A landmark randomized, placebo-controlled trial provides the clearest evidence to date that shunt surgery benefits patients with idiopathic normal-pressure hydrocephalus (iNPH), a condition long clouded by skepticism.

An international clinical trial has delivered the clearest evidence yet that brain shunt surgery can restore walking ability and reduce fall risk in older adults with iNPH, a condition long clouded by skepticism and controversy.

Investigators found that patients with the condition who received active shunts walked faster, steadier, and reported fewer falls compared with those whose shunts were set to a placebo mode.

“Although shunts have been used for 50 years, skepticism persisted because of the placebo effect and the risks of operating in frail, elderly patients,” Mark Luciano, MD, PhD, principal investigator and director of the Johns Hopkins Hydrocephalus and Cerebral Fluid Center, Baltimore, said in a statement.

“This study shows surgery is both effective and has an acceptable safety profile. We’ve proven this treatment works, definitively and safely, in the most rigorous type of study possible,” Luciano said.

The study was published online on September 16 in The New England Journal of Medicine.

Controversial Surgery

iNPH, first described in 1965, is caused by a buildup of cerebrospinal fluid (CSF) in the brain that leads to walking and balance difficulties, urinary incontinence, and cognitive decline. Left untreated, it can cause falls, loss of independence, and progressive disability.

Shunt surgery, in which CSF is diverted from the brain into the abdomen, has been used for decades, but its efficacy has been disputed. Some neurologists have even called for a moratorium on the procedure, citing inconsistent results, the risks of brain surgery in frail patients, and the possibility that improvements were largely due to the placebo effect.

The Placebo-Controlled Effectiveness in iNPH Shunting (PENS) trial sought to resolve that uncertainty. Conducted at 17 centers in the US, Canada, and Sweden, the study enrolled 99 patients (mean age, 75 years; 52% men) eligible for shunt surgery after showing short-term improvement in gait with temporary CSF drainage.

All patients underwent implantation of a noninvasively adjustable shunt — set to an open-shunt valve setting (opening pressure 110 mm of water) or a placebo valve setting (opening pressure > 400 mm of water which precluded drainage). Neither patients nor clinicians knew which group they were in.

Gait velocity at 3 months after surgery (the primary outcome) increased by an average of 0.23 m/sec in the open-shunt group and was essentially unchanged in the placebo shunt group (+0.03 m/sec).

A total of 37 of 46 patients with available data in the open-shunt group (80%) and 10 of 42 with available data in the placebo group (24%) had a change in gait velocity of at least 0.1 m/sec, which exceeded the threshold of substantial meaningful change of 0.10 m/sec in older adults, the investigators said.

Balance, measured by the Tinetti scale, also improved significantly more in the open-shunt group than in the placebo group at 3 months (2.9 points vs 0.5 points; P = .003). Patients with open shunts also had fewer falls (24% vs 46%; P = .03).

However, there were no significant differences at 3 months in the Montreal Cognitive Assessment scores or the Overactive Bladder Questionnaire.

“It has always been suspected that cognitive changes take longer to improve and need more detailed testing. At the 12-month period, we’ll have more to say not only about cognition, but how much patients are improving in their activities and independence,” Luciano said.

Shunt surgery was not without risks. Subdural hematoma was more common in the open-shunt group than in the placebo group (12% vs 2%), as was orthostatic headache (59% vs 28%), and with no difference in cerebral bleeding (2% in both groups).

The PENS trial strengthens the case for considering shunt surgery in appropriately selected older patients, the authors said.

“You’re suggesting surgery in people who are elderly. Of course, there is going to be increased risk. I have to feel like there’s an 80%-90% chance of an improvement and a very low risk before I would even suggest the surgery,” Luciano told Medscape Medical News.

Support With Caveats

Reached for a comment, James Liao, MD, PhD, neurologist at the Center for Neurological Restoration, Cleveland Clinic, Cleveland, who wasn’t involved in the study, noted that “even though shunting has been standard of care for years, the effectiveness of shunting in NPH [normal-pressure hydrocephalus] has not been proven in a randomized control trial until now.”

“NPH is an interesting condition where the diagnosis is largely based on whether symptoms respond to CSF removal. This process is prone to the placebo effect. The PENS study is the first large, randomized, double-blinded, placebo-controlled trial of shunting in NPH, and it shows that the benefit of shunting is significant and not due to placebo,” Liao told Medscape Medical News.

Liao cautioned, however, that “there is a great need for research to improve how we identify patients who would respond to shunting.”

Clifford Saper, MD, PhD, professor of neurology and neuroscience, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, also offered comment on the research.

Saper cautioned that there is no indication that the patients had an opening pressure of 180 mm of water or less at the time of initial lumbar puncture.

“This is critical as there are a small number of patients with communicating hydrocephalus who have high pressures and therefore do not have iNPH. Including even a small number of such patients in the ‘iNPH’ group would potentially bias the results toward a positive outcome (given that the actual improvements were so small),” he told Medscape Medical News.

Saper also noted that patients who were randomized to placebo did not have the shunt valve turned off but rather set to a relatively high setting. “You could argue whether the pressure of > 400 mm H₂O is sufficiently high as to constitute a real placebo,” he said.

Additionally, there was no improvement in either cognitive function or urinary continence, “the key reasons that PCPs [primary care physicians] refer a patient with large ventricles for evaluation of iNPH,” Saper said.

He added that he would consider the improvements in gait and balance to be “minimal, and even if statistically significant, probably not biologically significant enough to incur the risks associated with the surgery.”

In 2016, Saper wrote an editorial on shunt surgery for iNPH when he was an editor-in-chief of Annals of Neurology, pointing out the lack of placebo-controlled randomized trials and high rate of complications, and calling for a moratorium on the surgery until there was sufficient evidence from randomized controlled trials.

Saper said the PENS study and another one ongoing at the University of Washington, Seattle, are partially in response to his questioning the procedure.

“The current results are a signature advance given that no truly equivalent trials have been performed,” Allan Ropper, MD, with the Department of Neurology, Brigham and Woman’s Hospital, Boston, wrote in an accompanying editorial.

Ropper said he is looking forward to 12-month results to determine durability of the effect of shunting on gait and examine the effect on cognition. “Gait improvement could wane,” he noted.

The PENS trial was funded by the National Institute of Neurological Disorders and Stroke. Luciano and Saper reported having no relevant disclosures. Liao is a member of the Adult Hydrocephalus Clinical Research Network committee, as are many of the study’s authors. Ropper is an associate editor of The New England Journal of Medicine.