Inhibition of peridural fibrosis after laminectomy using low-dose external beam radiation in a dog model.

OBJECTIVE: Clinical studies have demonstrated a significant association between the presence of extensive postlumbar discectomy peridural scar formation and the recurrence of low back and radicular pain. Low-dose perioperative radiation therapy has previously been demonstrated to inhibit peridural fibrosis after laminectomy in a rat model. The current study was designed to measure the effect of low-dose radiation on postlaminectomy peridural fibrosis development in a larger animal model. METHODS: Three dogs underwent a total of 12 lumbar hemilaminectomies. For each animal, two levels received 1) external beam radiation 24 hours before surgery, 2) surgery alone, or 3) radiation alone. Radiation was administered in a single fraction of 700 cGy using computed tomographic guidance for dosimetry planning. The isodose distribution was such that the dose conformed to the posterior epidural space with minimal exit dose to normal tissue. Port films were used to confirm the correct levels. Gadolinium-enhanced magnetic resonance imaging (MRI) of the lumbar spines was obtained before the animals were killed 12 weeks after surgery. The spines were harvested, and axial sections through the laminectomy defect were stained with hematoxylin and eosin and Masson's trichrome. All specimens were evaluated for extent of fibrosis along the dura, density of fibrosis, nerve root entrapment, and sublaminar fibrosis. RESULTS: There were no complications from the surgery, and no new neurological deficits were noted. There was a statistically significant difference between the irradiated and nonirradiated groups regarding the extent of fibrosis (P = 0.001) and the density of fibroblasts (P = 0.001). There was also a marked difference in nerve root entrapment (P = 0.182) and the presence of sublaminar fibrosis (P = 0.061) between the treatment and control groups. MRI revealed less gadolinium enhancement at the irradiated levels compared with the nonirradiated levels, confirming the usefulness of MRI in predicting the degree of epidural fibrosis. CONCLUSION: Low-dose external beam radiation therapy administered 24 hours before laminectomy in a dog model significantly decreased the extent and density of peridural fibrosis as well as nerve root entrapment and sublaminar fibrosis. This treatment strategy may be efficacious in patients with recurrent radicular pain after lumbar discectomy that is thought to be secondary to peridural fibrosis on the basis of gadolinium-enhanced MRI studies, and who might benefit from reoperation for nerve root decompression.

Inhibition of peridural fibrosis after laminectomy using low-dose external beam radiation in a rat model.

OBJECTIVE: Clinical studies have revealed a significant association between the presence of extensive postlumbar discectomy peridural scar formation and the reoccurrence of low back and radicular pain. Low-dose perioperative radiation therapy has been shown to inhibit scar formation. Its effect on peridural fibrosis, however, has not been studied. METHODS: Thirty male Sprague-Dawley rats underwent L5 laminectomies. Ten rats each received a single fraction of 700-cGy external beam radiation to the lumbar spine 24 hours before surgery; 10 rats each received 700 cGy 24 hours after surgery. The remaining 10 rats served as a control group. All of the rats were killed 30 days after surgery. The spines were harvested, and axial histological sections through the laminectomy defect were evaluated. Each specimen was scored for extent, density, and arachnoidal involvement by fibrosis. RESULTS: There was a statistically significant difference between the treatment and control groups regarding the extent of fibrosis along the dura (P < 0.001), the density of fibroblasts (P < 0.005), and the arachnoid involvement (P < 0.01). There was no difference in fibrosis reduction between the groups receiving pre- and postlaminectomy radiation. CONCLUSION: Low-dose external beam radiation therapy administered before or after laminectomy in a rat model significantly decreases the extent, density, and arachnoidal involvement of peridural fibrosis. This technique may improve the outcome of patients who undergo reoperations for recurrent radicular and/or low back pain after successful lumbar discectomy in whom there is a significant amount of peridural fibrosis.

Anterior thoracic corpectomy without sternotomy: a strategy for malignant disease of the upper thoracic spine.

BACKGROUND: With increasing frequency, spine surgeons are being asked to provide decompression and stabilization in patients with spinal metastases. While no region of the spine is easily treated, the upper thoracic spine is perhaps the least accessible. Traditional approaches to this region involve either thoracotomy or at least limited sternotomy. The authors present an approach to anterior pathology of the upper thoracic spine that obviates the need for sternotomy. METHODS: Within the past two years, two patients with cervicothoracic metastases underwent anterior decompression and fusion without sternotomy. In both patients, the bodies of C7, T1, and T2 were removed. While both patients were prepared and draped for sternotomy, each required a neck dissection only. In both patients, left-sided incisions were made along the leading edge of the sternocleidomastoid. The platysma was divided with the overlying skin. With further dissection, the strap muscles were tagged and divided approximately one centimeter above their sternal attachments. The loose areolar tissue of the superior mediastinum was then bluntly dissected. Along the entire length of the incision, the vascular plane medial to the carotid sheath was developed to facilitate exposure of the anterior spine. A Farley-Thompson retractor system was then employed to retract and protect the superior mediastinal structures. With this exposure, corpectomies were carried out using a high speed drill. Fusion was accomplished through insertion of Steinmann pins into the adjacent intact bodies above and below. This was followed by application of methyl methacrylate. Both patients had immediate postoperative stability with preservation of spinal cord function. Both patients subsequently underwent removal of dorsally located tumor with posterior fusion. CONCLUSIONS: The goal of cancer surgery is to provide for increased functional survival without undue morbidity. The authors feel that when possible, the pain of sternal and clavicular osteotomies should be avoided. The described approach works well in conjunction with a methyl methacrylate/Steinmann pin construct. Because of the intact sternum, the surgeon has a downward angle to access the superior endplate of T3. With adequate soft tissue dissection and retraction as described, however, T3 and perhaps even T4 are easily accessible. While this downward angle would likely not permit an anterior plating procedure, it lends itself nicely to Steinmann pin/methyl methacrylate fusion and spares the patient the pain and potential morbidity of sternotomy.

Thoracic discectomy using video assisted thoracoscopy.

STUDY DESIGN: The applicability of using video assisted thoracoscopic surgery (VATS) to resect thoracic discs was investigated. A laboratory study was conducted using two human cadavers and three live pigs as surgical specimens. A total of nine thoracic levels were decompressed. OBJECTIVE: To study the feasibility of performing thoracic resections using VATS. SUMMARY OF BACKGROUND DATA: VATS has been used by thoracic surgeons since 1991 to resect pulmonary lesions. As far as we know, VAT has not yet been used to resect thoracic discs. METHODS: Surgical resections of nine disc levels were carried out in two human cadavers and three anesthetized pigs. VATS was used to provide the surgeon with visualization of the surgical site. Large thoracotomy incisions were not necessary. RESULTS: Five of seven cadaver disc spaces and two of three porcine disc spaces were adequately evacuated of disc material. One episode of dural violation occurred. One animal died during the procedure from an anesthetic complication. CONCLUSION: VATS provides a useful means of performing thoracic discectomies using a small thoracotomy incision. The decrease in invasiveness provided by this new technology may reduce operative morbidity, hospitalization time, and costs. More work is needed, especially in the design of instrumentation, before this becomes a viable alternative to current surgical procedures.

Pathology of cerebral atherosclerosis. Influence of age, race, and gender.

Age, race, and gender are among the logical variables to investigate in determining the natural history of disease. In this connection, the pathological lesions of cerebral atherosclerosis have been particularly difficult to investigate. The limitations of autopsy populations, time-consuming dissections of the intracranial and extracranial arteries, and numerous technical problems associated with specimen preparation, shipment, and storage are among the impediments. Long-term repeated studies and grading systems, and the validation thereof, are also elements adding to the complexity of these studies. In the currently available publications, the most systematic studies, several of which are international in scope and from diverse medical centers, permit some tentative conclusions: (1) There is no reliable evidence of a qualitative difference in the lesions of cerebral atherosclerosis among diverse autopsy populations. (2) Quantitative differences exist in lesion severity among different age groups and races and between males and females. (3) Quantitative differences in intracranial versus extracranial atherosclerosis exist that are related to age and to race (white versus black versus Asian) and gender. (4) The role of hypertension as a factor leading to more severe and more complicated lesions is most obvious in black and Japanese autopsy populations, but its influence is not a simple one to decipher. Complicated lesions refer to stenosis, ectasias and aneurysms, thrombosis, ulceration, and calcification and hemorrhages in plaques. (5) Diet is one obvious variable differing in the populations studied so far. Cigarette smoking is probably an important factor in several populations.(ABSTRACT TRUNCATED AT 250 WORDS)

Amplification of epidermal growth factor receptor gene in gliomas: histopathology and prognosis.

In order to evaluate the incidence and prognostic significance of gene amplification in primary brain neoplasms we measured the number of gene copies per cell of three oncogenes (epidermal growth factor receptor [EGFR] gene, N-myc, C-myc) and syntenic control genes in 40 specimens using quantitative DNA dot blots. We observed EGFR gene amplification in astrocytomas and anaplastic astrocytomas with approximately the same incidence as in glioblastoma multiforme (33%), although large amplifications were only seen in glioblastoma multiforme. Fourteen patients had a supratentorial glioblastoma multiforme; six had EGFR gene amplification and eight had either normal EGFR gene copy number or elevated EGFR copy number attributable to extra copies of chromosome 7. Patients with gene amplification had shorter survival than patients without gene amplification (p = 0.01). The observed difference in survival was not likely to be due to group differences in age, sex, treatment, or histopathology.

Brain enzyme levels in CSF after cardiac arrest and resuscitation in dogs: markers of damage and predictors of outcome.

Levels of brain creatine kinase (CK), aspartate aminotransferase (ASAT), and lactate dehydrogenase (LD) in CSF after cardiac arrest were studied in dog models. Ventricular fibrillation cardiac arrest lasting 10 min or asphyxiation cardiac arrest lasting 0-10 min was followed by cardiopulmonary resuscitation and 96-h intensive care. Outcome was scored as neurologic deficit (0% = normal, 100% = brain death) and overall performance category (1 = normal, 5 = death). Both measures correlated with EEG return time after asphyxiation cardiac arrest, but not after ventricular fibrillation cardiac arrest. Peak activity of enzymes in CSF at 48-72 h post arrest correlated with outcome, and CK was the best predictor. Brain histopathologic damage score at autopsy 96 h post arrest correlated with CK level in CSF (r = 0.79, n = 39) and neurologic deficit (r = 0.70, n = 50). Ischemic neuronal changes occurred after ventricular fibrillation cardiac arrest of 10 min, and neuronal changes plus microinfarcts occurred after asphyxiation cardiac arrest of 1.5-10 min. Brain enzymes were decreased at 6 h post arrest in regions with worst histologic damage (gray matter of neocortex, hippocampus, caudate nucleus, cerebellum). Brain CK decreased further, ASAT remained low, and LD increased at 72 h after arrest. The temporal changes in CK level paralleled the temporal ischemic neuronal changes in the brain, and time to peak activity was unaffected by the severity of the ischemic insult. Peak activity of individual enzymes in CSF was determined predominantly by the brain concentration, but was also influenced by rate of decomposition. This "chemical brain biopsy method" represents a useful adjunctive tool to predict permanent, severe brain damage during comatose states after cardiac arrest and resuscitation.

Conus medullaris nerve root avulsions.

The association of avulsive lesions and pain has been well established in avulsions of the brachial plexus from the cervical spinal cord, but avulsive lesions of the conus medullaris have not previously been recognized or documented by direct observation. Six patients with intractable lower-extremity pain due to avulsion of nerve roots from the conus medullaris were treated by thoracolumbar laminectomy and dorsal root entry zone (DREZ) lesions. Patients with avulsion of lumbosacral roots from the conus medullaris have a characteristic clinical presentation. They are usually young men who, as a result of a motorcycle accident, have suffered multiple pelvic or long-bone fractures or traumatic amputation of part of the lower extremity. Early in their course there is pain not directly attributable to the injured part. The pain is described as intense and burning, with episodic radiation and electric shock-like sensations in the injured or phantom limb. If the leg is intact, there is usually a dermatomal pattern to the distribution of the pain and neurological deficit. A myelogram often reveals a traumatic pseudomeningocele similar to those seen in the cervical region after avulsion of the brachial plexus. Surgical exploration of the conus medullaris usually reveals the extent of nerve root avulsion, and an appropriate DREZ operation can be performed.

Detrimental cerebrometabolic effects of hyperoxia in newborn rats.

To investigate the pathogenesis of oxygen toxicity in the newborn brain, we exposed one-day-old Sprague-Dawley albino rats to 100% O(2) and measured whole-brain high-energy phosphates, glucose, lactate, and free fatty acids (FFA) after 0, 15, 30, 60 and 120 min. Whole-brain adenosine triphosphate and creatine phosphate fell significantly from about 4.5 to 2.5 ?mol-mg(?1) protein. Brain lactate remained at about 0.3 ?mol.mg(?1) protein in hyperoxic rats, but increased in normoxic rats, from 0.3 to 1.3 ?mol.mg(?1) protein at 120 min. Total FFA decreased from 30 to 15 nmol.mg(?1) protein during normoxia, but increased to 40 nmol.mg(?1) protein during hyperoxia. Undetectable in normoxic rats, arachidonic acid increased to between 4 and 6 nmol.mg(?1) protein during hyperoxia while oleic acid increased by two to threefold. In normoxia, palmitate decreased by 70% from 12 to 4 nmol.mg(?1) protein whereas in hyperoxia it remained at 10 nmol.mg(?1) protein. Normobaric 100% O(2) has detrimental metabolic effects on the neonatal brain which cannot be attributed to cerebral vasospasm or seizure-induced cerebral anoxia because lactic acidosis was not observed. FFA changes suggest that a likely explanation is membrane lipid peroxidation from O(2)-induced free radicals.

Factors in predicting outcome from operation in patients with prolactin-secreting pituitary adenomas.

Forty female patients with pituitary adenomas were studied retrospectively to determine whether factors could be identified that would help predict outcome from operation. Twenty-five patients had a normal prolactin level (less than or equal to 30 ng/ml) during the early postoperative period (less than or equal to 3 months) and 15 patients had persistent disease (prolactin greater than 30 ng/ml). Nine of the 25 patients who initially had normal prolactin levels during the early postoperative period were found to have elevated prolactin levels during the late postoperative period (greater than 3 months). As has been shown previously, tumor size and preoperative prolactin levels were important factors in predicting surgical outcome. Patients with smaller (Hardy Grades I and II) tumors had significantly better outcome than those with larger (Hardy Grades III and IV) tumors. Patients with successful surgical outcomes had significantly lower preoperative prolactin values (204 ng/ml) than those with operative failures (524 ng/ml). In addition to the known factors, the patient's age at the time of operation, the length of amenorrhea, and the patient's growth hormone response to insulin hypoglycemia were newly identified as factors that helped predict surgical outcome. Patients who were less than or equal to 26 years of age and who had had amenorrhea for less than or equal to 6 years at the time of operation had significantly better surgical outcomes. Patients with normal growth hormone responses to stimulation testing had significantly better surgical outcomes than those with a blunted preoperative growth hormone response. The data suggest that prolactin-secreting pituitary tumors may cause a progressive disorder for which operative cure may be obtained only early in the disease.

Malfunctioning ventriculoperitoneal shunts. Clinical and pathological features.

The authors report the clinical and pathological findings in 201 procedures to revise malfunctioning ventriculoperitoneal shunts in 114 patients. The revised shunts were predominantly Raimondi three-piece systems. Collection of fluid along the shunt track is emphasized as an early clinical feature of shunt malfunction. In some cases, there are few clinical signs of shunt infection, and the first indication of it may be shunt malfunction. Ventricular catheter obstruction was caused by tissue from the region in addition to choroid plexus. Inflammation was frequently found around both ventricular and peritoneal catheters. Foreign bodies (cotton fiber, hair, and talc) and granulomatous inflammation were often present at both ends. Tissues obstructing the peritoneal catheter included embolic neoplastic cells, choroid plexus, and leptomeninges. There was also more necrotic debris at the peritoneal end. To prevent shunt malfunction, attention should be directed to: 1) optimal placement of ventricular catheters inside the lateral ventricle; 2) prevention of infection; 3) avoidance of contamination by cotton fibers, hair, or talc; and 4) improvements in the biocompatibility of the implanted materials.

Further studies on the effects of lesions in the rostral hypothalamus on gonadotropin secretion in the female rhesus monkey (Macaca mulatta).

Bilateral radiofrequency lesions were stereotaxically placed in the rostral hypothalamus of four adult female rhesus monkeys. These lesions resulted in extensive destruction of the ventromedial preoptic-anterior hypothalamic area (POA-AHA) and included the suprachiasmatic nucleus as well as, with the exception of one animal, the organum vasculosum of the lamina terminalis. In three of these four animals, gonadotropin surges similar to those observed before surgery were elicited in response to either a spontaneous increment in serum estrogen concentration or an estradiol benzoate injection. This stimulatory action of estradiol on LH and FSH release was not demonstratable in the remaining lesioned animal, but estradiol benzoate injections also failed to elicit a gonadotropin discharge in one of a series of five normal control animals. These findings fail to support the view that destruction of the ventromedial POA-AHA in this species compromises the ability of the hypothalamicohypophysial apparatus to respond to the positive feedback action of estradiol. The diurnal variation in serum cortisol concentration was not interrupted by placement of the lesions in the ventromedial POA-AHA.

Anencephaly and heterotopic central nervous tissue in lungs.

Heterotopic nodules of moderately well-differentiated central nervous tissue were seen in the lung of a full-term anencephalic baby who survived a few hours after birth. The pathogenesis of this rare condition remains obscure. None of the three hypotheses previously offered to explain this condition can be accepted or rejected on the basis of the recorded evidence.

The arcuate nucleus and the control of gonadotropin and prolactin secretion in the female rhesus monkey (Macaca mulatta).

Attempts were made to destroy selectively the arcuate nucleus with radiofrequency current in adult female rhesus monkeys as a first step in identifying the areas of the mediobasal hypothalamus (MBH) that are responsible for the neural control of gonadotropin secretion in this species. Extensive or complete destruction of the arcuate region was produced in three animals and in two of these the lesion was confined primarily to the arcuate region and the dorsal aspect of the posterior median eminence. These lesions resulted in the cessation of LH and FSH secretion and blocked the positive feedback action of estradiol on gonadotropin release but did not appear to influence grossly basal thyroid and adrenocortical function, or to abolish GH discharge in response to insulin hypoglycemia. Adenohypophysial infarcts were not observed and exogenous LHRH and TRH induced marked discharges of the appropriate anterior pituitary hormones. In two additional animals with large hypothalamic lesions, destruction of the arcuate region was incomplete. In this group only partial inhibition of gonadotropin secretion was observed. LH and FSH secretion did not appear to be influenced in one animal bearing a large MBH lesion that entirely spared the arcuate region. Although serum prolactin remained at pre-lesion control levels after placement of the two relatively discrete lesions confined to the arcuate region, unambiguous increases in the secretion of this hormone were observed when the area of destruction encompassed tissue anterior and/or dorsal to the arcuate region. These observations suggest that the arcuate region is the primary structure mediating the hypothalamic control of gonadotropin secretion in the rhesus monkey. They also suggest that, in this species, the regions of the MBH involved with the regulation of gonadotropin release and those which control prolactin secretion are anatomically distinct.

Estrogen-induced gonadotropin surges in decerebrated female rhesus monkeys with medial basal hypothalamic peninsulae.

Hypothalamic peninsulae, posteriorly continuous with the brain stem, were produced in ovariectomized rhesus monkeys by aspiration of the cerebral hemispheres and all brain structures anterior and dorsal to the optic chiasm. These preparations were sustained for 52 to 60 h postoperatively by maintaining blood gases and pH within physiological limits and by supporting arterial blood pressure, when necessary, with infusions of rhesus monkey blood and norepinephrine. The sc injection of estradiol benzoate (EB) to such animals upon completion of the ablation procedure was followed 18-24 h later by unambiguous surges of serum LH and FSH with durations in the excess of 24 h. These findings are consonant with the view that the central components of the control system that initiates the preovulatory gonadotropin surge in the rhesus monkey are resident within the medial basal hypothalamic-hypophysial apparatus.