Skip Ribbon Commands
Skip to main content
SharePoint

PublicComment

 Feedback

If you have any comments or questions on this page, please email us at:
 

Stroke recognition

Question Type:
Intervention
Full Question:
Among adults with suspected acute stroke (P), does the use of a rapid stroke scoring system or scale (I), compared with standard first aid assessment (C), change time to treatment (eg door to balloon), recognition of acute injury or illness, discharge with favorable neurologic status, cognitive knowledge, survival with favorable neurologic outcome (O)?
Consensus on Science:
For the critical outcome of time to treatment, we identified 6 studies with 6 different stroke assessment systems studied: 1. For the Face (facial drooping), Arm (arm weakness), Speech (speech difficulty), Time (time to call 9-1-1/EMS) (FAST) scale (measured as number of patients with time from symptom onset to hospital arrival within 3 hours), we identified moderate-quality evidence from 1 observational study(Harbison 2003, 71) enrolling 356 patients showing benefit where 48.2% patients who had the scale applied arrived within 3 hours compared with 14.6% who did not have the scale applied (RR, 3.3; 95% CI, 2.29–4.75). 2. For the Kurashiki Prehospital Stroke Scale (KPSS; measured as number of patients with time from symptom onset to hospital arrival within 3 hours), we identified very-low-quality evidence (downgraded for risk of bias) from 1 observational study(Iguchi 2011, 51) enrolling 430 patients showing benefit where 62.9% patients who had the scale applied arrived within 3 hours compared with 52.3% who did not have the scale applied (RR, 1.2; 95% CI, 1.01–1.43). In the same study, the mean time was 2.1 hours for those who had a stroke screening scale applied compared with 2.7 hours for those who did not have a stroke screening scale applied (MD, −0.6; 95% CI, −2.45 to 1.25). 3. For the Ontario Prehospital Stroke Scale (OPSS; measured as number of patients with time from symptom onset to hospital arrival within 3 hours), we identified very-low-quality evidence (downgraded for risk of bias) from 1 observational study(Chenkin 2009, 153) enrolling 861 patients showing no significant benefit where 52.3% patients who had the scale applied arrived within 3 hours compared with 47.2% who did not have the scale applied (RR, 1.1; 95% CI, 0.96–1.28). 4. For the Los Angeles Prehospital Stroke Screen (LAPSS; measured in minutes from symptom onset to emergency department arrival time), we identified low-quality evidence from 1 observational study(Wojner-Alexandrov 2005, 1512) enrolling 1027 patients showing a mean time of 356 minutes for those who had a stroke screening scale applied compared with 359 minutes for those who did not have a stroke screening scale applied (SMD, 0.11; 95% CI, 0.02–0.24). 5. For the Cincinnati Prehospital Stroke Scale (CPSS; measured with EMS on-scene time), we identified low-quality evidence (downgraded for risk of bias) from 1 observational study(Frendl 2009, 754) enrolling 308 patients showing no benefit, as the mean on-scene time was 17 minutes for those who had a stroke screening scale applied compared with 19 minutes for those who did not have a stroke screening scale applied (MD, −2.00; 95% CI, −3.34 to 0.66). 6. For the Face, Arm, Speech, Time, Emergency Response (FASTER) protocol (measured with symptom onset to emergency department arrival [door] time), we identified very-low-quality evidence (downgraded for risk of bias) from 1 observational study(O'Brien 2012, 241) enrolling 115 patients showing no significant benefit where the mean time was 59 minutes for those who had a stroke screening scale applied compared with 76 minutes for those who did not have a stroke screening scale applied (P=0.180). For the important outcome of recognition of stroke (interventional studies, outcome defined as definitive stroke diagnosis or administration of thrombolytic/fibrinolytic; the publications varied in the term used), we identified 4 observational studies of 4 different stroke scales: 1. For FAST (measured as number of patients with confirmed stroke or transient ischemic attack), we identified moderate-quality evidence from 1 observational study(Harbison 2003, 71-76) enrolling 356 patients showing benefit where 48.2% patients who had the scale applied were diagnosed compared with 14.6% who did not have the scale applied (RR, 3.3; 95% CI, 2.29–4.75). 2. For KPSS (measured as number of patients who received fibrinolytic), we identified very-low-quality evidence (downgraded for risk of bias) from 1 observational study(Iguchi 2011, 51-56) enrolling 430 patients showing no benefit where 13.7% patients who had the scale applied were diagnosed compared with 14.4% who did not have the scale applied (RR, 0.95; 95% CI, 0.59–1.53). 3. For the FASTER scale (measured as number of patients who received thrombolytic), we identified very-low-quality evidence (downgraded for risk of bias) from 1 observational study(O'Brien 2012, 241) enrolling 34 patients showing benefit where 19.1% patients who had the scale applied received fibrinolytic compared with 7.5% who did not have the scale applied (RR, 0.87; 95% CI, 0.78–0.98). 4. For CPSS (measured with patients who received fibrinolytic), we identified moderate-quality evidence from 1 observational study(You 2013, 1699) enrolling 308 patients showing benefit where 45.7% patients who had the scale applied received fibrinolytic compared with 2.1% who did not have the scale applied (RR, 22.2%; 95% CI, 7.14–69.1). For the important outcome of recognition of stroke (diagnostic studies, outcome defined as correct stroke diagnosis), we identified low-quality evidence (all downgraded for risk of bias) from 22 observational studies(Kothari 1999, 373; Kidwell 2000, 71; Harbison 2003, 71; Bray 2005, 28; Nor 2005, 727; Wojner-Alexandrov 2005, 1512; Kleindorfer 2007, 2864; Nazliel 2008, 2264; Ramanujam 2008, 307; Buck 2009, 2027; Chenkin 2009, 153; Frendl 2009, 754; Bergs 2010, 2; Bray 2010, 1363; Whiteley 2011, 1006; Yock-Corrales 2011; Chen 2013, e70742; De Luca 2013, 513; Fothergill 2013, 3007; Studnek 2013, 348; Asimos 2014, 509; Jiang 2014, e109762) enrolling a total of 30 635 patients, studying 8 different stroke screening assessment systems, showing diagnostic performance across all stroke screening systems of sensitivity ranging from 0.41 to 0.97 and specificity ranging from 0.13 to 1.00. These studies were divided into subgroups based on whether the stroke scales included glucose measurement or not. For studies that included stroke scales with glucose measurement (LAPSS, OPSS, KPSS, and Recognition of Stroke in the Emergency Room [ROSIER]), the pooled sensitivity was 0.84 (95% CI, 0.82–0.85) and pooled specificity was 0.97 (95% CI, 0.97–0.97), compared with stroke scales without glucose measurement (FAST, Melbourne Ambulance Stroke Screen [MASS], Los Angeles Motor Scale [LAMS], CPSS, Medical Priority Dispatch System [MPDS]), which have pooled sensitivity of 0.82 (95% CI, 0.81–0.83) and pooled specificity of 0.48 (95% CI, 0.46–0.49). For the important outcome of increased public/layperson recognition of signs of stroke, very-low-quality evidence (downgraded for risk of bias) from 1 human study(Wall 2008, A49) enrolling 72 participants (members of the public) showing benefit where 76.4% of participants (55/72) were able to identify signs of stroke before training on a stroke screening assessment system compared with 94.4% (68/72) immediately after training (OR, 5.25; 95% CI, 1.67–16.52), and 96.9% of participants (63/65) were able to identify the signs of stroke 3 months after training (OR, 2.07; 95% CI, 0.36–11.69).
Treatment Recommendation:
We recommend that first aid providers use stroke assessment systems (such as FAST or CPSS) for individuals with suspected acute stroke (strong recommendation, low-quality evidence). We suggest the use of FAST or CPSS stroke assessment systems (weak recommendation, low-quality evidence). We suggest the use of stroke assessment systems that include blood glucose measurement, when available, such as LAPSS, OPSS, ROSIER, or KPSS, to increase specificity of stroke recognition (weak recommendation, low-quality evidence). In the absence of a glucometer, we suggest the use of FAST or CPSS stroke assessment systems compared with MASS, LAMS, or MPDS (weak recommendation, low-quality evidence). The literature search was rerun in January 2015 to capture the most updated evidence possible. Two additional studies were added(Asimos 2014, 509-515; Jiang 2014, e109762) and incorporated into the consensus on science and GRADE tables, both supporting this treatment recommendation. Values, Preferences, and Task Force Insights In making this recommendation, we place increased value on the benefits of early stroke recognition, which could lead to early treatment to minimize potentially devastating neurologic injury. Training first aid providers in stroke assessment systems outweighs the risks, largely limited to false-positive identification by first aid providers. The cost of the intervention is estimated to be low. In this review of the literature, the stroke assessment systems include various components, such as looking for specific signs and obtaining blood glucose levels. Our review found that stroke assessment systems that included blood glucose measurement had similar sensitivity and increased specificity to accurately identify stroke compared with those systems that did not include glucose measurement. We recognize that first aid providers may or may not have access to a properly calibrated glucose measurement device. Although use of these devices is not a standard component of first aid, glucose measurement devices are commonly available among the public. Ideal stroke assessment systems for first aid use are accurate, have few steps, are easily understood and remembered, and take minimal time to complete. Those developing local guidelines for first aid providers can use the results of this review to determine if the benefit of increased specificity with systems that include glucose measurement would be desirable in their settings, compared with using simpler stroke assessment systems that do not include glucose measurement, which have similar sensitivity but lower specificity.
CoSTR Attachments:
2015 01 16 GRADE GRID - Stroke Recognition_Final.docx    
2015 02 04 ROC image.bmp    
2015 02 10 ILCOR Data Collection.xls    
2015 02 16 StrokeCoSTR.docx    
Forest plot .pdf    
Methodological quality graph 1.pdf    
Methodological quality summary 2.pdf    
Risk of bias graph 1 intervention.pdf    
Risk of bias summary 2 intervention.pdf    
Stroke Table of Systems Evaluated.docx    

 Contact Us

 
If you have any comments or questions on this page, please email us at: