Citations

Object The authors reviewed outcomes after stereotactic radiosurgery for patients with acromegaly and analyzed factors associated with biochemical remission. Methods Retrospective analysis was performed for 46 consecutive cases of growth hormone (GH)–producing pituitary adenomas treated by radiosurgery between 1991 and 2004. Biochemical remission was defined as a fasting GH less than 2 ng/ml and a normal age- and sex-adjusted insulin-like growth factor–I (IGF-I) level while patients were not receiving any pituitary suppressive medications. The median follow up after radiosurgery was 63 months (range 22–168 months). Twenty-three patients (50%) had biochemical remission documented at a median of 36 months (range 6–63 months) after one radiosurgical procedure. The actuarial rates of biochemical remission at 2 and 5 years after radiosurgery were 11 and 60%, respectively. Multivariate analysis showed that IGF-I levels less than 2.25 times the upper limit of normal (hazard ratio [HR] 2.9, 95% confidence interval [CI] 1.2–6.9, p = 0.02) and the absence of pituitary suppressive medications at the time of radiosurgery (HR 4.2, 95% CI 1.4–13.2, p = 0.01) correlated with biochemical remission. The incidence of new anterior pituitary deficits was 10% at 2 years and 33% at 5 years. Conclusions Discontinuation of pituitary suppressive medications at least 1 month before radiosurgery significantly improved endocrine outcomes for patients with acromegaly. Patients with GH–producing pituitary adenomas should not undergo further radiation therapy or surgery for at least 5 years after radiosurgery because GH and IGF-I levels continue to normalize over that interval.

Object The aim of this study was to determine whether ictal single-photon emission computed tomography (SPECT) is useful in localizing the site of seizure onset in patients in whom surgery for intractable epilepsy failed and who are being considered for repeated surgery. Methods Subtraction ictal SPECT coregistered to magnetic resonance imaging (SISCOM) studies were retrospectively analyzed in 58 patients who were being evaluated for possible repeated resection for intractable partial epilepsy between January 1, 1996, and October 31, 1999. All patients had persistent seizures subsequent to an initial resection and underwent another excision. The SISCOM-demonstrated abnormalities were classified as concordant, discordant, or indeterminate, compared with the localization of the epileptogenic zone revealed on video electroencephalography monitoring. The ability of SISCOM to predict operative outcome was also determined in patients who had undergone repeated surgical procedures. The SISCOM studies revealed a localized hyperperfused alteration in 46 (79%) of 58 patients. Forty-one (89%) of these 46 patients had a SISCOM-demonstrated alteration in the hemisphere of the previous epilepsy surgery. Imaging changes in 33 (72%) of the 46 patients were at the site of the previous focal cortical resection. Eight (17%) of the 46 had SISCOM-demonstrated abnormalities remote from the lobe in which surgery had been performed but in the ipsilateral hemisphere. The hyperperfusion focus was in the contralateral hemisphere in the remaining five patients (11%). The site of the epileptogenic zone was concordant with the SISCOM focus in 32 (70%) of 46 patients. Twenty-six patients underwent repeated resection and were followed up for a mean of 44 months thereafter; 11 of these patients (42%) had a significant reduction in seizure tendency. Only five patients (19%) were seizure free. Ten (50%) of 20 patients with a concordant SISCOM focus compared with none (0%) of three patients with a discordant focus had a favorable surgical outcome (p = 0.23). Conclusions The SISCOM method might be useful in the evaluation of, and the surgical planning for, patients with intractable partial epilepsy in whom previous resective treatment has failed and who are being considered for reoperation.

Object The authors reviewed the long-term outcome of focal resection in a large group of patients who had intractable partial nonlesional epilepsy, including mesial temporal lobe sclerosis (MTS), and who were treated consecutively at a single institution. The goal of this study was to evaluate the long-term efficacy of epilepsy surgery and the preoperative factors associated with seizure outcome. Methods This retrospective analysis included 399 consecutive patients who underwent epilepsy surgery at Mayo Clinic in Rochester, Minnesota, between 1988 and 1996. The mean age of the patients at surgery was 32 ± 12 years (range 3–69 years), and the mean age at seizure onset was 12 ± 11 years (range 0–55 years). There were 214 female (54%) and 185 male (46%) patients. The mean duration of epilepsy was 20 ±12 years (range 1–56 years). The preceding values are given as the mean ± standard deviation. Of the 399 patients, 237 (59%) had a history of complex partial seizures, 119 (30%) had generalized seizures, 26 (6%) had simple partial seizures, and 17 (4%) had experienced a combination of these. Preoperative evaluation included a routine and video-electroencephalography recordings, magnetic resonance imaging of the head according to the seizure protocol, neuropsychological testing, and a sodium amobarbital study. Patients with an undefined epileptogenic focus and discordant preoperative studies underwent an intracranial study. The mean duration of follow up was 6.2 ± 4.5 years (range 0.6–15.7 years). Seizure outcome was categorized based on the modified Engel classification. Time-to-event analysis was performed using Kaplan–Meier curves and Cox regression models to evaluate the risk factors associated with outcomes. Among these patients, 372 (93%) underwent temporal and 27 (7%) had extratemporal resection of their epileptogenic focus. Histopathological examination of the resected specimens revealed MTS in 113 patients (28%), gliosis in 237 (59%), and normal findings in 49 (12%). Based on the Kaplan–Meier analysis, the probability of an Engel Class I outcome (seizure free, auras, or seizures related only to medication withdrawal) for the overall patient group was 81% (95% confidence interval [CI] 77–85%) at 6 months, 78% (CI 74–82%) at 1 year, 76% (CI 72–80%) at 2 years, 74% (CI 69–78%) at 5 years, and 72% (CI 67–77%) at 10 years postoperatively. The rate of Class I outcomes remained 72% for 73 patients with more than 10 years of follow up. If a patient was in Class I at 1 year postoperatively, the probability of seizure remission at 10 years postoperatively was 92% (95% CI 89–96%); almost all seizures occurred during the 1st year after surgery. Factors predictive of poor outcome from surgery were normal pathological findings in resected tissue (p = 0.038), male sex (p = 0.035), previous surgery (p < 0.001), and an extratemporal origin of seizures (p < 0.001). Conclusions The response to epilepsy surgery during the 1st follow-up year is a reliable indicator of the long-term Engel Class I postoperative outcome. This finding may have important implications for patient counseling and postoperative discontinuation of anticonvulsant medications.

Object The purpose of this study was to examine the prevalence of intracranial cavernous malformations (CMs) in a large series of predominantly Caucasian patients with spinal cord CMs. The authors also studied the natural history of spinal CMs in patients who were treated nonoperatively. Methods The medical records of 67 consecutive patients (32 female and 35 male patients) in whom a spinal CM was diagnosed between 1994 and 2002 were reviewed. The patients’ mean age at presentation was 50 years (range 13–82 years). Twenty-five patients underwent resection of the lesion. Forty-two patients in whom the spinal CM was diagnosed using magnetic resonance (MR) imaging were followed expectantly. Thirty-three (49%) of 67 patients underwent both spinal and intracranial MR imaging. All available imaging studies were reviewed to determine the coexistence of an intracranial CM. Fourteen (42%) of the 33 patients with spinal CMs who underwent intracranial MR imaging harbored at least one cerebral CM in addition to the spinal lesion. Six (43%) of these 14 patients did not have a known family history of CM. Data obtained during the long-term follow-up period (mean 9.7 years, total of 319 patient-years) were available for 33 of the 42 patients with a spinal CM who did not undergo surgery. Five symptomatic lesional hemorrhages (neurological events), four of which were documented on neuroimaging studies, occurred during the follow-up period, for an overall event rate of 1.6% per patient per year. No patient experienced clinically significant neurological deficits due to recurrent hemorrhage. Conclusions As many as 40% of patients with a spinal CM may harbor a similar intracranial lesion, and approximately 40% of patients with coexisting spinal and intracranial CMs may have the nonfamilial (sporadic) form of the disease. Patients with symptomatic spinal CMs who are treated nonoperatively may have a small risk of clinically significant recurrent hemorrhage. The findings will aid in evaluation of surveillance images and in counseling of patients with spinal CMs, irrespective of family history.

Object The suprascapular nerve may become entrapped as it travels deep to the suprascapular ligament, necessitating decompression. The present study was performed to verify the feasibility of a minimally invasive, endoscopically assisted technique for decompressing the suprascapular nerve in the supraspinous fossa. Methods The authors performed dissection and decompression of the suprascapular ligament using an endoscopically assisted technique via a 3-cm skin incision in 10 adult cadavers (20 sides). Measurements were also made of the depth from the skin to the suprascapular ligament. Results A mean depth of 4 cm was necessary to reach the suprascapular ligament from the skin surface. With the authors' approach, no obvious injury occurred to the suprascapular or other vicinal neurovascular structures (such as the spinal accessory nerve and suprascapular vessels). Conclusions The results of this cadaveric study demonstrate that access to the suprascapular nerve can be obtained endoscopically via a small suprascapular incision. This approach obviates a large incision, entry into the glenohumeral joint, and reduces the risk of spinal accessory nerve injury in the posterior cervical triangle, or atrophy of the trapezius or supraspinatus muscles from a standard larger dissection. To the authors' knowledge an endoscopically assisted approach to decompressing the suprascapular nerve as it courses deep to the suprascapular ligament has not been reported previously.