Decreased head circumference in shunt-treated compared with healthy children

J Neurosurg Pediatrics 12:483 490, 2013 AANS, 2013 Decreased head circumference in shunt-treated compared with healthy children Clinical article Daniel Nilsson, M.D., Ph.D., 1,2 Johanna Svensson, M.D., 1 Betül A. Korkmaz, M.D., 1 Helena Nelvig, M.D., 1 and Magnus Tisell, M.D., Ph.D. 1,2 1 Institute of Neuroscience and Physiology at the Sahlgrenska Academy, University of Gothenburg; and 2 Department of Neurosurgery, Sahlgrenska University Hospital, Gothenburg, Sweden Object. In this study, the authors goal was to compare head circumference in hydrocephalic children during the first 4 years of ventriculoperitoneal (VP) shunt treatment with data on healthy children and to investigate predictors of skull growth in children with a VP shunt. Methods. Children from western Sweden treated for hydrocephalus with an initial VP shunt insertion performed between 2001 and 2006 who were younger than 12 months of age at the time of surgery were included. Children with major brain malformations, craniofacial syndromes, large cysts, and tumors were excluded. Head circumference, weight, and height at 9 defined ages up to 4 years were obtained and compared with data from a reference population of 3650 healthy children using the standard deviation score (SDS). Predictors (length, weight, etiology of hydrocephalus, valve type, number of revisions, valve setting, number of adjustments, and time of first surgery) for head circumference SDS and changes in head circumference SDS from shunt insertion at 1 year to last measurement were analyzed using bivariate and multiple linear regression analysis. Results. Fifty children were included. The mean SDSs for head circumference in shunt-treated compared with healthy children were 1.95 ± 2.50 at shunt insertion (p < 0.001, n = 44), 0.38 ± 1.97 at 1 year (p = 0.27, n = 33), -0.96 ± 2.05 at 2 years (p = 0.046, n = 21), -1.39 ± 2.25 at 3 years (p = 0.026, n = 16), and 0.63 ± 3.34 at 4 years (p = 0.73, n = 4). Significant predictors for low head circumference SDS at 1 year of age were low weight (p = 0.002) and short height (p = 0.022) and at last measurement low weight (p < 0.0001), short height (p = 0.002), and 1 4 shunt revisions (p = 0.034). A significant predictor for change in head circumference SDS from shunt insertion to 1 year of age was the number of shunt valve revisions (p = 0.04) and at last measurement an etiology of intraventricular hemorrhage (p = 0.0058). Conclusions. Shunt-treated children have smaller head circumferences at 2 and 3 years of age than healthy children. Low weight, short height, etiology of intraventricular hemorrhage, and frequent shunt valve revisions are predictors for decreased head circumference. Prospective, randomized studies comparing skull growth using fixed and adjustable pressure-regulated shunt valves and flow-regulated valves are needed. (http://thejns.org/doi/abs/10.3171/2013.8.peds1370) Key Words hydrocephalus ventriculoperitoneal shunt skull growth head circumference Two-thirds of brain growth occurs during the first 2 years of life. Growth of the cranium is triggered by the pressure of the growing brain, and head circumference is an important indicator of brain development. 10 In shunt-treated children, abnormal, reduced skull growth is often noticed clinically and may be an early indicator of excessive drainage of CSF, or overdrainage. Overdrainage has long been recognized as a side effect of shunt treatment and may cause later complications such as premature closure of the cranial sutures (craniosynostosis), radiological findings of the slit ventricle, and slit ventricle syndrome. 5,13,17 In spite of improvements in Abbreviations used in this paper: ASD = antisiphon device; IVH = intraventricular hemorrhage; SDS = standard deviation score; VP = ventriculoperitoneal. J Neurosurg: Pediatrics / Volume 12 / November 2013 shunt valves, overdrainage and its associated complications are persistent in shunt-treated patients and remain a challenge in pediatric neurosurgery. Although taking repeated measurements of head circumference is a simple, risk-free method for monitoring head growth and the effects of shunt treatment in infants, there are very few systematic data of head growth in shunt-treated children. 5,7 In Sweden, the nationwide, systematic, postnatal health examination scheme for children includes control of head growth at predefined ages in a standardized way by trained personnel. This service is free of charge and enrolls more than 99% of children. The reference charts for normal head growth are based on a large healthy population of children from a defined geographic area in western Sweden. 18 This made it possible to compare head circumference in shunt-treated children and healthy children 483

D. Nilsson et al. over time in this population-based study. We investigated predictors for reduced head circumference, including the etiology of hydrocephalus, number of shunt revisions, valve pressure, type of shunt, weight, height, and sex. If predictors for reduced head circumference could be identified, it might be possible to identify patients at risk for overdrainage and to influence skull growth before overdrainage complications occur. The aim of this study was to investigate skull growth in shunt-treated children compared with healthy children during infancy and early childhood and to investigate predictors for decreased head circumference in children with a ventriculoperitoneal (VP) shunt. Fig. 1. Flowchart of patient inclusion. GH = growth hormone. Methods This population-based pediatric study was carried out at Sahlgrenska University Hospital, Gothenburg, Sweden. All patients requiring neurosurgical care from the southwestern region of Sweden (population approximately 2.1 million) are referred to this unit. The electronic patient chart was searched for a procedure code starting with AAF (intracranial shunt procedures) according to the Swedish version of the NOMESCO Classification of Surgical Procedures version 1.9. 12 Patients with hydrocephalus who underwent VP shunt insertion before 12 months of age between January 1, 2001, and December 31, 2006, at the Department of Neurosurgery at the Sahlgrenska University Hospital, Queen Silvia Children s Hospital, Gothenburg, were included. Children with major brain malformations, craniofacial syndromes, large cysts, tumors, growth hormone deficiency, or missing data were excluded (Fig. 1). Head circumference was obtained from medical records at 9 defined ages (0, 3, 6, 9, 12, 18, 24, 36, and 48 months). For a measurement to be registered it had to be noted within 1 month from the dates correlating to 0, 3, 6, 9, 12, 18, and 24 months of age or within 3 months from the dates correlating to 36 and 48 months of age. Head circumference was measured in a standardized way by a pediatric neurologist or pediatric nurse either in a hospital (at birth) or in an outpatient pediatric clinic. The head circumferences of shunt-treated children were compared with data from a reference population of 3650 healthy children who were born between 1973 and 1975. 18 Predictors for abnormal skull growth (etiology of hydrocephalus, number of shunt revisions, valve pressure, type of shunt, weight and height [at birth and then at each defined age], and sex) were extracted from the electronic patient chart and other medical records. A correction of the date of birth for infants born earlier than Week 39 + 6 days to the expected date of birth (Week 40 + 0 days) was made. No adjustments were made for infants born after Week 40 + 0 days age of gestation. The number of valve replacements and number of level adjustments since the date of last measurement were also registered at each defined age, and each of these variables was summed to obtain the total number of each procedure. The type of shunt valve and level setting were documented. The incidence of surgery for craniosynostosis and cranial vault expansion during the study period were recorded. The shunt surgery was performed by a Swedish boardcertified neurosurgeon, usually a consultant. Shunt types used were pressure-regulated adjustable valve (Strata, Medtronic) in 30 cases (60%) and pressure-regulated fixed valve with medium pressure (Delta, Medtronic) in 20 cases (40%). Postoperatively, head circumference was measured weekly initially at an outpatient pediatric clinic, then monthly during the first 6 months. Patients were also seen 4 6 weeks postoperatively, at 6 and 12 months at an outpatient clinic by a pediatric neurologist, and if there was any problem (for example, rapidly changing head circumference or subgaleal fluid) a neurosurgeon was consulted. If skull growth stopped and/or head circumference crossed 2 SD curves, we adjusted the valve to a higher performance level, if the patient had an adjustable valve. A low pressure setting (< 8 cm H 2 O) was avoided whenever possible. Statistical Analysis Standard deviation scores (SDSs) for head circumference for each time point for the shunt-treated population compared with the reference population were calculated, as described below. This was done continuously for age, separately for boys and girls, as there is a sex difference in the reference population. 18 The individual measurements 484 J Neurosurg: Pediatrics / Volume 12 / November 2013

Head circumference after ventriculoperitoneal shunt treatment of head circumference were interpolated or extrapolated to exact ages, such as 1.0, 2.0, and 3.0 years of age. Linear interpolation was performed based on the nearest values on both sides. If extrapolation was necessary, this was performed only if the nearest value was within 3 months. This made it possible to calculate an SDS for every 365th day after first shunt insertion. The SDSs were then compared with the normal population using the SDS and the Wilcoxon signed-rank test. Similarly, calculations for changes in SDS over time were made by comparing values for each year and for the last recorded measurement with values from the first shunt operation using the Wilcoxon signedrank test. Standard Deviation Score Calculation for Head Circumference Estimation of Growth Reference Values. Three methods were used to construct the reference mean and SD. The estimated mean and SD values are indicated by a subscript accordingly, that is, mean 1, mean 2, and mean 3, and SD 1, SD 2, and SD 3. Raw Values. The mean 1 and SD 1 were simply computed from the raw values for head circumference at each particular age. Interpolated Values to Exact Ages. It is commonly observed that the mean values for height, weight, and head circumference fluctuate in an unsmoothed way in childhood. One possible reason is that children are not measured at exactly the same age. To reduce this influence, the individual body measurements were interpolated to exact ages, such as 1.0, 2.0, and 3.0 years. First, the mean 1 of the measured values between 0 and 3, as well as between 3 and 18 years of age, were fitted by a polynomial function combined with a logistics term. The curve fitting included polynomial, exponential, and logistic terms to achieve the optimal curve fitting (R 2 > 0.99). The estimated functional values mean 2 and SD 2 could be computed from these functions for any age point. Smoothed Values. To reduce any irregular fluctuation due to the mixed longitudinal and cross-sectional nature of the study, that is, the fluctuation in sample size over the various ages, the same curve-fitting procedures were reapplied to the mean 2 derived from the interpolated values. Good curve fitting, as judged from the residual values, was obtained when 3 age intervals were treated separately, that is, 0 2, 3 10, and 11 18 years of age (R 2 > 0.99). The mean 3 values were simply estimated within the 3 age periods. The SD 3 value was estimated by a polynomial function from the curve fitting of all SD 1 values for the measured values from 0 to 18 years of age. For a complete description of the SDS calculation method, see Wikland et al. 18 Predictors (sex, etiology of hydrocephalus, number of shunt revisions before 2 years of age, valve pressure, type of shunt valve, and weight and height SDS at shunt insertion and then at each defined age) for SDS of head circumference were analyzed at age 1 year (n = 33 [28 original and 5 interpolated]) and at last measurement (n J Neurosurg: Pediatrics / Volume 12 / November 2013 = 49 [29 original, 20 extrapolated or interpolated]) using bivariate regression analyses. The same predictors as above were analyzed for changes in SDS of head circumference from shunt insertion to age 1 year and at last measurement using bivariate regression. Change in head circumference SDS complements head circumference SDS by quantifying the head circumference change over time, which better reflects growth. Significant predictors from the bivariate analysis were entered into a multiple stepwise regression analysis. This study was approved by the regional ethics review board of Gothenburg University, Sweden. Patient consent for inclusion was not possible to obtain, and it was not deemed necessary by the ethics review board. Results Fifty children were included in this study (21 girls and 29 boys). Descriptive statistics including the investigated predictors of the shunt-treated children are summarized in Table 1. Intraventricular hemorrhage (IVH) etiology was found in 39% of shunt-treated children at 1 year, 47% at 2 years, 50% at 3 years, and 39% at last measurement. The head circumference for shunt-treated girls and boys compared with the reference population are found in Figs. 2 and 3. The head circumference SDS compared with healthy children and change in head circumference from shunt insertion until ages 1, 2, 3, and 4 years are shown in Figs. 4 and 5. The significant predictors for head circumference SDS at age 1 year and last measurement can be found in Table 2. Weight SDS and height SDS were the only independent predictors of head circumference (p < 0.05), confirmed by multiple stepwise regression analysis. None of the remaining predictors of head circumference at 1 year of age and at last measurement were significant at the p < 0.05 level. The following predictors of change in head circumference SDS from shunt insertion to age 1 year and to last measurement were found to be significant using bivariate linear regression: at 1 year number of shunt valve revisions (b = 0.66, SE = 0.32, p = 0.043) and at last measurement cause of IVH (b = 2.07, SE 0.71, p = 0.0058). None of the remaining predictors of change in head circumference from shunt insertion to 1 year of age and to last measurement were significant at the p < 0.05 level. Three children (6%) had craniosynostosis (2 children had sagittal suture craniosynostosis and 1 had craniosynostosis of the sagittal and metopic sutures), all treated with strip craniectomy and dynamic remodeling with spring cranioplasty. 19 The indication for surgery was correction of the abnormal head shape. Cranial vault expansion for increased intracranial pressure was not carried out in this population. Discussion Factors Affecting Head Circumference in Shunt-Treated Children This is the first population-based study of head circumference in shunt-treated children compared with a 485

D. Nilsson et al. TABLE 1: Descriptive statistics of the 50 children treated with a VP shunt Parameter Value* no. of patients 50 sex male 29 (58) female 21 (42) cause of hydrocephalus aqueductal stenosis 4 (8) myelomeningocele 19 (38) IVH 20 (40) meningitis 2 (4) other 5 (10) chronological age at 1st shunt insertion (days) mean ± SD 76.7 ± 82.5 median (range) 41 (0 279) corrected (biological) age at 1st shunt insertion (days) mean ± SD 39.5 ± 77.3 median (range) 8 ( 36 to 279) no. of shunt valve changes before age 2 yrs mean ± SD 0.7 ± 1.14 median (range) 0 (0 4) no. of shunt revisions before age 2 yrs mean ± SD 1.4 ± 1.68 median (range) 1 (0 5) valve type during study Delta 18 (36) Strata 32 (64) valve level mean ± SD 1.55 ± 0.32 median (range) 1.5 (1 2.5) weight at 1st shunt insertion (g) mean ± SD 4523.8 ± 1675.5 median (range) 3845 (2620 9590) height at 1st shunt insertion (cm) mean ± SD 56.3 ± 7.0 median (range) 53 (48 70) head circumference at 1st shunt insertion (cm) mean ± SD 40.1 ± 4.2 median (range) 40 (33 49.2) valve type at 1st shunt insertion Strata 30 (60) Delta 20 (40) * Values are the number of patients (%) unless otherwise noted. healthy reference population. We found significantly reduced head circumference in shunt-treated children at 2 and 3 years of age. Preoperative skull size, as expected, was larger in the hydrocephalic children. Reduction in head circumference began after shunt insertion, and head size was normalized, compared with the reference population, by approximately 1 year of age. However, reduction in skull size continued until 3 years of age, where the mean SDS was -1.39 in the shunt-treated population, significantly lower than in the reference population. At 4 years, head size was close to that of the reference population, but this is based on very few measurements, and the difference was not significant. A possible explanation for the reduced skull size in shunt-treated children is the continuous drainage of CSF from the ventricles, resulting in reduced brain size and skull growth. However, other factors that may contribute to a smaller than normal skull are decreased brain size and brain atrophy resulting from the cause of the hydrocephalus, for example, IVH or myelomeningocele. It is not possible to find a reference population with matching brain volume and/or matching brain development to shunt-treated children, but by excluding children with conditions that may dramatically influence brain size, we minimized the influence of etiology on skull size. That weight and height were predictors of head circumference SDS was not surprising as weight and height are in general associated with head circumference, also in healthy children. As change in head circumference SDS reflects change in head circumference (that is, growth) better than head circumference SDS only, the number of revisions and IVH etiology as predictors for change in head circumference SDS may suggest that these 2 factors are affecting change in head circumference more than they affect the absolute value of head circumference. This study found IVH, and thus prematurity since all children with IVH were born prematurely, low weight, and short height at birth to be predictors for a small head. This is in accordance with a previous study by Kan et al., who found a young age at shunt insertion to be a risk factor for the development of slit-like ventricles. 9 Furthermore, these patients have various degrees of damage to the brain parenchyma, with secondary brain atrophy, which might predispose them to a smaller head. We found a large number of shunt revisions to be associated with small head circumference. This may reflect the fact that overdrainage often leads to intermittent shunt dysfunction. Influence of Valve Type on Head Circumference The main findings of this study are supported by 2 previous studies of head circumference in shunt-treated children. Faulhauer and Schmitz 5 reported their early experience of shunt treatment, using mainly an adjustable pressure-controlled shunt (Hakim, Codman) between 1964 and 1974 in 336 adult and pediatric patients (the number of pediatric patients was not reported) where 33 of the shunt-treated infants developed marked microcephalus (below the 5th percentile) and 18 of them had shuntinduced craniosynostosis. In a study of 436 extremely low birth weight (< 1000 g) infants with Grade IV IVH, the authors found head circumference below the 10th percentile in 61% of children treated with a shunt (type not specified), compared with 35% in children without a shunt. In children without IVH and without a shunt, only 24% had a head circumference below the 10th percentile. 1 In contrast to this, a study of head circumference after shunting using a flow-regulated valve showed head circumference 486 J Neurosurg: Pediatrics / Volume 12 / November 2013

Head circumference after ventriculoperitoneal shunt treatment Fig. 2. Head circumference for shunt-treated girls from birth to 4 years of age. The reference values are from healthy children from the study by Wikland et al. Four-year data may not be valid, as the number of patients is small. The green lines represent the normal curve for girls in Sweden with ± 1 and 2 SD. The red line represents the average for girls in this study. The mean line is calculated based on the time point of the planned visits. Fig. 3. Head circumference for shunt-treated boys from birth to 4 years of age. The reference values are from healthy children from the study by Wikland et al. Four-year data may not be valid, as the number of patients is small. The green lines represent the normal curve for boys in Sweden with ± 1 and 2 SD. The red line represents the average for boys in this study. The mean line is calculated based on the time point of the planned visits. J Neurosurg: Pediatrics / Volume 12 / November 2013 487

D. Nilsson et al. Fig. 4. Box plot of SD scores for head circumference in shunt-treated children compared with healthy children at 1, 2, 3, and 4 years. Dots outside boxes represent outliers. above the 97th percentile in all 24 infants treated with VP shunts. 8 The authors concluded that the flow-regulated valve resulted in a different pattern of head growth compared with pressure-regulated shunts and that the larger head in children treated with a flow-regulated shunt valve may mislead the clinician to suspect shunt dysfunction. This suggests that flow-regulated valves may result in less severe skull growth retardation than pressure-regulated Fig. 5. Box plot of SD scores for change in head circumference shunt insertion until ages 1, 2, 3, and 4 years. Dots outside boxes represent outliers. 488 J Neurosurg: Pediatrics / Volume 12 / November 2013

Head circumference after ventriculoperitoneal shunt treatment TABLE 2: Results from bivariate linear regression analysis of predictors for head circumference SDS at age 1 year (n = 33) and at last measurement (n = 49)* 1-Yr Measurement Last Measurement Predictor of HDC SDS at Specified Age β (SE) p Value β (SE) p Value weight SDS 0.90 (0.21) 0.002 0.92 (0.21) <0.0001 height SDS 0.61 (0.25) 0.022 0.74 (0.23) 0.0034 no. of shunt revisions (1 4) 0.93 (0.52) 0.083 (NS) 0.84 (0.39) 0.034 * HDC = head circumference; NS = not significant. valves. However, currently there are no guidelines regarding the optimal valve type or valve level setting in hydrocephalic children. Future studies of effects of shunt treatment should include prospective randomized studies of differential-pressure valves at different opening pressures, and comparison with flow-controlled valves, with regard to head size. Craniosynostosis After Shunt Treatment In this study, 3 children developed craniosynostosis that required correction, and shunt treatment was a probable cause of the craniosynostosis in these cases. The number of patients who will develop slit ventricle syndrome later in life is unknown. With more balanced drainage of the ventricles and more normal growth of the skull this may have been avoided. Data on the incidence of shunt-induced craniosynostosis necessitating surgery are scarce, but the incidence of slit ventricle syndrome has been reported to range from 0.9% to as high as 37%. 2,3,11,13 16 However, the studies with the highest incidences of slit ventricle syndrome (24% and 37%) were both carried out primarily before antisiphon devices (ASDs) and adjustable shunts were available (1966 1984). 2,14 In contrast, another randomized study comparing 3 valves (a standard differential pressure valve without ASD, a pressureregulated valve with an ASD, and a flow-controlled valve) did not show any difference in the occurrence of shunt complications among these shunt designs. 4 Specifically there was no difference in the occurrence of slit-like ventricles between shunt designs. 15 Study Limitations We recognize limitations of this study. As we wanted to follow the change in head circumference over time, we decided to include the 4-year data in spite of the low sample size. The etiology of hydrocephalus is diverse and may influence brain development and skull size. By excluding children with major brain malformations, brain tumors, and cysts, we intended to avoid including children for whom the etiology itself may significantly influence skull size, regardless of shunt treatment. One of the main strengths of this study is the fact that we had a control group with head circumference from healthy children from the same geographic area. However, the controls were measured in children born between 1973 1975, thus not during the same time period as the shunttreated children. However, because Wikland et al. showed in a longitudinal study of reference values that there are J Neurosurg: Pediatrics / Volume 12 / November 2013 only minor changes over time in head circumference in healthy Swedish children, we do not expect that there were any effects of this difference in time period for data collection. 18 It is possible that children with benign familiar macrocrania have been included in the reference population. There are few reports on the prevalence of benign familiar macrocrania. A study of incidental findings in 1618 children referred to a pediatric neurology practice showed a prevalence of benign familiar macrocrania of 0.7%. It is unlikely that the prevalence is higher in a normal population than in a population referred to a pediatric neurology center. 6 Thus, even if benign familiar macrocrania is included in the normal population, it would have a minimal effect on the reference values. We did not measure ventricle size. It could not be done systematically, as we only perform CT scans on strict medical indications, that is, suspicion of shunt dysfunction, to reduce radiation exposure. Monitoring differences in change in ventricle size during the first 6 months after VP shunt insertion between valves or valve settings using frequent ultrasound examinations would have been an option, and could be pursued in future studies. Conclusions This study found significantly smaller head circumference at 2 and 3 years of age in shunt-treated children than in healthy controls. Three of the 50 children had surgical correction of craniosynostosis. Low weight and short height at birth, etiology of IVH, and the large number of shunt revisions were predictors of developing a small head. Disclosure This study was supported by research grants from the Gothenburg Medical Society, Neuro-S Up ALF 77090 (Agreement concerning research and education of doctors) grants from Petter Silfverskiölds minnesfond (Daniel Nilsson) and from Insamlingsstiftelsen för neurologisk forskning (Johanna Svensson and Helena Nelvig). 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 to the study and manuscript preparation include the following. Conception and design: Nilsson, Tisell. Acquisition of data: Svensson, Korkmaz, Nelvig. Analysis and interpretation of data: all authors. Drafting the article: Svensson. 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: Nilsson. Statistical analysis: Nilsson. Study supervision: Tisell. 489

D. Nilsson et al. References 1. Adams-Chapman I, Hansen NI, Stoll BJ, Higgins R: Neurodevelopmental outcome of extremely low birth weight infants with posthemorrhagic hydrocephalus requiring shunt insertion. Pediatrics 121:e1167 e1177, 2008 2. Benzel EC, Reeves JD, Kesterson L, Hadden TA: Slit ventricle syndrome in children: clinical presentation and treatment. Acta Neurochir (Wien) 117:7 14, 1992 3. Di Rocco C: Is the slit ventricle syndrome always a slit ventricle syndrome? Childs Nerv Syst 10:49 58, 1994 4. 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 5. Faulhauer K, Schmitz P: Overdrainage phenomena in shunt treated hydrocephalus. Acta Neurochir (Wien) 45:89 101, 1978 6. Gupta SN, Belay B: Intracranial incidental findings on brain MR images in a pediatric neurology practice: a retrospective study. J Neurol Sci 264:34 37, 2008 7. Huggare JA, Kantomaa T, Rönning O, Serlo W: Craniofacial growth in shunt-treated hydrocephalics: a four-year roentgenocephalometric follow-up study. Childs Nerv Syst 8:67 69, 1992 8. Kaiser G, Bittel M: Preliminary experiences with the Orbis- Sigma-System as a ventriculo-peritoneal shunt. Eur J Pediatr Surg 2:186 187, 1992 9. Kan P, Walker ML, Drake JM, Kestle JR: Predicting slitlike ventricles in children on the basis of baseline characteristics at the time of shunt insertion. J Neurosurg 106 (5 Suppl): 347 349, 2007 10. Kiesler J, Ricer R: The abnormal fontanel. Am Fam Physician 67:2547 2552, 2003 11. Major O, Fedorcsák I, Sipos L, Hantos P, Kónya E, Dobronyi I, et al: Slit-ventricle syndrome in shunt operated children. Acta Neurochir (Wien) 127:69 72, 1994 12. Nordic Centre for Classifications in Health Care: NOMES- CO Classification of Surgical Procedures (NCSP), version 1.15. Copenhagen: Nordic Medico-Statistical Committee (NOMESCO) (http://nomesco-eng.nom-nos.dk/filer/publika tioner/ncsp%201_15.pdf) [Accessed August 12, 2013] 13. Pudenz RH, Foltz EL: Hydrocephalus: overdrainage by ventricular shunts. A review and recommendations. Surg Neurol 35:200 212, 1991 14. Serlo W, Saukkonen AL, Heikkinen E, von Wendt L: The incidence and management of the slit ventricle syndrome. Acta Neurochir (Wien) 99:113 116, 1989 15. Tuli S, O Hayon B, Drake J, Clarke M, Kestle J: Change in ventricular size and effect of ventricular catheter placement in pediatric patients with shunted hydrocephalus. Neurosurgery 45:1329 1335, 1999 16. Walker ML, Fried A, Petronio J: Diagnosis and treatment of the slit ventricle syndrome. Neurosurg Clin N Am 4:707 714, 1993 17. Weinzweig J, Bartlett SP, Chen JC, Losee J, Sutton L, Duhaime AC, et al: Cranial vault expansion in the management of postshunt craniosynostosis and slit ventricle syndrome. Plast Reconstr Surg 122:1171 1180, 2008 18. Wikland KA, Luo ZC, Niklasson A, Karlberg J: Swedish population-based longitudinal reference values from birth to 18 years of age for height, weight and head circumference. Acta Paediatr 91:739 754, 2002 19. Windh P, Davis C, Sanger C, Sahlin P, Lauritzen C: Springassisted cranioplasty vs pi-plasty for sagittal synostosis a long term follow-up study. J Craniofac Surg 19:59 64, 2008 Manuscript submitted March 14, 2013. Accepted August 8, 2013. Portions of this work were presented in proceedings form at the 39th Annual Meeting of the International Society of Pediatric Neurosurgery, in Goa, India, October 16 20, 2011; at the meeting of the European Society of Craniofacial Surgery, Göteborg, Sweden, September 27 29, 2012; and at the 4th Meeting of the International Society for Hydrocephalus and Cerebrospinal Fluid Disorders, October 19 22, 2012, Kyoto, Japan. Please include this information when citing this paper: published online September 13, 2013; DOI: 10.3171/2013.8.PEDS1370. Address correspondence to: Daniel Nilsson, M.D., Ph.D., Department of Neurosurgery, Sahlgrenska University Hospital, Blå str 5, 3 tr, SE-413 45 Göteborg, Sweden. email: daniel.nilsson@neuro. gu.se. 490 J Neurosurg: Pediatrics / Volume 12 / November 2013