Introduction

The increasing popularity of endoscopic third ventriculostomy (ETV) in the treatment of hydrocephalus over the last decades has improved the outcome of treatment, with the pertinent question presently centred on selection criteria that predict good outcome [1, 2]. Hydrocephalus is a devastating surgical disease which, when left untreated, causes impaired psychosocial and mental development and cosmetically disfiguring craniofacial disproportion and could ultimately lead to death. Early appropriate treatment is associated with a significant reduction in morbidity and mortality as well as improvement in the quality of life of children presenting with hydrocephalus [2, 3]. The treatment of hydrocephalus has evolved over time. Shunting has been the traditional approach in treating the condition with the attendant complications such as shunt infection and malfunction [3]. In the last two decades, there is an upsurge in the use of ETV in the treatment of hydrocephalus in selected patients with a reduction of complications that traditionally accompanying shunting [1, 3]. The endoscopic third ventriculostomy success score (ETVSS) has been validated to be a useful preoperative tool that predicts the outcome of ETV [4,5,6,7]. Despite the major advantage of being shunt free, ETV is associated with varying success rates ranging from 50 to 90% [5] and could lead to progression of hydrocephalus that would necessitate another cerebrospinal fluid (CSF) diversion. It is therefore appropriate to devise clear-cut selection criteria among children with hydrocephalus in order to derive maximal benefit from the procedure in view of the moderate failure rate of the procedure.

The development of ETVSS is an attempt to accurately predict the outcome of ETV and quantify the common prognostic factors such as patient’s age, aetiology of hydrocephalus and history of previous shunting [8]. The scoring system helps to identify independent variables that predict a good outcome. This study was specifically aimed at evaluating the role of ETVSS as a preoperative parameter in children with hydrocephalus at Lagos University Teaching Hospital (LUTH), Lagos, Nigeria.

Methodology

This was a prospective hospital-based cohort study of 68 eligible children who met the inclusion criteria and underwent ETV out of a total of 161 patients who presented with hydrocephalus at the LUTH, Lagos, Nigeria, during an 18-month study period from November 2015 to April 2016 after obtaining ethics clearance from the human and research ethics committee. The inclusion criteria involve children under 2 years who presented with clinical features and brain computed tomography scan findings of hydrocephalus and qualified for an ETV procedure by the exclusion criteria.

The exclusion criteria were children above two years with hydrocephalus and unconscious, comatose and brain-dead patients with hydrocephalus. Patients who were treated for hydrocephalus with techniques other than ETV and children with hydrocephalus without brain computed tomography (CT) scan were also excluded. Patients who had choroid plexus cauterization (CPC) in addition to ETV were excluded.

All parents and guardians of children who satisfied the inclusion criteria were given an informed consent form to fill and sign. Gestational age at birth, presumed aetiology of hydrocephalus, previous history of shunting and age at surgery were recorded. The presumed aetiology of hydrocephalus was confirmed by brain computed tomography (CT) scan as part of preoperative evaluation. Occipitofrontal circumference (OFC) was measured at presentation using the glabella and occiput as reference points. Developmental milestones (social smile, neck control, sitting with and without support, crawling, standing, walking and talking) were evaluated and documented.

Endoscopic set-up was arranged. The right lateral ventricle was cannulated, cerebrospinal fluid (CSF) flow was confirmed and a flexible endoscope was introduced into the frontal horn of the right lateral ventricle through Kocher’s point which gives a direct trajectory through the foramen of Monro to the floor of the third ventricle. The floor of the third ventricle was identified using known landmarks (the mammillary bodies, infundibular recess and clivus) with basilar artery complex lying at the posterior edge of the third ventricular floor at the level of the mammillary bodies. The third ventricular floor was fenestrated bluntly using electrocautery, and at the ventriculostomy opening, cerebrospinal fluid flow was seen.

The ETVSS was calculated from the addition of scores for age, presumed aetiology (from brain CT scan) and presence or absence of previous shunt insertion, and the calculated (predicted) ETVSS was documented for each patient.

Table 1 shows the ETVSS chart. ETVSS is calculated by addition of the scores of age of patient at surgery, presumed aetiology and whether or not there has been previous shunting.

Table 1 Endoscopic third ventriculostomy success score (ETVSS) [9]
Full size table

Data collection and analysis

Data collected include birth history, presumed aetiology of hydrocephalus, presence or absence of previous shunting, complications at presentation, developmental milestones, brain CT scan findings, treatment and outcome. Patient’s confidentiality was protected by assigning a serial number to each patient at entry. The data collected was analysed using the Statistical Package for Social Sciences (SPSS) version 22 (Illinois, USA).

Correlation was done to establish relationships between ETVSS and ETV outcome using Pearson’s correlation ranking order, while the analysis of variance (ANOVA) and Kaplan-Meier statistical tool were used for comparing means where applicable, and p value < 0.05 was taken as significant. Receiver operating characteristic (ROC) curve was used to determine the predictive strength of the variable. Follow-up was for a period of 6 months which was carried out via routine clinic visits and mobile phone chat, and OFC and developmental milestone assessed.

Results

Sixty-eight eligible patients with complete data were analysed. The age of the 68 patients ranged from 0 to 24 months with a mean of 5.45 ± 5.41 months. Fifty-seven percent of patients were less than 6 months old while another 20% were between 6 and 12 months old. Forty-seven (69.1%) patients were male, while 21 (30.9%) were female with a male to female ratio of 2.2:1. Fifty-three (77.9%) of the 68 patients with hydrocephalus had no previous shunting prior to the ETV procedure, while 15 (22.1%) had ventriculoperitoneal shunting prior to ETV.

At 6 months follow-up, the ETV was successful in children aged 6 to < 12 months (100%, p = 0.001), followed by those aged 12 to 24 months (80%, p = 0.05) and 1 to < 6 months (69.2%), while ETV failed in 80% of patients aged less than 1 month.

It shows a significant area under the curve above the reference level of 0.5. The area under the curve (AUC) was 0.871, and the sensitivity and specificity of predicted ETVSS were 84% and 88.9%, respectively.

Eighteen (26.5%) of the 68 patients had failed ETV which was the major complication, but other minor complications were seen in 3 patients with surgical site infection that was treated daily with wound care and antibiotics.

The survival period is defined as the measured period of success of ETV within which there is no indication for an alternative secondary surgical intervention, and Fig. 2 shows that the higher the predicted ETVSS, the higher is the survival period of ETV and vice versa.

Discussion

The decision to treat a child with hydrocephalus surgically can be tasking in view of the options available for the treatment of hydrocephalus. The decision is based on available treatment options, efficacy of such options, available expertise, equipment and choice of treatment by the patient’s family.

ETV is a treatment option that seems to have fewer complications compared with the more traditional method of shunting [1, 3, 9, 10]. However, the use of a well-tailored selection criteria like the ETVSS done preoperatively is essential to determine who will likely benefit from ETV and help reduce the need for repeat procedures following failures [8, 9, 11].

ETV success is defined as freedom from the subsequent surgical procedure for the purpose of cerebrospinal fluid diversion [6]. A failed ETV is defined as having occurred in cases that will require subsequent surgical intervention by CSF diversion after primary treatment with ETV or death within 6 months after first ETV [6].

Fani et al. [11] in their study involving 59 patients below 2 years of age showed findings similar to this study as regards age and sex distribution. Males accounted for 62.7% of their patients, and 55.95% of all patients were below 6 months compared to 64.48% in this study who were less than 6 months old. In our study population, the trend to improve outcome increases to 12 months, and thereafter, there was a slight drop in the percentage of successful procedure.

In comparing our study to that of Breimer et al. [12] who externally validated ETVSS in 104 paediatric patients, age stratification was slightly different as 43.3% of the 104 patients studied were between 1 and less than 10 years probably due to a higher age of study population in their study. This study revealed that majority of the patients—61.8%—had presumed aqueductal stenosis (Table 2). The aetiological distribution showed a similar pattern of distribution from some literature. Breimer et al. [12] showed that the majority had aqueductal stenosis which accounted for 26.9%. The higher number of study population when compared to this study may have accounted for a lesser percentage of patients with aqueductal stenosis. However, Fani et al. [11] with almost similar study population showed that aqueductal stenosis was the commonest cause in the age group less than 2 years of age. As expected, this study showed that the measured mean survival period (time of success without intervention) for each stratum of the predicted ETVSS increases as the grading increases.

Table 2 Aetiology of hydrocephalus
Full size table

The findings from this study were similar to those by Fani et al. [11] and Breimer et al. [12] which showed that the success rate of ETV has a linear association with the predicted ETVSS score. This similarity in the predictive value of ETVSS may be a result of similar demography and duration of follow-up following ETV.

Kulkarni et al. [13] in a multicenter cohort study showed that the mean predicted ETVSS was 57.9% with a closely related actual ETV success rate of 59.2%. Kulkarni et al. [13] validated the ETVSS model and showed that the area under the curve (AUC) was 0.68. Naftel et al. [14] validated ETVSS with a follow-up period of 6 months, and the AUC was 0.74. In comparison to the above studies, this study showed a slight difference as the AUC was 0.871 with a sensitivity of 84% and specificity of 88.9% (Fig. 1). The difference may be due to the difference in sample sizes and probably late presentation in patients with hydrocephalus in our environment (Table 3). However, the AUCs in these studies and ours showed a significant predictive strength of ETVSS. The mean predicted ETVSS from this study was 48.82 ± 19.20% with a mean success score of 56.20 ± 15.10% (Table 4). This finding is similar to that of Kulkarni et al. [15] who showed that the mean predicted ETVSS was 57.9% with the closely related actual ETV success rate of 59.2%.

Fig. 1
figure 1

Receiver operating characteristic (ROC) curve comparing sensitivity and specificity (diagonal segments are produced by ties)

Full size image
Table 3 Comparison of age at surgery with outcome of ETV using ANOVA
Full size table
Table 4 The relationship between predicted ETVSS and outcome of ETV
Full size table

The correlation coefficient between predicted ETVSS and ETV outcome was 0.65 with a significant p value less than 0.05. This is similar to the findings observed by Kulkani et al. [15] that showed a linear correlation between predicted ETVSS and actual success of ETV.

The mean survival period after ETV therefore increases with increase in ETVSS using the Kaplan-Meier statistic tool as shown in Fig. 2 compared to findings from the work of Kulkarni et al. [13]. A positive correlation coefficient of 0.65 was recorded between the predicted ETVSS and actual ETV outcome within the follow-up period of 6 months using Pearson’s correlation ranking order.

Fig. 2
figure 2

The Kaplan-Meier survival curve for ETV in 68 patients stratified by ETVSS

Full size image

Conclusion

From this study, preoperative ETVSS which predicts ETV outcome is a valuable tool in deciding which children with hydrocephalus will most likely benefit from ETV, since findings from this study revealed that predicted ETVSS closely mirrors and predicts actual ETV outcome up to 6 months of follow-up.

We recommend the following:

ETVSS should be used preoperatively to evaluate the likely outcome of ETV in children with hydrocephalus.

ETVSS should form the basis for making a decision as to which children with hydrocephalus will benefit from the ETV procedure.