In addition to obstructive hydrocephalus, brain tumors can also cause communicating hydrocephalus. This may involve several different mechanisms. For example, vestibular schwannomas can present with communicating hydrocephalus, as indicated by the fact that imaging studies show no evidence of ventricular compression and that the hydrocephalus persists after tumor resection [44, 45]. Although the mechanism underlying hydrocephalus in these cases is uncertain, some investigators have postulated that proteins secreted by the tumor decrease CSF absorption [44, 46, 47]. Choroid plexus tumors (i.e., choroid plexus papilloma and choroid plexus carcinoma) can also cause communicating hydrocephalus [48]. In part, this appears to be the result of an increase in CSF production, a unique property of tumors arising from the choroid plexus [8].

Communicating hydrocephalus can also be caused by infiltration of the meninges by tumor cells, a phenomenon known as leptomeningeal carcinomatosis or carcinomatous meningitis. Tumor cells reach the meninges by dissemination through the CSF [49]. Such meningeal involvement by tumor cells is thought to decrease CSF absorption, leading to hydrocephalus and elevated ICP. Carcinomatous meningitis has been estimated to occur in 3–8% of all cancer patients [49–52]. The two most common cancers involved are breast cancer and lung cancer [44, 53, 54]. In one study, 70% of patients with carcinomatous meningitis had one of these two types of cancer [53]. There is some evidence that piecemeal resection of brain metastases is associated with an increased risk of leptomeningeal dissemination of cancer when compared to en bloc resection of metastases or radiosurgery [55, 56]. This is presumably the result of spillage of cancerous cells into the CSF.

Presenting symptoms and signs of carcinomatous meningitis include headache, nausea, vomiting, cranial nerve palsy, motor dysfunction, mental confusion, transient episodes of loss of consciousness, and enlarged ventricles [49, 53, 57]. In addition to the history and physical examination, several diagnostic tests can be useful in identifying patients with carcinomatous meningitis. Because carcinomatous meningitis causes a communicating hydrocephalus, a lumbar puncture can be performed safely. The ICP is usually elevated. In addition, the CSF obtained from patients with carcinomatous meningitis often contains malignant cells on cytology, and it may also contain less glucose and more protein than the CSF of healthy patients [54, 58]. However, CSF cytology produces a false negative diagnosis in roughly 10% of patients with carcinomatous meningitis [49]. A useful adjunct to CSF analysis for the diagnosis of carcinomatous meningitis is contrast-enhanced MRI [50, 51, 59, 60]. In patients with carcinomatous meningitis, contrast-enhanced MRI may show diffuse meningeal enhancement or enhancing pial nodules scattered widely over the surface of the nervous system [44, 50, 51, 60]. Often (but not always), ventricular enlargement is present as well, and transependymal flow of CSF may be seen on T2-weighted images (Fig. 11.3a, b).

Patients who have central nervous system tumors or who have received radiation and/or chemotherapy for such tumors can develop a form of communicating hydrocephalus that is not associated with severe elevations in ICP. This form of communicating hydrocephalus, called normal pressure hydrocephalus (NPH) , is characterized by enlarged cerebral ventricles, gait difficulty, incontinence, and dementia. Although some investigators have reported increased ICP pulsatility in NPH, these patients do not display sustained elevations of ICP [61]. Consequently, they do not develop symptoms of elevated ICP such as headache, nausea, vomiting, decreased consciousness, or herniation.

The incidence of NPH among brain cancer patients is not known, in part because the disorder often goes unrecognized or is misdiagnosed as effects of tumor progression or direct effects of treatment on neuronal function. Inamasu and colleagues reported that 5 of 50 consecutive patients (10%) who underwent treatment for supratentorial malignant glioma developed communicating hydrocephalus [62]. A larger retrospective study of 124 patients who underwent surgery, radiation, and chemotherapy for glioblastoma identified 7 patients (5.6%) who subsequently underwent shunt placement for communicating hydrocephalus [63].

The mechanisms underlying the development of NPH after brain tumor treatment are poorly understood. Entry into the ventricle at the time of surgery and high CSF protein levels were found to be significant predictors of the development of communicating hydrocephalus in patients with GBM [63]. Fischer and colleagues also reported a statistically significant correlation between ventricular opening at the time of surgery and the development of communicating hydrocephalus after treatment for GBM [64]. In addition, an association between radiation and the development of hydrocephalus has been reported [65, 66]. Shunt placement after therapy-induced hydrocephalus improved symptoms and quality of life in these patients [63, 65].

The gait difficulty seen in NPH is characterized by imbalance and gait apraxia [67]. Patients typically are not weak, even though they may have difficulty rising from a seated position, initiating movement or regulating the amplitude and timing of their movements. The cognitive dysfunction of NPH has multiple components, but is most notable for deficits in executive functioning, visuospatial processing and the ability to encode new memories [68]. Together, these deficits suggest impairment of frontal, parietal and temporal lobe function.

The diagnosis of NPH should be considered in patients who present with slowly progressive gait difficulty, dementia and incontinence [69]. Cranial imaging shows evidence of ventricular enlargement in most cases (Fig. 11.4a, b). The diagnosis of NPH can be confirmed by a trial of CSF drainage to determine whether symptoms improve. Although a high volume (greater than 30 cc) lumbar puncture can lead to immediate improvement, studies indicate that this test is unreliable because of a high frequency of false negatives [70]. The high false negative rate of high volume lumbar puncture for NPH diagnosis occurs largely because many patients require several days of CSF drainage before symptoms improve. Consequently, many practitioners utilize an extended trial of lumbar CSF drainage over several days to increase the likelihood that an accurate assessment of the effect of CSF drainage on NPH symptoms is obtained [69].

One of the most commonly used and most effective approaches to decreasing elevated ICP caused by vasogenic edema is the use of glucocorticoids such as dexamethasone [71, 72]. Dexamethasone activates glucocorticoid receptors in the nucleus of cells to increase the expression and activation of tight junction proteins in cerebral vessels. Thus, dexamethasone decreases ICP by reducing blood–brain barrier incompetence and decreasing vasogenic edema [71]. However, the long-term use of dexamethasone is limited by its many (and sometimes serious) side effects. These include weight gain, ulcers, aseptic necrosis of the hip, osteopenia, muscle weakness, psychosis, and others [73].

A recent addition to the armamentarium of treatments used to decrease elevated ICP caused by vasogenic edema is the anti-VEGF antibody, bevacizumab [72]. This antibody neutralizes the activity of tumor-secreted VEGF, which is a primary angiogenic stimulus for the generation of leaky blood vessels in many types of brain tumors. Bevacizumab has been shown to decrease cerebral edema by “normalizing” leaky vessels, thereby decreasing mass effect and reducing ICP [74]. By reducing vasogenic edema due to tumor or radiation necrosis [75], bevacizumab has become a useful alternative to dexamethasone for controlling ICP.

The treatment of hydrocephalus in brain tumor patients consists primarily of surgical removal of the obstructive lesion, performance of endoscopic third ventriculostomy (ETV) to bypass an obstructive lesion that cannot be removed, or placement of a shunt for CSF diversion. As mentioned previously, tumor resection to reopen blocked CSF pathways is performed whenever possible, as this allows for therapeutic tumor removal and may avoid the need for a permanent CSF diversion procedure. ETV or placement of a ventriculostomy may be performed to divert the CSF before or during the operative procedure when appropriate [43, 83]. The ventriculostomy can then be weaned under controlled conditions postoperatively in order to ensure that the obstructive hydrocephalus has resolved. Occasionally, hydrocephalus and elevated ICP persist despite resection of the tumor, and shunt placement or ETV is necessary [44, 45].

In most cases of hydrocephalus, shunt placement leads to a rapid improvement in symptoms. Indeed, shunt placement can be a life-saving procedure in brain tumor patients with elevated ICP or mass effect due to hydrocephalus. Because it frequently alleviates the NPH symptoms of gait difficulty, dementia and incontinence, shunt placement can lead to a significant improvement in the quality of life for both brain tumor patients and their caregivers.

Although CSF drainage provides relief of hydrocephalus-associated symptoms in patients with carcinomatous meningitis, the overall prognosis for patients with this disorder remains very poor. Literature estimates of median survival for untreated patients range from 4 to 6 weeks [49, 54]. With treatment, median survival time increases to 3–6 months [49–51, 57]. Survival of patients with carcinomatous meningitis is usually determined by the site and extent of systemic disease. In one study involving 122 carcinomatous meningitis patients, liver metastases were associated with the worst prognosis [57].

Therapeutic approaches to the treatment of carcinomatous meningitis include high dose systemic or intrathecal chemotherapy and/or radiation to symptomatic disease sites [49, 50, 84]. The most commonly used chemotherapeutic agents for intraventricular therapy include methotrexate, cytarabine, and thio-TEPA [50, 51]. Ventriculoperitoneal shunts (VPS) have also been used in patients with carcinomatous meningitis and hydrocephalus [84–87]. One study demonstrated that the use of a CSF reservoir system in combination with shunt placement led to better outcomes than the use of the reservoir system alone [84]. Other studies have also demonstrated the effectiveness of shunt placement as a palliative measure in this patient population [86, 87].

Here we have reviewed the origins, presentation, and treatment of elevated ICP and hydrocephalus in brain tumor patients. Large tumor volume, vasogenic edema, obstructive hydrocephalus, communicating hydrocephalus, and carcinomatous meningitis can all lead to mass effect and elevated ICP. Although we discussed each of these factors separately, they are often found in combination. For example, patients with large tumors and mass effect can develop vasogenic edema and obstructive hydrocephalus that contributes to elevated ICP (Fig. 11.5a–d). Treatment strategies in such situations are often combinatorial. The plan for which treatments to use and in what order varies, depending upon the specific details of each case. Consequently, neurooncologists, radiation oncologists and neurosurgeons must work collaboratively to formulate an optimal treatment plan for each patient.