Longstanding Overt Ventriculomegaly in Adults: Pitfalls in Treatment With Endoscopic Third Ventriculostomy

Harold L. Rekate, M.D.

Disclosures

Neurosurg Focus. 2007;22(4):E6 

In This Article

Discussion

Before the development and popularization of valve-regulated shunts in the 1950s, very few babies with a diagnosis of hydrocephalus received treatment. Most patients died during infancy, and the cognitive outcome of survivors was dismal.[18–20] Ventricular shunting in babies did not become a routine treatment until the 1960s. At that time, hydrocephalus was difficult to diagnose and only done so in severe cases. Children with large heads but no overt signs of intracranial hypertension were rarely treated because the only way to determine the cause of the megaloencephaly was to perform invasive procedures such as air ventriculography or even angiography. In that era, moderate ventriculomegaly would not have been diagnosed in a large percentage of patients. Now in their 40s and 50s, these patients have lived in a compensated state for decades.

In 1973 the first CT unit was installed in the US. Soon thereafter real-time ultrasonography was developed to the point that the anatomy of the lateral ventricles could be defined noninvasively when the anterior fontanel was open. At this point neurosurgeons were faced with a large number of patients with moderate degrees of ventriculomegaly who had received no previous treatment. Potentially effective forms of therapy with shunts were available, however. The issues of how much hydrocephalus is too much and when a shunt would improve outcomes became major subjects for discussion and study.[30,33] Unfortunately, firm answers to these questions are still not readily available.

There is no nomenclatural consensus in terms of patients without shunts but with large ventricles and no overt signs of increased ICP. Neither has the threshold for intervention been defined. Megaloencephaly with ventriculomegaly is almost certainly the substrate for the later development of NPH in the elderly.[3,5,30] A significant percentage of patients with NPH who have responded to shunting have head circumferences above the 98th percentile. This finding is reassuring when examining potential candidates for the treatment of this condition.[3–5] In this situation, it seems clear that an abnormality in CSF absorption leads to enlarged ventricles and a large head. These findings indicate that the process had begun before the cranial sutures fused in early childhood.

This situation is a chronically compensated state, and for each patient the cost of this compensation must be assessed. If treatment of the hydrocephalus were without risk, all of these patients should undergo shunt placement or ETV.[30] The more we seek information regarding risks and benefits, however, the more we realize that intervention does involve significant risks. During infancy and early childhood, the initial shunt insertion is associated with an 8% risk of infection. Fewer than half of the patients who receive a first shunt during childhood have working shunts 2 years later.[16] Patients who undergo shunt placement in childhood have mortality rates of 1% per year, a rate that persists into adulthood.[7,8,12,21,22] Late deaths have also been reported in patients in whom hydrocephalus is treated with ETV.[12]

This analysis of risk encourages a conservative approach, which entails awaiting the development of NPH in the aging population, overt signs of cognitive or motor deterioration, or signs of increased ICP before recommending intervention. A careful study by authors in Sweden showed results indicating that apparently asymptomatic patients with ventriculomegaly benefited significantly from treatment.[17] The study data provide support for proactive treatment.

The group in the current study represents a broad spectrum of patients with the mildest form of chronically compensated hydrocephalus. A risk-benefit analysis in this group may therefore be the most difficult. At the other end of the spectrum are patients with severely dilated ventricles, developmental delay, and a grossly abnormal head size. Such patients were the subject of a brief, provocative communication by Lorber[21] entitled "Is your brain really necessary?," which specifically featured a CT scan that had been obtained in a mathematics graduate student with massive ventriculomegaly.[21] For this group of patients, Oi et al.[24] coined the term "LOVA" and found that the complication rate associated with shunting in these individuals was low.

The closely related syndrome of hydrocephalus in young and middle-aged adults[9] relates to children with significant ventriculomegaly, some of whom also have large heads. Some of these patients have previously received shunts and either had the shunt removed or had a failed shunt without that diagnosis being considered important or with it being missed altogether. These patients have presented with subtle symptoms of insidious deterioration in cognitive and motor function. Many have head aches suggestive of mild increases in ICP. These patients challenge the concept that it is possible to have shunt-independent arrest of hydrocephalus and that all of these patients may suffer late deterioration years or decades later.[31]

Aqueductal stenosis is presumed to be the cause of hydrocephalus in a large percentage of patients with hydrocephalus without a clear origin and involving the lateral and third ventricles but not the fourth ventricle. In the context of sex-linked aqueductal stenosis, the aqueduct is closed by an anatomical obstruction. The latter is a genetic disorder found only in boys. The aqueduct is forked and does not allow CSF to pass from the third to the fourth ventricle. The aqueduct can be occluded by ependymitis from infection and by tumors of the tectum of the midbrain.

In animal models such as the HT rat, closure of the aqueduct follows the inward displacement of the temporal horns, leading to secondary closure of the aqueduct and triventricular hydrocephalus.[25] This situation also occurs in humans whose aqueduct is opened after a shunt repair.[2,6,23] In two of the patients (Cases 1 and 6) in the present report, successful internal bypass of the aqueduct by ETV led to its subsequent opening. I have been unable to find another case report involving opening of the aqueduct after ETV.

If the aqueduct closure is related to a more distal CSF absorptive difficulty, where is the source of that obstruction and what would be the effect of the internal bypass of the aqueduct? The patient in Case 6 provides a clear answer to that question. Her hydrocephalus resulted from congenital stenosis of the transverse sinuses. After the ETV, the aqueduct was shown to be open but ICP did not return to normal. The venous anomalies remained.

This situation is similar to cases of pseudotumor cerebri or the related condition of "normal-volume hydrocephalus," a term used to refer to infantile hydrocephalus in older patients who become symptomatic at the time of shunt failure without ventriculomegaly.[10,28] In normal-volume hydrocephalus, the ventricles communicate freely with the cortical subarachnoid space. Patients can be treated using lumboperitoneal shunts.[32] Several of the patients treated with these shunts have undergone retrograde venography and showed increased pressure in the dural venous sinuses. This condition has been seen in patients with venous sinus stenosis due to skull base abnormalities such as Crouzon syndrome and achondroplasia. It has been postulated that all pseudotumor cerebri results from high pressure in the dural venous sinuses.[15]

Several authors have reported success in treating this condition by using venous stenting techniques.[13,14,26,27] The patient in Case 6 in the present study had quite high ICP after the ETV. She was found to have bilateral transverse sinus stenosis with a high pressure gradient between the transverse and sigmoid sinuses. A venous stent was placed in this patient, and the pressure gradient resolved. Subsequent ICP monitoring has shown that her ICP is normal when she is either recumbent or upright.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....