Introduction

Hydrocephalus is a common clinical condition affecting the brain with an incidence of 1/1,000 live births [6]. The advent of shunting device almost half a century back by Spitz and Holter began the era of modern neurosurgical management of hydrocephalus and improved the prognosis of this condition [24, 26]. In the current era, although children with hydrocephalus can achieve a near normal neurological status, it is still associated with high rate of shunt failure leading to poor outcome and higher mortality [20, 28, 29].

While several studies analyzing the etiology, clinical profile, management, and outcome of pediatric hydrocephalus have been published from the developed countries, there is a dearth of data on hydrocephalus in children from developing countries. The current study was done with the aim of reviewing the complete 18-year experience with pediatric hydrocephalus at one of the largest tertiary care centers in Pakistan, a southeast Asian country. The data regarding the incidence, etiology, clinical features, management, and outcome were analyzed to study the time trends and the age-related differences.

Methods

This was a retrospective cohort study based on case note review of all pediatric patients (age, 1 month to 15 years) with hydrocephalus who were diagnosed and managed at the Aga Khan University Hospital, Karachi, over an 18-year period (1988–2005). The institution's medical record database with the International Classification of Diseases, Ninth Revision, Clinical Modification coding system was used to identify all children with the diagnosis of hydrocephalus. The case notes of all identified patients were retrieved and reviewed. These patients were originally seen by a pediatrician and then referred to a neurosurgeon for management. The patient characteristics analyzed included: age, sex, etiology, length of hospital stay, clinical manifestations, laboratory data, radiographic findings, management, and outcome.

An etiology was recorded for each case based on clinical findings and imaging studies. Based on previous studies on pediatric hydrocephalus [4, 26], the following etiologies were defined: hemorrhage (usually intraventricular, IVH), infection (including bacterial and tuberculous meningitis), brain tumor (including type and location), spinal dysraphism (including myelocele, meningocele, myelomeningocele, and spina bifida), arachnoid cyst, idiopathic aqueduct stenosis, idiopathic communicating hydrocephalus, congenital hydrocephalus, and miscellaneous others (including Arnold–Chiari and Dandy–Walker malformations, encephalocele, lissencephaly, and holoprosencephaly).

The details of management, including pharmacotherapy and date and type of surgery, as well as any revision procedures, shunt infections, or blockage, were documented. The clinical status of all patients was documented based on the data from last clinical follow-up of each patient. The outcome was analyzed based on the presence or absence of neurological and/or cognitive deficits on last clinic follow-up.

For analyzing time trends, the study period was divided into two epochs, each of 9 years duration (period A, 1988–1996; period B, 1997–2005). Incidence, etiology, clinical features, management, and outcomes were compared for each time period. Similar comparisons were also done for two different age groups (0–12 months and 1–15 years).

All data were collected on a proforma specifically designed for the purpose. The data entry and statistical analysis were done using Statistical Package for Social Sciences (SPSS) version 17.0 (© SPSS Inc., 1989–2007). The data analysis included simple frequency determination. For quantitative data, means±standard deviation were calculated. For quantitative data with skewed distributions, medians and ranges are presented. Pearson's chi-square test or Fisher's exact test, where appropriate, was used for comparison of proportions. The independent sample t test or the Mann–Whitney U test (for skewed distribution) was used for comparison of means. For all comparisons, a P value of <0.05 was considered statistically significant.

Results

During the 18-year study period (1988–2005), a total of 338 cases of pediatric hydrocephalus were identified, including 228 (67.5%) males and 110 (32.5%) females. The median number of patients presenting each year was 17 (range, 12–30). The number of patients in period B (1997–2005) was significantly higher as compared to period A (1988–1996, P = 0.002). There were more patients under 1 year of age (group A) than in the 1–15 years age group (group B, P < 0.001).

Etiologies of hydrocephalus

Tables 1 and 2 present the etiologic details for each epoch and age group, respectively. Overall, the most common etiology of pediatric hydrocephalus was meningitis (119, 38.1%), followed by congenital hydrocephalus (69, 20.4%), brain tumors (28, 8.3%), spinal dysraphism (26, 7.7%), and aqueductal stenosis (24, 7.1%). The other etiologies (21, 6.2%) included encephalocele (7, 2.1%), subdural hygroma (4, 1.2%), Arnold–Chiari malformation (3, 0.95), Dandy–Walker syndrome (3, 0.9%), lissencephaly (3, 0.9%), holoprosencephaly (2, 0.6%), infarct (2, 0.6%), hydrocephaly (1, 0.3%), vascular malformation of the brain (1, 0.3%), and Werdnig–Hoffman syndrome (1, 0.3%). Congenital hydrocephalus (59, 29.5%) followed by bacterial meningitis (48, 24%) comprised the major etiologies in children less than 1 year of age, whereas tuberculous meningitis (51, 36.9%) and brain tumor (24, 17.4%) were the most common causes of hydrocephalus in children 1–15 years of age.

Table 1 Different etiologies of pediatric hydrocephalus during the two time periods
Full size table
Table 2 Various etiologies of hydrocephalus in two pediatric age groups
Full size table

Table 3 describes the different brain tumor types and locations presenting with pediatric hydrocephalus during the study period. The most common brain tumor was glioma (12, 42.9%) followed by medulloblastoma (6, 21.4%). Overall, the most common locations of the tumors were the posterior fossa (8, 28.6%) and brainstem (8, 28.6%).

Table 3 Tumor types and locations presenting with pediatric hydrocephalus during the study period
Full size table

Clinical presentation

Overall, the common presenting features were nausea/vomiting (113, 33.4%), fever (107, 31.7%), seizures (95, 28.1%), large head (83, 24.6%), rapid increase in head circumference (61, 18%), and drowsiness/altered level of consciousness (59, 17.5%). Tables 4 and 5 present the clinical features in children with hydrocephalus and compare them across the two time periods and age groups.

Table 4 Presenting clinical features in children with hydrocephalus (overall and in the two time periods)
Full size table
Table 5 Presenting clinical features in children with hydrocephalus (overall and in the two age groups)
Full size table

Management and outcomes

The mean duration of hospital stay was 13.5 ± 13.186 days. The duration of hospital stay in period A children (16.92 ± 15.879) was significantly higher than period B children (11.06 ± 10.219, P < 0.001). However, there was no significant difference in the duration of hospital stay between two age groups (P = 0.501).

Tables 6 and 7 provide a comparison of management and outcomes in two different epochs and age groups, respectively. The pharmacological management of patients included the use of antibiotics (236, 69.8%), anticonvulsants (77, 22.8%), antituberculous treatment (62, 18.3%), mannitol (58, 17.15%), and steroids (52, 15.4%). Surgical management, including external ventricular drain (61, 18.04%) and ventriculoperitoneal shunt (216, 63.9%), was done in 255 (75.4%) patients. Shunt infection and shunt blockage was seen in 38 (11.2%) and 54 (16.0%) children, respectively, and 67 (19.8%) of them required shunt revision. The etiologies associated with highest rates of shunt infection included bacterial meningitis (12, 17.6%) and aqueductal stenosis (3, 12.5%). Although the rate of shunt infection, shunt blockage, and shunt revision was slightly higher in children under 1 year of age as compared to the 1–15 years age group, the difference was not statistically significant (P = 0.218, P = 0.222, and P = 0.227, respectively). Similarly, no significant difference was observed in the rate of shunt infection, shunt blockage, and shunt revision between the two epochs (P = 0.766, P = 0.178, and P = 0.771, respectively).

Table 6 Management and outcome in children with hydrocephalus (overall and in the two time periods)
Full size table
Table 7 Management and outcome in children with hydrocephalus (overall and in the two age groups)
Full size table

The mean duration of follow-up was 410 ± 968 days (range 0–6,033 days). Overall, some extent of neurological and/or cognitive deficit, at last clinic follow-up, was observed in 135 (37.0%) patients. Neurological and/or cognitive deficits were more commonly seen in patients with brain tumor (17, 85.0%), communicating hydrocephalus (4, 66.7%), and IVH (12, 54.5%). The neurological and/or cognitive deficits were observed more frequently in children under 1 year of age as compared to the 1–15 years age group (P = 0.029); however, no significant difference was observed between the two time periods (P = 0.079). The overall frequency of mortality was 38 (11.2%), but there was no significant difference between the two epochs and age groups (P = 0.059 and P = 0.865, respectively). The highest rate of mortality was seen in patients with IVH (6, 23.1%), followed by brain tumors (6, 21.4%) and spinal dysraphism (3, 11.5%).

Discussion

Pediatric hydrocephalus carries a significant burden with an incidence of 1 in 1,000 in newborns [6]. The introduction and use of shunts for CSF diversion has decreased the morbidity and mortality associated with pediatric hydrocephalus and improved the outcome of these patients. Despite the improvement in surgical techniques and technological advancements in the types of shunts over the years, the procedure is still associated with a high rate of infection and shunt revision. Till date, very few studies have been conducted on elucidating the risk factors and patient characteristics associated with poor prognosis and have yielded variable results [6–9, 14, 29].

There are many etiologies of hydrocephalus including congenital hydrocephalus, spinal dysraphism, meningitis, brain tumors, etc. The most common etiology noted in our study was meningitis followed by congenital hydrocephalus and brain tumor. This is in contrast with figures reported in literature where cranio-spinal dysraphism and congenital hydrocephalus are the most common etiologies and infectious causes are lower down on the list [12, 18, 20]. This can be attributed to the fact that infectious diseases and their complications are more prevalent in our part of the world as compared to developed countries.

Our study provides a good comparison of the two different epochs and age groups of patients. It highlights that the number of patients presenting with hydrocephalus increased over time, however there was a significant decrease in the duration of hospital stay over two eras. Nowadays, even in the developing countries, patients have better access to health care facilities as compared to the past, and overall, there has been a great improvement in management strategies and types of shunts which lead to a reduction in hospital stay. Nonetheless, the increase in the frequency of pediatric hydrocephalus observed in our study is in contrast to the recent literature from various countries which describes the decrease in the incidence of pediatric hydrocephalus [17, 21, 22]. Massimi et al. recently reported a significant 8.8% decrease in the incidence of pediatric hydrocephalus between two time periods comparable to ours [15]. In their study, there was a significant reduction of hydrocephalus associated to myelomeningocele, aqueduct stenosis, and CNS infections. This is in contrast to our study where the frequency of spinal dysraphism and aqueductal stenosis has increased. The rate of meningitis has remained stable; however, as expected in a developing country like Pakistan, it contributes a major proportion of cases to the pediatric hydrocephalus cohort over the two time periods. The increase in the rate of neural tube defects in our study sample reflects a lot about the lack of proper surveillance and preventive measures for such malformations in developing countries. However, we believe that these data need to be interpreted in a prudent manner, and epidemiological trends should not be drawn from a single-center data.

The main complication associated with treatment of hydrocephalus is the high rate of shunt infection requiring multiple shunt revisions and leading to neurological deficits and mortality in patients. It has been well established in literature that shunt infections are a major cause of mortality in patients treated for hydrocephalus [5, 14, 19, 29]. In our study, the rate of shunt infection and shunt blockage was 11.2% and 16.0%, respectively, which is comparable with other studies reporting infection rates varying from 8% to 22% [10, 11, 18]. Notarianni et al. have reported an overall shunt revision rate as high as 78% during their 14-year study period [18]. Results from previous studies signify that the rate of shunt failure is higher in the younger age group [3, 16, 28]. This difference was also seen in our study but was not statistically significant. Etiologies related to higher rates of shunt infection and revision included bacterial meningitis and aqueductal stenosis. It is interesting to note that these findings are not in concordance with the findings of other studies which report higher rate of shunt complications and poor outcomes in patients with intraventricular hemorrhage, cranio-spinal dysraphism, and brain tumors [2, 13, 27, 29].

Mortality rates in pediatric hydrocephalus patients range from 1.6% to 21.7% [1, 4, 7, 8, 14, 18, 20, 23, 29]. The mortality rate in our patient cohort was 11.2%, and despite the reduction in hospital stay noted over two epochs, there was neither a significant reduction in mortality rate over the years nor a significant difference in mortality rates in the two age groups. The highest mortality rate was seen in patients with IVH and brain tumor, which is in concordance with the findings of other studies [18, 25].

The main limitations of our study are that it is a retrospective case note review and the follow-up duration of patients is short. Thus, long-term outcomes of these patients are not known. Additionally, no specific tool or questionnaire was used to assess functioning, quality of life, and outcome of patients. The use of such a tool would have helped in better quantification and standardization of results. Our study has shown some interesting patterns and associations in our patient cohort between etiologies and outcome of patients with hydrocephalus, and further prospective studies are required to validate these results and associations.

Conclusion

Hydrocephalus is a common problem in the pediatric population. The most common etiologies of hydrocephalus in our population include meningitis, congenital hydrocephalus, and brain tumor. Despite the advancements over years, hydrocephalus is still associated with high rate of shunt failure and mortality. Factors associated with poor outcome include younger age group and etiology of hydrocephalus.