Hydrocephalus associated with vestibular schwannoma is a well described entity and its frequency varies between 1.2–42% [2–4, 7, 8]. Although the literature describes widely preoperative hydrocephalus and its management, there is no sufficient data about the early postoperative hydrocephalus.
Postoperative ventricles enlargement
In our institution we routinely perform preoperative MRI and postoperative MRI during the first 3 days after the surgery, or a CT examination during the 1 POD, if there is no possibility to perform MRI. We noticed that postoperative ventricles were larger in 14 cases (12%). However, only in 8 cases (57% of that group, 7% regarding the whole cohort) there were disturbances of consciousness. In 6 cases (5%), the EVD was placed, however, only 2 patients were in a severe condition (1%) and one of them died (0.5%). Huang et al.  had 14 cases (1.19%) of acute postoperative hydrocephalus, 8 of which were caused by a hemorrhage and 6 by cerebellar edema. However, the authors provide no further information about the outcome and clinical analysis in that group of patients. Mahboubi et al.  analyzed the overall complication rate in a group of 6553 cases operated on in California. Hydrocephalus was found in 3.6% of patients. Unfortunately, no further information could be found in that paper.
We have noticed that 42% of patients with postoperative ventriculomegaly did not have any signs of intracranial hypertension. That group of patients was managed conservatively using steroids (8 mg per day, twice a day) and mannitol (up to 1 g/kg, repeated q 6–8 hours). The rest of the patients in that group developed some signs of intracranial hypertension. Those patients could be divided into 2 distinct groups. The first one (type I hydrocephalus) was characterized by mild consciousness disturbances and some degree of edema in the ipsilateral cerebellar hemisphere. We observed 6 patients in that category. What is more, in 2 of them antiedema therapy was sufficient to improve the consciousness state and to control the symptoms. In 4 cases, pharmacological therapy was not successful, hence surgical treatment was employed. Still, all these patients were evaluated as 12 points or more in the GCS. In each case EVD was placed with immediate improvement after the surgery. All EVDs were removed on the 5th POD without further complications. The second group (type II hydrocephalus) was characterized by a sudden and deep deterioration in the consciousness state, usually during the 1st POD. In our cohort, we observed one postoperative hematoma in the tumor bed and one case of large cerebellar hemisphere edema, probably due to venous stroke. In that group, there was a tendency for longer EVD duration and uncertain prognosis, depending on the cause of the hydrocephalus. In one case, there was a necessity to implant the VPS, because of posthemorrhagic communicating hydrocephalus.
Predictive factors for acute postoperative hydrocephalus evolution
Peritumoral edema is a reported radiological finding in VS. It could be visualized in 5–38% of cases [11–13]. Its role and influence on the treatment is under recent evaluation. Guo et al.  found that peritumoral cerebellar edema is an independent risk factor of postoperative hematoma and a risk of reoperation after the VS resection. They also revealed that it is a factor influencing the extent of resection. Samii et al.  evaluated a similar topic and found that the presence of edema tends to be associated with a higher risk of the postoperative hematoma. However, the differences were not significant statistically. Cases with PTE seem to have no arachnoidal plane between neurovascular structures and tend to be more hypervascular . Recent data  suggest that tumors with higher presentation of VEGF (Vascular Endothelial Growth Factor) are more prone to have edema in the surrounding nervous tissue. To our knowledge, there is no data about correlation between PTE and postoperative hydrocephalus. PTE was seen in 23% of cases, which fits in the cited data. However, we did not see a correlation between preoperative edema and postoperative hydrocephalus. What is more, other preoperative and intraoperative factors such as age, sex, size of the tumor, volume of the tumor, preoperative Evans score, extent of resection of the tumor, obstruction of the ventricular system, cystic appearance of the tumor did not increase the risk of postoperative hydrocephalus. Postoperative hematoma in the tumor bed is the only condition, which poses a risk for acute hydrocephalus.
Based on our results, it is impossible to predict which patients would be in the risk group of acute postoperative hydrocephalus. Surgical technique seems to play role only in prevention of the second type of hydrocephalus. Protection of venous structures and meticulous hemostasis are mandatory in those cases. However, we could not find any technical nuances preventing the hydrocephalus described as first type. We routinely preserve the arachnoid plane, whenever it is present. However, we did not find lack of the arachnoid plane as a risk factor of hydrocephalus.
In our institution we used an algorithm of clinical management in that group of patients. We routinely perform contrast-enhanced MRI within the first 72 h after the surgery, in most cases on the day after the surgery. If there is no possibility to assess MRI on that day, we provide a non-enhanced CT investigation. If the imaging studies reveal an enlargement of the ventricular system, any signs of posterior fossa edema or aqueduct obstruction and the patient scores more than 14 points in the GCS, we try to manage such cases conservatively with clinical observation and anti-edema medication. If there is no improvement in the neurological status during the 6–12 h we perform the EVD. If there is an improvement, we continue the wait and watch strategy for 3 days. After that time, we perform CT and try to taper the antiedema medication. If the patient is evaluated as less than 14 points we place EVD and provide antiedema medication. Patients in the second group, with sudden loss of consciousness or a failure to regain it after anesthesia are candidates for urgent CT examination and neurosurgical evaluation – EVD and posterior fossa decompression, or hematoma evacuation.
As it follows from our experience, the patients qualified into the first group of postoperative hydrocephalus have good prognosis. Placement of the EVD is a procedure with low risk of complications. They usually need a short term maintenance of the drainage. In all of the cases in that group, EVD was removed on the 5th POD. That kind of hydrocephalus seems to pose no risk for further CSF diversion surgeries. None of the cases needed a VP shunt in late observation. Patients in the second group have uncertain prognosis, depending on the cause of the hydrocephalus. In our cohort we observed 2 patients in that group. One case was fatal because of prominent cerebellar edema, probably caused by venous infarction. The other case was an example of hematoma in the tumor bed, which necessitated VP after 2 weeks. In the late follow-up, the patient was evaluated as grade 0 in the mRankin scale. Based on our results, further CSF diversion surgeries depend on the cause of the postoperative hydrocephalus. Benign cases in the first group need only a temporary diversion surgery. The second group seems to be inhomogeneous in the etiology of hydrocephalus.
To our best knowledge, there is no accepted definition of acute postoperative hydrocephalus after VS resection. Based on our results, we could point out some typical characteristics of that condition. First of all, it is always obstructive in its pathophysiological mechanism. Postoperative edema of the ipsilateral cerebellar hemisphere, or a hematoma-like extracerebellar mass causes obstruction of CSF flow at the level of the aqueduct or the foramen of Luschka. What is more, some degree of intracranial hypertension signs must be present to diagnose the hydrocephalus. It is crucial, because 42% of cases with marked postoperative ventriculomegaly are asymptomatic and do not need surgical treatment. The signs of intracranial hypertension could be benign, like persistent headache not responding to pain medications or mild disturbances of consciousness, or severe like coma. We did not observe the hydrocephalus after the first 4 POD. On the basis of these findings, we could propose a definition of the early postoperative hydrocephalus after VS surgery. It is a condition after resection of the VS with marked symptomatic postoperative ventriculomegaly, caused by obstruction of the CSF flow in the posterior fossa during the first four POD.