Comparing Cortical Bone Trajectory and Traditional Pedicle Screws in Transforaminal Lumbar Interbody Fusion: A Retrospective Cohort Study of One-Year Outcomes.

INTRODUCTION: This is a retrospective study of consecutive patients undergoing transforaminal lumbar interbody fusion (TLIF) at a single institution. The objective of this study was to compare the long-term results associated with cortical bone trajectory (CBT) and traditional pedicle screw (TPS) via posterolateral approach in TLIF. METHODS: Consecutive patients treated from November 2014 to March 2019 were included in the CBT TLIF group, while consecutive patients treated from October 2010 to August 2017 were included in the TPS TLIF group. Inclusion criteria comprised single-level or two-level TLIF for degenerative spondylolisthesis with stenosis and at least one year of clinical and radiographic follow-up. Variables of interest included pertinent preoperative, perioperative, and postoperative data. Non-parametric evaluation was performed using the Wilcoxon test. Fisher's exact test was used to assess group differences for nominal data. RESULTS: Overall, 140 patients met the inclusion criteria; 69 patients had CBT instrumentation (mean follow-up 526 days) and 71 patients underwent instrumentation placement via TPS (mean follow-up 825 days). Examination of perioperative and postoperative outcomes demonstrate comparable results between the groups with perioperative complications, length of stay, discharge destination, surgical revision rate, and fusion rates all being similar between groups (p = 0.1; p = 0.53; p = 0.091; p = 0.61; p = 0.665, respectively). CONCLUSIONS: CBT in the setting of TLIF offer equivalent outcomes to TPS with TLIF at both short- and long-term intervals of care.

Telemedicine in Spine Surgery: Outcomes for 138 Patients With Virtual Preoperative Assessment Compared to Historical Controls.

INTRODUCTION: COVID-19 has accelerated the use of telemedicine in all aspects of health care delivery, including initial surgical evaluation. No existing literature investigates the safety and efficacy of telemedicine to preoperatively evaluate spine surgery candidates. Our objectives were: (1) Compare the change in visual analogue scale (VAS) scores between the telemedicine preoperative visit and in-person preoperative visit groups. (2) Compare the average surgical time, estimated blood loss (EBL), length of hospital stay (LOS), rates of intraoperative complications, rates of readmission, and rates of reoperation between the telemedicine preoperative visit and in-person preoperative visit groups. METHODS: The previously stated metrics were collected for 276 patients, 138 who were exclusively evaluated preoperatively with telemedicine and 138 historical controls who were evaluated preoperatively in person. We used χ2 and independent samples t tests to determine significance. RESULTS: There were no significant differences in the mean change in VAS scores (-2.7 ± 3.1 telemedicine vs. -2.2 ± 3.7 in-person, P = 0.317), mean percentage change in VAS scores (-40.5% ± 54.3% vs. -39.5% ± 66.6%, P = 0.811), mean surgical time (2.4 ± 1.4 hours vs. 2.3 ± 1.3 ours, P = 0.527), mean EBL (150.4 ± 173.3 mL vs. 156.7 ± 255.0 mL, P = 0.811), mean LOS (3.3 ± 2.4 days vs. 3.3 ± 2.5 days, P = 0.954), intraoperative complication rates (0.7% vs. 1.4%, P = 0.558), reoperation rates (7.9% vs. 4.3%, P = 0.208), or readmission rates (10.1% vs. 5.1%, P = 0.091) between the telemedicine preoperative visit and in-person preoperative visit groups. CONCLUSIONS: Preoperative evaluation via telemedicine leads to the same short-term surgical outcomes as in-person evaluation with no increased risk of surgical complications.

Telemedicine in the Evaluation and Management of Neurosurgical Spine Patients: Questionnaire Assessment of 346 Consecutive Patients.

STUDY DESIGN: Retrospective questionnaire study of all patients seen via telemedicine during the COVID-19 pandemic at a large academic institution. OBJECTIVE: This aim of this study was to compare patient satisfaction of telemedicine clinic to in-person visits; to evaluate the preference for telemedicine to in-person visits; to assess patients' willingness to proceed with major surgery and/or a minor procedure based on a telemedicine visit alone. SUMMARY OF BACKGROUND DATA: One study showed promising utility of mobile health applications for spine patients. No studies have investigated telemedicine in the evaluation and management of spine patients. METHODS: An 11-part questionnaire was developed to assess the attitudes toward telemedicine for all patients seen within a 7-week period during the COVID-19 crisis. Patients were called by phone to participate in the survey. χ2 and the Wilcoxon Rank-Sum Test were performed to determine significance. RESULTS: Ninety-five percent were "satisfied" or "very satisfied" with their telemedicine visit, with 62% stating it was "the same" or "better" than previous in-person appointments. Patients saved a median of 105 minutes by using telemedicine compared to in-person visits. Fifty-two percent of patients have to take off work for in-person visits, compared to 7% for telemedicine. Thirty-seven percent preferred telemedicine to in-person visits. Patients who preferred telemedicine had significantly longer patient-reported in-person visit times (score mean of 171) compared to patients who preferred in-person visits (score mean of 137, P = 0.0007). Thirty-seven percent of patients would proceed with surgery and 73% would proceed with a minor procedure based on a telemedicine visit alone. CONCLUSION: Telemedicine can increase access to specialty care for patients with prolonged travel time to in-person visits and decrease the socioeconomic burden for both patients and hospital systems. The high satisfaction with telemedicine and willingness to proceed with surgery suggest that remote visits may be useful for both routine management and initial surgical evaluation for spine surgery candidates.Level of Evidence: 3.

Safety Profile and Radiographic and Clinical Outcomes of Stand-Alone 2-Level Anterior Lumbar Interbody Fusion: A Case Series of 41 Consecutive Patients.

OBJECTIVE: The use of stand-alone 2-level anterior lumbar interbody fusion (ALIF) for degenerative lumbar disease has been increasing as an alternative to routinely augmenting these constructs with posterior fixation or fusion. Despite the potential benefits of a stand-alone approach (decreased cost and operative time, decreased pain and early mobilization), there is a paucity of information regarding these operations in the literature. This investigation aimed to determine the safety profile, radiographic outcomes including fusion rates, improvement in preoperative pain, and spinopelvic parameter modification, for patients undergoing stand-alone 2-level ALIF. METHODS: This retrospective case series involved a chart review of all patients undergoing 2-level stand-alone ALIF at a single tertiary hospital from 2008 to 2018. Data included patient demographics, hospitalization, complications and radiological studies. Visual analog scale (VAS) back and leg scores were measured via patient-administered surveys preoperatively and up to 18 weeks postoperatively. RESULTS: Forty-one patients who underwent L4-S1 stand-alone ALIF were included. Sixteen (39%) of patients had undergone previous posterior lumbar surgery. Length of stay averaged 4.2 days. Complication rates were comparable to 1-level ALIF. Two patients required reoperation. Fusion rates were 100% for L4-5 and 94.4% for L5-S1. There was no significant change in lumbar lordosis (LL) or LL-pelvic incidence (PI), but there was improved segmental lordosis (SL) and disc height at L4-S1 on final follow-up imaging. There was also modest but statistically significant improvement in VAS back and leg scores. CONCLUSIONS: Stand-alone 2-level ALIF is an option for a surgeon to perform in the absence of significant instability, even in the setting of prior posterior surgery. These procedures increase SL and disc height, but do not have the same effect on LL or LL-PI.

Motion-preserving, 2-stage transoral and posterior treatment of an unstable Jefferson fracture in a professional football player.

The authors report the case of a patient who suffered a Jefferson fracture during a professional football game. The C-1 (atlas) fracture was widely displaced anteriorly, but the transverse ligament was intact. In an effort to enable a return to play and avoid intersegmental (C1-2) fusion, the patient underwent a transoral approach for open reduction and internal fixation of the fracture. The associated posterior ring fracture displacement widened after this procedure, and a subsequent posterior arthrodesis and fixation of the fracture site was performed 6 months later when the fracture failed to heal with rigid collar immobilization. The approach maintained the normal range of motion at the atlantoaxial and atlantooccipital joints, which would have been sacrificed by an atlantoaxial or occipitocervical fusion, as is traditionally performed. Ultimately, the patient decided not to return to the football field, but this approach could avoid the more significant loss of motion associated with atlantoaxial or occipitocervical fusion for unstable Jefferson fractures.

Percutaneous Instrumentation Without Arthrodesis for Thoracolumbar Flexion-Distraction Injuries: A Review of the Literature.

Background: Flexion-distraction injuries (FDI) represent 5% to 15% of traumatic thoracolumbar fractures. Treatment depends on the extent of ligamentous involvement: osseous/Magerl type B2 injuries can be managed conservatively, while ligamentous/Magerl type B1 injuries undergo stabilization with arthrodesis. Minimally invasive surgery without arthrodesis can achieve similar outcomes to open procedures. This has been studied for burst fractures; however, its role in FDI is unclear. Objective: To conduct a systematic review of the literature that examined minimally invasive surgery instrumentation without arthrodesis for traumatic FDI of the thoracolumbar spine. Methods: Four electronic databases were searched, and articles were screened using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines for patients with traumatic FDI of the thoracolumbar spine treated with percutaneous techniques without arthrodesis and had postoperative follow-up. Results: Seven studies with 44 patients met inclusion criteria. There were 19 patients with osseous FDI and 25 with ligamentous FDI. When reported, patients (n = 39) were neurologically intact preoperatively and at follow-up. Osseous FDI patients underwent instrumentation at 2 levels, while ligamentous injuries at approximately 4 levels. Complication rate was 2.3%. All patients had at least 6 mo of follow-up and demonstrated healing on follow-up imaging. Conclusion: Percutaneous instrumentation without arthrodesis represents a low-risk intermediate between conservative management and open instrumented fusion. This "internal bracing" can be used in osseous and ligamentous FDIs. Neurologically intact patients who do not require decompression and those that may not tolerate or fail conservative management may be candidates. The current level of evidence cannot provide official recommendations and future studies are required to investigate long-term safety and efficacy.

Anterior Cervical Discectomy and Fusion: Comparison of Fusion, Dysphagia, and Complication Rates Between Recombinant Human Bone Morphogenetic Protein-2 and Beta-Tricalcium Phosphate.

INTRODUCTION: Anterior cervical discectomy and fusion (ACDF) is one of the most common spinal procedures performed. A direct comparison of the fusion and complication rates between recombinant human bone morphogenetic protein-2 (rhBMP2) and beta-tricalcium phosphate (bTCP) has not been reported. METHODS: A retrospective study of 191 consecutive patients who underwent ACDF with polyetheretherketone plastic fusion spacers during a 2-year period with either rhBMP2 (n = 84, 46%) or bTCP (n = 107, 56%) was performed. Patients underwent 1- (35%), 2- (41%), 3- (20%), and 4- (4%) level operations. The primary outcome measure was mature arthrodesis, with secondary measures including clinical outcomes and complication occurrence. Fusion was graded on plain lateral radiographs, with median length of follow-up of 12 months. RESULTS: Rates of cervical fusion were significantly greater for patients treated with rhBMP2 than bTCP at both 6 months (70% vs. 26%, P = 0.000) and 12 months (99% vs. 85%, P = 0.000). Postoperative dysphagia was reported in 35 patients (18%), with no difference in dysphagia incidence between rhBMP2 and bTCP (20% vs. 17%, P = 0.5); however, dysphagia was more severe in the rhBMP2 group, with greater rates of readmission and steroid use (both P < 0.05). A multivariable sensitivity analyses to control for patient characteristics and number of spinal fusion levels showed no differences in dysphagia rate between rhBMP2 and bTCP. CONCLUSIONS: In our cohort, the rate of mature arthrodesis after ACDF was greater with rhBMP2 compared with bTCP with no increased incidence of postoperative dysphagia; however, dysphagia severity was greater in the rhBMP2 cohort.

Neurological outcomes following iatrogenic vascular injury during posterior atlanto-axial instrumentation.

BACKGROUND: Iatrogenic vascular injury is a feared complication of posterior atlanto-axial instrumentation. A better understanding of clinical outcome and management options following this injury will allow surgeons to better care for these patients. The object of the study was to systematically review the neurologic outcomes after iatrogenic vascular injury during atlanto-axial posterior instrumentation. METHODS: We performed a systematic review of the Medline database following PRISMA guidelines. In our analysis, we included any retrospective cohort studies, prospective cohort studies, case reports, cases series, or systematic reviews with patients who had undergone posterior atlanto-axial fusion via screw rod constructs (SRC) or transarticular screws (TAS) that reported a patient with an injury to an arterial vessel directly attributable to the surgical procedure. RESULTS: Sixty cases of vascular injury were reported in 2078 (2.9%) patients over 27 publications. The average age for this patient population was 55.7+/-17.9. Vascular injury following posterior C1/2 instrumentation resulted in ipsilateral stroke in 10.0% (n=6/60) and non-persistent neurologic deficit in 6.7% (n=4/60) of cases with the deficit being permanent (not including death) in 1.7% (n=1/60) of cases. Four patients (6.7%) died. Arteriovenous fistula or pseudoaneurysm occurred in 8.3% (n=5/60) and 3.3% (n=2/60) of cases, respectively. Eight patients (13.3%) underwent endovascular repair of the injury with no permanent deficit. CONCLUSION: Neurological morbidity after iatrogenic vascular injury during posterior C1/2 fixation is higher than previously reported in literature. Some patients may benefit from endovascular treatment. Surgeons should be aware of normal and anomalous vertebral artery anatomy to avoid this potentially catastrophic complication.

Esophageal perforation after anterior cervical spine surgery: a systematic review of the literature.

OBJECTIVE Esophageal perforation is a rare but well-known complication of anterior cervical spine surgery. The authors performed a systematic review of the literature to evaluate symptomatology, direct causes, repair methods, and associated complications of esophageal injury. METHODS A PubMed search that adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines included relevant clinical studies and case reports (articles written in the English language that included humans as subjects) that reported patients who underwent anterior spinal surgery and sustained some form of esophageal perforation. Available data on clinical presentation, the surgical procedure performed, outcome measures, and other individual variables were abstracted from 1980 through 2015. RESULTS The PubMed search yielded 65 articles with 153 patients (mean age 44.7 years; range 14-85 years) who underwent anterior spinal surgery and sustained esophageal perforation, either during surgery or in a delayed fashion. The most common indications for initial anterior cervical spine surgery in these cases were vertebral fracture/dislocation (n = 77), spondylotic myelopathy (n = 15), and nucleus pulposus herniation (n = 10). The most commonly involved spinal levels were C5-6 (n = 51) and C6-7 (n = 39). The most common presenting symptoms included dysphagia (n =63), fever (n = 24), neck swelling (n = 23), and wound leakage (n = 18). The etiology of esophageal perforation included hardware failure (n = 31), hardware erosion (n = 23), and intraoperative injury (n = 14). The imaging modalities used to identify the esophageal perforations included modified contrast dye swallow studies, CT, endoscopy, plain radiography, and MRI. Esophageal repair was most commonly achieved using a modified muscle flap, as well as with primary closure. Outcomes measured in the literature were often defined by the time to oral intake following esophageal repair. Complications included pneumonia (n = 6), mediastinitis (n = 4), osteomyelitis (n = 3), sepsis (n = 3), acute respiratory distress syndrome (n = 2), and recurrent laryngeal nerve damage (n = 1). The mortality rate of esophageal perforation in the analysis was 3.92% (6 of 153 reported patients). CONCLUSIONS Esophageal perforation after anterior cervical spine surgery is a rare complication. This systematic review demonstrates that these perforations can be stratified into 3 categories based on the timing of symptomatic onset: intraoperative, early postoperative (within 30 days of anterior spinal surgery), and delayed. The most common source of esophageal injury is hardware erosion or migration, each of which may vary in their time to symptomatic manifestation.

Treatment of cervical stenotic myelopathy: a cost and outcome comparison of laminoplasty versus laminectomy and lateral mass fusion.

OBJECT: Cervical stenotic myelopathy due to spondylosis or ossification of the posterior longitudinal ligament is often treated with laminoplasty or cervical laminectomy (with fusion). The goal of this study was to compare outcomes, radiographic results, complications, and implant costs associated with these 2 treatments. METHODS: The authors analyzed the records of 56 patients (age range 42­81 years) who were surgically treated for cervical stenosis. Of this group, 30 underwent laminoplasty and 26 underwent laminectomy with fusion. Patients who had cervical kyphosis or spondylolisthesis were excluded. An average of 4 levels were instrumented in the laminoplasty group and 5 levels in the fusion group (p < 0.01). Forty-two percent of the fusions crossed the cervicothoracic junction, but no laminoplasty instrumentation crossed the cervicothoracic junction, and it only reached C-7 in one-third of the cases. Preoperative and postoperative Nurick grades and modified Japanese Orthopaedic Association (mJOA) scores were obtained. Outcomes were also assessed with neck pain visual analog scale (VAS) scores and the Odom outcome criteria. Postoperative length of stay, complications, and implant costs were calculated. RESULTS: The mean duration of follow-up, average patient age, and length of hospital stay were similar for both groups. The mean Nurick scores were also similar in the 2 groups and improved an average of 1.4 points in both (p < 0.01 for preoperative-postoperative comparison in each group). The mean mJOA scores improved 2.7 points in laminoplasty patients and 2.8 points in fusion patients (p < 0.01 for each group). The mean VAS scores for neck pain did not change significantly in the laminoplasty cohort (3.2 ± 2.8 [SD] preoperatively vs 3.4 ± 2.6 postoperatively, p = 0.50). In the fusion cohort, the mean VAS scores improved from 5.8 ± 3.2 to 3.0 ± 2.3 (p < 0.01). Excellent or good Odom outcomes were observed in 76.7% of the patients in the laminoplasty cohort and 80.8% of those in the fusion cohort (p = 0.71). In the fusion group, complications were twice as common and implant costs were nearly 3 times as high as in the laminoplasty group. When cases involving fusions crossing the cervicothoracic junction were excluded, analysis showed similar complication rates in the 2 groups. CONCLUSIONS: Patients treated with laminoplasty and patients treated with laminectomy and fusion had similar improvements in Nurick scores, mJOA scores, and Odom outcomes. Patients who underwent fusion typically had higher preoperative neck pain scores, but their neck pain improved significantly after surgery. There was no significant change in the neck pain scores of patients treated with laminoplasty. Our series suggests cervical fusion significantly reduces neck pain in patients with stenotic myelopathy, but that the cost of the implant and rate of reoperation are greater than in laminoplasty.

Circumferential fusion for cervical kyphotic deformity.

OBJECT: The treatment of cervical kyphotic deformity is challenging. Few prior reports have examined combined anterior/posterior correction methods, and fusion rates and standardized outcomes are rarely cited in literature examining these techniques. The authors present their midterm results with cervical kyphosis correction. METHODS: The authors retrospectively reviewed the charts of 30 patients with cervical kyphotic deformity who underwent circumferential spine surgery between 2001 and 2007. The causes of the deformity included chronic fracture in 17 patients, degenerative disease in 10, and tumor in 3. Anterior procedures included discectomies and corpectomies/osteotomies at 1 or more levels with fusion. Posterior operations included decompression and/or osteotomies with lateral mass or pedicle fixation. Preoperative and postoperative Ishihara kyphosis indices, modified Japanese Orthopaedic Association (mJOA) scores, and Nurick grades were analyzed. Arthrodesis was assessed via dynamic radiographs, and CT scans were used to assess fusion in questionable cases. RESULTS: One patient was lost to follow-up. Two patients died within 1 month of surgery. The follow-up period in the remaining 27 patients ranged from 1 to 6.4 years (mean 2.6 years). Ishihara indices improved from a preoperative mean of -17.7 to a postoperative mean of +11.4. The mean Nurick grades improved from 3.2 preoperatively to 1.3 postoperatively. The mJOA scores improved from a preoperative mean of 10 to 15 postoperatively. All surviving patients who underwent follow-up showed postoperative fusion except 1 patient with renal failure and osteoporosis (95% fusion rate). The overall rate of complications (major and minor) was 33.3%. CONCLUSIONS: In cases of cervical kyphosis, management with decompression, osteotomy, and stabilization from both anterior and posterior approaches can restore cervical lordosis. Furthermore, such surgical techniques can produce measurable improvements in neurological function (as measured with Nurick grades and mJOA scores) and achieve high fusion rates. However, there is a significant rate of complications.

The safety and efficacy of anterior cervical discectomy and fusion with polyetheretherketone spacer and recombinant human bone morphogenetic protein-2: a review of 200 patients.

OBJECT: The goal in this study was to demonstrate the safety and efficacy of anterior cervical discectomy and fusion ([ACDF]; single- or multilevel procedure) performed using titanium plates and polyetheretherketone (PEEK) spacers filled with recombinant human bone morphogenetic protein-2 (rhBMP-2) impregnated in a type I collagen sponge to achieve fusion. METHODS: The authors retrospectively reviewed 200 patients who underwent a single- or multilevel ACDF with titanium plate fixation and PEEK spacer filled with a collagen sponge impregnated with low-dose rhBMP-2. Clinical outcomes were assessed using pre- and postoperative Nurick grades and the Odom criteria. Radiographic outcomes were assessed using dynamic radiographs and computed tomography (CT) scans. RESULTS: The follow-up period ranged from 8 to 36 months (mean 16.7 months). A single-level ACDF was performed in 96 patients, 2-level ACDF in 62 patients, 3-level ACDF in 36 patients, and 4-level ACDF in 6 patients. Long-term follow-up was available for 193 patients. The Odom outcomes were rated as good to excellent in 165 patients (85%), fair in 24 (12.4%), and poor in 4 (2%). Among patients with myelopathy, Nurick grades improved from a preoperative mean of 1.42 to a postoperative mean of 0.26. All patients (100%) achieved solid radiographic fusion on dynamic radiographs and CT scans. Fourteen patients (7%) in this series experienced clinically significant dysphagia, and 4 (2%) required repeated operation for hematoma or seroma. CONCLUSIONS: An ACDF performed using a PEEK spacer filled with rhBMP-2 leads to good to excellent clinical outcomes and solid fusion, even in multilevel cases and in patients who are smokers. The incidence of symptomatic dysphagia may be decreased with a lower dose of rhBMP-2 that is placed only within the PEEK spacer.

Flexible rods and the case for dynamic stabilization.

The widespread use of instrumentation in the lumbar spine has led to high rates of fusion. This has been accompanied by a marked rise in adjacent-segment disease, which is considered to be an increasingly common and significant consequence of lumbar or lumbosacral fusion. Numerous biomechanical studies have demonstrated that segments fused with rigid metallic fixation lead to significant amounts of supraphysiological stress on adjacent discs and facets. The resultant disc degeneration and/or stenosis may require further surgical intervention and extension of the fusion to address symptomatic adjacent-segment disease. Recently, dynamic stabilization implants and disc arthroplasty have been introduced as an alternative to rigid fixation. The scope of spinal disease that can be treated with this novel technology, however, remains limited, and these treatments may not apply to patients who still require rigid stabilization and arthrodesis. In the spectrum between rigid metallic fixation and motion-preserving arthroplasty is a semirigid type of stabilization in which a construct is used that more closely mirrors the modulus of elasticity of natural bone. After either interbody or posterolateral arthrodesis is achieved, the fused segments will not generate the same adjacent-level forces believed to be the cause of adjacent-segment disease. Although this form of arthrodesis does not completely prevent adjacent-segment disease, the dynamic component of this stabilization technique may minimize its occurrence. The authors report their initial experience with the use of posterior dynamic stabilization in which polyetheretherketone rods were used for a posterior construct. The biomechanics of dynamic stabilization are discussed, clinical indications are reviewed, and case studies for its application are presented.

Combined ventral and dorsal surgery for myelopathy and myeloradiculopathy.

The past decade has witnessed significant advances in the surgical treatment of cervical myelopathy and myeloradiculopathy. In this article, we discuss in detail the indications for combined ventral and dorsal surgery for the treatment of cervical myelopathy and myeloradiculopathy. In addition, the advances in surgical technique and instrumentation in cervical spine surgery are explored. Finally, complication avoidance and management strategies are discussed. Combined ventral and dorsal decompression, reconstruction, and instrumentation procedures are viable options in the treatment of a select group of patients with complex cervical myelopathy or myeloradiculopathy.

Infected vertebroplasty requiring 360 degrees spinal reconstruction: long-term follow-up review. Report of two cases.

Transpedicular vertebroplasty has been established as a safe and effective treatment of thoracic and lumbar compression fractures. Complications are rare, and infectious complications requiring surgical management have only been reported once in the literature. The authors present two cases of infectious complications requiring surgical management. They emphasize that systemic infection is a contraindication to the performance of vertebroplasty. The serious nature of these infections, their surgical management, and strategies for avoiding them are discussed.

Recurrent subdural hematoma caused by cerebrospinal fluid leakage. Case report.

The authors report a case of a recurrent subdural hematoma (SDH) that was caused by a persistent cerebrospinal fluid (CSF) leak from an L1-2 fistula. A 34-year-old man experienced severe headaches due to SDH, and he underwent aspiration of subdural fluid four times due to recurrent collections. Further evaluation with computerized tomography (CT) myelography demonstrated extradural extravasation of contrast through an L1-2 fistula. The patient underwent an L1-2 laminectomy; a small dural defect with CSF leakage at the left nerve root sleeve was found and was repaired. Following the repair, the patient had no further recurrence of SDH. Recurrent SDH, caused by spontaneous CSF leakage through a lumbar CSF fistula, is extremely rare. In cases of recurrent SDH, radiographic workup with spinal CT myelography should be considered.