Robot-guided Ventriculoperitoneal Shunt in Slit-like Ventricles
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.314585
Source of Support: None, Conflict of Interest: None
Keywords: Idiopathic intracranial hypertension, robotic guidance, ROSA, slit ventricles, small ventricles, VP shuntKey Message: Robotic guided placement of ventriculostomy catheter is a safe and accurate method even in the presence of very small ventricles. In comparison to the conventional stereotactic methods the robotic technique provides a stable arm and can be performed with just a plain CT brain.
Ventriculoperitoneal shunt (VPS) is the most common CSF diversion procedure for hydrocephalus due to various etiologies. This is done using the standard points described in the literature depending upon the site of entry. Frazier's point is used for occipital, Keens for parietal, and Kocher's for frontal horn ventriculostomy. In patients with hydrocephalus, ventriculostomy is commonly performed using the free-hand technique. It becomes challenging in the presence of slit-like/small ventricles. This situation is frequently encountered in cases of idiopathic intracranial hypertension (IIH) with small ventricles. In these situations, the standard practice is to insert lumbo-peritoneal shunts (LPS), albeit, with attendant complications like secondary Chiari and syringomyelia, very well-documented in the literature., With the advancements in stereotactic techniques, VPS can be accomplished even in difficult to cannulate ventricles.,,, We report here our experience and technique of VPS insertion using robotic guidance in three cases.
Robot-guided VPS was inserted in three patients between 2016 and 2019.
A 1-year-old female child presented with a progressively increasing subcutaneous swelling in the suboccipital region since birth. The swelling increased on crying, there was no history of discharge from the swelling. On examination, the swelling was cystic in consistency with intact skin over the swelling. The transillumination test was positive suggestive of an occipital meningocele.
Imaging revealed partial vermian agenesis, an occipital meningocele in continuity with the fourth ventricle. There was no evidence of tonsillar herniation or hydrocephalus noted.
This patient underwent meningocele repair and on day 5, following the repair, she developed a florid CSF leak from the operative site. CT of the brain showed no evidence of hydrocephalus but rather very small ventricles. She was reoperated and repair of the wound was done using fascia lata graft and lumbar drain was placed using an epidural catheter, as the child was very small. On day 2, following the second surgery, CSF leak continued to occur. With limited options available we planned to divert the ventricular CSF followed by wound repair simultaneously. In view of the small ventricles, robotic guidance for placing the external ventricular drain (EVD) following the repair of the wound was planned. The dural tissue was inflamed and friable and hence watertight closure could not be achieved, rather the edges were approximated loosely with an overlay of fascia lata graft. Right frontal EVD was placed using robotic guidance. Postoperative CT showed accurate positioning of the catheter tip within the ventricle [Figure 1]. The EVD was left in situ for a week with continuous CSF drainage. The meningocele wound had healed completely and the EVD was then internalized (converted into a VPS). The patient was discharged following this and is doing well at 1-year follow-up.
A 38-year-old male with complaints of recurrent headaches and vomiting for 2 months, presented with history of rapidly progressive vision loss over 15 days. On evaluation, He was found to have perception of light (PL+) only, in both the eyes. The fundus examination revealed grade 3 papilledema in both the eyes. MRI of the brain showed features suggestive of IIH with very small ventricles. Lumbar puncture revealed an opening CSF pressure of 30 cm CSF. This patient underwent, robot-guided right parietal, medium pressure VPS. The ventricle was hit in a single attempt and surgery was uneventful. The postoperative CT brain showed accurate positioning of the catheter into the frontal horn [Figure 2]. The patient's headache relieved following surgery with improvement in vision.
A 32 years old male patient presented with history of chronic headache of 6 months duration and rapidly progressive vision loss over a period of 3 weeks. Examination revealed finger counting close to the face and bilateral papilledema. The lumbar CSF opening pressure was 28 cm of CSF. MRI revealed features suggestive of IIH. He underwent robot-guided right parietal VPS with postoperative CT of the brain showing the accurate placement of the catheter in the ventricle [Figure 3].
Video Link: https://youtu.be/QNclX2XYkHw
Video timeline with audio transcript:
Acquisition of images
Once the CT brain has been acquired as per the protocol, they are imported into the robotic software system. The trajectory for VPS is planned on the robot in a 3-dimensional plane. This process is done prior to the patient being taken up for surgery so as to minimize the duration of surgery.
After induction with general anaesthesia, the patient is placed supine with the head rotated 90-degrees, contralateral to the site of VPS insertion in case of a parieto-occipital shunt and supine with head in neutral position for frontal horn ventriculostomy. Head is fixed with Mayfield head-clamp.
Once the patient is positioned, the robot is positioned for registration using the laser facial scanning. It works on the principle of frameless stereotaxy. All the surface landmarks over the face particularly the nose-tip, malar eminence, and the forehead are scanned. Following this process, the accuracy of the registration is checked. This technique permits preoperative selection and examination of the catheter trajectory. Using conventional triplanar images, as well as reconstructed planes perpendicular to and along the trajectory, the precise entry and target points are planned before commencing the operation. Once the trajectory is chosen, it can be inspected for proximate hazards using the probe view such as avoiding the sulci and vessels along the planned trajectory.
Following registration, the robot is driven to the destined trajectory with the laser pointer in situ, so that the laser point is seen on the skin (entry point). The standard curvilinear incision is marked around the laser point. Paramedian horizontal incision marked over the abdomen just above the umbilicus. The patient is prepped and draped in the standard fashion.
Skin incision over the scalp is made and flap elevated to expose the bone. Burr hole fashioned in the center of the exposure. The peritoneum was opened through the abdominal incision. Subcutaneous tunneling made connecting both the cranial and abdominal incisions. The distal end of the shunt is passed through the tunnel and secured. The dura is coagulated and incised in cruciate fashion. The robotic arm is then driven to the planned trajectory. Once the arm is positioned, the adapter of appropriate size is inserted into the arm and is moved closer to the burr hole in the axial plane. The distance to the target calculated by the ROSA (Zimmer Biomet Warsaw, Indiana) software, is noted . This distance is corresponds to the length from the top surface of the adapter to the target point planned on the software preoperatively. A ventricular catheter is inserted after measuring the adequate length of the catheter. Once the ventricle is hit, the catheter is advanced further to the measured length and stylet is withdrawn. The rest of the procedure is completed in the standard fashion.
The ventricles were hit in a single attempt using this method in all the cases. No procedure-related complications were encountered. Postoperative CT scans showed the accurate placement of the VPS in all the cases. At 3 months follow-up, both the patients with IIH had significant relief in headache and improvement in the vision. while the CSF leak subsided and wound healed in case-1.
Experience with both frame-based and frameless stereotaxy has been published in the literature for placing VPS. Most of the studies have focused on the accuracy of the placement of stereotactic-guided ventriculostomy catheters in patients with hydrocephalus and its association with proximal end obstruction. The literature on the number of passes, published on bedside ventriculostomies, performed by neurosurgical residents described a mean of 1.5 freehand passes; 53% of the patients in the study had hydrocephalus, and only 12% had slit ventricles. In another study, image-guided catheter ventriculostomy in 36 patients with IIH described a 10% incidence of multiple passes. In a meta-analysis, on image-guided ventricular catheterization, Nesvicket et al. showed that there was no statistically significant difference in accuracy between freehand versus navigation-guided shunt placement in enlarged ventricles. They also concluded that frameless stereotaxy has a definitive role in small/slit-like, dysmorphic ventricles.
Tulipan et al. published VPS insertion in slit-like ventricles accompanying IIH in 7 patients using a frame-based stereotaxy technique. Using the Kocher's point, the ventriculostomy catheter could be accurately inserted in a single pass. There were no proximal catheter obstructions noted and all patients except one had symptomatic relief. Prolonged surgical time and placement of rigid frame prior to performing the procedure were the major disadvantages. To overcome these disadvantages, the focus has shifted to using frameless stereotaxy (Optoelectric and Electromagnetic) navigation techniques. Four series involving more than 10 patients have demonstrated very good accuracies without any procedure-related complications with this technique. All these studies involved patients diagnosed with IIH in the presence of small ventricles. Symptomatic relief was achieved following the shunt surgery. Shunt obstructions due to distal catheter block were noted in these studies, however, there were no proximal catheter obstructions noted in these studies during the follow-up suggesting accurate positioning of the ventricular end.,,,
Neurosurgical procedures have become more safe and precise in recent years with advancements in technology. With the advent of Neurorobots, there has been a steady increment in the performance of stereotactic procedures for complex pathologies. The ideal situation for robotic use is for accessing deep-seated areas of the brain not suitable for open surgeries; like placement of Stereoelectroencephalography (SEEG) leads, deep brain stimulation (DBS), and biopsy of deep-seated brain lesions.,,, Since the introduction of the robotic system, there has been an expansion in the surgical indications, particularly endoscopic surgeries involving Sellar/Suprasellar, Intraventricular lesions, Endoscopic corpus callosotomies, and Hemispherotomy.,, Literature is replete with various neurosurgical procedures performed using robotic assistance. Lollis and colleagues published their experience on reservoir insertion into the ventricles for administering chemotherapy in patients with primary CNS lymphoma.,
At our institution, we have been using a robot (ROSA, Zimmer Biomet, Warsaw, Indiana) for performing SEEG, Hemispherotomies, Lobar/multi-lobar disconnections, Hypothalamic Hamartoma disconnection using radiofrequency thermal ablation, bilateral cingulotomy for refractory obsessive compulsive disorder, Endoscopic pituitary surgeries, excision of Intraventricular tumors and Biopsies involving deep brain lesions.,,,, We expanded the indication to the placement of ventricular shunts in patients with small ventricles. One patient with recalcitrant CSF leak and very small ventricles following repair of the occipital meningocele was taken up for ventricular shunt as a desperate measure. Similarly, the other two patients with IIH and very small ventricles were subjected to robot-guided VPS, this was decided against LPS, as we had an experience of symptomatic secondary Chiari with cervico-dorsal syrinx formation in a patient of IIH following LPS [Figure 5].
Grit et al. in a retrospective analysis compared the efficacy and complications between LPS and VPS. Although there was good symptomatic relief between both the groups, this was only for a short period in the LPS group. This was due to the fact that the failure rate was higher in the LPS compared to the VPS group, which was statistically significant. Johnston et al. reported the occurrence of symptomatic acquired Chiari 1 malformation with or without syrinx in 14 patients following LPS for various etiology. Similarly, Chumas et al. reported six cases of symptomatic Chiari-1 following LPS in 143 patients requiring suboccipital decompression. Although considered rare, this complication has been recognized increasingly with a better understanding of the pathophysiology involved in its formation.
A robotic device like ROSA (Zimmer Biomet Warsaw, Indiana) with its inbuilt navigation system aids in careful study and review of a surgical plan on its workstation before the operation begins. This planning can be done with just a plain CT head, without requiring MRI brain and is invaluable in identifying and minimizing all possible hazards. Other advantages include; the stability of the robotic arm and also the preplanned length of the catheter (distance from entry to target point) to be inserted is automatically calculated and displayed by the robotic software and the ease of use, along with all the other advantages of a conventional image-guidance system.,,,,, The disadvantages of this technique relate to the limited availability of the robot in most centers owing to its high cost. The robotic device definitely adds up to the neurosurgical armamentarium. This is a pilot study, aims to provide the technique and also a proof of concept for inserting VPS using ROSA in difficult to cannulate ventricles accurately. We intend to continue this study including more number of patients and a long-term follow up in future.
More studies can be expected in the future for validating this technique and also comparing it with the conventional navigation methods.
Robotic ventricular shunt placement is a safe, effective, and precise technique. The ventriculostomy catheter can be inserted through a parietal burr hole accurately, thereby avoiding the need for changing the position of the patient, as seen in frontal ventriculostomy. This can be effectively used in patients having slit-like or difficult to cannulate ventricles with ease just based on plain CT of the brain.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]