How to Diagnose Normal Pressure Hydrocephalus (NPH)
Begin with MRI brain without IV contrast as the first-line imaging study, looking specifically for ventriculomegaly with an Evans index >0.3, and confirm the diagnosis with a CSF tap test showing normal opening pressure (<250 mm H2O) combined with clinical improvement after CSF removal. 1
Clinical Recognition: The Classic Triad
NPH presents with three cardinal features that typically evolve in sequence:
- Gait disturbance (appears first): Look for magnetic, shuffling gait with wide base and difficulty initiating steps
- Urinary urgency/incontinence (appears second): Detrusor overactivity, not stress incontinence
- Cognitive decline (appears last): Frontal-subcortical pattern with psychomotor slowing, executive dysfunction, and apathy—not the memory-predominant pattern of Alzheimer's disease
The symptoms develop slowly and progressively, which often delays presentation. Patients typically present late in the disease course. 1
Initial Imaging: MRI vs CT
MRI brain without IV contrast is preferred over CT, though both are acceptable first-line options. 1, 2
Key MRI Findings for Probable NPH (all required):
Ventriculomegaly not attributable to atrophy or congenital enlargement
- Evans index >0.3 (maximal width of frontal horns ÷ maximal width of inner skull)
No macroscopic CSF flow obstruction
At least one of the following:
- Enlarged temporal horns
- Callosal angle <90 degrees (narrowed posterior callosal angle)
- Periventricular white matter changes (altered brain water content)
- Aqueductal or fourth ventricle flow void on MRI 2
Additional Supportive MRI Features:
- Effaced sulci along high convexities with widened Sylvian fissures
- Increased CSF flow void through cerebral aqueduct (correlates with good shunt response)
- Cine MRI showing hyperdynamic aqueductal CSF flow 2
CT Findings (if MRI unavailable):
CT can identify ventriculomegaly, narrowed posterior callosal angle, effaced high convexity sulci, widened Sylvian fissures, and periventricular hypoattenuation—but has lower sensitivity than MRI and cannot detect aqueductal flow voids or small obstructing lesions. 1
Confirmatory Testing: CSF Analysis
Lumbar puncture with opening pressure measurement and CSF tap test is essential for diagnosis. 1
What to Measure:
- Opening pressure: Should be normal (typically <250 mm H2O) or only mildly elevated
- CSF composition: Normal cell count, protein, and glucose (to exclude other causes)
- Volume removal: Remove 40-50 mL of CSF
Assessing Response to Tap Test:
Perform standardized assessments before and after CSF removal (ideally within 2-4 hours, and again at 24-48 hours):
- Gait testing: Timed 10-meter walk, number of steps, Timed Up and Go test
- Cognitive testing: Trail Making Test, Mini-Mental State Examination
- Urinary symptoms: Frequency, urgency episodes
A positive tap test (clinical improvement after CSF removal) has high positive predictive value for shunt response. However, a negative tap test does NOT exclude NPH—some patients still respond to shunting. 3, 4
When Tap Test is Equivocal or Negative
If imaging supports NPH but the single tap test shows no improvement, consider:
Repeat tap test 2-3 times on consecutive days (improves predictive value) 5
Extended lumbar drainage (3-4 days with external drain, removing 10-15 mL/hour): This is more invasive but has better sensitivity for predicting shunt response 6, 5
Critical Diagnostic Pitfalls
Comorbid Neurodegenerative Disease
75% of patients with NPH severe enough to require treatment have coexisting Alzheimer's disease, vascular dementia, or other neurodegenerative disorders. 6 This is detected in 20-57% of NPH patients. 1
Key distinguishing features:
- NPH: Gait disturbance predominates early; frontal-subcortical cognitive pattern; no hippocampal atrophy on MRI
- Alzheimer's: Memory loss predominates; hippocampal atrophy present
- Vascular dementia: Patchy hypoperfusion on SPECT; extensive white matter lesions with frontal predominance; cortical/lacunar infarcts
The presence of comorbid disease does NOT exclude NPH—if the patient improves after CSF drainage, they deserve shunt treatment regardless of coexisting pathology. 4
Distinguishing from Cerebral Atrophy
Use the combination of:
- Evans index >0.3
- Callosal angle <90 degrees
- Temporal horn enlargement disproportionate to hippocampal atrophy
These features distinguish NPH from simple age-related atrophy. 5
Ancillary Tests (Not Required for Diagnosis)
DTPA Cisternography:
Shows persistent radiotracer in lateral ventricles and absent activity over convexities on delayed imaging. Evidence is insufficient to base shunt decisions on cisternography alone—use it only as supplementary information. 1
Tests NOT Useful:
- MR spectroscopy: Cannot distinguish NPH from other dementias or predict shunt response 2
- FDG-PET: Insufficient evidence for routine use 1
- Contrast-enhanced MRI: Not needed for NPH diagnosis 2
Diagnostic Algorithm Summary
- Recognize clinical triad (gait > urinary > cognitive)
- Order MRI brain without contrast (or CT if MRI unavailable)
- Confirm imaging criteria: Evans index >0.3 + at least one additional feature
- Perform lumbar puncture: Normal opening pressure + remove 40-50 mL CSF
- Assess clinical response with objective gait/cognitive testing
- If positive tap test → Proceed to shunt evaluation
- If negative but high suspicion → Repeat tap test 2-3 times OR extended lumbar drainage
The diagnosis requires both characteristic imaging AND clinical improvement with CSF removal—neither alone is sufficient. 1, 2