ARTICLE IN BRIEF
Experts debate the validity of a diagnosis of idiopathic normal pressure hydrocephalus in response to a call for a mortatorium on shunting procedures — at least until more randomized trials of the procedure are done.
A new review focused on idiopathic normal pressure hydrocephalus (iNPH) has generated debate among neurologists about the diagnosis of the disease and the benefits of the treatment. The review, which was published September 11 in the online edition of Annals of Neurology, was accompanied by an editorial by the editor-in-chief calling for a moratorium on all shunting procedures for iNPH.
Not everyone interviewed for this article agreed with the moratorium. But there was consensus on this point: wide-ranging studies are needed to determine which patients would receive the best results from shunting — and to follow those patients to make sure there are long-term benefits to the fluid removal.
The review was authored by 13 physicians and headed by Alberto J. Espay, MD, MSc, FAAN, who is the director, professor and endowed chair of the University of Cincinnati James J. and Joan A. Gardner Family Center for Parkinson's Disease and Movement Disorders. The authors of the review found that iNPH remains both over-suspected on clinical grounds and under-confirmed when based on immediate and sustained response to cerebrospinal fluid diversion.
THE POINT OF ‘CLINICAL EQUIPOISE’
The study authors reviewed articles on NPH published from January 1990 to December 2016 as well as electronic records from the University of Cincinnati to identify NPH referrals with first clinical encounters from 2005 to 2015.
The disease is diagnosed based on a triad of presentations, specifically gait issues, urinary incontinence, and cognitive difficulties. However, researchers found that the full triad was present in less than 60 percent of patients, and they noted that each individual component was too non-specific and could be caused by other diseases.
They also questioned post-shunt benefits and suggested that that hydrocephalic disorders appearing in late adulthood may often result from initially unapparent parenchymal abnormalities. In their review of the literature, they said they found “near universal lack” of blinding and controls, absence of specific clinical, imaging, or pathological features.
“Thus, we have reached the point of clinical equipoise justifying a randomized, controlled trial of VPS [ventriculoperitoneal shunting] in NPH, as advocated by over 75 percent of physicians including neurosurgeons who treat NPH in one recent survey,” the review concluded. “A reassessment of NPH as an entity will also require better characterization of the parenchymal and hydrodynamic contributions to gait, urinary, and cognitive symptoms in the pleomorphic syndrome of communicating hydrocephalus of adults, recognizing that NPH can commonly be a presenting feature of neurodegenerative disorders... rather than a risk factor for them.”
Dr. Espay said he started the review because he was humbled by patients with clinical stories that were inconsistent with “classic” NPH and turned out to be responsive to fluid drainage, as well as patients he thought would benefit long term, only to declare themselves with neurodegenerative disorders within a couple of years after shunt implantation.
“This prompted me to look at the literature and to question the idea that these patients had two diseases — NPH and, say, Alzheimer's disease. Rather, a large proportion of those we suspect as having iNPH may, in fact, represent a ‘hydrocephalic’ form of neurodegenerative disorders,” he said.
The accompanying editorial, written by Clifford B. Saper, MD, PhD, FAAN, editor-in-chief of the Annals of Neurology, said the definition of iNPH has become progressively blurred and that the sole defining characteristic of this “supposed disorder” has become the response to surgery.
“Fifty years is too long for our profession to go on with recommending surgery for a disease whose very existence has never been subjected to a critical test. To have this situation in neurology, which I otherwise perceive to be the most critical and academic of medical disciplines, is disturbing. As physicians, we need to clean our house,” wrote Dr. Saper.
That may be an example of evidence-based medicine being carried a little too far, said James B. Golomb, MD, an assistant clinical professor of neurology and director of the NYU Langone Adult Hydrocephalus Program. He said that iNPH does exist, and that anyone who has worked with the issue has seen patients who have a “dramatic unequivocal response to the shunt procedure that cannot be attributed to a placebo response.”
His concern is that patients are not doing well long-term. It may be due to shunts failing and becoming occluded. It may be due to other conditions worsening, or it may be that shunts are simply less effective even with they are revised.
Even the cause of the disease is a challenge to identify, Dr. Golomb said.
“It could be that what is causing patients with NPH to get worse is an underlying clinical decline,” said Dr. Golomb, “or a simple matter of CSF fluid dynamics that may be occurring that is yet to be identified.”
Gary M. Abrams, MD, FAAN, professor of clinical neurology and director of the Neurorehabilitation Clinic at University of California, San Francisco Medical Center, said idiopathic NPH has been a topic of controversy for 40 years, and that some question whether the disease itself is valid.
“I think the syndrome exists, there are people who have these constellation of symptoms who respond to cerebrospinal fluid pressure diversion on a short-term basis,” said Dr. Abrams. “Now the constellation may not always constitute a disease per se, but may be factors seen in other diseases.”
Dr. Abrams is particularly interested in gait mobility, and he said that about 75 percent of his patients respond to fluid removal, but it was difficult to know if there was long-term success after a year.
“You have to go in with your eyes wide open, because there is no one definitive test to determine who will respond and who won't,” he said, adding that the number of patients who should have shunts is “very minimal,” and that even with careful selection, maybe one-third will still have a good response after a year or two.
Michael A. Williams, MD, FAAN, is a professor of neurology and neurological surgery, and director of adult and transitional hydrocephalus at the University of Washington Medical Center in Seattle. He formerly established the Adult Hydrocephalus Center at Johns Hopkins University.
He said he had no doubt that NPH was real and noted that the Annals of Neurology paper showed persistent improvement for three years or more in one-third of the patients in their own series. He added that most patients do not have known risk factors.
“The word idiopathic,” he said, “does not mean imaginary when it comes to NPH; it simply means that we cannot identify a risk factor.”
Because iNPH typically appears in patients over the age of 70, Dr. Williams said that doctors should investigate the differential diagnosis for each of the main symptoms — urinary incontinence, dementia or gait problems — before pursuing tests, such as CSF removal.
“I would love it if there was an accurate noninvasive test,” he said. “However, the gaps in existing diagnostic paradigms mean that we need more research to understand the biology of iNPH, and not that we should call into question the very existence of the disorder or that we should stop treating it.”
Dr. Williams took issue with how shunts were portrayed in the paper, which stated that more than 40 percent of patients experience at least one post-ventriculoperitoneal shunting (VPS) complication including infection, shunt revision, or low-pressure headaches.
He pointed to the 2012 European multicenter study in which only 15 percent of patients experienced shunt complications that required surgery, including one with subdural hematoma, one infection, and 13 with catheter obstruction or valve malfunction, which are known and unpreventable complications of shunts in children as well as adults.
He added that the accompanying editorial calling for moratorium on all shunting procedures for iNPH is not supported by the literature.
Dr. Williams noted that a multicenter blinded, placebo-controlled study will soon be underway through the Adult Hydrocephalus Clinical Research Network, which is chaired by Mark Hamilton, MDCM, FRCSC, FAANS, a professor of neurosurgery at the University of Calgary. The trial will be coordinated by Johns Hopkins, and includes Seattle, Calgary, the Cleveland Clinic, and the University of Umeå in Sweden.
The University of Calgary Adult Hydrocephalus Clinic has maintained a prospective registry of patients since 2012 to determine the long-term effects of shunt surgery, Dr. Hamilton told Neurology Today. Similar to children with hydrocephalus, he said, the most common problem with shunts is obstruction, which can be repaired with surgery after which most patients recover.
“It is extremely important to understand the other disorders that can mimic iNPH, and in those with no other clear diagnosis, physicians should proceed to evaluate responsiveness to cerebrospinal fluid (CSF) removal with either a lumbar puncture or lumbar drain. With this approach, as outlined in the 2005 International Guidelines, patients selected for shunt surgery have a high probability of significant clinical improvement.”
“We perform about 75 lumbar drains and 25 lumbar punctures per year as part of this process,” Dr. Hamilton said, “and by identifying the patients with a positive response to CSF removal, we get a very high response to treatment with a shunt.”
“If we get a positive response with a lumbar drain but did not after insertion of a shunt, we don't assume it's a diagnosis failure, but we make sure the shunt is working, and if we ensure the shunt is working, patients get better,” Dr. Hamilton said. “Further, as we follow these patients the shunt treatment effects last. We have been following patients in our registry for five years and see the majority of the patients still getting a significant shunt response five years later.”
Dr. Hamilton said the randomized, blinded trial that is to start in early 2018 should sort out the concerns associated with the diagnosis and treatment of patients with suspected iNPH. Shunts will initially either be turned on or off, and then four months after surgery all shunts will be will on. This will allow the investigators to evaluate the efficacy of shunt treatment in patients with suspected iNPH.
Dr. Espay, the author of the Annals of Neurology study, said he aimed for a middle-of-the road focus, not a moratorium, since it does help some patients — even if for a brief period.
“While most patients with Alzheimer's disease will not benefit (or be harmed) from [the procedure] for even a short nine-month period, VPS may be warranted in selected patients for whom a response of any duration may be preferable to no response, assuming that patients and families are counseled as to the uncertainties of the long-term outlook and the risks of VPS-related adverse effects,” he said.