Vahe Zohrabian
Normal pressure hydrocephalus (NPH), first described by Hakim and Adams in 1965,1,2 is a potentially reversible syndrome characterized by cognitive impairment, gait disturbance, and urinary urgency or incontinence, associated with enlargement of the ventricular system and normal CSF pressure. Originally considered “idiopathic” and classically presenting in the seventh and eighth decades of life, today, cases of chronic communicating hydrocephalus with a known cause are also categorized as idiopathic NPH (iNPH). The exact pathogenesis of iNPH is unknown, though many theories have been proposed. A rather crude hydrodynamic model involving excess CSF accumulation as a result of an imbalance among CSF production, circulation, and reabsorption is not sufficient to explain the pathophysiology of iNPH given that CSF pressures are not abnormally raised, even though episodes of increased CSF pressure may be found between repeated spinal taps (ie, intermittent pressure hydrocephalus). As prior studies have demonstrated that periventricular blood flow and cerebrovascular autoregulation are reduced in iNPH,3,4 this, in combination with tissue distortion caused by ventriculomegaly and CSF/interstitial fluid stasis, may lead to the accumulation of certain toxic metabolites in the brain, such as β-amyloid peptide and tau protein.3,5,6
The challenge in diagnosing iNPH lies in the fact that it shares signs and symptoms with other disorders of cognitive impairment, such as Alzheimer disease and small-vessel ischemic disease.7 Patients must have at least a portion of the clinical triad; gait disturbance is usually the first symptom to appear and it is at this early stage that shunting is most favorable. The accurate diagnosis of iNPH is important because CSF diversion is not without risk and has yielded varying results in this particular patient cohort.8 The CSF tap test, consisting of quantitative measurements of gait and cognitive dysfunction before and after the drainage of 40–50 mL of lumbar CSF, has been shown to predict the outcome of shunt surgery and the degree of clinical improvement, though it has a low sensitivity9,10; as such, the 2005 International iNPH Guidelines11 recommend repeating the test and removing a minimum of 30–40 mL of CSF on each of 3 consecutive days. Alternatively, patients may undergo continuous external lumbar drainage for 3–5 days, with a minimum of 150 mL of CSF drained daily, which has been shown to have a higher sensitivity and positive predictive value and is therefore considered the most effective method for identifying shunt-responsive iNPH.
Conventional CT and MRI have played an important role in iNPH diagnosis and prognosis. The International Guidelines11 have identified 1) ventricular enlargement not entirely attributable to cerebral atrophy or congenital enlargement (with an Evans index of >0.3 or a ratio between the maximal width of the frontal horns of the lateral ventricles and the maximal width of the inner table of the cranium at the same level)12 and 2) an absence of macroscopic obstruction to CSF flow as the key imaging features of iNPH, in the presence of at least 1 of the following complementary features: a) enlargement of the temporal horns of the lateral ventricles not entirely attributable to hippocampal atrophy; b) a callosal angle of 40° or greater; c) evidence of altered brain water content, including periventricular density or signal changes not attributable to small-vessel ischemia or demyelination; and d) an aqueductal or fourth ventricular flow void on MRI. The reliability of the Evans index was questioned in a prior study that demonstrated that values vary depending on the plane of measurement,13 though a more recent study showed that linear measurements of ventricular volume are reliable and reproducible.14 Periventricular signal changes suggestive of interstitial edema are typically minimal or absent in iNPH, corresponding to normal mean intraventricular pressure and low opening pressure on lumbar puncture, though periventricular changes compatible with leukoaraiosis are frequently present.15 The aqueductal flow void, characterized by signal loss or hypointensity on T2-weighted images, is accentuated in states of hyperdynamic CSF motion, and although it has been shown that CSF flow through the cerebral aqueduct is greater in patients with iNPH (despite normal CSF pressures),16 studies have been inconsistent on whether a correlation exists between the void sign on MRI and shunt results.15,17–19 This may partly be a result of more modern MRI techniques, such as fast or turbo spin-echo, being more intrinsically flow-compensated and less apt to demonstrate flow voids than early MRI. The Japanese Guidelines20 have additionally incorporated enlarged Sylvian fissures and basal cisterns, as well as narrowing of the sulci and subarachnoid spaces over the high convexity and midline surface of the brain (referred to as disproportionately enlarged subarachnoid space hydrocephalus), while lessening the importance of periventricular changes in the imaging diagnosis of iNPH.
Other noninvasive neuroimaging tools and techniques, though not routine in the selection of patients for shunt surgery, include but are not limited to PET, SPECT, MR CSF flow imaging, resting-state fMRI, MR elastography (MRE), and DTI. A prior [18F]FDG-PET/CT study demonstrated reductions in cerebral blood flow and metabolism, primarily in the frontobasal and anterior periventricular regions, in patients with iNPH.21 CSF flow imaging through cine phase-contrast MRI (PC-MRI) has allowed the quantification of aqueductal CSF stroke volume (ACSV), defined as the average volume flowing down during cardiac systole and up during diastole. In PC-MRI, the slice is positioned in an angled axial plane perpendicular to the cerebral aqueduct, and higher spatial resolution images are obtained at several prespecified encoding velocities using retrospective cardiac gating. Given that values of ACSV are highly machine- and technique-dependent, CSF flow studies should be performed on a number of healthy elderly patients without dilated ventricles (10 to 20) to determine what is normal on a particular scanner before diagnosing shunt-responsive iNPH.22 Whereas initial studies demonstrated that patients with iNPH who responded to shunting had at least twice the ACSV of healthy elderly patients,23 later studies did not.24,25 The combination of PC-MRI and the CSF tap test, by measuring the peak CSF flow velocity at the level of the aqueduct before and after CSF drainage, has been shown to be more reliable for predicting improvement after shunt surgery.26,27 More recent MRI techniques for CSF flow, such as time-spatial labeling inversion pulse, a CSF tagging technique similar to arterial spin-labeling, have demonstrated increased CSF displacement through the aqueduct in iNPH,28,29 though further investigation is needed to determine correlation, if any, with shunt response. MRE, a method to noninvasively measure tissue viscoelasticity in vivo, has also been utilized in iNPH.30–32 A prior MRE study has shown a loss of rigidity of up to 20% in brain tissues of patients with iNPH compared with controls, with changes most pronounced in the periventricular regions,31 whereas another study demonstrated increased stiffness in the cerebrum and in the occipital, parietal, and temporal lobes compared with age- and sex-matched healthy controls.30 Advances in DTI have also allowed for the delineation of microstructural white matter injury in iNPH and have been shown to aid in the differentiation of iNPH from Alzheimer disease and Lewy body dementia.33–35 DTI has demonstrated that fornix volume, cross-sectional area, and fractional anisotropy are smaller in both iNPH and Alzheimer disease relative to controls, though fornix length is significantly greater in iNPH, probably due to mechanic stretching caused by lateral ventricular dilation and corpus callosum deformation.36
Neuroimaging techniques and markers are becoming increasingly valuable in diagnosing iNPH and prognosticating response to therapies. This issue of the AJNR News Digest highlights several recent studies that expand upon these imaging tools and contribute to a better understanding of iNPH.