Kern H. Guppy

Contemporary management of subarachnoid hemorrhage and vasospasm: the UIC experience

BACKGROUND Cerebral vasospasm is a well-known and serious complication of aneurysmal subarachnoid hemorrhage. The means of monitoring and treatment of vasospasm have been widely studied. Each neurosurgical center develops a protocol based on their experience, availability of equipment and personnel, and cost, so as to keep morbidity and mortality rates as low as possible for their patients with vasospasm. METHODS At the University of Illinois at Chicago, we have developed algorithms for the diagnosis and management of cerebral vasospasm based on the experience of the senior authors over the past 25 years. This paper describes in detail our approach to diagnosis and treatment of aneurysmal subarachnoid hemorrhage and vasospasm. Our discussion is highlighted with data from a retrospective analysis of 324 aneurysm patients. RESULTS Over 3 years, 324 aneurysms were treated; 185 (57%) were clipped, 139 (43%) were coiled. The rate of vasospasm for the 324 patients was 27%. The rate of hydrocephalus was 32% for those patients who underwent clipping, and 29% for those coiled. The immediate outcomes for those who underwent clipping was excellent in 35%, good in 38%, poor in 15.5%, vegetative in 3%, and death in 8% of the patients. For those who underwent coiling the immediate outcome was excellent in 64%, good in 14.5%, vegetative in 2.5%, and death in 14.5% of the patients. These statistics include all Hunt and Hess grades. For those patients who underwent clipping, 51% were intact at 6 months follow-up, 15% had a permanent deficit, 10% had a focal cranial nerve deficit, and 2% had died from complications not directly related to the procedure. For those patients who had undergone coiling, 75% were intact at 6 months follow-up, 12.5% had a permanent deficit, and 12.5% had a cranial nerve deficit, with no deaths. CONCLUSIONS The morbidity and mortality of cerebral vasospasm is significant. A good outcome after aneurysmal subarachnoid hemorrhage is dependent upon careful patient management in the preoperative, perioperative, and postoperative periods. The timely work-up and aggressive treatment of neurological deterioration, whether or not it is because of vasospasm, is paramount.

Hemodynamic Evaluation of Basilar and Vertebral Artery Angioplasty

OBJECTIVE The postangioplasty evaluation of a stenotic vessel is often conducted by studying serial angiograms to determine the anatomic reduction in stenosis. In flow-limiting stenosis, the hemodynamic change that accompanies these anatomic changes is of great importance in evaluating the success of the angioplasty. The purpose of this article is to demonstrate the usefulness of phase contrast magnetic resonance angiography (PCMRA) in evaluating the hemodynamic changes that occur after angioplasty of the basilar and vertebral arteries. METHODS Between January 1998 and February 2000, PCMRA was performed for the hemodynamic evaluation of 130 patients who presented at our institution. Twenty-six patients were evaluated for vertebrobasilar insufficiency, and flow rates of their vertebral and basilar arteries were determined. In five patients, angioplasty was done on the basilar or vertebral arteries, and PCMRA was performed to determine flow rates before and after the procedure. RESULTS Of the five patients undergoing angioplasty, the average percentage of stenosis was 81%. The average increase in basilar artery flow rate was 46 ml/min (P < 0.05) after angioplasty. Two of these patients are described. One patient demonstrated the comparison of flow rates in the vertebral and basilar arteries after angioplasty. The second patient showed follow-up flow rates measured after angioplasty and up to 4 months later to predict restenosis. CONCLUSION The use of flow rate data before and after angioplasty is helpful not only to evaluate the treatment immediately after the procedure but also to evaluate the effectiveness of the treatment during a long period. PCMRA provides a noninvasive method for measuring arterial flow rates with far-reaching implications in neurosurgery.

Risk Factors Associated With 30-day Readmissions After Instrumented Spine Surgery in 14,939 Patients: 30-day readmissions after instrumented spine surgery

Study Design. A retrospective review of instrumented spine registry from an integrated US healthcare system. Objective. Investigate the 30-day readmission rate and risk factors after instrumented spine surgery. Summary of Background Data. Published readmission rates range from 2% to over 20%. We were interested in learning which patients were at greatest risk, when did readmissions occur, and why. Method. 30-day readmission rates were determined for 14,939 patients after an index spine procedure between 1/2009 and 3/2013. Data were analyzed with descriptive statistics, univariate, and multivariate logistic regression analysis. Result. The average age of the cohort was 59 (SD = 13.4) and 52% were female. The 30-day readmission rate was 5.5% (821/14,939). The temporal pattern for readmission was: 17% (140) at week 1, 48% (394) at week 2, 72% (591) at week 3, and 100% (821) at week 4. The leading causes were wound complications (infection, hematoma, dehiscence, seroma), sepsis, pain management, pneumonia, and pulmonary emboli/deep venous thrombosis. In a multivariate model, readmission risk factors were: malignancy (OR 2.99, 95% CI: 1.56, 5.73), operative time more than 400 minutes (OR 2.59, 95% CI: 1.66, 4.02), operative time 300–399 minutes (OR 2.33, 95% CI: 1.54–3.52), hospital stay 6–10 days (OR 2.03, 95% CI: 1.31–3.14), hospital stay more than 10 days (OR 1.85, 95% CI: 1.1, −3.08), surgical complications (OR 1.67, 95% CI: 1.18, 2.36), operative time 200–299 (OR 1.52, 95% CI: 1.04, 2.22), depression (OR 1.48, 95% CI: 1.14, 1.93), rheumatoid arthritis (OR 1.45, 95% CI: 1.05, 2.01), deficiency anemia (OR 1.30, 95% CI: 1.05, 1.61), and hypothyroidism (OR 1.29, 95% CI: 1.01, 1.64). Conclusion. Surgical complications (dural tear, deep infections, superficial infections, epidural hematoma), malignancy, lengthy operative times, and lengthy initial hospitalizations are all risk factors for 30-day readmission. These findings should be considered during preoperative assessment and surgical planning. Level of Evidence: 3

Perioperative management of neurosurgical patients with methicillin-resistant Staphylococcus aureus

OBJECT The emergence of methicillin-resistant Staphylococcus aureus (MRSA) has posed a challenge in the treatment of neurosurgical patients. The authors investigated the impact of MRSA colonization and infection in the neurosurgical population at a community-based, tertiary care referral center. METHODS Hospitalized patients under the care of the Kaiser Permanente inpatient neurosurgery service were prospectively entered into a database. In Phase I of the study, 492 consecutive patients were followed. Per hospital policy, the 260 patients from this group who were admitted to the intensive care unit (ICU) underwent screening for MRSA based on nasal swab cultures and a review of their medical history for prior MRSA infections. These patients were designated as either MRSA positive (17 patients, 6.5% of screened patients) or MRSA negative (243 patients). The 232 patients admitted to non-ICU nursing units did not undergo MRSA screening and were designated as unscreened. In Phase II of the study, the authors reviewed 1005 neurosurgical admissions and completed a detailed chart review in 62 MRSA-positive patients (6.2%). Eleven patients received nonoperative treatment. Five patients presented with community-acquired neurosurgical infections, and the causative organism was MRSA in 3 cases. Forty-six patients underwent 55 procedures, and the authors reviewed their perioperative management. RESULTS In Phase I of the study, the authors found that for the MRSA-positive, MRSA-negative, and unscreened groups, the rates of postoperative neurosurgical wound infections caused by all pathogens were 23.5, 4.1, and 1.3%, respectively. For MRSA wound infections, the rates were 23.5, 0.8, and 0%, respectively. In Phase II, patients with MRSA were noted to have the following clinical features: male sex in 63%, a malignancy in 39.1%, diabetes in 34.8%, prior MRSA infection in 21.7%, immunosuppressed state in 17.4%, and a traumatic injury in 15.2%. The rate of postoperative neurosurgical wound infection in patients who received MRSA-specific prophylactic antibiotic therapy (usually vancomycin) was 7.4% (27 procedures) compared with 32.1% (28 procedures) in patients who received the standard treatment (usually cefazolin) (p = 0.04). Wound care for ICU patients was standardized for postoperative Days 0-7 with chlorhexidine cleaning at bandage changes at 3-day intervals. Wound cultures from neurosurgical site infections in patients with prior MRSA colonization or infection grew MRSA in 7 of 11 patients. CONCLUSIONS Neurosurgical patients identified with MRSA colonization or a prior history of MRSA infections benefit from specific perioperative care, including prophylactic antibiotics active against MRSA (such as vancomycin) and postoperative wound care with coverings and chlorhexidine antisepsis to reduce MRSA wound colonization.

Hemodynamics of in-tandem stenosis of the internal carotid artery: when is carotid endarterectomy indicated?

BACKGROUND Recent publications have pointed out the importance of evaluating patients with in-tandem stenosis and in particular the association of moderate stenosis of the extracranial internal carotid artery (ICA) with moderate or severe stenosis of the intracranial internal carotid artery. Such evaluations are needed in symptomatic patients before planning carotid endarterectomies because observations have shown that in some cases the removal of an extracranial lesion does not necessarily improve these symptoms. This paper examines the hemodynamic effects of in-tandem stenosis in the internal carotid artery. METHODS Equations describing flow in arteries are modified to accommodate two regions of stenosis in tandem. An equivalent value of stenosis is derived such that two stenoses in tandem behave as a single stenosis with similar hemodynamic properties. The solution to this problem is solved mathematically and this was used to analyze the observations made in five studies published on in-tandem stenosis of the internal carotid artery. RESULTS Equivalent stenoses for various values of extracranial and intracranial stenoses are presented. It was found that two stenotic lesions in tandem are not equivalent to a simple summation of both values. A graphical solution is presented to show the hemodynamic effects of both stenoses. CONCLUSIONS The most critical determinant of hemodynamic compromise when two lesions are in tandem is the larger one. Hence removal of a more proximal lesion may have little effect on a larger distal lesion if the symptoms are due to hypoperfusion. It is important that one distinguish between hypoperfusion and thromboembolic causes of the symptoms. No conclusions about the risk of thromboembolic events after a carotid endarterectomy in the setting of a distal stenosis can be made from this study.

En bloc resection of a multilevel high-cervical chordoma involving C-2: new operative modalities: technical note.

En bloc resection of cervical chordomas has led to longer survival rates but has resulted in significant morbidities from the procedure, especially when the tumor is multilevel and located in the high-cervical (C1-3) region. To date, there have been only 5 reported cases of multilevel en bloc resection of chordomas in the high-cervical spine. In this technical report the authors describe a sixth case. A complete spondylectomy was performed at C-2 and C-3 with spinal reconstruction and stabilization, using several new modalities that were not used in the previous cases. The use of 1) preoperative endovascular sacrificing of the vertebral artery, 2) CT image-guidance, 3) an ultrasonic aspirator for skeletonizing the vertebral artery, and 4) the custom design of an anterior cage all contributed to absence of intraoperative or long-term (20 months) hardware failure and pseudarthrosis.

The use of intraoperative navigation for complex upper cervical spine surgery

Imaging guidance using intraoperative CT (O-arm surgical imaging system) combined with a navigation system has been shown to increase accuracy in the placement of spinal instrumentation. The authors describe 4 complex upper cervical spine cases in which the O-arm combined with the StealthStation surgical navigation system was used to accurately place occipital screws, C-1 screws anteriorly and posteriorly, C-2 lateral mass screws, and pedicle screws in C-6. This combination was also used to navigate through complex bony anatomy altered by tumor growth and bony overgrowth. The 4 cases presented are: 1) a developmental deformity case in which the C-1 lateral mass was in the center of the cervical canal causing cord compression; 2) a case of odontoid compression of the spinal cord requiring an odontoidectomy in a patient with cerebral palsy; 3) a case of an en bloc resection of a C2-3 chordoma with instrumentation from the occiput to C-6 and placement of C-1 lateral mass screws anteriorly and posteriorly; and 4) a case of repeat surgery for a non-union at C1-2 with distortion of the anatomy and overgrowth of the bony structure at C-2.

Management and outcomes of malignant posterior reversible encephalopathy syndrome

INTRODUCTION Recognition of severe forms of posterior reversible encephalopathy syndrome (PRES) has improved. Management of these patients remains challenging, particularly in patients with the combination of edema and hemorrhage. METHODS A prospective inpatient neuro-intensive care database was queried for patients with PRES. Malignant PRES was diagnosed by clinical assessments (GCS less than 8 and clinical decline despite standard medical management for elevated intracranial pressure) and radiographic criteria (edema with associated mass effect; brain hemorrhage exerting mass effect; effacement of basal cisterns, transtentorial, tonsillar, or uncal herniation). Malignant PRES was defined as: radiology studies consistent with PRES; GCS less than 8; and clinical decline despite standard elevated intracranial pressure management. RESULTS Five cases were identified over a 4 year interval. The following contributing conditions were also present: chemotherapy (1), systemic lupus erythematosis (2), pregnancy (1), and methamphetamines (1). Neurocritical care interventions included: hyperosmolar therapy (5), anticonvulsants (5), management of coagulopathy (5), and ventilatory support (5). Neurosurgical interventions included: craniectomy (5), hematoma evacuation (3), and external ventricular drain (4). Brain biopsy was performed in 5 patients and was negative for vasculitis, demyelinating disease, tumor, or infection. Cyclophosphamide was administered to the two patients with SLE. With long-term follow up, all patients achieved good functional outcomes (modified Rankin score 1-2). CONCLUSION In contrast to historical reports of high mortality rates (16-29%) for severe and hemorrhagic PRES variants, we had no fatalities and observed favorable functional outcomes with intracranial pressure monitoring and craniectomy for malignant PRES cases who fail medical ICP management.

The use of flexion-extension magnetic resonance imaging for evaluating signal intensity changes of the cervical spinal cord

The authors present 2 cases involving patients who presented with myelopathy. Magnetic resonance imaging of the cervical spine showed spinal cord signal changes on T2-weighted images without any spinal cord compression. Flexion-extension plain radiographs of the spine showed no instability. Dynamic MR imaging of the cervical spine, however, showed spinal cord compression on extension. Compression of the spinal cord was caused by dynamic anulus bulging and ligamentum flavum buckling. This report emphasizes the need for dynamic MR imaging of the cervical spine for evaluating spinal cord changes on neutral position MR imaging before further workup for other causes such as demyelinating disease.

Reoperation rates for symptomatic nonunions in posterior cervicothoracic fusions with and without bone morphogenetic protein in a cohort of 450 patients

OBJECTIVE Fusions across the cervicothoracic junction have been challenging because of the large biomechanical forces exerted resulting in frequent reoperations for nonunions. The objective of this study was to investigate a retrospective cohort using chart review of posterior cervicothoracic spine fusions with and without bone morphogenetic protein (BMP) and to determine the reoperation rates for symptomatic nonunions in both groups. METHODS Between January 2009 and September 2013, posterior cervicothoracic spine fusion cases were identified from a large spine registry (Kaiser Permanente). Demographics, diagnoses, operative times, lengths of stay, and reoperations were extracted from the registry. Reoperations for symptomatic nonunions were adjudicated via chart review. Logistic regression was used to estimate odds ratios and 95% confidence intervals. Kaplan-Meier curves for the non-BMP and BMP groups were generated and compared using the log-rank test. RESULTS In this cohort there were 450 patients (32.7% with BMP) with a median follow-up of 1.4 years (interquartile range [IQR] 0.5-2.7 years). Kaplan-Meier curves showed no significant difference in reoperation rates for nonunions using the log-rank test (p = 0.088). In a subset of patients with more than 1 year of follow-up, 260 patients were identified (43.1% with BMP) with a median follow-up duration of 2.4 years (IQR 1.6-3.3 years). There was no statistically significant difference in the symptomatic operative nonunion rates for posterior cervicothoracic fusions with and without BMP (0.0% vs 2.7%, respectively; p = 0.137) for more than 1 year of follow-up. CONCLUSIONS This study presents the largest series of patients using BMP in posterior cervicothoracic spine fusions. Reoperation rates for symptomatic nonunions with more than 1 year of follow-up were 0% with BMP and 2.7% without BMP. No statistically significant difference in the reoperation rates for symptomatic nonunions with or without BMP was found.