Cancer Dissemination, Hydrocephalus, and Survival After Cerebral Ventricular Entry During High-Grade Glioma Surgery: A Meta-Analysis

Cancer Dissemination, Hydrocephalus, and Survival After Cerebral Ventricular Entry During... Abstract BACKGROUND The consequences of ventricular entry during resection of high-grade gliomas (HGG) are uncertain and often not detectable clinically. OBJECTIVE To reveal odds of tumor dissemination, hydrocephalus, and mortality in adult patients who had ventricular entry during surgical resection of HGG. METHODS Titles and abstracts of published journals in the NCBI/NLM PubMed and OVID EMBASE databases were searched without language restriction and systematically screened. Outcomes extracted included the odds of leptomeningeal dissemination and hydrocephalus in patients with ventricular entry during HGG resection compared to without. They were analyzed using a random-effects model to calculate summary odds ratios (sORs). Overall survival data were also compared between patients with and without ventricular entry. RESULTS Twenty final studies with 2251 total patients were included from the 6910 retrieved. Patients with ventricular entry during HGG resection demonstrated higher odds of leptomeningeal dissemination (sOR: 3.91 [95% confidence interval (CI): 1.89-8.10]; P = .0002; 86/410 vs 57/847 patients in 9 studies) and hydrocephalus (sOR: 7.78 [95% CI: 3.77-16.05]; P < .00001; 58/431 vs 11/565 patients in 11 studies). They also had decreased survival (median survival: 16.8 vs 19.1 mo; 413 vs 322 patients in 10 studies; hazard ratio: 1.25 [95% CI: 1.05-1.48], P = .01). CONCLUSION The association between ventricular entry during HGG resection and tumor dissemination, hydrocephalus, and decreased survival invites investigations to understand this link. Neurosurgeons and neuro-oncologists must be aware of the consequences of ventricular entry during surgery for HGG. Glioma, Glioblastoma, Dissemination, Hydrocephalus, Survival, Ventricle, Lateral ventricle ABBREVIATIONS ABBREVIATIONS CIs confidence intervals CSF cerebrospinal fluid GBM glioblastoma HR hazard ratio NOS Newcastle–Ottawa Scale ORs odds ratios Surgical treatment of high-grade gliomas is focused on the challenging task of maximal but safe removal of the tumor from the brain primarily to limit local recurrence. Unlike most other cancers, high-grade gliomas uncommonly disseminate within the central nervous system, and even more rarely metastasize outside it.1,2 A commonly recognized form of high-grade glioma dissemination is leptomeningeal dissemination. It is diagnosed radiographically with abnormal enhancement or nodular deposits in the cerebral ventricular wall and the subarachnoid spaces of the brain and spinal cord.3 FIGURE 1. View largeDownload slide Meta-analysis of studies comparing A, incidences of leptomeningeal dissemination in patients with and without ventricular entry during their high-grade glioma resection with its B, respective funnel plot to assess publication bias. The lines in the forest plot represent 95% CI, and the size of its corresponding square marker reflects the relative weight of the study used in random-effects Mantel–Haenszel method. The diamond shows the overall effect. In the funnel plot, vertical line represents the summary OR with diagonal lines representing 95% CIs. Each study is represented by a circle. FIGURE 1. View largeDownload slide Meta-analysis of studies comparing A, incidences of leptomeningeal dissemination in patients with and without ventricular entry during their high-grade glioma resection with its B, respective funnel plot to assess publication bias. The lines in the forest plot represent 95% CI, and the size of its corresponding square marker reflects the relative weight of the study used in random-effects Mantel–Haenszel method. The diamond shows the overall effect. In the funnel plot, vertical line represents the summary OR with diagonal lines representing 95% CIs. Each study is represented by a circle. In the process of achieving maximal, safe resection of high-grade gliomas, the cerebral ventricles are often entered intraoperatively. Early reports noted an association of this ventricular breach with the occurrence of leptomeningeal dissemination.4,5 This led to the emergence of the hypothesis that neoplastic cells could enter into the cerebrospinal fluid (CSF) in the ventricles and then spread via its circulation throughout the nervous system.4,5 Furthermore, hydrocephalus6-9 was often also noted with CSF or leptomeningeal dissemination. Given the relatively rare incidences of leptomeningeal dissemination and hydrocephalus in high-grade gliomas, we conducted a meta-analysis of published data to test the hypothesized associations with ventricular entry during high-grade glioma surgery. Further, we analyzed survival of patients with and without ventricular entry. Here, we reveal an important and strong link between entry into the ventricle—a common intraoperative occurrence during high-grade glioma resections—and leptomeningeal dissemination, hydrocephalus, and survival. METHODS Search Strategy A focused study question was developed with the aid of the common evidence medicine framework PICO (Patient population, Intervention, Control, Outcome): Do adult patients with high-grade glioma (World Health Organization grade III or IV; patient population) who had surgical entry into the cerebral ventricles during resection of their glioma (intervention) have an increased incidence of leptomeningeal dissemination, hydrocephalus, and mortality (outcomes) compared to patients whose ventricles were not breached (control)? The preferred reporting items for systematic reviews and meta-analyses checklist10 were followed to conduct and report this meta-analysis. A prior review protocol that specifically addresses this question does not exist; therefore, authors conducted a search of titles and abstracts of published journals in the NCBI/NLM PubMed (from 1966 to October 30, 2017) and OVID EMBASE (from 1980 to October 30, 2017) databases without language restriction. The following string of terms was used: (astrocytoma* OR oligodendroglioma* OR oligodendroma* OR glioma* OR glioblastoma* OR GBM*) AND (disseminat* OR distant OR metasta* OR multifocal OR “multifocal” OR hydrocephalus OR shunt* OR ventriculomegaly). The retrieved journal articles were imported to the reference manager EndNote X7 (Thompson Reuters, Philadelphia, Pennsylvania) while removing duplicates. Study Selection and Quality Assessment Excluding unpublished studies and conference abstracts, the titles and abstracts of the retrieved journal articles were systematically screened (by A.M.M. and P.D.K.) to select studies for inclusion in the meta-analysis. We included studies with data on our outcomes of interest in patients with or without ventricular entry during surgical resection of their high-grade glioma. We excluded review articles, case series’ of less than 3 patients, studies or data of pediatric patients or patients with gliomatosis cerebri, or studies with incomplete data. Multifocal or multicentric glioblastomas (GBM) were not excluded; however, data or studies of distant parenchymal recurrences were excluded. The screening process yielded 20 final studies with 2251 total patients. The search and selection process are outlined in a Figure, Supplemental Digital Content 1. A bibliography search of these studies did not reveal any additional relevant studies. The Newcastle–Ottawa Scale (NOS) was used to assess the quality of each nonrandomized, case-controlled or cohort study included in the meta-analyses.11 The Cochrane risk of bias assessment tool12 was used to assess the risk of selection, performance, detection, and attrition bias in each study. Publication bias was also assessed (Meta-Analysis subsection of the Methods). Outcomes and Data Extraction A review of all 20 studies was performed (by A.M.M. and P.D.K.). From each final study, the total number of patients with and without ventricular entry during surgical resection of high-grade glioma was noted. In both of these groups, the number of patients with leptomeningeal dissemination and hydrocephalus were noted. Only data on patients who required treatment for their hydrocephalus in the form of CSF diversion were noted. When the cause of hydrocephalus was clear, it was noted to allow for sub-analysis. Additionally, any overall survival data of patients in these 2 groups were extracted. Corresponding authors of studies were contacted to seek clarification of data presented in their publications; clarifications in 2 studies13,14 were obtained and incorporated here. Meta-Analysis The Review Manager (RevMan Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014) software version 5.3 for Windows (Microsoft, Redmond, Washington) was used to conduct meta-analyses and generate figures. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated from dichotomous data on the incidences of leptomeningeal dissemination and hydrocephalus in each study. Using Mantel–Haenszel random effects model, meta-analyses for these outcomes were performed to calculate a summary OR with 95% CIs. I2, Tau2, and Chi2 (Cochrane's Q) statistics were used to assess heterogeneity among the studies included in the meta-analyses. Significant heterogeneity was deemed with I2 > 75%, Tau2 > 1, and Chi2P value < .05.12 Sensitivity analysis in each meta-analysis was conducted to assess for a significant change in the summary OR by (1) sequential exclusion of 1 study at a time; (2) limiting the meta-analyses to only studies of patients likely treated in the temozolomide era (2005 or later); (3) excluding studies that examined patients with recurrent GBM or WHO grade III glioma; and (4) excluding patients with documented obstructive hydrocephalus. To explore potential heterogeneity related to possible trends in the incidences of our outcomes that may accompany increased survival in patients over time, a meta-regression was conducted to assess the effects of the year of study publication and the maximum length of follow-up in the studies on the outcomes. Publication bias was assessed using the following 2 methods: a funnel plot plotting ORs against standard error and Begg and Mazumdar's and Egger's tests. An asymmetrical funnel plot or a P value meeting significance in the latter 2 tests indicated potential publication bias. Meta-regressions and statistical analyses for publication bias were conducted using The Comprehensive Meta-Analysis software version 2.2 (Biostat Inc, Englewood, New Jersey) for Windows. Statistical significance was identified with a 2-tailed P ≤ .05. Lastly, the quality of overall evidence was summarized with grades of low, medium, or high based on the 5 GRADE criteria:15 risk of bias, inconsistency (heterogeneity), indirectness (data representativeness), imprecision, and publication bias. Survival Analysis Survival data of patients with and without ventricular entry were presented heterogeneously in studies (ie, tabulated numerical data, Kaplan–Meier survival curves, and often without any survival effect size statistics). For a uniform analysis in the absence of published individual patient data, we digitized published Kaplan–Meier survival curves using WebPlotDigitizer (Version 3.12; June 2017; http://arohatgi.info/WebPlotDigitizer) to recreate individual patient data, an accepted methodology for meta-analysis of time-to-event data.16,17 The individual survival times obtained at censored and noncensored events and proportion of population surviving at those times were used to recreate published Kaplan–Meier survival curves. Accuracy of the reconstructed individual patient data was assessed by visually overlaying the reconstructed Kaplan–Meier curve with the original published curve. In all cases, near accurate curves were generated (refer to example in Figure, Supplemental Digital Content 2). All reconstructed and published individual patient survival data were pooled for Kaplan–Meier survival analysis. Log-rank (Mantel–Cox) text was used to compare survival of patients with ventricular entry to those without. Data are reported with a hazard ratio (HR) and 95% CI. Age, tumor grade, extent of resection, treatment with chemotherapy (including temozolomide), and radiation were noted for patients included in the survival analysis. Fisher's exact test was used to compare proportions between ventricular entry and no entry groups. Survival curve reconstructions, analyses, and Fisher's exact tests were performed using GraphPad Prism version 7.00 for Windows (GraphPad Software, San Diego, CA). Statistical significance was identified with a 2-tailed P ≤ 0.05. RESULTS Search Result and Study Characteristics Our search resulted in 6910 articles out of which 20 studies3,5,13,14,18-33 met our criteria for inclusion in the meta-analysis (Figure, Supplemental Digital Content 1). Specifically, 10 studies (References, Supplemental Digital Content 3) were excluded from the meta-analysis due to incomplete data. All were retrospective observational studies that included adult patients with high-grade glioma. A vast majority of the patients had glioblastoma. Overall, the case-controlled or cohort studies were assessed to be moderate in quality using the NOS (Table, Supplemental Digital Content 4). When specified, the lateral cerebral ventricles were most often the structures that were breached during resections. Association of Leptomeningeal Dissemination With Ventricular Entry We studied the association of leptomeningeal dissemination with ventricular entry during resections of high-grade gliomas. Leptomeningeal dissemination in all studies was diagnosed radiographically and defined as abnormal enhancement or nodular deposits in the cerebral ventricular wall and/or the subarachnoid spaces of the brain or spinal cord, except in 1 study28 where no clear definition was provided. Meta-analysis of the data revealed higher odds of developing leptomeningeal dissemination with a ventricular entry. The summary OR was calculated to be 3.91 [1.89-8.10] (P = .0002) from 9 studies (n = 86/410 vs 57/847 for patients with and without ventricular entry, respectively; Figure 1A). The funnel plot of the included studies (Figure 1B) was visually assessed to be mostly symmetric, implying no significant publication bias. Association of Hydrocephalus With Ventricular Entry Because hydrocephalus is commonly coexistent with leptomeningeal dissemination,3,6-9,24,25,34 we sought to reveal an association between ventricular entry during high-grade glioma resection and the development of hydrocephalus requiring CSF diversion. In studies where patients developed hydrocephalus (9 out of 11), insertion of a shunt was the most common form of treatment chosen (8 out of the 9 studies) for hydrocephalus, and specifically a ventriculoperitoneal shunt (specified in 5 out of the 9 studies). Meta-analysis of data from 11 studies demonstrated higher odds for developing hydrocephalus with a ventricular entry (summary OR of 7.78 [3.77-16.05]; P < .00001; n = 58/431 vs 11/565 for patients with and without ventricular entry, respectively; Figure 2A). The funnel plot of the included studies (Figure 2B) was visually assessed to be mostly symmetric, implying no significant publication bias. FIGURE 2. View largeDownload slide Meta-analysis of studies comparing A, incidences of hydrocephalus in patients with and without ventricular entry during their high-grade glioma resection with its B, respective funnel plot to assess publication bias. The lines in the forest plot of the meta-analysis represent 95% CI, and the size of its corresponding square marker reflects the relative weight of the study used in random-effects Mantel–Haenszel method. The diamond shows the overall effect. In the funnel plot, vertical line represents the summary OR with diagonal lines representing 95 % CIs. Each study is represented by a circle. a = ventricle entry was defined as ≥1 cm defect; b = data from initial surgeries with known ventricular breach status; c = judged as nonobstructive, external hydrocephalus; no treatment reported; d = data available for only frontal lobe malignant gliomas. FIGURE 2. View largeDownload slide Meta-analysis of studies comparing A, incidences of hydrocephalus in patients with and without ventricular entry during their high-grade glioma resection with its B, respective funnel plot to assess publication bias. The lines in the forest plot of the meta-analysis represent 95% CI, and the size of its corresponding square marker reflects the relative weight of the study used in random-effects Mantel–Haenszel method. The diamond shows the overall effect. In the funnel plot, vertical line represents the summary OR with diagonal lines representing 95 % CIs. Each study is represented by a circle. a = ventricle entry was defined as ≥1 cm defect; b = data from initial surgeries with known ventricular breach status; c = judged as nonobstructive, external hydrocephalus; no treatment reported; d = data available for only frontal lobe malignant gliomas. Heterogeneity, Sensitivity Analyses, and Bias Overall, no significant heterogeneity was noted in the 2 meta-analyses (Tau² ≤ 0.40; Chi² P value ≥ .11; I² ≤ 41%). Heterogeneity was further explored by examining time-dependent trends in the outcomes using meta-regression. Specifically, whether recently published studies reported lower incidences of the outcomes (perhaps reflecting adoption of more and better therapies), or if higher reported incidences of the outcomes trended directly with the maximum length of patient follow-up in the studies were examined. Meta-regressions of the year of study publication and the maximum length of follow-up in the studies with the ORs of the 2 outcomes (leptomeningeal dissemination and hydrocephalus) reported in the studies did not reveal any statistically meaningful relationships (Table, Supplemental Digital Content 5). The following sensitivity analyses conducted on the meta-analyses did not change the significance of the summary ORs: (1) sequential exclusion of 1 study at a time; (2) limiting the meta-analyses to only studies of patients likely treated in the temozolomide era; (3) excluding studies that had patients with recurrent GBM or WHO grade III glioma (Table, Supplemental Digital Content 6); and (4) excluding patients with documented obstructive hydrocephalus (there were only 2 such patients, 1 in each group, from Montano et al, 201128). Funnel plots generated were mostly visually symmetric (Figures 1B and 2B), and statistical analysis did not suggest publication bias in the meta-analyses (Table, Supplemental Digital Content 7). However, the included studies overall had a considerable risk of bias (Figure, Supplemental Digital Content 8). Effect of Ventricular Entry on Survival Although tumor dissemination and hydrocephalus are known to adversely impact survival, whether ventricular entry itself influences survival is debated. Some studies5,19,22,35,36 demonstrate no association, and others18,20,23,26 suggest its association with decreased survival. Therefore, we conducted a pooled survival analysis of published overall survival data for a total of 795 patients in 10 studies5,14,18-20,23,26,30,31,37 to glean the survival impact of ventricular entry. Kaplan–Meier survival analysis revealed a median survival time of 19.1 mo in 373 patients who did not have ventricular entry and 16.8 mo in 422 patients who had a ventricular entry (Figure 3). This survival difference was statistically significant (HR: 1.25 [1.05-1.48], P = .01). Analysis of the available data revealed that mean age and the proportions of patients who received gross total resection, subtotal (cytoreductive) resection, chemotherapy, including temozolomide, and radiation therapy did not significantly differ between these 2 groups (Table). Treatment information, however, was more often unavailable in the ventricular entry group. FIGURE 3. View largeDownload slide Kaplan–Meier overall survival curves of pooled survival data in patients reported in studies with and without ventricular entry during their high-grade glioma resection. HR represents log-rank hazard ratio. Tick marks below the curves represent censored values, representing alive patients or unknown living status. FIGURE 3. View largeDownload slide Kaplan–Meier overall survival curves of pooled survival data in patients reported in studies with and without ventricular entry during their high-grade glioma resection. HR represents log-rank hazard ratio. Tick marks below the curves represent censored values, representing alive patients or unknown living status. TABLE. Difference in the Commonly Reported Survival Predictors Between Patients With and Without Ventricular Entry During High-Grade Glioma Resection Ventricular entry (n = 422) No ventricular entry (n = 373) Fisher's exact test (P value) Age, mean, years, (total n)a 56.6 (n = 187) 57.1 (n = 186) a Tumor grade  WHO grade III (%) 19/286 9/270 .08  Unknownb (%) 136/422 103/373 .16 Extent of resection  GTR (%) 83/201 (41.3%) 90/209 (43.1%) .76  STR (%) 173/247 (70.0%) 159/249 (63.9%) .15  Unknown (%) 166/422 (39.3%) 124/373 (33.2%) .05 Chemotherapy  Temozolomide (%) 182/236 (77.1%) 193/249 (77.5%) 1.00  Any chemotherapy (%) 232/277 (83.8%) 227/270 (84.1%) 1.00  Unknown (%) 145/422 (34.4%) 103/373 (27.6%) .05 Radiation therapy  Any radiotherapy (%) 262/274 (95.6%) 260/270 (96.3%) .83  Unknown (%) 148/422 (35.1%) 103/373 (27.6%) .03 Ventricular entry (n = 422) No ventricular entry (n = 373) Fisher's exact test (P value) Age, mean, years, (total n)a 56.6 (n = 187) 57.1 (n = 186) a Tumor grade  WHO grade III (%) 19/286 9/270 .08  Unknownb (%) 136/422 103/373 .16 Extent of resection  GTR (%) 83/201 (41.3%) 90/209 (43.1%) .76  STR (%) 173/247 (70.0%) 159/249 (63.9%) .15  Unknown (%) 166/422 (39.3%) 124/373 (33.2%) .05 Chemotherapy  Temozolomide (%) 182/236 (77.1%) 193/249 (77.5%) 1.00  Any chemotherapy (%) 232/277 (83.8%) 227/270 (84.1%) 1.00  Unknown (%) 145/422 (34.4%) 103/373 (27.6%) .05 Radiation therapy  Any radiotherapy (%) 262/274 (95.6%) 260/270 (96.3%) .83  Unknown (%) 148/422 (35.1%) 103/373 (27.6%) .03 GTR, gross total resection; STR, subtotal (cytoreductive) resection; WHO, World Health Organization aMean age was calculated by summing the product of study means by their study size and dividing this sum by the sum of all sample sizes. Because data on variance were missing, a formal statistical t-test could not be performed for comparisons of the age means. bThe total number of WHO grade III patients is 54 among the 239 patients whose tumor grade distribution is unknown. View Large TABLE. Difference in the Commonly Reported Survival Predictors Between Patients With and Without Ventricular Entry During High-Grade Glioma Resection Ventricular entry (n = 422) No ventricular entry (n = 373) Fisher's exact test (P value) Age, mean, years, (total n)a 56.6 (n = 187) 57.1 (n = 186) a Tumor grade  WHO grade III (%) 19/286 9/270 .08  Unknownb (%) 136/422 103/373 .16 Extent of resection  GTR (%) 83/201 (41.3%) 90/209 (43.1%) .76  STR (%) 173/247 (70.0%) 159/249 (63.9%) .15  Unknown (%) 166/422 (39.3%) 124/373 (33.2%) .05 Chemotherapy  Temozolomide (%) 182/236 (77.1%) 193/249 (77.5%) 1.00  Any chemotherapy (%) 232/277 (83.8%) 227/270 (84.1%) 1.00  Unknown (%) 145/422 (34.4%) 103/373 (27.6%) .05 Radiation therapy  Any radiotherapy (%) 262/274 (95.6%) 260/270 (96.3%) .83  Unknown (%) 148/422 (35.1%) 103/373 (27.6%) .03 Ventricular entry (n = 422) No ventricular entry (n = 373) Fisher's exact test (P value) Age, mean, years, (total n)a 56.6 (n = 187) 57.1 (n = 186) a Tumor grade  WHO grade III (%) 19/286 9/270 .08  Unknownb (%) 136/422 103/373 .16 Extent of resection  GTR (%) 83/201 (41.3%) 90/209 (43.1%) .76  STR (%) 173/247 (70.0%) 159/249 (63.9%) .15  Unknown (%) 166/422 (39.3%) 124/373 (33.2%) .05 Chemotherapy  Temozolomide (%) 182/236 (77.1%) 193/249 (77.5%) 1.00  Any chemotherapy (%) 232/277 (83.8%) 227/270 (84.1%) 1.00  Unknown (%) 145/422 (34.4%) 103/373 (27.6%) .05 Radiation therapy  Any radiotherapy (%) 262/274 (95.6%) 260/270 (96.3%) .83  Unknown (%) 148/422 (35.1%) 103/373 (27.6%) .03 GTR, gross total resection; STR, subtotal (cytoreductive) resection; WHO, World Health Organization aMean age was calculated by summing the product of study means by their study size and dividing this sum by the sum of all sample sizes. Because data on variance were missing, a formal statistical t-test could not be performed for comparisons of the age means. bThe total number of WHO grade III patients is 54 among the 239 patients whose tumor grade distribution is unknown. View Large Quality of Evidence Quality of the data and results presented herein based on 5 GRADE criteria15 was assessed as follows: bias (low quality), inconsistency (high quality), indirectness (medium quality), imprecision (medium quality), and publication bias (high quality; Table, Supplemental Digital Content 9). Therefore, overall this work represents at best a “medium” level of quality of evidence. DISCUSSION A primary goal of the surgical treatment of high-grade glioma is maximal, safe resection of the tumor. In the process of achieving this, the ventricular system is not uncommonly entered. The consequences of ventricular entry are often not detectable clinically, and it is commonly assumed that entering the ventricles during high-grade glioma resection is not a source of major morbidity or mortality in these patients. Although few studies have reported a link between glioma dissemination and the development of hydrocephalus with ventricular breach during surgery, the low incidence of these events in postoperative course of high-grade glioma patients may have obscured a clear association. Here, in a quantitative meta-analysis, we reveal statistically compelling associations between them, which may have important clinical and biological implications. Hypotheses for Cancer Dissemination Following Ventricular Entry We demonstrate a strong association of leptomeningeal dissemination with ventricular entry. Our results support the prior proposed hypothesis4,5 that neoplastic cells spilled into the CSF in the ventricles may disseminate and seed elsewhere in the brain. Further biological investigations are warranted to test this hypothesis and to understand whether glioma cells (1) survive in the circulating CSF, (2) alter CSF fluid dynamics, and (3) seed in the leptomeninges and ventricular linings. The interface between the ventricular system and the brain contains the biologically relevant subventricular zone. It is a large, prominent neural stem cell zone that lies throughout the entire lateral walls of the lateral ventricles,38,39 and therefore, it is often violated by surgical entry into the ventricles. Both clinical40,41 and biological42,43 studies have shown that glioma cells preferentially migrate towards this growth factor rich “lake-front property.”44 Whether surgically disseminated glioma cells in the CSF display the same affinity or tropism for the subventricular zone is relevant to understanding the leptomeningeal dissemination that occurs after ventricular entry. Further, subventricular zone tumor contact may confound this association as any residual glioma cells after resection could continue to spread along the subventricular zone. Hypotheses for Hydrocephalus Development and Its Impact Communicating hydrocephalus is often noted in the presence of leptomeningeal dissemination.3,6-9,24,25,34 If ventricular entry is associated with the development of leptomeningeal dissemination, then by logical extension, we tested and found that it was also associated with the development of hydrocephalus. Hydrocephalus arises from increased pressure in the CSF often either from its increased production (by cells of the choroid plexus) or decreased absorption (by arachnoid granulations). The biological basis for the alteration in CSF dynamics following ventricular breach during high-grade glioma resection is unclear. Some have suggested the cause to be malabsorption by arachnoid granulations due to increased protein content and its precipitation in the CSF that accompanies the presence of tumor cells28,32,45 or surgical debris. Prevention of hydrocephalus is important not only from a survival perspective, but also in ensuring the best quality of life for patients, by avoiding additional procedures and delays in starting cancer therapy.28,32,45,46 Minimal Effect of Ventricular Entry on Overall Survival The pooled survival analysis of the available published overall survival data demonstrated decreased survival of high-grade glioma patients with ventricular entry during surgery. Before reaching to clinical implications of these results, we draw attention to a few considerations. First, although we demonstrated nonsignificant differences in age and treatment between patients with and without ventricular entry, the survival analysis is unadjusted due to the lack of individual patient data for their effects and other important prognostic variables (ie, patient performance status, MGMT promoter methylation or IDH1/2 mutation status of the gliomas). Compared to the survival effect sizes of these known predictors, the survival effect of ventricular entry, though significant, is smaller (median difference of 2.3 mo) and may not be a prominent contributor in predicting patient survival in a multivariable, adjusted analysis. Second, the primary cause of death from high-grade glioma is treatment resistance that manifests primarily as local recurrence.47-53 Glioma dissemination and hydrocephalus are uncommon. Therefore, presently, we believe the best surgical treatment is maximal and safe cytoreduction of the glioma that is currently supported by a significant body of scientific literature.54 The consequences of ventricular entry during surgery should not alter this primary goal. Rather, patients should be counseled, and surgeons and treating oncologists must be aware of the possibilities of tumor dissemination and hydrocephalus, so that they can be managed in time. If gross total resections can be achieved without ventricular entry, the pursuit of such approaches is reasonable. Limitations Despite adjustments with sensitivity analyses, assessments of heterogeneity and risk of bias, this meta-analysis is limited by 1) the inherent bias present in all studies related to their retrospective observational designs that may overestimate the effect sizes derived and 2) possible variations in ventricular entry and outcome assessments during the time-frame of the studies included. Some of these studies were not designed to explore the link between ventricular breach and our outcomes of interest. Further, the population of patients in these studies may be subject to reporting bias and a confirmatory prospective analysis is indicated to support the validity of the findings in this manuscript. Lastly, we recognize that time-to-event analysis with individual patient data adjusting for potential covariates, such as and importantly ventricular or subventricular zone contact—not possible in this study design—as necessary to reveal any time-dependent patterns in the occurrence of our outcomes, including overall survival (as discussed above), and to understand the contributions of the covariates in predicting these outcomes. CONCLUSION We uncover a strong association between cerebral ventricular entry that occurs during high-grade glioma resections and leptomeningeal dissemination, hydrocephalus, and overall survival. The quality of the data and results supporting this association are judged to be medium at best. This association encourages further investigations to unveil any confounding or causal links among these factors. Our results should not deter neurosurgeons from achieving a maximum possible glioma resection as supported by the current evidence54 to avoid ventricular entry. Treating physicians should become aware of and recognize the consequences of ventricular entry for their timely management. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Pietschmann S , von Bueren AO , Kerber MJ , Baumert BG , Kortmann RD , Muller K . An individual patient data meta-analysis on characteristics, treatments and outcomes of glioblastoma/gliosarcoma patients with metastases outside of the central nervous system . PLoS One . 2015 ; 10 ( 4 ): e0121592 . Google Scholar CrossRef Search ADS PubMed 2. Pietschmann S , von Bueren AO , Henke G , Kerber MJ , Kortmann RD , Muller K . An individual patient data meta-analysis on characteristics, treatments and outcomes of the glioblastoma/gliosarcoma patients with central nervous system metastases reported in literature until 2013 . J Neurooncol . 2014 ; 120 ( 3 ): 451 - 457 . Google Scholar CrossRef Search ADS PubMed 3. Mandel JJ , Yust-Katz S , Cachia D et al. Leptomeningeal dissemination in glioblastoma; an inspection of risk factors, treatment, and outcomes at a single institution . J Neurooncol . 2014 ; 120 ( 3 ): 597 - 605 . Google Scholar CrossRef Search ADS PubMed 4. Grabb PA , Albright AL , Pang D . Dissemination of supratentorial malignant gliomas via the cerebrospinal fluid in children . Neurosurgery . 1992 ; 30 ( 1 ): 64 - 71 . Google Scholar CrossRef Search ADS PubMed 5. Elliott JP , Keles GE , Waite M , Temkin N , Berger MS . Ventricular entry during resection of malignant gliomas: effect on intracranial cerebrospinal fluid tumor dissemination . J Neurosurg . 1994 ; 80 ( 5 ): 834 - 839 . Google Scholar CrossRef Search ADS PubMed 6. Awad I , Bay JW , Rogers L . Leptomeningeal metastasis from supratentorial malignant gliomas . Neurosurgery . 1986 ; 19 ( 2 ): 247 - 251 . Google Scholar CrossRef Search ADS PubMed 7. Delattre JY , Walker RW , Rosenblum MK . Leptomeningeal gliomatosis with spinal cord or cauda equina compression: a complication of supratentorial gliomas in adults . Acta Neurol Scand . 1989 ; 79 ( 2 ): 133 - 139 . Google Scholar CrossRef Search ADS PubMed 8. Onda K , Tanaka R , Takahashi H , Takeda N , Ikuta F . Symptomatic cerebrospinal fluid dissemination of cerebral glioblastoma . Neuroradiology . 1990 ; 32 ( 2 ): 146 - 150 . Google Scholar CrossRef Search ADS PubMed 9. Ito U , Tomita H , Yamazaki S , Takada Y , Inaba Y . CT findings of leptomeningeal and periventricular dissemination of tumors . Clin Neurol Neurosurg . 1986 ; 88 ( 2 ): 115 - 120 . Google Scholar CrossRef Search ADS PubMed 10. Moher D , Liberati A , Tetzlaff J , Altman DG , Group P . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement . PLoS Med . 2009 ; 6 ( 7 ): e1000097 . Google Scholar CrossRef Search ADS PubMed 11. Wells GA , Shea B , O’Connell D et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses . Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed September 13, 2017 . 12. Higgins JPT , Green S . Cochrane Handbook for Systematic Reviews of Interventions . Vol 4 . Hoboken, NJ : Wiley ; 2011 . 13. John JK , Robin AM , Pabaney AH et al. Complications of ventricular entry during craniotomy for brain tumor resection . J Neurosurg . 2017 ; 127 ( 2 ): 426 - 432 . Google Scholar CrossRef Search ADS PubMed 14. Chan DT , Hsieh SY , Kam MK , Cheung TC , Ng SC , Poon WS . Pattern of recurrence and factors associated with cerebrospinal fluid dissemination of glioblastoma in Chinese patients . Surg Neurol Int . 2016 ; 7 ( 1 ): 92 . Google Scholar CrossRef Search ADS PubMed 15. Ryan R , Hill S . The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed January 11, 2018 . 16. Wan X , Peng L , Li Y . A review and comparison of methods for recreating individual patient data from published Kaplan-Meier survival curves for economic evaluations: a simulation study . PLoS One . 2015 ; 10 ( 3 ): e0121353 . Google Scholar CrossRef Search ADS PubMed 17. Mistry AM , Hale AT , Chambless LB , Weaver KD , Thompson RC , Ihrie RA . Influence of glioblastoma contact with the lateral ventricle on survival: a meta-analysis . J Neurooncol . 2017 ; 131 ( 1 ): 125 - 133 . Google Scholar CrossRef Search ADS PubMed 18. Sonoda Y , Shibahara I , Matsuda KI et al. Opening the ventricle during surgery diminishes survival among patients with newly diagnosed glioblastoma treated with carmustine wafers: a multi-center retrospective study . J Neurooncol . 2017 ; 134 ( 1 ): 83 - 88 . Google Scholar CrossRef Search ADS PubMed 19. Kondo N , Barth RF , Miyatake SI et al. Cerebrospinal fluid dissemination of high-grade gliomas following boron neutron capture therapy occurs more frequently in the small cell subtype of IDH1R132H mutation-negative glioblastoma . J Neurooncol . 2017 ; 133 ( 1 ): 107 - 118 . Google Scholar CrossRef Search ADS PubMed 20. Behling F , Kaltenstadler M , Noell S et al. The prognostic impact of ventricular opening in glioblastoma surgery: A retrospective single center analysis . World Neurosurg . 2017 ; 106 : 615 - 624 . Google Scholar CrossRef Search ADS PubMed 21. Hasegawa Y , Iuchi T , Sakaida T , Yokoi S , Kawasaki K . The influence of carmustine wafer implantation on tumor bed cysts and peritumoral brain edema . J Clin Neurosci . 2016 ; 31 : 67 - 71 . Google Scholar CrossRef Search ADS PubMed 22. Adeberg S , Diehl C , Jung CS et al. Is a modification of the radiotherapeutic target volume necessary after resection of glioblastomas with opening of the ventricles? J Neurooncol . 2016 ; 127 ( 3 ): 581 - 587 . Google Scholar CrossRef Search ADS PubMed 23. Roelz R , Reinacher P , Jabbarli R et al. Surgical ventricular entry is a key risk factor for leptomeningeal metastasis of high grade gliomas . Sci Rep . 2016 ; 5 ( 1 ): 17758 . Google Scholar CrossRef Search ADS 24. Fischer CM , Neidert MC , Peus D et al. Hydrocephalus after resection and adjuvant radiochemotherapy in patients with glioblastoma . Clin Neurol Neurosurg . 2014 ; 120 : 27 - 31 . Google Scholar CrossRef Search ADS PubMed 25. Onuma K , Ishikawa E , Matsuda M et al. Clinical characteristics and neuroimaging findings in 12 cases of recurrent glioblastoma with communicating hydrocephalus . Neurol Med Chir (Tokyo) . 2013 ; 53 ( 7 ): 474 - 481 . Google Scholar CrossRef Search ADS PubMed 26. Tejada-Solis S , Aldave-Orzaiz G , Pay-Valverde E , Marigil-Sanchez M , Idoate-Gastearena MA , Diez-Valle R . Prognostic value of ventricular wall fluorescence during 5-aminolevulinic-guided surgery for glioblastoma . Acta Neurochir . 2012 ; 154 ( 11 ): 1997 - 2002 ; discussion 2002 . Google Scholar CrossRef Search ADS PubMed 27. Konishi Y , Muragaki Y , Iseki H , Mitsuhashi N , Okada Y . Patterns of intracranial glioblastoma recurrence after aggressive surgical resection and adjuvant management: retrospective analysis of 43 cases . Neurol Med Chir (Tokyo) . 2012 ; 52 ( 8 ): 577 - 586 . Google Scholar CrossRef Search ADS PubMed 28. Montano N , D’Alessandris QG , Bianchi F et al. Communicating hydrocephalus following surgery and adjuvant radiochemotherapy for glioblastoma . J Neurosurg . 2011 ; 115 ( 6 ): 1126 - 1130 . Google Scholar CrossRef Search ADS PubMed 29. Della Puppa A , Rossetto M , Ciccarino P et al. Carmustine wafer implantation when surgical cavity is communicating with cerebral ventricles: technical considerations on a clinical series . World Neurosurg . 2011 ; 76 ( 1-2 ): 156 - 159 ; discussion 167-158 . Google Scholar CrossRef Search ADS PubMed 30. Bock HC , Cohnen J , Keric N , Kantelhardt SR , Giese A . Occlusion of surgical opening of the ventricular system with fibrinogen-coated collagen fleece: a case collection study . Acta Neurochir . 2011 ; 153 ( 3 ): 533 - 539 . Google Scholar CrossRef Search ADS PubMed 31. Bae JS , Yang SH , Yoon WS , Kang SG , Hong YK , Jeun SS . The clinical features of spinal leptomeningeal dissemination from malignant gliomas . J Korean Neurosurg Soc . 2011 ; 49 ( 6 ): 334 - 338 . Google Scholar CrossRef Search ADS PubMed 32. Marquardt G , Setzer M , Lang J , Seifert V . Delayed hydrocephalus after resection of supratentorial malignant gliomas . Acta Neurochir (Wien) . 2002 ; 144 ( 3 ): 227 - 231 ; discussion 231 . Google Scholar CrossRef Search ADS PubMed 33. Buhl R , Barth H , Hugo HH , Hutzelmann A , Mehdorn HM . Spinal drop metastases in recurrent glioblastoma multiforme . Acta Neurochir (Wien) . 1998 ; 140 ( 10 ): 1001 - 1005 . Google Scholar CrossRef Search ADS PubMed 34. Arita N , Taneda M , Hayakawa T . Leptomeningeal dissemination of malignant gliomas. Incidence, diagnosis and outcome . Acta Neurochir . 1994 ; 126 ( 2-4 ): 84 - 92 . Google Scholar CrossRef Search ADS PubMed 35. Adeberg S , Konig L , Bostel T et al. Glioblastoma recurrence patterns after radiation therapy with regard to the subventricular zone . Int J Radiat Oncol Biol Phys . 2014 ; 90 ( 4 ): 886 - 893 . Google Scholar CrossRef Search ADS PubMed 36. Shibahara I , Sonoda Y , Saito R et al. The expression status of CD133 is associated with the pattern and timing of primary glioblastoma recurrence . Neuro-oncol . 2013 ; 15 ( 9 ): 1151 - 1159 . Google Scholar CrossRef Search ADS PubMed 37. Saito R , Kumabe T , Kanamori M , Sonoda Y , Tominaga T . Distant recurrences limit the survival of patients with thalamic high-grade gliomas after successful resection . Neurosurg Rev . 2017 ; 40 ( 3 ): 469 - 477 . Google Scholar CrossRef Search ADS PubMed 38. Sanai N , Tramontin AD , Quinones-Hinojosa A et al. Unique astrocyte ribbon in adult human brain contains neural stem cells but lacks chain migration . Nature . 2004 ; 427 ( 6976 ): 740 - 744 . Google Scholar CrossRef Search ADS PubMed 39. Sanai N , Nguyen T , Ihrie RA et al. Corridors of migrating neurons in the human brain and their decline during infancy . Nature . 2011 ; 478 ( 7369 ): 382 - 386 . Google Scholar CrossRef Search ADS PubMed 40. Mistry AM , Dewan MC , White-Dzuro GA et al. Decreased survival in glioblastomas is specific to contact with the ventricular-subventricular zone, not subgranular zone or corpus callosum . J Neurooncol . 2017 ; 132 ( 2 ): 341 - 349 . Google Scholar CrossRef Search ADS PubMed 41. Parsa AT , Wachhorst S , Lamborn KR et al. Prognostic significance of intracranial dissemination of glioblastoma multiforme in adults . J Neurosurg . 2005 ; 102 ( 4 ): 622 - 628 . Google Scholar CrossRef Search ADS PubMed 42. Qin EY , Cooper DD , Abbott KL et al. Neural precursor-derived pleiotrophin mediates subventricular zone invasion by glioma . Cell . 2017 ; 170 ( 5 ): 845 - 859.e19 . Google Scholar CrossRef Search ADS PubMed 43. Goffart N , Kroonen J , Di Valentin E et al. Adult mouse subventricular zones stimulate glioblastoma stem cells specific invasion through CXCL12/CXCR4 signaling . Neuro-oncol . 2015 ; 17 ( 1 ): 81 - 94 . Google Scholar CrossRef Search ADS PubMed 44. Ihrie RA , Alvarez-Buylla A . Lake-front property: a unique germinal niche by the lateral ventricles of the adult brain . Neuron . 2011 ; 70 ( 4 ): 674 - 686 . Google Scholar CrossRef Search ADS PubMed 45. Inamasu J , Nakamura Y , Saito R et al. Postoperative communicating hydrocephalus in patients with supratentorial malignant glioma . Clin Neurol Neurosurg . 2003 ; 106 ( 1 ): 9 - 15 . Google Scholar CrossRef Search ADS PubMed 46. Esquenazi Y , Moussazadeh N , Link TW et al. Thalamic glioblastoma: clinical presentation, management strategies, and outcomes . Neurosurgery . 2017 . doi: 10.1093/neuros/nyx349. [Published online ahead of print ]. 47. Brandes AA , Tosoni A , Franceschi E et al. Recurrence pattern after temozolomide concomitant with and adjuvant to radiotherapy in newly diagnosed patients with glioblastoma: correlation with MGMT promoter methylation status . J Clin Oncol . 2009 ; 27 ( 8 ): 1275 - 1279 . Google Scholar CrossRef Search ADS PubMed 48. Rapp M , Baernreuther J , Turowski B , Steiger HJ , Sabel M , Kamp MA . Recurrence pattern analysis of primary glioblastoma . World Neurosurg . 2017 ; 103 : 733 - 740 . Google Scholar CrossRef Search ADS PubMed 49. Gebhardt BJ , Dobelbower MC , Ennis WH , Bag AK , Markert JM , Fiveash JB . Patterns of failure for glioblastoma multiforme following limited-margin radiation and concurrent temozolomide . Radiat Oncol . 2014 ; 9 ( 1) : 130 . Google Scholar CrossRef Search ADS PubMed 50. Sherriff J , Tamangani J , Senthil L et al. Patterns of relapse in glioblastoma multiforme following concomitant chemoradiotherapy with temozolomide . Br J Radiol . 2013 ; 86 ( 1022 ): 20120414 . Google Scholar CrossRef Search ADS PubMed 51. Milano MT , Okunieff P , Donatello RS et al. Patterns and timing of recurrence after temozolomide-based chemoradiation for glioblastoma . Int J Radiat Oncol Biol Phys . 2010 ; 78 ( 4 ): 1147 - 1155 . Google Scholar CrossRef Search ADS PubMed 52. Chamberlain MC . Radiographic patterns of relapse in glioblastoma . J Neurooncol . 2011 ; 101 ( 2 ): 319 - 323 . Google Scholar CrossRef Search ADS PubMed 53. Zhou X , Liao X , Zhang B et al. Recurrence patterns in patients with high-grade glioma following temozolomide-based chemoradiotherapy . Mol Clin Oncol . 2016 ; 5 ( 2 ): 289 - 294 . Google Scholar CrossRef Search ADS PubMed 54. Brown TJ , Brennan MC , Li M et al. Association of the extent of resection with survival in glioblastoma . JAMA Oncol . 2016 ; 2 ( 11 ): 1460 - 1469 . Google Scholar CrossRef Search ADS PubMed Supplemental digital content is available for this article at www.neurosurgery-online.com. Supplemental Digital Content 1. Figure. PRISMA Flow Diagram. PRISMA Flow Diagram represents the process of searching, identifying, and screening published studies for inclusion in the meta-analysis. The template utilized is from Moher et al.10 Supplemental Digital Content 2. Figure. An example comparing published and reconstructed Kaplan–Meier survival curves after digitization. The original Kaplan–Meier survival curve is reprinted by permission from Springer Science + Business Media New York: Springer Nature, Journal of Neuro-Oncology18, Opening the ventricle during surgery diminishes survival among patients with newly diagnosed glioblastoma treated with carmustine wafers: a multi-center retrospective study, Sonoda Y et al, © 2017. Supplemental Digital Content 3. References. Excluded Studies. Studies with incomplete data for inclusion in meta-analysis. Supplemental Digital Content 4. Table. Newcastle-Ottawa Assessment. Quality assessment of studies included in the meta-analyses using the Newcastle-Ottawa Assessment Scale. Supplemental Digital Content 5. Table. Meta-Regression Analyses. Results of meta-regressions of the year of study publication and the maximum length of study follow-up in the studies with the OR of the 3 outcomes studies in the 3 meta-analyses. Supplemental Digital Content 6. Table. Recurrent high-grade and WHO Grade III gliomas. List of studies with the number and proportions of patients with recurrent high-grade gliomas and WHO Grade III gliomas. Supplemental Digital Content 7. Table. Publication bias. Visual and statistical funnel plot analyses to assess publication bias. Supplemental Digital Content 8. Figure. Analyses of risk of bias in studies included in meta-analyses. The risk of several types of biases in each study included in the meta-analyses was assessed by denoting a red circle with a minus sign for high risk and green circle with plus sign for low risk. The risk of bias is assessed based on studies’ intended outcomes of interest. Supplemental Digital Content 9. Table. Quality of Evidence Assessment. Each of the 5 “GRADE” criteriaa is graded on a low, medium, high quality. COMMENTS The authors report a thoughtfully conducted meta-analysis with purposeful sensitivity analyses aiming to better understand the impact of entering the ventricular system during high-grade glioma surgery. Building upon recent case series, this is a timely study conducted with sound methodological principles and with an honest assessment of limitations, including lack of information on MGMT and IDH1 status. Importantly, the authors report rates of gross total resection did not significantly differ by ventricular entry, although distal parenchymal recurrence, leptomeningeal spread, hydrocephalus, overall survival were worse in patients with ventricular entry. This study reminds us that tumor spread and hydrocephalus may be unintended sequela of entering the ventricular system during surgical resection. Entry into the ventricular system may additionally impact visual fields, particular around the lateral aspect of the temporal horns, although this particular outcome was not catalogued in this study. With increasing interest in performing “supra-total” resection for high-grade gliomas, the findings of this meta-analysis provide an important counter-balance, causing the neurosurgeon to think carefully about the consequences of more aggressive surgical resection. We have increasingly relied on image-guidance systems, intraoperative neuromonitoring including cortical and sub-cortical stimulation, and awake craniotomy techniques to maximize extent of resection while minimizing morbidity. As our field continues to use these advanced technologies to push the boundaries of safe surgical resection for high-grade gliomas, we need to be aware of our surgical technique in the periventricular zone, minimize inadvertent entry into the ventricular system, and thoughtfully counsel our patients when such an occurrence is expected. Debraj Mukherjee Pittsburgh, Pennsylvania The authors address a not uncommon surgical circumstance of ventricular entry during the resection of malignant gliomas as reported in the literature in a retrospective fashion. They have worked to assess survival, hydrocephalus, and leptomeningeal spread via descriptive statistics and attempted to extract more sweeping conclusions via meta-analysis. Limitations to the work are the retrospective analysis of a very unevenly reported event in the literature leaving the data they have subject to a number of biases. They do come up with a suggestion that ventricular entry or subventricular zone contact results in more leptomeningeal dissemination and hydrocephalus, and lessened overall survival. Given the limits in the quality of the data this meta-analysis is based upon, the results do not imply a need to change in techniques used for resection of high-grade gliomas. Ideally these results will serve as a bench mark for the knowledge on this topic and spur a prospective analysis of a surgical series that contains a more homogeneous reporting mechanism regarding ventricular entry so as to preclude the biases encountered in this study. Jeffrey J. Olson Atlanta, Georgia Neurosurgery Speaks (Audio Abstracts) Listen to audio translations of this paper's abstract into select languages by choosing from one of the selections below. Chinese: Liang Chen, MD. Department of Neurosurgery Huashan Hospital Shanghai, China Chinese: Liang Chen, MD. Department of Neurosurgery Huashan Hospital Shanghai, China Close French: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France French: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France Close English: William W. Ashley, MD, PhD, MBA. Department of Neurological Surgery Sinai Hospital and LifeBridge Health System Baltimore, Maryland English: William W. Ashley, MD, PhD, MBA. Department of Neurological Surgery Sinai Hospital and LifeBridge Health System Baltimore, Maryland Close Russian: Natalia Denisova, MD. Novosibirsk Federal Centre of Neurosurgery Novosibirsk, Russia Russian: Natalia Denisova, MD. Novosibirsk Federal Centre of Neurosurgery Novosibirsk, Russia Close Italian: Daniele Bongetta, MD. Department of Neurosurgery Fondazione IRCCS Policlinico San Matteo Pavia, Italy Italian: Daniele Bongetta, MD. Department of Neurosurgery Fondazione IRCCS Policlinico San Matteo Pavia, Italy Close Spanish: Alvaro Campero, MD. Department of Neurosurgery Hospital Zenón Santillán San Miguel de Tucumán Tucumán, Argentina Spanish: Alvaro Campero, MD. Department of Neurosurgery Hospital Zenón Santillán San Miguel de Tucumán Tucumán, Argentina Close Portuguese: Hugo Leonardo Doria-Netto. Department of Micro-Neurosurgery CNC-Centro de Neurociasncia Sao Paulo, Brazil Portuguese: Hugo Leonardo Doria-Netto. Department of Micro-Neurosurgery CNC-Centro de Neurociasncia Sao Paulo, Brazil Close Japanese: Soichi Oya, MD, PhD. Department of Neurosurgery Saitama Medical Center/University Saitama, Japan Japanese: Soichi Oya, MD, PhD. Department of Neurosurgery Saitama Medical Center/University Saitama, Japan Close Korean: Sun Ha Paek, MD, PhD Department of Neurosurgery Seoul National University College of Medicine Seoul, Republic of Korea Korean: Sun Ha Paek, MD, PhD Department of Neurosurgery Seoul National University College of Medicine Seoul, Republic of Korea Close Greek: George Georgoulis, MD Department of Neurosurgery University Hospital of Ioannina Ioannina, Greece Greek: George Georgoulis, MD Department of Neurosurgery University Hospital of Ioannina Ioannina, Greece Close Copyright © 2018 by the Congress of Neurological Surgeons This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Neurosurgery Oxford University Press

Cancer Dissemination, Hydrocephalus, and Survival After Cerebral Ventricular Entry During High-Grade Glioma Surgery: A Meta-Analysis

Loading next page...
 
/lp/ou_press/cancer-dissemination-hydrocephalus-and-survival-after-cerebral-z3nLC5MQWu
Publisher
Congress of Neurological Surgeons
Copyright
Copyright © 2018 by the Congress of Neurological Surgeons
ISSN
0148-396X
eISSN
1524-4040
D.O.I.
10.1093/neuros/nyy202
Publisher site
See Article on Publisher Site

Abstract

Abstract BACKGROUND The consequences of ventricular entry during resection of high-grade gliomas (HGG) are uncertain and often not detectable clinically. OBJECTIVE To reveal odds of tumor dissemination, hydrocephalus, and mortality in adult patients who had ventricular entry during surgical resection of HGG. METHODS Titles and abstracts of published journals in the NCBI/NLM PubMed and OVID EMBASE databases were searched without language restriction and systematically screened. Outcomes extracted included the odds of leptomeningeal dissemination and hydrocephalus in patients with ventricular entry during HGG resection compared to without. They were analyzed using a random-effects model to calculate summary odds ratios (sORs). Overall survival data were also compared between patients with and without ventricular entry. RESULTS Twenty final studies with 2251 total patients were included from the 6910 retrieved. Patients with ventricular entry during HGG resection demonstrated higher odds of leptomeningeal dissemination (sOR: 3.91 [95% confidence interval (CI): 1.89-8.10]; P = .0002; 86/410 vs 57/847 patients in 9 studies) and hydrocephalus (sOR: 7.78 [95% CI: 3.77-16.05]; P < .00001; 58/431 vs 11/565 patients in 11 studies). They also had decreased survival (median survival: 16.8 vs 19.1 mo; 413 vs 322 patients in 10 studies; hazard ratio: 1.25 [95% CI: 1.05-1.48], P = .01). CONCLUSION The association between ventricular entry during HGG resection and tumor dissemination, hydrocephalus, and decreased survival invites investigations to understand this link. Neurosurgeons and neuro-oncologists must be aware of the consequences of ventricular entry during surgery for HGG. Glioma, Glioblastoma, Dissemination, Hydrocephalus, Survival, Ventricle, Lateral ventricle ABBREVIATIONS ABBREVIATIONS CIs confidence intervals CSF cerebrospinal fluid GBM glioblastoma HR hazard ratio NOS Newcastle–Ottawa Scale ORs odds ratios Surgical treatment of high-grade gliomas is focused on the challenging task of maximal but safe removal of the tumor from the brain primarily to limit local recurrence. Unlike most other cancers, high-grade gliomas uncommonly disseminate within the central nervous system, and even more rarely metastasize outside it.1,2 A commonly recognized form of high-grade glioma dissemination is leptomeningeal dissemination. It is diagnosed radiographically with abnormal enhancement or nodular deposits in the cerebral ventricular wall and the subarachnoid spaces of the brain and spinal cord.3 FIGURE 1. View largeDownload slide Meta-analysis of studies comparing A, incidences of leptomeningeal dissemination in patients with and without ventricular entry during their high-grade glioma resection with its B, respective funnel plot to assess publication bias. The lines in the forest plot represent 95% CI, and the size of its corresponding square marker reflects the relative weight of the study used in random-effects Mantel–Haenszel method. The diamond shows the overall effect. In the funnel plot, vertical line represents the summary OR with diagonal lines representing 95% CIs. Each study is represented by a circle. FIGURE 1. View largeDownload slide Meta-analysis of studies comparing A, incidences of leptomeningeal dissemination in patients with and without ventricular entry during their high-grade glioma resection with its B, respective funnel plot to assess publication bias. The lines in the forest plot represent 95% CI, and the size of its corresponding square marker reflects the relative weight of the study used in random-effects Mantel–Haenszel method. The diamond shows the overall effect. In the funnel plot, vertical line represents the summary OR with diagonal lines representing 95% CIs. Each study is represented by a circle. In the process of achieving maximal, safe resection of high-grade gliomas, the cerebral ventricles are often entered intraoperatively. Early reports noted an association of this ventricular breach with the occurrence of leptomeningeal dissemination.4,5 This led to the emergence of the hypothesis that neoplastic cells could enter into the cerebrospinal fluid (CSF) in the ventricles and then spread via its circulation throughout the nervous system.4,5 Furthermore, hydrocephalus6-9 was often also noted with CSF or leptomeningeal dissemination. Given the relatively rare incidences of leptomeningeal dissemination and hydrocephalus in high-grade gliomas, we conducted a meta-analysis of published data to test the hypothesized associations with ventricular entry during high-grade glioma surgery. Further, we analyzed survival of patients with and without ventricular entry. Here, we reveal an important and strong link between entry into the ventricle—a common intraoperative occurrence during high-grade glioma resections—and leptomeningeal dissemination, hydrocephalus, and survival. METHODS Search Strategy A focused study question was developed with the aid of the common evidence medicine framework PICO (Patient population, Intervention, Control, Outcome): Do adult patients with high-grade glioma (World Health Organization grade III or IV; patient population) who had surgical entry into the cerebral ventricles during resection of their glioma (intervention) have an increased incidence of leptomeningeal dissemination, hydrocephalus, and mortality (outcomes) compared to patients whose ventricles were not breached (control)? The preferred reporting items for systematic reviews and meta-analyses checklist10 were followed to conduct and report this meta-analysis. A prior review protocol that specifically addresses this question does not exist; therefore, authors conducted a search of titles and abstracts of published journals in the NCBI/NLM PubMed (from 1966 to October 30, 2017) and OVID EMBASE (from 1980 to October 30, 2017) databases without language restriction. The following string of terms was used: (astrocytoma* OR oligodendroglioma* OR oligodendroma* OR glioma* OR glioblastoma* OR GBM*) AND (disseminat* OR distant OR metasta* OR multifocal OR “multifocal” OR hydrocephalus OR shunt* OR ventriculomegaly). The retrieved journal articles were imported to the reference manager EndNote X7 (Thompson Reuters, Philadelphia, Pennsylvania) while removing duplicates. Study Selection and Quality Assessment Excluding unpublished studies and conference abstracts, the titles and abstracts of the retrieved journal articles were systematically screened (by A.M.M. and P.D.K.) to select studies for inclusion in the meta-analysis. We included studies with data on our outcomes of interest in patients with or without ventricular entry during surgical resection of their high-grade glioma. We excluded review articles, case series’ of less than 3 patients, studies or data of pediatric patients or patients with gliomatosis cerebri, or studies with incomplete data. Multifocal or multicentric glioblastomas (GBM) were not excluded; however, data or studies of distant parenchymal recurrences were excluded. The screening process yielded 20 final studies with 2251 total patients. The search and selection process are outlined in a Figure, Supplemental Digital Content 1. A bibliography search of these studies did not reveal any additional relevant studies. The Newcastle–Ottawa Scale (NOS) was used to assess the quality of each nonrandomized, case-controlled or cohort study included in the meta-analyses.11 The Cochrane risk of bias assessment tool12 was used to assess the risk of selection, performance, detection, and attrition bias in each study. Publication bias was also assessed (Meta-Analysis subsection of the Methods). Outcomes and Data Extraction A review of all 20 studies was performed (by A.M.M. and P.D.K.). From each final study, the total number of patients with and without ventricular entry during surgical resection of high-grade glioma was noted. In both of these groups, the number of patients with leptomeningeal dissemination and hydrocephalus were noted. Only data on patients who required treatment for their hydrocephalus in the form of CSF diversion were noted. When the cause of hydrocephalus was clear, it was noted to allow for sub-analysis. Additionally, any overall survival data of patients in these 2 groups were extracted. Corresponding authors of studies were contacted to seek clarification of data presented in their publications; clarifications in 2 studies13,14 were obtained and incorporated here. Meta-Analysis The Review Manager (RevMan Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014) software version 5.3 for Windows (Microsoft, Redmond, Washington) was used to conduct meta-analyses and generate figures. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated from dichotomous data on the incidences of leptomeningeal dissemination and hydrocephalus in each study. Using Mantel–Haenszel random effects model, meta-analyses for these outcomes were performed to calculate a summary OR with 95% CIs. I2, Tau2, and Chi2 (Cochrane's Q) statistics were used to assess heterogeneity among the studies included in the meta-analyses. Significant heterogeneity was deemed with I2 > 75%, Tau2 > 1, and Chi2P value < .05.12 Sensitivity analysis in each meta-analysis was conducted to assess for a significant change in the summary OR by (1) sequential exclusion of 1 study at a time; (2) limiting the meta-analyses to only studies of patients likely treated in the temozolomide era (2005 or later); (3) excluding studies that examined patients with recurrent GBM or WHO grade III glioma; and (4) excluding patients with documented obstructive hydrocephalus. To explore potential heterogeneity related to possible trends in the incidences of our outcomes that may accompany increased survival in patients over time, a meta-regression was conducted to assess the effects of the year of study publication and the maximum length of follow-up in the studies on the outcomes. Publication bias was assessed using the following 2 methods: a funnel plot plotting ORs against standard error and Begg and Mazumdar's and Egger's tests. An asymmetrical funnel plot or a P value meeting significance in the latter 2 tests indicated potential publication bias. Meta-regressions and statistical analyses for publication bias were conducted using The Comprehensive Meta-Analysis software version 2.2 (Biostat Inc, Englewood, New Jersey) for Windows. Statistical significance was identified with a 2-tailed P ≤ .05. Lastly, the quality of overall evidence was summarized with grades of low, medium, or high based on the 5 GRADE criteria:15 risk of bias, inconsistency (heterogeneity), indirectness (data representativeness), imprecision, and publication bias. Survival Analysis Survival data of patients with and without ventricular entry were presented heterogeneously in studies (ie, tabulated numerical data, Kaplan–Meier survival curves, and often without any survival effect size statistics). For a uniform analysis in the absence of published individual patient data, we digitized published Kaplan–Meier survival curves using WebPlotDigitizer (Version 3.12; June 2017; http://arohatgi.info/WebPlotDigitizer) to recreate individual patient data, an accepted methodology for meta-analysis of time-to-event data.16,17 The individual survival times obtained at censored and noncensored events and proportion of population surviving at those times were used to recreate published Kaplan–Meier survival curves. Accuracy of the reconstructed individual patient data was assessed by visually overlaying the reconstructed Kaplan–Meier curve with the original published curve. In all cases, near accurate curves were generated (refer to example in Figure, Supplemental Digital Content 2). All reconstructed and published individual patient survival data were pooled for Kaplan–Meier survival analysis. Log-rank (Mantel–Cox) text was used to compare survival of patients with ventricular entry to those without. Data are reported with a hazard ratio (HR) and 95% CI. Age, tumor grade, extent of resection, treatment with chemotherapy (including temozolomide), and radiation were noted for patients included in the survival analysis. Fisher's exact test was used to compare proportions between ventricular entry and no entry groups. Survival curve reconstructions, analyses, and Fisher's exact tests were performed using GraphPad Prism version 7.00 for Windows (GraphPad Software, San Diego, CA). Statistical significance was identified with a 2-tailed P ≤ 0.05. RESULTS Search Result and Study Characteristics Our search resulted in 6910 articles out of which 20 studies3,5,13,14,18-33 met our criteria for inclusion in the meta-analysis (Figure, Supplemental Digital Content 1). Specifically, 10 studies (References, Supplemental Digital Content 3) were excluded from the meta-analysis due to incomplete data. All were retrospective observational studies that included adult patients with high-grade glioma. A vast majority of the patients had glioblastoma. Overall, the case-controlled or cohort studies were assessed to be moderate in quality using the NOS (Table, Supplemental Digital Content 4). When specified, the lateral cerebral ventricles were most often the structures that were breached during resections. Association of Leptomeningeal Dissemination With Ventricular Entry We studied the association of leptomeningeal dissemination with ventricular entry during resections of high-grade gliomas. Leptomeningeal dissemination in all studies was diagnosed radiographically and defined as abnormal enhancement or nodular deposits in the cerebral ventricular wall and/or the subarachnoid spaces of the brain or spinal cord, except in 1 study28 where no clear definition was provided. Meta-analysis of the data revealed higher odds of developing leptomeningeal dissemination with a ventricular entry. The summary OR was calculated to be 3.91 [1.89-8.10] (P = .0002) from 9 studies (n = 86/410 vs 57/847 for patients with and without ventricular entry, respectively; Figure 1A). The funnel plot of the included studies (Figure 1B) was visually assessed to be mostly symmetric, implying no significant publication bias. Association of Hydrocephalus With Ventricular Entry Because hydrocephalus is commonly coexistent with leptomeningeal dissemination,3,6-9,24,25,34 we sought to reveal an association between ventricular entry during high-grade glioma resection and the development of hydrocephalus requiring CSF diversion. In studies where patients developed hydrocephalus (9 out of 11), insertion of a shunt was the most common form of treatment chosen (8 out of the 9 studies) for hydrocephalus, and specifically a ventriculoperitoneal shunt (specified in 5 out of the 9 studies). Meta-analysis of data from 11 studies demonstrated higher odds for developing hydrocephalus with a ventricular entry (summary OR of 7.78 [3.77-16.05]; P < .00001; n = 58/431 vs 11/565 for patients with and without ventricular entry, respectively; Figure 2A). The funnel plot of the included studies (Figure 2B) was visually assessed to be mostly symmetric, implying no significant publication bias. FIGURE 2. View largeDownload slide Meta-analysis of studies comparing A, incidences of hydrocephalus in patients with and without ventricular entry during their high-grade glioma resection with its B, respective funnel plot to assess publication bias. The lines in the forest plot of the meta-analysis represent 95% CI, and the size of its corresponding square marker reflects the relative weight of the study used in random-effects Mantel–Haenszel method. The diamond shows the overall effect. In the funnel plot, vertical line represents the summary OR with diagonal lines representing 95 % CIs. Each study is represented by a circle. a = ventricle entry was defined as ≥1 cm defect; b = data from initial surgeries with known ventricular breach status; c = judged as nonobstructive, external hydrocephalus; no treatment reported; d = data available for only frontal lobe malignant gliomas. FIGURE 2. View largeDownload slide Meta-analysis of studies comparing A, incidences of hydrocephalus in patients with and without ventricular entry during their high-grade glioma resection with its B, respective funnel plot to assess publication bias. The lines in the forest plot of the meta-analysis represent 95% CI, and the size of its corresponding square marker reflects the relative weight of the study used in random-effects Mantel–Haenszel method. The diamond shows the overall effect. In the funnel plot, vertical line represents the summary OR with diagonal lines representing 95 % CIs. Each study is represented by a circle. a = ventricle entry was defined as ≥1 cm defect; b = data from initial surgeries with known ventricular breach status; c = judged as nonobstructive, external hydrocephalus; no treatment reported; d = data available for only frontal lobe malignant gliomas. Heterogeneity, Sensitivity Analyses, and Bias Overall, no significant heterogeneity was noted in the 2 meta-analyses (Tau² ≤ 0.40; Chi² P value ≥ .11; I² ≤ 41%). Heterogeneity was further explored by examining time-dependent trends in the outcomes using meta-regression. Specifically, whether recently published studies reported lower incidences of the outcomes (perhaps reflecting adoption of more and better therapies), or if higher reported incidences of the outcomes trended directly with the maximum length of patient follow-up in the studies were examined. Meta-regressions of the year of study publication and the maximum length of follow-up in the studies with the ORs of the 2 outcomes (leptomeningeal dissemination and hydrocephalus) reported in the studies did not reveal any statistically meaningful relationships (Table, Supplemental Digital Content 5). The following sensitivity analyses conducted on the meta-analyses did not change the significance of the summary ORs: (1) sequential exclusion of 1 study at a time; (2) limiting the meta-analyses to only studies of patients likely treated in the temozolomide era; (3) excluding studies that had patients with recurrent GBM or WHO grade III glioma (Table, Supplemental Digital Content 6); and (4) excluding patients with documented obstructive hydrocephalus (there were only 2 such patients, 1 in each group, from Montano et al, 201128). Funnel plots generated were mostly visually symmetric (Figures 1B and 2B), and statistical analysis did not suggest publication bias in the meta-analyses (Table, Supplemental Digital Content 7). However, the included studies overall had a considerable risk of bias (Figure, Supplemental Digital Content 8). Effect of Ventricular Entry on Survival Although tumor dissemination and hydrocephalus are known to adversely impact survival, whether ventricular entry itself influences survival is debated. Some studies5,19,22,35,36 demonstrate no association, and others18,20,23,26 suggest its association with decreased survival. Therefore, we conducted a pooled survival analysis of published overall survival data for a total of 795 patients in 10 studies5,14,18-20,23,26,30,31,37 to glean the survival impact of ventricular entry. Kaplan–Meier survival analysis revealed a median survival time of 19.1 mo in 373 patients who did not have ventricular entry and 16.8 mo in 422 patients who had a ventricular entry (Figure 3). This survival difference was statistically significant (HR: 1.25 [1.05-1.48], P = .01). Analysis of the available data revealed that mean age and the proportions of patients who received gross total resection, subtotal (cytoreductive) resection, chemotherapy, including temozolomide, and radiation therapy did not significantly differ between these 2 groups (Table). Treatment information, however, was more often unavailable in the ventricular entry group. FIGURE 3. View largeDownload slide Kaplan–Meier overall survival curves of pooled survival data in patients reported in studies with and without ventricular entry during their high-grade glioma resection. HR represents log-rank hazard ratio. Tick marks below the curves represent censored values, representing alive patients or unknown living status. FIGURE 3. View largeDownload slide Kaplan–Meier overall survival curves of pooled survival data in patients reported in studies with and without ventricular entry during their high-grade glioma resection. HR represents log-rank hazard ratio. Tick marks below the curves represent censored values, representing alive patients or unknown living status. TABLE. Difference in the Commonly Reported Survival Predictors Between Patients With and Without Ventricular Entry During High-Grade Glioma Resection Ventricular entry (n = 422) No ventricular entry (n = 373) Fisher's exact test (P value) Age, mean, years, (total n)a 56.6 (n = 187) 57.1 (n = 186) a Tumor grade  WHO grade III (%) 19/286 9/270 .08  Unknownb (%) 136/422 103/373 .16 Extent of resection  GTR (%) 83/201 (41.3%) 90/209 (43.1%) .76  STR (%) 173/247 (70.0%) 159/249 (63.9%) .15  Unknown (%) 166/422 (39.3%) 124/373 (33.2%) .05 Chemotherapy  Temozolomide (%) 182/236 (77.1%) 193/249 (77.5%) 1.00  Any chemotherapy (%) 232/277 (83.8%) 227/270 (84.1%) 1.00  Unknown (%) 145/422 (34.4%) 103/373 (27.6%) .05 Radiation therapy  Any radiotherapy (%) 262/274 (95.6%) 260/270 (96.3%) .83  Unknown (%) 148/422 (35.1%) 103/373 (27.6%) .03 Ventricular entry (n = 422) No ventricular entry (n = 373) Fisher's exact test (P value) Age, mean, years, (total n)a 56.6 (n = 187) 57.1 (n = 186) a Tumor grade  WHO grade III (%) 19/286 9/270 .08  Unknownb (%) 136/422 103/373 .16 Extent of resection  GTR (%) 83/201 (41.3%) 90/209 (43.1%) .76  STR (%) 173/247 (70.0%) 159/249 (63.9%) .15  Unknown (%) 166/422 (39.3%) 124/373 (33.2%) .05 Chemotherapy  Temozolomide (%) 182/236 (77.1%) 193/249 (77.5%) 1.00  Any chemotherapy (%) 232/277 (83.8%) 227/270 (84.1%) 1.00  Unknown (%) 145/422 (34.4%) 103/373 (27.6%) .05 Radiation therapy  Any radiotherapy (%) 262/274 (95.6%) 260/270 (96.3%) .83  Unknown (%) 148/422 (35.1%) 103/373 (27.6%) .03 GTR, gross total resection; STR, subtotal (cytoreductive) resection; WHO, World Health Organization aMean age was calculated by summing the product of study means by their study size and dividing this sum by the sum of all sample sizes. Because data on variance were missing, a formal statistical t-test could not be performed for comparisons of the age means. bThe total number of WHO grade III patients is 54 among the 239 patients whose tumor grade distribution is unknown. View Large TABLE. Difference in the Commonly Reported Survival Predictors Between Patients With and Without Ventricular Entry During High-Grade Glioma Resection Ventricular entry (n = 422) No ventricular entry (n = 373) Fisher's exact test (P value) Age, mean, years, (total n)a 56.6 (n = 187) 57.1 (n = 186) a Tumor grade  WHO grade III (%) 19/286 9/270 .08  Unknownb (%) 136/422 103/373 .16 Extent of resection  GTR (%) 83/201 (41.3%) 90/209 (43.1%) .76  STR (%) 173/247 (70.0%) 159/249 (63.9%) .15  Unknown (%) 166/422 (39.3%) 124/373 (33.2%) .05 Chemotherapy  Temozolomide (%) 182/236 (77.1%) 193/249 (77.5%) 1.00  Any chemotherapy (%) 232/277 (83.8%) 227/270 (84.1%) 1.00  Unknown (%) 145/422 (34.4%) 103/373 (27.6%) .05 Radiation therapy  Any radiotherapy (%) 262/274 (95.6%) 260/270 (96.3%) .83  Unknown (%) 148/422 (35.1%) 103/373 (27.6%) .03 Ventricular entry (n = 422) No ventricular entry (n = 373) Fisher's exact test (P value) Age, mean, years, (total n)a 56.6 (n = 187) 57.1 (n = 186) a Tumor grade  WHO grade III (%) 19/286 9/270 .08  Unknownb (%) 136/422 103/373 .16 Extent of resection  GTR (%) 83/201 (41.3%) 90/209 (43.1%) .76  STR (%) 173/247 (70.0%) 159/249 (63.9%) .15  Unknown (%) 166/422 (39.3%) 124/373 (33.2%) .05 Chemotherapy  Temozolomide (%) 182/236 (77.1%) 193/249 (77.5%) 1.00  Any chemotherapy (%) 232/277 (83.8%) 227/270 (84.1%) 1.00  Unknown (%) 145/422 (34.4%) 103/373 (27.6%) .05 Radiation therapy  Any radiotherapy (%) 262/274 (95.6%) 260/270 (96.3%) .83  Unknown (%) 148/422 (35.1%) 103/373 (27.6%) .03 GTR, gross total resection; STR, subtotal (cytoreductive) resection; WHO, World Health Organization aMean age was calculated by summing the product of study means by their study size and dividing this sum by the sum of all sample sizes. Because data on variance were missing, a formal statistical t-test could not be performed for comparisons of the age means. bThe total number of WHO grade III patients is 54 among the 239 patients whose tumor grade distribution is unknown. View Large Quality of Evidence Quality of the data and results presented herein based on 5 GRADE criteria15 was assessed as follows: bias (low quality), inconsistency (high quality), indirectness (medium quality), imprecision (medium quality), and publication bias (high quality; Table, Supplemental Digital Content 9). Therefore, overall this work represents at best a “medium” level of quality of evidence. DISCUSSION A primary goal of the surgical treatment of high-grade glioma is maximal, safe resection of the tumor. In the process of achieving this, the ventricular system is not uncommonly entered. The consequences of ventricular entry are often not detectable clinically, and it is commonly assumed that entering the ventricles during high-grade glioma resection is not a source of major morbidity or mortality in these patients. Although few studies have reported a link between glioma dissemination and the development of hydrocephalus with ventricular breach during surgery, the low incidence of these events in postoperative course of high-grade glioma patients may have obscured a clear association. Here, in a quantitative meta-analysis, we reveal statistically compelling associations between them, which may have important clinical and biological implications. Hypotheses for Cancer Dissemination Following Ventricular Entry We demonstrate a strong association of leptomeningeal dissemination with ventricular entry. Our results support the prior proposed hypothesis4,5 that neoplastic cells spilled into the CSF in the ventricles may disseminate and seed elsewhere in the brain. Further biological investigations are warranted to test this hypothesis and to understand whether glioma cells (1) survive in the circulating CSF, (2) alter CSF fluid dynamics, and (3) seed in the leptomeninges and ventricular linings. The interface between the ventricular system and the brain contains the biologically relevant subventricular zone. It is a large, prominent neural stem cell zone that lies throughout the entire lateral walls of the lateral ventricles,38,39 and therefore, it is often violated by surgical entry into the ventricles. Both clinical40,41 and biological42,43 studies have shown that glioma cells preferentially migrate towards this growth factor rich “lake-front property.”44 Whether surgically disseminated glioma cells in the CSF display the same affinity or tropism for the subventricular zone is relevant to understanding the leptomeningeal dissemination that occurs after ventricular entry. Further, subventricular zone tumor contact may confound this association as any residual glioma cells after resection could continue to spread along the subventricular zone. Hypotheses for Hydrocephalus Development and Its Impact Communicating hydrocephalus is often noted in the presence of leptomeningeal dissemination.3,6-9,24,25,34 If ventricular entry is associated with the development of leptomeningeal dissemination, then by logical extension, we tested and found that it was also associated with the development of hydrocephalus. Hydrocephalus arises from increased pressure in the CSF often either from its increased production (by cells of the choroid plexus) or decreased absorption (by arachnoid granulations). The biological basis for the alteration in CSF dynamics following ventricular breach during high-grade glioma resection is unclear. Some have suggested the cause to be malabsorption by arachnoid granulations due to increased protein content and its precipitation in the CSF that accompanies the presence of tumor cells28,32,45 or surgical debris. Prevention of hydrocephalus is important not only from a survival perspective, but also in ensuring the best quality of life for patients, by avoiding additional procedures and delays in starting cancer therapy.28,32,45,46 Minimal Effect of Ventricular Entry on Overall Survival The pooled survival analysis of the available published overall survival data demonstrated decreased survival of high-grade glioma patients with ventricular entry during surgery. Before reaching to clinical implications of these results, we draw attention to a few considerations. First, although we demonstrated nonsignificant differences in age and treatment between patients with and without ventricular entry, the survival analysis is unadjusted due to the lack of individual patient data for their effects and other important prognostic variables (ie, patient performance status, MGMT promoter methylation or IDH1/2 mutation status of the gliomas). Compared to the survival effect sizes of these known predictors, the survival effect of ventricular entry, though significant, is smaller (median difference of 2.3 mo) and may not be a prominent contributor in predicting patient survival in a multivariable, adjusted analysis. Second, the primary cause of death from high-grade glioma is treatment resistance that manifests primarily as local recurrence.47-53 Glioma dissemination and hydrocephalus are uncommon. Therefore, presently, we believe the best surgical treatment is maximal and safe cytoreduction of the glioma that is currently supported by a significant body of scientific literature.54 The consequences of ventricular entry during surgery should not alter this primary goal. Rather, patients should be counseled, and surgeons and treating oncologists must be aware of the possibilities of tumor dissemination and hydrocephalus, so that they can be managed in time. If gross total resections can be achieved without ventricular entry, the pursuit of such approaches is reasonable. Limitations Despite adjustments with sensitivity analyses, assessments of heterogeneity and risk of bias, this meta-analysis is limited by 1) the inherent bias present in all studies related to their retrospective observational designs that may overestimate the effect sizes derived and 2) possible variations in ventricular entry and outcome assessments during the time-frame of the studies included. Some of these studies were not designed to explore the link between ventricular breach and our outcomes of interest. Further, the population of patients in these studies may be subject to reporting bias and a confirmatory prospective analysis is indicated to support the validity of the findings in this manuscript. Lastly, we recognize that time-to-event analysis with individual patient data adjusting for potential covariates, such as and importantly ventricular or subventricular zone contact—not possible in this study design—as necessary to reveal any time-dependent patterns in the occurrence of our outcomes, including overall survival (as discussed above), and to understand the contributions of the covariates in predicting these outcomes. CONCLUSION We uncover a strong association between cerebral ventricular entry that occurs during high-grade glioma resections and leptomeningeal dissemination, hydrocephalus, and overall survival. The quality of the data and results supporting this association are judged to be medium at best. This association encourages further investigations to unveil any confounding or causal links among these factors. Our results should not deter neurosurgeons from achieving a maximum possible glioma resection as supported by the current evidence54 to avoid ventricular entry. Treating physicians should become aware of and recognize the consequences of ventricular entry for their timely management. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Pietschmann S , von Bueren AO , Kerber MJ , Baumert BG , Kortmann RD , Muller K . An individual patient data meta-analysis on characteristics, treatments and outcomes of glioblastoma/gliosarcoma patients with metastases outside of the central nervous system . PLoS One . 2015 ; 10 ( 4 ): e0121592 . Google Scholar CrossRef Search ADS PubMed 2. Pietschmann S , von Bueren AO , Henke G , Kerber MJ , Kortmann RD , Muller K . An individual patient data meta-analysis on characteristics, treatments and outcomes of the glioblastoma/gliosarcoma patients with central nervous system metastases reported in literature until 2013 . J Neurooncol . 2014 ; 120 ( 3 ): 451 - 457 . Google Scholar CrossRef Search ADS PubMed 3. Mandel JJ , Yust-Katz S , Cachia D et al. Leptomeningeal dissemination in glioblastoma; an inspection of risk factors, treatment, and outcomes at a single institution . J Neurooncol . 2014 ; 120 ( 3 ): 597 - 605 . Google Scholar CrossRef Search ADS PubMed 4. Grabb PA , Albright AL , Pang D . Dissemination of supratentorial malignant gliomas via the cerebrospinal fluid in children . Neurosurgery . 1992 ; 30 ( 1 ): 64 - 71 . Google Scholar CrossRef Search ADS PubMed 5. Elliott JP , Keles GE , Waite M , Temkin N , Berger MS . Ventricular entry during resection of malignant gliomas: effect on intracranial cerebrospinal fluid tumor dissemination . J Neurosurg . 1994 ; 80 ( 5 ): 834 - 839 . Google Scholar CrossRef Search ADS PubMed 6. Awad I , Bay JW , Rogers L . Leptomeningeal metastasis from supratentorial malignant gliomas . Neurosurgery . 1986 ; 19 ( 2 ): 247 - 251 . Google Scholar CrossRef Search ADS PubMed 7. Delattre JY , Walker RW , Rosenblum MK . Leptomeningeal gliomatosis with spinal cord or cauda equina compression: a complication of supratentorial gliomas in adults . Acta Neurol Scand . 1989 ; 79 ( 2 ): 133 - 139 . Google Scholar CrossRef Search ADS PubMed 8. Onda K , Tanaka R , Takahashi H , Takeda N , Ikuta F . Symptomatic cerebrospinal fluid dissemination of cerebral glioblastoma . Neuroradiology . 1990 ; 32 ( 2 ): 146 - 150 . Google Scholar CrossRef Search ADS PubMed 9. Ito U , Tomita H , Yamazaki S , Takada Y , Inaba Y . CT findings of leptomeningeal and periventricular dissemination of tumors . Clin Neurol Neurosurg . 1986 ; 88 ( 2 ): 115 - 120 . Google Scholar CrossRef Search ADS PubMed 10. Moher D , Liberati A , Tetzlaff J , Altman DG , Group P . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement . PLoS Med . 2009 ; 6 ( 7 ): e1000097 . Google Scholar CrossRef Search ADS PubMed 11. Wells GA , Shea B , O’Connell D et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses . Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed September 13, 2017 . 12. Higgins JPT , Green S . Cochrane Handbook for Systematic Reviews of Interventions . Vol 4 . Hoboken, NJ : Wiley ; 2011 . 13. John JK , Robin AM , Pabaney AH et al. Complications of ventricular entry during craniotomy for brain tumor resection . J Neurosurg . 2017 ; 127 ( 2 ): 426 - 432 . Google Scholar CrossRef Search ADS PubMed 14. Chan DT , Hsieh SY , Kam MK , Cheung TC , Ng SC , Poon WS . Pattern of recurrence and factors associated with cerebrospinal fluid dissemination of glioblastoma in Chinese patients . Surg Neurol Int . 2016 ; 7 ( 1 ): 92 . Google Scholar CrossRef Search ADS PubMed 15. Ryan R , Hill S . The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed January 11, 2018 . 16. Wan X , Peng L , Li Y . A review and comparison of methods for recreating individual patient data from published Kaplan-Meier survival curves for economic evaluations: a simulation study . PLoS One . 2015 ; 10 ( 3 ): e0121353 . Google Scholar CrossRef Search ADS PubMed 17. Mistry AM , Hale AT , Chambless LB , Weaver KD , Thompson RC , Ihrie RA . Influence of glioblastoma contact with the lateral ventricle on survival: a meta-analysis . J Neurooncol . 2017 ; 131 ( 1 ): 125 - 133 . Google Scholar CrossRef Search ADS PubMed 18. Sonoda Y , Shibahara I , Matsuda KI et al. Opening the ventricle during surgery diminishes survival among patients with newly diagnosed glioblastoma treated with carmustine wafers: a multi-center retrospective study . J Neurooncol . 2017 ; 134 ( 1 ): 83 - 88 . Google Scholar CrossRef Search ADS PubMed 19. Kondo N , Barth RF , Miyatake SI et al. Cerebrospinal fluid dissemination of high-grade gliomas following boron neutron capture therapy occurs more frequently in the small cell subtype of IDH1R132H mutation-negative glioblastoma . J Neurooncol . 2017 ; 133 ( 1 ): 107 - 118 . Google Scholar CrossRef Search ADS PubMed 20. Behling F , Kaltenstadler M , Noell S et al. The prognostic impact of ventricular opening in glioblastoma surgery: A retrospective single center analysis . World Neurosurg . 2017 ; 106 : 615 - 624 . Google Scholar CrossRef Search ADS PubMed 21. Hasegawa Y , Iuchi T , Sakaida T , Yokoi S , Kawasaki K . The influence of carmustine wafer implantation on tumor bed cysts and peritumoral brain edema . J Clin Neurosci . 2016 ; 31 : 67 - 71 . Google Scholar CrossRef Search ADS PubMed 22. Adeberg S , Diehl C , Jung CS et al. Is a modification of the radiotherapeutic target volume necessary after resection of glioblastomas with opening of the ventricles? J Neurooncol . 2016 ; 127 ( 3 ): 581 - 587 . Google Scholar CrossRef Search ADS PubMed 23. Roelz R , Reinacher P , Jabbarli R et al. Surgical ventricular entry is a key risk factor for leptomeningeal metastasis of high grade gliomas . Sci Rep . 2016 ; 5 ( 1 ): 17758 . Google Scholar CrossRef Search ADS 24. Fischer CM , Neidert MC , Peus D et al. Hydrocephalus after resection and adjuvant radiochemotherapy in patients with glioblastoma . Clin Neurol Neurosurg . 2014 ; 120 : 27 - 31 . Google Scholar CrossRef Search ADS PubMed 25. Onuma K , Ishikawa E , Matsuda M et al. Clinical characteristics and neuroimaging findings in 12 cases of recurrent glioblastoma with communicating hydrocephalus . Neurol Med Chir (Tokyo) . 2013 ; 53 ( 7 ): 474 - 481 . Google Scholar CrossRef Search ADS PubMed 26. Tejada-Solis S , Aldave-Orzaiz G , Pay-Valverde E , Marigil-Sanchez M , Idoate-Gastearena MA , Diez-Valle R . Prognostic value of ventricular wall fluorescence during 5-aminolevulinic-guided surgery for glioblastoma . Acta Neurochir . 2012 ; 154 ( 11 ): 1997 - 2002 ; discussion 2002 . Google Scholar CrossRef Search ADS PubMed 27. Konishi Y , Muragaki Y , Iseki H , Mitsuhashi N , Okada Y . Patterns of intracranial glioblastoma recurrence after aggressive surgical resection and adjuvant management: retrospective analysis of 43 cases . Neurol Med Chir (Tokyo) . 2012 ; 52 ( 8 ): 577 - 586 . Google Scholar CrossRef Search ADS PubMed 28. Montano N , D’Alessandris QG , Bianchi F et al. Communicating hydrocephalus following surgery and adjuvant radiochemotherapy for glioblastoma . J Neurosurg . 2011 ; 115 ( 6 ): 1126 - 1130 . Google Scholar CrossRef Search ADS PubMed 29. Della Puppa A , Rossetto M , Ciccarino P et al. Carmustine wafer implantation when surgical cavity is communicating with cerebral ventricles: technical considerations on a clinical series . World Neurosurg . 2011 ; 76 ( 1-2 ): 156 - 159 ; discussion 167-158 . Google Scholar CrossRef Search ADS PubMed 30. Bock HC , Cohnen J , Keric N , Kantelhardt SR , Giese A . Occlusion of surgical opening of the ventricular system with fibrinogen-coated collagen fleece: a case collection study . Acta Neurochir . 2011 ; 153 ( 3 ): 533 - 539 . Google Scholar CrossRef Search ADS PubMed 31. Bae JS , Yang SH , Yoon WS , Kang SG , Hong YK , Jeun SS . The clinical features of spinal leptomeningeal dissemination from malignant gliomas . J Korean Neurosurg Soc . 2011 ; 49 ( 6 ): 334 - 338 . Google Scholar CrossRef Search ADS PubMed 32. Marquardt G , Setzer M , Lang J , Seifert V . Delayed hydrocephalus after resection of supratentorial malignant gliomas . Acta Neurochir (Wien) . 2002 ; 144 ( 3 ): 227 - 231 ; discussion 231 . Google Scholar CrossRef Search ADS PubMed 33. Buhl R , Barth H , Hugo HH , Hutzelmann A , Mehdorn HM . Spinal drop metastases in recurrent glioblastoma multiforme . Acta Neurochir (Wien) . 1998 ; 140 ( 10 ): 1001 - 1005 . Google Scholar CrossRef Search ADS PubMed 34. Arita N , Taneda M , Hayakawa T . Leptomeningeal dissemination of malignant gliomas. Incidence, diagnosis and outcome . Acta Neurochir . 1994 ; 126 ( 2-4 ): 84 - 92 . Google Scholar CrossRef Search ADS PubMed 35. Adeberg S , Konig L , Bostel T et al. Glioblastoma recurrence patterns after radiation therapy with regard to the subventricular zone . Int J Radiat Oncol Biol Phys . 2014 ; 90 ( 4 ): 886 - 893 . Google Scholar CrossRef Search ADS PubMed 36. Shibahara I , Sonoda Y , Saito R et al. The expression status of CD133 is associated with the pattern and timing of primary glioblastoma recurrence . Neuro-oncol . 2013 ; 15 ( 9 ): 1151 - 1159 . Google Scholar CrossRef Search ADS PubMed 37. Saito R , Kumabe T , Kanamori M , Sonoda Y , Tominaga T . Distant recurrences limit the survival of patients with thalamic high-grade gliomas after successful resection . Neurosurg Rev . 2017 ; 40 ( 3 ): 469 - 477 . Google Scholar CrossRef Search ADS PubMed 38. Sanai N , Tramontin AD , Quinones-Hinojosa A et al. Unique astrocyte ribbon in adult human brain contains neural stem cells but lacks chain migration . Nature . 2004 ; 427 ( 6976 ): 740 - 744 . Google Scholar CrossRef Search ADS PubMed 39. Sanai N , Nguyen T , Ihrie RA et al. Corridors of migrating neurons in the human brain and their decline during infancy . Nature . 2011 ; 478 ( 7369 ): 382 - 386 . Google Scholar CrossRef Search ADS PubMed 40. Mistry AM , Dewan MC , White-Dzuro GA et al. Decreased survival in glioblastomas is specific to contact with the ventricular-subventricular zone, not subgranular zone or corpus callosum . J Neurooncol . 2017 ; 132 ( 2 ): 341 - 349 . Google Scholar CrossRef Search ADS PubMed 41. Parsa AT , Wachhorst S , Lamborn KR et al. Prognostic significance of intracranial dissemination of glioblastoma multiforme in adults . J Neurosurg . 2005 ; 102 ( 4 ): 622 - 628 . Google Scholar CrossRef Search ADS PubMed 42. Qin EY , Cooper DD , Abbott KL et al. Neural precursor-derived pleiotrophin mediates subventricular zone invasion by glioma . Cell . 2017 ; 170 ( 5 ): 845 - 859.e19 . Google Scholar CrossRef Search ADS PubMed 43. Goffart N , Kroonen J , Di Valentin E et al. Adult mouse subventricular zones stimulate glioblastoma stem cells specific invasion through CXCL12/CXCR4 signaling . Neuro-oncol . 2015 ; 17 ( 1 ): 81 - 94 . Google Scholar CrossRef Search ADS PubMed 44. Ihrie RA , Alvarez-Buylla A . Lake-front property: a unique germinal niche by the lateral ventricles of the adult brain . Neuron . 2011 ; 70 ( 4 ): 674 - 686 . Google Scholar CrossRef Search ADS PubMed 45. Inamasu J , Nakamura Y , Saito R et al. Postoperative communicating hydrocephalus in patients with supratentorial malignant glioma . Clin Neurol Neurosurg . 2003 ; 106 ( 1 ): 9 - 15 . Google Scholar CrossRef Search ADS PubMed 46. Esquenazi Y , Moussazadeh N , Link TW et al. Thalamic glioblastoma: clinical presentation, management strategies, and outcomes . Neurosurgery . 2017 . doi: 10.1093/neuros/nyx349. [Published online ahead of print ]. 47. Brandes AA , Tosoni A , Franceschi E et al. Recurrence pattern after temozolomide concomitant with and adjuvant to radiotherapy in newly diagnosed patients with glioblastoma: correlation with MGMT promoter methylation status . J Clin Oncol . 2009 ; 27 ( 8 ): 1275 - 1279 . Google Scholar CrossRef Search ADS PubMed 48. Rapp M , Baernreuther J , Turowski B , Steiger HJ , Sabel M , Kamp MA . Recurrence pattern analysis of primary glioblastoma . World Neurosurg . 2017 ; 103 : 733 - 740 . Google Scholar CrossRef Search ADS PubMed 49. Gebhardt BJ , Dobelbower MC , Ennis WH , Bag AK , Markert JM , Fiveash JB . Patterns of failure for glioblastoma multiforme following limited-margin radiation and concurrent temozolomide . Radiat Oncol . 2014 ; 9 ( 1) : 130 . Google Scholar CrossRef Search ADS PubMed 50. Sherriff J , Tamangani J , Senthil L et al. Patterns of relapse in glioblastoma multiforme following concomitant chemoradiotherapy with temozolomide . Br J Radiol . 2013 ; 86 ( 1022 ): 20120414 . Google Scholar CrossRef Search ADS PubMed 51. Milano MT , Okunieff P , Donatello RS et al. Patterns and timing of recurrence after temozolomide-based chemoradiation for glioblastoma . Int J Radiat Oncol Biol Phys . 2010 ; 78 ( 4 ): 1147 - 1155 . Google Scholar CrossRef Search ADS PubMed 52. Chamberlain MC . Radiographic patterns of relapse in glioblastoma . J Neurooncol . 2011 ; 101 ( 2 ): 319 - 323 . Google Scholar CrossRef Search ADS PubMed 53. Zhou X , Liao X , Zhang B et al. Recurrence patterns in patients with high-grade glioma following temozolomide-based chemoradiotherapy . Mol Clin Oncol . 2016 ; 5 ( 2 ): 289 - 294 . Google Scholar CrossRef Search ADS PubMed 54. Brown TJ , Brennan MC , Li M et al. Association of the extent of resection with survival in glioblastoma . JAMA Oncol . 2016 ; 2 ( 11 ): 1460 - 1469 . Google Scholar CrossRef Search ADS PubMed Supplemental digital content is available for this article at www.neurosurgery-online.com. Supplemental Digital Content 1. Figure. PRISMA Flow Diagram. PRISMA Flow Diagram represents the process of searching, identifying, and screening published studies for inclusion in the meta-analysis. The template utilized is from Moher et al.10 Supplemental Digital Content 2. Figure. An example comparing published and reconstructed Kaplan–Meier survival curves after digitization. The original Kaplan–Meier survival curve is reprinted by permission from Springer Science + Business Media New York: Springer Nature, Journal of Neuro-Oncology18, Opening the ventricle during surgery diminishes survival among patients with newly diagnosed glioblastoma treated with carmustine wafers: a multi-center retrospective study, Sonoda Y et al, © 2017. Supplemental Digital Content 3. References. Excluded Studies. Studies with incomplete data for inclusion in meta-analysis. Supplemental Digital Content 4. Table. Newcastle-Ottawa Assessment. Quality assessment of studies included in the meta-analyses using the Newcastle-Ottawa Assessment Scale. Supplemental Digital Content 5. Table. Meta-Regression Analyses. Results of meta-regressions of the year of study publication and the maximum length of study follow-up in the studies with the OR of the 3 outcomes studies in the 3 meta-analyses. Supplemental Digital Content 6. Table. Recurrent high-grade and WHO Grade III gliomas. List of studies with the number and proportions of patients with recurrent high-grade gliomas and WHO Grade III gliomas. Supplemental Digital Content 7. Table. Publication bias. Visual and statistical funnel plot analyses to assess publication bias. Supplemental Digital Content 8. Figure. Analyses of risk of bias in studies included in meta-analyses. The risk of several types of biases in each study included in the meta-analyses was assessed by denoting a red circle with a minus sign for high risk and green circle with plus sign for low risk. The risk of bias is assessed based on studies’ intended outcomes of interest. Supplemental Digital Content 9. Table. Quality of Evidence Assessment. Each of the 5 “GRADE” criteriaa is graded on a low, medium, high quality. COMMENTS The authors report a thoughtfully conducted meta-analysis with purposeful sensitivity analyses aiming to better understand the impact of entering the ventricular system during high-grade glioma surgery. Building upon recent case series, this is a timely study conducted with sound methodological principles and with an honest assessment of limitations, including lack of information on MGMT and IDH1 status. Importantly, the authors report rates of gross total resection did not significantly differ by ventricular entry, although distal parenchymal recurrence, leptomeningeal spread, hydrocephalus, overall survival were worse in patients with ventricular entry. This study reminds us that tumor spread and hydrocephalus may be unintended sequela of entering the ventricular system during surgical resection. Entry into the ventricular system may additionally impact visual fields, particular around the lateral aspect of the temporal horns, although this particular outcome was not catalogued in this study. With increasing interest in performing “supra-total” resection for high-grade gliomas, the findings of this meta-analysis provide an important counter-balance, causing the neurosurgeon to think carefully about the consequences of more aggressive surgical resection. We have increasingly relied on image-guidance systems, intraoperative neuromonitoring including cortical and sub-cortical stimulation, and awake craniotomy techniques to maximize extent of resection while minimizing morbidity. As our field continues to use these advanced technologies to push the boundaries of safe surgical resection for high-grade gliomas, we need to be aware of our surgical technique in the periventricular zone, minimize inadvertent entry into the ventricular system, and thoughtfully counsel our patients when such an occurrence is expected. Debraj Mukherjee Pittsburgh, Pennsylvania The authors address a not uncommon surgical circumstance of ventricular entry during the resection of malignant gliomas as reported in the literature in a retrospective fashion. They have worked to assess survival, hydrocephalus, and leptomeningeal spread via descriptive statistics and attempted to extract more sweeping conclusions via meta-analysis. Limitations to the work are the retrospective analysis of a very unevenly reported event in the literature leaving the data they have subject to a number of biases. They do come up with a suggestion that ventricular entry or subventricular zone contact results in more leptomeningeal dissemination and hydrocephalus, and lessened overall survival. Given the limits in the quality of the data this meta-analysis is based upon, the results do not imply a need to change in techniques used for resection of high-grade gliomas. Ideally these results will serve as a bench mark for the knowledge on this topic and spur a prospective analysis of a surgical series that contains a more homogeneous reporting mechanism regarding ventricular entry so as to preclude the biases encountered in this study. Jeffrey J. Olson Atlanta, Georgia Neurosurgery Speaks (Audio Abstracts) Listen to audio translations of this paper's abstract into select languages by choosing from one of the selections below. Chinese: Liang Chen, MD. Department of Neurosurgery Huashan Hospital Shanghai, China Chinese: Liang Chen, MD. Department of Neurosurgery Huashan Hospital Shanghai, China Close French: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France French: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France Close English: William W. Ashley, MD, PhD, MBA. Department of Neurological Surgery Sinai Hospital and LifeBridge Health System Baltimore, Maryland English: William W. Ashley, MD, PhD, MBA. Department of Neurological Surgery Sinai Hospital and LifeBridge Health System Baltimore, Maryland Close Russian: Natalia Denisova, MD. Novosibirsk Federal Centre of Neurosurgery Novosibirsk, Russia Russian: Natalia Denisova, MD. Novosibirsk Federal Centre of Neurosurgery Novosibirsk, Russia Close Italian: Daniele Bongetta, MD. Department of Neurosurgery Fondazione IRCCS Policlinico San Matteo Pavia, Italy Italian: Daniele Bongetta, MD. Department of Neurosurgery Fondazione IRCCS Policlinico San Matteo Pavia, Italy Close Spanish: Alvaro Campero, MD. Department of Neurosurgery Hospital Zenón Santillán San Miguel de Tucumán Tucumán, Argentina Spanish: Alvaro Campero, MD. Department of Neurosurgery Hospital Zenón Santillán San Miguel de Tucumán Tucumán, Argentina Close Portuguese: Hugo Leonardo Doria-Netto. Department of Micro-Neurosurgery CNC-Centro de Neurociasncia Sao Paulo, Brazil Portuguese: Hugo Leonardo Doria-Netto. Department of Micro-Neurosurgery CNC-Centro de Neurociasncia Sao Paulo, Brazil Close Japanese: Soichi Oya, MD, PhD. Department of Neurosurgery Saitama Medical Center/University Saitama, Japan Japanese: Soichi Oya, MD, PhD. Department of Neurosurgery Saitama Medical Center/University Saitama, Japan Close Korean: Sun Ha Paek, MD, PhD Department of Neurosurgery Seoul National University College of Medicine Seoul, Republic of Korea Korean: Sun Ha Paek, MD, PhD Department of Neurosurgery Seoul National University College of Medicine Seoul, Republic of Korea Close Greek: George Georgoulis, MD Department of Neurosurgery University Hospital of Ioannina Ioannina, Greece Greek: George Georgoulis, MD Department of Neurosurgery University Hospital of Ioannina Ioannina, Greece Close Copyright © 2018 by the Congress of Neurological Surgeons This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

NeurosurgeryOxford University Press

Published: May 22, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off