Preoperative Third Ventricle Floor Bowing is Associated with Increased Surgical Success Rate in Patients Undergoing Endoscopic Third Ventriculostomy – A Systematic Review and Meta-analysis
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.344680
Source of Support: None, Conflict of Interest: None
Keywords: Cerebrospinal fluid, endoscopic third ventriculostomy, ETVSS, hydrocephalus, neuroradiology, third ventricular floor bowing
Endoscopic third ventriculostomy (ETV) is a procedure that involves devising an opening in the third ventricle floor, allowing cerebrospinal fluid (CSF) to flow into the prepontine cistern and the subarachnoid space. The concept of this procedure has been formed in the early to mid-20th century, interest has resurfaced due to advancement in fiberoptic and lens technology allowing a smaller neuroendoscope with added features and superior optic resolution. ETV is generally ineffective in patients with communicating hydrocephalus, and out of all patients who underwent ETV 20% still require shunting., Principally, ETV relies on CSF absorption within the neuraxis, and subsequently, the classification of hydrocephalus, as well as the distinction between intraventricular and interventricular obstruction, becomes a crucial part in deciding the need for ETV. This is shown by much higher rates of ETV success in intraventricular obstructive cases, however, ETV is known to help cases of extraventricular hydrocephalus.
A scoring system exists to predict the early success of ETV, ETV Success Score (ETVSS). This score was developed and validated using a dataset of 618 consecutive ETV procedures performed at 12 international institutions. This scoring system covers age, etiology, and history of previous shunting. However, there is still a room for improvement to this scoring because a theoretically ideal candidate may suffer surgical failure. Furthermore, the cause of hydrocephalus is often difficult to assess, due to the coexisting combination of intraventricular and extraventricular aspect of the obstruction, therefore limiting the usefulness of the factors that are weighed in the scoring system. As ETV is effective in treating intraventricular obstruction, Third ventricular floor bowing (TVFB) serves as an indicator of intraventricular obstruction in hydrocephalus and existence of pressure gradient across third ventricular floor, which is the prerequisite of a successful ETV. Patients often present without apparent obstruction but actually has an obstructive component, TVFB is postulated to be able to identify patients that will benefit from ETV. In this systematic review and meta-analysis, we aimed to synthesize the latest evidence on the TVFB as a marker of surgical success in patients undergoing ETV.
A systematic literature search was performed on topics that assesses the association of TVFB with the surgical success in patients undergoing ETV with keywords [“ventricle floor bowing”] and their synonyms from beginning of time until December 2019 by using PubMed, Cochrane Central Database, EuropePMC, ProQuest, ScienceDirect, and ClinicalTrials.gov. Inclusion and exclusion criteria were then applied to the records. Initial search was performed by two independent researchers, if discrepancies arose, would be resolved by discussion. The literature search is outlined in [Figure 1].
All studies that assess the association of TVFB with surgical success in patients undergoing ETV are included in the study. Clinical researches/original articles were included and animal studies, case reports, review articles were excluded.
Two independent authors performed data extraction from the included studies followed by quality assessment. A standardized extraction form was used to ensure uniform data extraction. The data of interest were authors of the study, the year of publication, study design, sample size, subject characteristics, the proportion of TVFB (+), corresponding success rate, and follow-up length.
RevMan v5.3 (Cochrane Collaboration) and STATA v16.0 were used to perform the statistical analysis. Odds ratio (OR) and its 95% confidence interval generated using Mantel–Haenzsel formula were used for measuring dichotomous data. Inconsistency index (I2) test was used for assessing heterogeneity between the included studies, the value ranges from 0% to 100%, and an I2 >50% or P < 0.10 indicates statistically significant heterogeneity. A fixed-effect model was used for meta-analysis, and a random-effect model was used in case of heterogeneity if appropriate. Subgroup analysis was performed for pediatric patients. P values were two-tailed with a statistical significance set at <0.05 except for heterogeneity.
Initial search leads to 1433 potential articles, in which 1151 records were obtained after removal of duplicates. We exclude 1142 articles after titles and abstracts screening. Full-text of three studies were then explored and three studies were excluded because 1) Quantitative measurement of third ventricular floor deformity (n = 1), 2) Did not assess the predictive power for third ventricular floor bowing (n = 2). A total of six cohort studies were included for qualitative synthesis and meta-analysis [Figure 1], [Table 1]. There was a total of 568 subjects from six studies.,,,,,
Characteristics of the included studies
The definition of successful surgery in the studies was the resolution of symptoms and required no further procedures for hydrocephalus/CSF diversion procedure. There were two studies that enroll patients ≤16 years only. The remaining four studies enroll adults and children.
Third ventricle floor bowing and surgical success rate
TVFB (+) was associated with 85% (81-88%) ETV success [Figure 2]a. Compared to TVFB (-), TVFB (+) was associated with OR 4.13 [2.59, 6.60], P < 0.001; I2: 6%, P = 0.37 for ETV success [Figure 2]b. Subgroup analysis on pediatric patients showed 86% (82-91%) success rate.
In terms of value for ETV success compared to ETVSS, a high ETVSS does not significantly differ (P = 0.31) from TVFB (+) [Figure 3]a and TVFB (+) was associated with OR 3.14 [1.72, 5.73], P < 0.001; I2: 69%, P = 0.07 [Figure 3]b compared to intermediate/moderate ETVSS.
There were three studies by Foroughi et al., Krejci et al., and Kehler et al. that reported resolution of TVFB after surgery. Foroughi et al. showed that TVFB resolution occurred in 33/36 in the successful surgery group and 3/4 in the failed surgery group. Krejci et al. reported resolution of bowing 48/50 in successful surgery group and 1/5 in failure. Meanwhile, Kehler et al. showed that in general, the third ventricular floor reverted into normal position, however, the number of patients was not reported.
Funnel plot analysis showed an asymmetrical funnel plot [Figure 4]a due to the presence of an outlier (Foroughi et al.). Sensitivity analysis was performed by removing Foroughi et al. study, the OR was 3.62 [2.22, 5.89], P < 0.001; I2: 0%, P = 0.88 for successful surgery in TVFB (+). Regression-based Harbord's test showed no statistically significant presence of small-study effects (P = 0.396). [Figure 4]b
TVFB was associated with an increased rate of successful surgery in patients undergoing ETV. However, it's prognostic performance did not surpass EVTSS score with current evidence. There is also no statistically significant difference between TVFB (+) and TVFB (-) in patients <6 months old.
ETV regains interest and becoming more feasible due to the modernization of neuroendoscopic procedures. It provides the possibility of shunt free procedure, allowing the patients to avoid shunt-related complications. Theoretically, ETV is effective for obstructive hydrocephalus, not in that of communicating hydrocephalus due to suboptimal CSF absorption or immature absorption capacity. TVFB portend a significant pressure gradient between the prepontine and third ventricle system, which may be caused by intraventricular or intracisternal obstruction, such pressure is required to maintain ETV patency., Obstruction distal to the prepontine system was thought to not significantly increase the pressure gradient along the floor of third ventricle which may signify lesion unfavorable for ETV., Hence, TVFB is a favorable sign linked ETV success. Our meta-analysis has shown that TVFB is associated with ETV success across adults and children. However, the presence of TVFB is not shown to predict ETV success in patients < 6 months old according to a study. It was stated that the ineffectiveness was due to immature absorption capacity, cranial elasticity, and low-pressure gradient, along with the possibility of spontaneous closure of the stoma.,,, ETV in infants < 6 months old has been shown to have a low success rate of only 20.8%, which rises to 55% in patients aged 6-12 months old, indicating immature absorption capability and spontaneous closure of stoma hinders the effectiveness of ETV. Premature infants have poorer result compared to full term normal birth weight infants. However, information on TVFB performance in prognostication in patients aged <6 months old can only be drawn from one study, because the remaining studies did not perform subgroup analysis on patients <6 months old. Further investigations are required before drawing a conclusion.
ETVSS is a 6 months predictive model developed in 2011 to assess the prognosis of patients undergoing ETV. Our meta-analysis indicates that there is no statistically significant difference between the surgical success rate in TFVB (+) compared to high ETVSS. There are radiological findings that were associated with failure in patients with high ETVSS, these findings are not assessed by ETVSS. This suggests that there is a room for improvement to the ETVSS, TFVB is one of the simple radiological findings that may be used as an adjunct to ETVSS. We encourage further investigation to integrate this radiological finding into a prediction model or as an adjunct to ETVSS.
Although the results of earlier studies found that ETV to be generally ineffective for communicating hydrocephalus, results of newer studies have found ETV to be useful even in patients with communicating hydrocephalus.,, In this context, TVFB serves as an indice for an obstruction within the ventricular system, which will then be used as grounds to perform ETV. In conclusion, the dichotomy between communicating and noncommunicating hydrocephalus in deciding whether or not to perform ETV is likely to be harmful, because ETV has been shown not only to be effective generally in noncommunicating hydrocephalus but also in communicating hydrocephalus, suggesting that arachnoidal absorption dysfunction is not the sole cause of obstruction in this subset of condition. Furthermore, the results of the study by Tomas et al., showed that the preoperative absence of bowing does not automatically indicate ETV failure, with 47.6% bowing negative patients in the study successfully undergoing ETV. This might be explained by the fact that bowing itself is influenced by varying degrees of cerebral compliance. Finally, safety and effectiveness of the procedure may be dependent on anatomic variants which require tailored ventriculostomy approach in order to obtain best result.
The limitation of this systematic review includes publication bias in which a positive study is likely to be reported and published in the body of literature. The sample size is small in comparison for TFVB vs high ETVSS and also very limited to assess patients <6 months old. Also a substantial proportion of the studies were retrospective cohorts.
TVFB was associated with an increased rate of successful surgery in adults and children undergoing ETV. However, it was not applicable to patients <6 months old. We encourage further investigation to include this radiological finding to a prediction model.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
Not applicable for systematic review and meta-analysis
[Figure 1], [Figure 2], [Figure 3], [Figure 4]