbance was the most reliable distinguishing sign, while dementia alone did not reliably separate NPH from other causes. Urinary incontinence tended to show up later in the disease.² The practical takeaway: if an older person develops a progressive walking problem that looks “magnetic” (feet seem stuck to the floor, short shuffling steps, wide stance) and then starts showing cognitive or bladder problems, NPH belongs on the list of possibilities.

How Doctors Decide Whether a Shunt Will Help

This is the trickiest part of the whole NPH picture. Imaging can show enlarged ventricles, but enlarged ventricles also occur in brain atrophy from Alzheimer’s and other neurodegenerative diseases. Clinicians use several imaging markers to tell the two apart, including the Evans index (a ratio measuring ventricle width), the callosal angle (measuring how compressed a specific brain structure is), and the DESH pattern, which describes a characteristic mismatch between tight brain sulci at the top of the head and widened spaces lower down.³ These imaging features help with diagnosis, but on their own they do not reliably predict who will improve after shunt surgery. One study found no significant difference in preoperative DESH scores, Evans index, or callosal angle between patients who improved after surgery and those who did not.⁴ Another study did find a correlation between higher DESH scores and favorable surgical response, so the picture is not fully settled.⁵

Because imaging alone is imperfect, most centers rely on a CSF removal test to help predict shunt response. The simplest version is the tap test: a standard lumbar puncture removes a volume of CSF, and clinicians measure whether the patient’s gait improves afterward. A systematic review found the tap test has a very high positive predictive value (around 92%), meaning that if someone improves after the tap, they are very likely to improve with a permanent shunt. The catch is its low negative predictive value (around 37%), so many patients who fail the tap test would still benefit from a shunt.⁶ One interesting finding: the amount of CSF removed during the tap does not appear to matter much. A study examining the relationship between fluid volume drained and gait improvement found no significant correlation.⁷

For patients who test negative on the tap test but still look clinically suspicious, an extended lumbar drain (ELD) offers a more thorough assessment. A catheter is placed in the spinal canal and drains CSF continuously over several days. This test predicted shunt responsiveness in about 96% of patients in one study and was reasonably accurate at distinguishing future shunt responders from non-responders.⁸ Combining imaging markers with the tap test also improves prediction. A multicenter study found that adding the callosal angle and DESH sign to tap test results pushed sensitivity up to 85% and improved the overall accuracy of predicting who would respond to shunt surgery.⁹

Types of Shunts Used for NPH

The most common type is the ventriculoperitoneal (VP) shunt, which runs a thin catheter from a ventricle in the brain, under the skin behind the ear, down the neck and chest, and into the abdominal cavity where excess CSF is reabsorbed. The alternative is a lumboperitoneal (LP) shunt, which draws CSF from the spinal canal instead of directly from the brain. A systematic review comparing the two found that both improve symptoms to a similar degree. VP shunts tend to provide faster symptom relief but carry a higher risk of infection. LP shunts have a safer immediate profile but need more frequent revision surgery because of mechanical problems like catheter kinking or migration.¹⁰ A comparative study confirmed the trade-off: LP shunts avoid the risk of brain hemorrhage from the ventricular puncture that VP placement requires, but they had a higher rate of reoperation.¹¹ In practice, VP shunts remain the default in most centers, with LP shunts offered when patient anatomy or surgical considerations favor the lumbar approach.

Why Programmable Valves Changed the Game

Every shunt has a valve that controls how much CSF it drains. The earliest shunts used fixed-pressure valves set at a single drainage rate during manufacturing. The problem is that the ideal drainage rate for any given patient is hard to predict in advance, and it can change over time. Drain too much and the ventricles shrink too fast, potentially tearing bridging veins and causing subdural fluid collections or bleeding. Drain too little and symptoms persist.

Programmable valves can be adjusted after surgery using an external magnetic device held against the scalp. This turned out to be a substantial improvement. In one cohort study, patients with programmable valves had a complication rate roughly five times lower than those with fixed-pressure valves (about 10% versus 53%).¹² A larger study found that programmable valves cut overall revision rates roughly in half (about 13% versus 24%) and virtually eliminated the need for multiple revision surgeries.¹³

Overdrainage-related complications remain the most common problem with any shunt, occurring in roughly a third of patients in one cohort. But in patients who had programmable valves, overdrainage problems like subdural fluid collections could be managed simply by dialing the valve to a higher resistance setting, avoiding reoperation entirely. Among patients with fixed valves, many had to undergo a second surgery to swap in a programmable valve.¹⁴ Some shunt systems also include anti-siphon devices, designed to prevent excess drainage when the patient stands up. However, at least one study found that anti-siphon devices did not significantly reduce overdrainage complications in practice.¹⁵ A large registry study of over 1,400 shunted NPH patients confirmed that anti-siphon devices did not prevent subdural hematoma.¹⁶

What Improvement Looks Like After Surgery

Walking is the symptom most likely to improve and the one that improves most quickly. In one study of 90 patients, median walking speed increased by 25% within months of surgery, and nearly 80% of patients took fewer steps to cover the same distance.¹⁷ Another study tracking gait parameters over time found that improvements appeared within the first three months and plateaued at around six months.¹⁸ A case series of 116 patients found that VP shunt placement was well tolerated and associated with improved outcomes in at least the short term, with a shorter duration of gait problems before surgery predicting greater improvement.¹⁹

Long-term durability is the more complicated story. Gait tends to hold up best over time. A study following patients for up to seven years found that walking improvement was maintained in about 87% of patients at last follow-up.²⁰ Cognitive gains and bladder control improvements, however, are more fragile. Another long-term study confirmed that cognition and urinary incontinence improved in the early period after surgery but showed declines during extended follow-up, while gait improvement remained more sustained.²¹ This pattern makes sense intuitively: if some of the cognitive decline is from coexisting Alzheimer’s pathology (which is common in the same age group), a shunt can relieve the hydrocephalus component but cannot reverse a separate degenerative process. The earlier in the disease course someone is diagnosed, the better their odds. Patients shunted at an early stage reported symptom improvement about 65% of the time versus 50% for those treated at a later stage.²²

Complications and Revision Surgery

Shunts are mechanical devices living inside the body, and like all such devices, they can fail. A multicenter prospective study recorded an overall complication rate of about 9% and a reoperation rate of about the same. The most common complication was subdural hematoma from overdrainage, occurring in about 5% of patients. Most of those resolved by adjusting the programmable valve to its maximum pressure setting. Shunt malfunction requiring surgical revision occurred in about 9% of cases, split roughly equally between valve failure and problems at the peritoneal end of the catheter.²³ A single-institution study tracking long-term revisions found an overall revision rate of about 11%, with the vast majority (about 86%) due to mechanical malfunction rather than infection.²⁴

Subdural hematoma after shunting deserves special mention because it is the complication most unique to this surgery. When the ventricles decompress, the brain shifts slightly, and bridging veins on the brain’s surface can stretch and tear. Treatment depends on severity. In a case series, two patients improved just from having their programmable valve reset to maximum pressure, while six others needed surgical drainage of the blood collection. Most of those who needed drainage had also first undergone valve adjustment before the decision was made to operate.²⁵ The registry study of over 1,400 patients found that lower initial valve settings were associated with higher subdural hematoma risk, reinforcing the strategy of starting conservatively and adjusting downward.²⁶

Endoscopic Third Ventriculostomy as an Alternative

Not every NPH patient needs hardware implanted permanently. Endoscopic third ventriculostomy (ETV) is a procedure where a small hole is made in the floor of the third ventricle, creating a new pathway for CSF to drain internally without any implanted device. This eliminates shunt-related complications entirely. A meta-analysis comparing ETV to VP shunting found no significant difference in symptom improvement rates (about 68% for ETV versus 73% for VP shunts) and no significant difference in failure rates. ETV had a dramatically lower complication rate, at about 8% compared to roughly 51% for VP shunts.²⁷

That sounds like a clear win for ETV, but the evidence remains limited. A Cochrane review examining the question found only one randomized trial to analyze. The results were too imprecise to determine whether there was any real difference between the two approaches at three or twelve months.²⁸ ETV also does not work for every patient; its success depends on the specific anatomy of the obstruction, and some NPH cases involve CSF absorption problems rather than flow obstruction, which ETV cannot fix. Most neurosurgeons still consider VP shunting the standard treatment, but ETV is gaining traction as an option worth discussing, particularly for patients at higher risk for implant-related complications.

Improving Surgical Accuracy with Navigation

Placing the ventricular catheter accurately matters more than you might expect. The catheter tip ideally sits in a specific location near the foramen of Monro, the opening between the brain’s ventricles. Misplacement can lead to catheter obstruction, reduced shunt longevity, and the need for early revision. Traditionally, surgeons place the catheter using anatomical landmarks alone, but electromagnetic (EM) navigation systems are changing that. One study found that EM-guided placement achieved optimal catheter position (grade 1) in 92% of patients compared to 71% with freehand placement, and the catheter tip ended up significantly closer to its target.²⁹ Another study concluded that EM-guided navigation reduces the proximal-end failures caused by malpositioning, obstruction, and infection.³⁰ These systems add time and cost to the procedure, so they are not yet universal, but the data supporting their use is growing.

Rehabilitation After Shunt Placement

A shunt addresses the mechanical problem of excess fluid, but the months or years of impaired walking before surgery often leave patients deconditioned, with weak muscles and poor balance. Physical rehabilitation after shunt placement is an area where the research is still catching up. A randomized trial of a structured high-intensity exercise program after shunt surgery did not find a clear additional benefit on an overall NPH symptom scale, possibly because so many participants dropped out. However, the exercise group did show long-term improvements in balance and higher achievement of personal goals.³¹

A scoping review found that one specific approach stood out: dual-task exercise with gait training, which combines walking practice with a simultaneous cognitive or motor task. The group receiving this type of training showed a large improvement in a functional gait assessment and, strikingly, saw fall incidence drop from about 73% to 10% at follow-up, compared to more modest reductions in groups receiving standard exercise or no exercise.³² Falls are one of the biggest practical dangers for NPH patients even after shunting, so targeted rehabilitation may be just as important as the surgery itself in terms of real-world outcomes.

Is Shunt Surgery Cost-Effective?

Given that NPH primarily affects older adults who are already likely to need substantial healthcare and social support, the economics of shunt surgery matter. A cost-utility analysis estimated that shunt treatment for idiopathic NPH gained patients about 2.2 additional life years and 1.7 quality-adjusted life years, at a cost of roughly €7,500 per quality-adjusted life year. That figure is well below the threshold most healthcare systems use to define a cost-effective treatment.³³ Another analysis based on Japanese trial data found that the investment shows a positive return in as soon as 18 months for VP shunts, primarily because improved function reduces caregiving and institutional care costs.³⁴ The implication is that underdiagnosis of NPH is not just a clinical problem but a financial one: patients who could regain independence are instead absorbing the costs of progressive disability.

Telemetric Monitoring and the Future of Shunt Management

One of the persistent frustrations of shunt management is that there is no easy way to know whether it is working properly at any given moment. If a patient’s symptoms return, it could be shunt failure, disease progression, or an unrelated problem. Traditionally, checking requires imaging and sometimes an invasive pressure measurement. Telemetric intracranial pressure sensors are a newer technology that lets clinicians check CSF pressure noninvasively through the skin. A systematic review identified two main devices in use since 2009 and 2015, respectively, and noted growing interest in the field. These sensors allow continuous or repeated pressure monitoring without a hospital visit, which could speed up the detection of shunt malfunction and allow more precise valve adjustments.³⁵ The technology is still relatively early in adoption, but the direction is clear: shunt management is moving from a reactive model (wait for symptoms, then investigate) toward proactive monitoring that catches problems before they cause setbacks.