Bruce A. Kaufman

Practice Guidelines for the Management of Bacterial Meningitis

Allan R. Tunkel, Barry J. Hartman, Sheldon L. Kaplan, Bruce A. Kaufman, Karen L. Roos, W. Michael Scheld, and Richard J. Whitley Drexel University College of Medicine, Philadelphia, Pennsylvania; Weill Cornell Medical Center, New York, New York; Baylor College of Medicine, Houston, Texas; Medical College of Wisconsin, Milwaukee; Indiana University School of Medicine, Indianapolis; University of Virginia School of Medicine, Charlottesville; and University of Alabama at Birmingham

Neuroblastoma: Effect of genetic factors on prognosis and treatment

Background and Methods. Genetic analysis of tumor tissue has provided considerable insight into mechanisms of malignant transformation and progression. Neuroblastomas have been studied by cytogenetics, flow cytometry, and molecular genetic techniques, and these studies have identified several specific abnormalities that allow subclassification of these tumors into genetic/ clinical subtypes.

Sagittal craniosynostosis outcome assessment for two methods and timings of intervention.

A retrospective quantitative analysis of 40 infants who underwent surgery for sagittal craniosynostosis was conducted to determine whether any difference in outcome, with respect to cranial index (cranial width/cranial length x 100), could be associated with either the age at surgery or the extent of the operation. Children < or = 13 months old at surgery and for whom there were archived computed tomography digital data preoperatively, perioperatively, and 1 year postoperatively were studied. For statistical analysis, the operation was classified as either extended strip craniectomy or subtotal calvarectomy, and the age at operation was either < or = 4 months or > 4 months. Twenty-eight patients underwent extended strip craniectomy at a mean age of 5.1 months. Their mean cranial index preoperatively was 67 versus 71 at 1 year postoperatively (p < 0.0001). Of extended strip craniectomy patients, 15 were operated on at age < or = 4 months (mean = 2.9 months) and 13 at age > 4 months (mean = 7.6 months). Mean cranial indices for age at operation groups did not achieve age-appropriate normal range values 1 year postoperatively for either group, and there was no significant difference between the mean percentages of improvement achieved (p = 0.143). Twelve patients underwent subtotal calvarectomy at a mean age of 5.2 months. Their mean cranial index preoperatively was 66 versus 74 at 1 year postoperatively (p < 0.0001). The mean cranial index in this group reached age-appropriate normal range values 1 year postoperatively. The percentage improvement in cranial index 1 year after subtotal calvarectomy was greater than after extended strip craniectomy (p = 0.003). Extended strip craniectomy for sagittal craniosynostosis does not achieve normal cranial width:length proportions, even when performed before 4 months of age. Subtotal calvarectomy for sagittal craniosynostosis does achieve normal cranial width:length proportions in the majority of the children, at least when performed within the first 13 months of life.

Prospective study of recovery following neonatal brachial plexus injury.

The prognosis for recovery from brachial plexus injury sustained at or before birth is generally favorable. However, roughly 10% of these infants remain profoundly weak and later exhibit functional disability in the affected arm. Early identification of these at-risk infants would be helpful in selecting patients for surgical management. In our prospective study, 80 infants with brachial plexus injury were examined on a monthly basis. Complete recovery occurred in 53 (66%); in 9 (11%), mild weakness persisted. In each child, recovery to antigravity strength in the biceps, triceps, and deltoid was noted by 6 months of age. Moderate arm weakness persisted in 7 children (9%); none had antigravity strength in the deltoid at age 6 months. Eleven children (14%) had severe permanent weakness (mean follow-up: 4.4 years). At age 6 months, these individuals exhibited at best 2/5 strength proximally and typically 0-1/5 strength in the wrist and finger extensors. Our results demonstrate that detailed strength testing up to 6 months of age predicts not only complete recovery of neonatal brachial plexus injury but also those children destined for long-term severe disability. (J Child Neurol 2001;16:488-492).

Neural tube defects

Defects of development of the neural tube can result in a number of seemingly different malformations. Understanding the abnormal embryology helps one understand the malformations and their surgical treatments. The clinical presentations and the follow-up of these patients require attention to various end organs besides the nervous system. For most of these conditions, long-term follow-up is necessary regardless of initial treatment. A decline in function is not a part of the natural history of these malformations and requires prompt evaluation and treatment.

Quadriplegia after chiropractic manipulation in an infant with congenital torticollis caused by a spinal cord astrocytoma

An infant with congenital torticollis underwent chiropractic manipulation, and within a few hours had respiratory insufficiency, seizures, and quadriplegia. A holocord astrocytoma, with extensive acute necrosis believed to be a result of the neck manipulation, was found and resected. We believe that every child with torticollis, regardless of age, should undergo neurologic and radiologic evaluation before any form of physical treatment is instituted.

Selective Lumbosacral Dorsal Rhizotomy Immediately Caudal to the Conus Medullaris for Cerebral Palsy Spasticity

We describe a variation of selective dorsal rhizotomy for spastic cerebral palsy that involves sectioning of the dorsal spinal roots immediately caudal to the conus medullaris. The operation entails an L1-L2 laminectomy, ultrasonographic localization of the conus medullaris, and partial deafferentation of the L1-S2 roots with electromyographic testing under an operating microscope. In 66 children with cerebral palsy, the operation reduced spasticity in the lower extremity without complications, e.g., motor weakness, neurogenic bladder, and sensory loss. It offers several important advantages over alternative techniques.

Ventriculoperitoneal Shunt Infections with Gram-Negative Bacteria

Infection causes major morbidity and mortality in patients with cerebrospinal fluid (CSF) shunts. The prognosis of CSF shunt infections caused by Gram-negative bacteria (GNB) has been thought to be particularly poor. The authors reviewed all GNB shunt infections treated at Childrens Memorial Hospital from January 1986 to January 1990 (n = 23). Of these infections 20 (87%) occurred within 4 weeks after shunt revision (median, 10 days). The most frequent symptoms were fever, lethargy, and irritability; the illness was not severe in the majority of these patients. Escherichia coli was isolated from 12 of 23 patients (52%), Klebsiella pneumoniae from 5 (22%), and mixed GNB from 3 (13%) patients. Initial treatment always included immediate shunt removal, externalized ventricular drainage, and intravenous antibiotics. Extraventricular drainage revision and/or intraventricular antibiotics were required in four patients whose CSF cultures were persistently positive for GNB. At admission, these patients had CSF glucose levels of < 10 mg/dl and CSF positive for GNB by Grams stain. The overall cure rate was 100%, and no recurrence was observed; however, a subsequent infection with a different organism developed in four patients. Only 2 of 19 patients (11%) who were followed up suffered apparent CNS damage. One patient died of unrelated causes shortly after treatment. Our findings indicate that 1) patients with GNB CSF shunt infections often appear relatively well at presentation; 2) CSF positive for GNB by Grams stain and very low CSF glucose levels predict continued positive CSF cultures, despite appropriate antibiotic therapy; and 3) GNB CSF shunt infections can be successfully treated by prompt shunt removal, extraventricular drainage, and intravenous antibiotics.(ABSTRACT TRUNCATED AT 250 WORDS)

Outcome of surgically treated birth-related brachial plexus injuries in twenty cases.

Birth-related brachial plexus injury occurs in 0.19-2.5 per 1,000 live births, of which 70-92% improve with conservative management. With the advent of microsurgical techniques, patients who fail expectant treatment may benefit from brachial plexus exploration and reconstruction. From 1991 to 1996, 87 patients were referred to the multidisciplinary brachial plexus clinic at St. Louis Childrens Hospital. Twenty patients were selected for surgical management. The average age at surgery was 10.5 months (range 3-35, median = 8), with an average follow-up of 23.9 months (range 7-45, median = 24). Two patients were lost to follow-up. Surgical procedures included neurolysis (n = 8), neurotization (n = 2), nerve grafting (n = 5), and a combination (n = 3) of the above. Two patients underwent exploration without repair. Intercostal nerves, pectoral nerves, and C4 roots were used for neurotizations, and the sural nerve was used for nerve grafting. Results from 18 patients were available for follow-up review. Fifteen patients (83% demonstrated clinical improvement postoperatively. Of the 3 patients without improvement, 2 underwent exploration without repair, and one underwent neurolysis of the axillary nerve. Of patients undergoing reconstruction, 93% had improved strength postoperatively. No subjects had worsening neurologic status, and there were no complications. These results suggest that surgery for birth-related brachial plexus injury may show favorable outcomes if patients are selected appropriately. Patients undergoing neurolysis and nerve grafting had more favorable outcomes than those undergoing neurotization.

Normal Diameter of Filum terminate in Children: In vivo Measurement

The hallmarks of tethered cord syndrome are a low-lying conus medullaris and a thick filum terminale. In diagnosing the tethered cord syndrome, the thick filum terminale is often defined as that greater than 2 mm in diameter. The cutoff of 2 mm in diameter was derived from myelographic measurements a few decades ago, and the true normal diameter of the filum terminale diameter in children or adults remains unknown. We measured the diameters of the filum terminale in vivo in the operating room on 31 children (age range = 2-14 years; mean age = 5 years) undergoing selective dorsal rhizotomy for spastic cerebral palsy. None of them had clinical evidence of tethered cord syndrome. The conus medullaris and filum terminale were videotaped intraoperatively and images were transferred to an image analyzer; the filum diameters at 10 and 15 mm caudal to the conus medullaris were then measured extraoperatively using the computer graphics system. The diameter of the filum at 10 and 15 mm caudal to the conus was 1,211 +/- 209 and 1,163 +/- 245 microm (mean +/- SD), respectively. In all children except one, the conus medullaris ended above the L2 level. The data indicate that filum terminales greater than 2 mm in diameter in children are abnormally thick.