Brian M. Corliss and Brian L. Hoh
William C. Newman, Dan W. Neal and Brian L. Hoh
Comorbidities have an impact on risk stratification for outcomes in analyses of large patient databases. Although the Charlson Comorbidity Index (CCI) and the Elixhauser Comorbidity Index (ECI) are the most commonly used comorbidity indexes, these have not been validated for patients with unruptured cerebral aneurysms; therefore, the authors created a comorbidity index specific to these patients.
The authors extracted all records involving unruptured cerebral aneurysms treated with clipping, coiling, or both from the Nationwide Inpatient Sample (2002–2010). They assessed the effect of 37 variables on poor outcome and used the results to create a risk score for these patients. The authors used a validation data set and bootstrapping to evaluate the new index and compared it to CCI and ECI in prediction of poor outcome, mortality, length of stay, and hospital charges.
The index assigns integer values (−2 to 7) to 20 comorbidities: neurological disorder, renal insufficiency, gastrointestinal bleeding, paralysis, acute myocardial infarction, electrolyte disorder, weight loss, metastatic cancer, drug abuse, arrhythmia, coagulopathy, cerebrovascular accident, psychosis, alcoholism, perivascular disease, valvular disease, tobacco use, hypothyroidism, depression, and hypercholesterolemia. Values are summed to determine a patient's risk score. The new index was better at predicting poor outcome than CCI or ECI (area under the receiver operating characteristic curve [AUC] 0.814 [95% CI 0.798–0.830], vs 0.694 and 0.712, respectively, for the other indices), and it was also better at predicting mortality (AUC 0.775 [95% CI 0.754–0.792], vs 0.635 and 0.657, respectively, for CCI and ECI).
This new comorbidity index outperforms the CCI and ECI in predicting poor outcome, mortality, length of stay, and total charges for patients with unruptured cerebral aneurysm. Reevaluation of other patient cohorts is warranted to determine the impact of more accurate patient stratification.
R. Loch Macdonald
William C. Newman, Paul S. Kubilis and Brian L. Hoh
Comorbidities have a significant effect on patient outcomes. Accounting for this effect is especially important in retrospective reviews of large databases; overpowered studies are at risk for finding significant results because of inaccurate patient risk stratification. The authors previously created a neurovascular comorbidities index (NCI) for patients with an unruptured intracranial aneurysm and found that the model’s ability to predict patient outcomes was statistically significantly improved over that of the routinely used Charlson Comorbidity Index (CCI) and Elixhauser Comorbidity Index (ECI). In this study, the authors aimed to validate use of the NCI over that of the CCI and ECI for risk stratification of patients with other neurovascular diseases.
The authors queried the National (Nationwide) Inpatient Sample database for the years 2002–2012 to compare the accuracy of the previously validated NCI with that of the CCI and ECI with respect to predicting outcomes for patients who had an arteriovenous malformation, a ruptured intracranial aneurysm, carotid artery stenosis, or dural arteriovenous fistula and who underwent surgical intervention.
For patients with an arteriovenous malformation, the NCI outperformed the CCI and ECI in predicting poor outcome, hospital length of stay (LOS), and total cost but was equivalent to the CCI in predicting death. For patients with a ruptured intracranial aneurysm, the NCI outperformed the ECI and CCI in predicting death, poor outcome, LOS, and total cost. For patients with carotid artery stenosis, the NCI outperformed the ECI and CCI in predicting LOS, but it was equivalent to the ECI in predicting death and total cost and inferior to the CCI in predicting poor outcome (p < 0.002 for all). An insufficient number of patients with dural arteriovenous fistula who underwent surgical intervention were available for analysis (n < 10), and they therefore were excluded from study. For 11 of 12 metrics, the NCI was the significantly more efficient model.
The NCI outperforms the CCI and ECI by providing more appropriate and efficient risk stratification of patients regarding death, outcome, LOS, and cost. Given this finding, the NCI should be used for retrospective reviews of patient outcomes instead of the CCI or ECI.
Kyle M. Fargen, Dan Neal, Spiros L. Blackburn, Brian L. Hoh and Maryam Rahman
The Agency for Healthcare Research and Quality patient safety indicators (PSIs) and the Centers for Medicare and Medicaid Services hospital-acquired conditions (HACs) are publicly reported quality metrics linked directly to reimbursement. The occurrence of PSIs and HACs is associated with increased mortality and hospital costs after stroke. The relationship between insurance status and PSI and HAC rates in hospitalized patients treated for acute ischemic stroke was determined using the Nationwide Inpatient Sample (NIS) database.
The NIS was queried for all hospitalizations involving acute ischemic stroke between 2002 and 2011. The rate of each PSI and HAC was determined by searching the hospital records for ICD-9 codes. The SAS statistical software package was used to calculate rates and perform multivariable analyses to determine the effects of patient variables on the probability of developing each indicator.
The NIS query revealed 1,507,336 separate patient admissions that had information on both primary payer and hospital teaching status. There were 227,676 PSIs (15.1% of admissions) and 42,841 HACs reported (2.8%). Patient safety indicators occurred more frequently in Medicaid/self-pay/no-charge patients (19.1%) and Medicare patients (15.0%) than in those with private insurance (13.6%; p < 0.0001). In a multivariable analysis, Medicaid, self-pay, or nocharge patients had significantly longer hospital stays, higher mortality, and worse outcomes than those with private insurance (p < 0.0001).
Insurance status is an independent predictor of patient safety events after stroke. Private insurance is associated with lower mortality, shorter lengths of stay, and improved clinical outcomes.
Kyle M. Fargen and Brian L. Hoh
Maryam Rahman, Gregory J. Velat, Brian L. Hoh and J Mocco
Cerebral venous sinus thrombosis (CVST) is an increasingly diagnosed disease with a wide range of symptoms, ranging from a mild headache to cerebral herniation. A potentially devastating syndrome, CVST has been associated with a mortality rate of 6–10%. In prospective studies, the overall rate of death and dependency from CVST ranges from 8.8 to 44.4%. Systemic anticoagulation remains the first-line treatment. However, a percentage of patients deteriorate despite medical therapy. These cases have resulted in the development of thrombolysis or endovascular treatment for CVST. Initial reports of the use of endovascular treatment of CVST have been promising. However, enthusiasm for the use of endovascular thrombolysis and thrombectomy should be tempered by an understanding of possible risks such as intracerebral hemorrhage and/or vessel dissection. The authors review the literature regarding endovascular treatment of CVST with a description of the chemical and mechanical thrombolytic techniques.
Kyle M. Fargen and Brian L. Hoh
Brian M. Corliss, Adam J. Polifka, Neil S. Harris, Brian L. Hoh and W. Christopher Fox
Inhibition of platelet aggregation is vital to preventing thromboembolic complications related to stent placement in endovascular neurosurgery, but excessive inhibition potentiates hemorrhagic complications. Recent evidence suggests an ideal inhibition range of 70–150 P2Y12 response units (PRU) as measured on the VerifyNow assay, which relies on photometric measurements of platelet aggregation. Thromboelastography (TEG) with platelet mapping (PM) is an alternative assay that directly measures clot formation and mechanical strength. This study compares the results of PRU to TEG-PM.
Patients with simultaneous or near-simultaneous PRU and TEG-PM results who underwent cervical carotid artery stenting, intracranial stent-assisted aneurysm coiling, or flow diversion at the authors’ institution between August 2015 and November 2016 were identified. PRU results were compared with the TEG maximal amplitude (MA) attributable to adenosine diphosphate (ADP) activity (MA-ADP) as measured by TEG-PM. Platelet inhibition was considered therapeutic for MA-ADP values < 50 mm or PRU < 194. The Pearson correlation coefficient was calculated, and the sensitivity and specificity of PRU were calculated assuming that the results of TEG-PM reflected the true degree of platelet inhibition.
Twenty-three patients were identified with a total of 37 matched sets of TEG-PM and PRU. Three of these pairs were excluded due to anemia outside of the PRU manufacturer’s recommended range. The Pearson correlation coefficient for these values was 0.50 (p = 0.0026). The prevalence of clopidogrel nonresponders determined by TEG-PM (9%) matched reported rates (5%–12%); PRU demonstrated much higher prevalence (39%). For detecting a therapeutic level of platelet inhibition, PRU demonstrated a sensitivity of 0.59, specificity of 0.50, positive predictive value of 0.95, and negative predictive value of 0.07. Ideal inhibition was concordant in only 25% of observations in which at least one of the results was ideal.
Agreement between TEG-PM and PRU regarding the degree of platelet inhibition is poor. PRU likely overestimates clopidogrel resistance, as 93% of patients with PRU > 194 demonstrate a therapeutic level of platelet inhibition on TEG.
Brian L. Hoh, Christopher M. Putman, Ronald F. Budzik, Bob S. Carter and Christopher S. Ogilvy
Object. Certain intracranial aneurysms, because of their fusiform or complex wide-necked structure, giant size, or involvement with critical perforating or branch vessels, are unamenable to direct surgical clipping or endovascular coil treatment. Management of such lesions requires alternative or novel treatment strategies. Proximal and distal occlusion (trapping) is the most effective strategy. In lesions that cannot be trapped, alteration in blood flow to the “inflow zone,” the site most vulnerable to aneurysm growth and rupture, is used.
Methods. From 1991 to 1999 the combined neurosurgical—neuroendovascular team at the Massachusetts General Hospital (MGH) managed 48 intracranial aneurysms that could not be clipped or occluded. Intracavernous internal carotid artery aneurysms were excluded from this analysis. By applying a previously described aneurysm rupture risk classification system (MGH Grades 0–5) based on the age of the patient, aneurysm size, Hunt and Hess grade, Fisher grade, and whether the aneurysm was a giant lesion located in the posterior circulation, the authors found that a significant number of patients were at moderate risk (MGH Grade 2; 31.3% of patients) and at high risk (MGH Grades 3 or 4; 22.9%) for treatment-related morbidity. The lesions were treated using a variety of strategies—surgical, endovascular, or a combination of modalities. Aneurysms that could not be trapped or occluded were treated using a paradigm of flow alteration, with flow redirected from either native collateral networks or from a surgically performed vascular bypass. Overall clinical outcomes were determined using the Glasgow Outcome Scale (GOS). A GOS score of 5 or 4 was achieved in 77.1%, a GOS score of 3 or 2 in 8.3%, and death (GOS 1) occurred in 14.6% of the patients. Procedure-related complications occurred in 27.1% of cases; the major morbidity rate was 6.3% and the mortality rate was 10.4%. Three patients experienced aneurysmal hemorrhage posttreatment; in two patients this event proved to be fatal. Aneurysms with MGH Grades 0, 1, 2, 3, and 4 were associated with favorable outcomes (GOS scores of 5 or 4) in 100%, 92.8%, 71.4%, 50%, and 0% of instances, respectively.
Conclusions. Despite a high incidence of transient complications, intracranial aneurysms that cannot be clipped or occluded require alternative surgical and endovascular treatment strategies. In those aneurysms that cannot safely be trapped or occluded, one approach is the treatment strategy of flow alteration.