Search

Neuroendoscopic resection of colloid cysts has gained tremendous popularity over the last 2 decades because of good clinical outcomes and a low complication profile. However, in comparison to microsurgical resections, endoscopic resection has a lower rate of gross-total resection, which leaves the patient at risk for cyst recurrence. At present, there is still ongoing debate as to the best surgical approach for colloid cysts. Endoscopic resection as a technique has to compete with the good outcomes of microsurgical resections with respect to a long-term recurrence-free outcome. It is the authors’ belief that gross-total resection should be the aim of endoscopic cyst resection. In this technical note, they describe their surgical technique for achieving safe gross-total resection of colloid cysts by using a ventriculoscopic system. The surgical technique includes a far anterolateral entry point, navigational guidance, bimanual sharp dissection, use of the endoscopic sheath as a retractor, the small-chamber irrigation technique, and the dry-field technique for hemostasis.

OBJECTIVE Mineralized or desiccated colloid cysts pose some unique challenges to endoscopic removal. The extrusion of the solid matrix into the intraventricular compartment has not been previously reported and, as such, no guidance exists regarding its predilection, prevention, and fate. METHODS Postoperative imaging studies in a registry of patients undergoing endoscopic removal of colloid cyst were reviewed to detect any solid matrix within the ventricular compartment. Preoperative images and operative notes were used to determine if any features were predictive. Serial postoperative images and clinical notes were used to characterize the implications of these findings. RESULTS From a review of 94 patients, 10 (10.6%) patients had evidence of an extruded intraventricular solid fragment (median follow-up 4 months; range 0.5–115 months). Of the evaluable patients, 7 of 9 patients had T1-weighted hyperintense and T2-weighted hypointense cysts on preoperative scans. Seventy-eight percent of the extrusions were on the same side as the endoscopic entry. Three patients demonstrated early fragment migration, but not after 8 months of radiological follow-up. All evaluable patients demonstrated improvement in their hydrocephalus, and none suffered a complication attributable to the intraventricular extruded fragments. CONCLUSIONS Intraventricular extruded colloid fragments can occur after endoscopic resection, with the possible risk demonstrated as cyst hypointensity on preoperative T2-weighted images. The finding does not seem to result in any clinical morbidity, and radiographic involution is the rule. Migratory capacity, however, does exist and justifies a more frequent imaging surveillance schedule and consideration for removal.

Intraventricular rupture of a colloid cyst is a rare phenomenon and has been proposed as a mechanism for sudden death in patients with colloid cysts. Imaging of a colloid cyst during rupture has been described in only one other instance. The authors report a highly unusual case of a 53-year-old man who presented with acute onset headaches and imaging findings of hydrocephalus caused by a colloid cyst originating from the septum pellucidum and superior surface of the roof of the third ventricle. Interestingly, the colloid cyst revealed imaging signs of intraventricular rupture characterized by a tail-like drainage of cystic contents into the occipital horn of the lateral ventricle. The patient was surgically treated with a craniotomy and transcallosal approach to the colloid cyst, where it was noted that the cyst wall was spontaneously open. This rare case highlights unique imaging findings of a rare event in an infrequent pathology confirmed with intraoperative microscopy. The authors further document the process of cyst rupture and speculate on its pathomechanisms.

OBJECTIVEEndoscopic resection of third-ventricle colloid cysts is technically challenging due to the limited dexterity and visualization provided by neuroendoscopic instruments. Extensive training and experience are required to master the learning curve. To improve the education of neurosurgical trainees in this procedure, a synthetic surgical simulator was developed and its realism, procedural content, and utility as a training instrument were evaluated.METHODSThe simulator was developed based on the neuroimaging (axial noncontrast CT and T1-weighted gadolinium-enhanced MRI) of an 8-year-old patient with a colloid cyst and hydrocephalus. Image segmentation, computer-aided design, rapid prototyping (3D printing), and silicone molding techniques were used to produce models of the skull, brain, ventricles, and colloid cyst. The cyst was filled with a viscous fluid and secured to the roof of the third ventricle. The choroid plexus and intraventricular veins were also included. Twenty-four neurosurgical trainees performed a simulated colloid cyst resection using a 30° angled endoscope, neuroendoscopic instruments, and image guidance. Using a 19-item feedback survey (5-point Likert scales), participants evaluated the simulator across 5 domains: anatomy, instrument handling, procedural content, perceived realism, and confidence and comfort level.RESULTSParticipants found the simulator’s anatomy to be highly realistic (mean 4.34 ± 0.63 [SD]) and appreciated the use of actual instruments (mean 4.38 ± 0.58). The procedural content was also rated highly (mean 4.28 ± 0.77); however, the perceived realism was rated slightly lower (mean 4.08 ± 0.63). Participants reported greater confidence in their ability to perform an endoscopic colloid cyst resection after using the simulator (mean 4.45 ± 0.68). Twenty-three participants (95.8%) indicated that they would use the simulator for additional training. Recommendations were made to develop complex case scenarios for experienced trainees (normal-sized ventricles, choroid plexus adherent to cyst wall, bleeding scenarios) and incorporate advanced instrumentation such as side-cutting aspiration devices.CONCLUSIONSA patient-specific synthetic surgical simulator for training residents and fellows in endoscopic colloid cyst resection was successfully developed. The simulator’s anatomy, instrument handling, and procedural content were found to be realistic. The simulator may serve as a valuable educational tool to learn the critical steps of endoscopic colloid cyst resection, develop a detailed understanding of intraventricular anatomy, and gain proficiency with bimanual neuroendoscopic techniques.

OBJECTIVERemoval of colloid cysts of the third ventricle using a purely endoscopic method has been established as a safe and advantageous technique. It is hypothesized that endoscopic removal in recurrent cases might pose more technical challenges and result in less success. The objective of this study was to assess the feasibility and outcomes of using a purely endoscopic approach for the management of recurrent colloid cysts compared to primary cysts.METHODSA retrospective cohort study was performed on patients who underwent purely endoscopic removal of their colloid cyst. Descriptive statistics were compared for patients undergoing surgery for a recurrent cyst and those for a control cohort undergoing surgery for a primary cyst. Bivariate analysis was conducted using a Fisher’s exact test for categorical variables and Mann-Whitney U-test for continuous variables.RESULTSIn total, 121 patients had a primary colloid cyst endoscopically removed and 10 patients had a total of 11 recurrent cysts removed. Recurrence or progression after surgery occurred in 3 (2.5%) cases in the primary cyst group and 2 (18.2%) cases in the recurrent cyst group. Symptomatic presentation during the follow-up period occurred in 6 (54.5%) cases in the recurrent cyst group versus 75 (62%) cases in the primary cyst group (p = 0.749). Two patients (20%) in the recurrent group had a second recurrence in a mean period of 30 months (1 patient at 15 and 1 patient at 45 months). One of these patients required a tertiary endoscopic removal 8 years after the second resection. No immediate postoperative complications or new morbidities were observed after repeat endoscopic surgery. The authors’ findings indicated a nonsignificant trend toward a higher recurrence rate (18.2% vs 2.5%, p = 0.055) and a decreased proportion of complete removal (90.9% vs 81.8%, p = 0.296) in the recurrent cyst group compared to the primary cyst group. However, a significantly higher rate of preoperative hydrocephalus was observed in the primary cyst group compared with the recurrent cyst group (63.6% vs 18.2%, p = 0.007).CONCLUSIONSPurely endoscopic approaches for the removal of recurrent colloid cysts of the third ventricle are feasible and equally safe compared with endoscopic removal of primary cysts. The study’s findings did not show a statistically significant difference in the rate of recurrence between the 2 groups. The proportion of patients with symptomatic cysts on presentation was lower in patients with recurrent cysts than in patients with primary cysts. Due to the high rate of complete removal with negligible morbidity, the authors continue to advocate for an endoscopic removal at the time of cyst recurrence.

The authors report the case of a 15-year-old girl with a third ventricle colloid cyst. She presented with prolonged headache, nausea, vomiting, and loss of visual acuity with bilateral papilledema. Computed tomography and MRI revealed severe biventricular hydrocephalus with transependymal periventricular fluid and a minimally enhancing cystic mass of the third ventricle. The patient was diagnosed with a colloid cyst and obstructive hydrocephalus, and endoscopic resection with ablation of the cyst remnant was performed. While attempting to extricate the cyst from the patient's head, control of the cyst was lost and the cyst fell into the lateral ventricle beyond the surgeon's view. Postoperative imaging showed that the cyst had settled in the right occipital horn. After 3 years of follow-up, imaging suggests growth of the cyst in its new position without necrosis or displacement on prone imaging.

The finding of a colloid cyst on neuroimaging is often incidental. These lesions are usually located at the foramen of Monro, are hyperdense on CT scans, and generally demonstrate signal intensity of water on MR images, although this depends on their content. When symptomatic, they frequently present with headaches and nausea due to an obstructive hydrocephalus. The authors describe a case of a giant colloid cyst in a patient presenting with complete left-sided vision loss and progressive memory loss, two very atypical findings in colloid cyst presentation. Imaging findings were also atypical, and this case proved to be a diagnostic dilemma because of its clinical and radiological presentation. Histopathological investigation was of utmost importance in the final diagnosis of a colloid cyst. To the authors' knowledge this colloid cyst is larger than any other described in the literature.

Object An optimal entry point and trajectory for endoscopic colloid cyst (ECC) resection helps to protect important neurovascular structures. There is a large discrepancy in the entry point and trajectory in the neuroendoscopic literature. Methods Trajectory views from MRI or CT scans used for cranial image guidance in 39 patients who had undergone ECC resection between July 2004 and July 2010 were retrospectively evaluated. A target point of the colloid cyst was extended out to the scalp through a trajectory carefully observed in a 3D model to ensure that important anatomical structures were not violated. The relation of the entry point to the midline and coronal sutures was established. Entry point and trajectory were correlated with the ventricular size. Results The optimal entry point was situated 42.3 ± 11.7 mm away from the sagittal suture, ranging from 19.1 to 66.9 mm (median 41.4 mm) and 46.9 ± 5.7 mm anterior to the coronal suture, ranging from 36.4 to 60.5 mm (median 45.9 mm). The distance from the entry point to the target on the colloid cyst varied from 56.5 to 78.0 mm, with a mean value of 67.9 ± 4.8 mm (median 68.5 mm). Approximately 90% of the optimal entry points are located 40–60 mm in front of the coronal suture, whereas their perpendicular distance from the midline ranges from 19.1 to 66.9 mm. The location of the “ideal” entry points changes laterally from the midline as the ventricles change in size. Conclusions The results suggest that the optimal entry for ECC excision be located at 42.3 ± 11.7 mm perpendicular to the midline, and 46.9 ± 5.7 mm anterior to the coronal suture, but also that this point differs with the size of the ventricles. Intraoperative stereotactic navigation should be considered for all ECC procedures whenever it is available. The entry point should be estimated from the patient's own preoperative imaging studies if intraoperative neuronavigation is not available. An estimated entry point of 4 cm perpendicular to the midline and 4.5 cm anterior to the coronal suture is an acceptable alternative that can be used in patients with ventriculomegaly.

This article describes a new endoscopic cutting tool that acts like a guillotine. It enables safe cutting of membranes overlying neurovascular structures because the footplate protects these structures from inadvertent injury. The footplate is introduced under the membrane, elevating it slightly, and then the membrane is cut while pushing the blade down into the footplate. The guillotine knife has been used in various endoscopic procedures such as septostomies, arachnoid cyst fenestrations, colloid cysts resections, and fenestrations of other intraventricular membranes. The authors think that the guillotine knife is a useful addition to the neuroendoscopic armamentarium.

Colloid cysts are histologically benign lesions whose primary goal of treatment should be complete resection to avoid recurrence and sudden death. Open surgery is traditionally considered the standard approach, but, recently, the endoscopic technique has been recognized as a viable and safe alternative to microsurgery. The endoscopic approach to colloid cysts of the third ventricle is usually performed through the foramen of Monro. However, this route does not provide adequate visualization of the cyst attachment on the tela choroidea. The combined endoscopic transforaminal-transchoroidal approach (ETTA), providing exposure of the entire cyst and a better visualization of the tela choroidea, could increase the chances of achieving a complete cyst resection. Between April 2005 and February 2011, 19 patients with symptomatic colloid cyst of the third ventricle underwent an endoscopic transfrontal-transforaminal approach. Five of these patients, harboring a cyst firmly adherent to the tela choroidea or attached to the middle/posterior roof of the third ventricle, required a combined ETTA. Postoperative MRI documented a gross-total resection in all 5 cases. There were no major complications and only 1 patient experienced a transient worsening of the memory deficit. To date, no cyst recurrence has been observed. An ETTA is a minimally invasive procedure that can allow for a safe and complete resection of third ventricle colloid cysts, even in cases in which the lesions are firmly attached to the tela choroidea or located in the middle/posterior roof of the third ventricle.

Object Incidental colloid cysts are frequently managed with surveillance imaging rather than surgical excision. This approach is born out of their purported indolent growth pattern and the surgical morbidity associated with microsurgical removal. The advent of endoscopic colloid cyst removal may offer renewed assessment of these patients who carry a risk of acute neurological deterioration. An evidence-based recommendation should weigh the risks of operative treatment. Thus far, there has been no concentrated assessment of cyst removal in patients with incidental colloid cysts. The major objective in this study was to define the risks associated with the endoscopic surgical removal of incidentally diagnosed colloid cysts Methods A retrospective review of the medical records was performed to search for patients evaluated for a colloid cyst between the years 1996 and 2012. Eighty-seven patients underwent colloid cyst resection, and 34 were managed with nonoperative surveillance imaging. Microsurgical resections, endoscopic resections of residual or recurrent colloid cysts, and cases with unknown preoperative symptomatic status were excluded from further analysis. Seventy-seven cases of primary endoscopic resections were identified. Twenty resections were performed in patients with an incidental diagnosis and 57 in symptomatic individuals. Presenting characteristics and surgical outcomes were compared between the incidental and symptomatic groups. Results The mean age at surgery was 39.65 years for the incidental and 43.31 years for the symptomatic group (p = 0.36). The median maximal cyst diameter was 9.7 mm (range 3–31 mm) for the incidental and 12 mm (range 5–34 mm) for the symptomatic group. The mean frontal and occipital horn ratio was 0.3928 for the incidental and 0.4445 for the symptomatic group (p = 0.002). Total resection was achieved in 90% of the incidental and 82.3% of the symptomatic cases (p = 0.49). The median hospital stay was 1 day for incidental and 2 days for symptomatic cases (p = 0.006). There were no deaths. There was one case of aseptic meningitis in the incidental group. In the symptomatic group there were 3 complications: one patient with subjective memory impairment, one with transient short-term memory deterioration, and another with a superficial wound infection treated with operative debridement. Two patients from the symptomatic group needed a CSF diversion procedure, and no shunting was needed in the incidental group. There were two recurrences in the symptomatic group (78 and 133 months postoperatively) and none in the incidental group (p = 1). Conclusions Age and cyst diameter were not correlated with the absence or presence of symptoms in patients with a colloid cyst of the third ventricle. Operative results were highly favorable in both groups and did not reveal a higher risk of morbidity in the patient presenting with an incidental lesion. The results support endoscopic resection as a legitimate therapeutic option for patients with incidental colloid cysts. Generalization of the operative results should be cautiously made, since this is a limited series and the results may depend on the degree of neuroendoscopic experience.

Colloid cysts of the third ventricle presenting with acute obstructive hydrocephalus due to intracystic and intraventricular hemorrhage are extremely rare. The authors report a case of a 43-year-old man with a hemorrhagic colloid cyst that was treated using endoscopic surgery. A small colloid cyst of the third ventricle was initially diagnosed in the patient, and he was treated conservatively at that time. On admission to the authors' institution he presented with sudden headache onset without neurological deficits. Computed tomography and MRI demonstrated a round hemorrhagic mass lesion in the third ventricle with bilateral intraventricular hemorrhage. Endoscopic resection was performed using a flexible videoscope. Only partial removal of the cyst was performed because of a tough cyst wall with highly viscous, hemorrhagic cystic contents. Histological examination revealed a typical colloid cyst wall and hemorrhage mixed within a mucinous substance. Postoperative serial neuroimaging demonstrated a gradual reduction in the residual cyst size and normalization in the lateral ventricle size.

Intraoperative neuromonitoring utilizing electroencephalography (EEG) is rarely performed during neuroendoscopy. The authors present a case in which this monitoring modality was used for a patient with a colloid cyst in preparation for an open craniotomy should an endoscopic approach fail. In this case, EEG serendipitously captured near-complete cessation of electrocerebral activity that occurred during intraventricular irrigation in response to ventricular collapse and resulted in no postoperative deficits. To the authors' knowledge, this is the first reported case of severe suppression of electrical activity captured by EEG during neuroendoscopy. Although they describe a transient phenomenon that resulted in no residual cognitive or neurological deficits, the importance of cautious introduction of ventricular irrigation, the need to carefully monitor intracranial pressure during neuroendoscopic procedures, and the need to pay close attention to irrigation temperature and composition should not be underestimated. Additional studies regarding the utility of EEG in alerting neurosurgeons to adverse electrical cerebral activity during neuroendoscopy are warranted.

Object Endoscopic surgery has been reported to be more cost-effective and safer than open craniotomy for resection of colloid cysts, despite a 5–10% conversion rate to craniotomy, a 5% recurrence rate, a 5–10% ventricular shunting rate, a 5–10% epilepsy rate, and a 3–4 day hospital stay. In 1985, the authors developed a interhemispheric, transcallosal, subchoroidal, fornix-sparing approach that allowed safe total resection of the colloid cyst and that appeared to be superior to the endoscopic approach. The long-term results are analyzed and compared with findings in the literature. Methods Fifty-seven consecutive colloid cysts were totally removed via a 3 ×3–in paramedian craniotomy flap and a microscopic interhemispheric, transcallosal, subchoroidal approach sparing the ipsilateral fornix. The length of the callosotomy was 1.5–2 cm in all patients. The mean follow-up duration was 12 years (range 2–22 years). A retrospective analysis comparing the authors' results with those reported in the endoscopic literature was performed. Results All patients had 1-year postoperative imaging studies (CT or MR imaging) documenting gross-total resection with no deaths, infection, hemiparesis, seizures, or disconnection syndrome. One surgery was complicated by bilateral subdural hematomas, which were successfully treated. There has been a zero recurrence rate. Three patients required a permanent ventriculoperitoneal shunt (including 2 who required emergency ventriculostomy before surgery). The mean hospital stay was 4.8 days (range 2–24 days). There was 1 patient with permanent short-term memory loss who presented with a herniation syndrome requiring emergency ventriculostomy. Conclusions The interhemispheric, transcallosal, subchoroidal, fornix-sparing approach to gross-total resection of colloid cysts is safe and led to a zero recurrence rate with no permanent neurological sequelae including epilepsy, and these results are superior to any reported results with endoscopy.

OBJECTIVE Colloid cysts are rare, histologically benign lesions that may result in obstructive hydrocephalus and death. Understanding the natural history of colloid cysts has been challenging given their low incidence and the small number of cases in most reported series. This has complicated efforts to establish reliable prognostic factors and surgical indications, particularly for asymptomatic patients with incidental lesions. Risk factors for obstructive hydrocephalus in the setting of colloid cysts remain poorly defined, and there are no grading scales on which to develop standard management strategies. METHODS The authors performed a single-center retrospective review of all cases of colloid cysts of the third ventricle treated over nearly 2 decades at Washington University. Univariate analysis was used to identify clinical, imaging, and anatomical factors associated with 2 outcome variables: symptomatic clinical status and presentation with obstructive hydrocephalus. A risk-prediction model was defined using bootstrapped logistic regression. Predictive factors were then combined into a simple 5-point clinical scale referred to as the Colloid Cyst Risk Score (CCRS), and this was evaluated with receiver-operator characteristics. RESULTS The study included 163 colloid cysts, more than half of which were discovered incidentally. More than half of the incidental cysts (58%) were followed with surveillance neuroimaging (mean follow-up 5.1 years). Five patients with incidental cysts (8.8%) progressed and underwent resection. No patient with an incidental, asymptomatic colloid cyst experienced acute obstructive hydrocephalus or sudden neurological deterioration in the absence of antecedent trauma. Nearly half (46.2%) of symptomatic patients presented with hydrocephalus. Eight patients (12.3%) presented acutely, and there were 2 deaths due to obstructive hydrocephalus and herniation. The authors identified several factors that were strongly correlated with the 2 outcome variables and defined third ventricle risk zones where colloid cysts can cause obstructive hydrocephalus. No patient with a lesion outside these risk zones presented with obstructive hydrocephalus. The CCRS had significant predictive capacity for symptomatic clinical status (area under the curve [AUC] 0.917) and obstructive hydrocephalus (AUC 0.845). A CCRS ≥ 4 was significantly associated with obstructive hydrocephalus (p < 0.0001, RR 19.4). CONCLUSIONS Patients with incidentally discovered colloid cysts can experience both lesion enlargement and symptom progression or less commonly, contraction and symptom regression. Incidental lesions rarely cause acute obstructive hydrocephalus or sudden neurological deterioration in the absence of antecedent trauma. Nearly one-half of patients with symptomatic colloid cysts present with obstructive hydrocephalus, which has an associated 3.1% risk of death. The CCRS is a simple 5-point clinical tool that can be used to identify symptomatic lesions and stratify the risk of obstructive hydrocephalus. External validation of the CCRS will be necessary before objective surgical indications can be established. Surgical intervention should be considered for all patients with CCRS ≥ 4, as they represent the high-risk subgroup.

OBJECTIVE Many colloid cyst patients present with obstructive hydrocephalus that resolves after resection of the cyst. However, a proportion of patients with these cysts will require cerebrospinal fluid shunting after tumor resection, despite resolution of the obstruction at the foramina of Monro. The goal of this study was to determine if colloid cyst size or preoperative ventricular volume predicted the need for postresection shunting. METHODS In a retrospective study design, ICD-9 codes 742.2 (colloid cyst) and 348.0 (brain cyst) were used to identify patients who had undergone resection of a colloid cyst at the University of Florida over the last 20 years. Preoperative imaging (CT or MRI) with a stereotactic software program developed at the University of Florida was used to measure volumes of the colloid cyst and the lateral ventricles. The relationships among ventricular volume, colloid cyst volume, and postoperative shunting were analyzed. RESULTS The number of patients included in the study was 67, and their mean age was 37.7 years. Forty percent of the patients were female. Overall, 49.2% of the patients had a transcallosal approach, 35.8% a transcortical approach, and 14.9% an endoscope-assisted surgery. Mean preoperative ventricular volume was 76.5 cc in patients who never received a ventriculoperitoneal shunt (VPS) and 98.1 cc in those who were eventually treated with a VPS (p = 0.305). Patients with a postoperative VPS had an initial mean colloid cyst volume of 1.8 cc compared with 0.9 cc in patients without a VPS postoperatively (p = 0.019). Patients with colloid cysts larger than 0.6 cc (1-cm diameter) had a 12.8 increased odds of needing a VPS postoperatively (95% CI 1.81–275). CONCLUSIONS Larger colloid cysts are associated with an increased need for postresection shunting independent of preoperative ventricular size. Prospective studies of patients with colloid cysts are necessary to further identify risks of permanent hydrocephalus.

Object Endoscopic removal of intraventricular brain tumors is well established for cystic tumors such as colloid cysts. Aspiration followed by removal or ablation of the membranous wall is possible given the constituent features of these tumors. It is generally expected that endoscopic removal of solid brain tumors from the intraventricular compartment would impose additional technical demands. In this paper, the feasibility and safety of endoscopic removal of solid intraventricular brain tumors is evaluated. Methods Eighty-one patients who underwent endoscopic management of an intraventricular brain tumor were identified from a prospective database. Of these patients, seven underwent attempted endoscopic surgical removal of a solid primary brain tumor. Patient selection, surgical technique, procedure-related morbidity, and extent of removal were reviewed. Five patients underwent complete resection of a solid intraventricular brain tumor, a treatment option that was based on intraoperative assessment and confirmed by postoperative imaging. No patient experienced any procedure-related morbidity. Of the individuals in whom a total endoscopic resection was successful, there has been no symptomatic or radiological evidence of recurrence (mean follow up 20 months). Maximum tumor diameter ranged from 0.5 to 1.8 cm for patients who underwent complete resection, whereas maximum tumor diameter measured 2.4 and 2.5 cm in the two patients in whom a subtotal excision was performed. Conclusions In select patients, complete endoscopic removal of solid intraventricular brain tumors is possible and safe. Factors that influence the ability of a surgeon to perform a complete endoscopic resection include tumor size, composition, and vascularity. The procedure requires careful patient selection, the use of refined endoscopic instrumentation, and a disciplined surgical technique.

Object. Patients harboring colloid cysts of the third ventricle can present with acute neurological deterioration, or the first indication of the lesion may appear when the patient suddenly dies. The risk of such an occurrence in a patient already identified as harboring a colloid cyst is unknown. The goal of this stud was to estimate the risk of acute deterioration in patients with colloid cysts. Methods. A retrospective study was made of a cohort of patients with newly diagnosed colloid cysts who were recruited in The Netherlands between January 1, 1993, and December 31, 1997. Seventy-eight patients were identified, all of whom displayed symptoms. Twenty-five patients (32%) presented with symptoms of acute deterioration; four patients died suddenly and the cysts were discovered at autopsy. The overall mortality rate was 12%. Results of a multivariate logistic regression analysis demonstrated that no subgroup of patients presenting without acute deterioration could be identified on the basis of patient age, duration of symptoms, cyst size, or the presence of hydrocephalus. The national incidence of colloid cysts in The Netherlands is 1/10(6) person-years; the prevalence was estimated to be 1800 asymptomatic colloid cysts. Conclusions. Acute deterioration was a frequent presentation among a national cohort of Dutch patients harboring symptomatic colloid cysts. The risk of acute deterioration in a symptomatic patient with a colloid cyst in The Netherlands is estimated to be 34%. The estimated risk for an asymptomatic patient with an incidental colloid cyst is significantly lower. These results strongly advocate the selection of surgical treatment for patients with symptomatic colloid cysts.

Object. Patients harboring colloid cysts of the third ventricle can present with acute neurological deterioration, or the first indication of the lesion may appear when the patient suddenly dies. The risk of such an occurrence in a patient already identified as harboring a colloid cyst is unknown. The goal of this study was to estimate the risk of acute deterioration in patients with colloid cysts. Methods. A retrospective study was made of a cohort of patients with newly diagnosed colloid cysts who were recruited in The Netherlands between January 1, 1993, and December 31, 1997. Seventy-eight patients were identified, all of whom displayed symptoms. Twenty-five patients (32%) presented with symptoms of acute deterioration; four patients died suddenly and the cysts were discovered at autopsy. The overall mortality rate was 12%. Results of a multivariate logistic regression analysis demonstrated that no subgroup of patients presenting without acute deterioration could be identified on the basis of patient age, duration of symptoms, cyst size, or the presence of hydrocephalus. The national incidence of colloid cysts in The Netherlands is 1/106 person-years; the prevalence was estimated to be 1800 asymptomatic colloid cysts. Conclusions. Acute deterioration was a frequent presentation among a national cohort of Dutch patients harboring symptomatic colloid cysts. The risk of acute deterioration in a symptomatic patient with a colloid cyst in The Netherlands is estimated to be 34%. The estimated risk for an asymptomatic patient with an incidental colloid cyst is significantly lower. These results strongly advocate the selection of surgical treatment for patients with symptomatic colloid cysts.

Object. Frameless computerized neuronavigation has been increasingly used in intracranial endoscopic neurosurgery. However, clear indications for the application of neuronavigation in neuroendoscopy have not yet been defined. The purpose of this study was to determine in which intracranial neuroendoscopic procedures frameless neuronavigation is necessary and really beneficial compared with a free-hand endoscopic approach. Methods. A frameless infrared-based computerized neuronavigation system was used in 44 patients who underwent intracranial endoscopic procedures, including 13 third ventriculostomies, nine aqueductoplasties, eight intraventricular tumor biopsy procedures or resections, six cystocisternostomies in arachnoid cysts, five colloid cyst removals, four septostomies in multiloculated hydrocephalus, four cystoventriculostomies in intraparenchymal cysts, two aqueductal stent placements, and fenestration of one pineal cyst and one cavum veli interpositi. All interventions were successfully accomplished. In all procedures, the navigational system guided the surgeons precisely to the target. Navigational tracking was helpful in entering small ventricles, in approaching the posterior third ventricle when the foramen of Monro was narrow, and in selecting the best approach to colloid cysts. Neuronavigation was essential in some cystic lesions lacking clear landmarks, such as intraparenchymal cysts or multiloculated hydrocephalus. Neuronavigation was not necessary in standard third ventriculostomies, tumor biopsy procedures, and large sylvian arachnoid cysts, or for approaching the posterior third ventricle when the foramen of Monro was enlarged. Conclusions. Frameless neuronavigation has proven to be accurate, reliable, and extremely useful in selected intracranial neuroendoscopic procedures. Image-guided neuroendoscopy improved the accuracy of the endoscopic approach and minimized brain trauma.

✓ Colloid cysts are rare intracerebral lesions that are preferentially encountered within the third ventricle. There are only a few reports in which colloid cysts are described in other locations such as the fourth ventricle.A symptomatic intracerebellar colloid cyst in a 45-year-old woman is described. The patient presented with headache, gait disturbance, and nausea. Neuroradiological imaging revealed compression of the fourth ventricle, hydrocephalus, and an intracerebellar cystic lesion measuring 4 × 5 cm that had a small peripheral solid portion. The cyst was successfully removed via a paramedian suboccipital approach. Postoperatively, the patient recovered quickly.The findings in the present report represent an additonal example of the broad spectrum of cystic lesions encountered in the cerebellum.

✓ The authors present the first reported case of a hemangiopericytoma (HPC) occurring in the third ventricle. Most of these lesions are based in the meninges. There is only one other reported case of an intraventricular HPC; in that case the lesion was found in the lateral ventricle.A 40-year-old right-handed man presented with a 3-month history of headaches. Clinical evaluation, including computerized tomography and magnetic resonance imaging studies, revealed a 1-cm enhancing lesion in the third ventricle. Given the findings on the preoperative imaging studies, the lesion was not consistent with some of the more commonly occurring tumors of the third ventricle, namely colloid cysts. A transcortical approach and resection of the lesion was performed without complication. The final pathological findings were consistent with those of an HPC.Hemangiopericytomas rarely occur in the ventricles and may pose a difficult diagnostic dilemma based on their radiographic and gross appearances, as shown in this case. Because of this difficulty, histological confirmation is required to make a definitive diagnosis. These lesions have a propensity to recur and metastasize in the central nervous system and periphery, thus making the goal of treatment a complete surgical resection followed by postoperative radiation therapy in most cases.

✓ The surgical technique for the endoscopic evacuation of colloid cysts of the third ventricle in 13 patients is described. The authors conclude that endoscopic resection of these lesions is a useful addition to the current surgical repertoire and a viable alternative to stereotactic aspiration or open craniotomy.

✓ A continuous follow-up review of colloid cysts including aspects of natural history and evaluation of treatment options is necessary to optimize individual treatment. Thirty-seven consecutive patients with colloid cyst of the third ventricle seen at Karolinska Hospital between 1984 and 1995 were reviewed. Five patients were admitted in a comatose state, and two died despite emergency ventriculostomy. Three had recurrent cysts following previous aspiration procedure. During the study period, patients underwent a total of 10 ventriculostomies, 10 aspirations, 26 microsurgical operations, and two shunt operations. Twenty-four of 26 microsurgical operations were transcallosal and two were transcortical. Twenty-four operations (22 transcallosal and two transfrontal approaches) without permanent morbidity were performed by four surgeons. Transient memory deficit from forniceal traction was noted in 26%. The remaining two transcallosal operations, which led to permanent morbidity or mortality, were performed by two different surgeons. Aspiration of cysts performed by four different surgeons carried a 40% risk of transient memory deficit (10% permanent) and an 80% recurrence rate. One patient was found to be cured on radiological studies obtained at the 5-year follow-up review. Seven cysts were followed by means of radiological studies with no treatment for 6 to 37 months. Five of these cysts grew, indicating that younger patients with colloid cysts will probably need surgical treatment. The main causes of unfavorable results were: 1) failure to investigate symptoms that proved fatal; 2) subtotal resection; and 3) surgical complications. Transcallosal microsurgery produced excellent results when performed by experienced surgeons. A colloid cyst of the foramen of Monro is a disease that should be detected before permanent neurological damage has occurred. Permanent morbidity or mortality should not be accepted in modern series of third ventricle colloid cysts.

✓ Two cases are reported of third ventricle masses that were clinically and radiographically indistinguishable from pure colloid cysts. A 21- and a 36-year-old man presented with 5-year and 10-day histories of headache, respectively. Magnetic resonance (MR) imaging revealed smooth, homogeneous masses in the anterior third ventricle that were iso- to hyperintense on T1-weighted MR images and hyperintense on T2-weighted images. There was little enhancement with intravenous contrast material. In both patients, craniotomies were performed and histopathological examination revealed xanthogranulomas of the choroid plexus with only microscopic foci of colloid cyst-like structures. These cases illustrate that xanthogranulomas of the third ventricle may clinically and radiologically mimic pure colloid cysts, that a range of MR imaging signals can be seen, and that craniotomy rather than stereotactic aspiration is the indicated treatment.

✓ It is still not determined which is the best surgical option for third ventricle colloid cysts. Since 1990, the authors have used a steerable fiberscope to remove colloid cysts in seven patients and have performed microsurgery via a transcallosal approach in eight patients. The two techniques were compared for operating time, length of hospital stay, incidence of complications, recurrence, and hydrocephalus, and days spent recuperating before return to work to determine if endoscopic removal of colloid cysts is a safe and effective alternative to microsurgery. Statistical analysis was adjusted for age, sex, and presenting symptoms. Microsurgical cases averaged 206 minutes of operating time whereas endoscopic cases averaged 127 minutes (p = 0.01). For combined days spent in the intensive care unit and on the ward, the patients averaged 9.5 days after microsurgery and 4 days after endoscopy (p = 0.05). Postoperative complications occurred in five of eight patients after microsurgery and in one of seven patients after endoscopy (p = 0.09); complications were transient and primarily related to short-term memory loss. In all patients, preoperative symptoms resolved and the cysts have not recurred. Postoperatively, one patient required a ventriculoperitoneal shunt after microsurgery but all patients were shunt-independent after endoscopy. Patients returned to work an average of 59 days after discharge following microsurgery compared with an average of 26 days after endoscopy (p = 0.05). Compared with transcallosal microsurgery for the removal of colloid cysts, these preliminary results show that a steerable endoscope reduced operating time and that patients spent fewer days in the hospital and returned to work sooner after endoscopy.

✓ Sixteen patients treated between 1969 and 1989 for a colloid cyst of the foramen of Monro by stereotactically guided aspiration (not stereotactic extirpation) were evaluated to assess the long-term outcome of the procedure. Thirteen of these patients required reoperation due to an acute comatose state, failure to achieve permanent reduction of the cyst, or symptomatic hydrocephalus. Of these 13, six were treated twice and two were treated three times by stereotactic aspiration. Five patients underwent microsurgical extirpation and three had a shunt placed following a failed aspiration. Failure of the first procedure was detected within the first 2 months after treatment in eight patients and after more than 6 years in seven. Following stereotactic aspiration, three patients experienced a temporary memory deficit and confusion and one patient suffered a central pain syndrome. Eleven of the 26 procedures were followed by a recurrence 6 to 15 years after treatment; seven recurrent cysts were detected after more than 8 years. Of the patients with recurrences, three did not undergo repeat surgery but showed an increase in cyst size at the latest follow-up examination. It is suggested that radical removal by open or stereotactically guided microsurgery is the method of choice since stereotactic aspiration fails to offer a radical or permanent treatment for colloid cysts of the third ventricle.

✓ Xanthogranulomas involving the choroid plexus of the lateral ventricles are generally asymptomatic lesions. The case is reported of a 50-year-old man in whom a xanthogranuloma of the choroid plexus had occluded the left trigone, causing unilateral hydrocephalus of the left temporal horn and neurological symptoms. A review of the literature shows that xanthogranulomas of the glomus of the lateral ventricles differ from the xanthomatous cystic lesions of the third ventricle, which are probably akin to colloid cysts.

✓ Stereotactic aspiration is a valuable surgical alternative for colloid cysts when used alone or in conjunction with microsurgical resection. Since 1981, the authors have performed computerized tomography (CT)-guided stereotactic aspiration as the initial procedure in 22 patients with colloid cysts; stereotactic aspiration alone was successful in 11 patients (50%). Of the 11 patients in whom aspiration failed, stereotactic endoscopic resection was attempted in three and was successful in one. Seven patients required a craniotomy and microsurgical removal of the cyst performed via a transcortical approach. The preoperative CT appearance in eight cases of a hypodense or isodense cyst correlated favorably with successful aspiration of the cyst in six patients. A hyperdense appearance on the preoperative CT scan in 14 cases was associated with subtotal aspiration in 13 patients; five required craniotomy for removal. Preoperative magnetic resonance (MR) imaging in eight patients provided excellent anatomical definition of the cyst and its relationship to other structures of the third ventricle, but it was not possible to correlate successful aspiration with cyst appearance on MR images with short or long relaxation time sequences. The authors' 9-year experience suggests that preoperative CT studies accurately determine size, predict viscosity, and help to define a group of colloid cyst patients for whom stereotactic cyst aspiration will likely be successful. Unsuccessful stereotactic aspiration was related to two features: the high viscosity of the intracystic colloid material (nine patients), or deviation of the cyst away from the aspiration needle due to small cyst volume (two patients). Because of its simplicity and low risk, stereotactic surgery can be offered to selected patients as the initial procedure of choice. Craniotomy can be reserved for those whose imaging studies predict failure or for those whose cyst cannot be aspirated.

✓ The authors report the case of a patient harboring a posterior fossa neuroepithelial cyst who presented with positional facial weakness and syncope. The patient recovered rapidly after cyst fenestration and placement of an internal cyst-cisternal shunt. The pathogenesis and principles of diagnosis and management of these rare lesions are reviewed.

✓ Initial experience with a new arc-radius design of stereotactic frame that interfaced with the existing components of the Brown-Roberts-Wells instrument is reported. Over a 6-month period, 32 procedures were performed on 23 males and nine female patients (mean age 32 years); these included 27 stereotactic biopsy procedures, two stereotactic implantations of cyst catheter reservoirs, two ventriculoscopic aspirations of third ventricular colloid cysts, and one stereotactic aspiration of a craniopharyngioma. In all cases successful targeting was achieved and verified by postoperative computerized tomography. There were no operation-related complications. This new frame offers rapid and accurate targeting and is a useful adjunct to the stereotactic armamentarium.

✓ The authors recently encountered a unique case of anterior third ventricular neurocysticercosis in which the cyst exhibited an unusually high signal on T1-weighted magnetic resonance imaging. The lesion's signal intensity and location made differentiation from colloid cyst difficult. Intraventricular neurocysticercosis should be included in the differential diagnosis of a colloid cyst.

✓ The authors present a newly designed device for ultrasonic guidance of neuroendoscopic procedures. It consists of a puncture adapter that attaches to a rigid endoscope having an outer diameter of 6 mm and is mounted on a small, bayonet-shaped ultrasound probe. This adapter directs the movement of the endoscope precisely within the ultrasonic field of view. The targeted region is identified by transdural insonation via an enlarged single burr-hole approach, and the endoscope is tracked in real time throughout its approach to the target. The procedure has been performed in 10 patients: endoscopic ventriculocystostomy in four cases; removal of a colloid cyst of the third ventricle in two cases; and intraventricular tumor biopsy, intraventricular tumor resection, third ventriculostomy, and removal of an intraventricular hematoma in one case each. The endoscope was depicted on ultrasonograms as a hyperechoic line without disturbing echoes and, consequently, the target (cyst, ventricle, or tumor) was safely identified in all but one case, in which intraventricular air hid a colloid cyst in the foramen of Monro. The method presented by the authors proved to be very effective in the guidance and control of neuroendoscopic procedures. Combining this method with image guidance is recommended to define the entry point of the endoscope precisely.

Object Frameless computerized neuronavigation has been increasingly used in intracranial endoscopic neurosurgery. However, clear indications for the application of neuronavigation in neuroendoscopy have not yet been defined. The purpose of this study was to determine in which intracranial neuroendoscopic procedures frameless neuronavigation is necessary and really beneficial compared with a free-hand endoscopic approach. Methods A frameless infrared-based computerized neuronavi- gation system was used in 44 patients who underwent intracranial endoscopic procedures, including 13 third ventriculostomies, nine aqueductoplasties. eight intraventricular tumor biopsy procedures or resections, six cystocisternostomies in arachnoid cysts, five colloid cyst removals, four septostomies in multiloculated hydrocephalus, four cystoventriculostomies in intraparenchymal cysts, two aqueductal stent placements, and fenestration of one pineal cyst and one cavum veli interpositi. All interventions were successfully accomplished. In all procedures, the navigational system guided the surgeons precisely to the target. Navigational tracking was helpful in entering small ventricles, in approaching the posterior third ventricle when the foramen of Monro was narrow, and in selecting the best approach to colloid cysts. Neuronavigation was essential in some cystic lesions lacking clear landmarks, such as intraparenchymal cysts or multiloculated hydrocephalus. Neuronavigation was not necessary in standard third ventriculostomies, tumor biopsy procedures, and large sylvian arachnoid cysts, or for approaching the posterior third ventricle when the foramen of Monro was enlarged. Conclusions Frameless neuronavigation has proven to be accurate, reliable, and extremely useful in selected intracranial neuroendoscopic procedures. Image-guided neuroendoscopy improved the accuracy of the endoscopic approach and minimized brain trauma.

✓ Colloid cysts appear most commonly in the third ventricle; the occurrence of a colloid cyst in the brainstem is very unusual. The authors report on a patient with an intrapontomesencephalic colloid cyst. This 15-year-old girl complained of a headache associated with diplopia. Her neurological examination revealed right-sided sixth nerve paresis and a mild left hemiparesis. Radiological investigations revealed an intraparenchymal pontomesencephalic cystic mass. Surgical removal of the lesion was achieved via the pterional transsylvian approach and the patient experienced an excellent recovery. Histopathological examination revealed that the lesion was a typical colloid cyst. To the best of the authors' knowledge, this is the first case in which an intraparenchymal upper brainstem colloid cyst was surgically excised totally. In addition to describing this case, the authors also review other brainstem neuroepithelial cysts described in the literature and briefly discuss the concept of their origin.

✓ In an effort to shed light upon the nature of the colloid cyst, the immunohistochemical properties of 21 examples of this lesion were compared with those of other neuraxial cysts and choroid plexus epithelium. The neuraxial cysts included the following: eight Rathke's cleft cysts, 25 pituitaries containing follicular cysts of the pars intermedia, and four enterogenous cysts. Fifteen examples of normal choroid plexus and 12 choroid plexus papillomas were studied as well. These lesions were examined for localization of the following antigens: cytokeratins, epithelial membrane antigen, secretory component, carcinoembryonic antigen, prealbumin, vimentin, glial fibrillary acidic protein (GFAP), S-100 protein, neuron-specific enolase, 68-kD neurofilament protein, chromogranin, serotonin, and lysozyme, and with Leu-7 monoclonal antibodies. Five colloid cysts were immunostained with monoclonal antibodies that were specific for Clara-cell antigens and surfactant, respectively. Sugar moieties were localized using Ulex europaeus I, and Ricinus communis agglutinin I lectins. All Rathke's cleft cysts and follicular cysts of the pars intermedia as well as three selected colloid cysts were examined for pituitary hormones. The epithelial cells of colloid and enterogenous cysts, as well as those lining follicular and Rathke's cleft cyst, showed uniformly strong reactivity for cytokeratins, epithelial membrane antigen, secretory component, and vimentin, and bound Ulex europaeus lectin. Occasional cells in colloid cysts were positive for Clara cell-specific antigens. Reaction for carcinoembryonic antigen was present on the apical surface of scattered cells of colloid, follicular, and Rathke's cleft cysts. Many cells of follicles in the pars intermedia as well as individual cells of five Rathke's cleft cysts were also immunoreactive for chromogranin, S-100 protein, GFAP, and pituitary hormones. Colloid and enterogenous cysts were negative for prealbumin, S-100 protein, GFAP, and neuron-specific enolase; in all but a few instances, they failed to bind Ricinus communis agglutinin. In contrast, normal choroid plexus and choroid plexus papillomas were positive for prealbumin, S-100 protein, neuron-specific enolase, cytokeratin, vimentin, and Ricinus communis agglutinin receptors; they lacked Ulex europaeus lectin, 56/66-kD cytokeratins, and epithelial membrane antigen. Unlike normal choroid plexus, choroid plexus papillomas were often GFAP-positive. All tissues studied were nonreactive for lysosome, serotonin, and neurofilament, and with Leu-7 antibodies. This study indicates that the immunophenotype of epithelium lining colloid cysts is similar to that of other cysts showing endodermal or ectodermal differentiation and to respiratory tract mucosa. Epithelium of colloid cysts is immunohistochemically different from that of normal or neoplastic choroid plexus. These findings indicate an endodermal rather than neuroepithelial nature for colloid cysts.

✓ The authors report on a case of solitary neurocysticercosis located in the anterior third ventricle of a 21-year-old Hispanic man who presented with progressive headache, nausea, and mental decline. The appearance of the lesion on computerized tomography and magnetic resonance studies was typical of that of a colloid cyst, which was the presumptive preoperative diagnosis. A transcortical-transventricular microsurgical approach was used, and inspection of the anterior third ventricle via the ipsilateral foramen of Monro revealed no lesion. Inspection of the floor of the ipsilateral lateral ventricle revealed the intact cyst, which had migrated since the preoperative studies had been obtained. The lesion was removed intact, and histopathological examination confirmed the diagnosis of neurocysticercosis. The authors review the radiographic findings in this case and further illustrate the potential for neurocysticercosis to masquerade as other intracranial lesions.

✓ A device to facilitate the stereotactic aspiration of cystic lesions is described. The device anchors the cyst wall, allows penetration of the lesion without displacement, and assures proper positioning of the tip of the aspiration needle within the cyst so that the contents may be evacuated without spillage.

Colloid cysts of the third ventricle are well described in the neuroscience literature; however, we were unable to find any reports in which hemorrhagic colloid cysts were described. This 60-year-old man had a 5-year history of follicular mixed lymphoma. After completing chemotherapy, he experienced acute nonlymphocytic leukemia with chronic thrombocytopenia. Magnetic resonance imaging of the head was obtained because the patient complained of persistent nausea and vomiting (Fig. 1). One day later, the patient died unexpectedly from causes unrelated to his malignancy or intracranial disease. On gross examination, the patient’s brain revealed a well-circumscribed dark-brown and black hemorrhagic mass adherent to the roof of the third ventricle with no intraventricular blood. Microscopic sections revealed a well-circumscribed, fibrotic, hemorrhagic colloid cyst containing large amounts of hemorrhage of varying ages (Fig. 2). Freshly extravasated red blood cells were seen, as well as abundant deposits of hemosiderin. At the periphery, focally preserved patches of well-differentiated, mucin-producing, simple cuboidal epithelium were observed (Fig. 3). Also, a few cystic cavities, which were lined by similar epithelium and contained periodic acid/Schiff–positive colloid, were found in the fibrotic capsule. There was no evidence of infiltration of the colloid cyst or other sections of brain by lymphoma. Colloid cysts of the third ventricle have accounted for 0.5 to 2% of

✓ The therapeutic strategies employed in the management of anterior third-ventricular mass lesions remain controversial. Resection by conventional craniotomy, whether via a transcallosal or transcortical approach, carries well-known risks and limitations. Alternatively, in this region traditional stereotaxy has been relegated to use with biopsy only or cyst aspiration procedures. Combining aspects of both conventional and stereotaxic techniques has allowed total removal of 12 colloid cysts in six women and six men ranging in age from 25 to 71 years. No mortality and minimal morbidity have been associated with the procedures. There has been no evidence of recurrence in an average follow-up period of 19 months. By coupling the benefits of stereotaxic precision and localization to the microsurgical management of colloid cysts, several rewards have been realized: 1) only a limited cortical dissection is needed; 2) the hazards of callosal or forniceal injury can be avoided; 3) the lesion is easily localized regardless of ventricular size; 4) hemostasis can be readily achieved with bipolar cautery or defocused laser power; and 5) most importantly, a total resection is possible with little risk to the patient. Stereotaxic microsurgical laser craniotomy provides a new option for the management of colloid cysts and other anterior third-ventricular lesions.