Hydrocephalus by definition is an excessive accumulation of CSF in the ventricular system. What leads to this condition can be a host of causes ranging from haemorrhage to neoplasm. However, its aetiology can be roughly divided into two categories: non-communicating, wherein there is a blockage in the ventricular system (e.g. aqueductal stenosis); and communicating, which can result either from a decrease in CSF resorption (scarring after subarachnoid haemorrhage) or from CSF overproduction (e.g. choroid plexus papilloma). In the realm of geriatrics, it is virtually never a consequence of CSF overproduction. As far as NPH is concerned, it is commonly accepted that fluid build-up is due to impairment of the brain’s CSF resorptive capacity—an idea that is known as the bulk flow theory. This can occur at the level of the arachnoid granulations along the skull base and over the hemispheres.8 Some authors feel that resorption also occurs in the brain parenchyma through capillaries and venules9, 10 and resorptive problems can also be at this level. Among the known causes of NPH are subarachnoid haemorrhage, trauma and meningitis, all of which can lead to scarring of the arachnoid granulations in the subarachnoid space. Such conditions are commonly referred to as secondary NPH.

In over half of NPH patients, however, there is no identifiable cause. The development of idiopathic NPH remains unclear, but it appears to involve the interplay of several factors such as cerebral ischaemia, decreased vascular compliance, decreased cerebral blood flow to periventricular regions and brain atrophy. MR flow quantification studies showed lower venous compliance in NPH patients, which could lead to increased resistance to CSF outflow and cerebral ischaemia in the concerned areas.11 Indeed, it has been shown that after shunting, cortical vein compliance is improved in patients who respond to the procedure.12 The bulk flow theory is easily understandable but others feel that it is an oversimplification. It appears inadequate in explaining certain characteristics such as why CSF pressure can remain normal and why some patients improve with endoscopic third ventriculostomy (ETV), which does not involve diverting fluid volume away from the intracranial space. An alternative theory known as the hydrodynamic concept of hydrocephalus has been gaining traction. It holds that a decrease in intracranial compliance allows an increased transmission of systolic pressure to the brain parenchyma, which in turn becomes distended against the unforgiving skull.13 Creating a hole on the floor of the third ventricle (ETV) thus allows for ‘venting’ of the increased systolic pressure and CSF to the subarachnoid space. The thought is that the aqueduct of Sylvius is too narrow and inadequate for venting.

There are other factors that influence the pathophysiology of NPH. Cerebral blood flow has been shown to be decreased in patients with idiopathic NPH, particularly in periventricular regions14 and also the basal ganglia and thalamus.15 Idiopathic NPH patients have also been found to have brain atrophy in addition to hydrocephalic features.16 More recently, there has been an increasing awareness of the role that metabolic disturbances play in the genesis of NPH symptoms. For instance, accumulation of beta-amyloid and disturbance in cholinergic neurotransmission may contribute to cognitive and memory decline.17 This would explain why cognitive function sometimes does not improve after shunting. It is also suggested that deterioration of striatal GABAergic neurons and lower dopamine levels in the substantia nigra could lead to motor dysfunction. This idea would account for the observation that patients who initially show gait improvement after shunting later regress despite proof that the shunt is working or even after valve pressure adjustments. Although we do not fully understand the interplay of the multiple factors that affect NPH patients, it is clear that NPH symptoms do not simply evolve from problems in CSF dynamics.

Although NPH has traditionally been associated with neurosurgeons because of its ultimate treatment, the evaluation and workup that lead to surgery require the involvement of many other medical professionals. It may be the primary care provider who casts the first suspicion or the geriatric specialist who pushes for further workup. Then there is the neuropsychologist who delineates the condition by performing neurocognitive evaluations. However, it can be argued that it is the neurologist who holds the central position as he or she has the deepest knowledge of other similar conditions that allow for better differentiation and diagnosis of NPH.18

It is easy to say that NPH presents with the classic triad of gait disturbance, urinary incontinence and mental decline. However, it can be challenging to assess the character of each of these symptoms since there are other conditions that may produce NPH’s components. For instance, gait disturbance can be seen in movement disorders and a common differential is Parkinson’s disease. Memory or cognitive decline can be seen in Alzheimer’s disease; in fact, it is the more common consideration. As for bladder incontinence, there are a host of urological and neurological reasons that can be suspected. Having said all that, there are typical characteristics that we can look for when evaluating a patient who presents with these symptoms.

The most prominent symptom is usually the progressive difficulty in ambulation. It is typically described as a wide-based gait that is slow and unsteady. The patient usually walks in small steps and has poor floor clearance, that is, they do this with their feet clearing the floor at a low height. Others describe it as a feeling that the feet are glued to the floor, commonly referred to as ‘magnetic feet’. The degree of gait disturbance depends on the severity of the NPH and patients who are in the early stages of the disease can walk almost normally except for some unsteadiness. It is not uncommon to encounter patients who are still highly functional and simply report a gait disturbance on walking a certain distance or against an incline. Conversely, patients with severe NPH can be so debilitated that they are unable to stand without support. Despite the advanced and severe state, their response to shunting can be very dramatic so they are not to be dismissed or given up on.

Mental decline is usually in the form of problems with memory, particularly short-term memory. It may also involve a decrease in mental alertness and an overall slowing of thought processes. The process may be so gradual that nobody notices it or it is attributed to ageing. It is important to ask family members aside from the patient because it may not be apparent to the patient himself or herself. The physician may also find it difficult to detect during the finite visit, whereas family members are able to share their observations of the patient over a longer period of time and across varying situations. Family members are crucial to the history-taking in this sense.

Urinary incontinence may start as progressive urgency, but many will have frank incontinence by the time of consultation. It is a symptom that the examiner needs specifically to ask about because it may not be recognized by the elderly patient or family members as significant. They may not realize that it is related to the overall neurological condition or the patient may be somewhat embarrassed to say that he or she is already on adult diapers. Of the clinical triad of symptoms, urinary incontinence is usually the last to develop. It is considered to be a symptom of late-stage NPH.19

When dealing with the geriatric population, it is not hard to imagine that NPH symptomatology in an elderly patient can be easily mistaken for, and brushed aside as, part of the ageing process. Awareness of this condition is not as prevalent as it should be and many patients do not come to medical attention. Indeed, one can only wonder how many patients have gone undiagnosed and have potentially missed out on a better quality of life had they been shunted.

Another challenge to identifying NPH is related to its differential diagnoses. These include Parkinson’s disease, Alzheimer’s disease, vascular dementia and spinal stenosis. Most commonly, it can be difficult to differentiate from Parkinson’s and Alzheimer’s disease. Gait similarities in NPH and Parkinson’s disease include short steps and leg rigidity.20 The typical tremor of Parkinson’s disease may help make this diagnosis, but sometimes patients with NPH can also have confusing tremors. In general, it is possible to differentiate between the two by the gait patterns. NPH is characterized by reduced stride length, reduced step height and balance difficulty. Parkinson’s disease patients have a shuffling gait with impaired arm swing and walk with their hips and knees slightly flexed and trunk bent forward.21 In comparing Alzheimer’s disease and NPH, the dementia is not as pronounced as in Alzheimer’s disease. However, it is not uncommon to find NPH patients with concurrent Alzheimer’s disease.22 It is important to identify patients with both Alzheimer’s disease and NPH, because this particular subset tends to respond poorly to shunting.23 Given these challenges in differential diagnosis, we again emphasize the crucial role that neurologists and neuropsychologists play in the evaluation of potential NPH patients.

Numerous tests have been developed over the years to help in the diagnosis of NPH. Some have stuck and become incorporated into regular practice whereas others have passed on to the realm of historical interest. It is fascinating to see how the evolution of these tests reflects the dynamic ideas that have been elucidated with regard to its pathophysiology over the past five decades. As in many medical conditions, diagnostic tests are not meant to replace good history taking and physical examination. In fact, clinical acumen and computed tomography/magnetic resonance imaging (CT/MRI) play a primary role in clinching the diagnosis. However, it is important to note that the diagnostic workup for NPH is as much about diagnosis as it is about determining whether a patient will respond well to surgery. For this reason, a number of other diagnostic modalities have been developed to predict shunt responsiveness. There is still significant variability in practice as far as usage is concerned, depending on ideological and logistical differences from place to place.

CT/MRI