Factors associated with 30-day ventriculoperitoneal shunt failure in pediatric and adult patients

Ian A. Anderson FRCS (Neuro.Surg), Louise F. Saukila MRCS, James M. W. Robins MRCS, Christopher Y. Akhunbay-Fudge MRCS, John R. Goodden FRCS (Neuro.Surg), Atul K. Tyagi FRCS(SN), Nick Phillips PhD, FRCS (SN) and Paul D. Chumas MD, FRCS (SN)
View More View Less
  • Department of Neurosurgery, Leeds General Infirmary, Leeds, United Kingdom
Full access

OBJECTIVE

The aim of this study was to provide a comprehensive benchmark of 30-day ventriculoperitoneal (VP) shunt failure rates for a single institution over a 5-year study period for both adult and pediatric patients, to compare this with the results in previously published literature, and to establish factors associated with shunt failure.

METHODS

A retrospective database search was undertaken to identify all VP shunt operations performed in a single, regional neurosurgical unit during a 5-year period. Data were collected regarding patient age, sex, origin of hydrocephalus, and whether the shunt was a primary or secondary shunt. Operative notes were used to ascertain the type of valve inserted, which components of the shunt were adjusted/replaced (in revision cases), level of seniority of the most senior surgeon who participated in the operation, and number of surgeons involved in the operation. Where appropriate and where available, postoperative imaging was assessed for grade of shunt placement, using a recognized grading system. Univariate and multivariate models were used to establish factors associated with early (30-day) shunt failure.

RESULTS

Six hundred eighty-three VP shunt operations were performed, of which 321 were pediatric and 362 were adult. The median duration of postoperative follow-up for nonfailed shunts (excluding deaths) was 1263 days (range 525–2226 days). The pediatric 30-day shunt failure rates in the authors’ institution were 8.8% for primary shunts and 23.4% for revisions. In adults, the 30-day shunt failure rates are 17.7% for primary shunts and 25.6% for revisions. In pediatric procedures, the number of surgeons involved in the operating theater was significantly associated with shunt failure rate. In adults, the origin of hydrocephalus was a statistically significant variable. Primary shunts lasted longer than revision shunts, irrespective of patient age.

CONCLUSIONS

A benchmark of 30-day failures is presented and is consistent with current national databases and previously published data by other groups. The number of surgeons involved in shunt operations and the origin of the patient’s hydrocephalus should be described in future studies and should be controlled for in any prospective work. The choice of shunt valve was not a significant predictor of shunt failure. Most previous studies on shunts have concentrated on primary shunts, but the high rate of early shunt failure in revision cases (in both adults and children) is perhaps where future research efforts should be concentrated.

ABBREVIATIONS BMI = body mass index; CI = confidence interval; EVD = external ventricular drain; IIH = idiopathic intracranial hypertension; NPH = normal pressure hydrocephalus; OR = odds ratio; VP = ventriculoperitoneal.

OBJECTIVE

The aim of this study was to provide a comprehensive benchmark of 30-day ventriculoperitoneal (VP) shunt failure rates for a single institution over a 5-year study period for both adult and pediatric patients, to compare this with the results in previously published literature, and to establish factors associated with shunt failure.

METHODS

A retrospective database search was undertaken to identify all VP shunt operations performed in a single, regional neurosurgical unit during a 5-year period. Data were collected regarding patient age, sex, origin of hydrocephalus, and whether the shunt was a primary or secondary shunt. Operative notes were used to ascertain the type of valve inserted, which components of the shunt were adjusted/replaced (in revision cases), level of seniority of the most senior surgeon who participated in the operation, and number of surgeons involved in the operation. Where appropriate and where available, postoperative imaging was assessed for grade of shunt placement, using a recognized grading system. Univariate and multivariate models were used to establish factors associated with early (30-day) shunt failure.

RESULTS

Six hundred eighty-three VP shunt operations were performed, of which 321 were pediatric and 362 were adult. The median duration of postoperative follow-up for nonfailed shunts (excluding deaths) was 1263 days (range 525–2226 days). The pediatric 30-day shunt failure rates in the authors’ institution were 8.8% for primary shunts and 23.4% for revisions. In adults, the 30-day shunt failure rates are 17.7% for primary shunts and 25.6% for revisions. In pediatric procedures, the number of surgeons involved in the operating theater was significantly associated with shunt failure rate. In adults, the origin of hydrocephalus was a statistically significant variable. Primary shunts lasted longer than revision shunts, irrespective of patient age.

CONCLUSIONS

A benchmark of 30-day failures is presented and is consistent with current national databases and previously published data by other groups. The number of surgeons involved in shunt operations and the origin of the patient’s hydrocephalus should be described in future studies and should be controlled for in any prospective work. The choice of shunt valve was not a significant predictor of shunt failure. Most previous studies on shunts have concentrated on primary shunts, but the high rate of early shunt failure in revision cases (in both adults and children) is perhaps where future research efforts should be concentrated.

ABBREVIATIONS BMI = body mass index; CI = confidence interval; EVD = external ventricular drain; IIH = idiopathic intracranial hypertension; NPH = normal pressure hydrocephalus; OR = odds ratio; VP = ventriculoperitoneal.

Ventriculoperitoneal (VP) shunt failure is a source of frustration and anxiety for patients, caregivers, and neurosurgeons alike. The reason for this dissatisfaction is clear: shunt insertion is a commonly performed operation, and collectively neurosurgeons now have more than half a century of experience in performing shunt surgery, yet shunts remain as prone to failure now as they were decades ago.4,25

During this time, different catheters and valves have been introduced, but without high-level evidence of any particular device’s superiority.7,15 Similarly, there have been numerous studies examining the effects that surgical adjuncts (including neuronavigation) have on shunt failure rates, but systematic review of these papers has again failed to establish high-level evidence to support their routine usage.9,10

If shunt survival rates are to be further investigated and hopefully improved upon, it is important to identify which episodes of shunt failure should be considered relatively avoidable and which should be viewed as relatively unavoidable, given the limitations of current shunt technology. To this end, a group from our institution (Al-Tamimi et al.) proposed that a 30-day shunt failure rate represents a reasonable and valid barometer of surgical outcomes (and therefore of relatively avoidable shunt failures) and that this should be used as a separate outcome measure in future shunt trials.2

To validate the data produced in future trials (and in the Al-Tamimi et al. paper) it is important that sufficient background data be published with which others can compare their results. In this study we present a single-center experience of all adult and pediatric VP shunts inserted over a 5-year period, including 30-day and 1-year failure rates, as well as examine our own data for any significant factors associated with shunt failure.

Methods

Study Population

A retrospective database search was undertaken to identify all VP shunt operations performed in a single, regional neurosurgical center over a 5-year period (January 1, 2010, to December 31, 2014). The date range was selected so as not to overlap with previously published multicenter data that included patients from our institution.2 To be able to draw direct comparison with previously published data, patients were classified as pediatric if they were less than 19 years old on the date of surgery.2 Ventriculoatrial shunts, ventriculopleural shunts, lumboperitoneal shunts, external ventricular drains (EVDs), and endoscopic third ventriculostomy were all excluded from the study.

There is no standardized operating protocol for shunt insertion at our institution, but all patients undergoing VP shunt surgery are given intravenous flucloxacillin and gentamicin prophylaxis at induction of anesthesia (teicoplanin if there is a penicillin allergy). It is routine to use Bactiseal (DePuy Synthes) proximal and distal catheters for all VP shunts, to have the minimum number of people in the operating theater during a case, and for the theater doors to carry signs forbidding entry during the case.

Data Collection

Data were collected regarding patient age, sex, origin of hydrocephalus, and whether the shunt was a primary shunt (i.e., first-ever VP shunt; previous external ventricular drain, lumboperitoneal shunt or endoscopic third ventriculostomy allowed) or a secondary shunt (all subsequent shunt insertions or revisions after the primary shunt). In cases in which a shunt failed and a subsequent shunt was inserted during the study period, that subsequent shunt was entered as a new entity within our study. Operative notes were used to acquire data on the type of valve inserted, which components of the shunt were adjusted/replaced (revision cases), level of seniority of the most senior surgeon who participated in the operation, and number of surgeons involved in the operation (coded as single surgeon or multiple surgeons).

The diagnosis resulting in shunt insertion was categorized as congenital malformation, arachnoid cyst, communicating hydrocephalus, dysraphism, vascular (secondary to hemorrhage), idiopathic intracranial hypertension (IIH), infection, mass lesion, normal pressure hydrocephalus (NPH), or trauma.

All available electronic records (electronic operation notes database, electronic patient notes, electronic imaging software, and the local microbiology and biochemistry results database) were searched for evidence of subsequent shunt failure and the reason for the failure, if known. Shunt failure was defined as “any event that culminated in an operation to revise the shunt.” In most cases, the reason for shunt failure was apparent in the operative note for the revision of said shunt.

Where there was no evidence to suggest shunt failure, it was assumed that the inserted shunt was still functioning at the time of the study, unless the patient had died. When interrogated for any information, our databases automatically flag patients who have died, and the date and cause of death was noted for these patients also. For cases in which the valve inserted was documented, this information was also noted.

An “intention to treat” attitude was used when considering infection in this study. Any shunt revised because of preoperative clinical suspicion of infection and resulting in the shunt being removed and an EVD inserted was classified as an infected case, even if no infection was ever proven. Examples of shunts coded as infected would therefore include not only those where pus was identified or where wounds had become erythematous, but would also include cases in which the patient was pyrexial or had abnormal inflammatory markers and in which a decision was made to operate because of a suspicion of shunt infection. For all cases of reoperation due to shunt infection, the microbiology database was checked for 5-day CSF culture results from shunt taps or subsequent intraoperative samples from EVD insertion. Where a causative organism was identified, cases were termed “confirmed infection,” and where cultures were negative, “suspected infection.” For the purposes of this study, both confirmed and suspected infected cases were still classified as shunt failures due to infection.

Finally, for each operation in which a ventricular catheter was inserted or revised, postoperative imaging (where available) was assessed for grade of placement, using a recognized grading system (Table 1).10

TABLE 1.

Grading of ventricular catheter placement

Catheter PlacementDefinition
Grade 1Catheter tip floating in CSF equidistant from ventricular walls, away from choroid & a straight trajectory from the burr hole
Grade 2Catheter tip touching ventricle wall or choroid
Grade 3Part of catheter tip w/in parenchyma or failure to cannulate ventricle completely

From Hayhurst et al.10

Statistical Analysis

Statistical analysis was undertaken by an independent statistician (Statsconsultancy Ltd). As per the guidelines of The Leeds (East) Research Ethics Committee, ethical committee approval is not required for an evaluation of service such as this (the collation and analysis of patient data collected on a routine basis during normal clinical care), on the understanding that the data are pooled and therefore nonidentifiable.

Results

Six hundred eighty-three VP shunt operations were identified, of which 321 were classified as pediatric (patients < 19 years old on the date of surgery) and 362 as adult. The median duration of postoperative follow-up for nonfailed shunts (excluding deaths) was 1263 days (range 525–2226 days). Thirty-nine patients died during the study period (10 in the pediatric cohort, 29 in the adult). For each death, the cause of death was reviewed and each of these cases was judged by the authors not to be directly related to the shunt surgery or to shunt failure. Postoperative imaging was available such that ventricular catheter placement grading was possible for 487 of the shunts inserted. Table 2 summarizes the baseline patient and procedural details for both the pediatric and adult cohorts.

TABLE 2.

Baseline patient and surgical characteristics of the pediatric and adult groups

VariablePediatric Patients (n = 321)Adult Patients (n = 362)
Mean age ± SD (yrs)4.2 ± 5.247.3 ± 18.3
Sex
 Female141 (43.9)192 (53.0)
 Male180 (56.1)170 (47.0)
Origin of hydrocephalus
 Congenital malformation71 (22.2)31 (8.6)
 Arachnoid cyst8 (2.5)2 (0.6)
 Communicating0 (0)6 (1.7)
 Dysraphism85 (26.6)6 (1.7)
 Vascular87 (27.2)93 (25.8)
 IIH4 (1.3)18 (5.0)
 Infection17 (5.3)34 (9.4)
 Mass lesion39 (12.2)124 (34.3)
 NPH0 (0)32 (8.9)
 Trauma9 (2.8)15 (4.2)
Type of shunt
 Primary126 (39.3)187 (51.7)
 Secondary195 (60.7)175 (48.3)
Valve inserted*
 Codman fixed56 (21.9)260 (92.5)
 Programmable8 (3.1)8 (2.8)
 Medtronic fixed68 (26.6)1 (0.4)
 Miethke gravitational111 (43.3)9 (3.2)
 Integra OSV II flow regulated13 (5.1)3 (1.1)
Consultant involved
 No161 (50.2)301 (83.1)
 Yes160 (49.8)61 (16.9)
No. of surgeons
 Multiple236 (73.5)249 (68.8)
 Single85 (26.5)113 (31.2)

Data given as number of patients (%), except for age.

Where applicable; no new valve inserted in some revision cases.

There were 65 operations in the pediatric group for which either no valve type was listed or for whom no valve was inserted during the study period (i.e., they had a revision of a shunt in which the valve was not exchanged). The log-rank test was used to compare the overall time to failure between the different valves for the remaining 256 procedures; however, consistent with previously published series, no significant differences were identified (Supplementary Tables 1 and 2, Supplementary Fig. 1). Specifically, there was no advantage in shunt survival when gravitational valves were used as compared with fixed-pressure valves. Of the 281 cases in the adult group for which information on the shunt valve inserted was available, 260 (92.5%) were Codman fixed-pressure valves, making further comparison futile.

The surgeon-documented causes of the shunt failure for revisions within 30 days were noted (Supplementary Table 3) and were found to be broadly consistent with that which has previously been published.

Pediatric Shunt Procedures

There were a total of 321 pediatric shunt procedures identified in the initial search, however, there were 4 patients who died of non–shunt-related failure within 30 days of surgery and a further 4 patients who died of non–shunt-related failure between 30 days and 1 year after surgery. These patients were therefore excluded from the relevant analyses, leaving 317 patients in the 30-day analysis and 313 in the 1-year analysis. The mean age of the patients undergoing pediatric shunt surgery in this series was 4.2 years (similar to previously published series3), but the mean age for primary shunts was substantially lower (2.0 years) than for revisions (5.6 years). In 180 (56.1%) of the pediatric operations, the child was male. One hundred thirty four (41.7%) of our pediatric cases were performed for patients less than 2 years old.

The overall 30-day shunt failure rate for pediatric shunts was 17.7% (56/317; Table 3). When subdivided, the 30-day shunt failure rate for primary shunts was 8.8% (11/125) compared with 23.4% (45/192) for revision operations (p = 0.001). The overall 1-year shunt failure rate was 40.9% (128/313); 26.0% (32/123) for the primary cohort and 50.5% (96/190) for the revision group (p < 0.001).

TABLE 3.

Univariate analysis of factors associated with 30-day shunt failure (pediatric patients)

Variable30-Day Failures (%)OR (95% CI)p Value
Age1.04 (0.98–1.09)0.19
Sex0.83
 Female24/140 (17.1)1
 Male32/177 (18.1)1.07 (0.60–1.91)
Origin of hydrocephalus*0.28
 Congenital malformation9/71 (12.7)1
 Arachnoid cyst1/8 (12.5)0.98 (0.10–8.96)
 Dysraphism23/85 (27.1)2.56 (1.09–5.96)
 Vascular15/87 (17.2)1.43 (0.59–3.51)
 Infection3/17 (17.6)1.48 (0.35–6.17)
 Mass lesion4/36 (11.1)0.86 (0.25–3.01)
 Trauma1/8 (12.5)0.98 (0.11–8.96)
Type of shunt0.001
 Primary11/125 (8.8)1
 Secondary45/192 (23.4)3.17 (1.57–6.41)
Consultant involved0.02
 No36/158 (22.8)1
 Yes20/159 (12.6)0.49 (0.27–0.89)
No. of surgeons0.006
 Multiple33/234 (14.1)1
 Single23/83 (27.7)2.33 (1.27–4.28)
Catheter placement0.05
 Grade 14/47 (8.5)1
 Grade 219/100 (19.0)2.52 (0.86–7.83)
 Grade 316/60 (26.7)3.91 (1.21–12.6)

For the categorical variables, the number and percentage of patients with a failure within 30 days are reported. Additionally, the effect size of each factor on the outcome is given in the form of ORs. These give the odds of a failure within 30 days in one situation relative to another. For the categorical variables, these represent the odds of failure in each category relative to the odds in a baseline category. For the 1 continuous variable (age), the OR represents the change in the odds of failure for a 1-year increase in age. Boldface type indicates statistical significance.

Analysis of origin of hydrocephalus excluded 4 patients with IIH, who experienced no failures. These patients were excluded due to mathematical problems calculating ORs.

The log-rank test was used to compare the overall time to failure between the two shunt types. The difference was found to be highly statistically significant (p < 0.001) and is represented in a Kaplan-Meier plot (Fig. 1).

Fig. 1.
Fig. 1.

Kaplan-Meier survival curves for primary and secondary shunts in pediatric patients. Figure is available in color online only.

Univariate analysis was undertaken to explore the association between a number of patient/shunt characteristics and 30-day shut failure; the results are summarized in Table 3. Three factors were identified as being associated with significantly increased odds of shunt failure: secondary shunts (as described previously), lack of consultant involvement in the operating theater, and single-surgeon operations. Consultant involvement was associated with a lower shunt failure rate in the pediatric group, with the risk of 30-day failure being 12.6% (20/159) in which consultants were involved, as compared with 22.8% (36/158) in which procedures were performed without consultant involvement (odds ratio [OR] 0.49, 95% confidence interval [CI] 0.27–0.89). When subdivided into primary (OR 0.60, 95% CI 0.14–2.46) and secondary (OR 0.65, 95% CI 0.31–1.36) procedures, the result ceased to achieve statistical significance.

Performing pediatric shunt surgery as a single surgeon was associated with a significantly higher failure rate at 30 days when multiple surgeons were involved (14.1%, 33/234), rising to 27.7% (23/83) if surgery was performed by a solo surgeon (OR 2.33, 95% CI 1.27–4.28). This effect was less pronounced for primary shunts (OR 1.59, 95% CI 0.31–8.11) than it was for secondary shunts (OR 6.03, 95% CI 2.83–12.87).

A backward selection procedure was performed to retain only the statistically significant variables in a final model. The results of this multivariate analysis suggested that, in fact, only type of shunt (primary/secondary) and number of surgeons involved (single/multiple) were independently associated with 30-day failure rates. The multivariate analysis found the odds of shunt failure for secondary procedures are 2.75 times the odds of primary procedures (95% CI 1.35–5.65, p = 0.006) and that the odds of shunt failure for shunts inserted by a single surgeon are 1.87 times the odds for shunts inserted by multiple surgeons (95% CI 1.00–3.49, p = 0.05). After adjusting for each of these factors, there was no additional significant effect of a consultant being involved.

Adult Shunt Procedures

There were a total of 362 adult shunt procedures. Four patients died of non–shunt-related failure within 30 days of surgery and an additional 19 patients died of non–shunt-related failure between 30 days and 1 year after surgery. These exclusions left 358 patients in the 30-day analysis and 339 in the 1-year analysis.

The overall 30-day shunt failure rate for adult shunts was 21.5% (77/358; Table 4). When subdivided, the 30-day shunt failure rate for primary shunts was 17.7% (33/186) and 25.6% (44/172) for revision operations, but this difference did not quite reach statistical significance (p = 0.07). The overall 1-year shunt failure rate was 34.8% (118/339): 27.7% (48/173) in the primary cohort and 42.2% (70/166) in the revision group (p = 0.005). When the log-rank test was used to compare the overall time to failure between the two shunt types, the difference was found to be highly statistically significant (p = 0.002) and is represented in a Kaplan-Meier plot (Fig. 2).

TABLE 4.

Univariate analysis of factors associated with 30-day shunt failure (adult patients)

Variable30-Day Failures (%)OR (95% CI)p Value
Age*0.89 (0.78–1.03)0.11
Sex0.12
 Female47/190 (24.7)1
 Male30/168 (17.9)0.66 (0.40–1.11)
Origin of hydrocephalus<0.001
 Congenital malformation2/39 (5.1)1
 Communicating1/6 (16.7)3.70 (0.8–48.6)
 Vascular13/93 (14.0)3.01 (0.64–14.0)
 IIH5/18 (27.8)7.11 (1.23–41.2)
 Infection15/34 (44.1)14.6 (3.02–70.6)
 Mass lesion34/120 (28.3)7.31 (0.25–3.01)
 NPH3/32 (9.4)1.91 (0.30–12.2)
 Trauma4/15 (26.7)6.73 (1.08–41.8)
Type of shunt0.07
 Primary33/186 (17.7)1
 Secondary44/172 (25.6)1.60 (0.96–2.65)
Consultant involved0.97
 No64/298 (21.5)1
 Yes13/60 (21.7)1.01 (0.52–1.98)
No. of surgeons0.97
 Multiple53/247 (21.5)1
 Single24/111 (21.6)1.01 (0.59–1.74)
Catheter placement0.27
 Grade 16/39 (15.4)1
 Grade 235/166 (21.1)1.47 (0.57–3.78)
 Grade 320/71 (28.2)2.16 (0.78–5.93)

Boldface type indicates statistical significance.

Odds ratio represented for a 10-year increase in age.

Due to the small number of patients in the arachnoid cyst and dysraphism categories, these were included in the congenital malformation category for analysis.

Fig. 2.
Fig. 2.

Kaplan-Meier survival curves for primary and secondary shunts in adult patients. Figure is available in color online only.

Univariate analysis was undertaken to explore the association between a number of patient/shunt characteristics and 30-day shut failure; these results are summarized in Table 4. Only one factor (origin of hydrocephalus; p < 0.001) was identified as being associated with significantly increased odds of 30-day shunt failure. Failure was lowest in patients whose hydrocephalus was due to congenital malformation (5.1% failed within 30 days) and NPH (9.4% failed within 30 days). The highest failure rates were observed in patients whose shunts were inserted for hydrocephalus secondary to infection (44.1%). The results of the multivariate analysis confirmed that only origin of hydrocephalus was a statistically significant factor in 30-day shunt failure.

Additional Analysis

During the study period, there were a total of 25 shunt revisions due to suspected or confirmed infection: 5 in the pediatric group and 20 in the adult group. Of these 25 shunts, an organism was cultured in 72% (18/25) of cases and the most common organism identified was coagulase-negative Staphylococcus (10 cases). The median interval from shunt insertion to revision in infected cases was 26 days (range 0–942 days). At 30 days, the infection rates were 2.5% (8/317) in the pediatric and 1.4% (5/358) in the adult groups. These rates rose to 6.4% (20/313) in the pediatric and 3.2% (11/339) in the adult groups at the 1-year mark.

Discussion

This study has demonstrated that the 30-day shunt failure rate for pediatric shunts inserted at our institution over a 5-year period was 8.8% for primary shunts and 23.4% for revision operations. These results are comparable to the rates that have been quoted for the United Kingdom and Ireland as a whole.2 For adult patients, the 30-day shunt failure rate was 17.7% for primary shunts and 25.6% for revisions.

There are numerous previously published papers that have focused on VP shunt failure and the causes thereof, but few have quoted a 30-day failure rate, especially where the adult population is concerned. A summary of some of the data that is available for early shunt failure rates is displayed in Table 5, but it is difficult to draw direct comparisons because several of the studies reported results for heterogeneous cohorts containing either pediatric and adult patients or containing primary and revision shunt cases.

TABLE 5.

Previously published series of shunt failure rates

Authors & YearShunt Failure Rates (%)
PrimarySecondary 
30-Day1-Year30-Day1-Year 
Current study
 Pediatric8.826.023.450.5
 Adult17.734.825.642.2
Al-Tamimi et al., 2014
 Pediatric13.328.823.042.3
Piatt, 2014
 Pediatric7.410.9–16.2*
Park et al., 2015
 All ages9.6
Venable et al., 2016
 All ages15.2§
Drake et al., 1998
 Pediatric1439
Kestle et al., 2000
 Pediatric1634–42**
Khan et al., 2015
 Adult11
Reddy et al., 2014
 All ages29

Secondary procedures subdivided by which shunt component was revised.

Outcomes quoted for 1 month, rather than 30 days.

Includes 27.7% revision cases.

Includes 23.3% primary insertions.

Extracted 30-day failure rates.

Thirty-four percent in nonendoscopic cohort, 42% in endoscopic cohort.

This study validates the previously described finding that the failure rates for revision shunts are worse than those for primary shunts, something that was true in both our pediatric and adult cohorts.2 Clearly, future studies examining shunt survival should divide patients in a similar fashion to allow for fair comparisons to be drawn.

Origin of Hydrocephalus

The underlying cause for hydrocephalus was shown to have a significant association with 30-day shunt failure rates in our adult population, but in the pediatric cohort the difference did not achieve statistical significance. That etiology is associated with outcome in VP shunt surgery is logical and has previously been described.14,27 Clearly, the reasons for origin of hydrocephalus affecting VP shunt revision rates will be multifactorial. While knowledge of etiology-related failure rates may aid clinicians in appropriate patient counseling, the primary cause of the hydrocephalus will always represent a nonmodifiable risk factor.

Number of Surgeons and Consultant Involvement

In addition to primary versus revision surgery, the results of the multivariate analysis suggest that only the number of surgeons performing the procedure is independently associated with 30-day failure rates in pediatric shunt operations. Many shunt operations are performed acutely, during off hours, and surgery performed during off hours may have poorer outcomes, although this issue remains contentious.1,8,19,27 Off-hours neurosurgical operations in our hospital are (in general) more likely to have been performed by the single neurosurgical trainee on-call. It would therefore be likely that there would have been a greater number of off-hours cases in the single surgeon group, however, data regarding time of surgery was not collected and therefore this hypothesis cannot be verified using the available data.

The insertion of a VP shunt system typically requires a minimum of two separate incisions/approaches (cranial and abdominal). If two surgeons are in the operating theater together, it is natural for them to each perform one of these components of the operation simultaneously. The consequent reduction in operative time would potentially reduce the risk of infection.16,23 Furthermore, if surgeons have a good working relationship, they will operate in a collaborative fashion at potentially difficult stages in the procedure, such as holding retractors or clips during the abdominal approach, improving vision for the operator.

Direct pediatric neurosurgical consultant involvement in operations was also associated with a better outcome at 30 days in univariate analysis, but this was not proven in the multivariate analysis. In adult shunt operations, consultant involvement in the procedure had no effect on outcome and, interestingly, neither did the number of surgeons.

An association between pediatric neurosurgical consultant involvement and better shunt survival has been previously described, specifically for revision shunt procedures.2 Furthermore, there is evidence to suggest that more experienced surgeons have better 6-month shunt survival rates for primary shunts than do their less-experienced colleagues.6 The question of consultant involvement remains debatable for any operation; it will always be argued that consultants would tend to be more actively involved in the management of more difficult cases and therefore their own success rates may be adversely affected by selection bias.

Choice of Valve

The Shunt Design Trial remains the best-known comparative study between valve types (standard differential pressure, Delta [PS Medical-Medtronic], and Sigma [NMT Cordis] valves). Both short- and long-term follow-up from this multicenter, randomized trial have failed to demonstrate a significant difference between the valve types.7,12 Another randomized trial found no difference in outcome between a simple, inexpensive Slit N Spring valve shunt and the Codman-Hakim Micro Precision valve, a device that costs more than 18 times more.29 It is perhaps not surprising, therefore, that in our study there was no difference between valve types.

Shunt Catheter Position

Hayhurst et al. offer a simple, user-friendly grading system for assessing ventricular catheter placement, but it should be noted that their results did not achieve statistical significance.10 In our series, there was a trend that ventricular catheter position is associated with 30-day outcome but the result did not achieve statistical significance in either age cohort.

Given the association between poor catheter placement and shunt failure, it would appear logical to presume that neuronavigation would improve catheter placement accuracy and thereby shunt failure rates, with poorly placed catheters being more susceptible to blockage than those that are free from parenchymal or ependymal contact. In fact, the use of navigation did not improve shunt failure rates in the Hayhurst et al. paper, nor in a more recent systematic review and meta-analysis.10,20

Thirteen of the 77 adult shunt failures were due to abdominal catheter malposition (Supplementary Table 3). This would appear to represent a high rate for this particular complication and further investigation is warranted locally to ascertain the reasons for this finding. Body habitus might be a factor in distal complications. Our study did not collect data on patient body habitus specifically, but the body mass index (BMI) was calculated post hoc for these 13 cases. One patient had a normal BMI (20.9 kg/m2), 2 were overweight (25–29.9 kg/m2), and the remaining 10 cases were obese (≥ 30 kg/m2). While the BMI for the other patients in the series was not calculated, it is possible that body habitus does have a role in abdominal shunt failure rates.

Infection

In the pediatric population a 30-day infection rate has previously been quoted at 2.0%, which was lower than our own pediatric 30-day infection rate of 3.2% using our definition of shunt infection. At 1 year, the infection rates in the literature are more readily described but are also more variable. The finding that our 1-year pediatric infection rate (4.9%) is higher than our adult rate (3.2%) is consistent with that which has been previously described.12,18 Clearly the definition of infection is key to calculating infection rates and clear and consistent criteria should be used when declaring these in the literature. Previous authors have described potential definitions for shunt infection.12,22,26

From the patient’s perspective, the morbidity (and increased length of stay) comes from having the shunt removed, an EVD inserted, and being exposed to a second operation, irrespective of the subsequent CSF results. We therefore suggest that a case should be classified as failure due to infection whenever a shunt is revised or removed based upon the suspicion of infection, irrespective of whether the CSF cultures are positive or not.

Numerous risk factors have been identified as associated with shunt infection.15,17 It has also been reported that the inception of specialist hydrocephalus networks and standardized operating protocols can be effective in reducing the incidence of VP shunt infection.12,13 There is also a current randomized controlled trial that is investigating the potential benefits of silver- and antibiotic-impregnated shunt tubing systems.11

Limitations of the Study

The 30-day results of 5 years of VP shunts are presented to add to the available published data and to provide a benchmark for future studies. As with any retrospective nonrandomized series, our results are subject to bias and any differences between groups may be subject to various confounding factors, not least of which is case selection.

Furthermore, for some categories such as IIH and NPH, our total number of patients is small, which prevented detailed statistical analysis. It could be argued that these patients should be excluded or analyzed separately due to their unique problems (small ventricles in the former and difficulty ascertaining shunt failure in the latter).

The data collected were based on searching all available clinical databases and electronic records, but there remains the potential for operations that were incorrectly coded and therefore may have been missed from the analysis. Certainly, the quality of coding with the National Health Service hospitals has previously been found to be suboptimal.5 There may also be patients in whom we inserted a shunt who subsequently experienced shunt failure at another hospital and therefore we did not capture said failure episode in our results.

In this study, we included all shunt operations occurring within our department to most accurately represent our performance. The decision that once a shunt had failed, the subsequent shunt was then added as a separate entity in the database means that while our series comprises 683 shunt procedures, there will be some patients in this series who have contributed multiple shunt revisions to the failure rates. These “problem patients” could potentially skew the failure rates, but it was not felt representative to exclude cases after first failure.

In retrospect, it would have been useful to compare results for shunts performed within normal working hours and those performed overnight to see whether the outcomes were affected, but these data were not part of our data collection tool.

Conclusions

In this paper, we present a single-institution’s 30-day shunt failure rates for all VP shunts inserted over a 5-year period. It is hoped that this will act as a benchmark with which future studies can be compared. If future publications were to include 30-day failure rates as an outcome, this would enable more appropriate direct comparisons to be drawn between them. This study confirms the previous finding that primary VP shunts have far better survival than do revision procedures. This high early failure rate in revision shunt surgery warrants more studies in both the adult and pediatric populations to determine which factors associated with failure may be avoidable. In addition, this study highlights specific risk factors for shunt failure for which future studies should control: the number of surgeons involved in the operating theater (for pediatric cases) and the origin of hydrocephalus (for adults).

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Anderson, Chumas. Acquisition of data: Anderson, Saukila, Robins, Akhunbay-Fudge, Phillips. Analysis and interpretation of data: Anderson, Phillips, Chumas. Drafting the article: Anderson, Chumas. Critically revising the article: Anderson, Goodden, Tyagi, Chumas. Reviewed submitted version of manuscript: Anderson, Goodden, Tyagi, Chumas. Approved the final version of the manuscript on behalf of all authors: Anderson. Statistical analysis: Anderson. Administrative/technical/material support: Anderson.

Supplemental Information

Online-Only Content

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

Previous Presentations

Some of the outline results from this paper were presented orally at the 42nd Annual Meeting of the International Society for Pediatric Neurosurgery in Rio de Janeiro, Brazil, in November 2014, and at the North of England Neurological Society Meeting in Harrogate, United Kingdom, in November 2016.

References

  • 1

    Ahmed N, Devitt KS, Keshet I, Spicer J, Imrie K, Feldman L, : A systematic review of the effects of resident duty hour restrictions in surgery: impact on resident wellness, training, and patient outcomes. Ann Surg 259:10411053, 2014

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2

    Al-Tamimi YZ, Sinha P, Chumas PD, Crimmins D, Drake J, Kestle J, : Ventriculoperitoneal shunt 30-day failure rate: a retrospective international cohort study. Neurosurgery 74:2934, 2014

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3

    Beuriat PA, Puget S, Cinalli G, Blauwblomme T, Beccaria K, Zerah M, : Hydrocephalus treatment in children: long-term outcome in 975 consecutive patients. J Neurosurg Pediatr 20:1018, 2017

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Bondurant CP, Jimenez DF: Epidemiology of cerebrospinal fluid shunting. Pediatr Neurosurg 23:254259, 1995

  • 5

    Burns EM, Rigby E, Mamidanna R, Bottle A, Aylin P, Ziprin P, : Systematic review of discharge coding accuracy. J Public Health (Oxf) 34:138148, 2012

  • 6

    Cochrane DD, Kestle JRW: The influence of surgical operative experience on the duration of first ventriculoperitoneal shunt function and infection. Pediatr Neurosurg 38:295301, 2003

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Drake JM, Kestle JR, Milner R, Cinalli G, Boop F, Piatt J Jr, : Randomized trial of cerebrospinal fluid shunt valve design in pediatric hydrocephalus. Neurosurgery 43:294305, 1998

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Faiz O, Banerjee S, Tekkis P, Papagrigoriadis S, Rennie J, Leather A: We still need to operate at night! World J Emerg Surg 2:29, 2007

  • 9

    Flannery AM, Duhaime AC, Tamber MS, Kemp J: Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 3: Endoscopic computer-assisted electromagnetic navigation and ultrasonography as technical adjuvants for shunt placement. J Neurosurg Pediatr 14 (Suppl 1):2429, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Hayhurst C, Beems T, Jenkinson MD, Byrne P, Clark S, Kandasamy J, : Effect of electromagnetic-navigated shunt placement on failure rates: a prospective multicenter study. J Neurosurg 113:12731278, 2010

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11

    Jenkinson MD, Gamble C, Hartley JC, Hickey H, Hughes D, Blundell M, : The British antibiotic and silver-impregnated catheters for ventriculoperitoneal shunts multi-centre randomised controlled trial (the BASICS trial): study protocol. Trials 15:4, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Kestle J, Drake J, Milner R, Sainte-Rose C, Cinalli G, Boop F, : Long-term follow-up data from the Shunt Design Trial. Pediatr Neurosurg 33:230236, 2000

  • 13

    Kestle JR, Holubkov R, Douglas Cochrane D, Kulkarni AV, Limbrick DD Jr, Luerssen TG, : A new Hydrocephalus Clinical Research Network protocol to reduce cerebrospinal fluid shunt infection. J Neurosurg Pediatr 17:391396, 2016

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Khan F, Rehman A, Shamim MS, Bari ME: Factors affecting ventriculoperitoneal shunt survival in adult patients. Surg Neurol Int 6:25, 2015

  • 15

    Klimo P Jr, Thompson CJ, Baird LC, Flannery AM: Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 7: Antibiotic-impregnated shunt systems versus conventional shunts in children: a systematic review and meta-analysis. J Neurosurg Pediatr 14 (Suppl 1):5359, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Kontny U, Höfling B, Gutjahr P, Voth D, Schwarz M, Schmitt HJ: CSF shunt infections in children. Infection 21:8992, 1993

  • 17

    Kulkarni AV, Drake JM, Lamberti-Pasculli M: Cerebrospinal fluid shunt infection: a prospective study of risk factors. J Neurosurg 94:195201, 2001

  • 18

    Langley JM, LeBlanc JC, Drake J, Milner R: Efficacy of antimicrobial prophylaxis in placement of cerebrospinal fluid shunts: meta-analysis. Clin Infect Dis 17:98103, 1993

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    National Confidential Enquiry into Patient Outcome and Death: Investigation of Out of Hours Cases in the NHS. London: National Confidential Enquiry into Patient Outcome and Death, 2003 (http://www.ncepod.org.uk/2003report/Downloads/03_s08.pdf) [Accessed November 3, 2017]

    • Search Google Scholar
    • Export Citation
  • 20

    Nesvick CL, Khan NR, Mehta GU, Klimo P Jr: Image guidance in ventricular cerebrospinal fluid shunt catheter placement: a systematic review and meta-analysis. Neurosurgery 77:321331, 2015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Park MK, Kim M, Park KS, Park SH, Hwang JH, Hwang SK: A retrospective analysis of ventriculoperitoneal shunt revision cases of a single institute. J Korean Neurosurg Soc 57:359363, 2015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Piatt JH Jr: Thirty-day outcomes of cerebrospinal fluid shunt surgery: data from the National Surgical Quality Improvement Program-Pediatrics. J Neurosurg Pediatr 14:179183, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Ratanalert S, Musikawat P, Oearsakul T, Saeheng S, Chowchuvech V: Non-shaved ventriculoperitoneal shunt in Thailand. J Clin Neurosci 12:147149, 2005

  • 24

    Reddy GK, Bollam P, Caldito G: Long-term outcomes of ventriculoperitoneal shunt surgery in patients with hydrocephalus. World Neurosurg 81:404410, 2014

  • 25

    Stein SC, Guo W: Have we made progress in preventing shunt failure? A critical analysis. J Neurosurg Pediatr 1:4047, 2008

  • 26

    Thomas R, Lee S, Patole S, Rao S: Antibiotic-impregnated catheters for the prevention of CSF shunt infections: a systematic review and meta-analysis. Br J Neurosurg 26:175184, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Tuli S, Drake J, Lawless J, Wigg M, Lamberti-Pasculli M: Risk factors for repeated cerebrospinal shunt failures in pediatric patients with hydrocephalus. J Neurosurg 92:3138, 2000

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Venable GT, Rossi NB, Morgan Jones G, Khan NR, Smalley ZS, Roberts ML, : The Preventable Shunt Revision Rate: a potential quality metric for pediatric shunt surgery. J Neurosurg Pediatr 18:715, 2016

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29

    Warf BC: Comparison of 1-year outcomes for the Chhabra and Codman-Hakim Micro Precision shunt systems in Uganda: a prospective study in 195 children. J Neurosurg 102 (4 Suppl):358362, 2005

    • PubMed
    • Search Google Scholar
    • Export Citation

If the inline PDF is not rendering correctly, you can download the PDF file here.

Contributor Notes

Correspondence Ian Anderson: Leeds General Infirmary, Leeds, United Kingdom. ian.anderson4@nhs.net.

INCLUDE WHEN CITING Published online March 9, 2018; DOI: 10.3171/2017.8.JNS17399.

Disclosures The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

  • View in gallery

    Kaplan-Meier survival curves for primary and secondary shunts in pediatric patients. Figure is available in color online only.

  • View in gallery

    Kaplan-Meier survival curves for primary and secondary shunts in adult patients. Figure is available in color online only.

  • 1

    Ahmed N, Devitt KS, Keshet I, Spicer J, Imrie K, Feldman L, : A systematic review of the effects of resident duty hour restrictions in surgery: impact on resident wellness, training, and patient outcomes. Ann Surg 259:10411053, 2014

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2

    Al-Tamimi YZ, Sinha P, Chumas PD, Crimmins D, Drake J, Kestle J, : Ventriculoperitoneal shunt 30-day failure rate: a retrospective international cohort study. Neurosurgery 74:2934, 2014

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3

    Beuriat PA, Puget S, Cinalli G, Blauwblomme T, Beccaria K, Zerah M, : Hydrocephalus treatment in children: long-term outcome in 975 consecutive patients. J Neurosurg Pediatr 20:1018, 2017

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Bondurant CP, Jimenez DF: Epidemiology of cerebrospinal fluid shunting. Pediatr Neurosurg 23:254259, 1995

  • 5

    Burns EM, Rigby E, Mamidanna R, Bottle A, Aylin P, Ziprin P, : Systematic review of discharge coding accuracy. J Public Health (Oxf) 34:138148, 2012

  • 6

    Cochrane DD, Kestle JRW: The influence of surgical operative experience on the duration of first ventriculoperitoneal shunt function and infection. Pediatr Neurosurg 38:295301, 2003

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Drake JM, Kestle JR, Milner R, Cinalli G, Boop F, Piatt J Jr, : Randomized trial of cerebrospinal fluid shunt valve design in pediatric hydrocephalus. Neurosurgery 43:294305, 1998

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Faiz O, Banerjee S, Tekkis P, Papagrigoriadis S, Rennie J, Leather A: We still need to operate at night! World J Emerg Surg 2:29, 2007

  • 9

    Flannery AM, Duhaime AC, Tamber MS, Kemp J: Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 3: Endoscopic computer-assisted electromagnetic navigation and ultrasonography as technical adjuvants for shunt placement. J Neurosurg Pediatr 14 (Suppl 1):2429, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Hayhurst C, Beems T, Jenkinson MD, Byrne P, Clark S, Kandasamy J, : Effect of electromagnetic-navigated shunt placement on failure rates: a prospective multicenter study. J Neurosurg 113:12731278, 2010

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11

    Jenkinson MD, Gamble C, Hartley JC, Hickey H, Hughes D, Blundell M, : The British antibiotic and silver-impregnated catheters for ventriculoperitoneal shunts multi-centre randomised controlled trial (the BASICS trial): study protocol. Trials 15:4, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Kestle J, Drake J, Milner R, Sainte-Rose C, Cinalli G, Boop F, : Long-term follow-up data from the Shunt Design Trial. Pediatr Neurosurg 33:230236, 2000

  • 13

    Kestle JR, Holubkov R, Douglas Cochrane D, Kulkarni AV, Limbrick DD Jr, Luerssen TG, : A new Hydrocephalus Clinical Research Network protocol to reduce cerebrospinal fluid shunt infection. J Neurosurg Pediatr 17:391396, 2016

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Khan F, Rehman A, Shamim MS, Bari ME: Factors affecting ventriculoperitoneal shunt survival in adult patients. Surg Neurol Int 6:25, 2015

  • 15

    Klimo P Jr, Thompson CJ, Baird LC, Flannery AM: Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 7: Antibiotic-impregnated shunt systems versus conventional shunts in children: a systematic review and meta-analysis. J Neurosurg Pediatr 14 (Suppl 1):5359, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Kontny U, Höfling B, Gutjahr P, Voth D, Schwarz M, Schmitt HJ: CSF shunt infections in children. Infection 21:8992, 1993

  • 17

    Kulkarni AV, Drake JM, Lamberti-Pasculli M: Cerebrospinal fluid shunt infection: a prospective study of risk factors. J Neurosurg 94:195201, 2001

  • 18

    Langley JM, LeBlanc JC, Drake J, Milner R: Efficacy of antimicrobial prophylaxis in placement of cerebrospinal fluid shunts: meta-analysis. Clin Infect Dis 17:98103, 1993

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    National Confidential Enquiry into Patient Outcome and Death: Investigation of Out of Hours Cases in the NHS. London: National Confidential Enquiry into Patient Outcome and Death, 2003 (http://www.ncepod.org.uk/2003report/Downloads/03_s08.pdf) [Accessed November 3, 2017]

    • Search Google Scholar
    • Export Citation
  • 20

    Nesvick CL, Khan NR, Mehta GU, Klimo P Jr: Image guidance in ventricular cerebrospinal fluid shunt catheter placement: a systematic review and meta-analysis. Neurosurgery 77:321331, 2015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Park MK, Kim M, Park KS, Park SH, Hwang JH, Hwang SK: A retrospective analysis of ventriculoperitoneal shunt revision cases of a single institute. J Korean Neurosurg Soc 57:359363, 2015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Piatt JH Jr: Thirty-day outcomes of cerebrospinal fluid shunt surgery: data from the National Surgical Quality Improvement Program-Pediatrics. J Neurosurg Pediatr 14:179183, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Ratanalert S, Musikawat P, Oearsakul T, Saeheng S, Chowchuvech V: Non-shaved ventriculoperitoneal shunt in Thailand. J Clin Neurosci 12:147149, 2005

  • 24

    Reddy GK, Bollam P, Caldito G: Long-term outcomes of ventriculoperitoneal shunt surgery in patients with hydrocephalus. World Neurosurg 81:404410, 2014

  • 25

    Stein SC, Guo W: Have we made progress in preventing shunt failure? A critical analysis. J Neurosurg Pediatr 1:4047, 2008

  • 26

    Thomas R, Lee S, Patole S, Rao S: Antibiotic-impregnated catheters for the prevention of CSF shunt infections: a systematic review and meta-analysis. Br J Neurosurg 26:175184, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Tuli S, Drake J, Lawless J, Wigg M, Lamberti-Pasculli M: Risk factors for repeated cerebrospinal shunt failures in pediatric patients with hydrocephalus. J Neurosurg 92:3138, 2000

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Venable GT, Rossi NB, Morgan Jones G, Khan NR, Smalley ZS, Roberts ML, : The Preventable Shunt Revision Rate: a potential quality metric for pediatric shunt surgery. J Neurosurg Pediatr 18:715, 2016

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29

    Warf BC: Comparison of 1-year outcomes for the Chhabra and Codman-Hakim Micro Precision shunt systems in Uganda: a prospective study in 195 children. J Neurosurg 102 (4 Suppl):358362, 2005

    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 968 47 0
Full Text Views 778 468 41
PDF Downloads 489 231 22
EPUB Downloads 0 0 0