✓The authors present the case of a child with an untreated lipomyelomeningocele who developed an acquired Chiari malformation Type I (CM-I) with a large syrinx over the course of 3 years. To the best of the authors' knowledge, this is the first report to document a case in which an acquired CM-I evolved in a patient with an untreated tethered cord.
Ben Waldau, Gerald Grant, and Herbert Fuchs
Report of two cases
Ciaran J. Powers, Timothy George, and Herbert E. Fuchs
✓Ventriculoperitoneal (VP) shunt failure is a common problem encountered by pediatric neurosurgeons. The majority of such failures are due to obstruction of the device. Conditions in which intraabdominal pressure is chronically elevated, such as pregnancy, have been associated with shunt failure. Chronic constipation may also result in abnormally elevated intraabdominal pressure and may be an underrecognized cause of distal VP shunt failure. The authors describe the cases of two children who presented with clinical and imaging evidence of VP shunt failure and who were also severely constipated. Treatment of their constipation resulted in both clinical and imaging-documented resolution of their shunt failure.
Srinivasan Mukundan, Herbert Fuchs, Michael J. Alexander, and Gerald A. Grant
✓The authors report the first clinical use of 3-tesla dynamic contrast-enhanced magnetic resonance (MR) angiography for the diagnosis of a vascular malformation in a pediatric patient. The supply and drainage of an arteriovenous malformation were accurately demonstrated on MR angiography, which was performed without sedating the patient. This lesion was confirmed on catheter angiography, and definitive treatment via embolization was undertaken in a single session. The patient's therapeutic response will be followed with surveillance dynamic MR imaging.
Carrie R. Muh, Naomi D. Chou, Shervin Rahimpour, Jordan M. Komisarow, Tracy G. Spears, Herbert E. Fuchs, Sandra Serafini, and Gerald A. Grant
To determine resection margins near eloquent tissue, electrical cortical stimulation (ECS) mapping is often used with visual naming tasks. In recent years, auditory naming tasks have been found to provide a more comprehensive map. Differences in modality-specific language sites have been found in adult patients, but there is a paucity of research on ECS language studies in pediatric patients. The goals of this study were to evaluate word-finding distinctions between visual and auditory modalities and identify which cortical subregions most often contain critical language function in a pediatric population.
Twenty-one pediatric patients with epilepsy or temporal lobe pathology underwent ECS mapping using visual (n = 21) and auditory (n = 14) tasks. Fisher’s exact test was used to determine whether the frequency of errors in the stimulated trials was greater than the patient’s baseline error rate for each tested modality and subregion.
While the medial superior temporal gyrus was a common language site for both visual and auditory language (43.8% and 46.2% of patients, respectively), other subregions showed significant differences between modalities, and there was significant variability between patients. Visual language was more likely to be located in the anterior temporal lobe than was auditory language. The pediatric patients exhibited fewer parietal language sites and a larger range of sites overall than did adult patients in previously published studies.
There was no single area critical for language in more than 50% of patients tested in either modality for which more than 1 patient was tested (n > 1), affirming that language function is plastic in the setting of dominant-hemisphere pathology. The high rates of language function throughout the left frontal, temporal, and anterior parietal regions with few areas of overlap between modalities suggest that ECS mapping with both visual and auditory testing is necessary to obtain a comprehensive language map prior to epileptic focus or tumor resection.
Shahid M. Nimjee, Ciaran J. Powers, Roger E. McLendon, Gerald A. Grant, and Herbert E. Fuchs
Cerebrospinal fluid overproduction resulting in communicating hydrocephalus is observed in patients who have choroid plexus papilloma or choroid plexus carcinoma. Less often, patients with these conditions have diffuse villous hyperplasia. Prior studies report CSF production greater than 3 L per day in these patients. These patients are treated with CSF shunting or by either unilateral choroid plexectomy or staged bilateral choroid plexectomy. The authors present a patient who had a number of congenital anomalies and a karyotype that revealed balanced translocations, 5 to 7 and 9 to 11. She presented with hydrocephalus and had CSF production of 5 L per day, greater output than ever previously reported. She was treated with a single-stage bilateral choroid plexectomy. Histopathological analysis revealed a bilateral choroid plexus papilloma. Postoperatively, the patient responded well clinically and showed radiographic improvement of her hydrocephalus. Bilateral choroid plexus papilloma has been reported in the literature as a cause for neonatal and congenital hydrocephalus. It can result in high CSF output and can be successfully treated with a single-stage bilateral choroid plexectomy. Further studies are ongoing to identify genes involved in embryogenesis of the choroid plexus.
Stephen R. Parker, Peggy Harris, Thomas J. Cummings, Timothy George, Herbert Fuchs, and Gerald Grant
Posterior fossa decompression with duraplasty for Chiari malformation Type I (CM-I) is a common pediatric neurosurgery procedure. Published series report a complication rate ranging from 3% to 40% for this procedure. Historically, many dural substitutes have been used, including bovine grafts, human cadaveric pericardium, synthetic dura, and autologous pericranium. The authors hypothesized that a recently observed increase in complications was dependent on the graft used.
Between January 2004 and January 2008, 114 consecutive patients ≤ 18 years old underwent primary CM-I decompression using duraplasty. Records were retrospectively reviewed for short- and intermediate-term complications and operative technique, focusing on the choice of duraplasty graft with or without application of a tissue sealant.
The average age of the patients was 8.6 years. The dural graft used was variable: 15 were treated with cadaveric pericardium, 12 with Durepair, and 87 with EnDura. Tisseel was used in 75 patients, DuraSeal in 12, and no tissue sealant was used in 27 patients. The overall complication rate was 21.1%. The most common complications included aseptic meningitis, symptomatic pseudomeningocele, or a CSF leak requiring reoperation. The overall complication rates were as follows: cadaveric pericardium 26.7%, Durepair 41.7%, and EnDura 17.2%; reoperation rates were 13%, 25%, and 8.1%, respectively. Prior to adopting a different graft product, the overall complication rate was 18.1%; following the change the rate increased to 35%. Complication rates for tissue sealants were 14.8% for no sealant, 18.7% for Tisseel, and 50% for DuraSeal. Nine patients were treated with the combination of Durepair and DuraSeal and this subgroup had a 56% complication rate.
Complication rates after CM-I decompression may be dependent on the dural graft with or without the addition of tissue sealant. The complication rate at the authors' institution approximately doubled following the adoption of a different graft product. Tissue sealants used in combination with a dural substitute to augment a duraplasty may increase the risk of aseptic meningitis and/or CSF leak. The mechanism of the apparent increased inflammation with this combination remains under investigation.
Sharad Rajpal, R. Shane Tubbs, Timothy George, W. Jerry Oakes, Herbert E. Fuchs, Mark N. Hadley, and Bermans J. Iskandar
Children with spina bifida occulta require early surgery to prevent neurological deficits. The treatment of patients with a congenitally tethered cord who present in adulthood remains controversial.
The authors studied the medical records of 61 adult patients who underwent surgical untethering for spina bifida occulta at three institutions between 1994 and 2003. Patients who had undergone prior myelomeningocele repair or tethered cord release surgery were excluded.
The most common intraoperative findings were lipomyelomeningocele (41%) and a tight terminal filum (36%). The follow-up duration ranged from 10.8 to 149.5 months. Of the 34 patients with back pain, status improved in 65%, worsened in 3%, remained unchanged in 18%, and improved and later recurred in 15%. Lower-extremity pain improved in 16 patients (53%), remained unchanged in 23%, improved and then recurred in 17%, and worsened in 7%. Lower-extremity weakness improved in 47%, remained unchanged in 47%, and improved and then recurred in 5%. Finally, of the 17 patients with lower-extremity sensory changes, status improved in 35%, remained unchanged in 35%, and the information on five patients was unavailable. Surgical complications included three wound infections, one cerebrospinal fluid leak, and two pseudomeningoceles requiring surgical revision. One patient developed acute respiratory distress syndrome and sepsis postoperatively and died several days later.
Adult-age presentation of a congenital tethered cord is unusual. Despite a slight increase in postoperative neurological injury in adults, surgery has relatively low risk and offers good potential for neurological improvement or stabilization. As they do in children, the authors recommend early surgery in adults with this disorder. The decision to undertake surgery, however, should be modulated by other factors such as a patient's general medical condition and risk posed by anesthesia.
Ben Waldau, Leahthan F. Domeshek, Fawn A. Leigh, Kristian C. Lum, Herbert E. Fuchs, Jeffrey R. Marcus, Srinivasan Mukundan, and Gerald A. Grant
The case of a 3-year-old patient with tuberous sclerosis and a 13-mm Chiari malformation Type I that spontaneously disappeared over the course of 4 years is presented. Using morphometric measurements of the posterior fossa and cerebellum in this patient, the authors show that the volume of the posterior fossa at the time of initial evaluation was consistent with that reported as normal in the literature (180.24 cm3; normal volume 132–198 cm3). Moreover, the patient showed a normal rate of growth of his posterior fossa over the period of observation (201.05 cm3; normal range 153–230 cm3). Cerebellar volumes were found to increase only minimally during this time period, which is compatible with observations in healthy controls. The posterior fossa volume, on the other hand, was shown to increase significantly more than that of the cerebellum (p = 0.0185). This differential growth may permit the tonsils to ascend back up into the posterior fossa. Therefore, pediatric patients with normal posterior fossa volumes and normal development may have a spontaneous resolution of their asymptomatic Chiari malformation Type I.
Karen Monuszko, Michael Malinzak, Lexie Zidanyue Yang, Donna Niedzwiecki, Herbert Fuchs, Carrie R. Muh, Krista Gingrich, Robert Lark, and Eric M. Thompson
Patients with shunted hydrocephalus often accumulate high levels of radiation over their lifetimes during evaluation of hardware integrity. Current practice involves the use of a series of conventional radiographs for this purpose. Newer low-dose EOS radiography is currently used to evaluate scoliosis but has not been explored to evaluate shunt integrity on a large scale. The goal of this study was to compare the quality of imaging using EOS low-dose radiography to conventional radiography to evaluate shunt tubing.
A retrospective chart review was performed on 57 patients who previously had both conventional radiographs and low-dose EOS images of their cerebral shunt tubing from 2000 to 2018. Patient demographics (age, sex, type of shunt tubing, primary diagnosis) were collected. Conventional radiographic images and low-dose EOS images were independently analyzed by a neurosurgeon and neuroradiologist in three categories: image quality, delineation of shunt, and distinction of shunt compared to adjacent anatomy.
All patients had shunted hydrocephalus due to spina bifida and Chiari type II malformation. Ratings of EOS and conventional radiographic images by both raters did not differ significantly in terms of image quality (rater 1, p = 0.499; rater 2, p = 0.578) or delineation of shunt (p = 0.107 and p = 0.256). Conventional radiographic images received significantly higher ratings than EOS on the ability to distinguish the shunt versus adjacent anatomy by rater 1 (p = 0.039), but not by rater 2 (p = 0.149). The overall score of the three categories combined was not significantly different between EOS and conventional radiography (rater 1, p = 0.818; rater 2, p = 0.186). In terms of cost, an EOS image was less costly than a conventional radiography shunt series ($236–$366 and $1300–$1547, respectively). The radiation dose was also lower for EOS images, with an effective dose of 0.086–0.140 mSv compared to approximately 1.6 mSv for a similar field of view with conventional radiography.
The image quality of low-dose EOS radiography does not significantly differ from conventional radiography for the evaluation of cerebral shunts. In addition, EOS affords a much lower radiation dose and a lower cost.
Albert Moghrabi, Henry S. Friedman, David M. Ashley, Krystal S. Bottom, Tracy Kerby, Elizabeth Stewart, Carol Bruggers, James M. Provenzale, Martin Champagne, Linda Hershon, Melody Watral, Janis Ryan, Karima Rasheed, Shelley Lovell, David Korones, Herbert Fuchs, Timothy George, Roger E. McLendon, Allan H. Friedman, Edward Buckley, and Darryl C. Longee
In this study, the authors sought to investigate the response rate and toxicity of carboplatin in patients with progressive low-grade glioma (LGG). Thirty-two patients with progressive LGG were treated with carboplatin at a dosage of 560 mg/m2. Treatment was given at 4-week intervals and continued until the disease progressed, unacceptable toxicity supervened, or for 12 additional courses after achieving maximal response. Patients with stable disease were treated with a total of 12 cycles. All patients were treated as outpatients. Patients were evaluated for response to treatment and toxicity.
All patients received a minimum of two cycles of carboplatin, and were examined for response. A partial response was achieved in nine patients (28%) and a minimal response in two (6%), for an overall response rate of 34% (11 of 32 patients). Eighteen patients (56%) had stable disease. A partial response was achieved in the nine patients after a median of six cycles (range 4-11 cycles), a minimal response was achieved in the two patients after five cycles. Glioma progression was noted in three patients after three, five, and five cycles, respectively. The 11 patients in whom some response was achieved had either an optic pathway tumor or a juvenile pilocytic astrocytoma. Twenty-six of the 32 patients had those characteristics, making the response rate in that group 42% (11 of 26 patients). Thirty-two patients received a total of 387 cycles of chemotherapy. Hematological toxicity was moderate. Twenty-one patients developed thrombocytopenia (platelet count < 50,000/μl); three patients required one platelet transfusion each. Nine patients developed neutropenia (absolute neutrophil count < 500/μl); one developed fever and required administration of antibiotic agents. One dose adjustment in each of the patients prevented further thrombocytopenia and neutropenia. Two patients with stable disease died of respiratory complications. One patient developed Grade III ototoxicity after receiving five cycles, one patient developed hypersensitivity to carboplatin, and none developed nephrotoxicity.
Carboplatin given at a dosage of 560 mg/m2 every 4 weeks has activity in patients with progressive LGG. This drug regimen is relatively simple and well tolerated. Further investigation and longer follow-up study are warranted.