✓ Larsen syndrome is a rare genetic disorder of the connective tissue that is characterized by multiple joint dislocations, distinctive deformities of the hands and feet, characteristic facial features, kyphoscoliosis, and segmentation anomalies of the vertebrae. Diverse treatment options, including conservative observation and surgical correction, have been reported for patients who present with cervical spine pathophysiology. Differences in surgical approaches, timing of the correction, and pre- or postoperative bracing have been reported. According to the authors, the present case is the first report of a pediatric patient with Larsen syndrome in whom an asymptomatic cervical instability was treated before neurological deterioration with synchronous anterior decompression and fixation, posterior fusion and fixation, and halo placement. A review of the literature on similar patients reveals the variety of practices associated with a diagnosis of Larsen syndrome.
Marci Madera, Alvin Crawford and Francesco T. Mangano
James L. Leach, Reem Awwad, Hansel M. Greiner, Jennifer J. Vannest, Lili Miles and Francesco T. Mangano
Diagnostic criteria for hippocampal malrotation (HIMAL) on brain MRI typically include a rounded hippocampus, vertical collateral sulcus, and architectural blurring. Relationship to epileptogenesis remains speculative, and usefulness for surgical guidance is unknown. The study was performed to determine the prevalence of hippocampal rotational anomalies in a cohort of pediatric patients with intractable epilepsy undergoing evaluation for surgery and to determine the significance of this finding in the context of surgical planning.
Forty-eight surgically treated children with intractable epilepsy were compared with matched healthy subjects; reviewers were blinded to surgical side. Each temporal lobe was evaluated for rounded hippocampus, blurring, vertical collateral sulcus, wide choroidal fissure, enlarged temporal horn, low fornix, hippocampal signal, and findings of hippocampal sclerosis. A mesial temporal lobe (MTL) score was calculated by summing the number of features, and the collateral sulcus angle (CSA) was measured in each temporal lobe. Surgical side, pathological diagnosis, and imaging findings elsewhere in the brain were tabulated. Presence of HIMAL, associated imaging features, and MTL score were compared between sides, between epilepsy and control groups, in relationship to side of surgery, and in relationship to postoperative outcome.
Only 3 epilepsy patients (6.2%) and no controls exhibited all 3 features of HIMAL (p = 0.12). Eight of 48 (16.7%) epilepsy versus 2 of 48 (4.6%) control subjects had both a rounded hippocampus and vertical collateral sulcus (suggesting HIMAL) (p = 0.045). In control and epilepsy subjects, most findings were more prevalent on the left, and the left CSA was more vertical (p < 0.0001). Epilepsy subjects had higher MTL scores (z = −2.95, p = 0.002) and more acute CSAs (p = 0.04) than controls. Only lateralizing raw MTL score had a significant association with surgical side (p = 0.03, OR 7.33); however, this was not significant when hippocampal sclerosis cases were excluded. HIMAL findings were more prevalent and MTL scores were higher in patients with resections involving the temporal lobes. On group analysis, HIMAL findings did not predict eventual surgical side and did not predict outcome, although the numbers are small. In 4 patients the abnormally rotated hippocampus was resected and showed hippocampal sclerosis and/or dysplastic changes on histopathology. All of these patients had a good outcome after surgery.
While increased in prevalence in children with intractable epilepsy, imaging findings of HIMAL did not have preoperative lateralizing utility in this group. Findings of HIMAL (including round hippocampus, architectural blurring, and vertical collateral sulcus) did not predict outcome after surgery, although the small number of patients with these findings limits evaluation. In the small number of patients in which the malrotated hippocampus was removed, outcome was good. Further research is needed to continue to define this association in children with intractable epilepsy, focusing on a temporal lobe cohort.
Ellen L. Air, Yashar M. Ghomri, Rachana Tyagi, Andrew W. Grande, Kerry Crone and Francesco T. Mangano
Vagal nerve stimulators (VNSs) have been used successfully to treat medically refractory epilepsy. Although their efficacy is well established, appropriate management of infections is less clearly defined. In the authors' experience, patients who have gained a benefit from VNS implantation have been reluctant to have the device removed. The authors therefore sought conservative management options to salvage infected VNS systems.
The authors performed a retrospective review of 191 (93 female and 98 male) consecutive patients in whom VNS systems were placed between 2000 and 2007.
They identified 10 infections (5.2%). In 9 of 10 patients the cultured organism was Staphylococcus aureus. Three (30%) of 10 patients underwent early removal (within 1 month) of the VNS as the initial treatment. The remaining 7 patients were initially treated with antibiotics. Two (28.6%) of these patients were successfully treated using antibiotics without VNS removal. Patients in whom conservative treatment failed were given cephalexin as first-line antibiotic treatment. All patients recovered completely regardless of treatment regimen.
This study confirms the low rate of infection associated with VNS placement and suggests that, in the case of infection, treatment without removal is a viable option. However, the authors' data suggest that oral antibiotics are not the best first-line therapy.
Weihong Yuan, Scott K. Holland, Blaise V. Jones, Kerry Crone and Francesco T. Mangano
Diffusion tensor (DT) imaging was used in children with supratentorial tumors to evaluate the anisotropic diffusion properties between different tumor grades and between tumors and adjacent and contralateral white matter.
In this retrospective review, the authors review the cases of 16 children (age range 1–18 years) who presented to their institution with supratentorial tumors and were treated between 2004 and 2007. Eleven patients had low-grade and 5 had high-grade tumors. Fractional anisotropy (FA), mean diffusivity, and axial (λ∥) and radial (λ⊥) eigenvalues within selected regions were studied. Mitotic index, necrosis, and vascularity of the tumors were compared with DT imaging parameters.
The mean diffusivity was significantly higher in low-grade than in high-grade tumors (p = 0.04); the 2 tumor grades also significantly differed for both λ∥ (p < 0.05) and λ⊥ (p < 0.05). Mean diffusivity values in low-grade tumors were significantly higher than in adjacent normal-appearing white matter (NAWM; p = 0.0004) and contralateral NAWM (p = 0.0001). In both low- and high-grade tumors, the FA was significantly lower than in NAWM (p < 0.0001 and p < 0.03, respectively) and contralateral NAWM (p < 0.0001 and p < 0.003, respectively). Tumor cellularity highly correlated with mean diffusivity and λ∥and λ⊥.
Diffusion tensor imaging is a useful tool in the evaluation of supratentorial tumors in children. The mean diffusivity appears to be a significant marker in differentiating tumors grades. Findings related to λ∥ and λ⊥ within tumor groups and between tumors and NAWM may be an indirect manifestation of the combined effects of axonal injury, demyelination, and tumor mass within the cranial compartment.
Francesco T. Mangano and Matthew D. Smyth
Joshua K. Schaffzin, Katherine Simon, Beverly L. Connelly and Francesco T. Mangano
Surgical site infections (SSIs) are costly to patients and the health care system. Pediatric neurosurgery SSI risk factors are not well defined. Intraoperative protocols have reduced, but have not eliminated, SSIs. The effect of preoperative intervention is unknown. Using quality improvement methods, a preoperative SSI prevention protocol for pediatric neurosurgical patients was implemented to assess its effect on SSI rate.
Patients who underwent a scheduled neurosurgical procedure between January 2014 and December 2015 were included. Published evidence and provider consensus were used to guide preoperative protocol development. The Model for Improvement was used to test interventions. Intraoperative and postoperative management was not standardized or modified systematically. Staff, family, and overall adherence was measured as all-or-nothing. In addition, SSI rates among eligible procedures were measured before and after protocol implementation.
Within 4 months, overall protocol adherence increased from 51.3% to a sustained 85.7%. SSI rates decreased from 2.9 per 100 procedures preintervention to 0.62 infections postintervention (p = 0.003). An approximate 79% reduction in SSI risk was identified (risk ratio 0.21, 95% CI 0.08–0.56; p = 0.001).
Clinical staff and families successfully collaborated on a standardized preoperative protocol for pediatric neurosurgical patients. Standardization of the preoperative phase of care alone reduced SSI rates. Attention to the preoperative in addition to the intraoperative and postoperative phases of care may lead to further reduction in SSI rates.
James T. Rutka
Ellen L. Air, Weihong Yuan, Scott K. Holland, Blaise V. Jones, Karin Bierbrauer, Mekibib Altaye and Francesco T. Mangano
The goal in this study was to compare the integrity of white matter before and after ventriculoperitoneal (VP) shunt insertion by evaluating the anisotropic diffusion properties with the aid of diffusion tensor (DT) imaging in young children with hydrocephalus.
The authors retrospectively identified 10 children with hydrocephalus who underwent both pre- and postoperative DT imaging studies. The DT imaging parameters (fractional anisotropy [FA], mean diffusivity, axial diffusivity, and radial diffusivity) were computed and compared longitudinally in the splenium and genu of the corpus callosum (gCC) and in the anterior and posterior limbs of the internal capsule (PLIC). The patients' values on DT imaging at the pre- and postshunt stages were compared with the corresponding age-matched controls as well as with a large cohort of healthy children in the database.
In the gCC, 7 of 10 children had abnormally low preoperative FA values, 6 of which normalized postoperatively. All 3 of the 10 children who had normal preoperative FA values had normal FA values postoperatively as well. In the PLIC, 7 of 10 children had abnormally high FA values, 6 of which normalized postoperatively, whereas the other one had abnormally low postoperative FA. Of the remaining 3 children, 2 had abnormally low preoperative FA values in the PLIC; this normalized in 1 patient after surgery. The other child had a normal preoperative FA value that became abnormally low postoperatively. When comparing the presurgery frequency of abnormally low, normal, and abnormally high FA values to those postsurgery, there was a statistically significant longitudinal difference in both gCC (p = 0.02) and PLIC (p = 0.002).
In this first longitudinal DT imaging study of young children with hydrocephalus, DT imaging anisotropy yielded abnormal results in several white matter regions of the brain, and trended toward normalization following VP shunt placement.
Francesco T. Mangano, Jose A. Menendez, Tracy Habrock, Prithvi Narayan, Jeffrey R. Leonard, Tae Sung Park and Matthew D. Smyth
The use of adjustable differential pressure valves has been recommended to improve ventriculoperitoneal (VP) shunt performance in selected patients; however, published data are scarce regarding their clinical reliability. Recently, the identification of a number of malfunctioning programmable valves during shunt revision surgery in children prompted a retrospective review of valve performance in this patient cohort.
The authors performed a retrospective chart analysis of 100 patients with programmable valve shunts and 89 patients with nonprogrammable valve shunts implanted at the St. Louis Children's Hospital between April 2002 and June 2004. They noted the cause of hydrocephalus, the type of shunt malfunction, and cerebrospinal fluid (CSF) protein levels. Regular clinical follow up ranged from 1 to 26 months, with a mean follow-up time of 9.75 months for patients with programmable valves and 10.4 months for patients with nonprogrammable valves.
Patient ages ranged from 2 weeks to 18 years. One hundred patients had 117 programmable valves implanted, and 35 of these patients (35%) underwent shunt revision because of malfunction. The programmable valve itself malfunctioned in nine patients who had undergone shunt revision (11.1%/year of follow up). The nonprogrammable valve group had no valve malfunctions. The overall VP shunt revision rate in the nonprogrammable valve group was 20.2%. No significant differences were identified when CSF protein levels and specific malfunction types were compared within the programmable valve and nonprogrammable valve groups.
In this study the authors demonstrated an annualized intrinsic programmable valve malfunction rate of 11.1%, whereas during the same period no intrinsic valve malfunctions were noted with nonprogrammable valve systems for similar causes of hydrocephalus. The CSF protein levels did not correlate with observed valve malfunction rates. Further evaluation in a prospective, randomized fashion will elucidate specific indications for programmable valve systems and better determine the reliability of these valves in the pediatric population.
James M. Johnston Jr., Francesco T. Mangano, Jeffrey G. Ojemann, Tae Sung Park, Edwin Trevathan and Matthew D. Smyth
The purpose of this study was to better define the incidence of complications associated with placement of subdural electrodes for localization of seizure foci and functional mapping in children.
The authors retrospectively reviewed the records of 112 consecutive patients (53 boys, 59 girls; mean age 10.9 years, range 10 months–21.7 years) with medically intractable epilepsy who underwent invasive monitoring at the Pediatric Epilepsy Center at St. Louis Children’s Hospital between January 1994 and July 2005. There were 122 implantation procedures (85 grids and strips, 32 strips only, five grids only, four with additional depth electrodes), with a mean monitoring period of 7.1 days (range 2–21 days). Operative complications included the need for repeated surgery for additional electrode placement (5.7%); wound infection (2.4%); cerebrospinal fluid leak (1.6%); and subdural hematoma, symptomatic pneumocephalus, bone flap osteomyelitis, and strip electrode fracture requiring operative retrieval (one patient [0.8%] each). There were four cases of transient neurological deficit (3.3%) and no permanent deficit or death associated with invasive monitoring.
Placement of subdural grid and strip electrodes for invasive video electroencephalographic monitoring is generally well tolerated in the pediatric population. The authors found that aggressive initial electrode coverage was not associated with higher rates of blood transfusion or perioperative complications, and reduced the frequency of repeated operations for placement of supplemental electrodes.