Normal Pressure Hydrocephalus

I. What every physician needs to know.

Normal pressure hydrocephalus (NPH) is a clinical diagnosis based on a chronic, progressive history of gait instability, subcortical dementia, and urinary incontinence accompanied by a confirmatory physical examination and characteristic imaging studies.

It is often considered in patients with a newly identified dementia or new gait instability, including falls. There is great interest in identifying these patients because some improve dramatically after a ventriculoperitoneal (VP) shunting procedure, making it a rare cause of reversible gait instability and/or dementia.

The pathophysiology of NPH is poorly understood but likely involves obstruction to CSF reabsorption over the cerebral convexities into the venous system. Thus, it is a communicating hydrocephalus; the aqueduct of Sylvius is patent. Recent studies have implicated microbleeds and cerebrovascular accidents as an etiology of NPH. In fact a large population based study noted an association between hypertension, diabetes mellitus and white matter lesions with clinical and radiologic evidence of NPH. Lack of adenosine or more precisely, downregulation of adenosine receptors, has also been discovered in a study on patients with idiopathic NPH. Given the protective effect of adenosine on vasculature and in modulation of inflammation, involvement of the adenosine system could be the missing link in discovering the pathophysiology of this disease. In addition, pro-inflammatory cytokine markers such as IL-1beta, IL-8, TGF-Beta, and TNF-alpha, have been noted to be high in the CSF of patients with NPH.

Continue Reading

Imaging shows enlarged lateral ventricles without cerebral atrophy; the presumed indolent nature of the disease results in enlarged ventricles with normal CSF pressures on lumbar puncture. This can be understood using Pascal’s law (force = pressure x area). Initially, the obstruction to reabsorption increases the ICP (pressure), and the ventricles expand, exerting pressure on the cortex, which is stretched in response (area). This chronic stretch results in a larger subcortical surface area and thus pressure normalizes in response to the constant CSF production (force) by the choroid plexus, resulting in enlarged ventricles and subcortical disruption but normal CSF pressure on lumbar puncture.

How this causes the clinical syndrome is similarly unclear; it is hypothesized that the stretching of the subcortex results in disruption of white matter tracts relevant for memory, executive function, postural stability, and regulation of urinary function. Several recent studies using magnetic resonance imaging (MRI) with diffusion-tensor imaging have supported this idea, identifying subtle white matter disease in the anterior frontal white matter.

It is clear from the literature that not all patients with NPH have shunt-responsive disease. Thus, the focus must be on identifying whether NPH is present, and then identifying whether or not the patient has shunt-responsive disease, since supportive treatment is all that is currently available to those who would not respond to shunting, and the procedure itself is associated with significant morbidity.

II. Diagnostic Confirmation: Are you sure your patient has Normal Pressure Hydrocephalus?

Dementia, urinary incontinence, and gait instability are each among the most common complaints of elderly patients and none are specific for NPH; thus, a complete history and examination looking for other causes of these symptoms is prudent.

As a dementia syndrome, it can be difficult to differentiate from Alzheimer’s Disease (AD), Lewy body dementia (LBD), or multi-infarct dementia. Clues that a given patient does not have Alzheimer’s Disease include prominent early gait disturbance (usually before prominent disturbances of memory), and lack of cortical atrophy on imaging. Imaging in early AD may not show cortical atrophy; in this case, MRI with specific imaging of the hippocampus may be useful which is usually normal in NPH but atrophic in AD. However, it must be noted that NPH and AD are not mutually exclusive diseases and studies have identified patients with clinical NPH who have neurofibrillary plaques and tangles consistent with AD on pathology.

In contrast to LBD, delirium or hallucinations are not usually found in NPH. Multi-infarct dementia will tend to have more microangiopathic changes on imaging and usually does not show hydrocephalus on imaging; additionally, the history of NPH is of gradual decline, rather than stepwise descent in function as might be seen in multi-infarct dementia.

Gait instability and incontinence merit consideration as independent entities as well; the patient may have spinal stenosis, peripheral neuropathy, arthritis, prostate enlargement or overactive bladder. It cannot be emphasized enough that these possibilities must be investigated since only symptoms relating to NPH have a chance of improvement with a shunting procedure.

An incidental finding of hydrocephalus on imaging may cause the clinician to question whether a patient has NPH; careful consideration should be given to whether there is any evidence of significant cortical atrophy (hydrocephalus ex vacuo) and whether the aqueduct of Sylvius is closed; the presence of either rules out the diagnosis.

A. History Part I: Pattern Recognition:

In the largest case studies to date, gait instability or falls is usually the reason for initial evaluation; dementia is an uncommon initial presentation and usually is clinically recognized later in the course. Depending on the case series, in fact, the classic triad of symptoms may in fact be uncommon and most patients will have two of the three, with gait instability and dementia being most specific. Asking about changes in gait, falls, and need for support when walking may prove fruitful.

Patients typically have symptoms of urinary frequency and urgency accompanying their incontinence. Some studies indicate that overactive bladder is seen in many cases and questioning along these lines to help differentiate prostatic disease from detrusor hyperreflexia helps differentiate the conditions.

Consider the diagnosis if the patient has symptoms of subcortical dementia on questioning, particularly evidence of impaired recall, difficulty with higher-level tasks (paying bills, organizing daytime activities), and psychomotor slowing. Impairment in attention is also common and history-taking may prove difficult without family members and/or caretakers present during the interview.

Again, it must be noted that on the largest pathology studies to date, up to 75% of patients with NPH also have plaques consistent with Alzheimer’s Disease and in such patients treatment may result in improved gait but not improved cognition. For example, in one study, brain biopsies were performed when VP shunts were inserted; presence of Alzheimer’s Disease-associated pathology was common and predicted poor outcome in response to shunting.

B. History Part 2: Prevalence:

Because of lack of standardized criteria for diagnosis, estimations of prevalence rates have varied. In series in Sweden and Germany, approximately 1-2 per 100,000 in the population underwent shunting procedures for NPH over a 5-year period. However, in Olmstead County over 4 years, no reversible cases were identified that presented with dementia, pointing to the possible temporal relationship between early recognition of the disease (when gait rather than dementia is present) and symptomatic improvement. There is no evidence currently of familial inheritance.

Some patients with NPH have a history of a disease which could cause scarring of the basilar meninges which might obstruct CSF flow. Such conditions include meningitis/encephalitis, subarachnoid hemorrhage, and head trauma – patients with these conditions have secondary hydrocephalus and are variably included in the published literature. Guidelines suggest they should not be considered to have true NPH.

C. History Part 3: Competing diagnoses that can mimic Normal Pressure Hydrocephalus.

Alzheimer’s disease – frequently present along with NPH; distinguishing characteristics are gait symptoms and degree of cortical impairment.

Lewy body dementia – as above, more frequently with Parkinsonism and delirium with hallucinations.

Multi-infarct dementia – the course may suggest stepwise rather than primary progressive disease and tends to have prominent dementia rather than abnormal gait.

Parkinson’s Disease – as with LBD though again patients with NPH can have presenting symptoms and signs that appear Parkinsonian, distinguishing characteristics are response to levodopa and less frequent cogwheel rigidity, bradykinesia, and more impairment of subcortical function in NPH.

Hydrocephalus due to obstruction (tumor, infection, congenital stricture of aqueduct of Sylvius) – the distinction is revealed on imaging.

Spinal stenosis – may be responsible for gait instability but not dementia; unlike NPH usually associated with significant pain and improvement with standing.

D. Physical Examination Findings.

Gait apraxia – apraxia signifies the inability to produce normal gait despite normal strength and sensation, abnormalities of either strength or sensation is not characteristic of NPH. The gait is typically shuffling, wide-based (in contrast to the narrow base of Parkinson’s Disease), slow, with a relatively normal arm-swing (again in contrast to PD). Some patients have the typical “magnetic” gait as if their feet were pulled by a magnetic field to the floor. Gait typically does not improve with levodopa, though there are patients who seem to have more overlap with Parkinson’s Disease and a Parkinsonian gait exam does not exclude of the diagnosis of NPH.

Urinary incontinence – patients may complain of pain or urgency with suprapubic palpation. A prostate exam to detect prostate enlargement in males is recommended.

Dementia – because the Mini-Mental State Exam was designed for cortical dementias, it may not be sensitive for a subcortical dementia such as NPH, though patients should have difficulty with recall and clock-drawing. For this reason, experts recommend using the HIV Dementia Scale as a quick screening tool. The presence of aphasia or agnosia suggests cortical pathology not consistent with NPH.

E. What diagnostic tests should be performed?

There are no physical findings that confirm the diagnosis. The diagnosis of NPH relies on a combination of symptoms, physical exam findings, and imaging studies in a patient over the age of 40 with a slowly progressive course.

1. What laboratory studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?

Lumbar puncture is indicated to rule out other causes of increased intracranial pressure; caution should be taken if there are any signs of midline shift or abnormalities in the gray or white matter which would be unusual for NPH and could raise the risk of herniation. CSF examination is notable for an opening pressure in the high-normal range (for this reason, LP must be performed with patient in the decubitus position rather than sitting up) but normal glucose, protein, and cell count studies.

The American Academy on Neurology recommends counseling patient with NPH and presence of elevated CSF pressure gradient regarding the increased chances of responding to shunts which is not negated by age.

Large-volume lumbar puncture has been advocated by some as a means to clinch a diagnosis of NPH and to evaluate shunt responsiveness. 40 to 50mL of CSF are removed and the patient is examined before, immediately after, and 4-6 hours after. Improvement in neurologic exam (particularly gait) is consistent with shunt-responsive NPH. However, the literature indicates this test has low sensitivity (a negative test does not rule out shunt-responsive hydrocephalus). Some authors advocate using continuous drainage extended lumbar cerebrospinal fluid drainage; studies indicate a positive response is associated with a higher positive predictive value of shunt-responsive NPH. Newer studies have used better markers of gait improvement such computerized quantitative gait analysis as a better predictor of improvement. This eliminates the subjective error from patient reporting.

Extensive evaluation of CSF and brain tissue in autopsies has revealed a population of NPH with concurrent Alzheimers disease. Tau proteins particular to Alzheimers were found in such cases and were associated with lack of improvement and often worse outcomes after shunting. In a large study, a majority of these patients with NPH and AD both, went on to experience no relief with shunting despite improvement after HVLP. Often symptoms actually worsened after shunting.

2. What imaging studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?

Magnetic resonance imaging (MRI) of the brain is the definitive diagnostic imaging test; computed tomography (CT) is unable to examine the aqueduct of Sylvius in full and does not image the hippocampus well. Characteristic findings include rounding or bowing of the lateral and third ventricles, thinning of the corpus callosum, and a flow void in the aqueduct of Sylvius. Periventricular hyperintensities can be seen in NPH or in microvascular disease. Evidence of cortical or hippocampal atrophy is not consistent with the diagnosis and degenerative diseases should be considered instead. Neither study requires IV contrast.

One group has evaluated MRI characteristics that predict shunt responsiveness and has found tight high-convexity and medial subarachnoid spaces and enlarged Sylvian fissures with ventriculomegaly has prognostic importance. In a study comparing MRIs of patients with NPH, patients with NPH and concurrent AD, and patients with AD alone, the z-axial expansion of the lateral ventricle and compression of the brain just above the ventricles were distinctive features for NPH.

See Figure 1.

Figure 1.

A. Sagittal T1-weighted MR image demonstrates dilatation of the lateral ventricle and stretching of the corpus callosum (arrows), depression of the floor of the third ventricle (single arrowhead), and enlargement of the aqueduct (double arrowheads). Note the diffuse dilatation of the lateral, third, and fourth ventricles with a patent aqueduct, typical of communicating hydrocephalus. B. Axial T2-weighted MR images demonstrate dilatation of the lateral ventricles. This patient underwent successful ventriculoperitoneal shunting. Source: Fauci AS, Kasper DL, Braunwald E, Hauser SL, Longo DL, Jameson JL, Loscalzo J. Harrison’s Principles of Internal Medicine: 17th Edition.

F. Over-utilized or “wasted” diagnostic tests associated with this diagnosis.

Measuring CSF pressure by any alternative means than LP is unlikely to change management. CT scanning is supportive of the diagnosis but if the patient can tolerate an MRI, more information that can confirm or reject the diagnosis is gained.

III. Default Management.

The only therapy proven to improve symptoms in selected patients is ventriculoperitoneal shunting. Otherwise, management is supportive. A neurosurgery service with experience with shunting procedures should be consulted towards this end. The only elements shown by multivariate analysis to be related to better outcomes were shorter duration of symptoms and gait impairment as the primary symptom.

A. Immediate management.

Given the chronic nature of the disease, intervention is not time-limited. The most immediate management is assuring that the patient is safe from falls.

B. Physical Examination Tips to Guide Management.

In addition to monitoring for clinical improvement (monitoring gait and mental status testing daily), the hospitalist should monitor for complications of a shunting procedure if undertaken. These include subdural hematoma and shunt infection. While both are likely to result in alteration in mental status (including stupor, confusion, and coma), a subdural hematoma may have more focal findings, while shunt infection will usually be accompanied by fever and leukocytosis.

Improvement on gait is best measured using a standardized timed walking test to measure distance patient can cover in a discrete period (such as 6-minute walk test) of time or time required to cover a certain distance. More recent studies recommend a quantitative gait assessment using a computerized analysis.

Careful daily neurologic examination cannot be underestimated in importance as response to therapy and complications related to therapy will only be able to be identified via clinical testing.

C. Laboratory Tests to Monitor Response To, and Adjustments in, Management.

Response to management is measured clinically. A CBC to monitor for elevations in white blood cell count related to shunt infection as well as monitoring of electrolytes, particularly sodium, in response to changing intracranial pressures can be useful though is not required.

D. Long-term management.

Unfortunately, as there is no gold standard for identifying patients with NPH, there is also no gold standard which identifies those with the disease that benefit from shunting to relieve the hydrocephalus. Case series vary on response to shunting depending on how patients are selected for the procedure, but generally find between 1/3 and 2/3 of patients have objective improvement in symptoms. Gait tends to improve more than memory; despite being commonly known as a “reversible dementia,” studies are mixed with respect to whether patients substantially improve their cognition after shunting.

Some series suggest that patients gradually improve over a longer time than initially thought (up to 75% response when tested at 18 months), though other authors believe transient improvements are the norm. Surveys of patients who have undergone shunt placement, although prone to responder bias, show substantial improvements in quality of life. Patient with NPH and concurrent Alzheimers had poorer outcomes with shunting despite response after high volume CSF tap.

Around 1/3 of patients in series have complications of the shunt requiring revision and 2-6% unfortunately have clinically significant subdural hematomas as a result.

Patients who have not improved at all by 6 months after shunting deserve re-evaluation of potential alternative diagnoses. Re-imaging for ventricular size is not useful as patients with dramatic clinical responses may have no change in imaging. If two of the classic three symptoms are improving but the third does not, an investigation should be undertaken for other causes of the remaining symptom.

Two types of shunts are used: programmable (variable resistance, can change amount of flow using external magnets), and fixed (requires surgical replacement if different degree of flow needed). The former are susceptible to magnetic fields, such as MRI scanners, and must be seen after MRI to recalibrate the shunt. The latter are not susceptible to magnetic fields. Gradual reductions in resistance were associated with better clinical outcomes than fixed low resistance shunts. Both types had the same rate of complications.

A certain group of patients with NPH, who also have concurrent Parkinson’s disease may benefit from dopamine therapy.

Ventriculoatrial shunting has also been evaluated in some small studies and found to be just as safe as ventriculoperitoneal shunting, but more studies are needed.

Endoscopic third ventriculostomy does not have enough evidence to support its clinical use.

E. Common Pitfalls and Side-Effects of Management.

Around 1/3 of patients in series have complications of the shunt requiring revision and 2-6% unfortunately have symptomatic subdural hematomas as a result. The most common side effects of the shunting procedure are over-drainage (usually presenting with postural hypotension, positional headache, or when severe with focal signs and symptoms of hematoma), or under-drainage (usually as a result of obstruction to flow). The latter is the most common complication of shunts, occurring in up to 1/3 of patients. It usually presents with return of NPH symptoms (if there was improvement after procedure) or lack of any improvement since time of shunt procedure despite maximizing flow (which should prompt a search for other causes as outlined above once the shunt is verified to be working normally).

If there is suspicion of obstruction to shunt flow, plain radiographs should be obtained of the proximal catheter, valve mechanism, and peritoneal distal end to look for disconnections. If none are found, the next step is usually a nuclear medicine shunt patency study, which should show brisk flow through the catheter and dispersal all over the peritoneum; restriction of diffuse flow in the peritoneum can identify obstruction around the catheter tip.

If the shunt is functioning normally but the patient is declining, a search must be undertaken for other causes given the age and usual comorbidity of such patients. An analogy can be made to the demented Alzheimer’s patient who presents with worsening altered mental status in which toxic, metabolic, infectious, and other causes should be investigated.

There is good evidence that can be extrapolated from pediatric shunting that antibiotic-impregnated shunts may decrease the rate of shunt-related infections.

IV. Management with Co-Morbidities.

A. Renal Insufficiency.

No change in standard management.

B. Liver Insufficiency.

No change in standard management.

C. Systolic and Diastolic Heart Failure.

The patient will need careful attention to volume status prior to undergoing VP shunt placement.

D. Coronary Artery Disease or Peripheral Vascular Disease.

The patient should have a preoperative history and physical to determine readiness for anesthesia for shunt placement.

E. Diabetes or other Endocrine issues.

No change to standard management.

F. Malignancy.

Patients with end-stage malignant disease may not wish to undergo the possible morbidity of shunt placement; studies have shown patients with malignant disease have more complications from shunt placement and require more revisions than patients without.

G. Immunosuppression (HIV, chronic steroids, etc).

No change to standard management.

H. Primary Lung Disease (COPD, Asthma, ILD).

No change to standard management other than assessment of readiness for the operating room.

I. Gastrointestinal or Nutrition Issues.

No change to standard management.

J. Hematologic or Coagulation Issues.

Shunt placement requires intact coagulation because of the risk of hematoma surrounding placement. Coagulopathy and thrombocytopenia should be corrected to the extent possible prior to shunt placement. Patients who are unable to be corrected may sustain intolerable morbidity and mortality from intracranial hemorrhage.

K. Dementia or Psychiatric Illness/Treatment.

Management as above, with the understanding that pre-existing dementia or psychiatric disease not related to NPH will not improve after shunt placement.

V. Transitions of Care.

A. Sign-out considerations While Hospitalized.

These patients must have an updated complete neurologic exam documented on signout for comparison by the covering hospitalist or team since hemorrhage, infection, and shunt obstruction may present acutely and require urgent decision-making using the clinical exam as a guide.

B. Anticipated Length of Stay.

Once a diagnosis is made, the shunt procedure should be completed as an inpatient given need for frequent initial monitoring. Most patients can be safely discharged within 48 hours of shunt placement if uncomplicated.

C. When is the Patient Ready for Discharge.

Stable neurologic exam after shunt placement without evidence of infection is satisfactory for safe discharge.

D. Arranging for Clinic Follow-up.

1. When should clinic follow up be arranged and with whom.

The patient should be seen shortly in the neurosurgery clinic for re-evaluation of shunt function. Many patients benefit from being followed by a neurologist as well who can perform objective serial exams to monitor improvement (or lack thereof) after shunt placement.

2. What tests should be conducted prior to discharge to enable best clinic first visit.


3. What tests should be ordered as an outpatient prior to, or on the day of, the clinic visit.

If the patient is clinically worsening or has not improved after 6 months, plain films of the shunt are indicated, as is a nuclear medicine study to evaluate flow along the shunt. Otherwise, no tests are necessary.

E. Placement Considerations.

Studies show the vast majority of patients require home health support with VNA.

F. Prognosis and Patient Counseling.

Extensive counseling needs to be undertaken once diagnosis is confirmed due to the risks and invasive nature of the treatment for people with shunt-responsive disease. For patients with NPH and concurrent Alzheimers, the prognosis is much poorer even with evidence of shunt responsive disease.

They should be counseled that in the best series, while 1/3 to 2/3 of patients show objective improvement in at least one symptom of NPH (and some show dramatic improvement), 1/3 of patients also have complications of shunt placement requiring revision, and a substantially smaller percentage have severe complications resulting in permanent neurologic damage or death.

VI. Patient Safety and Quality Measures.

A. Core Indicator Standards and Documentation.

There are no JCAHO core indicators for this disease.

B. Appropriate Prophylaxis and Other Measures to Prevent Readmission.

Regular visits with a primary care physician and ensuring that the patient is monitored continuously at home can help to identify shunt malfunction early in the course.

Yamada, S, Ishikawa, M, Yamamoto, K. “Comparison of CSF Distribution between Idiopathic Normal Pressure Hydrocephalus and Alzheimer Disease”. . 2016 Feb 25.

Mihalj, M, Dolić, K, Kolić, K, Ledenko, V. “CSF tap test – Obsolete or appropriate test for predicting shunt responsiveness? A systemic review”. . vol. 362. 2016 Mar 15. pp. 78-84.

Moussa, WM, Mohamed, MA. “Efficacy of postoperative antibiotic injection in and around ventriculoperitoneal shunt in reduction of shunt infection: A randomized controlled trial”. . vol. 143. 2016 Apr. pp. 144-9.

Liu, A, Sankey, EW, Jusué-Torres, I, Patel, MA, Elder, BD, Goodwin, CR, Hoffberger, J, Lu, J, Rigamonti, D. “Clinical outcomes after ventriculoatrial shunting for idiopathic normal pressure hydrocephalus”. . vol. 143. 2016 Apr. pp. 34-8.

Broggi, M, Redaelli, V, Tringali, G, Restelli, F, Romito, L, Schiavolin, S, Tagliavini, F, Broggi, G. “Normal pressure hydrocephalus and parkinsonism: preliminary data on neurosurgical and neurological treatment”. . 2016 Mar 9.

Schniepp, R, Trabold, R, Romagna, A, Akrami, F, Hesselbarth, K, Wuehr, M, Peraud, A, Brandt, T, Dieterich, M, Jahn, K. “Walking assessment after lumbar puncture in normal-pressure hydrocephalus: a delayed improvement over 3 days”. . vol. 18. 2016 Mar. pp. 1-10.

Casati, M, Arosio, B, Gussago, C, Ferri, E, Magni, L, Assolari, L, Scortichini, V, Nani, C, Rossi, PD, Mari, D. “Down-regulation of adenosine A1 and A2A receptors in peripheral cells from idiopathic normal-pressure hydrocephalus patients”. . vol. 361. 2016 Feb 15. pp. 196-9

Halperin, JJ, Kurlan, R, Schwalb, JM, Cusimano, MD, Gronseth, G, Gloss, D. “Practice guideline: Idiopathic normal pressure hydrocephalus: Response to shunting and predictors of response: Report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology”. . vol. 85. 2015 Dec 8. pp. 2063-71.

Pomeraniec, IJ, Bond, AE, Lopes, MB, Jane, JA. “Concurrent Alzheimer's pathology in patients with clinical normal pressure hydrocephalus: correlation of high-volume lumbar puncture results, cortical brain biopsies, and outcomes”. . vol. 124. 2016 Feb. pp. 382-8.

Jaraj, D, Agerskov, S, Rabiei, K, Marlow, T, Jensen, C, Guo, X, Kern, S, Wikkelsø, C, Skoog, I. “Vascular factors in suspected normal pressure hydrocephalus: A population-based study”. . vol. 86. 2016 Feb 16. pp. 592-9.

Sæhle, T, Farahmand, D, Eide, PK, Tisell, M, Wikkelsö, C. “A randomized controlled dual-center trial on shunt complications in idiopathic normal-pressure hydrocephalus treated with gradually reduced or "fixed" pressure valve settings”. . vol. 121. 2014 Nov. pp. 1257-63.

Lim, TS, Choi, JY, Park, SA, Youn, YC, Lee, HY, Kim, BG, Joo, IS, Huh, K, Moon, SY. “Evaluation of coexistence of Alzheimer's disease in idiopathic normal pressure hydrocephalus using ELISA analyses for CSF biomarkers”. . vol. 14. 2014 Apr 1. pp. 66

Jump to Section