HONOLULU – Applying the American Heart Association’s "Target: Stroke" best practices more than doubled the percentages of patients receiving CT imaging and lifesaving, clot-busting therapy in a study of 2,338 patients.
The percentage of patients who underwent CT imaging within 25 minutes of hospital arrival increased from 12.6% to 27.7% post intervention, while receipt of intravenous tissue plasminogen activator (TPA) within 60 minutes of arrival jumped from 32% to 70.3% (both P less than .001), Dr. Ilana Ruff reported at the International Stroke Conference.
"Target: Stroke best practices are feasible to implement but require a dedicated team effort, including the ED [emergency department], to be most successful," said Dr. Ruff, a vascular neurology fellow at Massachusetts General Hospital in Boston.
The preintervention data are somewhat surprising given that hospitals with high annual volumes of TPA-treated patients have been shown to be associated with lower door-to-needle times, but Dr. Ruff points out that they are not alone. The pivotal University of California, Los Angeles, study that prompted the creation of Target: Stroke in 2010 reported that hospitals had door-to-needle times within 60 minutes in only 29.1% of patients treated with intravenous TPA in 2009, with a range of 0% to 79.2% (Circulation 2011;123:750-8).
The newly updated American Heart Association/American Stroke Association acute ischemic stroke guidelines put a strong emphasis on the 60-minute treatment window for TPA and the need to streamline initial testing to avoid delays in its administration.
The guidelines also feature a new recommendation calling on health care institutions to create a multidisciplinary quality improvement committee to review and monitor stroke care quality benchmarks, evidence-based practices, and outcomes.
Massachusetts General Hospital implemented a new institutional acute stroke care model in 2007 that included 10 best practices, all of which were later included in the Target: Stroke program. Some of those changes included sending stroke patients to the acute area of the ED; using a group paging system to notify the entire stroke team, including radiology technicians, nurses, pharmacy, and the laboratory; sending all labs as "super stat"; storing TPA in an ED drug-dispensing machine or a TPA box; having neuroradiologists in the ED bay; having a pharmacist at the bedside during the entire stroke code to mix TPA for eligible patients; and holding monthly case reviews, Dr. Ruff said.
Structurally, the hospital also shifted sole ownership of stroke patients from the stroke fellow by naming the ED attending the team captain to handle door-to-CT flow and the stroke fellow to facilitate CT-to-TPA administration.
"We realized that the emergency department is best equipped to handle emergency protocols," she said.
The investigators analyzed the median ED door-to-CT and door-to-needle within 60 minutes (DTN60) times in 2,338 patients who arrived within 4.5 hours of an acute ischemic stroke and were enrolled in the hospital’s Get with the Guidelines Stroke registry from 2003 to 2011.
In all, 281 patients received TPA, of which 155 were treated within 60 minutes and 126 beyond 60 minutes.
Patients’ characteristics for those treated with TPA were similar, including average National Institutes of Health Stroke Scale (NIHSS) score (mean 13; range 8-19) and prevalence of hypertension (about 75%) and diabetes mellitus (about 25%). The median age was 72.
Among all patients, the median time from the ED to CT decreased from 71 minutes in the 2003-2006 preintervention period to 59 minutes in the 2008-2011 postintervention period (P less than .001), Dr. Ruff said. Among TPA-treated patients, the median time to CT dropped from 26 minutes to 18 minutes over the same time period (P less than .001).
Median DTN60 times decreased from 70 minutes before intervention to 47 minutes post intervention (P less than .001). Specifically, the median time to TPA was 51.5 minutes in 2007 and reached a low of 40 minutes in 2010 before an uptick to 51 minutes in 2011 (both P less than .001).
It’s unclear why the DTN60 time increased in 2011, although it’s possible, but unlikely, that a new neurology unit further from the ED may have played a role, she said in an interview.
In univariate analysis, patients with a DTN60 did not differ significantly in age, gender, race, comorbidities or NIHSS scores. Only calendar year was associated with achieving DTN60, suggesting that changes at the hospital level, rather than patient-related factors, are driving the improvement, she said at the meeting sponsored by the American Heart Association.
Dr. Ruff and her coauthors reported no disclosures.