Neurosurgery Blog

Neurosurgery 72:590–595, 2013

Although exceedingly rare, wrong-site surgery (WSS) remains a persistent problem in the United States. The incidence is thought to be 2 to 3 per 10 000 craniotomies and about 6 to 14 per 10 000 spine surgeries. In July 2004, the Joint Commission mandated the Universal Protocol (UP) for all accredited hospitals.

OBJECTIVE: To assess the effect of UP implementation on the incidence of neurosurgical WSS at the University of Illinois College of Medicine at Peoria/Illinois Neurological Institute.

METHODS: The Morbidity and Mortality Database in the Department of Neurosurgery was reviewed to identify all recorded cases of WSS since 1999. This was compared with the total operative load (excluding endovascular procedures) of all attending neurosurgeons to determine the incidence of overall WSS. A comparison was then made between the incidences before and after UP implementation.

RESULTS: Fifteen WSS events were found with an overall incidence of 0.07% and Poisson 95% confidence interval of 8.4 to 25. All but one of these were wrong-level spine surgeries (14/15). There was only 1 recorded case of wrong-side surgery and this occurred after implementation of the UP. A statistically greater number of WSS events occurred before (n = 12) in comparison with after (n = 3) UP implementation (P < .001).

CONCLUSION: A statistically significant reduction in overall WSS was seen after implementation of the UP. This reduction can be attributed to less frequent wronglevel spine surgery. There was no case of wrong procedure or patient surgery and the 1 case of wrong-side surgery occurred after UP implementation.

Neurosurgery 72:566–572, 2013

The incidence of radiation-induced complications is increasingly part of the informed consent process for patients undergoing neuroendovascular procedures. Data guiding these discussions in the era of modern radiation-minimizing equipment is lacking.

OBJECTIVE: To quantify the rates of skin and hair effects at a modern high-volume neurovascular center, and to assess the feasibility of accurately quantifying the risk of future central nervous system (CNS) tumor formation.

METHODS: We reviewed a prospectively collected database of endovascular procedures performed at our institution in 2008. The entrance skin dose and brain dose were calculated. Patients receiving skin doses .2 Gy were contacted to inquire about skin and hair changes. We reviewed several recent publications from leading radiation physics bodies to evaluate the feasibility of accurately predicting future cancer risk from neurointerventional procedures.

RESULTS: Seven hundred two procedures were included in the study. Of the patients receiving .2 Gy, 39.6% reported subacute skin or hair changes following their procedure, of which 30% were permanent. Increasing skin dose was significantly associated with permanent hair loss. We found substantial methodological difficulties in attempting to model the risk of future CNS tumor formation given the gaps in our current understanding of the brain’s susceptibility to low-dose ionizing radiation.

CONCLUSION: Radiation exposures exceeding 2 Gy are common in interventional neuroradiology despite modern radiation-minimizing technology. The incidence of side effects approaches 40%, although the majority is self-limiting. Gaps in current models of brain tumor formation after exposure to radiation preclude accurately quantifying the risk of future CNS tumor formation.

J Neurosurg Spine 18:372–377, 2013

The aims of this study were to evaluate a large series of posterior C-1 lateral mass screws (LMSs) to determine accuracy based on CT scanning findings and to assess the perioperative complication rate related to errant screw placement.

Methods. Accuracy of screw placement was evaluated using postoperative CT scans obtained in 196 patients with atlantoaxial instability. Radiographic analysis included measurement of preoperative and postoperative CT scans to evaluate relevant anatomy and classify accuracy of instrumentation placement. Screws were graded using the following definitions: Type I, screw threads completely within the bone (ideal); Type II, less than half the diameter of the screw violates the surrounding cortex (safe); and Type III, clear violation of transverse foramen or spinal canal (unacceptable).

Results. A total of 390 C-1 LMSs were placed, but 32 screws (8.2%) were excluded from accuracy measurements because of a lack of postoperative CT scans; patients in these cases were still included in the assessment of potential clinical complications based on clinical records. Of the 358 evaluable screws with postoperative CT scanning, 85.5% of screws (Type I) were rated as being in the ideal position, 11.7% of screws (Type II) were rated as occupying a safe position, and 10 screws (2.8%) were unacceptable (Type III). Overall, 97.2% of screws were rated Type I or II. Of the 10 screws that were unacceptable on postoperative CT scans, there were no known associated neurological or vertebral artery (VA) injuries. Seven unacceptable screws erred medially into the spinal canal, and 2 patients underwent revision surgery for medial screws. In 2 patients, unilateral C-1 LMSs penetrated the C-1 anterior cortex by approximately 4 mm. Neither patient with anterior C-1 penetration had evidence of internal carotid artery or hypoglossal nerve injury. Computed tomography scanning showed partial entry of C-1 LMSs into the VA foramen of C-1 in 10 cases; no occlusion, associated aneurysm, or fistula of the VA was found. Two patients complained of postoperative occipital neuralgia. This was transient in one patient and resolved by 2 months after surgery. The second patient developed persistent neuralgia, which remained 2 years after surgery, necessitating referral to the pain service.

Conclusions. The technique for freehand C-1 LMS fixation appears to be safe and effective without intraoperative fluoroscopy guidance. Preoperative planning and determination of the ideal screw insertion point, the ideal trajectory, and screw length are the most important considerations. In addition, fewer malpositioned screws were inserted as the study progressed, suggesting a learning curve to the technique.

J Neurosurg 118:243–249, 2013

Awake craniotomy for removal of intraaxial tumors within or adjacent to eloquent brain regions is a well-established procedure. However, awake craniotomy failures have not been well characterized. In the present study, the authors aimed to analyze and assess the incidence and causes for failed awake craniotomy.

Methods. The database of awake craniotomies performed at Tel Aviv Medical Center between 2003 and 2010 was reviewed. Awake craniotomy was considered a failure if conversion to general anesthesia was required, or if adequate mapping or monitoring could not have been achieved.

Results. Of 488 patients undergoing awake craniotomy, 424 were identified as having complete medical, operative, and anesthesiology records. The awake craniotomies performed in 27 (6.4%) of these 424 patients were considered failures. The main causes of failure were lack of intraoperative communication with the patient (n = 18 [4.2%]) and/or intraoperative seizures (n = 9 [2.1%]). Preoperative mixed dysphasia (p < 0.001) and treatment with phenytoin (p = 0.0019) were related to failure due to lack of communication. History of seizures (p = 0.03) and treatment with multiple antiepileptic drugs (p = 0.0012) were found to be related to failure due to intraoperative seizures. Compared with the successful awake craniotomy group, a significantly lower rate of gross-total resection was achieved (83% vs 54%, p = 0.008), there was a higher incidence of short-term speech deterioration postoperatively (6.1% vs 23.5%, p = 0.0017) as well as at 3 months postoperatively (2.3% vs 15.4%, p = 0.0002), and the hospitalization period was longer (4.9 ± 6.2 days vs 8.0 ± 10.1 days, p < 0.001). Significantly more major complications occurred in the failure group (4 [14.8%] of 27) than in the successful group (16 [4%] of 397) (p = 0.037).

Conclusions. Failures of awake craniotomy were associated with a lower incidence of gross-total resection and increased postoperative morbidity. The majority of awake craniotomy failures were preventable by adequate patient selection and avoiding side effects of drugs administered during surgery.

Neurosurgery 72:160–169, 2013

Tailored craniotomies for awake procedures limit cortical exposure. Recently we demonstrated that the identification of eloquent areas increased the risk of postoperative deficits. However, it was not clear whether the observed neurological deficits were caused by proximity of functional cortex to the tumor [cortical injury] or subcortical injury.

OBJECTIVE: We hypothesize that subcortical injury during tumor resection is an important predictor of postoperative neurological deficits compared to cortical injury.

METHODS: A retrospective review of 214 patients undergoing awake craniotomy was carried out in whom preoperative functional magnetic resonance imaging (fMRI) and cortical mapping (CM) were performed. A radiologist blinded to the clinical data reviewed and graded the postoperative changes on diffusion-weighted MR-imaging (DWI).

RESULTS: Of the 40 cases who developed new intraoperative neurological deficit, 36 (90%) occurred during subcortical dissection, 3 (7.5%) during both subcortical and cortical dissection, and 1 (2.5%) during cortical dissection. Neurological dysfunction acquired during subcortical dissection was an independent predictor of postoperative deficits both in the immediate postoperative period (P< .001) and at the 3-month follow-up (P < .001). Significant DWI restriction in the subcortical white matter was predictive of neurological deficits both immediately and at 3 months, P = .011 and P < .001, respectively. New or worsening deficits were seen in 38% of patients; however, at 3 months 13% had a mild persistent neurological deficit.

CONCLUSION: Subcortical injury with significant DWI changes result in postoperative neurological decline despite our efforts to preserve cortical areas of function. This underscores the importance of preserving subcortical fiber tracts during awake craniotomy procedures.

DOI: 10.3171/2012.11.SPINE12755

The minimally invasive lateral transpsoas approach has become an increasingly popular means of fusion. The most frequent complication is related to lumbar plexus nerve injuries; these can be diagnosed based on distribution of neurological deficit following the motor and/or sensory nerve injury. However, the literature has failed to provide a clinically relevant description of these complications. With accurate clinical diagnosis, spine practitioners can provide more precise prognostic and management recommendations to include observation, nerve blocks, neurodestructive procedures, medications, or surgical repair strategies. The purpose of this study was to standardize the clinical findings of lumbar plexopathies and nerve injuries associated with minimally invasive lateral retroperitoneal transpsoas lumbar fusion.

Methods. A thorough literature search of the MEDLINE database up to June 2012 was performed to identify studies that reported lumbar plexus and nerve injuries after the minimally invasive lateral retroperitoneal transpsoas approach. Included studies were assessed for described neurological deficits postoperatively. Studies that did attempt to describe nerve-related complications clinically were excluded. A clinically relevant assessment of lumbar plexus nerve injury was derived to standardize early diagnosis and outline prognostic implications.

Results. A total of 18 studies were selected with a total of 2310 patients; 304 patients were reported to have possible plexus-related complications. The incidence of documented nerve and/or root injury and abdominal paresis ranged from 0% to 3.4% and 4.2%, respectively. Motor weakness ranged from 0.7% to 33.6%. Sensory complications ranged from 0% to 75%. A lack of consistency in the descriptions of the lumbar plexopathies and/or nerve injuries as well as a lack of diagnostic paradigms was noted across studies reviewed. Sensory dermal zones were established and a standardized approach was proposed.

Conclusions. There is underreporting of postoperative lumbar plexus nerve injury and a lack of standardization of clinical findings of neural complications related to the minimally invasive lateral retroperitoneal transpsoas approach. The authors provide a diagnostic paradigm that allows for an efficient and accurate classification of postoperative lumbar plexopathies and nerve injuries.

Neurosurgery 72:65–69, 2013

Microsurgical clip obliteration remains a time-honored and viable option for the treatment of select aneurysms with very low rates of recurrence.

OBJECTIVE: We studied previously clipped aneurysms that were found to have recurrences to better understand the patterns and configurations of these rare entities.

METHODS: A retrospective review was performed of 2 prospectively maintained databases of aneurysm treatments from 2 institutions spanning 14 years to identify patients with recurrence of previously clipped intracranial aneurysms.

RESULTS: Twenty-six aneurysm recurrences were identified. Three types of recurrence were identified: type I, proximal to the clip tines; type II, distal; and type III, lateral. The most common type of recurrence was that arising distal to the clip tines (46.1%), and the least frequently encountered recurrence was that arising proximal to the tines (19.2%). Laterally located recurrences were found in 34.6% of cases.

CONCLUSION: We describe 3 different patterns of aneurysm recurrence with respect to clip application: those occurring proximal, distal, or lateral to the clip tines.

Eur Spine J (2012) 21:2573–2579

Dynamic stabilization of the spine was developed as an alternative to rigid fusion in chronic back pain to reduce the risk of adjacent segment degeneration. Dynamic neutralization system (Dynesys, Zimmer CH) is one of the most popular systems available, but some midterm studies show revision rates as high as 30 %. Some late infectious complications in our patients prompted us to review them systematically. Propionibacterium recently has been shown to cause subtle infections of prosthetic material.

Materials and methods Here, we report on a consecutive series of 50 Dynesys implants. In a median follow-up of 51 months (range 0–91), we identified 12 infectious and 11 non-infectious complications necessitating reoperation or removal of the implant in 17 patients.

Results Material infections occurred after a median of 52 months (2–77) and were due to Propionibacterium alone (n = 4) or in combination (n = 3) in seven out of 11 patients. Clinical presentation combines new or increasing pain associated with signs of screw loosening on conventional X-rays; however, as many as 73.5 % of patients present some degree of screw loosening without being at all symptomatic of infection.

Conclusion The high rate of late infections with lowgrade germs and the frequency of screw loosening signs made us suspect a lack of integration at the bone-screw interface. Surgeons should be suspicious if the patient presents a combination of new or increasing pain and signs of screw loosening, and aggressive revision is recommended in these cases.

Neurosurgery 71[ONS Suppl 2]:ons302–ons307, 2012

Primary closure of the dura in posterior fossa (p-fossa) surgeries is technically difficult and usually requires the use of a dural substitute. A variety of substitutes are currently available and data suggest that autologous materials are preferred in comparison with nonautologous substitutes.

OBJECTIVE: To report our experience using locally harvested autologous pericranium as a dural substitute in patients who underwent p-fossa surgeries.

METHODS: Retrospective analysis of patients who had undergone p-fossa craniotomies between 2005 and 2011. All patients received locally harvested autologous pericranium for duraplasty augmented with a dural sealant. Data were reviewed for complications including: surgical site infection, meningitis, cerebrospinal fluid leak, the radiographic formation of a pseudomeningocele, and any new neurological symptoms related to the incision or repair.

RESULTS: One hundred patients were identified. Indications for surgery included tumor, vascular lesions, or hemorrhage requiring surgical intervention, symptomatic Chiari I malformation, microvascular decompression for trigeminal neuralgia, and trauma requiring surgical decompression. The complication rate was 1% with 1 patient developing an nonsteroidal anti-inflammatory drug-induced aseptic meningitis and graft dehiscence requiring surgical revision.

CONCLUSION: Autologous pericranium with dural sealant augmentation is an effective way to repair the durotomy in p-fossa surgeries. To the best of our knowledge, this is currently the largest study using this technique in the adult neurosurgical literature. Our results report a much lower rate of complications in comparison with other duraplasty studies.

Acta Neurochir (2012) 154:2063–2068

The impact of brain shift on deep brain stimulation surgery is considerable. In DBS surgery, brain shift is mainly caused by CSF loss. CSF loss can be estimated by post-surgical intracranial air. Different approaches and techniques exist to minimize CSF loss and hence brain shift. The aim of this survey was to investigate the extent and dynamics of CSF loss during DBS surgery, analyze its impact on final electrode position, and describe a simple and inexpensive method of burr hole closure.

Methods Sixty-six patients being treated with deep brain stimulation were retrospectively analyzed for this treatise. During surgery, CSF loss was minimized using bone wax as a burr hole closure. Intracranial air volume was calculated based on early post-surgery stereotactic 3D CT and correlated with duration of surgery and electrode deviations derived from post-surgery image fusion.

Results Median early post-surgery intracranial air was 2.1 cm3 (range 0–35.7 cm3, SD 8.53 cm3). No correlation was found between duration of surgery and CSF-loss (R0 0.078, p00.534), indicating that CSF loss mainly occurs early during surgery. Linear regression analysis revealed no significant correlations regarding volume of intracranial air and electrode displacement in any of the three principal axes. No significant difference regarding electrode deviations between first and second side of surgery were observed.

Conclusions CSF loss mainly occurs during the early phase of DBS surgery. CSF loss during a later phase of surgery can be effectively averted by burr hole closure. Postoperative intracranial air volumes up to 35 cm3 did not result in significant electrode displacement in our series. Comparing our results to studies previously published on this subject, burr hole closure using bone wax is highly effective.

J Neurosurg 117:955–961, 2012

The authors undertook this study to investigate the incidence, cause, and clinical influence of short circuits in patients treated with deep brain stimulation (DBS).

Methods. After the incidental identification of a short circuit during routine follow-up, the authors initiated a policy at their institution of routinely evaluating both therapeutic impedance and system impendence at every outpatient DBS follow-up visit, irrespective of the presence of symptoms suggesting possible system malfunction. This study represents a report of their findings after 1 year of this policy.

Results. Implanted DBS leads exhibiting short circuits were identified in 7 patients (8.9% of the patients seen for outpatient follow-up examinations during the 12-month study period). The mean duration from DBS lead implantation to the discovery of the short circuit was 64.7 months. The symptoms revealing short circuits included the wearing off of therapeutic effect, apraxia of eyelid opening, or dysarthria in 6 patients with Parkinson disease (PD), and dystonia deterioration in 1 patient with generalized dystonia. All DBS leads with short circuits had been anchored to the cranium using titanium miniplates. Altering electrode settings resulted in clinical improvement in the 2 PD cases in which patients had specific symptoms of short circuits (2.5%) but not in the other 4 cases. The patient with dystonia underwent repositioning and replacement of a lead because the previous lead was located too anteriorly, but did not experience symptom improvement.

Conclusions. In contrast to the sudden loss of clinical efficacy of DBS caused by an open circuit, short circuits may arise due to a gradual decrease in impedance, causing the insidious development of neurological symptoms via limited or extended potential fields as well as shortened battery longevity. The incidence of short circuits in DBS may be higher than previously thought, especially in cases in which DBS leads are anchored with miniplates. The circuit impedance of DBS should be routinely checked, even after a long history of DBS therapy, especially in cases of miniplate anchoring.

Neurosurgery 71:869–876, 2012

Although the Fisher scale is commonly used to grade vasospasm risk in aneurysmal subarachnoid hemorrhage (aSAH) patients, it fails to account for increasing subarachnoid hemorrhage (SAH) thickness.

OBJECTIVE: We developed a simple quantitative scale based on maximal SAH thickness and compared its reproducibility and ability to predict symptomatic vasospasm against the Fisher scale.

METHODS: The incidence of radiographic and symptomatic vasospasm among 250 aSAH patients treated at our institution was investigated. Admission head computed tomography scans were graded according to the Fisher scale and the proposed scale, which assigns a score from 1 to 5 based on a single measurement of maximum SAH thickness. We calculated vasospasm risk per grade for the Fisher scale and the proposed scale, and compared inter- and intraobserver variability for both scales.

RESULTS: Forty-five patients (20.6%) developed symptomatic vasospasm. On the proposed scale, grade 5 patients were at highest risk, with an odds ratio for symptomatic vasospasm of 11 (95% confidence interval [CI] 2.27-53.37). Odds ratios for proposed grades 4 and 3 were 4.63 (95% CI 1.10-19.59) and 3.04 (95% CI 0.85-10.90), respectively. The odds ratio for Fisher grade 3 was 3.3 (0.96-11.30). Mean inter- and intraobserver agreement was greater for the proposed scale in comparison with the Fisher scale (k0.65 and k0.81 vs k0.51 and k0.35, respectively).

CONCLUSION: The new scale accounted for increasing SAH thickness and was superior to the Fisher scale in inter- and intraobserver agreement and in predicting symptomatic vasospasm, particularly among the highest-risk patients.

Neurosurg Focus 33 (3):E14, 2012

Various studies report outcomes of vestibular schwannoma (VS) surgery, but few studies have compared outcomes across the various approaches. The authors conducted a systematic review of the available data on VS surgery, comparing the different approaches and their associated complications.

Methods. MEDLINE searches were conducted to collect studies that reported information on patients undergoing VS surgery. The authors set inclusion criteria for such studies, including the availability of follow-up data for at least 3 months, inclusion of preoperative and postoperative audiometric data, intraoperative monitoring, and reporting of results using established and standardized metrics. Data were collected on hearing loss, facial nerve dysfunction, persistent postoperative headache, CSF leak, operative mortality, residual tumor, tumor recurrence, cranial nerve (CN) dysfunction involving nerves other than CN VII or VIII, and other neurological complications. The authors reviewed data from 35 studies pertaining to 5064 patients who had undergone VS surgery.

Results. The analyses for hearing loss and facial nerve dysfunction were stratified into the following tumor categories: intracanalicular (IC), size (extrameatal diameter) < 1.5 cm, size 1.5–3.0 cm, and size > 3.0 cm. The middle cranial fossa approach was found to be superior to the retrosigmoid approach for hearing preservation in patients with tumors < 1.5 cm (hearing loss in 43.6% vs 64.3%, p < 0.001). All other size categories showed no significant difference between middle cranial fossa and retrosigmoid approaches with respect to hearing loss. The retrosigmoid approach was associated with significantly less facial nerve dysfunction in patients with IC tumors than the middle cranial fossa method was; however, neither differed significantly from the translabyrinthine corridor (4%, 16.7%, 0%, respectively, p < 0.001). The middle cranial fossa approach differed significantly from the translabyrinthine approach for patients with tumors < 1.5 cm, whereas neither differed from the retrosigmoid approach (3.3%, 11.5%, and 7.2%, respectively, p = 0.001). The retrosigmoid approach involved less facial nerve dysfunction than the middle cranial fossa or translabyrinthine approaches for tumors 1.5–3.0 cm (6.1%, 17.3%, and 15.8%, respectively; p < 0.001). The retrosigmoid approach was also superior to the translabyrinthine approach for tumors > 3.0 cm (30.2% vs 42.5%, respectively, p < 0.001). Postoperative headache was significantly more likely after the retrosigmoid approach than after the translabyrinthine approach, but neither differed significantly from the middle cranial fossa approach (17.3%, 0%, and 8%, respectively; p < 0.001). The incidence of CSF leak was significantly greater after the retrosigmoid approach than after either the middle cranial fossa or translabyrinthine approaches (10.3%, 5.3%, 7.1%; p = 0.001). The incidences of residual tumor, mortality, major non-CN complications, residual tumor, tumor recurrence, and dysfunction of other cranial nerves were not significantly different across the approaches.

Conclusions. The middle cranial fossa approach seems safest for hearing preservation in patients with smaller tumors. Based on the data, the retrosigmoid approach seems to be the most versatile corridor for facial nerve preservation for most tumor sizes, but it is associated with a higher risk of postoperative pain and CSF fistula. The translabyrinthine approach is associated with complete hearing loss but may be useful for patients with large tumors and poor preoperative hearing.

Neurosurg Focus 33 (3):E14, 2012

Various studies report outcomes of vestibular schwannoma (VS) surgery, but few studies have compared outcomes across the various approaches. The authors conducted a systematic review of the available data on VS surgery, comparing the different approaches and their associated complications.

Methods. MEDLINE searches were conducted to collect studies that reported information on patients undergoing VS surgery. The authors set inclusion criteria for such studies, including the availability of follow-up data for at least 3 months, inclusion of preoperative and postoperative audiometric data, intraoperative monitoring, and reporting of results using established and standardized metrics. Data were collected on hearing loss, facial nerve dysfunction, persistent postoperative headache, CSF leak, operative mortality, residual tumor, tumor recurrence, cranial nerve (CN) dysfunction involving nerves other than CN VII or VIII, and other neurological complications. The authors reviewed data from 35 studies pertaining to 5064 patients who had undergone VS surgery.

Results. The analyses for hearing loss and facial nerve dysfunction were stratified into the following tumor categories: intracanalicular (IC), size (extrameatal diameter) < 1.5 cm, size 1.5–3.0 cm, and size > 3.0 cm. The middle cranial fossa approach was found to be superior to the retrosigmoid approach for hearing preservation in patients with tumors < 1.5 cm (hearing loss in 43.6% vs 64.3%, p < 0.001). All other size categories showed no significant difference between middle cranial fossa and retrosigmoid approaches with respect to hearing loss. The retrosigmoid approach was associated with significantly less facial nerve dysfunction in patients with IC tumors than the middle cranial fossa method was; however, neither differed significantly from the translabyrinthine corridor (4%, 16.7%, 0%, respectively, p < 0.001). The middle cranial fossa approach differed significantly from the translabyrinthine approach for patients with tumors < 1.5 cm, whereas neither differed from the retrosigmoid approach (3.3%, 11.5%, and 7.2%, respectively, p = 0.001). The retrosigmoid approach involved less facial nerve dysfunction than the middle cranial fossa or translabyrinthine approaches for tumors 1.5–3.0 cm (6.1%, 17.3%, and 15.8%, respectively; p < 0.001). The retrosigmoid approach was also superior to the translabyrinthine approach for tumors > 3.0 cm (30.2% vs 42.5%, respectively, p < 0.001). Postoperative headache was significantly more likely after the retrosigmoid approach than after the translabyrinthine approach, but neither differed significantly from the middle cranial fossa approach (17.3%, 0%, and 8%, respectively; p < 0.001). The incidence of CSF leak was significantly greater after the retrosigmoid approach than after either the middle cranial fossa or translabyrinthine approaches (10.3%, 5.3%, 7.1%; p = 0.001). The incidences of residual tumor, mortality, major non-CN complications, residual tumor, tumor recurrence, and dysfunction of other cranial nerves were not significantly different across the approaches.

Conclusions. The middle cranial fossa approach seems safest for hearing preservation in patients with smaller tumors. Based on the data, the retrosigmoid approach seems to be the most versatile corridor for facial nerve preservation for most tumor sizes, but it is associated with a higher risk of postoperative pain and CSF fistula. The translabyrinthine approach is associated with complete hearing loss but may be useful for patients with large tumors and poor preoperative hearing.

Eur Spine J (2012) 21:1633–1639 DOI 10.1007/s00586-012-2298-8

Study design Analysis of the Nationwide Inpatient Sample (NIS) from 1998 to 2008.

Objective To analyze the most recent available and nationally representative data for risk factors contributing to in-hospital mortality after primary lumbar spine fusion.

Summary of background data The total number of lumbar spine fusion surgeries has increased dramatically over the past decades. While the field of spine fusion surgery remains highly dynamic with changes in perioperative care constantly affecting patient care, recent data affecting rates and risk for perioperative mortality remain very limited.

Methods We obtained the NIS from the Hospital cost and utilization project. The impact of patient and health care system related demographics, including various comorbidities as well as postoperative complications on the outcome of in-hospital mortality after spine fusion were studied. Furthermore, we analyzed the timing of in-hospital mortality.

Results An estimated total of 1,288,496 primary posterior lumbar spine fusion procedures were performed in the US between 1998 and 2008. The average mortality rate for lumbar spine fusion surgery was 0.2 %. Independent risk factors for in-hospital mortality included advanced age, male gender, large hospital size, and emergency admission. Comorbidities associated with the highest in-hospital mortality after lumbar spine fusion surgery were coagulopathy, metastatic cancer, congestive heart failure and renal disease. Most lethal complications were cerebrovascular events, sepsis and pulmonary embolism. Furthermore, we demonstrated that the timing of death occurred relatively early in the in-hospital period with over half of fatalities occurring by postoperative day 9.

Conclusion This study provides nationally representative information on risk factors for and timing of perioperative mortality after primary lumbar spine fusion surgery. These data can be used to assess risk for this event and to develop targeted intervention to decrease such risk.

Neurosurgery 71:365–380, 2012 DOI: 10.1227/NEU.0b013e31825c3426

Foramen magnum decompression is widely accepted as the treatment of choice for Chiari I malformation. However, important surgical details of the procedure are controversial.

OBJECTIVE: This study analyzes 371 decompressions focusing on intraoperative findings, analysis of complications, and long-term outcomes.

METHODS: Among 644 patients between 1985 and 2010, 359 patients underwent 371 decompressions. Surgery for symptomatic patients consisted of suboccipital craniectomy, C1 laminectomy, arachnoid dissection, and duraplasty. Short-term results were determined after 3 months; long-term outcomes were evaluated with Kaplan-Meier statistics.

RESULTS: The mean age was 40 ± 16 years; mean follow-up was 49 ± 56 months; 75.8% demonstrated syringomyelia. The complication rate was 21.8% with permanent surgical morbidity of 3.2% and surgical mortality of 1.3%. Of the patients, 73.6% reported improvement after 3 months; 21% were unchanged. Overall, 14.3% demonstrated a neurological deterioration within 5 years and 15.4% within 10 years. The severity of neurological symptoms correlated with the grade of arachnoid pathology. Outcome data correlated with the number of previous decompressions, severity of arachnoid pathology, handling of the arachnoid, type of duraplasty, and surgical experience. Firsttime decompressions with arachnoid dissection and an alloplastic duraplasty resulted in surgical morbidity for 2.0%, a 0.9% mortality rate, postoperative improvement after 3 months for 82%, and neurological recurrence rates of 7% after 5 years and 8.7% after 10 years.

CONCLUSION: Arachnoid pathology in Chiari I malformation has an impact on clinical symptoms and postoperative results. Decompressions with arachnoid dissection and an alloplastic duraplasty performed by surgeons experienced with this pathology offer a favorable long-term prognosis.

Neurosurg Focus 33 (2):E14, 2012

Postoperative visual loss (POVL) after spine surgery performed with the patient prone is a rare but devastating postoperative complication. The incidence and the mechanisms of visual loss after surgery are difficult to determine. The 4 recognized causes of POVL are ischemic optic neuropathy (approximately 89%), central retinal artery occlusion (approximately 11%), cortical infarction, and external ocular injury.

There are very limited guidelines or protocols on the perioperative practice for “prone-position” surgeries. However, new devices have been designed to prevent mechanical ocular compression during prone-position spine surgeries.

The authors used PubMed to perform a literature search for devices used in prone-position spine surgeries. A total of 7 devices was found; the authors explored these devices’ features, advantages, and disadvantages. The cause of POVL seems to be a multifactorial problem with unclear pathophysiological mechanisms. Therefore, ocular compression is a critical factor, and eliminating any obvious compression to the eye with these devices could possibly prevent this devastating perioperative complication.

J Neurosurg Spine 17:30–36, 2012

Incidental durotomies (IDs) are an unfortunate but anticipated potential complication of spinal surgery. The authors surveyed the frequency of IDs for a single spine surgeon and analyzed the major risk factors as well as the impact on long-term patient outcomes.

Methods. The authors conducted a prospective review of elective spinal surgeries performed over a 15-year period. Any surgery involving peripheral nerve only, intradural procedures, or dural tears due to trauma were excluded from analysis. The incidence of ID was categorized by surgery type including primary surgery, revision surgery, and so forth. Incidence of ID was also examined in the context of years of physician experience and training. Furthermore, the incidence and types of sequelae were examined in patients with an ID.

Results. Among 3000 elective spinal surgery cases, 3.5% (104) had an ID. The incidence of ID during minimally invasive procedures (3.3%) was similar, but no patients experienced long-term sequelae. The incidence of ID during revision surgery (6.5%) was higher. There was a marked difference in incidence between cervical (1.3%) and thoracolumbar (5.1%) cases. The incidence was lower for cases involving instrumentation (2.4%). When physician training was examined, residents were responsible for 49% of all IDs, whereas fellows were responsible for 26% and the attending for 25%. Among all of the cases that involved an ID, 7.7% of patients went on to experience a neurological deficit as compared with 1.5% of those without an ID. The overall failure rate of dural repair was 6.9%, and failure was almost 3 times higher (13%) in revision surgery as compared with a primary procedure (5%).

Conclusions. The authors established a reliable baseline incidence for durotomy after spine surgery: 3.5%. They also identified risk factors that can increase the likelihood of a durotomy, including location of the spinal procedure, type of procedure performed, and the implementation of a new procedure. The years of physician training or resident experience did not appear to be a major risk for ID.

Neuroradiology (2012) 54:719–726 DOI 10.1007/s00234-011-0950-3

The authors present a series of patients in whom early rebleeding occurred after coiling for ruptured aneurysms. We investigated the incidence and possible mechanisms of early rebleeding.

Methods This study consisted of 1,167 consecutive patients who underwent coiling for a ruptured saccular aneurysm. Clinical and radiological data were collected retrospectively from three institutions. Early rebleeding was defined as occurrence of further bleeding within 30 days after coiling with worsening of the patient’s condition. We divided early rebleeding into hyperacute, subacute, and delay groups depending on the timing of rebleeding after coil embolization.

Results Incidence of early rebleeding after coiling of a ruptured saccular aneurysm was 1.1% (13 of 1,167), and mortality was 31% (4 of 13) in our series. Out of ten patients in hyperacute group, three (30%) had incomplete occlusion result and six patients (60%) underwent intraarterial (IA) infusion of abciximab or tirofiban during the procedures. Seven patients (70%) had an intracerebral hemorrhage (ICH) on initial computed tomography. Four patients died, another four sustained severe disabilities, and the others had good recovery. All three patients in subacute and delay group showed recanalization on post-rebleeding angiography and made an excellent recovery.

Conclusion Early rebleeding was associated with high mortality and morbidity. IA abciximab infusion or thrombolytic interventions during the procedure, maintenance of anticoagulation after the procedure, incomplete treatment of the aneurysms, and presence of ICH seemed to be related to hyperacute early rebleeding after coiling. Increased aneurysmal size and coil compaction could induce subacute and delayed early rebleeding.

Neurosurg Pediatrics 10:39–43, 2012. http://thejns.org/doi/abs/10.3171/2012.4.PEDS1222

Electrolyte and endocrinological complications of endoscopic third ventriculostomy (ETV) are infrequent but serious events, likely due to transient hypothalamic-pituitary dysfunction. While the incidence of diabetes insipidus is relatively well known, hyponatremia is not often reported. The authors report on a series of 5 patients with post-ETV hyponatremia.

Methods. The records of patients undergoing ETV between 2008 and 2010 were reviewed. All ETVs were performed with a rigid neuroendoscope via a frontal bur hole, standard third ventricle floor blunt perforation, Fogarty catheter dilation, and intermittent normal saline irrigation. Postoperative MR images were evaluated for endoscope tract injury as well as the trajectory from the bur hole center to the fenestration site.

Results. Thirty-two patients (20 male and 12 female) underwent ETV. Their median age was 6 years (range 3 weeks–28 years). Hydrocephalus was most commonly due to nontumoral aqueductal stenosis (43%), nontectal tumor (25%), or tectal glioma (13%). Five patients (16%) had multicystic/loculated hydrocephalus. Five patients (16%) developed hyponatremia between 1 and 8 days following ETV, including 2 patients with seizures (1 of whom was still hospitalized at the time of the seizure and 1 of whom was readmitted as a result of the seizure) and 3 patients who were readmitted because of decline in their condition following routine discharge. No hypothalamic injuries were noted on imaging. Univariate risk factors consisted of age of 2 years or less (p = 0.02), presence of cystic lesions (p = 0.02), and ETV trajectory angle 10° or more from perpendicular (p = 0.001).

Conclusions. Endoscopic third ventriculostomy is a well-tolerated procedure but can result in serious complications. Hyponatremia is rare and may be more likely in younger patients or those with cystic loculations. Patients with altered craniometry may be at particular risk with a rigid endoscopic approach requiring greater manipulation of subforniceal or hypothalamic structures.