Management of Shunt Malfunction

The ASPN surgeons who answered the survey prefer 3 varieties of shunt valves: 41% primarily use differential pressure valves, 29% use differential pressure valves with an antisiphon device (ASD), and 27% mostly use programmable valves. A small percentage of surgeons (3%) prefer the flow-regulating Orbis-Sigma valve (OSV or OSV-II) as their primary choice (Fig. 1).

Primary shunt valve preference. The respondents reported which valve they predominantly use. Multiple-choice options included differential pressure valve, differential pressure valve with ASD, programmable valve, and flow-regulating OSV.

The majority of respondents (98%) agree that obstructive shunt failure occurs most commonly at the ventricular catheter, although a few (2%) report a higher incidence of distal catheter obstruction. Surgeons were then asked what, in their experience, was the most common source of ventricular catheter obstruction among those cited in the literature, that is, CSF debris, choroid plexus, ventricle collapse, and loculations. Most respondents (66%) consider ingrowth of the choroid plexus to be the most common cause of proximal shunt obstruction, whereas 18% favor CSF contents (for example, blood, cellular debris), 8% ventricular collapse, and < 1% loculations (Table 1). Ten surgeons reported other sources of ventricular obstruction: ependymal or scar tissue (8 respondents), undetermined cellular ingrowth (1 respondent), or some combination of the above categories (1 respondent). One surgeon reported uncertainty.

A total of 109 surgeons offered their opinions in free-text format regarding the most likely underlying cause of ventricular obstructions. These fell into 6 general categories: 1) small ventricle size, 2) catheter position, 3) choroid plexus migration or adherence, 4) siphoning or overdrainage, 5) accumulation of CSF debris, and 6) reactive or inflammatory processes. Several of these explanations fit into multiple domains. The most commonly cited explanation for proximal shunt obstruction involved adherence, growth, or migration of choroid plexus (41 comments), followed by siphoning or overdrainage (32 comments), poor catheter position (25 comments), and/or small ventricles (22 comments). Few surgeons cited CSF debris (12 comments) or reactive/inflammatory processes (5 comments) as the most likely cause. Representative free-text descriptions are shown in Table 2.

Management of Chronic Shunt Overdrainage

When asked to estimate the rate of chronic shunt overdrainage in their practices, most respondents (86%) reported that no more than 15% of their patients suffer from it (Fig. 2). Estimates ranged from 0% to 75%, with a median of 5%.

Prevalence of chronic shunt overdrainage. Respondents estimated the approximate percentage of patients in their practice with chronic shunt overdrainage. Responses were collected as any number between 0 and 100. Categorized estimates are shown, which range from 0% to 75%, with a median of 5%.

If a diagnosis of chronic overdrainage was made, most respondents reported managing the problem with a combination of shunt valve replacement for pressure incompatibility, addition of ASDs, or valve-setting adjustments (Fig. 3). Specifically, 45% of respondents reported often or always placing an ASD, 41% reported adjusting valve settings, and 30% reported replacing the valve. Fifty percent of surgeons who reported replacing the valve primarily used programmable valves with or without an ASD, although differential pressure with ASD (25%) and OSV (19%) were also popular valve choices (Fig. 4). Few surgeons (3%) said they use differential pressure valves alone. A small number reported placing a shunt assistant (2 respondents) or distal slit catheter (1 respondent). Overall, abdominal binders and placement of an additional valve or shunt assistant are unpopular overdrainage management strategies with more than two-thirds of surgeons reporting never or rarely using them (Fig. 3).

Treatment of chronic shunt overdrainage. Respondents were asked about their preferred treatment for chronic shunt overdrainage. Response rate was 95%–98%.

Valve preference for chronic shunt overdrainage. Surgeons who reported treating chronic CSF overdrainage with valve replacement (119 [92%] of 130) were asked their valve preference.

In an optional free-response box, some respondents suggested treating overdrainage by other methods, including cranial expansion (4 respondents) and endoscopic third ventriculostomy (ETV; 2 respondents). Moreover, in the optional free-response section, 3 surgeons expressed concerns about the definition of overdrainage. For example, one suggested that absolute overdrainage that results in postural headaches and subdural collections is rare and should be distinguished from the relative overdrainage that predisposes to SVS. However, others doubted whether shunt overdrainage exists at all (2 respondents).

Management of Chronic Headaches in Shunt Patients

The ASPN surgeons most often attribute chronic headaches in shunt patients to medical reasons, for example, tension or migraine. Forty-nine percent of respondents considered these reasons “very likely” or “extremely likely.” Stress, CSF underdrainage, and CSF overdrainage were not considered likely causes, with 18% or fewer respondents considering them “very likely” or “extremely likely” (Fig. 5). Overall, however, respondents seemed to consider overdrainage a more likely etiology of chronic headaches than underdrainage. Other offered explanations included altered CSF or vascular dynamics (5 comments). These included CSF pulsation absorption, hyperpulsatility in microvascular flow, and poor physiological match of the shunt to intracranial pressure (ICP) pulsations.

Causes of chronic headaches. Respondents were asked the most likely reasons for chronic headaches in patients with shunts. Response rate was 94%–97% for each category.

When shunt patients present with chronic headaches, most ASPN surgeons provide reassurance (51% “often” or “always”). Referral to a pain clinic is also used (24% “often” or “always;” Fig. 6). More aggressive measures such as shunt revision and valve replacement are less popular, with 65% and 57% of surgeons reporting “rarely” or “never” using those strategies, respectively. In an optional free-text box, 25 surgeons (19%) reported that they offer ICP monitoring to these patients. Surgeons also mentioned referral to neurology or headache clinics (18 respondents [14%]) and the use of other imaging studies to ensure shunt patency (7 respondents [5%]).

Treatment of chronic headaches. Respondents were asked how often they use each of the reported treatment strategies to manage chronic headaches in shunt patients. Response rate was 84%–88% for all categories.

Slit Ventricle Syndrome

For the purposes of this study, SVS was defined as “symptomatic intermittent ICP elevation with small ventricles,” as stated in the survey prior to all the SVS-related questions. Beyond patient history and imaging studies, 51% of respondents said they “often” or “always” diagnose SVS with the assistance of prolonged ICP monitoring (Fig. 7). Shunt tap and lumbar puncture are less commonly employed. In the optional free-response box, 5 surgeons suggested shunt exploration surgery. Four respondents suggested specific imaging studies, for example, CSF flow studies.

Tests used to diagnose SVS. Respondents were asked which diagnostic tests they employ to confirm the diagnosis of SVS (in addition to history and imaging studies). Response rate was 92%–97% for all categories.

To assess SVS treatment methodology, participants were given a list of treatment options and asked to select all the procedures they perform. An optional free-text box was also provided for respondents to add treatments not listed. In an effort to understand general treatment strategy for patients with SVS, participants were then asked to rank the treatments they selected from most often to least often used.

The most commonly selected treatments for managing patients with SVS among 127 respondents included ventriculoperitoneal shunt revision (88%), cranial expansion (57%), addition of an ASD (53%), and ETV (42%). The least popular treatments were placement of an additional valve or shunt assistance device (30%), placement of a lumboperitoneal (LP) shunt (25%), and abdominal binder (16%; Table 3). Other treatments included in the optional free-text box generally fit into the above categories, although 2 surgeons included medical treatments (for example, propranolol) in their registry of SVS treatments. The corresponding ranking of each treatment approach, compared with other treatments specific to each respondent, is also shown in Table 3.

Treatment methodology is shown in Table 4. Among surgeons who elect to revise the shunt for SVS, 62% rank it as their first-line treatment approach. Other first-line approaches include medical management (2 respondents) and cisternal catheter placement (2 respondents). Placement of an ASD is most often used as second-line therapy (34%). Endoscopic third ventriculostomy and cranial expansion are less commonly used. Among surgeons performing ETV, 28% rank it as third line, closely followed by cranial expansion (24%). When 4 or more treatments were selected (53 respondents), cranial expansion was the only significant fourth-line treatment (42%). Other treatments such as LP shunting, placement of an additional valve or shunt assistant, and abdominal binders are not significant first-, second-, third-, or fourth-line treatments as they were selected by fewer than 16% of respondents across each of these categories.

Survey participants were also asked why they think SVS occurs. Free-text responses from 98 respondents were reviewed and categorized. Responses generally fell into 4 groups: chronic CSF overdrainage (42 comments), abnormal brain compliance (35 comments), small skull size (22 comments), and shunt placement early in life (18 comments). Individual responses often fit into multiple domains. Table 5 shows representative free-text responses within each of these categories. Other comments included poor shunt valve design, altered venous pressure or compliance, abnormally elevated ICP pulsation, and ventricular adhesions.

Shunt Obstruction

Shunt obstruction is the most common problem encountered in the care of the patient with hydrocephalus. Prospective studies have shown shunt failure rates of up to 40% after 1 year and 50% after 2 years.¹⁶ Randomized controlled trials have demonstrated no difference in shunt longevity when the most commonly used valve types were compared.^1,8,16 Consistent with the lack of evidence for a superior shunt valve, surgeons of the ASPN are equally divided among 3 varieties of shunt valves: differential pressure valve, differential pressure valve with ASD, and programmable valve.

According to prospective studies, most shunt malfunctions are due to obstruction at the level of the ventricular catheter.^8,15,16 Other well-known, noninfectious causes of shunt malfunction include mechanical failure, CSF over- and underdrainage, ventricular loculations, and drainage site complications.^3,4 Our survey data are consistent with the literature suggesting that proximal obstruction accounts for most shunt malfunctions. However, survey respondents provide different explanations for such obstruction. While most (66%) consider choroid plexus to be the most common source of obstruction, a significant number of respondents (18%) consider CSF debris to be the main cause. Few respondents favored other causes. Free-text explanations of the underlying pathophysiology are similarly distributed, with many respondents citing adherence, growth, or migration of the choroid plexus, often in combination with poor catheter positioning or CSF siphoning.

Classically, the choroid plexus is cited as the dominant tissue obstructing the proximal catheter, a notion based on several early histological observations.^6,11 However, subsequent work has demonstrated that proximal shunt obstructions are primarily related to a variety of biological materials that lodge in ventricular catheters including, but not limited to, ependyma, connective tissue, neural tissue, macrophages, multinucleated giant cells, and aggregates of blood cells or debris.^7,17,23 In turn, clinical reports have shown that the choroid plexus is responsible for only a minority of ventricular catheter obstructions.^2,22,24 More recently, researchers investigating the underlying pathophysiology of proximal shunt obstruction have proposed a reactive and/or inflammatory process occurring in response to the foreign catheter material.^12–14 Respondent opinions are inconsistent with these data. Among 109 free-text explanations, only 5 surgeons cited reactive or inflammatory processes as the most likely cause of proximal catheter obstruction.

Chronic CSF Overdrainage

Shunt overdrainage is another well-known complication of shunt-dependent hydrocephalus.⁵ Although acute shunt overdrainage is a recognized but uncommon CSF shunt complication, our survey aims to investigate current understandings of chronic overdrainage. Most respondents consider chronic overdrainage a rare problem among their patients, while a few contend that most of their patients suffer from it. Reports in the literature suggest that about 10% of patients suffer from overdrainage.^16,18 More than half of the survey respondents’ estimates are well below this figure. These discrepancies could be attributed to differences in definition and diagnosis. For example, some surgeons may use more stringent criteria to define overdrainage, such as the presence of specific symptoms (for example, positional headaches) or radiographic findings (for example, subdural hematomas).

Despite the lack of evidence for valve superiority in treating shunt overdrainage,¹ many surgeons elect to replace or adjust the shunt hardware in patients suffering from this condition. In this context, respondents tend to use predominantly programmable and flow-regulating valves. However, given the apparent variability in the definition and diagnosis of overdrainage, the results are difficult to interpret and highlight the need to develop better classifications to describe this problem.

Chronic Headaches in Shunt Patients

Patient surveys demonstrate a startlingly high prevalence of severe headaches among those with CSF shunts (29% of children and 42%–44% of adolescents and young adults),²¹ but few studies have investigated the pathophysiology of this challenging problem. Results of the present survey show that surgeons consider medical reasons to be the most likely cause of chronic headaches in this patient population. Yet, there is no evidence in the literature to support this contention. Moreover, despite suggestions in the literature that have linked headaches to CSF overdrainage,^10,18,20,25 our survey data show that most surgeons do not consider this a likely cause. Interestingly, several surgeons suggest that changes in CSF flow dynamics, particularly pulsatility, are responsible for the headaches.

General headache treatment strategies among the respondents are consistent with opinions about headache etiology, with reassurance and referral to neurology or pain clinics being the most common practice. Aggressive invasive and surgical (shunt-related) measures to address headaches have been suggested by some authors^20,21 but were uncommon among survey respondents. Considering that the incidence of headaches in the shunt patient population is probably much higher than average,²¹ it is troubling that no evidence-based methods exist to address this problem. Furthermore, although medical and psychosocial interventions may be effective in the non-shunt population, it is unclear if such interventions succeed in shunt patients. It is also logical to assume that surgeons at medical centers with comprehensive headache clinics are more likely to develop referral patterns to multidisciplinary headache clinics than surgeons at centers lacking these services. Still, no evidence-based methods exist to address headaches in shunt patients. While the present survey does provide a rigorous comparison of general management strategies among survey participants, it raises important questions to address in future studies.

Slit Ventricle Syndrome

Slit ventricle syndrome is a well-recognized complication of shunt-dependent hydrocephalus. While SVS is considered a rare complication of CSF shunting,^3,20 patients with SVS undergo a disproportionate share of shunt revisions. The precise definition of SVS is probably physician and/or institution dependent. Using chronic ICP monitoring data, Rekate has described 5 different classifications of the condition, which include intracranial hypotension, intermittent proximal obstruction with small ventricles, shunt failure without ventricular enlargement, increased ICP with a working shunt, and shunt-related migraine.^19,20 For the current survey, SVS was defined as symptomatic intermittent ICP elevation with small ventricle sizes. This definition is commonly used in the literature and considered to be the most common variant among the subclassifications proposed by Rekate.

As with proximal shunt obstruction, there is a wide range of opinions about the cause of SVS. Some respondents suggest that overdrainage is a central mechanism, but others inculpate altered compliance and duration of shunting. Data from the present survey indicate that most surgeons diagnose the condition with the assistance of ICP monitoring. Treatment philosophy among respondents generally involves, in order of preference, shunt revision and ASD placement, followed by ETV and cranial expansion. Yet, given the complexity of caring for these patients, practice patterns are difficult to compare and quite variable.

Study Limitations

This opinion-based survey has several limitations. Respondents represent only a portion (64%) of the ASPN membership. Accordingly, presented findings are subject to nonresponse bias. Additionally, the reported philosophies and management styles represent those of board-certified pediatric neurosurgeons predominantly practicing at large medical centers. Thus, the data and opinions may or may not reflect all North American pediatric neurosurgical care. Furthermore, despite efforts to create clear and objective questions, the words and phrases in the questionnaire could have been interpreted differently by respondents. Importantly, multiple-choice questions in the survey included a limited number of options, which underestimates the complexity of these clinical issues. While optional free-text boxes were included for most questions, only a few surgeons used this feature. Lastly, interpreting clinical management questions is challenging, especially since, in the absence of evidence-based guidelines, most surgeons provide individualized approaches to their patients’ care. Despite these limitations, the study provides an overview of expert attitudes and opinions on shunt-dependent hydrocephalus and highlights the need to improve our understanding and management of these critical problems. Accordingly, the discrepancies illustrated in this report should be considered in the design of future studies and registries.

Online-Only Content

Supplemental material is available with the online version of the article.

Questionnaire. https://thejns.org/doi/suppl/10.3171/2017.2.PEDS16265.