Application of neuroendoscopy in the treatment of obstructive hydrocephalus secondary to hypertensive intraventricular hemorrhage
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.91366
Source of Support: None, Conflict of Interest: None
Objective: Neuroendoscopy has become an integral part of neurosurgery, in particular in the ventricular system. Obstructive hydrocephalus secondary to intraventricular hemorrhage (IVH) is a good indication for neuroendoscopic surgery. We evaluated its efficacy and limitations in these patients. Materials and Methods: During a 5-year period, 13 patients with obstructive hydrocephalus secondary to hypertensive IVH were studied. Patients with IVH with no evidence of obstructive hydrocephalus or with a large parenchymal hematoma or IVH of vascular origin were excluded. Rigid endoscope was used to evacuate hematoma in lateral ventricles and third ventricle in all patients. Glasgow Coma Scale (GCS), Graeb score and ventriculo-cranial ratio were evaluated before and after endoscopic intervention and Glasgow Outcome Scale (GOS) was appraised at 1 month and 12 months, postoperatively. Results: Of the 13 patients, eight (61.5%) patients had thalamus hemorrhage. A successful endoscopic removal of intraventricular hematoma was achieved in all patients. Complications observed included, fornix contusion (1) and meningitis (1) and there was no procedure-related mortality. The mean Graeb score reduced from 8.69 ± 1.89 to 6.00 ± 2.68 (P=0.001) and ventriculo-cranial ratio decreased from 0.41 ±0.05 to 0.39 ± 0.05 (P=0.085) following the procedure. The mean GOS scores at 1 month and 12 months both were 2.7, but a bipolar distribution was observed at 12 months. The Graeb score changed significantly with positive correlation to GCS score change (=0.565 and P<0.05). Conclusions: Endoscopic management of severe IVH with obstructive hydrocephalus allows effective reduction of the amount of ventricular blood and improves level of consciousness. Future refinement in instrumentation and discreet case selection may make this method more applicable and effective.
Keywords: Endoscope, hydrocephalus, hypertension, intraventricular hemorrhage
Spontaneous intraventricular hemorrhage (IVH) secondary to hypertensive intracerebral hemorrhage (ICH) is associated with high mortality and poor functional outcome. ,, The volume of IVH is an important determinant of the outcome in supratentorial ICH. , Obstructive hydrocephalus due to obstruction of cerebrospinal fluid (CSF) circulation in hypertensive IVH patients is often present on admission and has been a predictive factor of outcome. , Communicating hydrocephalus, at the subsequent stage, is not fatal but often requires shunt procedures. Hence, the goal of management of IVH should thus be to evacuate IVH at the acute stage as much as possible. 
Surgical methods of reducing hematoma volume and relieving obstruction to flow of CSF include placement of external ventricular drain (EVD) and stereotactic computed tomography (CT) guided aspiration with thrombolysis. , However, these procedures are associated with ventricular catheter obstruction, hemorrhage, procedure-related infection, and CSF pleocytosis. , Use of endoscopy in the field of neurosurgery has been increasing and with well developed techniques and devices; various procedures of endoscopic removal of IVH have been reported. ,,, In the present study, we performed endoscopic evacuation of IVH in 13 patients presented with obstructive hydrocephalus, and retrospectively examined the functional outcomes and intraoperative findings.
Study design and patient data
The study included 13 patients with hypertensive IVH and hydrocephalus who underwent neuroendoscopic evacuation of IVH between June 2004 and July 2009 at the Department of Neurosurgery, Chang Gung Memorial Hospital in Keelung, Taiwan. Patients with IVH but no evidence of obstructive hydrocephalus or a large parenchymal hematoma (midline shift over 5 mm with an extension to subcortical surface) or IVH of aneurismal or arteriovenous malformation origin were excluded. Glasgow Outcome Scale (GOS) was performed at 1 and 12 months, postoperatively. Final status evaluation of all the patients was done either during follow-up outpatient visit or via telephone interview.
The clinical features of the patients are summarized in [Table 1]. All patients had their preoperative diagnosis confirmed by a computerized tomography (CT) scan at admission. Severity of IVH was graded as proposed by Graeb et al [Table 2]. Degree of ventricular dilatation was measured by calculating the ventriculo-cranial ratio FH/ID (where FH is the largest width of frontal horns (FH), and ID is the internal diameter (ID) at the same level) from pre- and postoperative CT scans.
Endoscopic equipment and general surgical technique
All patients underwent bilateral external ventriculostomy placed just anterior to the coronal suture within 24 hours of diagnosis. The endoscopic procedure was performed under direct vision using rigid Aesculap 0° and 30° scope (MINOP system, Aesculap AG and Co. KG, Tuttlingen, Germany). The side contralateral to major hematoma was approached first. Once the endoscope reached lateral ventricle, confirmed by an overflow of blood-mixed CSF, continuous lactated Ringer's solution irrigation and an intermittent manual aspiration with a rigid sheath connected to the operating channel of the endoscope was commenced. Intraoperative visibility improved with continuous irrigation and gentle suction and allowed for an initial orientation in the lateral ventricle. The UNITRAC® Holding System (Aesculap AG and Co. KG, Tuttlingen, Germany) was used for positioning and holding of instruments in a stable manner over long periods. The instrument was advanced into third ventricle after choroidal plexus and foramen of Monro were identified; then the procedure was repeated for removal of clot in the third ventricle and as a way to aqueduct as much as possible. More attention was needed while inspecting the aqueduct as local increased pressure might be induced from over-irrigation. Subsequently, the side of the major hematoma was approached. The hematoma was evacuated by repeated irrigation using lactated Ringer's solution and aspiration. Grasping forceps for dispersing larger clots and bipolar forceps for cauterizing visible bleeders were used through the endoscope's operating channel if needed. The infant feeding tube for approaching sharp-angle regions such as temporal horn were used for carefully aspirating the clot, under the assistance of a rigid Aesculap 30° scope. A unilateral or bilateral EVD was left in the lateral ventricle to monitor intracranial pressure (ICP) and also drainage of blood in all the patients. EVD was removed after normalization of ICP.
Postoperative assessment included the necessity of shunt diversion and occurrence of ventriculitis. A CT scan was obtained within 24-hours after surgery for recalculation of the Graeb score and the ratio FH/ID; another CT scan was obtained at 6-8 weeks after surgery to confirm the resolution of hydrocephalus and IVH. The GOS score at 1 month and 12 months postoperative were documented.  Changes in Graeb score and FH/ID ratio were statistically analyzed by the Wilcoxon matched-pairs test. The relation between the GCS score change (postoperative score minus preoperative score) and the Graeb score change (preoperative score minus postoperative score) was statistical analyzed by using the Pearson's correlation coefficient.
The mean age of the patients was 66.9 years (range 44-80 years). Of the 13 patients, 8 (61.5%) patients had thalamic ICH. Preoperative GCS was ≤8 in 9 (69.2%) patients and 9-12 in the remaining 4 patients [Table 1]. In all these 13 patients hydrocephalus was due to obstruction of CSF circulation as evidenced by radiological findings and endoscopic procedure was considered as reasonable treatment option.
Successful endoscopic removal of IVH was achieved in all the patients and no procedure had to be abandoned. In all the patients, the intraoperative view was blurred by a blood clot or bloody CSF and vigorous irrigation was required. Complications included fornix lesion as defined by intraoperative observation of a contusion of the fornix area at the Monro foramen (1) and meningitis (1). The patient with fornix contusion expired 2 weeks postoperative as a result of multiorgan failure. The patient with meningitis had risk factors for infection like diabetes mellitus and poor neurological status at admission. Other than these two complications there were no procedure-related permanent morbidity and mortality. Of the 13 patients, 3 patients required placement of ventriculoperitoneal shunt for persistent elevated ICP.
Radiological and outcome evaluation
All the patients underwent pre- and postoperative CT. Preoperatively, all patients had enlargement of the ventricles filling from blood clots. The mean Graeb score changed from 8.69±1.89 to 6.00±2.68 (P=0.001). However, the range of the FH/ID ratio decreased only from 0.41± 0.05 to 0.39± 0.05 (P=0.085). Significant reduction of the intraventricular blood clot and maintaining clarity of the CSF pathway (the foramens of Monro, third ventricle, and cerebral aqueduct) were achieved [Figure 1].
Two (15.4%) patients died during hospital-stay because of the associated comorbidities, end-stage renal disease, diabetes mellitus, and atrial fibrillation and severe ICH [preoperative Glasgow Coma Scale (GCS) score <8]. The mean GOS score at 1 month was 2.7. Of the patients who survived, 4 (30.7%) patients had a score of 2, 3 (23.3%) patients had a score of 3, and 4 (30.7%) patients had a score of 4. The GOS scores at 12 months: 5 (38.4%) patients died, 1 (7.7%) patient had a score of 2 and 1 (7.7%) patient had a score of 3, 5 (38.4%) patients had a score of 4, and 1 (7.7%) patient had a score of 5. There was a significant positive correlation between GCS score change (postoperative score minus preoperative score) and the Graeb score change (preoperative score minus postoperative score) (=0.565 and P<0.05) [Figure 2].
Obstructive hydrocephalus secondary to a hypertensive IVH represents a clinicopathological entity with a poor prognosis. A rise in intracranial pressure with initial bleeding is the primary cause of neurological deterioration and may contribute to a decrease in cerebral blood flow, subsequently leading to cerebral ischemia.  Furthermore, blocking of CSF and mass effect of blood clots within the ventricles may cause further deterioration.  Removal of blood clots and restoration of CSF pathway are the most essential aspects of treatment for this disorder. The analysis of our data revealed a positive, statistically significant relation between pre and postoperative Graeb score and GCS changes. These findings reflect the importance of reducing of ventricular blood volume. 
As of now placement of an EVD system remains the primary treatment in patients with IVH and obstructive hydrocephalus. This procedure can be performed quickly and easily with only minimal surgical risks. However, prolonged EVD use with frequent change of the external drain has been shown to be associated with an increased rate of infection. Malfunctioning of the EVD due to obstruction by blood clots or malpositioning of the catheter tip is also not infrequent.  Several reports of use of various endoscopic procedures in the treatment of IVH with hydrocephalus with variable functional outcomes and mortality have been reported. ,,,,,, Although, fiberscopes have been used in some reports, rigid endoscopes are commonly used because of the higher visual quality and ease in orientation. Yadav et al,  had demonstrated the safety of use of rigid scope in the management of severe IVH. Use of rigid endoscopic evacuation was associated with significant reduction in the amount of ventricular blood and good outcome in patients with a preoperative GCS of >9. Rigid endoscopes were also successfully used in our patient with hypertensive IVH and obstructive hydrocephalus.
In our series, we reached the goal of removing most of the ventricular blood and restoration of CSF circulation with the use of rigid endoscope [Figure 1] with good functional outcomes. However, there are distinct advantages in the form of flexibility and accessibility with fiberscopes. Our experience suggests that irrigating the aqueduct may result in local increased pressure. Therefore, it is important to closely monitor the hemodynamic status and keep the output channel of the endoscope free from obstructions since the channel may be obstructed by small blood clots. In addition, we feel that an endoscopic third ventriculostomy (ETV) may not be attempted by an inexperienced neurosurgeon, especially under a blurred field of vision and distorted ventricular anatomy.  Besides, communicating hydrocephalus due to absorption defects of CSF in the chronic stage of hypertensive IVH patients is traditionally considered a contraindication for endoscopic third ventrinulostomy (ETV). We suggest the use of neuroendoscopy in the management of obstructive hydeocephalus secondary to IVH in the acute stage as it can re-establish CSF circulation paths and remove blood clots.
It will be difficult to compare our results with the reported results in the literature for several reasons: case mix of both hypertensive IVH and IVH due to vascular malformation; small case series; interventions included both endoscopic procedures and third ventriculostomy; and variable outcome measures. In our series, the percentage of patients with a favorable outcome (GOS scores 3-5 at 12 months) was 53.9%, whereas Longatti et al, reported 61.5% favorable outcome. This series included both patients with hypertensive IVH and also IVH due to vascular origin and the endoscopes used were also different. 
Neuroendoscopic intervention for obstructive hydrocephalus secondary to IVH has some advantages over traditional surgical approaches, such as increased efficiency over EVD, and it is less invasive than craniotomy. In certain aspects, the endoscopic approach still has its limitations. First, these operative skills need a steep learning curve. After sufficient experience, one can shorten the operation time and the complications rate will be less. Second, one should carefully move the endoscope within the ventricles because it may contuse the vital structures of the ventricular wall, as happened in one of our patients, fornix contusion. A holder for fixation of the scope intraoperatively is advantageous. Third, some distal targets such as blood clots filling the fourth ventricle and anterior temporal horn are difficult to approach, although gentle irrigation and feeding tube aspiration can clean up most targets. Use of a flexible endoscope probably can solve this problem. 
[Figure 1], [Figure 2]
[Table 1], [Table 2]