Exophytic ependymoma of the thoracic spine.

Exophytic ependymomas of the spinal cord are very rare outside the filum or conus region. We present a patient with a thoracic spinal cord intradural extramedullary and intramedullary, World Health Organization grade II ependymoma. Gross total resection of the extramedullary component with subtotal resection of the intramedullary tumor was achieved, since there was no clear distinction between cord and tumor. The patient received postoperative external beam radiotherapy for residual tumor, and at a 2-year follow-up he is ambulatory without evidence of tumor recurrence.

Piriformis and related entrapment syndromes: myth & fallacy.

This article explores the myths surrounding piriformis syndrome. It looks at the syndrome's history ranging from early hypothesis to acceptance; then from disfavor to resurrection as a diagnosis for patients with leg pain devoid of objective neurologic deficits, without disc herniations. It includes a critical review of the clinical literature and treatment strategies. It calls for a renaming of the syndrome to "nonlocalizing sciatica" and restriction of surgical resection of the piriformis to cases where all other treatment has failed. This is part of a Point-Counterpoint discussion with Dr. Aaron Filler's presentation of "Diagnosis and Management".

Tumors of the brachial plexus.

Tumors of the brachial plexus are relatively rare and present a clinical challenge for the neurosurgeon. The management of these tumors therefore requires not only an understanding of the complex anatomy of the brachial plexus but also an appreciation of the appropriate surgical approach to the various tumors that may be encountered. Over a 30-year period (1969-1999), 226 patients with brachial plexus tumors were evaluated and surgically treated by the senior authors (R.L.T., D.G.K.). In the present paper they review the most common benign and malignant brachial plexus tumors and discuss management and surgical principles established through their experience at the Louisiana State University Health Sciences Center.

Gunshot wounds involving the brachial plexus: surgical techniques and outcomes.

This paper presents the management and outcomes for two consecutive operative series of gunshot wounds (GSWs) involving the brachial plexus. The cases were from Louisiana State University Health Sciences Center (LSUHSC) and were obtained by retrospective chart reviews. Series 1 includes patients with injuries managed between 1968 and 1980 and series 2, from 1981 to 1998. Pre- and postoperative motor function was assessed using the LSUHSC grading system. The outcomes for each surgical technique for each series are presented. Fewer cases of brachial plexus elements injured by GSWs in series 2 may be due to decreased firearm-related injuries between 1993-1997, i.e., 39,595 versus 32,436 deaths, respectively. The graft repair increase in series 2 may have been the result of more severe injuries, since documented trends between 1971 and 1997 show the use of larger caliber bullets. This paper shows that with the outlined management and surgical techniques, surgery on certain GSWs of the brachial plexus is worthwhile.

Lacerations to the brachial plexus: surgical techniques and outcomes.

The charts of patients with 201 brachial plexus elements sustaining operative lacerations and managed at Louisiana State University Health Sciences Center (LSUHSC) were reviewed retrospectively. Results for elements injured by sharp transections and undergoing suture repairs performed within 72 hr, as well as secondary suture and secondary graft repairs are documented. Similarly, results for secondary end-to-end suture anastomosis and secondary graft repairs for elements sustaining blunt transections are reviewed. Results for neurolysis, end-to-end suture anastomosis, and graft repairs for plexus elements in continuity despite the laceration injury are reviewed. Outcomes for the LSUHSC series of brachial plexus lacerations are one of the best of all LSUHSC plexus injuries, even for elements generally viewed as unfavorable for repair. Lesions in continuity with positive nerve action potentials (NAPs) had the best outcomes.

Management and outcomes of 42 surgical suprascapular nerve injuries and entrapments.

OBJECTIVE: Retrospective chart reviews of 42 patients with surgical suprascapular nerve (SSN) injury/entrapment were performed. Presenting symptoms, findings, operative approach, and results are documented. METHODS: Forty-two patients with SSN injuries/entrapments underwent operations between 1970 and 2002. Charts were retrospectively reviewed for the presence of shoulder pain; spinati muscle function was evaluated with the Louisiana State University Health Sciences Center grading system. Side of lesion and sex were equally represented; mean follow-up was 18 months (range, 12-48 mo). SSN injuries/entrapments were associated with occupational overuse, sports-related injury, direct trauma and ganglion cysts. Thirty-one (79%) of 39 patients with suprascapular notch SSN injuries/entrapments, excluding ganglion cysts, presented with mild to moderate shoulder pain and spinati weakness. RESULTS: Motor function for these 31 patients was graded on a scale of 0 to 5. Preoperatively, patients had supraspinatus function Grades 0 to 2 and infraspinatus function Grades 0 to 2. Supraspinatus function improved postoperatively to Grade 4 or better in 28 patients (90%) and to Grades 2 to 3 in 3 patients (10%). Infraspinatus function improved to better than Grade 3 in 10 patients (32%), to Grades 2 to 3 in 14 patients (45%), and to Grade 1 in 7 patients (23%). Preoperatively, eight (21%) of 39 patients presenting with persistent severe pain had Grade 3 spinati strength. Of these eight patients, seven (88%) had an improvement in pain postoperatively. Strength in this group remained the same or improved to Grade 4. Postoperatively, three patients with ganglion cysts had good improvement in spinati function. CONCLUSION: Although SSN injury/entrapment is rare, 42 patients are presented who responded well to SSN release. Supraspinatus muscle improvement was as good as or better than that achieved in the infraspinatus.

Surgical management of 33 ilioinguinal and iliohypogastric neuralgias at Louisiana State University Health Sciences Center.

OBJECTIVE: This is a retrospective review of 33 charts of patients with ilioinguinal and iliohypogastric neuralgias who underwent a neurectomy at Louisiana State University Health Sciences Center between 1967 and 2000. Operations associated with neuralgias and postoperative pain outcomes were analyzed. METHODS: There were 23 ilioinguinal and 10 combined ilioinguinal-iliohypogastric neuralgias, and the side of the lesion and sex of the patient were found to be evenly distributed in this group. Nerve blocks must have resulted in a complete or substantial decrease in pain before a neurectomy was recommended. Twenty-nine (88%) of 33 neuralgia patients had injuries from iatrogenic causes, and 4 (12%) injuries were caused by blunt trauma. In the 23 isolated ilioinguinal neuralgias, the operation associated with neuralgias in 13 (57%) of 23 patients was a herniorrhaphy. This was followed by 4 (17%) neuralgias after an appendectomy and 3 (13%) after a hysterectomy. Three (13%) patients had neuralgias resulting from blunt trauma. Nine (90%) of 10 ilioinguinal-iliohypogastric lesions were caused by iatrogenic causes, and 1 (10%) neuralgia resulted from blunt trauma. A neurectomy was performed in all patients. RESULTS: The neurectomy resulted in considerable pain relief in 21 (91%) of 23 patients with ilioinguinal lesions and 9 (90%) of 10 patients with ilioinguinal-iliohypogastric lesions. Postoperative side effects were persistent numbness below the resected nerve and loss of the cremasteric reflex. Minor postoperative complications consisted of two superficial skin infections. CONCLUSION: Ilioinguinal and ilioinguinal-iliohypogastric neuralgias are infrequent conditions; however, 33 patients from the Louisiana State University Health Sciences Center were accrued and analyzed in this study, and most had significant pain relief after neurectomy.

A series of 397 peripheral neural sheath tumors: 30-year experience at Louisiana State University Health Sciences Center.

OBJECT: This is a retrospective review of 397 benign and malignant peripheral neural sheath tumors (PNSTs) that were surgically treated between 1969 and 1999 at the Louisiana State University Health Sciences Center (LSUHSC). The surgical techniques and adjunctive treatments are presented, the tumors are classified with respect to type and prevalence at each neuroanatomical location, and the management of malignant PNSTs is reviewed. METHODS: There were 361 benign PNSTs (91%). One hundred forty-one benign lesions were brachial plexus tumors: 54 schwannomas (38%) and 87 neurofibromas (62%), of which 55 (63%) were solitary neurofibromas and 32 (37%) were neurofibromatosis Type 1 (NF1)-associated neurofibromas. Among the brachial plexus lesions supraclavicular tumors predominated with 37 (69%) of 54 schwannomas; 34 (62%) of 55 solitary neurofibromas; and 19 (59%) of 32 NF1-associated neurofibromas. One hundred ten upper-extremity benign PNSTs consisted of 32 schwannomas (29%) and 78 neurofibromas (71%), of which 45 (58%) were sporadic neurofibromas and 33 (42%) were NF1-associated neurofibromas. Twenty-five benign PNSTs were removed from the pelvic plexus. Lower-extremity PNSTs included 32 schwannomas (38%) and 53 neurofibromas (62%), of which 31 were solitary neurofibromas and 22 were NF1-associated neurofibromas. There were 36 malignant PNSTs: 28 neurogenic sarcomas and eight other sarcomas (fibro-, spindle cell, synovial, and perineurial sarcomas). CONCLUSIONS: The majority of tumors were benign PNSTs from the brachial plexus region. Most of the benign PNSTs in all locations were neurofibromas, with sporadic neurofibromas predominating. Similar numbers of schwannomas were found in the upper and lower extremities, whereas neurofibromas were more prevalent in the upper extremities. Despite aggressive limb-ablation or limb-sparing surgery plus adjunctive therapy, malignant PNSTs continue to be associated with high morbidity and mortality rates.

A series of 146 peripheral non-neural sheath nerve tumors: 30-year experience at Louisiana State University Health Sciences Center.

OBJECT: This is a retrospective review of 146 surgically treated benign and malignant peripheral non-neural sheath tumors (PNNSTs). Tumor classifications with patient numbers, locations of benign PNNSTs, and surgical techniques and adjunctive treatments are presented. The results of a literature review regarding tumor frequencies are presented. METHODS: One hundred forty-six patients with 111 benign and 35 malignant PNNSTs were treated between 1969 and 1999 at the Louisiana State University Health Sciences Center (LSUHSC). The benign tumors included 33 ganglion cysts, 16 cases of localized hypertrophic neuropathy, 12 lipomas, 12 tumors of vascular origin, and 11 desmoid tumors. There were four each of lipofibrohamartomas, myositis ossificans, osteochondromas, and ganglioneuromas; two each of meningiomas, cystic hygromas, myoblastoma or granular cell tumors, triton tumors, and lymphangiomas; and one epidermoid cyst. The locations of benign PNNSTs were the following: 33 in the brachial plexus region, 39 in an upper extremity, one in the pelvic plexus, and 38 in a lower extremity. The malignant PNNSTs included 35 surgically treated carcinomas, 15 of which originated in the breast and nine in the lung. There were two melanomas metastatic to nerve and one tumor each that had metastasized from the bladder, rectum, skin, head and neck, and thyroid, and from a primary Ewing sarcoma. There was a single lymphoma that had metastasized to the radial nerve and one chordoma and one osteosarcoma, each of which had metastasized to the brachial plexus. CONCLUSIONS: There were more benign PNNSTs than malignant ones. Benign tumors were relatively equally distributed in the brachial plexus region and upper and lower extremities, with the exception of the pelvic plexus, which had only one tumor.

Surgical management of 10 genitofemoral neuralgias at the Louisiana State University Health Sciences Center.

OBJECTIVE: This is a retrospective review of the charts of 10 patients with genitofemoral neuralgia who underwent neurectomy at the Louisiana State University Health Sciences Center between 1967 and 2000. Operations associated with these neuralgias and postoperative pain outcomes were analyzed. METHODS: The charts of 10 patients with genitofemoral neuralgias were analyzed retrospectively. RESULTS: The distribution of the 10 genitofemoral neuralgias with regard to right or left side and sex was found to be equal. L1 and L2 nerve blocks had resulted in a complete or substantial decrease in pain before neurectomy was recommended. Of six iatrogenic injuries (60%), gynecological surgery, including two hysterectomy procedures, resulted in a total of three genitofemoral neuralgias (50%), and vasectomy procedures antedated two nerve injuries (33%). Four (40%) of the 10 patients had injury to the genitofemoral nerve after blunt abdominal trauma. Genitofemoral neurectomy was performed in all genitofemoral neuralgia patients after conservative therapy had failed. This procedure resulted in considerable pain relief in all 10 patients, whether their injury was the result of iatrogenic causes or trauma. CONCLUSION: Genitofemoral neuralgias are infrequent conditions; however, 10 patients were accrued and analyzed in this study, and most had considerable or complete pain relief after neurectomy.

Direct plexus repair by grafts supplemented by nerve transfers.

This article reviews the Louisiana State University Health Sciences Center experience with direct repair of brachial plexus lacerations, gunshot wounds, and stretch/contusive/avulsive injuries. In the stretch category, limited outcomes with direct repair have led to addition of nerve transfers rather than their exclusive use. It is important to per-form direct plexus repair in conjunction with nerve transfers in the same patient when-ever possible. The intent of such a "pants-over-vest" approach is to maximize axonal input to denervated structures.

Elevated ankyrin G in a plexiform neurofibroma and neuromas associated with pain.

UNLABELLED: Ankyrin G has recently been shown to be responsible for activation of sodium channels in the developing and regenerating axonal membrane. Via this sodium channel mechanism, elevated ankyrin G levels have been linked with spontaneous ectopic hyperexcitability and thus with pain phenomena in nervous tissue. Ankyrin G, a transmembrane, structural protein of the axon, was examined in four conditions: (a) painful plexiform neurofibroma; (b) painful neuroma; (c) non-painful neuromas; (d) normal nerve. Neurofibroma tissue was obtained from an 18-year old male patient who developed an intensely painful, plexiform neurofibroma of the posterior femoral cutaneous nerve and subsequently underwent surgery. Sample proteins were separated by PAGE and labeled with anti-ankyrin G antibodies in a Western blot procedure. RESULTS: The ankyrin G band density (mug) of protein for the painful neurofibroma was 6014 and was 3557 for the painful neuroma as compared to 3041, 1988 and 606 (mean+/-SD=1878+/-1221) for the three non-painful neuromas. Ankyrin G expression in normal nerves (8 specimens from 7 patients) was comparatively less (mean+/-SD=411+/-339). CONCLUSION: Our results may represent the first evidence for abnormally increased levels of ankyrin G protein with painful neurofibromas. Due to ankyrin G's multifunctional role in the development and remodeling of excitable membranes, it can be hypothesized that the significant increase contributes to the development of hyperexcitable axonal membranes in neurofibromas and potentially other peripheral pain conditions.

Mechanisms of injury in operative brachial plexus lesions.

OBJECT: The authors focus on injury mechanisms involved in 1019 operative brachial plexus injuries (BPIs) managed between 1968 and 1998 at Louisiana State University Health Sciences Center (LSUHSC). METHODS: Data regarding these mechanisms of injury were obtained via retrospective chart reviews of patients who had undergone operations at LSUHSC. Five main mechanisms of injury to the brachial plexus occurred in the series. These included 509 stretch/contusion injuries (49%) with four patterns of presentation in 366 patients: 208 C5-T1 nerve injuries; 75 C5-7, 55 C5-6 injuries; and 28 involving the C8-T1 or C7-T1 nerves. Stretch/contusion injury was followed in frequency by gunshot wound (GSW), resulting in 118 injuries (12%). Most of the 293 involved plexus elements had some gross continuity when surgically exposed. Seventy-one lacerations involved the brachial plexus (7%), including 83 sharp lacerations caused by knives or glass; 61 blunt transections due to automobile metal, fan, and motor blades, chain saws, or animal bites. Nontraumatic BPIs included 160 cases of thoracic outlet syndrome or 16% of the total of 1019 BPIs. There were 161 tumors (16%) of neural sheath origin including 55 solitary neurofibromas (34%), 32 neurofibromas associated with von Recklinghausen disease (20%), 54 schwannomas (34%), and 20 malignant nerve sheath tumors (20%) removed. Obstetrical BPI was not included in the original series; however, the current literature is reviewed in this paper. CONCLUSIONS: The conclusion of this study is that the brachial plexus can be injured by multiple mechanisms of which stretch/contusion injury is the most frequently encountered, followed by GSWs.

Penetrating injuries due to gunshot wounds involving the brachial plexus.

The authors review 118 operative brachial plexus gunshot wounds (GSWs), causing 293 element injuries that were managed over a 30-year period at Louisiana State University Health Sciences Center (LSUHSC). Retrospective chart reviews were performed. Using the LSUHSC grading system for motor sensory function, each element's grades were combined and averaged. Most of the 293 injured elements were found to have gross continuity at operation and of 202 elements with complete neurological loss, only 16 (8%) exhibited total disruption. Of 293 injuries, 128 elements with complete or incomplete loss were not only in continuity when explored but also had positive intraoperative nerve action potentials (NAPs). After neurolysis, 120 of 128 elements in continuity (94%) improved to greater than or equal to Grade 3 function. Elements not regenerating early usually required repair. One hundred fifty-six of 202 completely or incompletely injured elements (77%) required resection and suture or graft repair based on intraoperative NAPs. Neurolysis achieved greater than or equal to Grade 3 results in 42 (91%) of 46 elements with complete loss. Suture repair resulted in good outcomes in 14 (67%) of 21 and in 73 (54%) of 135 undergoing graft repairs (1 to 3.5 cm length) and presenting with complete loss. Of 91 incomplete elements, intraoperative NAPs were positive in 82 (90%) and 78 of 82 had good results. Nine had negative NAPs and six elements required suture repair. Three required grafts with results of greater than or equal to Grade 3 in five (83%) of six and two (67%) of three, respectively. Based on 118 patient results with 293 injured elements, guidelines for the management of GSWs were established as described in this paper.

Infraclavicular brachial plexus stretch injury.

OBJECT: The authors report the surgery-related results obtained in 143 patients with stretch-induced infraclavicular brachial plexus injuries (BPIs). The entire series comprised 1019 operative BPIs managed at the Louisiana State University Health Sciences Center between 1968 and 1998. METHODS: Infraclavicular lesions represented 143 (28%) of the total of 509 stretch injuries involving both the infra- and supraclavicular brachial plexus, of which 366 (72%) were supraclavicular lesions. The operative approach is thoroughly outlined, and common patterns and combinations of involvement of nerves peculiar to the infraclavicular area are presented. Overall, the results of suture and graft repair were favorable for the lateral and posterior cord and their outflows. Repair of medial cord-median nerve also yielded acceptable results. The results of medial cord and medial cord-ulnar nerve, however, were poor. The incidence of associated injuries in the infraclavicular as opposed to the supraclavicular area, including shoulder dislocation and fracture and humeral fractures as well as vascular injuries including axillary artery injury was higher. Results of a literature search supported the finding that vascular injuries were increased due to the juxtaposition of vessels among the brachial plexus elements. CONCLUSIONS: Thus, although less common than their supraclavicular counterpart, infraclavicular stretch injury lesions when they occur are technically more difficult to treat and are associated with a higher incidence of vascular and dislocation/fraction injuries. Favorable results were obtained for lateral and posterior cord lesions and their outflows, with acceptable outcome after medial cord-median nerve stretch injury repair. The results of medial cord and medial cord to ulnar nerve, however, were poor.

Intrapelvic and thigh-level femoral nerve lesions: management and outcomes in 119 surgically treated cases.

OBJECT: The authors present a retrospective analysis of 119 surgically treated femoral nerve lesions at intrapelvic and thigh levels seen at the Louisiana State University Health Sciences Center. METHODS: Femoral nerve lesions treated between 1967 and 2000, (89 traumatic injuries and 30 tumors and cystic lesions) were evaluated for injury mechanisms, resulting lesions, surgical management, and postoperative functional outcomes by using retrospective chart reviews. The most common injury mechanism was iatrogenic (52 cases), which occurred after hernia and hip operations (10 each), followed by arterial bypass and gynecological procedures (eight each), angiography (seven), abdominal surgery (five), appendectomy (two), a laparoscopy, and a lumbar sympathectomy. Other injury mechanisms included hip or pelvic fractures (19), gunshot wounds (10), and lacerations (eight). The 30 femoral nerve tumors and cystic lesions consisted of neurofibromas (16), schwannomas (nine), ganglionic cysts (two), neurogenic sarcomas (two), and a leiomyosarcoma. Forty-four patients underwent neurolysis. Some had recordable nerve action potentials (NAPs) across their lesions in continuity, despite severe distal loss. Others with recordable NAPs had mild loss, but also experienced a pain problem, which was helped in some by neurolysis. In 36 patients, in whom repairs were performed using long sural grafts for mostly proximal pelvic-level injuries, recovery of useful function occurred. Eight of nine thigh-level suture repairs led to improvement to good functional levels. Most of the tumors and cystic lesions were resected, with preservation of preoperative function. CONCLUSIONS: The majority of femoral nerve injuries resulted in lesions in continuity, and intraoperative NAP recordings were essential in evaluating axonal regeneration across these lesions. Despite severe and frequently proximal injury levels requiring repairs with long grafts, femoral nerve lesion repairs resulted in good functional recovery.

Operative outcomes of 546 Louisiana State University Health Sciences Center peripheral nerve tumors.

The surgical management of benign PNSTs and some other benign tumors can result in successful outcomes. Schwannomas and nonplexiform neurofibromas can be resected with minimal deficit by sparing all but the fascicles entering and exiting the tumor. These fascicles, if not functional by NAP testing, can be resected, and the tumor can be removed. Surgery to remove other benign lesions, such as intraneural ganglion cysts, hemangiomas,and ganglioneuromas, has become more timely. The desmoid tumor, although microscopically benign, is locally and regionally invasive, and chemotherapy and radiation therapy may need to be used as adjunctive therapy. Neurogenic sarcomas and other malignancies have high morbidity and mortality despite aggressive limb ablation or limb-sparing surgery with adjunctive therapy. Thus, surgery involving decompression as well as the most complete resection possible remains the essential initial step in the management of most malignancies.

Management and outcomes in 318 operative common peroneal nerve lesions at the Louisiana State University Health Sciences Center.

OBJECTIVE: This study analyzes 318 operative knee-level common peroneal nerve lesions managed at the Louisiana State University Health Sciences Center between 1967 and 1999. METHODS: Each patient was retrospectively evaluated for injury mechanism, preoperative neurological status, electrophysiological studies, lesion type, and operative technique, i.e., neurolysis, suture, or graft repair. All lesions in continuity had intraoperative nerve action potential recordings. RESULTS: There were 141 stretch/contusions without fracture/dislocations (44%), 39 lacerations (12%), 40 tumors (13%), 30 entrapments (9%), 22 stretch/contusions with fracture/dislocations (7%), 21 compressions (7%), 13 iatrogenic injuries (4%), and 12 gunshot wounds (4%). After neurolysis, 107 (88%) of 121 knee-level common peroneal nerve lesions with recordable intraoperative nerve action potentials recovered useful function. Nineteen patients underwent end-to-end suture repair, and 16 (84%) of these achieved good recovery by 24 months. Graft repair was performed in 138 peroneal injuries. Thirty-six patients (26%) had grafts less than 6 cm long, of which 27 (75%) achieved Grade 3 or greater peroneal function. Twenty-four (38%) of 64 patients with 6- to 12-cm grafts, and only 6 (16%) of 38 patients with 13- to 24-cm grafts, attained good peroneal function. Longer grafts correlated with more severe injuries and thus poorer outcomes. Thirty-two (80%) of 40 tumors were resected with preservation of preoperative clinical function. CONCLUSION: Surgical exploration and repair of peroneal nerve lesions achieved good results with timely operations and thorough intraoperative evaluations. Useful function was achieved in 27 (75%) of 36 patients with grafts less than 6 cm in length and in only 88 (44%) of 202 patients with grafts greater than 6 cm in length.