Search

Background: Improved clinical outcomes after intravenous tissue plasminogen activator (IV tPA) are time dependent. Participation in a telestroke program allows the spoke hospitals 24/7 access to a vascular fellowship trained neurologist for a telestroke consult, as well as educational partnership with the hub site, shared protocols, and access to quality improvement feedback. We sought to assess the effects of continued participation in a telestroke program on times to administration of IV tPA. Methods: Our institutional telestroke program began in March 2011 and consists of an academic hub (comprehensive stroke center) that serves 8 community spoke hospitals. We retrospectively reviewed acute ischemic stroke patients treated with IV tPA via the telestroke program. We compared 2 cohorts of patients: Period 1 (July 2011 to June 2013) and Period 2 (July 2013 to July 2014). We collected data on demographics, National Institutes of Health Stroke Scale (NIHSS), and times from initiation of telestroke consult to IV tPA administration. Results: Among 259 consecutive stroke patients (mean: 69.6 years, 56% female) treated with IV tPA via telestroke, the median NIHSS score was 11.8, and 41.7% of patients were transferred to the hub. The mean time from initiation of telestroke consult to IV tPA administration was 42.2 minutes. Period 1 included 129 patients and Period 2 included 130 patients, and the two groups did not differ by age (p=0.2), gender (p=0.3), or NIHSS score (p=0.3). Time from initiation of telestroke consult to IV tPA administration improved from Period 1 to Period 2 (35 vs. 49.9 minutes, p Conclusions: Maturation of a telestroke program is associated with improvement in the timeliness of IV tPA delivery, possibly due to a learning effect that continues the longer the sites participate in the program. This improvement is due to faster responses in both the hub site (recommending IV tPA earlier) and spoke site (administering IV tPA quicker). Further studies aimed at improving delivery of IV tPA in telestroke program are warranted.

Background: A serious complication of intravenous tissue plasminogen activator (tPA) in acute ischemic stroke is hemorrhage. Alterations in fibrinogen and fibrinogen degradation products, labeled “early fibrinogen degradation coagulopathy” (EFDC), are well established after tPA. However, tPA is not recognized to cause markedly abnormal international normalized ratio (INR). Methods: With IRB approval, we retrospectively reviewed 338 acute ischemic stroke patients who received tPA. We defined post-tPA coagulopathy as a documented elevation of INR ≥ 1.5 after IVtPA within 24 hours of tPA without a known cause. Patients with recent warfarin use (within 3 weeks of presentation) were excluded. Results: Among the 328 acute ischemic stroke patients who had received tPA, we identified 21 (7.4%) patients with an elevated INR post-thrombolysis. The mean age was 68.3 years (SD +/- 11.9) and 57% were male. The mean initial NIHSS (pre-tPA) was 15.8 (range, 4 to 35). Liver disease or alcohol abuse was noted in 19%. There were 2 tPA protocol violations who received >90mg tPA. The mean post-tPA INR was 2.03 (range, 1.5 to 4.7) and the elevation in INR was documented within a mean 5.4 hours (range, 1 to 15) after tPA initiation. Repeat INR levels returned to normal during their hospital stay in 19 patients. Hypofibrinogenemia was noted in 10 of 12 patients who had fibrinogen levels drawn within 48 hours after tPA initiation (7 of these fibrinogen levels were drawn the same time as the elevated INR). Among the 6 patients with bleeding complications, 2 patients had symptomatic intracerebral hemorrhage. Conclusions: We report an under-recognized early transient coagulopathy associated with elevated INR in stroke patients after treatment with tPA. Post-thrombolytic INR elevation may represent a severe manifestation on the spectrum of EDFC rather than a distinct phenomenon

Background: Acute stroke treatment is recognized to be time dependent. We sought to assess the effects of a quality improvement (QI) initiative called “Stroke60” on reducing time to intra-arterial therapy (IAT) for acute ischemic stroke patients transferred from an outside hospital to an academic center. Methods: We reviewed acute ischemic stroke transfer patients who underwent attempted IAT for suspected large vessel occlusion. The time to IAT (tIAT) was defined as time from arrival at our institution to groin puncture. Stroke60 QI protocol entailed early notification of parallel service lines prior to patient arrival. We compared 2 cohorts of patients: pre-Stroke60 (January 1, 2010 to May 1, 2012) and post-Stroke 60 (May 1, 2012 to August 1, 2013). We collected data on demographics, National Institutes of Health Stroke Scale (NIHSS), frequency of neuroimaging (CT, CT angiogram/perfusion, or MRI brain) prior to IAT, and tIAT. Results: Seventy-nine transfer patients underwent urgent cerebral angiogram for acute ischemic stroke (52 pre-Stroke60, 27 post-Stroke60). The mean age was 60.5 (range, 17 to 84 years) and the mean NIHSS was 19.5. Comparing post-Stroke60 with pre-Stroke60, there was no significant difference in age (p= 0.1), gender (p= 0.2), race (p= 0.8), or NIHSS (0.7). Stroke60 implementation resulted in a reduction of mean tIAT by 18 minutes (52.7 vs. 34.3 min, p= 0.04). Among the entire cohort, 25% (20/79) underwent neuroimaging prior to IAT. Neuroimaging prior to IAT was associated with increased delay of mean tIAT (74 min vs. 37 min, p= 0.0001). The post-Stroke60 cohort was more likely to have imaging prior to IAT compared to the pre-Stroke60 cohort (10% vs. 56%, p< 0.0001). Conclusion: Despite being associated with an increased use of imaging from 10% to 56%, the “Stroke60” QI initiative was associated with a significant reduction in tIAT for acute ischemic stroke transfer patients. Further studies aimed at improving time to acute stroke therapies are warranted.

Background: Controversy exists on the role of perfusion imaging-based selection of patients with acute ischemic stroke (AIS) for endovascular reperfusion therapy (ERT). We hypothesized that perfusion imaging based selection would improve functional outcomes at 3 months compared to time based selection alone. Methods: We reviewed consecutive patients with anterior circulation strokes, baseline NIHSS score > 8, and treated with ERT at 4 centers between 2006 and 2011. Data were collected on demographics, initial NIHSS score, intravenous tissue plasminogen activator (IV tPA), use of perfusion imaging (CT or MRI), time from onset to groin puncture, reperfusion grade using the modified TICI scale, symptomatic hemorrhage, and functional outcome using the modified Rankin Scale (mRS) at 3 months. Using univariable statistics and multivariable logistic regression, we assessed whether patients who underwent perfusion imaging had better outcomes than those selected based on time alone. Results: One-hundred eighty-five patients (mean age 66.7 years; median NIHSS score 19; MCA occlusion 73% and ICA occlusion 27%) were included. TICI 2b or 3 reperfusion grade was achieved in 49.7% while symptomatic hemorrhage (PH1, PH2, or perforation) occurred in 10.8%. Perfusion imaging was used in 69 (37.3%) patients (45 CT and 24 MRI) and was associated with increased onset-to-groin puncture time (359 vs. 298 minutes, P=0.019). Patients who underwent perfusion imaging were also older (73 vs. 63 years, P Conclusions: In this multi-center observational study, patients with AIS who underwent perfusion imaging were over 2-fold more likely to have good outcomes following ERT despite a delay in time to treatment and advanced age compared to time-based selected patients. Further studies should continue to address the optimal perfusion imaging criteria for patient selection for ERT.

Background: The International Cooperative Study on the Timing of Aneurysm Surgery suggested that delayed cerebral vasospasm is a significant contributor to mortality (at 6 months) after subarachnoid hemorrhage (SAH). However, the Cincinnati population based study did not find cerebral vasospasm to be a significant contributor to short-term mortality. Methods: We reviewed 350 consecutive non-traumatic SAH patients who were hospitalized at our institution from August 1, 2006 to June 30, 2010. Data were collected on demographics, Hunt and Hess grade, presence of cerebral infarct, symptomatic vasospasm, and in-hospital mortality. Symptomatic vasospasm was defined as cerebral vasospasm that required intra-arterial therapy (pharmacologic vasodilators or angioplasty). Cerebral infarct was defined as any new radiographic infarct seen on CT or MRI (of any cause). Results: Among 350 patients with non-traumatic SAH, the mean age was 56.7 years (range, 21.5 to 89.1) and 68% were aneurysmal. The overall in-hospital mortality rate was 20.2% (71 patients). The causes of in-hospital mortality were withdrawal of support in 53 (75%), brain death in 14 (20%), and cardiopulmonary death in 4 (8%). The mean time to death was 8.1 (range 0 to 31) days, with 36% of deaths occurring within the first 2 days of hospitalization. Among patients with symptomatic cerebral vasospasm or cerebral infarct , 19% (26/139) died. In univariate and multivariate analyses, symptomatic cerebral vasospasm and cerebral infarct were not associated with in-hospital mortality. When the analysis was restricted to patients with aneurysmal SAH, cerebral vasospasm and cerebral infarct were still not predictive of in-hospital mortality. In multivariable analysis, the variables most associated with in-hospital mortality were Hunt and Hess score (HH) (p Conclusions: In-hospital mortality after SAH is associated with initial severity of hemorrhage, rebleeding, and age. Symptomatic cerebral vasospasm and cerebral infarct were not significant contributors of overall in-hospital mortality, and approximately one-third of in-hospital mortality after SAH occurs within the first 2 days of hospitalization. By design, studies that exclude patients with severe SAH tend to over-estimate the contribution of cerebral vasospasm on short-term mortality.

Introduction: Rapid availability to stroke specialists remains a critical hurdle in thrombolysis for acute ischemic stroke in many regions of the United States. Remote consultation is often utilized with varying success. Telemedicine use has been increasing but it is unclear if this is cost effective when compared to simple telephone consultation. Hypothesis: We assessed the hypothesis that telemedicine is cost effective for acute ischemic stroke patients. Methods: We created a decision tree model to compare telemedicine (with videoconferencing and image sharing) to telephone consultation. We estimated probabilities and utilities from the available literature and calculated incremental cost-effectiveness ratios (ICER). We used actual costs of telemedicine equipment, installation, and physician fees based on our own experience ($50,900 per hospital per year). We assumed equal access to standard medical treatments between the two approaches; an 8-hospital spoke system; and 60 patients assessed for acute stroke per year at each spoke hospital. However, we estimated higher tissue plasminogen activator (tPA) utilization among telemedicine-treated patients (28% vs 23%) based on the STRokE-DOC trial. We performed sensitivity analyses to assess all assumptions in our model. An ICER < $50,000/QALY was considered cost-effective. Results: In the first year after stroke, telemedicine resulted in a gain of 0.00459 QALY, but with an additional cost of $799 compared to telephone only (ICER = $174,074/QALY). In the long term (over 30 years), telemedicine resulted in a gain of 0.0282 QALY at an additional cost of $497 (ICER = $17,624/QALY). Using a long term sensitivity analysis, ICER remained < $50,000/QALY when the absolute difference in tPA utilization (telemedicine vs. telephone) was > 2.5%. When the absolute difference in tPA use is > 12%, telemedicine had cost-saving effects (ICER < 0). Cost-effectiveness was also sensitive to number of patients: 30 or more patients must be evaluated with telemedicine per hospital annually to remain cost-effective. Telemedicine remained cost-effective when varying its program costs from $500 to $1,500 per patient. Conclusions: Assuming that telemedicine consultations increase the use of tPA for acute ischemic stroke patients by at least 2.5%, we found that it is cost-effective compared to telephone only consultations. However, 30 patients must be assessed annually per spoke hospital to remain cost-effective. Telemedicine stroke networks need to consider these aspects when assessing their societal impact and when justifying their considerable costs.

Introduction: A subset of patients with acute ischemic stroke (AIS) who are treated with intravenous tissue plasminogen activator (IV tPA) may benefit from further intra-arterial (IA) treatment but are initially treated at hospitals without endovascular resources. This necessitates emergent transfer (“drip-and-ship” strategy) to a facility with IA capabilities. The safety and utility of this method has not been sufficiently evaluated. Hypothesis: We assessed the hypothesis that the “drip-and-ship” strategy can be safely combined with IA therapy and leads to improved outcomes following acute ischemic stroke. Methods: We retrospectively reviewed consecutive AIS patients from July 2007 to May 2011 who were treated with IV tPA at community hospitals and then emergently transferred to our comprehensive stroke center. Demographic, clinical, and radiographic data were retrieved from an electronic registry. National Institutes of Health Stroke Scale scores were abstracted from the medical records prior to tPA administration and at hospital discharge (amongst survivors). Protocol violations and treatment-related complications were abstracted from the medical record. Results: Among 65 “drip-and-ship” IV tPA patients (mean age 62 years, 57% male) admitted to our stroke center, the median NIHSS score was 16 and median onset-to-treatment time was 160 (IQR 125-193) minutes; 72% were treated in the 3-hour window. There were 8 (12.3%) protocol violations: 1 treated > 4.5 hours from onset, 1 treated with cardiac dose of tPA, 1 treated in 3-4.5hr window despite NIHSS score > 25, 2 despite prior stroke and diabetes history, and 3 despite age > 80 years. Twenty-five patients (38%) underwent subsequent IA therapy. Overall, there were 12 (18.5%) intracranial hemorrhages (IV-9.8%, IV + IA-33.3%, p=0.024), of which 3 (4.6%) were sICH (IV-2.4%, IV + IA-8.3%; p=0.274). In-hospital mortality was higher among IV-IA patients (16.7 vs. 4.9%, p=0.183) than IV tPA alone. Among survivors (n=59), the median NIHSS improved from admission to discharge for IV tPA alone (13 to 6, p < 0.001) and following IV-IA treatment (20 to 13, p < 0.001). Conclusions: “Drip-and-ship” thrombolysis appears safe despite an increase in protocol violations. A combined approach with IA treatment leads to improvements in the NIHSS score at discharge and has a safety profile comparable to previous studies. A combined “drip-and-ship” model for acute ischemic stroke may be a safe and effective strategy to increase access to IA therapy among eligible patients presenting to community hospitals with large artery occlusion.

Objective: Non-stroke center hospitals often utilize a “drip-and-ship” (DS) method in which the acute ischemic stroke (AIS) patients are emergently transferred to stroke centers after treatment with intravenous tissue plasminogen activator (IV tPA). In October 2008, a new ICD-9 ″V-code″ was approved by the Centers for Medicare and Medicaid Services to identify these patients. We analyzed the use of this V-code in the Nationwide Inpatient Sample (NIS) in order to characterize rates of use, regional differences, and safety among AIS patients in the United States (US). Methods: We searched the NIS, a database representative of 20% of inpatient admissions in the US, to identify patients with primary diagnosis of stroke using ICD-9 codes (430, 431, 433.01, 433.11, 433.21, 433.31, 433.81, 433.91, 434.01, 434.11, 434.91, or 436) in the last quarter of 2008 and all of 2009. We searched for “drip and ship” coding using the V-code (v45.88) among any of the 15 listed diagnostic codes and for secondary diagnoses of brain hemorrhage (430 or 431) complicating ischemic stroke. We included primary diagnoses of hemorrhage (430 or 431) only if ischemic stroke treatment was provided (DS, thrombolysis on day 0, or any mechanical embolectomy by procedure codes: v45.88, 99.10, and 39.72). We compared demographics, hospital characteristics, and outcomes in DS vs. non-DS thrombolysis patients using univariable statistics. Results: Of the 140,698 AIS discharges from October 2008 to December 2009, 4314 (3.1%) received intravenous thrombolysis and 929 (0.66%) were DS patients based on V-code reporting. V-code reporting increased in each quarter (0.5 to 0.8%, p Conclusion: In our analysis, we found that the “drip-and-ship” method is not uncommon and appropriate coding is increasing in the US. The practice is most prevalent in the Midwest and most patients are transferred to teaching hospitals. The safety of this method appears similar to non-DS tPA use for ischemic stroke. Further studies are needed to validate our findings.