Key-words:
Cerebrospinal fluid diversion - endoscopic third ventriculostomy - endoscopic third
ventriculostomy complications - fornix injury - hydrocephalus - image guidance
Introduction
Cerebrospinal fluid (CSF) diversion through shunts has been used for both communicating
and noncommunicating hydrocephalus. The procedure constitutes one of the most complex
challenges in the neurosurgical practice. Shunt hardware has seen significant advances
in the last decade. Diversion by endoscopic third ventriculostomy (ETV) has been in
practice for nearly a century now.[[1]] It has recently been rejuvenated with advancements in endoscopic techniques including
three-dimensional and high-definition images and other optical enhancement measures.
In 1923, Mixter, a urologist, performed the first ETV using a urethroscope for the
procedure.[[2]] However, such primitive attempts at the procedure were encountered with failure
and were not deemed ideal for therapeutic intervention. This changed with recent advances
in endoscopic equipment and instruments when ETV was brought to focus.
The procedure involves introducing a rigid endoscope typically through a burrhole
placed 0.5 cm anterior to the coronal suture and 2.5 cm lateral to the midline.[[3]],[[4]],[[5]] This is maneuvered to approach the floor of the third ventricle through the foramen
of Monro and is perforated between the infundibular recess anteriorly and mammillary
bodies posteriorly. The aim of the surgery is to visualize the contents of interpeduncular
cisterns, including the basilar artery, the brainstem, and dura of the clivus.[[6]]
The complications of the procedure include hemorrhage secondary to basilar artery
rupture and injury to neural structures, including the hypothalamus, pituitary, and
fornix. Several studies have reported anterograde amnesia secondary to forniceal injury.[[7]] All these complications must be taken into consideration while opting for an endoscopic
procedure for the management of hydrocephalus.
The aim of our study was to see whether image guidance would alter the incidence of
neural and hemorrhagic iatrogenic injuries when trying to reach the third ventricle
through foramen of Monro. Image guidance can thus help young neurosurgeons to overcome
the learning curve associated with this procedure.
Materials and Methods
This was a prospective, descriptive, and observational study conducted in the neurosurgery
department of Liaquat National Hospital in Karachi, Pakistan. A total of 43 patients
were included in the study who underwent ETV between June 2015 and December 2018.
The recorded information included age, sex, cause of hydrocephalus, and intraoperative
complications. All patients with a previous history of head injury and those who underwent
ETV without the image guidance trajectory planning were excluded from the study.
The intraoperative complications were divided into three major groups, and the results
were analyzed using the SPSS software version 23 (IBM Corp. Released 2015. IBM SPSS
Statistics for Windows, Version 23.0. Armonk, NY: IBM Corp.). The results of this
study were compared with the complications reported in historical studies conducted
on ETV with and without intraoperative image guidance usage [[Table 1]] and [[Table 2]].
Table 1: Comparison of our complications with studies conducted on endoscopic third ventriculostomy
procedures without image guidance
Table 2: Comparison of complication rates of image-guided endoscopic third ventriculostomy
Patients were divided into three groups according to the complications developed.
Group A included patients with hemorrhage including rupture of basilar artery or its
branches. Group B included patients developing venous hemorrhages, whereas Group C
included patients with neural injuries. Group C was further divided into C1 (injury
to oculomotor nerve), C2 (injury to hypothalamus), and C3 (injury to fornix). Memory
assessment was done with Mini-Mental State Examination (MMSE) score. Pre- and post-operative
scores were assessed and compared. The normal range of MMSE score was established
in the range of 25–30.[[8]] Null hypothesis and alternative hypothesis were put in place. Alternative hypothesis:
MMSE scores stay within the normal range postoperatively. Null hypothesis: MMSE scores
do not stay within normal range postoperatively. P < 0.5 was considered statistically
significant.
Surgical technique
A preoperative magnetic resonance imaging (MRI) of the brain with image-guided protocol
was done and uploaded on the neuronavigation system. All patients were operated under
general anesthesia in the supine position. The ideal trajectory and incision site
were identified before incision thus making sure that neural structures along the
path were not manipulated. Basilar artery was identified in the sagittal section of
MRI. A center point of third ventricular floor was pointed out as site of ventriculostomy.
Care was taken to avoid making the stoma close to basilar artery and thus avoid rupture.
A line was drawn from this point to the foramen of Monro and then extended up to the
scalp. This point marks the optimal entry point for the procedure. Once we had the
burr hole site, the rest of the procedure was done based on the description provided
by Sainte-Rose and Chumas.[[9]] An operating sheath was inserted, and a 0° endoscope was introduced through it
after removing the stylet. Anatomical structures including the foramen of Monro, choroid
plexus, thalamostriate, and septal veins were identified, and the sheath was advanced
through the foramen down to the floor of the third ventricle. It is perforated using
a blunt instrument such as bipolar probe, between infundibular recess in front, and
mammillary bodies behind. Bipolar diathermy is used with caution when the floor of
the third ventricle is opaque. This is to avoid iatrogenic injury to underlying structures.
A wide enough opening in the floor of the third ventricle must be made. An endoscope
is inserted through the prepontine cistern to see the presence of a secondary membrane.
If present, a blunt probe can be used to puncture this membrane. Judicial use of a
Fogarty balloon catheter may be done to dilate the stoma. It is important to visualize
the basilar artery and cranial nerves. Underlying secondary arachnoid membranes or
adhesions are opened up in the same fashion.
Image guidance is used at all stages of surgery, starting from the burr hole entry
through the third ventricle to navigating inside the brain. While removing the endoscope,
pulsatile movement of the surgical stoma was identified to confirm the free-flowing
CSF through fenestration connecting the third ventricle and the prepontine subarachnoid
space. Special attention was directed toward looking for contusions or injuries at
the time of procedure and any difficulty in moving the scope through the foramen of
Monro.
Results
We included 43 patients with ETV performed between June 2015 and December 2018. There
were 34 male and 9 female patients. The mean age of the patients was 31.3 years.
The intraoperative complications encountered among these patients are illustrated
in [[Table 1]]. The study reveals that Group C3 (injury to fornix) included all of the complications
encountered during this study. No other complications were noted in any of our patients
operated.
None of the patients developed postoperative memory deficit. The average preoperative
MMSE score of 43 patients was 27.023. The average postoperative MMSE score was 27.79.
Data were analyzed, and a statistically significant P < 0.5 was found thus rejecting
the null hypothesis. No patients scored below 25 indicating that none developed memory
impairment.
Discussion
Mixter reported the use of an endoscope to perform third ventriculosotomy back in
1923.[[2]] Since then there has been a significant modification in endoscopes and their usage,
technological advances in endoscopic surgery and MRI functionality has improved the
safety of surgical procedures in neurosurgery. Safe ETV, in children and adults both,
is a fine example of such advancements.[[10]]
Intraoperative complications are a common occurrence during ETV. These complications
occur secondary to intraoperative maneuvers including intra-operative endoscopic manipulations,
irrigation, and perforation of the third ventricle. There have been several studies
demonstrating early complications of ETV. Short term (<6 months) and intermediate
term (<3 years) success have been shown to be quite high, but it depends upon several
risk factors including the age of the patient, previous shunt, and the etiology of
the hydrocephalus.[[3]],[[7]],[[10]],[[11]],[[12]]
Kulkarni et al. have done extensive work on the short-term success of the ETV and
calculated their results using ETV success score and has shown good outcome.[[10]] Matthew et al. has shown a comparable result with a 61% and 47% success rate within
6 months and 3 years of follow-up, respectively.[[13]]
In order to reduce the complications and increase the success rate of ETV several
innovations were introduced over the time. The use of stereotactic techniques while
performing third ventriculostomy was first described by Hoffman et al.[[14]] and was modified through computed tomography by Kelly[[15]],[[16]] in the 80's. However, MRI-guided stereotactic techniques for this procedure have
not been described frequently in the literature.
We noticed that the standard entry point was not giving us an ideal trajectory and
thus increasing the risk of complications. The introduction of image guidance significantly
alters our entry point [[Figure 1]] and improved outcomes. It provides a safe trajectory for the endoscope to advance
through the cortex into the lateral ventricle and then navigate safely through the
foramen of Monro.
Figure 1: Image guidance significantly alters the entry point from the conventional burr hole
site. Shaded line: Indicates the coronal suture. The centre of incision marks the
Kocker’s point where an EVD was done in this patient previously and is also the site
of conventional endoscopic third ventriculostomy incision. Cross mark indicates the
entry point for endoscopic third ventriculostomy defined by the image guidance system
which is significantly altered in relation to Kocker's point
The technique involves conducting an MRI on image-guided protocol and then performing
the standard endoscopic surgery with the addition of image guidance. The system diverts
the trajectory of the endoscope from the usual pericoronal approach to avoid the possibility
of iatrogenic injury in that path. It defines a new altered trajectory for the endoscope
to follow, rendering it safer and more accurate. The conventional entry point usually
utilized for this surgery lies at the Kocker's point, 1 cm anterior to the coronal
suture on the mid pupillary line.[[17]] In our cases the image guidance altered this trajectory to a minimum of 1 cm behind
the coronal suture, thus avoiding frequent complications.
Wrong trajectories secondary to the conventional blind procedure can cause iatrogenic
complications [[Figure 2]]. An anterior trajectory might cause damage to structures underlying the floor of
the third ventricle. A posterior trajectory, on the other hand, might cause iatrogenic
injury to the motor cortex and fornix. Similarly, a lateral entry point can cause
iatrogenic injury to the fornix. Image guidance can significantly reduce the number
of such complications as shown in this study.
Figure 2: Too anterior entry point can cause injury to fornix and brainstem. Too posterior
entry point can cause injury to fornix and motor cortex. Correct entry point is essential
to avoid such complications
Previously, intra-operative complications without using any stereotactic technique
have been reported between 5% and 30%, and mortality has been reported up to 1%.[[18]] Our study, however, reveals minimal complications occurring in the patients undergoing
image-guided ETV with only 2 (4.65%) patients developing fornix contusion [[Figure 3]]. In one of these patients, the diameter of Foramen of Monro was narrower than the
endoscope. In the second case, the burr hole was made more than 1 cm anterior to the
proposed site secondary to the iatrogenic error. None of the patients developed any
other neural iatrogenic injury or hemorrhagic complications. These injuries occur
due to the long-learning curve as the neuroendoscopic procedure involves complicated
maneuvers to be carried out.[[19]] Regular usage of image guidance can help young neurosurgeons develop skills to
avoid such complications and their clinical implications.
Figure 3: Endoscopic view of endoscopic third ventriculostomy showing forniceal Grade 1 contusion
Literature reports incidents of multiple other complications intraoperatively. Hemodynamic
disturbance secondary to iatrogenic injury to posterior hypothalamus which regulates
cardiovascular functions has been reported most commonly.[[18]] This was not seen in our patients.
Magnetic resonance image guidance to plan the trajectory of the endoscope proved to
be a safe and useful technique in our study compared to other studies conducted on
ETV being performed with the blind endoscopic approach [[Table 1]]. Oertel et al.[[20]] have reported a 16.4% incidence of fornix contusion in patients undergoing ETV
without image guidance compared to 4.65% in our study. They reported a 2.3% oculomotor
nerve injury while no such complications were observed in our study. The navigational
technique significantly alters the plan of trajectory thus creating a route with least
possibility of iatrogenic injury. Martínez-Moreno et al. developed advanced navigational
protocol for ETV and applied it on their patients. They had a fornix contusion without
clinical correlation in one patient [[Table 2]].[[21]]
The left fornix is functionally involved in the verbal memory, while the right fornix
is involved in the visuospatial memory.[[22]] The fornix carries fibres from both the caudal and rostral ends of the hippocampus.
Both these fibers are separately distributed along the length of the fornix with caudal
fibres being carried on the medial aspect and rostral fibres being more lateral. The
lateral fibres are concerned with interoceptive signals that are involved with learning,
emotional and motivational memory. The medial fibers are mostly involved with exteroceptive
processing that relates to scene learning (recognition of objects within a spatial
context.[[22]] Considering these functional parameters, we assessed the patients postoperatively
for neurocognitive dysfunction through MMSE. None of the patients experienced memory
impairment.
Our study incites the need for further large-scale studies to be conducted to establish
safety and efficacy of image guidance in ETV procedure.
Conclusion
Our study reports no major neurological complications secondary to image-guided ETV.
It is thus concluded that the use of image guidance to plan the trajectory for ETV
improves safety and accuracy of the procedure compared to conducting ETV without the
use of image guidance system (IGS). Young neurosurgeons can benefit from this technique
to overcome the learning curve and develop better surgical skills to avoid iatrogenic
complications. We recommend the use of IGS for all cases of ETV, when available.
