See the corresponding editorial in this issue, p 461. J Neurosurg Pediatrics 9:000 000, 9:462 467, 2012 Nonprogrammable and programmable cerebrospinal fluid shunt valves: a 5-year study Clinical article Timothy J. Hatlen, B.A., 1 David B. Shurtleff, M.D., 1 John D. Loeser, M.D., 2 Jeffrey G. Ojemann, M.D., 2 Anthony M. Avellino, M.D., M.B.A., 2 and Richard G. Ellenbogen, M.D. 2 Departments of 1 Pediatrics and 2 Neurological Surgery, University of Washington School of Medicine, Seattle, Washington Object. Programmable valves (PVs) for shunting CSF have increasingly replaced nonprogrammable valves (NPVs). There have been only a few longer-term studies ( 5 years) conducted that have compared the effectiveness of NPVs with that of PVs for children with hydrocephalus, and only 1 study has reported NPVs as being favorable over PVs. The objective of this retrospective study was to compare the long-term survival of these 2 types of shunt valves. Methods. The authors collected data for all patients who underwent CSF shunt insertion or revision between January 1, 2000, and December 31, 2008. Patients underwent follow-up for a minimum of 2 years postoperatively. Statistical analyses were done using chi-square, Kaplan-Meier survival curve, and multivariate analyses. Results. A total of 616 valves were implanted, of which 313 were PVs and 303 were NPVs. Of these, 253 were original shunt implantations and 363 were revisions. The proportion of 5-year survival for NPVs (45.8%) was significantly higher than that for PVs (19.8%) (p = 0.0005, log-rank). The NPVs that survived longer than 6 months also survived through the 5th year better than the PVs (p = 0.0001). Conclusions. The authors data suggest that NPVs survive longer than PVs in children, but there is a need for prospective, case-control studies to confirm these data. (http://thejns.org/doi/abs/10.3171/2012.1.peds10482) Key Words hydrocephalus nonprogrammable valve shunt programmable valve Abbreviations used in this paper: HR = hazard ratio; NPV = nonprogrammable valve; PV = programmable valve. Pediatric hydrocephalus, affects 125,000 children in the US. 30 Since the development of the first balland-spring shunt mechanism in 1957, 28 many have attempted to improve the function of valve systems to achieve optimum control of hydrocephalus. 7 9,23 Most recently, PVs have become widely used. However, investigators have not agreed on a clearly superior valve when used in a broad range of children with hydrocephalus. 1,7 9,12,13,16,20 22,29,32 Conventional NPVs require surgeons to select an optimal opening pressure before implantation. Inaccurate estimates of the shunt valve function in situ may necessitate replacement of the valve to prevent over- or underdrainage. Also, when a child matures, the valve functions may need to be changed, requiring replacement. 2,22 In 1973, Hakim 11 introduced a servo valve that would maintain optimal intracranial pressure without surgical intervention. With complications due in part to open revisions, there would be a theoretical clinical advantage to avoiding invasive procedures as well as providing potential cost savings as reported by some using PVs. 22,34 For the past decade, clinicians have reported the efficacy of PVs compared with NPVs. Pollack et al. 22 reported the first prospective study that demonstrated equal safety and efficacy of a Codman Hakim (Codman, Inc.) PV compared with an NPV. However, they reported no difference in reduction of ventricle size and an equal rate of survival before revision during a 2-year follow-up period. Since then, PVs have been cited as effectively decreasing the ventricle size through pressure adjustments, 4,24 but they have not been associated with a lower frequency 462 J Neurosurg: Pediatrics / Volume 9 / May 2012
Nonprogrammable and programmable CSF shunt valves of slit ventricles than standard valves. 32 Programmable valves have also been reported to be more beneficial than NPVs for the treatment of a hygroma 2 and for patients with an arachnoid cyst. 27 Reports have shown that PVs are more successful than NPVs in the slow reduction in valve pressure to achieve permanent removal of the valve in certain subsets of patients 31 and are more successful in treating patients with postmeningitis hydrocephalus. 11 McGirt et al. 18 reported fewer proximal catheter obstructions using PVs than using NPVs. Many authors have reported that noninvasive pressure adjustments of a PV are capable of replacing an NPV s revision and thus are cost effective, 1,2,15,33,34 particularly during the first 6 months after insertion. 22 Our clinical experience suggests that there is utility in programmable shunt valves when patients complain of symptoms that are related to excessive intracranial pressure or overdrainage that can be relieved by reprogramming the valve. However, there have been exceptions to the proposed benefits of PVs. Rates of infection, type and frequency of shunt complications, and 2-year shunt survival curves have all shown no significant difference between the 2 types of valves. 7 10,13,20,21,24 Moritake et al. 20 reported that PVs have a lower risk for a combination of postoperative infection with malfunction over a 3-month period than NPVs, but neither infection nor malfunction alone reaches significance. The literature contains analyses of preoperative CSF protein and bacterial cultures and their correlation with valve malfunction. 15,19 Authors have reported that protein level has no direct relationship to valve failure. 5 7,15,19 Using scanning electron microscopy, Brydon et al. 5,6 reported that 13% of their 102 explanted valves studied after shunt failure were coated by white, red, or degenerating cells. Most were coated with protein, but the thin coating was not thought to be the cause of shunt failure. In 2004, Sgouros and Dipple 25 reported that 2 weeks after implantation, deposits of chemicals (sodium, potassium, calcium, and chloride) or protein were present on the valve, especially in the areas of high CSF flow turbulence. They also reported clumps of red blood cells from bleeding at the time of proximal limb placement were observed to occlude PVs at the narrow interface between the valve and diaphragm. Our retrospective study was designed to compare the longer-term effectiveness of PVs with NPVs. Methods With institutional review board approval, we retrospectively reviewed the medical records of 523 patients who underwent 616 shunt implantations or valve revisions between 2000 and 2006 in a single center, Seattle Children s Hospital. The 2-year minimum follow-up extended the study to December 2008. Data were collected from the patient data management system (a prospectively collected comprehensive database of patients with shunt procedures seen at our institution) and our hospital s computerized information services patient records. Forty-three patients diagnosed with a tumor were excluded from this study. We also excluded 117 patients with inadequate data in their records or less than a 2-year follow-up. J Neurosurg: Pediatrics / Volume 9 / May 2012 Operations were performed by the neurosurgeon on call on the day of admission. Twelve different surgeons performed surgery in 363 patients between 2000 and 2006. There was no preplanned strategy for utilizing PVs or NPVs. Shunt hardware was selected by the attending surgeon without an agreed set of criteria, and selection biases were unique to each surgeon. The PVs used included Strata (Medtronic, Inc.) and Codman Hakim valves, whereas the NPVs used included Heyer Schulte (Integra Neurosciences, Inc.), Spetzler (Integra Neurosciences, Inc.), Delta (Medtronic, Inc.), and Medtronic (Medtronic, Inc.) valves. Independent antisiphon devices were not used. Valve survival was defined as ending with removal or replacement of the shunt valve. Valves were replaced when the patient had signs and symptoms of either overdrainage or increased CSF pressure with functional proximal and distal limbs. In the presence of proximal catheter obstruction and slit ventricles, we routinely replaced the valve. Revisions of only the distal or proximal limb were not included as complications since they were considered non valve related invasive procedures when no subsequent revision was immediately required. Removal of a functioning shunt due to shunt infection was not considered shunt failure. Shunt infection was defined by demonstration of the organism by culture of the CSF, shunt equipment, overlying wound, or distal drainage from the shunt. One patient died of unknown causes and was considered in the survival curve as lost to follow-up. Therefore, it did not have an influence on the survival curves. The transfer of patient care to another institution was recorded as lost to follow-up for survival analysis. Statistical Analysis All analyses were performed using VassarStats Statistical Computation (http://faculty.vassar.edu/lowry/vassar Stats.html [Accessed February 13, 2012]) and other statistical analyses using MedCalc (version 11.3.3.0, MedCalc for Windows) at an alpha value set at 0.05. We used logrank, Student t-test, chi-square, and, when numbers were too small, Fisher exact probability tests to determine if the populations were comparable or not. To determine if the PV and the NPV populations were comparable, we analyzed the clinical characteristics of the 2 cohorts. These characteristics are listed in Tables 1 and 2. We then compared the characteristics of the 4 populations who underwent surgery performed by the 4 surgeons who operated on 70% of the patients. We used the same tests and for the same reasons as for the first set of analyses. These characteristics included type of valve by analyzing all NPV types by comparing brands and types to one another and all PV types to one another, time of day of the operation, duration of the operation, presence of a trainee, development of proximal limb obstruction, and type and time of the administration of antibiotic prophylaxis. To determine if any characteristics affected outcome within the 2 PV and NPV cohorts, we used the same tests noted above in addition to Kaplan-Meier survival probability estimates. These characteristics included time to 463
T. J. Hatlen et al. TABLE 1: Comparison of first implantations of PVs and NPVs* Type of Shunt Parameter PV NPV no. of shunts 81 85 M/F 33:45 51:34 age (yrs) mean 2.74 1.73 range 0 16.9 0 14.8 no. of valves w/ pressure adjustment 39 NA no. of valve reprogramming procedures mean 0.945 NA range 0 6 NA * Values represent the number of patients unless stated otherwise. Abbreviation: NA = not applicable. shunt failure, valve survival time by the patient s underlying etiology, age (0 6 months, 6 months to 10 years, and 10 years), type of valve (PV or NPV) and manufacturer (Codman or Strata), presence of a trainee at surgery, time of day of the operation (morning, afternoon, or night), duration of the operation (longer or shorter than 1 hour), and type and time of the administration of prophylaxis. The final 2 analyses were then performed using Kaplan-Meier survival probability estimates and HRs between the PV and NPV cohorts to determine valve survival time. We calculated 2 survival curves. The first analysis consisted of all patients with either initial insertion of a PV or an NPV. To include our total experience during the 5 years of this study, we next included valves used as a replacement during a revision and included only cases that had an adjustment or revision 6 months after either an initial or subsequent implantation. Reprogramming episodes were so few that they made no difference in calculating the survival outcomes. We therefore considered only the surgical removal or replacement of an NPV or PV in the final 2 shunt survival estimates. (Detailed data and analyses can be obtained on request from the corresponding author.) Results Tables 1 and 2 demonstrate the clinical characteristics and complications in the 2 cohorts. Age was significantly younger in the NPV cohorts than in the PV cohorts. One patient in the PV cohort died, and none died in the NPV group. The only significant differences between the NPV and PV cohorts were the longer average shunt survival and the longer 5-year survival of the NPV cohort. When analyzing the characteristics of the 4 cohorts of the surgeons noted in Statistical Analysis, there were no significant differences between either the 2 cohorts (those with PVs and NPVs) or the 4 primary surgical cohorts. Proximal obstructions defined by only replacement of the proximal limb were recorded in 14 (47%) of 44 operations in the PV cohort and in 37 (40%) of 91 in the NPV cohort (chi-square with Yates correction = 0.65; p = 0.4). There were no statistically significant influences on TABLE 2: Demographics of all shunts surviving longer than 6 months* Type of Shunt Parameter PV NPV no. of shunts 313 303 no. of pts 162 201 no. of NPVs surviving >6 mos 220 no. of pts w/ NPVs surviving >6 mos 193 no. of PVs needing reprog after 6 mos (total 56 reprog) total no. of reprog sessions after 6 mos 152 no. of pts w/ PV w/ reprog after 6 mos 53 M/F 31:22 110:83 mean age at insertion (yrs) 8.01 ± 6.53 6.5 ± 6.16 etiology intraventricular hemorrhage 15 53 myelomeningocele 14 56 congenital 7 32 unknown 5 12 bacterial meningitis 1 11 cyst 4 11 head trauma 3 1 other 4 17 complications death 1 0 no. of shunt infections 2 11 shunt infections (per 1000 person-yrs) 11.15 10.42 * pts = patients; reprog = reprogramming. Age at insertion of the valve either at the time of initial insertion or replacement of the valve (neither proximal nor distal limb of shunt revision is included in this number). Subarachnoid or ependymal. Nontumor. shunt survival within the PV and NPV cohorts by the surgical characteristics listed in Statistical Analysis. In the final analyses there were 81 PV and 85 NPV first implantations. Figure 1 illustrates the longer survival of the NPV compared with the PV cohort through the 5th year (HR 0.63 [95% CI 0.44 0.89], p = 0.001). The average shunt lifetime was shorter for PVs (2.94 months [range 0.3 7.86 months]) than for NPVs (4.04 months [range 0.1 8.5 months]) (p = 0.12, Student t-test). Preoperative characteristics of these 2 cohorts were similar (Tables 1 and 3), and there were no significant influences on survival between the NPV and PV first implantation cohorts when compared by etiological group (Table 3). The rate of infection for the 2 cohorts was equivalent (see Table 3; 7 of the 81 cases of PVs and 5 of the 85 NPVs; p = 0.56, chi-square test). A total of 303 NPVs were inserted initially or during a revision after removal of a prior valve, of which 220 (72.6%) survived 6 months or longer. Of the 313 PVs inserted initially or during a revision after removal of a prior valve, only 56 (17.9%) survived 6 months or longer. 464 J Neurosurg: Pediatrics / Volume 9 / May 2012
Nonprogrammable and programmable CSF shunt valves TABLE 3: Results of first implantations of PVs and NPVs Fig. 1. Kaplan-Meier survival curve for PVs and NPVs including those that were first-insertion cohorts at our institution and underwent follow-up for a minimum of 2 years. Preoperative and postoperative characteristics of the initial insertion and total cohorts were similar, and survival curves for these 2 cohorts were essentially the same as for Fig. 1 with the same degree of significance (HR 0.48 [95% CI 0.32 0.73], p = 0.0001), thus we have not reproduced this second set of curves. The 1- and 2-year survivals for PVs and NPVs were comparable but the proportion of 5-year survivals for NPVs (45.8%) was significantly higher than for PVs (19.8%) (p = 0.0005). Therefore, the average shunt lifetime was shorter for PVs (2.94 months [range 0.3 7.86 months]) than for NPVs (4.04 months [range 0.1 8.51 months). Discussion Our study has some limitations because bias can be introduced in a retrospective review that does not have randomized, prospectively matched groups. Bias could have also been introduced by selection of a PV for more severe cases. Other confounding factors could be the operative decisions and techniques of the different surgeons and the difficulty in determining the site of shunt obstruction. The primary goal of this study was to evaluate the relative long-term (5 years or longer) survival of PVs compared with NPVs. We concur with Richards et al. (Richards H, Seeley H, Pickard J: Are adjustable valves effective in all ages of patient? Paper presented at the 54th International Conference, Society for Research into Hydrocephalus and Spina Bifida, Vancouver, British Columbia, Canada, July 9, 2010), who reported a better long-term survival of NPVs than of PVs for children. The majority of past reports only demonstrate an advantage to using PVs in the short term. 1 3, 8,11,13,15,18,22,32,34 We excluded the first 6-month experience in the second analysis because we could not distinguish between fine tuning and an adjustment that would have substituted for a needed revision. Pollack et al. 22 observed that close to 70% of pressure adjustments for fine tuning occur in the first 6 months. Our first set of 3 analyses described cohorts that were essentially very similar with differences in surgical characteristics having no effect on the survival curves of cohorts with either PVs or NPVs. As mentioned in Results, the frequency of proximal revisions was similar in patients with NPVs or PVs, suggesting that proximal obstructions J Neurosurg: Pediatrics / Volume 9 / May 2012 Parameter PV NPV p Value* no. of shunts 81 85 shunt lifetime (yrs) 0.012 mean 2.94 4.04 range 0.03 7.86 0.01 8.94 overall valve lifetime (yrs) 283.5 343.25 complications no. of infections 7 5 0.56 no. of nonvalve invasive ops/ 18/13 32/23 no. of valves valve survival in yrs (%) 1 68.0 71.8 0.62 2 56.8 65.9 0.29 3 48.2 61.2 0.12 4 38.3 50.6 0.12 5 19.8 45.8 0.0005 * All values in this column are based on the chi-square test, except for shunt lifetime, which is based on the t-test. The number of shunt revisions performed that involved only the proximal and/or distal limbs of the shunt system. The valves were evaluated as functioning and were not replaced. The number of years that the percentage of valves in each cohort functioned without replacement. did not influence the outcome of the valve survival in these 2 cohorts differently. The exception is the younger age of the cohort with NPVs, a characteristic that should have led to a shorter survival time than the older cohort with PVs. These older ages can be explained by 18 patients older than 3 years in the PV cohort with an average age of 10.18 years (range 3.33 23.44 years). The NPV cohort includes 13 patients with an average age of 8.79 years (range 3.51 14.76 years). The younger age of the NPV cohort is partially explained by concern that the firmer and higher profile of the PVs would compromise the tenuous skin in small patients. Furthermore, at our institution shunt insertion is not typically performed in the acute setting of inflammatory causes such as intraventricular hemorrhage, which is particularly common in newborns, or infection. We use serial ventriculostomies with or without a reservoir or subgaleal drainage to reduce protein content in the CSF prior to inserting CSF shunts and then only if the patients remain symptomatic. In addition, we routinely perform cesarean sections for all myelomeningocele cases diagnosed in utero to reduce shunt exposure to inflammation. 14 Thus, comparisons between our outcomes may not be comparable to other reports. 14,21 A study of removed valves after shunt complication by Sgouros and Dipple 25 reinforced the possibility of deposits interfering with the PV s hydrodynamic function. Other studies mention biofilm or proteinaceous debris on the valve surface upon removal. 5,6,9,10,17 However, other studies analyzing protein counts in the CSF both in vitro and after complication have shown no significant correlation with shunt failure. 5,6,19 The increased technological 465
T. J. Hatlen et al. advancements made to valve systems (like the staircase mechanism of the Codman Medos valve), have provided additional CSF pathways that are turbulent and susceptible to areas of buildup of debris that may be a reason for the higher failure rate of PVs compared with NPVs 25 in the 2 long-term studies. Conclusions Our study indicates that PVs do not improve longterm valve survival when compared with NPVs. However, other data 1,2,11,13 16,18,22,26,27,33,34 and our experience attest to the ability of PVs to obtain optimum pressures nonoperatively in the outpatient department in the first 6 months of life as an indication for continued use. This current study does not address this short postoperative period. Our new finding of the better survival of NPVs compared with PVs in this study and in the report by Richards et al. (Richards H, Seeley H, Pickard J: Are adjustable valves effective in all ages of patient? Paper presented at the 54th International Conference, Society for Research into Hydrocephalus and Spina Bifida, Vancouver, British Columbia, Canada, July 9, 2010) and the lack of adequate cost-effectiveness data, suggests the need for a long-term study that explores these variables. Such a study should be a prospective and case-control one, and randomized by type and severity of diagnoses with identified criteria for surgical selection of valves and surgical methods. Disclosure This study was supported, in part, by the National Foundation March of Dimes and gifts to the Birth Defects Research Program in the Department of Pediatrics at the University of Washington and Seattle Children s Hospital. Author contributions to the study and manuscript preparation include the following. Conception and design: Shurtleff, Hatlen. Acquisition of data: Shurtleff, Hatlen. Analysis and interpretation of data: Shurtleff, Hatlen, Loeser. Drafting the article: Shurtleff, Hatlen, Loeser, Ojemann. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Shurtleff. Statistical analysis: Shurtleff, Hatlen. Administrative/technical/material support: Shurtleff, Ojemann. Study supervision: Shurtleff. References 1. Ahn ES, Bookland M, Carson BS, Weingart JD, Jallo GI: The Strata programmable valve for shunt-dependent hydrocephalus: the pediatric experience at a single institution. Childs Nerv Syst 23:297 303, 2007 2. Arnell K, Eriksson E, Olsen L: The programmable adult Codman Hakim valve is useful even in very small children with hydrocephalus. A 7-year retrospective study with special focus on cost/benefit analysis. Eur J Pediatr Surg 16:1 7, 2006 3. Berry JG, Hall MA, Sharma V, Goumnerova L, Slonim AD, Shah SS: A multi-institutional, 5-year analysis of initial and multiple ventricular shunt revisions in children. Neurosurgery 62:445 454, 2008 4. Black PM, Hakim R, Bailey NO: The use of the Codman- Medos Programmable Hakim valve in the management of patients with hydrocephalus: illustrative cases. Neurosurgery 34:1110 1113, 1994 5. Brydon HL, Bayston R, Hayward R, Harkness W: The effect of protein and blood cells on the flow-pressure characteristics of shunts. Neurosurgery 38:498 505, 1996 6. Brydon HL, Keir G, Thompson EJ, Bayston R, Hayward R, Harkness W: Protein adsorption to hydrocephalus shunt catheters: CSF protein adsorption. J Neurol Neurosurg Psychiatry 64:643 647, 1998 7. Christiansen CM: Understanding hydrocephalus: a review of diagnosis and management strategies. Physician Assist 26: 30 36, 2002 8. Drake JM: The surgical management of pediatric hydrocephalus. Neurosurgery 62 (Suppl 2):633 642, 2008 9. Drake JM, Kestle JR, Milner R, Cinalli G, Boop F, Piatt J Jr, et al: Randomized trial of cerebrospinal fluid shunt valve design in pediatric hydrocephalus. Neurosurgery 43:294 305, 1998 10. Drake JM, Kulkarni AV: Cerebrospinal fluid shunt infections. Neurosurg Q 3:283 294, 1993 11. Hakim S: Hydraulic and mechanical mis-matching of valve shunts used in the treatment of hydrocephalus: the need for a servo-valve shunt. Dev Med Child Neurol 15:646 653, 1973 12. Kamida T, Uchida S, Abe T, Fijiki M, Kobayashi H: Efficacy of a programmable valve shunt system for hydrocephalic patients with various diseases: comparison with a nonprogrammable differential pressure valve. Neurosurg Q 15:223 226, 2005 13. Kestle JR, Walker ML: A multicenter prospective cohort study of the Strata valve for the management of hydrocephalus in pediatric patients. J Neurosurg 102 (2 Suppl):141 145, 2005 14. Luthy DA, Wardinsky T, Shurtleff DB, Hollenbach KA, Hickok DE, Nyberg DA, et al: Cesarean section before the onset of labor and subsequent motor function in infants with meningomyelocele diagnosed antenatally. N Engl J Med 324:662 666, 1991 15. Mangano FT, Menendez JA, Habrock T, Narayan P, Leonard JR, Park TS, et al: Early programmable valve malfunctions in pediatric hydrocephalus. J Neurosurg 103 (6 Suppl): 501 507, 2005 16. Martínez-Lage JF, Almagro MJ, Del Rincón IS, Pérez-Espejo MA, Piqueras C, Alfaro R, et al: Management of neonatal hydrocephalus: feasibility of use and safety of two programmable (Sophy and Polaris) valves. Childs Nerv Syst 24:549 556, 2008 17. Mauer UM, Schuler J, Kunz U: The Hakim programmable valve: reasons for reprogramming failures. J Neurosurg 107: 788 791, 2007 18. McGirt MJ, Buck DW II, Sciubba D, Woodworth GF, Carson B, Weingart J, et al: Adjustable vs set-pressure valves decrease the risk of proximal shunt obstruction in the treatment of pediatric hydrocephalus. Childs Nerv Syst 23:289 295, 2007 19. McLaurin RL: Ventricular shunts: complications and results, in McLaurin RL, Schut L, Venes JL, et al (eds): Pediatric Neurosurgery: Surgery of the Developing Nervous System, ed 2. Philadelphia: WB Saunders, 1989, pp 219 229 20. Moritake K, Nagai H, Miyazaki T, Nagasako N, Yamasaki M, Sakamoto H, et al: Analysis of a nationwide survey on treatment and outcomes of congenital hydrocephalus in Japan. Neurol Med Chir (Tokyo) 47:453 461, 2007 21. Notarianni C, Vannemreddy P, Caldito G, Bollam P, Wylen E, Willis B, et al: Congenital hydrocephalus and ventriculoperitoneal shunts: influence of etiology and programmable shunts on revisions. Clinical article. J Neurosurg Pediatr 4: 547 552, 2009 22. Pollack IF, Albright AL, Adelson PD: A randomized, controlled study of a programmable shunt valve versus a conventional valve for patients with hydrocephalus. Neurosurgery 45:1399 1411, 1999 23. Pople IK: Hydrocephalus and shunts: what the neurologist should know. J Neurol Neurosurg Psychiatry 73 (Suppl 1): i17 i22, 2002 24. Reinprecht A, Dietrich W, Bertalanffy A, Czech T: The Medos Hakim programmable valve in the treatment of pediatric hydrocephalus. Childs Nerv Syst 13:588 594, 1997 466 J Neurosurg: Pediatrics / Volume 9 / May 2012
Nonprogrammable and programmable CSF shunt valves 25. Sgouros S, Dipple SJ: An investigation of structural degradation of cerebrospinal fluid shunt valves performed using scanning electron microscopy and energy-dispersive x-ray microanalysis. J Neurosurg 100:534 540, 2004 26. Sikorski CW, Rosen DS, Frim DM: Adjustable shunt valve reprogramming at home: safety and feasibility. Neurosurgery 60:333 337, 2007 27. Sindou M, Guyotat-Pelissou I, Chidiac A, Goutelle A: Transcutaneous pressure adjustable valve for the treatment of hydrocephalus and arachnoid cysts in adults. Experiences with 75 cases. Acta Neurochir (Wien) 121:135 139, 1993 28. Spitz EB: Neurosurgery in the prevention of exogenous mental retardation. Pediatr Clin North Am 6:1215 1235, 1959 29. Stein SC, Guo W: Have we made progress in preventing shunt failure? A critical analysis. J Neurosurg Pediatr 1:40 47, 2008 30. Stein SC, Guo W: The prevalence of shunt-treated hydrocephalus: a mathematical model. Surg Neurol 72:131 137, 2009 31. Takahashi Y: Withdrawal of shunt systems clinical use of the programmable shunt system and its effect on hydrocephalus in children. Childs Nerv Syst 17:472 477, 2001 32. Weinzierl MR, Rohde V, Gilsbach JM, Korinth M: Management of hydrocephalus in infants by using shunts with adjustable valves. J Neurosurg Pediatr 2:14 18, 2008 33. Yamashita N, Kamiya K, Yamada K: Experience with a programmable valve shunt system. J Neurosurg 91:26 31, 1999 34. Zemack G, Romner B: Do adjustable shunt valves pressure our budget? A retrospective analysis of 541 implanted Codman Hakim programmable valves. Br J Neurosurg 15:221 227, 2001 Manuscript submitted October 22, 2010. Accepted January 17, 2012. Portions of this work were presented in abstract form at the 53rd International Conference, Society of Research into Hydrocephalus and Spina Bifida, Belfast, Ireland, June 25, 2009. Please include this information when citing this paper: DOI: 10.3171/2012.1.PEDS10482. Address correspondence to: David B. Shurtleff, M.D., M/S A7938, P.O. Box 5371, Seattle, Washington 98105. email: david. shurtleff@seattlechildrens.org. J Neurosurg: Pediatrics / Volume 9 / May 2012 467