Gregory P. Lekovic

Advances in the treatment and outcome of brainstem cavernous malformation surgery: a single-center case series of 300 surgically treated patients.

BACKGROUND: Brainstem cavernous malformations (BSCMs) are relatively uncommon, low-flow vascular lesions. Because of their relative rarity, relatively little data on their natural history and on the efficacy and durability of their treatment. OBJECTIVE: To evaluate the long-term durability of surgical treatment of BSCMs and to document patient outcomes and clinical complications. METHODS: The charts of all patients undergoing surgical treatment of BSCM between 1985 and 2009 were reviewed retrospectively. The study population consisted of 300 patients who had surgery for BSCM. Forty patients were under 19 years of age at surgery; pediatric BSCMs have been reported separately. Patient demographics, lesion characteristics, surgical approaches, and patient outcomes were examined. RESULTS: The study population consisted of 260 adult patients with a female-to-male ratio of 1.5 and mean age of 41.8 years. Of the 260 patients, 252 presented with a clinical or radiographic history of hemorrhage. The mean follow-up in 240 patients was 51 months. The mean Glasgow Outcome Scale on admission, at discharge, and at last follow-up was 4.4, 4.2, and 4.6. Postoperatively, 137 patients (53%) developed new or worsening neurological symptoms. Permanent new deficits remained in 93 patients 3(36%). There were perioperative complications in 74 patients (28%); tracheostomy, feeding tube placement, and cerebrospinal fluid leakage were most common. Eighteen patients (6.9%) experienced 20 rehemorrhages. Twelve patients required reoperation for residual/recurrent BSCM. The overall annual risk of postoperative rehemorrhage was 2%/patient. CONCLUSION: Although BSCM surgery has significant associated risks, including perioperative complications, new neurological deficits, and death, most patients have favorable outcomes. Overall, surgery markedly improved the risk of rehemorrhage and related symptoms and should be considered in patients with accessible lesions.

TREATMENT OPTIONS FOR THIRD VENTRICULAR COLLOID CYSTS: COMPARISON OF OPEN MICROSURGICAL VERSUS ENDOSCOPIC RESECTION

OBJECTIVEWe retrospectively reviewed our experience treating third ventricular colloid cysts to compare the efficacy of endoscopic and transcallosal approaches. METHODSBetween September 1994 and March 2004, 55 patients underwent third ventricular colloid cyst resection. The transcallosal approach was used in 27 patients; the endoscopic approach was used in 28 patients. Age, sex, cyst diameter, and presence of hydrocephalus were similar between the two groups. RESULTSThe operating time and hospital stay were significantly longer in the transcallosal craniotomy group compared with the endoscopic group. Both approaches led to reoperations in three patients. The endoscopic group had two subsequent craniotomies for residual cysts and one repeat endoscopic procedure because of equipment malfunction. The transcallosal craniotomy group had two reoperations for fractured drainage catheters and one operation for epidural hematoma evacuation. The transcallosal craniotomy group had a higher rate of patients requiring a ventriculoperitoneal shunt (five versus two) and a higher infection rate (five versus none). Intermediate follow-up demonstrated more small residual cysts in the endoscopic group than in the transcallosal craniotomy group (seven versus one). Overall neurological outcomes, however, were similar in the two groups. CONCLUSIONCompared with transcallosal craniotomy, neuroendoscopy is a safe and effective approach for removal of colloid cysts in the third ventricle. The endoscope can be considered a first-line treatment for these lesions, with the understanding that a small number of these patients may need an open craniotomy to remove residual cysts.

A comparison of two techniques in image-guided thoracic pedicle screw placement: a retrospective study of 37 patients and 277 pedicle screws

OBJECT The goal of this study was to compare the accuracy of thoracic pedicle screw placement aided by two different image-guidance modalities. METHODS The charts of 40 consecutive patients who had undergone stabilization of the thoracic spine between January 2003 and January 2005 were retrospectively reviewed. Three patients were excluded from the study because, on the basis of preoperative findings, small pedicle diameter precluded the use of pedicle screws. Thus, a total of 37 patients had 277 screws placed with the aid of either virtual fluoroscopy or isocentric C-arm 3D navigation. The indications for surgery included trauma, degenerative disease, and tumor, and were similar in both groups. All 37 patients underwent postoperative computed tomography scanning, and an independent reviewer graded all screws based on axial, sagittal, and coronal projections for a full determination of the placement of the screw in the pedicle. RESULTS The rate of unintended perforations was found to depend on pedicle diameter (p < 0.0001). There were no statistical differences between groups with regard to rate or grade of cortical perforations. Overall, the rate and grade of perforations was low, and there were no neurological or vascular complications. CONCLUSIONS The authors have shown that either image-guidance system may be used with a high degree of accuracy and safety. Because both systems were found to be comparably safe and accurate, the choice of image-guidance modality may be determined by the level of surgeon comfort and/or availability of the system.

Supracerebellar infratentorial approach to cavernous malformations of the brainstem: surgical variants and clinical experience with 45 patients.

OBJECTIVEThe supracerebellar infratentorial (SCIT) approach can be performed at the midline (median variant), lateral to the midline (paramedian variant), or at the level of the angle formed by the transverse and sigmoid sinuses (extreme lateral variant). We analyzed our experience with SCIT approaches for the surgical treatment of cavernous malformations of the brainstem (CMBs). METHODSDemographic, clinical, radiologic, and surgical data from 45 patients (20 males and 25 females; mean age, 36.2 years) with CMBs surgically removed through SCIT approaches were reviewed retrospectively. Anatomic information was explored using cadaver head dissection. RESULTSTwenty-three lesions were in the midbrain, 3 were at the midbrain and extended to the thalamus, 9 were at the pontomesencephalic junction, and 10 were in the upper pons. All patients presented with hemorrhage. The median variant was used in 13 patients, the paramedian variant in 9, and the extreme lateral variant in 23. Intraoperatively, all CMBs were associated with a developmental venous anomaly. At last follow-up, 88% of the patients were the same or better. After a mean follow-up of 20 months, their mean Glasgow Outcome Scale score was 4.1. CONCLUSIONSCIT approaches provide excellent exposure to CMBs located at the posterior incisural space, not only in the midline but also in the posterolateral surface of the upper pons and midbrain. Careful preoperative planning and neuronavigational assistance are needed to determine the best angle of attack and trajectory for SCIT approaches. Refined microsurgical techniques are paramount to achieve safe surgical removal of CMBs with good outcomes.

Anatomical and Biomechanical Analyses of the Unique and Consistent Locations of Sacral Insufficiency Fractures

Study Design. Correlation of locations of sacral insufficiency fractures is made to regions of stress depicted by finite element analysis derived from biomechanical models of patient activities. Objective. Sacral insufficiency fractures occur at consistent locations. It was postulated that sacral anatomy and sites of stress within the sacrum with routine activities in the setting of osteoporosis are foundations for determining patterns for the majority of sacral insufficiency fractures. Summary of Background Data. The predominant vertical components of sacral insufficiency fractures most frequently occur bilaterally through the alar regions of the sacrum, which are the thickest and most robust appearing portions of the sacrum instead of subjacent to the central sacrum, which bears the downward force of the spine. Methods. First, the exact locations of 108 cases of sacral insufficiency fractures were catalogued and compared to sacral anatomy. Second, different routine activities were simulated by pelvic models from CT scans of the pelvis and finite element analysis. Analyses were done to correlate sites of stress with activities within the sacrum and pelvis compared to patterns of sacral insufficiency fractures from 108 cases. Results. The sites of stress depicted by the finite element analysis walking model strongly correlated with identical locations for most sacral and pelvic insufficiency fractures. Consistent patterns of sacral insufficiency fractures emerged from the 108 cases and a biomechanical classification system is introduced. Additionally, alteration of walking mechanics and asymmetric sacral stress may alter the pattern of sacral insufficiency fractures noted with hip pathology (P = 0.002). Conclusion. Locations of sacral insufficiency fractures are nearly congruous with stress depicted by walking biomechanical models. Knowledge of stress locations with activities, cortical bone transmission of stress, usual fracture patterns, intensity of sacral stress with different activities, and modifiers of walking mechanics may aid medical management, interventional, or surgical efforts.

Litigation of missed cervical spine injuries in patients presenting with blunt traumatic injury.

BACKGROUNDApproximately 800,000 cervical spines are cleared in emergency departments each year. Errors in diagnosis of cervical spine injury are a potentially huge medicolegal liability, but no established protocol for clearance of the cervical spine is known to reduce errors or delays in diagnosis. METHODSThe Lexis-Nexis, Westlaw, and Medline databases were queried for cases of missed cervical injury. Errors were categorized according to a novel system of classification. Type I errors occurred when inadequate or improper tests were ordered. Type II errors occurred when adequate tests were ordered, but were either misread or not read. Type III errors occurred when adequate tests were ordered and read accurately, but the ordered test was not sensitive enough to detect the injury. RESULTSTwenty cases of missed or delayed diagnosis of cervical spine injury were found in 10 jurisdictions. Awards averaged

Developmental venous anomaly, cavernous malformation, and capillary telangiectasia: spectrum of a single disease.

2.9 million (inflation adjusted to 2002 dollars). Eight cases resulted in verdicts in favor of the defendant, but none of these cases involved an alleged Type II error. CONCLUSIONFear of lawsuits encourages defensive medicine and complicates the process of clearing a patients cervical spine. This analysis adds medicolegal support for the judicious use of imaging studies in current cervical spine clearance protocols. However, exposure to significant liability suggests that a low threshold for computed tomography is a reasonable alternative.

Cavernous malformations of the brainstem presenting in childhood: surgical experience in 40 patients.

SummaryDevelopmental venous anomalies (DVAs), cavernous malformations, and capillary telangiectasias are related vascular malformations of the central nervous system. Mixed lesions of the central nervous system vasculature have been reported in a host of combinations, including many possible concomitant combinations of cavernous malformations, venous anomalies, capillary telangiectasias, and arteriovenous malformations (AVMs).We describe the natural history of disease in a female with developmental venous anomaly, cavernous malformation, and capillary telangiectasias appearing in sequence.

Endoscopic resection of hypothalamic hamartoma using a novel variable aspiration tissue resector.

BACKGROUND: Brainstem cavernous malformations (BSCMs) are believed to compose 9% to 35% of all cerebral cavernous malformations, but these lesions have been reported in children in very limited numbers. OBJECTIVE: To review surgical outcomes of pediatric patients with BSCMs treated at 1 institution. METHODS: We retrospectively analyzed the course of 40 pediatric patients (19 males, 21 females; age range, 10 months to 18.9 years; mean, 12.3 years) who underwent surgery between 1984 and 2009. Age, sex, presenting symptoms, location of lesion, surgical approach, new postoperative deficits, Glasgow Outcome Scale score, recurrences, and resolution of baseline symptoms were recorded. RESULTS: Thirty-nine patients experienced hemorrhage before surgery. Lesion locations included the pons (n = 22), midbrain (n = 4), midbrain and thalamus (n = 4), pontomesencephalic junction (n = 3), medulla (n = 3), pontomedullary junction (n = 3), and cervicomedullary junction (n = 1). Mean lesion size was 2.3 cm. Mean length of hospital stay was 10.7 days. The average clinical follow-up was 31.9 months in 36 patients with follow-up after discharge. At last follow-up, 5 patients had experienced symptoms and/or imaging consistent with rehemorrhage, either from a residual that enlarged or true recurrence (5.25% annual rebleed risk per patient after surgery); 2 required reoperation for further resection of cavernoma. Mean Glasgow Outcome Scale score was 4.2 on admission, 4.05 at discharge, and 4.5 at latest follow-up. Preoperative symptoms and deficits improved in 16 patients (40%). New neurological deficits developed in 19 patients (48%) and resolved in 9, leaving 10 patients (25%) with new permanent deficit. CONCLUSION: Compared with adults, pediatric patients with BCSMs tend to have larger lesions and higher rates of recurrence (regrowth of residual lesion). Given the greater life expectancy of children, surgical treatment seems warranted in those with surgically accessible lesions that have bled. Outcomes were similar to those in our adult series of patients with BSCMs.

Incidence and pattern of direct blunt neurovascular injury associated with trauma to the skull base.

OBJECTIVE: We present a novel variable aspiration tissue resector for use with neuroendoscopy. METHODS: Two patients, 4 and 14 years old, respectively, presented with intractable gelastic seizures refractory to maximal medical therapies. Magnetic resonance imaging showed mass lesions of the third ventricle consistent with hypothalamic hamartoma. RESULTS: The patients underwent magnetic resonance imaging wand-guided endoscopic resection of the tumor with the Suros novel variable aspiration tissue resector. There were no device-associated complications or adverse events. The hamartoma was disconnected in one patient, and gross total resection was achieved in the other. CONCLUSION: Endoscopy for tumor resection is still frustrated by the lack of surgical tools, such as ultrasonic aspirators, comparable with those available for use during open procedures. The variable-aspiration tissue resector reported here can be used to resect tumor tissue safely. These two cases demonstrate that gross total resection of small hypothalamic hamartomas is feasible with minimal morbidity through an endoscopic approach.