J Neurosurg 116:54–65, 2012. DOI: 10.3171/2011.9.JNS11177

The object of this study was to define the long-term outcomes and risks of arteriovenous malformation (AVM) management using 2 or more stages of stereotactic radiosurgery (SRS) for symptomatic large-volume lesions unsuitable for surgery.

Methods. In 1992, the authors prospectively began to stage the treatment of anatomical components to deliver higher single doses to AVMs with a volume of more than 10 cm3. Forty-seven patients with such AVMs underwent volume-staged SRS. In this series, 18 patients (38%) had a prior hemorrhage and 21 patients (45%) underwent prior embolization. The median interval between the first-stage SRS and the second-stage SRS was 4.9 months (range 2.8–13.8 months). The median target volume was 11.5 cm3 (range 4.0–26 cm3) in the first-stage SRS and 9.5 cm3 in the second-stage SRS. The median margin dose was 16 Gy (range 13–18 Gy) for both stages.

Results. In 17 patients, AVM obliteration was confirmed after 2–4 SRS procedures at a median follow-up of 87 months (range 0.4–209 months). Five patients had near-total obliteration (volume reduction > 75% but residual AVM). The actuarial rates of total obliteration after 2-stage SRS were 7%, 20%, 28%, and 36% at 3, 4, 5, and 10 years, respectively. The 5-year total obliteration rate after the initial staged volumetric SRS with a margin dose of 17 Gy or more was 62% (p = 0.001). Sixteen patients underwent additional SRS at a median interval of 61 months (range 33–113 months) after the initial 2-stage SRS. The overall rates of total obliteration after staged and repeat SRS were 18%, 45%, and 56% at 5, 7, and 10 years, respectively. Ten patients sustained hemorrhage after staged SRS, and 5 of these patients died. Three of 16 patients who underwent repeat SRS sustained hemorrhage after the procedure and died. Based on Kaplan-Meier analysis (excluding the second hemorrhage in the patient who had 2 hemorrhages), the cumulative rates of AVM hemorrhage after SRS were 4.3%, 8.6%, 13.5%, and 36.0% at 1, 2, 5, and 10 years, respectively. This corresponded to annual hemorrhage risks of 4.3%, 2.3%, and 5.6% for Years 0–1, 1–5, and 5–10 after SRS. Multiple hemorrhages before SRS correlated with a significantly higher risk of hemorrhage after SRS. Symptomatic adverse radiation effects were detected in 13% of patients, but no patient died as a result of an adverse radiation effect. Delayed cyst formation did not occur in any patient after SRS.

Conclusions. Prospective volume-staged SRS for large AVMs unsuitable for surgery has potential benefit but often requires more than 2 procedures to complete the obliteration process. To have a reasonable chance of benefit, the minimum margin dose should be 17 Gy or greater, depending on the AVM location. In the future, prospective volumestaged SRS followed by embolization (to reduce flow, obliterate fistulas, and occlude associated aneurysms) may improve obliteration results and further reduce the risk of hemorrhage after SRS.

Neurosurgery 69:1255–1260, 2011 DOI: 10.1227/NEU.0b013e31822ba315

Patients with medically unresponsive trigeminal neuralgia (TN) who are >70 years of age often undergo operations that typically provide pain relief for <5 years despite having a life expectancy that can exceed 15 years.

OBJECTIVE: To review the safety and efficacy of posterior fossa exploration (PFE) for TN patients >70 years of age.

METHODS: From 1999 to 2009, 67 TN patients >70 years of age (median, 74 years) underwent a PFE. Thirty-seven patients (55%) had failed ≥1 prior surgeries (median, 2). Fifty-nine patients (88%) had a microvascular decompression, and 8 patients (12%) underwent a partial sensory rhizotomy. Follow-up (median, 40 months) was censored at the time of last contact (n = 51), additional surgery (n = 12), or death (n = 4).

RESULTS: Complete pain relief (no pain, no medications) was 87% at 1 year and 78% at 5 years. Facial pain outcomes did not correlate with patient age, sex, prior surgery, or pain duration. Postoperative complications were noted in 10 patients (15%) and included ataxia (10%), hearing loss (5%), trigeminal dysesthesias (5%), facial weakness (3%), aseptic meningitis (2%), and pulmonary embolus (2%). Factors associated with postoperative complications were prior PFE (P = .01) and neurovascular compression from a dolicoectatic basilar artery (P = .03).

CONCLUSION: Posterior fossa exploration is safe and effective for physiologically healthy TN patients .70 years of age. It should be deferred in older patients with TN secondary to a dolicoectatic basilar artery and patients who have persistent/recurrent pain after a previous PFE unless simpler procedures prove ineffective at controlling their facial pain.

J Neurosurg 115:151–158, 2011. DOI: 10.3171/2011.1.JNS101492

Traditional ventriculoperitoneal (VP) shunt surgery involves insertion of the distal catheter by minilaparotomy. However, minilaparotomy may be a significant source of morbidity during shunt surgery. Laparoscopic insertion of the distal catheter is an alternative technique that may simplify and improve the safety of shunt surgery.

Methods. The authors performed a retrospective review of hospital records of all patients undergoing new VP shunt insertion at a tertiary care center between 2004 and 2009. Patient characteristics and outcomes were compared between patients undergoing open or laparoscopic insertion of the distal catheter. Independent variables in the analysis included age, sex, race, body mass index, surgical technique, previous VP shunt placement, previous abdominal procedures, American Society of Anesthesiology (ASA) score, and indication for shunt placement. Dependent variables included the occurrence of shunt failure, cause of shunt failure, complications, length of stay (LOS), LOS after shunt placement, estimated blood loss, and operative time.

Results. The authors identified 810 patients who met the inclusion criteria; open or laparoscopic distal catheter insertion was performed in 335 and 475 patients, respectively. There were no significant differences between the groups regarding age, race, ASA score, or indication for shunt placement. The most common indication was hydrocephalus due to subarachnoid hemorrhage, followed by tumor-associated hydrocephalus, normal pressure hydrocephalus (NPH), and hydrocephalus due to trauma. The incidence of shunt failure was not statistically different between cohorts, occurring in 20.0% of laparoscopic and 20.9% of open catheter placement cases (p = 0.791). With analysis of causes of shunt failure, shunt obstruction occurred significantly more often in the open surgery cohort (p = 0.012). In patients with a known cause shunt obstruction, distal obstruction occurred in 35.7% of the open cohort obstructions and 4.8% of the laparoscopic cohort obstructions (p = 0.014). The relative risk of distal obstruction in open cases compared with laparoscopic cases was 7.50. Infections occurred in 8.2% of laparoscopic cases compared with 6.6% of open cases (p = 0.419). Within the NPH subgroup, the laparoscopically treated patients had significantly more overdrainage (p = 0.040), whereas those in the open cohort experienced significantly more shunt obstructions (p = 0.034). Laparoscopically treated patients had shorter operative times (p < 0.0005), inpatient LOS (p < 0.001), and inpatient LOS after VP shunt placement (p = 0.01) as well as less blood loss (p = 0.058).

Conclusions. To our knowledge this is the largest reported comparison of distal VP shunt catheter insertion techniques. Compared with minilaparotomy, the laparoscopic approach was associated with decreased time in the operating room and a decreased LOS. Moreover, laparoscopy was associated with fewer distal shunt obstructions. Laparoscopic shunt surgery is a viable alternative to traditional shunt surgery.

Neurosurgery 69:38–44, 2011 DOI: 10.1227/NEU.0b013e3182134171

Incidental durotomy is a familiar encounter during surgery for lumbar spinal stenosis. The impact of durotomy on long-term outcomes remains a matter of debate.

OBJECTIVE: To determine the impact of durotomy on the long-term outcomes of patients in the Spine Patient Outcomes Research Trial (SPORT).

METHODS: The SPORT cohort participants with a confirmed diagnosis of spinal stenosis, without associated spondylolisthesis, undergoing standard, first-time, open decompressive laminectomy, with or without fusion, were followed up from baseline at 6 weeks, and 3, 6, and 12 months and yearly thereafter at 13 spine clinics in 11 US states. Patient data from this prospectively gathered database were reviewed. As of May 2009, the mean follow-up among all analyzed patients was 43.8 months.

RESULTS: Four hundred nine patients underwent first-time open laminectomy with or without fusion. Thirty-seven of these patients (9%) had an incidental durotomy. No significant differences were observed with or without durotomy in age; sex; race; body mass index; the prevalence of smoking, diabetes mellitus, and hypertension; decompression level; number of levels decompressed; or whether an additional fusion was performed. The durotomy group had significantly increased operative duration, operative blood loss, and inpatient stay. There were, however, no differences in incidence of nerve root injury, mortality, additional surgeries, or primary outcomes (Short Form-36 Bodily Pain or Physical Function scores or Oswestry Disability Index) at yearly follow-ups to 4 years.

CONCLUSIONS: Incidental durotomy during first-time lumbar laminectomy for spinal stenosis did not impact long-term outcomes in affected patients.

Neurosurgery 68:1259–1269, 2011 DOI: 10.1227/NEU.0b013e31820c0441

In order to weigh the risks of surgery against the presumed advantages, it is important to have specific knowledge about complication rates.

OBJECTIVE: To study the surgical mortality and rate of reoperations for hematomas and infections after intracranial surgery for brain tumors in a large, contemporary, single-institution consecutive series.

METHODS: All adult patients from a well-defined population of 2.7 million inhabitants who underwent craniotomies for intracranial tumors at Oslo University Hospital from 2003 to 2008 were included (n = 2630). The patients were identified from our prospectively collected database and their charts studied retrospectively. Follow-up was 100%.

RESULTS: The overall surgical mortality, defined as death within 30 days of surgery, was 2.3% (n = 60). The mortality rates for high- and low-grade gliomas, meningiomas, and metastases were 2.9%, 1.0%, 0.9%, and 4.5%, respectively. Age >60 (odds ratio 1.84, P< 0.05) and biopsy compared with resection (odds ratio 4.67, P <0.01) were significantly positively associated with increased surgical mortality. Hematomas accounted for 35% of the surgical mortality. Postoperative hematomas needing evacuation occurred in 2.1% (n = 54). Age >60 was significantly correlated to increased risk of postoperative hematomas (odds ratio 2.43, P < 0.001). A total of 39 patients (1.5%) were reoperated for postoperative infection. Meningiomas had an increased risk of infections compared with high-grade gliomas (odds ratio 4.61, P < 0.001).

CONCLUSION: The surgical mortality within 30 days of surgery was 2.3%, with age >60 and biopsy vs resection being the 2 factors significantly associated with increased mortality. Postoperative hematomas caused about one third of the surgical mortality.

Neurosurgery 68:491–495, 2011 DOI: 10.1227/NEU.0b013e3181ffa21c

Cerebrospinal fluid (CSF) diversionary methods are fraught with complications (eg, infection, obstruction, and CSF malabsorption at the distal site).

INTRODUCTION: The authors investigated the sternum, specifically the manubrium, as a potential CSF receptacle for patients with hydrocephalus.

METHODS: Five fresh adult human cadavers had the manubrium cannulated in a suprasternal location. Tap water was infused via a metal trocar for approximately 60 minutes. Additionally, morphometric examination of the manubrium from 40 adult human skeletons was performed. Next, 4 anesthetized rhesus monkeys underwent cannulation of the manubrium: 2 were infused with 50 mL of saline over approximately 1 hour, and 2 were infused by gravity drip of saline over 24 hours. Finally, 2 adult pigs underwent long-term ventriculosternal tube placement with analysis for function and potential development of osteomyelitis.

RESULTS: Thirty liters of water were injected into all cadaveric specimens without overflow or noticeable edema. No fluid accumulation was identified. The manubrium had a mean length, width, and thickness of 5.1 cm, 5.0 cm, and 1 cm, respectively. The animals that underwent infusion of 50 mL of saline and the animals that underwent gravity drip tolerated the procedure without vital sign changes or evidence of saline leakage into the pleural cavity. The 2 pigs did not show any vital sign changes, and, 2 weeks post procedure, they had no findings of osteomyelitis.

CONCLUSION: Based on our studies, the manubrium of the sternum appears to be an ideal location for the placement of the distal end of a CSF diversionary shunt when other anatomic receptacles are not an option. In vivo human studies are now required to verify our findings.

Spine 2010;35: S322–S330

Study Design. Prospective multicenter nonrandomized institutional review board-approved observational study of clinical and radiographic outcomes of the extreme lateral interbody fusion (XLIF) procedure in adult scoliosis.

Objective. Perioperative measures from this longitudinal study were compiled to identify the short-term results and complications of the procedure.

Summary of Background Data. The surgical treatment of adult scoliosis presents a treatment challenge. Neural decompression with combined anterior/posterior instrumented fusion is often performed. These procedures have been reported to carry a high risk of complication, particularly in the elderly patient population. Over the past decade, less invasive surgical approaches to neural decompression and fusion have been popularized and have recently been applied in the treatment of degenerative scoliosis. To date, there has been little published data evaluating these treatment approaches.

Methods. A total of 107 patients who underwent the XLIF procedure with or without supplemental posterior fusion for the treatment of degenerative scoliosis were prospectively studied. Intraoperative data collection included surgical procedural details, operative time, estimated blood loss, and surgical complications. Postoperative complications, length of hospital stay, and neurologic status were recorded. For this report, perioperative data (inclusive of outcomes through the 6-week postoperative clinic visit) were evaluated.

Results. In all, 107 patients (mean age, 68 years; range, 45–87) were treated with XLIF; 28% had at least 1 comorbidity. A mean of 4.4 levels (range, 1–9) were treated per patient. Supplemental pedicle screw fixation was used in 75.7% of patients, 5.6% had lateral fixation, and 18.7% had stand-alone XLIF. Mean operative time and blood loss were 178 minutes (58 minutes/level) and 50 to 100 mL. Mean hospital stay was 2.9 days (unstaged), 8.1 day (staged, 16.5%), 3.8 days overall. Five patients (4.7%) received a transfusion, 3 (2.8%) required intensive care unit admission, and 1 (0.9%) required rehabilitation services. Major complications occurred in 13 patients (12.1%): 2 (1.9%) medical, 12 (11.2%) surgical. Of procedures that involved only less invasive techniques (XLIF stand-alone or with percutaneous instrumentation), 9.0% had one or more major complications. In those with supplemental open posterior instrumentation, 20.7% had one or more major complication. Early reoperations (3) (all for deep wound infections) were associated with open posterior instrumentation procedures.

Conclusion. The morbidity in adult scoliosis surgery is minimized with less invasive techniques. The rate of major complications in this study (12.1%) compares favorably to that reported from other studies of surgery for degenerative deformity.

Neurosurgery 68:117–124, 2011 DOI: 10.1227/NEU.0b013e3181fcf14e

Unintended durotomy is a common complication of spinal surgery. However, the incidences reported in the literature vary widely and are based primarily on relatively small case numbers from a single surgeon or institution.

OBJECTIVE: To provide spine surgeons with a reliable incidence of unintended durotomy in spinal surgery and to assess various factors that may influence the risk of durotomy.

METHODS: We assessed 108.478 surgical cases prospectively submitted by members of the Scoliosis Research Society to a deidentified database from 2004 to 2007.

RESULTS: Unintended durotomy occurred in 1.6% (1745 of 108.478) of all cases. The incidence of unintended durotomy ranged from 1.1% to 1.9% on the basis of preoperative diagnosis, with the highest incidence among patients treated for kyphosis (1.9%) or spondylolisthesis (1.9%) and the lowest incidence among patients treated for scoliosis (1.1%). The most common indication for spine surgery was degenerative spinal disorder, and among these patients, there was a lower incidence of durotomy for cervical (1.0%) vs thoracic (2.2%; P = .01) or lumbar (2.1%, P < .001) cases. Scoliosis procedures were further characterized by etiology, with the highest incidence of durotomy in the degenerative subgroup (2.2% vs 1.1%; P < .001). Durotomy was more common in revision compared with primary surgery (2.2% vs 1.5%; P < .001) and was significantly more common among elderly (. 80 years of age) patients (2.2% vs 1.6%; P = .006). There was a significant association between unintended durotomy and development of a new neurological deficit (P < .001).

CONCLUSION: Unintended durotomy occurred in at least 1.6% of spinal surgeries, even among experienced surgeons. Our data provide general benchmarks of durotomy rates and serve as a basis for ongoing efforts to improve safety of care.

SPINE Volume 36, Number 1, pp 26–33

Study Design. Prospective analysis of 600 extreme lateral interbody fusion (XLIF) approach procedures for intraoperative and perioperative complications.

Objective. To delineate and describe complications in a large, prospective series of minimally invasive lateral lumbar fusion procedures (XLIF).

Summary of Background Data. While some small series of lateral lumbar fusion have discussed complications, no results from large studies have been reported.

Methods. A total of 600 patients were treated with a lateral approach to fusion (XLIF) for degenerative spinal conditions. Data were collected prospectively on all patients and analyzed for demographic, diagnostic, and hospitalization information to identify operative and early postoperative complications. Documented complication types and rates in this large series were compared with smaller prior reports on lateral approach fusions, as well as other minimally invasive (mini-anterior lumbar interbody fusion and minimally invasive surgical [MIS] transforaminal lumbar interbody fusion) and more traditional fusion approaches (posterior intertransverse fusion, anterior lumbar interbody fusion, posterior lumbar interbody fusion, transforaminal lumbar interbody fusion).

Results. Seven hundred forty-one levels were treated, 80.8% single level, 15.0% 2 level, 4.0% 3 level, 0.2% 4 level; 59.3%, including the L4 to L5 levels. A total of 99.2% included supplemental internal fi xation; 83.2% included pedicle screw fi xation (predominantly unilateral). Hemoglobin change from pre- to postoperation averaged 1.38. Hospital stay averaged 1.21 days. The overall incidence of perioperative complications (intraoperation and out to 6 weeks postoperation) was 6.2%: 9 (1.5%) in-hospital surgery-related events, 17 (2.8%) in-hospital medical events, 6 (1.0%) out-of-hospital surgery-related events, and 5 (0.8%) out-of-hospital medical events. There were no wound infections, no vascular injuries, no intraoperative visceral injuries, and 4 (0.7%) transient postoperative neurologic defi cits. Eleven events (1.8%) resulted in additional procedures/reoperation.

Conclusions. Compared with traditional open approaches, the MIS lateral approach to fusion by using the XLIF technique resulted in a lower incidence of infection, visceral and neurologic injury, and transfusion as well as markedly shorter hospitalization. Complications in MIS XLIF compare favorably with those from other MIS fusion procedures; duration of hospitalization is shorter than with any previously reported technique.