Continuous intracranial pressure monitoring: a last resort in pseudotumor cerebri.

I n patients with pseudotumor cerebri, it is critical to intervene surgically if visual loss continues despite medical therapy (1). This sounds like a simple rule to follow, but in reality, the decision to recommend surgery can be fraught with uncertainty. The essential problem is that if one gives medical therapy a chance to work, it may not be possible to snatch the bacon from the fire. By the time it has become clear that a patient has not responded to pressure-lowering drugs or lost weight, irreversible visual loss may have already occurred (2). For this reason, it is crucial to monitor patients with great vigilance who suffer from acute or progressive visual loss, so that optic nerve damage does not go too far. To monitor patients with pseudotumor cerebri is quite straightforward: check the visual acuity, obtain computer-assisted perimetry, and examine the fundi to assess the severity of papilledema (3–5). It is useful to sample the cerebrospinal fluid when the disease is diagnosed, to exclude other etiologies and to confirm that the intracranial pressure is elevated. However, it is rarely advisable or necessary to perform subsequent spinal taps on a regular basis. If the patient’s signs and symptoms are well controlled with medical therapy, then the intracranial pressure is moot. By the same token, if the patient is losing vision from papilledema, surgery is needed regardless of the intracranial pressure. There are, to be sure, exceptions to these principles. For instance, serial lumbar punctures to drain fluid may be a useful strategy in patients with tetracycline-induced intracranial hypertension. In such cases, the intracranial pressure returns to normal about a month or so after stopping the antibiotic, making it possible to avoid a shunt by using frequent lumbar punctures as a temporizing measure (6). What about patients who are unreliable, unable to perform visual field examinations, or even to cooperate with visual acuity testing? In such cases, the clinician is forced to rely primarily on the appearance of the optic discs to judge whether the patient can be managed safely without surgery. This approach works well in most patients, but the rub comes if there is already significant optic atrophy. Dead or damaged nerves do not swell, at least not much. Patients can lose their remaining vision with little evidence of papilledema. This is a potential drawback of optic nerve sheath fenestration: disc edema is reduced, but the optic nerves remain exposed to high intracranial pressure. As a result, visual loss can sometimes progress after surgery, even though papilledema is nearly absent (7). For this reason, among a host of others, it is usually better to implant a shunt than to perform optic nerve sheath fenestration when surgical intervention is necessary (8–10). Patients who have optic atrophy and perform unreliably on sensory tests pose a quandary. Dysplastic optic discs can also be difficult to evaluate. In such individuals, repeated measurement of the intracranial pressure may be the only dependable way to guide management. There is nothing else reliable to hang one’s hat on. Unfortunately, patients with pseudotumor cerebri are difficult to tap, prompting many neurologists to refer them for fluoroscopy-guided lumbar puncture. Inaccurate measurements are common due to incorrect positioning, partial needle obstruction, fluid loss, or patient straining. For example, a Valsalva maneuver can increase the pressure from a mean of 146 mm water to a mean of 323 mm water (11). Perhaps the biggest problem with lumbar puncture is that it captures a pressure reading at only one moment in time. Just like blood pressure, the intracranial pressure fluctuates widely, depending on posture and activity. Recall that lumbar subarachnoid pressure is equal to intracranial pressure only