Introduction

Hydrocephalus can occur in association with other pathologies of the cerebrospinal fluid (CSF) system, e.g. cerebellar and posterior fossa anomalies (Dandy–Walker complex, Blake’s pouch, and megacisterna magna), arachnoid cysts, Chiari malformation, and spina bifida. The etiology of hydrocephalus as a result of impaired CSF circulation when coexisting with these pathologies still challenges our understanding of CSF dynamics, as cerebellar anomalies and or arachnoid cysts, for example, not infrequently occur as stand-alone pathologies; in other words, they can exist with and without hydrocephalus.

The aim of this chapter is not to provide a comprehensive overview or textbook information on all these pathologies, but rather to strengthen our appreciation of how complex the treatment of hydrocephalus and associated CSF disorders can be, across ages. For despite many advances in neurosurgery, these pathologies still provide a challenging complexity for those involved in diagnosing and treating hydrocephalus in both pediatric and adult patients.There is an element of disturbance of CSF circulation in all of these; however, the mechanism of how and why symptoms occur is still not clearly understood, and not always as obvious as it is with Chiari I and the resolution of syringomyelia after sufficient decompression. As congenital pathologies, they occur during the development of the central nervous system, and are therefore traditionally deemed incidental findings and rarely thought to be symptomatic. And yet, at the time of diagnosis, patients may exhibit symptoms that challenge the decision about clinical relevance and treatment indication, particularly in the adult or elderly population. The chronicity of the diseases is usually associated with insidious onset of a disease process, and as such, considered “benign.” If symptomatic, they might have significant impact on quality of life and function in an individual. Beyond that, little is still known about pathophysiology of the CSF system in the prenatal period. Furthermore, surgical treatment is associated with a high rate of recurrence or unsatisfactory results, for example in arachnoid cysts, and contemporary approaches need to utilize multimodal concepts; traditional shunt placement and microsurgical approaches are now partnered by more contemporary and “noninvasive” endoscopic treatment. There is no state-of-the-art treatment strategy available.

Clinical scenarios will be presented that support the complex nature of the disease, the challenging decision-making about whether or not to surgically intervene, and the difficulties with choosing the most efficient type of surgical intervention. The discussion will provoke suggestions for contemporary management, e.g. the importance of interdisciplinary management of patients, the role of modern magnetic resonance imaging (MRI) in the detection of CSF circulation and pathologies, and multimodal surgical strategies. The presentation of an 86-year-old patient with hydrocephalus associated with a symptomatic cerebellar arachnoid cyst should increase awareness that these pathologies might also be becoming more prevalent in the elderly population, where aging might provide a risk factor not only for normal pressure hydrocephalus but also for hydrocephalus associated with congenital CSF pathologies.

Basic and clinical research has to address improvement in diagnosis, management, and outcome of CSF disorders. Hopefully, we will be able to establish and maintain a pool of researchers and clinicians who are keen to work in the traditionally neglected area of hydrocephalus and associated disorders. Here, the recently formed International Society of Hydrocephalus and Cerebrospinal Fluid disorders (ISHCSF) and the International Hydrocephalus Imaging Working Group (IHIWG) are exciting initiatives to encourage and provide platforms for young researchers. These initiatives will be introduced in the conclusion of this chapter.

Case 1. Hydrocephalus associated with a posterior fossa cyst

A 4-month-old female presented with significant change in head circumference percentiles reaching above the 97th percentile, but otherwise reaching milestones. A first child, uncomplicated term vaginal delivery of a healthy couple without contributive history. Because of the concern about the increased head circumference, a brain MRI was ordered that showed massive enlargement of the lateral and third ventricles associated with posterior fossa enlargement of the cisterna magna and atrophic cerebellum without clear mass effect. Cine flow studies indicated disrupted CSF motion at the level of the foramen magnum and additional CISS sequences were suggestive of an arachnoid cyst. The cerebral aqueduct appeared patent though narrowed in the lower segment (Figure 24.1a–c).

Figure 24.1 Hydrocephalus associated with posterior fossa cyst. (a–c) Preoperative MRI findings of hydrocephalus in an axial T2-weighted image (a) and a sagittal T1-weighted image showing posterior fossa pathology (b), which was suggestive of an arachnoid cyst based on web-like structures (white arrow) seen in the CISS sequence (c). (d–f) Corresponding postoperative MRI findings after microsurgical cyst fenestration with improved hydrocephalus (d) and decompression of the posterior fossa cyst and expansion of the cerebellum (e) with flow void signal at the level of the foramen of Magendie and the cisterna magna indicating restoration of CSF circulation (f; white arrow).

Notably, the parents reported that despite appropriate milestones being reached, they were always concerned with a lack of head control when comparing their daughter with others in the playgroup. However, despite the increase in head circumference, the clinical examination did not reveal any obvious signs of increased intracranial pressure. There was a full but not a tense fontanelle and absence of splayed sutures, leaving the impression of a chronic condition.

Given the clinical concern and the radiographic findings of a posterior fossa arachnoid cyst in the presence of a patent aqueduct, a midline posterior fossa craniotomy with microsurgical fenestration of the arachnoid cyst into the cisterna magna and the basolateral cisterns was selected and performed in an attempt to eventually also alleviate the hydrocephalus. A temporary external ventricular drain was placed, however, which could be weaned at postsurgical day three. The histopathological findings confirmed fragments of dense, fibrous connective tissue and adjacent pigmented arachnoid lining consistent with an arachnoid cyst, further supported by EMA (epithelial membrane antigen) positive immunohistochemistry of the specimen.

The postoperative course was uneventful and the 1-year postoperative MRI findings show resolution of posterior fossa pathology with improved but persistent hydrocephalus (Figure 24.1d, e). Clinically, the head circumference improved to the 90th percentile and seemed to plateau at that level. Pediatric neurology is following the patient bi-annually and stated that the now 2-year-old girl is developing appropriately without motor and mental delay.

Case 2. Posterior fossa cyst without hydrocephalus

A 16-year-old female was admitted emergently for failure to thrive, nausea and vomiting, and an additional history of migraines and psychiatric illness. A brain MRI showed a posterior fossa cystic lesion with mild to moderate mass effect on the cerebellum. Additional CISS sequences suggested the presence of an arachnoid cyst that had some focal communication with the spinal canal associated with CSF turbulences (Figure 24.2a–c). There was no hydrocephalus. Given the teenager’s complicated and pre-existing history of migraines and psychiatric illness, the findings were initially deemed incidental. The patient continued to have excesses of headaches associated with episodes of vomiting and dry heaving during the hospital stay, and pediatric neurology was consulted and raised concerns of postural headaches associated with posterior fossa pathology.

Figure 24.2 Posterior fossa cyst without hydrocephalus. (a–c) Preoperative MRI findings showing a retrocerebellar cystic lesion with some mass effect on the cerebellum as indicated in a midline sagittal T1-weighted image (a). The axial T2-weighted image (b) shows a flow void signal in the cisterna magna suggestive of CSF turbulences (white arrow), which was supported by a sagittal CISS sequence focusing on that area that indicated a focal communication inferiorly, inferior to the fourth ventricle, suggestive of a “ball-valve” mechanism causing the cyst to expand (c, black arrow). (d–e) Postoperative MRI shows reduction of the cyst and expansion of the cerebellum in a sagittal T1-weighted image as well as restored CSF flow at the level of the foramen magnum as indicated by physiological flow void signals at the foramen of Magendie, and between the cisterna magna and the cervical spinal CSF compartment (e, black arrow). The intraoperative microscopic view during microsurgical fenestration of the cyst supports the finding of an arachnoid cyst with a pre-existing focal opening (black arrow) to support the “ball-valve” mechanism as suggested by the radiographic findings (f).

Given the prolonged disease state despite all medical and critical care efforts, the interdisciplinary decision was made to explore the cyst. A midline craniotomy with microsurgical fenestration of the arachnoid cyst was performed, and the intraoperative finding supported a “ball-valve” mechanism possibly causing intermittent expansion of the cyst by a “jet flow” of CSF from the spinal compartment into and out of the cyst triggered by cardiac cycles (Figure 24.2f). A wide microsurgical fenestration of the cyst into the spinal compartment was performed.

Postoperatively, the patient fully recovered within 14 days; however, she developed a pseudomeningocele with recurrence of positional headaches (Figure 24.2e). The pseudomeningocele was repaired with an autologous fat graft a month later and the last postoperative MRI 3 months post-microsurgical fenestration shows resolution of the pseudomeningocele and a significantly decreased size of posterior fossa cyst and overall improved posterior fossa conditions (Figure 24.2d). The patient, 1 year after surgery, despite some occasional episodes of headaches similar to her previous baseline and history of migraines, resumed full activity without any further episodes of positional headaches and nausea and vomiting.

Case 3. Hydrocephalus associated with Chiari I malformation

A 7-year-old female with a known history of attention deficit hyperactivity disorder (ADHD) became more symptomatic with deterioration in her grades despite medical management of ADHD. For her reported “occipital headaches” she was also diagnosed with “juvenile migraines.” Given the concern about her decline in grades starting from A grades now down to C and D grades as well as the increased complaints about occipital headaches, the pediatrician ordered an MRI that disclosed significant ventriculomegaly and a 7 mm Chiari malformation (Figure 24.3a–c). MRI of the spine did not show any associated syringohydromyelia or other spinal pathology. The cine flow study found absent flow signal at the level of the foramen magnum (Figure 24.3c).

Figure 24.3 Hydrocephalus associated with Chiari I malformation. (a–c) Preoperative MRI findings show hydrocephalus in an axial T2-weighted image (a) and a 7 cm tonsillar descent in a sagittal T1-weighted image (b) with absent CSF flow signal posterior to the tonsils supported by cine flow studies (c). (d–f) Two years after shunt surgery (d, white arrow), there is slight improvement of hydrocephalus though persistent and but slightly worse appearance of Chiari malformation, now reported as 1.3 cm, with restricted CSF flow signal on cine MRI studies (e).

Upon further evaluation it became clear that in addition to the decline in grades and the daily headaches, the girl also had a history of falls, balance problems, clumsiness, and dizziness, and the mother as well as the school teacher felt that “she spaces out” more, though it was initially attributed to her known ADHD.

The pediatric neurology department was involved and found appropriate milestones and suggested physiological intoeing as a potential cause of her clumsiness. Also, there was a positive family history of ADHD and learning disabilities on the maternal side. Her head circumference leveled at the 90th percentile. Given the above findings, observation and neurocognitive testing was indicated in the presence of a normal neurological exam.

The neurocognitive baseline testing disclosed impairment in attention and behavioral regulation, in maintaining focus and executive function, consistent with “Attention-Deficit/Hyperactivity Disorder but R/O Specific Learning Disorder (e.g. Hydrocephalus) deficits is mandated.” Despite a stable neurocognitive exam on repeated examination 6 months after initial diagnosis, there was ongoing decline in school grades and more distress with episodes of the girl holding her head for reports of headaches and “head pain.” Due to the parents’ as well as teachers’ concern, an interdisciplinary decision was made for placement of a ventriculoperitoneal shunt with an adjustable valve system. It was felt that the progressive symptoms of cognitive and balance decline and the neurocognitive profile were attributable to the significant chronic hydrocephalus and that shunt treatment should be given priority. Though there was absent flow signal in the cine studies, the Chiari abnormality was not deemed severe enough to suggest a causal relation with the hydrocephalus, and it was felt that the Chiari I was coexisting with rather than being the cause of it. At one year after shunt placement, the girl improved remarkably with improved school performance (reports of B grades) as well as full recovery from her falling spells and significant reduction of headache episodes from daily to occasionally. The girl maintained improvement within the 2-year postoperative follow-up. However, the postoperative images at 2 years post-surgery do not show significant changes in ventriculomegaly and the Chiari I malformation (Figure 24.3d–f).

Case 4. Normal pressure hydrocephalus associated with cerebellar arachnoid cyst

An 86-year-old female had a history of a left-sided symptomatic cerebellar arachnoid cyst when she was 30 years old. The cyst was microsurgically fenestrated, and finally successfully shunted with one revision. She had been doing well since then though some residual balance problems remained after some difficulties with managing the cyst.

In the previous year, the patient experienced much more unsteadiness to a point where she could no longer walk independently and displayed overall weakness and lack of initiative and interest. Extensive work-up for systemic illnesses given her age was initiated by the primary care physician with negative results. A head CT scan was performed showing increase in ventricular size with transependymal signal suggestive of hydrocephalus. The ventricular size was markedly enlarged compared to previous scans performed a couple of years previously while the appearance of the cerebellar cyst was stable (Figure 24.4a, b). Due to vessel clips in the area of the posterior fossa cyst presumably applied during her first surgery with microsurgical fenestration, the patient was not able to undergo MRI assessment (Figure 24.4c).

Figure 24.4 Normal pressure hydrocephalus associated with cerebellar arachnoid cyst. (a, b) A non-contrast CT scan shows findings consistent with normal pressure hydrocephalus associated with a cerebellar arachnoid cyst that was shunted 50 years previously after an unsuccessful attempt of microsurgical fenestration. It appears that the proximal catheter has moved out and is located in a cyst septum. There is no obvious mass effect from the cyst (b). (c) Anteroposterior X-ray film shows a right ventriculoperitoneal shunt placed for normal pressure hydrocephalus and also vessel clips (white arrow) that were placed 30 years ago on a first attempt to fenestrate the arachnoid cyst microsurgically. As such, an MRI was not possible. (d, e) Post-shunt placement, the hydrocephalus improved; however, the arachnoid cyst increased in size with mass effect as seen by displacement and compression of the fourth ventricle (e). CT scan after placement of a new cystoperitoneal shunt (f).

Given the clinical and radiological findings of progressive ventriculomegaly and functional decline, ventricular shunt surgery was performed. Though it was clear that the old cyst shunt was located within a membrane that separated a small portion of the cyst indicating a non-functioning position, there was no indication of mass effect from the cyst to suggest that the cyst was symptomatic (Figure 24.4b). After ventriculoperitoneal shunt placement, the patient did well for about a month with improved walking and balance; however, she started to develop symptoms of loss of appetite and nausea.

Upon emergent admission to the hospital for failure to thrive, the patient was investigated again for coexisting systemic illnesses, but all tests were negative. Also, the CT scan showed resolution of hydrocephalus. However, an increase in the size of the cerebellar arachnoid cyst was noted when compared to the CT scan that was done before placement of the ventriculoperitoneal shunt 6 months prior to the onset of the above symptoms (Figure 24.4d, e). There was now a subtle mass effect on the cerebellum and fourth ventricle and brainstem. Medical management of her failure to thrive and her nausea was maximized; however, no improvement was seen and instead there was further deterioration. Given her age and the equivocal scenario, the neurology department was involved over the question of a coexisting neurodegenerative process; however, the concern with the increased size of the arachnoid cyst persisted and placement of a new cystoperitoneal shunt was performed with reduction of the cyst size and resolution of mass effect (Figure 24.4f). The patient did not show any further episodes of vomiting and nausea and had a full appetite at her 1-month and 6-month follow-up post surgery.

Case 5. Hydrocephalus with multiple arachnoid cysts

A 4-month-old girl was admitted emergently with failure to thrive, irritability, and lethargy along with increasing head circumference. On examination, the child displayed clear signs of increased intracranial pressure with bulging fontanelle, splayed sutures, sun setting signs, and distended scalp veins. There was no prior complicated history in this term vaginal delivery first child of a healthy couple without contributory history.

The MRI revealed massive hydrocephalus with T2 hyperintensities around the frontal and occipital horns indicating transependymal CSF signal. Of note, too, is that there were bilateral sizable temporal arachnoid cysts and enlargement of the cisterna magna. There was no indication of aqueductal stenosis as an underlying cause of the hydrocephalus (Figure 24.5a–c).

Figure 24.5 Hydrocephalus with multiple arachnoid cysts. (a–c) Preoperative T2-weighted MRI images showing severe hydrocephalus with transependymal flow signal around the frontal horns (a). A sagittal CISS sequence showed posterior fossa abnormalities suggestive of megacisterna magna or Dandy–Walker variant with a patent aqueduct (b). Additional findings were bilateral temporal arachnoid cysts (c). (d–f) Two years after shunt surgery for hydrocephalus, ventricular size has normalized as seen on T2-weighted axial MRI (d) with resolution of the posterior fossa abnormalities (e) and stable bilateral arachnoid cysts (f); however, a new finding of a suprasellar arachnoid cyst is seen in the sagittal CISS sequence (e, asterisk).

Given the clinical and radiological impression of acute hydrocephalus, shunt surgery was performed and the girl did well without signs of hydrocephalus on clinical examination and improved ventricular size on follow-up MRI scan. The girl got lost from follow-up and presented at 2.5 years of age with concerns for new irritability, headaches, and visual disturbances. A follow-up MRI ordered by the pediatrician showed sufficiently drained lateral ventricles and stable sized temporal arachnoid cysts. However, a large suprasellar arachnoid cyst as a new finding was suggestive as the cause of her new clinical symptoms of headaches and visual disturbances. Of note, the enlargement of the cisterna magna had subsided (Figure 24.5d–f). The patient recently underwent successful endoscopic transventricular fenestration of the arachnoid cyst through a right frontal burr hole approach.

Case 6. Hydrocephalus associated with a giant suprasellar arachnoid cyst diagnosed in a fetus

A level III ultrasound revealed massive hydrocephalus associated with a posterior fossa pathology most likely an arachnoid cyst or a Dandy–Walker variant in a 24-week-old fetus of a twin gestation (Figure 24.6a). The couple have a 5-year-old healthy girl. A fetal MRI was performed that suggested hydrocephalus associated with posterior fossa arachnoid cyst.

Figure 24.6 Hydrocephalus associated with a giant suprasellar arachnoid cyst diagnosed in a fetus. (a–d) Preoperative findings of a fetal ultrasound showing hydrocephalus and infratentorial pathology, suggestive of a posterior fossa cyst or Dandy–Walker malformation (a, white asterisk). T2-weighted axial MRI (B) and sagittal CISS sequences (c, d) at day 3 post-partum show a giant arachnoid cyst extending into the supra- and infratentorial compartment with multiple septi; however, the CISS sequences suggest one compartment cyst of suprasellar origin, and endoscopic fenestration into the spinal canal and opening of the caudal cyst membrane (d, white arrow) was attempted. A left frontal transventricular endoscopic approach was performed and fenestration of the cyst into the spinal compartment was accomplished as seen by the intraoperative endoscopic view (e; C = clivus, B = brainstem basilar artery, asterisk = cyst membrane, white arrow = spinal subdural space). (f–h) Postoperative corresponding MRI findings show significant reduction of cyst size and improved hydrocephalus in both axial and sagittal T2-weighted and CISS sequences (f, g) as well as CSF flow signal suggestive of communication of the lateral ventricles with the spinal compartment shown by T2 CSF space imaging (h).

Both fetuses were stable throughout pregnancy and cesarean section was scheduled at 38 weeks of pregnancy. The twin girls were stable, and initially, there were no signs of increase of intracranial pressure in the affected twin. An MRI of the brain and spine was done on day two, and showed a massive arachnoid cyst, likely of suprasellar origin extending into the prepontine cistern, the third ventricle, and lateral ventricles with massive displacement of midbrain, brainstem structures, and the cerebellum (Figure 24.6b–d).

Due to the stable condition and the fact that the baby was thriving, it was decided to delay treatment. Repeated MRI including dedicated CISS sequences showed that the cyst most likely consisted of one compartment, with extension into the lateral and third ventricle, as well as the potential of fenestrating the caudal portion of the cyst into the spinal CSF compartment (Figure 24.6d). Therefore endoscopic fenestration through a left frontal transventricular approach was deemed feasible. During surgery, fenestration of the inferior prepontine portions of the cyst into the spinal subarachnoid space was also accomplished (Figure 24.6e). Six months after endoscopic treatment, the child suffered from strabismus, but was otherwise well, and developmentally about 3 months behind her twin. Imaging showed improvement of the cyst size and CSF flow (Figure 24.6f–h).

Case synopsis and discussion

The above cases all demonstrate clinical scenarios in patients with congenital pathologies of the CSF system with and without hydrocephalus suggestive of disturbed CSF dynamics causing the patients’ symptoms or problems. The case mix of patients shows that clinicians are confronted with these conditions at all stages of life, and that age-associated comorbidity and the individual spectrum of symptoms provides a difficult clinical scenario, e.g. in the teenager and the 86-year-old woman where despite all medical efforts and no improvement being seen, finally surgical intervention was sought by the primary care provider and the family. Sometimes treatment had to be offered in the equivocal scenario, despite incomplete understanding of how and whether the cyst ultimately caused the patient’s symptoms and how surgical intervention would affect outcome. The case of the 7-year-old girl with the Chiari I and the associated normal pressure hydrocephalus type of radiological and clinical picture shows that interdisciplinary management including cognitive testing might be superior in the decision-making than the traditional methods of intracranial pressure monitoring or measurements of intracranial compliance and resistance to outflow in these complex conditions [1–3]. In particular, it is not always clear whether and how the hydrocephalus is related to the associated finding, for example in the 4-month-old girl with the posterior fossa cyst, where successful decompression and establishment of flow did not result in resolution of the hydrocephalus. It is likely that the hydrocephalus coexists with the posterior fossa cyst and will eventually need treatment once subtle symptoms that appear chronic in nature occur. This child will need thorough and close follow-up of motor and mental milestones in her young life and any decline should warrant intervention on her hydrocephalus given increasing evidence of the late life impact untreated hydrocephalus may have on cognitive outcome, e.g. as known from the LOVA concept [4,5]. Also the patient with the Chiari I plus hydrocephalus indicates that the normal pressure hydrocephalus condition with elements of the triad (i.e. slow progression of gait and balance problems, cognitive decline, and incontinence) is prevalent in toddlers and infants, and is probably still underdiagnosed in the pediatric population [6,7]. In the differential diagnosis of ADHD in this age group, hydrocephalus should be considered. The sequelae of chronic hydrocephalus with (delayed) onset of cognitive and functional decline despite stable ventriculomegaly indeed suggests a progressive neurodegenerative process associated with chronic hydrocephalus, as suggested not only by experimental studies [8–10] but also by studies of metabolism and cerebral blood flow in normal pressure hydrocephalus [10–12]. The cases also show that any intervention in the CSF spaces may unpredictably change CSF dynamics and pathology is still incompletely understood, as in the 4-month-old patient who developed a previously absent large suprasellar cyst that occurred after successful shunting of hydrocephalus associated with bilateral temporal arachnoid cysts; or the increase of the pre-existing but previously stable cerebellar arachnoid cyst in the 86-year-old woman 6 months after shunting of normal pressure hydrocephalus. Part of this might certainly indicate some of the existing limitations of imaging of hydrocephalus and the associated impairment of CSF spaces and dynamics. Ultimately, those complex scenarios also question our traditional thinking about the association of intracranial pressure, compliance and CSF absorption and hydrocephalus [13,14], and suggest that we should appreciate that our traditional and dogmatic ways of measuring ICP and CSF dynamics with compliance and R_out values are limited in helping our understanding of disease associated with CSF dynamic disturbances [15–19]. Also, the fact that despite enormous pathology, e.g. in the fetus with the massive suprasellar arachnoid cyst, the condition is stable, whereas in a teenager or an elderly person little pathology or only a small change in pathology can push the patient’s health over the edge; age-associated susceptibility and comorbidity is probably one of the most important factors determining the disease and the outcome of pathology associated with hydrocephalus and associated CSF pathology/disorder [20–22]. Age as a complicating factor in hydrocephalus has been suggested by the LOVA concept as well as observations of congenital origin in elderly patients presenting with normal pressure hydrocephalus. However, this understanding should probably be translated to all pathologies that are associated with congenital CSF circulatory disorders. For example in Chiari I, borderline conditions more frequently seem to affect the third and fourth decades of life [23].

Needless to say that despite the difficult decision of whether to treat, the mode of treatment is very individual given the above incomplete understanding of pathology and clinical symptoms in these circumstances. Traditionally, most of the findings, particularly all kinds of arachnoid cyst as well as Chiari I, have initially been considered incidental findings when diagnosed with imaging, more or less regardless of the patient’s complaints. Unless the patient exhibits focal signs that match with a potential mass effect or irritation of structures in the location of the cyst or potential CSF dynamic perturbance (e.g. in syringomyelia [24–26]), treatment is usually not offered or it is suggested that the finding has nothing to do with the patient’s complaints. Particularly in non-focal or non-lesion-related signs, such as headaches, cognitive decline or decline in school performance, clumsiness, failure to thrive, behavioral outbursts, depression, and emotional problems, there is almost always for any of the above pathological circumstances and for any age group a comorbid condition at hand that might also explain and account for the patient’s problem.

Only recently, again stemming from treatment of Chiari I patients, has there been more appreciation of nonspecific findings, particularly certain cognitive elements that indicate that CSF dynamic disturbances might have systemic effects on CSF homeostasis with an impact on cognitive and behavioral outcome [27,28]. Advanced concepts in the understanding of diseases associated with CSF disorders, e.g. arachnoid cysts, posterior fossa anomalies, Chiari I, and spina bifida with and without hydrocephalus, are needed including more standardized and multidisciplinary approaches [15,29]. For example, the 7-year-old girl with Chiari and hydrocephalus did show remarkable improvement in headaches, balance, and school grades, but no change in ventriculomegaly and Chiari after ventriculoperitoneal shunting. This demonstrates that, in hydrocephalus and in the CSF disorders addressed in this chapter, we often see remarkable improvement despite no or little change in pathology. This basically raises interesting research questions that have not yet been answered: (i) does shunting affect brain metabolism and CSF dynamics beyond what we can see and document by imaging; and (ii) to what extent is there a placebo effect of our interventions?

Increased accessibility and use of brain and spine imaging modalities with MRI techniques will likely discover more congenital pathology, particularly in patients at later stages in life as disability is no longer accepted due to a generally higher longevity and the demands of modern society [19,30,31].

The following paragraphs will suggest that more standardized approaches and management in those complex and sometimes exceptional conditions of hydrocephalus and associated CSF disorders across ages can be accomplished if one extrapolates recent and advanced knowledge and concepts in the diagnosis and management of hydrocephalus. The concepts include the role of cognitive testing in the diagnosis and prognosis of disease, advances in imaging of CSF physiology and pathology, multimodal and multidisciplinary neurosurgical techniques, as well as improved classification schemes of CSF disorders.

The role of cognition and cognitive assessment

In normal pressure hydrocephalus, the role of a certain cognitive profile and its importance in diagnosing the disease as well as in the understanding of the disease burden has been established and seminal work was published in the last three decades of the twentieth century. Deficits in attention, visuospatial function, visual over verbal impairment, concentration, and impaired wakefulness have been found as unique features throughout and across different investigators [30]. Despite these efforts, cognitive tests have never traditionally been used in the decision-making as to whether to shunt or not to shunt a patient or, in other words, used as predictors of shunt treatment. Mainly, neurocognitive test batteries of various origins have been used to support outcome assessment in more systematic studies. Only very few but non-systematic attempts have been made to identify prognostic values, however, and have yet not been able to identify cognitive tests with sufficient accuracy to predict the response to shunting [32,33]. Also, well-designed and systematic studies and trials from expert centers published on intracranial pressure measurement, measurement of resistance to CSF outflow and compliance, and other adjunctive testing have ultimately failed to accurately predict shunt response [19].

The recent European multicenter trial conducted on the prediction of outcome in idiopathic NPH revealed the importance of a neurocognitive test battery with accuracy of diagnosis and prognosis of shunt response and 1-year outcome of shunt treatment far superior to the high-volume CSF tap test and R_out measurement [34–36]. The findings showed that a battery of three cognitive tests, the Stroop test, Rey Auditory Verbal Learning Test (RAVLT), and Grooved Pegboard test, were highly discriminative and diagnostically accurate in both probable and possible NPH when comparing the performance of iNPH patients with that of 108 healthy individuals (HI). iNPH patients performed significantly worse than HI on all of the neuropsychological measures at entry. The discriminative capacities of the eight variables were similar, with areas under the curve (AUC; ROC analysis) ranging between 0.86 (Delayed Recall) and 0.95 (Grooved Pegboard). The most usable test was RAVLT (Learning and Delayed Recall), administered to ≥90% of patients on all occasions. However, the Grooved Pegboard and the Stroop test were more sensitive to treatment effects. The three neuropsychological tests used in the European iNPH study are expedient, highly diagnostically discriminative, and well suited to evaluate changes following shunt treatment [37].

Little is known about the potential of cognitive tests and their diagnostic and prognostic accuracy in other CSF pathologies with potential CSF dynamic disturbance, e.g. Chiari I, arachnoid cysts, spina bifida, and posterior fossa anomalies. As discussed above, some evidence exists for Chiari I malformation and associated cognitive decline to suggest that any abnormal anatomy of the CSF spaces in fact has an impact on cognitive function [23,27,28,38–40], e.g. behavioral and autistic problems indicating higher cognitive impairment. The above supports and stresses the importance of cognitive disorder in CSF dynamic diseases and the potential value of cognitive batteries to improve the diagnosis and prognosis. Systematic prospective studies in hydrocephalus and associated CSF disorders are encouraged to identify cognitive domains most likely affected by these disorders and which show change after treatment and intervention and have the potential to increase the diagnostic as well as prognostic accuracy of cognitive test batteries [33].

The role of neuroimaging and MRI

As stated above, invasive CSF dynamic testing, e.g. R_out and compliance measurements, have failed to improve selection of shunt candidates, and have not helped our understanding of disease processes and pathophysiology. For example, variable and only weak correlation of resistance to outflow or compliance and CSF biomarkers of neurodegeneration and ischemia (e.g. neurofilament and sulfatide) has been shown [41,42]. Furthermore, other metabolic studies done in chronic hydrocephalus have rarely shown any correlation with CSF dynamics. This questions the significance of those measures and of CSF dynamic testing for the disease and the pathophysiology of chronic hydrocephalus [43,44].

MRI has evolved in the past 5 to 10 years with improved dynamic imaging (e.g. MR phase-contrast imaging) enabling assessment of CSF pulsations in relation to arterial pulsation and visualization of CSF pulsation in critical areas of interest, e.g. the cerebral aqueduct or at the foramen magnum [45].

Certainly, to date, these measures, too, lack evidence as to the definitive diagnostic and prognostic value, as seen with the battle around the value of the aqueductal stroke volume in normal pressure hydrocephalus [46–48]. But aqueductal stroke volume and its evolution is also an example of encouraging and ongoing research to improve MRI techniques, accuracy, and also ways of quantifying CSF volume and CSF spaces and compartments using MR imaging [49]. An example is the research provided by Alperin et al. on the ICP-MRI [50,51]. Though still a subject of controversy it leaves one with encouragement that future diagnosis of hydrocephalus and CSF disorders will be noninvasive and superior to classical invasive CSF dynamic testing. Besides new ways of quantification it allows assessment of dynamics with anatomical details in various brain and CSF compartments in conjunction with measures of metabolism and anatomical structures, which will eventually give us a more comprehensive understanding as they might reflect both brain biomechanics and biochemistry.

With regard to anatomical detailing of CSF spaces, most exciting are the T2-weighted 3D-volume-based detailed imaging of CSF spaces as obtained by the CISS and FIESTA sequences that are increasingly being used to assess cysts, Chiari, and other types of obstruction of CSF flow in hydrocephalus [52]. They have the potential to change our understanding and classification as well as management of hydrocephalus as they might shed light on the causes of CSF flow disruption or noncommunication in conditions previously thought to be communicating hydrocephalus [53–58]. Also, the better anatomical detail of membranes and webs in association with physiological CSF spaces will eventually change and improve the reasoning behind, approaches to, and techniques of surgical treatment.

The role of multimodal surgical management

Endoscopic approaches to arachnoid cysts and hydrocephalus have dramatically increased in the past 10 years. For example, the indication of third ventriculostomy has expanded and is increasingly used with success in non-classical conditions such as normal pressure hydrocephalus as well as in post-hemorrhagic and/or post-meningitic hydrocephalus and hydrocephalus associated with spina bifida and interhemispheric arachnoid cysts [59,60]. This further challenges our full understanding of where and how CSF dynamics fail in hydrocephalus and associated CSF disorders.

Also, in arachnoid cysts, more endoscopic treatment routes and management is published in locations that were thought to be the domain of microsurgical fenestration, e.g. posterior fossa, cerebellopontine angle and/or supracerebellar cysts and spine [61].

Particularly in pediatric neurosurgery, the “minimally invasive” aspect is promoted, though in most cases and not just the complex ones, the limitations of all the available surgical approaches, e.g. shunting versus endoscopy versus microsurgical techniques, should be appreciated. A more thorough and evidence-based approach to the pros and cons and the efficacy of these measures both as stand-alone and in combination should be considered in collaborations among surgeons who are dealing with complex CSF pathology. Also, information about long-term outcome in various endoscopic approaches is still lacking. Certainly, neuronavigation and modern MR imaging techniques will allow multimodal approaches utilizing microsurgical and neuroendoscopic techniques, but knowledge about disease mechanisms, indications, and timing of treatment and outcome needs standardization in both the adult and pediatric population.

One aspects needs attention, which is the fetal aspect of intervention as recently challenged by the results of the MOMs trial, evidencing improved outcome of Chiari I and hydrocephalus as well as ambulation in the prenatal repair group. The MOMs trial was initiated in February 2003 as a multicenter trial involving the Children’s Hospital of Philadelphia (CHOP), University of Vanderbilt, Tennessee, and University of San Francisco, California (UCSF). The trial aimed to assess the benefits of prenatal repair in spina bifida, and the prospective protocol applied very strict criteria on the selection of women and their fetuses diagnosed with spina bifida. One of the selection criteria was that the defect should be below L3. Here, one has to acknowledge the advances made in fetal MRI and its assessment and interpretation. Despite enormous difficulties with recruitment, one of which was the fact that families had to commit to staying in close proximity to the centers from the time that fetal surgery had been performed around week 26, and also mothers had to commit to accepting cesarean section for any future pregnancies, the trial showed clear and stunning benefits for the child and due to the significance of the benefit, the trial was halted prematurely for evidence of success; for example, at 30 months 42% of the children that had undergone prenatal repair were walking independently compared with 21% in the postnatal repair group. Furthermore, 36% in the prenatal group were free of Chiari versus only 4% in the postnatal group and, finally, only 40% of children received a shunt compared to 82% in the postnatal group [62,63]. Soon the results of MOMs II will shed further light on the long-term consequences and effects, e.g. differences in cognitive outcome or prevalence of tethered spinal cord and complications of LOVA with reluctance and potential bias to treat hydrocephalus. Despite some criticism and concerns, mainly with the maternal complications, there are important aspects with the MOMs trial that deserve attention particularly in line with the message of this chapter.

As mentioned above, difficulties with recruitment had led to skepticism that any definite result based on the calculated number of surgeries needed would ever be available anywhere in a reasonable time frame. With multicenter collaboration utilizing a well-designed prospective study protocol backed up with scientific and surgical expertise and endurance based on a multidisciplinary (e.g. prenatal team) program and hospital supports, revolutionary results can be obtained. This has happened even in rare or complex diseases and/or diseases where further advances were not thought likely, but, in the end advances came, to the surprise and excitement of an entire medical community and generation, e.g. the results of the MOMs trial of fetal repair of spina bifida.

The role of disease classification

It is important also to think about new dimensions of classification of hydrocephalus and associated abnormalities of the central nervous system (CNS) at various ages. This is even more important considering the fact that with increased knowledge of molecular science, new treatment modalities for central nervous disorders will become available (e.g. pharmacological CNS applications and stem-cell treatment). A very neat classification of posterior fossa anomalies had been proposed by King et al. in the Journal of Neurosurgery: Pediatrics in 2010 [61]. Due to the known uncertainties and the disease spectrum that complicates the clinical and radiological differential diagnosis of congenital posterior fossa anomalies that occur in Dandy–Walker, megacisterna magna, Blake’s pouch, and posterior fossa arachnoid cysts with and without hydrocephalus, King et al. subsumed these pathologies based on their anatomical location under the term “giant retrocerebellar cysts.” They identified pathological as well as distinct imaging criteria and listed the frequency of those criteria observed in each of the traditional entities, e.g. choroid plexus is absent in Dandy–Walker cyst, while ependymal lining is never seen in arachnoid cyst. While imaging does not discriminate between the anatomy of the posterior fossa and appearance of the occipital bone, all can show normal size to enlarged and thinned occipital bone, and variable mass effect is also present in all entities. Arachnoid cyst and Blake’s pouch rarely show other associated CNS malformations with a preserved falx cerebelli, whereas in Dandy–Walker cyst the falx cerebelli is absent but the combination with other CNS malformations is frequent. The differentiation between Blake’s pouch and arachnoid cyst on imaging is the mild vermian hypoplasia, which certainly still provides difficulties; however, the proposal from King et al. [61] is a very nice example that should encourage stepping back from traditional entities and providing classification schemes that are based on advanced molecular, pathological, and imaging criteria and on disease and pathology rather than names.

Another important step forward in this direction, though still criticized among researchers, is the recent proposal for a new classification scheme for hydrocephalus that tries to overcome the limitation of thinking in terms of communicating versus noncommunicating hydrocephalus and to clarify that disease classification should ultimately be linked to treatment efficacy and patient outcome and cure [17].

Conclusions

The ISHCSF (www.ishcsf.com) was founded in 2006 during an exciting meeting hosted by Carsten Wikkelsö in Gothenburg, Sweden. With HYDROCEPHALUS 2006, Carsten Wikkelsö fathered the concept and initiated a group dedicated to building an international society to provide a platform of exchange between senior and young researchers to assure continuity of education, clinical and basic research in hydrocephalus, CSF disorders and related fields. International and interdisciplinary collaboration should support guidelines, standardized research methods, and funding structures.

Since 2006, successful meetings have followed up on this. In Hanover, Germany, in 2008, the society was inaugurated, and the first “official” meeting of the ISHCSF was hosted in Baltimore, Maryland in 2009. In 2010, a meeting was held in Crete, Greece, in conjunction with the 5th International Hydrocephalus Workshop and the annual meeting of the Hellenic Neurosurgical Society. The society decided to implement the Anthony Marmarou Memorial Lecture, the first of which was held at the meeting, in Copenhagen, Denmark, in 2011.

Also, as an important step forward in this direction, Hydrocephalus 2012, held in Kyoto, Japan, is the first meeting officially combining with the bi-annual meeting of the IHIWG (International Hydrocephalus Imaging Working Group, http://ihiwg.org/).

The IHIWG was founded in 2010 to advance the study of hydrocephalus utilizing a full range of imaging modalities. As such, its members so far include a multi- and transdisciplinary team of neuroradiology, biophysics/engineering, basic science and clinical neurology and neurosurgery. The goal is to get these disciplines with special knowledge in their field to communicate to allow a more rapid evolution of the understanding in diagnosing hydrocephalus, and improve management and understanding of the neurological and pathophysiological sequelae of hydrocephalus across age groups. Meanwhile, this group consists of about 30 members of dedicated scientists. Two successful meetings have already been held, in April 2011 in Great Neck in New York, and a more recent one in association with the Annual Meeting of the American Society of Neuroradiology (ASNR) in April 2012 in Manhattan, NY, hosted by Professor William Bradley, Chairman, Department of Radiology, Professor of Radiology, UCSD School of Medicine, University of California, San Diego, USA. Yet only a few colleagues from neuroradiology usually attend focused meetings on hydrocephalus, and vice versa, hydrocephalus researchers usually do not attend neuroradiological meetings to enable face-to-face interdisciplinary contact and exchange. The main prospectus of the IHIWG meetings is implementation of evidentiary work and later guidelines for noninvasive assessment of hydrocephalus with neuroimaging, and recommendations for how to proceed and conduct clinical research studies in the field of neuroimaging of hydrocephalus.

Acknowledgments

I would like to acknowledge the collaboration with Dr. Jeffrey Rogg and Dr. Jerrold Boxermann, Department of Radiology, Division of Neuroradiology, Rhode Island Hospital and Warren Alpert Medical School of Brown University. Their contribution to imaging these patients and profound knowledge and expertise in understanding CSF diseases provided optimal diagnosis of pathology, and ultimately guided management and surgical approach in the presented patients and in many others. For collaboration in the patients’ care I would like to thank Dr. David Mandelbaum, Director of Child Neurology, Hasbro Children’s Hospital, Providence.