Efficacy of translamina terminalis ventriculostomy tube in prevention of chronic hydrocephalus after aneurysmal subarachnoid hemorrhage.

BACKGROUND: Chronic shunt-dependent hydrocephalus is still a common complication after aneurysmal SAH (aSAH) and is associated with increased morbidity. Pathology of chronic shunt-dependent hydrocephalus after aSAH is complex and multifactorial which makes its prevention challenging. We thought to evaluate whether external ventricular drainage (EVD) through fenestrated lamina terminalis would decrease the rate of chronic shunt-dependent hydrocephalus after aSAH.
METHODS: A retrospective analysis of 68 consecutive patients with aSAH who underwent microsurgical clipping of the ruptured aneurysm. Patients were divided into two groups: Group A included patients with lamina terminalis fenestration without insertion of ventriculostomy tube and Group B included patients with EVD through fenestrated lamina terminalis. Demographic, clinical, radiological, and outcome variables were compared between groups.
RESULTS: Group A comprised 29 patients with mean age of 47.8 years and Group B comprised 39 patients with mean age of 46.6 years. Group B patients had statistically significant (P < 0.05) lower incidence of chronic shunt- dependent hydrocephalus than Group A patients (30.8% vs. 55.2%, respectively).
CONCLUSION: EVD through fenestrated lamina terminalis is safe and may be effective in decreasing the incidence of chronic shunt-dependent hydrocephalus after aSAH. Copyright:

INTRODUCTION

Chronic shunt-dependent hydrocephalus is a common consequence after aneurysmal subarachnoid hemorrhage (aSAH) with reported incidence of 8.9–48%. It contributes largely to patient long-term morbidity, hospital readmission, and health-care economic burden occurring after aSAH. Several factors have been identified to predict the development of chronic hydrocephalus after aSAH including radiological findings as hydrocephalus at presentation, intraventricular blood, intracerebral hematoma and high Fisher grade, clinical data as high Hunt and Hiss score, and laboratory data as sustained systemic inflammatory response. The rate of development of chronic hydrocephalus after aSAH was affected by treatment modality used to secure the ruptured aneurysm, whether microsurgical clipping or endovascular coiling, in some reports and not affected in other ones.[1,2,9,11,13,15-19,23,29,30,32,33,46,47,49]

It has been reported that maneuvers accelerating the clearance of blood in the basal cisterns could halt the development of symptomatic ventriculomegaly and shunt dependence after aSAH.[30] It was also suggested that fenestration of lamina terminalis during microsurgery for anterior circulation aneurysms decreases the incidence of chronic hydrocephalus and shunt dependence after aSAH.[29] The objective of our study was to evaluate whether external ventricular drainage through a catheter inserted into fenestrated lamina terminalis during microsurgery for clipping of ruptured anterior circulation aneurysms was associated with reduced rate of development of chronic hydrocephalus and shunt dependence.

MATERIALS AND METHODS

The present study is conducted with the approval of our university ethical committee. This is a retrospective analysis of prospectively collected data on patients with intracranial aneurysms (IAs) who were managed at our university emergency hospital between July 2013 and July 2019. We enrolled in this study patients with ruptured anterior circulation IAs who underwent microsurgical clipping of their ruptured IAs. Our plan was to assess the effect of performing EVD through a translamina terminalis tube inserted during clipping surgery on the development of chronic hydrocephalus after aSAH. We divided our patients into two groups; Group A: patients who underwent aneurysm clipping with lamina terminalis fenestration only without inserting EVD tube and Group B: patients in whom EVD through a translamina terminalis tube was done during clipping surgery. We excluded patients in whom EVD was done by conventional ventriculostomy through Kocher burr hole before surgery or through the Paine’s point during surgery and patients who presented by Hunt and Hess score V.

The general surgical steps for clipping of the ruptured IAs were essentially the same in both groups. Either lateral supraorbital or pterional craniotomy was done in all cases. After dural opening, CSF is released from the basal cisterns and from the lamina terminalis to lax the brain. The ruptured aneurysm is then clipped under temporary proximal control. In Group B, a ventriculostomy tube is inserted, under direct microscopic visualization, inside the third ventricle through the fenestrated lamina terminalis to come out beneath the basal brain surface [Figure 1]. The tube is externalized through a craniotomy burr hole and a separate skin puncture and connected to external drainage system for CSF release and daily cytochemical analysis. Weaning of EVD starts on day 7–10 after surgery. The external reservoir is gradually elevated over 2 days to maximum of 20 cm water then clamped for 1–2 days with serial clinical examination and CT scans to monitor any clinical deterioration or ventriculomegaly. Prophylactic intravenous broad-spectrum antibiotics are given to all patients in both groups from day of hospital admission till discharge.

Figure 1:

(a) computed tomographic (CT) axial scan, obtained at admission, demonstrating subarachnoid hemorrhage (SAH) and intracerebral hematoma (ICH) in a 54-year-old woman with Hunt and Hess Grade IV. CT angiogram (b) axial view (c) sagittal view demonstrating a small aneurysm at the anterior communicating artery. (d) intraoperative microscopic image showing fenestrated lamina terminalis (arrow). (e) intraoperative microscopic image showing the catheter inserted inside the fenestrated lamina terminalis after clipping of the ruptured aneurysm. Postoperative CT angiogram (f) axial view (g) coronal view (h) sagittal view demonstrating the tip of the catheter in the third ventricle.

Statistical analysis

We assessed the distribution and frequency of all demographic, radiologic, and clinical variables. Covariates were compared between Group A and Group B using the Chi-squared test for categorical variables, the Student’s t-test for parametric continuous variables, and Mann–Whitney U-test for nonparametric continuous variables. P < 0.05 was considered statistically significant. All statistics were calculated using SPSS software version 25 (SPSS, Inc., Chicago, Illinois).

RESULTS

A total of 83 consecutive patients underwent microsurgical clipping for ruptured anterior circulation IAs at the Neurosurgery Department in Tanta University Hospitals, Egypt, between July 2013 and July 2019. Two of them had Hunt and Hess Grade V. Ten patients required insertion of ventriculostomy tube on admission before the definitive clipping surgery. In three patients, the intracranial pressure was extremely high and CSF release from basal cisterns and lamina terminalis was not possible except after inserting a ventriculostomy tube inside the lateral ventricle from Paine’s point. These 15 patients were excluded from our study. The remaining 68 patients were included in our study where lamina terminalis fenestration was done during clipping surgery without other ventriculostomy. We divided our patients into two groups; Group A operated on between July 2013 and June 2016 with lamina terminalis opening but without inserting ventriculostomy tube during clipping surgery and Group B patients operated on between July 2016 and July 2019 with lamina terminalis opening and with inserting a translamina terminalis EVD tube during the clipping surgery.

Demographic information for patients in both groups are summarized in [Table 1]. There was no significant difference between patients in both groups as regard age or gender of patients, associated comorbidities, or locations of the ruptured aneurysm. [Table 2] shows clinical and radiological finding of patients in both groups. There was also no significant difference in the clinical grade, assessed by Hunt and Hess scale, of patients in both groups. There was also no significant difference in the CT finding of hydrocephalus or the amount of SAH, evaluated by Fisher’s grading system, of patients of both groups. Furthermore, there was no significant difference in timing of surgery for patients in both groups.

Table 1:

Demographic information of patients in both groups.

Table 2:

Clinical and radiological variables on admission in both groups.

Group A comprised 29 patients; 18 females and 11males, their age ranged from 30 to 65 years with a mean age of 47.8 years, 8 patients were smokers, 10 were hypertensive, 12 patients had diabetes mellitus, and 1 case had positive family history of aneurysmal subarachnoid hemorrhage [Table 1].

On admission, 7 patients had H&H classification Grade II, 15 patients Grade III, and 7 patients Grade IV. Admission brain CT showed subarachnoid hemorrhage in all patients, hydrocephalus in 5 patients, intracerebral hematoma in 16 patients, and intraventricular hemorrhage in 14 patients. The location of the ruptured IA, in this group, was clearly defined by preoperative cerebral CTA. The ruptured aneurysm was located on the anterior communicating artery in 14 patients, on the MCA in 9 patients, the posterior communicating segment of ICA in 3 patients, on the ophthalmic segment of the ICA in 2 patients, and on the ICA bifurcation in 1 patient [Table 2].

Most patients in this group underwent clipping surgery in the early period after aneurysm rupture. Seventeen patients were operated on within 24 h after aneurysm rupture, 7 patients in the 2nd day after rupture, and 5 patients were operated on late because of delayed referral and/or bad general condition at admission. In all patients, in this group, the lamina terminalis was fenestrated early in the operation with no associated complication [Table 2].

During the early, the first 21 postoperative days, follow-up: nine patients had hydrocephalus and conventional VP shunt was inserted, two patients died (one from vasospasm and one from ventilator associated pneumonia), one patient had hemiparesis, and another patient was aphasic. Twenty-seven patients were scheduled for 6 months clinical and radiological follow-up. By the end of the 6 months follow-up period, two patients were lost, 7 patients developed hydrocephalus, and a ventriculoperitoneal shunt was inserted.

Group B comprised 39 patients; 22 females and 17 males, their age ranged from 28 to 66 years with a mean age of 46.6 years, 12 patients were smokers, 15 were hypertensive, 14 had diabetes mellitus, and 1 case had positive family history of aneurysmal subarachnoid hemorrhage [Table 1].

On admission, 10 patients had H&H classification Grade II, 18 patients Grade III, and 11 patients Grade IV. Admission brain CT showed subarachnoid hemorrhage in all patients, hydrocephalus in 7 patients, intracerebral hematoma in 16 patients, and intraventricular hemorrhage in 22 patients. The location of the ruptured IA, in this group, was clearly defined by preoperative cerebral CTA. The ruptured aneurysm was located on the anterior communicating artery in 15 patients, on the MCA in 15 patients, the posterior communicating segment of ICA in 5 patients, on the ophthalmic segment of the ICA in 2 patients, and on the ICA bifurcation in 2 patients [Table 2].

Most patients in this group underwent clipping surgery in the early period after aneurysm rupture. Twenty-six patients were operated on within 24 h after aneurysm rupture, 8 patients in the 2nd day after rupture, and 5 patients were operated on late because of delayed referral and/or bad general condition at admission. In all patients, in this group, a translamina terminalis, EVD tube was inserted regardless presence or absence of hydrocephalus. We encountered no complication related to lamina terminalis fenestration or tube insertion.

During the early, the first 21 postoperative days, follow- up: eight patients had positive CSF findings for infection during daily CSF analysis and were treated by systemic and intraventricular antibiotic injection. Infection subsided in all patients. Eight patients developed hydrocephalus; the tube was obstructed by blood clots in two patients and slipped in two patients. Three patients failed weaning with evidence of neurologic decline and ventriculomegaly during tube occlusion. One patient developed hydrocephalus and neurologic decline 2 days after tube removal in the 12th postoperative day. Conventional VP shunt was ultimately inserted for those eight patients. One patient had lower limb monoparesis, another patient had 3rd nerve palsy, and one patient was aphasic. Two patients died (one from cardiopulmonary problems and one from electrolyte disturbances). Thirty-seven patients were scheduled for 6 months clinical and radiological follow-up. By the end of the 6 months follow-up period, two patients were lost, four patients developed hydrocephalus, and a ventriculoperitoneal shunt was inserted.

We used the modified Rankin score to document clinical outcome of our patients; Group A included 10 patients symptom free (Grade 0), seven patients Grade 1, three patients Grade 2, two patients Grade 3, and three patients Grade 5, while Group B included 18 patients Grade 0, 12 patients Grade 1, one patient Grade 2, two patients Grade 3, and two patients Grade 5. Two cases were lost from follow-up in each group and two cases have expired during their hospital stay in each group. Statistically, there was no significant difference in outcome of both groups. The incidence of permanent VP shunt dependence was significantly lower in Group B than in Group A patients (30.8% vs. 55.2%, respectively) [Table 3].

Table 3:

Outcome at the end of 6 months follow-up

DISCUSSION

Hydrocephalus is a common finding with aSAH and a major contributor to the occurrence of vasospasm with subsequent poor clinical outcomes.[9] The origin of development of early hydrocephalus after aneurysmal subarachnoid hemorrhage may be increased CSF viscosity causing communicating hydrocephalus or obstruction of CSF pathway by thick blood clots. Hydrocephalus can also develop after latent period due to arachnoid scarring decreasing CSF absorption and/ or causing its pathway obstruction. All efforts should be directed to treat and prevent hydrocephalus to improve clinical outcome.[18] The incidence of late hydrocephalus and shunt dependence following aSAH in the literature is highly variable.[7,9,16,18,21,36,38,41,43,45]

The lamina terminalis is a thin sheet of gray matter and pia mater stretches upward between the optic chiasm and rostrum of the corpus callosum. It forms the anterior wall of the third ventricle.[50] Its opening allows entrance to the third ventricle and release of significant amount of CSF, this facilitates aneurysm dissection without excessive brain retraction.[42] Microscopic connection of the third ventricle to the basal cisterns through the lamina terminalis in its anterior wall is similar to connection of the third ventricle to subarachnoid spaces through its floor through endoscopic third ventriculostomy and can be considered a solution for hydrocephalus.[22] The effect of lamina terminalis fenestration on reducing the incidence of development of chronic hydrocephalus after aSAH has been supported by some reports and refused by others.[3,5,12,16,22,29,44]

The cornerstone in operating on ruptured IAs is to have slack brain for minimal brain retraction at the same time of maintaining adequate cerebral perfusion. In some situations of massive subarachnoid hemorrhage (SAH) or presence of a large space-occupying intracerebral hematoma with angry brain and awkward neuroanesthetic measures, CSF release is the way of choice to decrease the intracranial pressure and minimize brain retraction.[37] CSF can be released from the lumbar subarachnoid space through an external lumbar drainage. It can also be released from the basal cisterns around the optic nerve and internal carotid artery. All these maneuvers may still be insufficient in some situations of angry brain.[3]

We used to open the lamina terminalis in all cases of ruptured anterior circulation aneurysms treated through pterional or lateral supraorbital approaches. We believe this allows release of significant amount of CSF from the third ventricle rendering the brain lax and facilitating aneurysm dissection and clipping with minimal brain retraction. Furthermore, it is considered similar to endoscopic third ventriculostomy and helps in management of hydrocephalus in the acute stage. Since July 2016, we started to leave a silicon tube inside the third ventricle through the fenestrated lamina terminalis and connect to EVD system to wash bloody CSF from the ventricles to outside of the body aiming to ultimately decrease arachnoid scarring. We typically open basal cisterns around the carotid artery and optic nerves then fenestrate the lamina terminalis early before starting dissection of the aneurysm for clipping. In one case of ruptured anterior communicating aneurysm, the dome was projecting downward into the lamina terminalis. Hence, lamina terminalis fenestration was performed after securing the aneurysm. The lamina terminalis tube, when used, was always inserted after final clipping of the aneurysm not to obscure the operative field.

The technique of lamina terminalis ventriculostomy during microsurgery for ruptured IAs has provoked some criticism. It was claimed that a conventional ventriculostomy through Kocher burr hole before surgery or intraoperative ventriculostomy through the Paine’s point would be more reasonable to avoid retraction of swollen brain, especially when there is no need to approach the midline. The big advantage of lamina terminalis ventriculostomy is that it is done under direct vision which avoids all possible complications of other blind ventriculostomy techniques. We also avoid performing extra burr holes and the tube can be inserted safely and CSF drained regardless the size of the ventricle. Furthermore, frontal lobe is retracted for only short period till lamina terminalis is reached and there was no complication from this brief retraction.[24,28]

Lehto et al.[24] reported their experience with insertion of lamina terminalis tube during microsurgery of 78 patients with ruptured anterior circulation aneurysms. They had no procedure-related complications, about 10% of their patients needed revision because of either tube obstruction (7 patients) or tube dislodgement (one patient). They documented positive CSF culture for 7 patients (9%) who needed antibiotic treatment. Twenty-two patients (28%) in their series required a permanent shunt.

The results of Lehto et al.[24] are close to the results of lamina terminalis tube insertion in Group B of our patients. Although we used the regular ventricular catheter which is thicker and more rigid than the special thin catheter they used, we did not have any complication related to lamina terminalis tube insertion. We had higher rate of positive CSF culture (21%) despite our routine use of prophylactic antibiotics while they did not use antibiotic prophylaxis. This might be related to higher temperature and humidity in our country. About 18% of our patients required revision and ultimate shunt insertion in the early postoperative period because of tube obstruction in two patients, tube dislodgement in two patients, failed weaning in three patients, and hydrocephalus in one patient 2 days after tube removal. About 31% of our patients ultimately needed permanent CSF shunt which is very close to results of their series (28%).

In the literature, there is broad variation in the incidence of ventriculostomy-related infection ranging from 1% to 27%. There is also wide difference in management strategies. Among these strategies are the catheter exchange which cannot be simply done for the lamina terminalis ventriculostomy tube, while it can be simply done for conventional ventriculostomy tubes. The reported data do not justify this catheter exchange for infection prophylaxis as it does not reduce the risk of CSF infection.[4,6,10,14,20,25-27,34,35,39,40,48,52] We used to give prophylactic systemic antibiotic to all our patients with ventriculostomy and when CSF culture is positive, we start intraventricular antibiotic injection through the catheter. We did not have neurological sequelae or death related to catheter infection.

Cerebral CTA was our imaging modality of choice for diagnosis and surgical planning in this series. We believe that multislice helical CTA is more practical than digital subtraction angiography because it is highly sensitive, easily available at all times, less time consuming and is less invasive. CT images can easily demonstrate hydrocephalus, intracerebral hematoma, intraventricular hemorrhage, and even calcification in the vessel wall or the aneurysm. The 3D reconstruction helps to show the surgical view and aneurysm relation to bony landmarks as the anterior clinoid process.

There have been some features reported to be associated with increased incidence of development of chronic hydrocephalus after aSAH including acute stage hydrocephalus, poor Fisher’s grade, and poor Hunt and Hess grade.[3,16,22,36,41-43,45]

We had no significant difference (P > 0.05) between both groups regarding these factors which might influence the rate of development of chronic hydrocephalus [Table 2]. This strongly suggests that the difference in incidence of chronic hydrocephalus between both groups was related to the effect of lamina terminalis EVD. In Group A, 7 patients (43.8%), of the 16 cases who passed the acute stage without shunt insertion and who were available at the end of 6 months follow-up, developed hydrocephalus and permanent shunt was inserted. In Group B, 4 patients (14.8%), of the 27 cases who passed the acute stage without shunt insertion and who were available at the end of 6 months follow-up, developed hydrocephalus and permanent shunt was inserted. Insertion of lamina terminalis tube was associated with statistically significant (P < 0.05) lower incidence of late hydrocephalus. Finally, Group B patients had statistically significant (P < 0.05) lower incidence of chronic shunt-dependent hydrocephalus than Group A patients (30.8% vs. 55.2%, respectively) [Table 3].

Reviewing the literature regarding the histopathological basis[8,21,51], clinical and radiological predictors[11,15,18,19,30,33,36,41,47] of chronic hydrocephalus after aSAH suggest that the disease is complex and multifactorial requiring innovations to prevent and treat hydrocephalus without resorting to CSF shunting. Although there have been some reports suggesting that some drugs as cilostazol[31] or some surgical maneuvers as fenestration of lamina terminalis[5,22,29] or tandem fenestration of lamina terminalis and Lilliquist membrane[3,42] can reduce the development chronic shunt-dependent hydrocephalus after aSAH, the situation is still challenging. We believe that some extra few minute consuming steps, during open microsurgical clipping, to reduce the clot burden and to accelerate drainage of subarachnoid blood products, could decrease the rate of chronic shunt-dependent hydrocephalus after aSAH without exposing the patients to more complications.

Limitations

The results of the present work must be interpreted in the light of small number of cases and the retrospective nature of the study. Prospective studies with larger number of cases are recommended to overcome these limitations in forthcoming research.

CONCLUSION

EVD through fenestrated lamina terminalis is safe and may be effective in decreasing the incidence of chronic shunt- dependent hydrocephalus after aSAH.