How to rapidly and accurately identify an acute stroke patient in a prehospital setting?

Does your hospital publish their diagnosis results? Or they don't know them? INCOMPETENCE BY INACTION OR INCOMPETENCE BY DESIGN? 

Lots of statistics here but completely unusable.

Just maybe you want extremely fast diagnosis.

Hats off to Helmet of Hope - stroke diagnosis in 30 seconds   February 2017

 

Microwave Imaging for Brain Stroke Detection and Monitoring using High Performance Computing in 94 seconds March 2017

 

New Device Quickly Assesses Brain Bleeding in Head Injuries - 5-10 minutes April 2017

 The latest here:

How to rapidly and accurately identify an acute stroke patient in a prehospital setting?

Acute stroke is a time-dependent condition. If it happens in a prehospital setting, practitioners have to know how to treat the patient in the best and fastest way as possible. Here the results of a pilot study in Genova (Italy).

This article will report a pilot study led by Dr Andrea Furgani, MD Policlinico San Martino, Genova which aims to identify how to rapidly recognize and treat an acute stroke in a prehospital setting and what difference there is with the neurologist in-hospital stroke assessment.

Why is important to immediately identify a stroke both in the prehospital and in-hospital setting?

Dr Furgani, as the entire scientific community, declares in his study, that early identification and treatment of the stroke critically reduces immediate damage and long-term disability. The challenge to won is the rapid and accurate identification of a stroke patient.
The definitive treatment for an acute stroke is the lysis of the thrombus carried out as soon as possible after the onset. The use of EMS, compared to the spontaneous presentation at the Emergency Room (ER) of the patient, improves time measures and scores on indexes. According to the National Institutes of Health Stroke Scale (NIHSS) and Barthel Index, the emergency care teams dispatch may reduce complications and mortality. On the other hand, it also reduces time to administration of tissue plasminogen activator.
In 2019, several trials have shown the efficacy of endovascular therapy (ET) with stent retrievers versus IV t-PA alone in patients with Large Vessel Occlusion (LVO) who generally presented NIHSS (National Institute of Health Stroke Scale) scores greater than or equal to 6.9.

What does the literature say about this?

The current literature indicates that strokes are only identified by emergency call-takers about one-third to one- half of the time. Dr Furgani explains that the Stroke Genova Network uses a first telephone “checkpoint” made using MPDS (Medical Priority Dispatch System). Then, when the rescuers are with the patient, they perform the Cincinnati Stroke Scale. If this second “checkpoint” is positive, the Emergency Medical Communication Center (EMCC) activates the “Stroke Team” during transport to the hospital.
By telephone, the team communicates the sex and age of the patient, presumed time of onset of symptoms, and estimated time of arrival. Is important for the network to find a correlation between the Medical Priority Dispatch System Stroke Diagnostic Tool (SDxT) and NIHSS because, with high probability, the patients with NIHSS > 10 must be subjected to ET. During an emergency call, it is absolutely important to detect patients with probable NIHSS > 10. This would allow the network to send the patient to the hospital capable of providing the best therapy, saving time and brain (San Martino hospital, in the case of Genova).
It was used the MPDS (Priority Dispatch Corp.’, MPDS version 12.1, 2012, Salt Lake City, UT, USA) at the time of information gathering. The Emergency Medical Dispatch Quality Assurance (EMD-Q) replayed and reviewed, with two specific objectives, stroke cases confirmed by neurologists but unidentified during the emergency call. They proceeded this way to determine whether the selected Chief Complaint was correct (using the International Academies of Emergency Dispatch’—IAED~—standards, version 9a), and to determine whether any spontaneous information related to the stroke was provided by the caller during the call. Essential is the consideration of the population studies. The Genova 118 EMS covers a region of 736,235 inhabitants (52.4% female) and an area of 1,127.41 square kilometers (653 people/square kilometer); 28.2% of the population served is 65 years or older.

Rapidly identify an acute stroke. What are the results?

For analysis and plots they carried out SPSS’ Statistics software (IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp.). They assessed statistical significance was using the Kruskal-Wallis test for the independent sample, using a 0.05 cut-off level of significance. For the analysis of NIHSS values the average, standard deviation, and confidence interval (Cl) were used, while for the analysis of the time intervals the median, with 25th and 75th percentile expressed in brackets, was used.
Among the results, they found out that of 438 suspected strokes included in the registry, 353 cases (80.6%) called EMS. Other cases included: self-presentation, 64 cases (14.6%); sent from other hospitals, 21 cases (4.8%). The patients who called EMS had an NIHSS on arrival at the hospital of 10.9 (Cl: 9.5 – 12.3), as opposed to 6.0 (Cl: 2.0 – 10.0) for self-presentation at first aid, and 15.1 (Cl: 9.3 – 20.9) for patients transferred from other hospitals (Fig. 1). Of the patients who called EMS, 205 (58.1%) were identified as suspected strokes by the EMD during the emergency call.
Of the remaining 148 cases, in 104 the suspicion of stroke was posed by ambulance rescuers, and in 44 of those cases the Chief Complaint was missing at dispatch. In the 104 cases the most frequent Chief Complaints were Sick Person (n=31, 29.8%), Unconscious/Fainting (n=28, 26.9%), Unknown Problem (n=16,15.4%), and Falls (n=15; 14.4%) (Table 1). The SDxT was used in 129 (62.9%) cases: 5 (3.9%) no evidence; 87 (67.4%) PARTIAL evidence; 5 (3.9%) STRONG evidence; and 32 (24.8%) CLEAR evidence.
In 76 cases, the SDxT was not used or was not completed. Time of onset, as collected in the SDxT, was classified as follows: less than 4 hours 93 cases (72.1%); between 4 and 6 hours 4 cases (3.1%); more than 6 hours 10 cases (7.8%); unknown 22 cases (17.1%)
The neurologist at the hospital confirmed 260 cases out of 353 (73.7%); of these, 91.5% (n=238) were ischemic, and 8.5% (n=22) were hemorrhagic. Of the 205 cases identified by the EM D, 154 (75.1%) were confirmed by neurologists, while of the 104 cases identified by rescuers, 78 (75.0%) were confirmed at the hospital (Fig. 2). The report of symptom onset time during the emergency call was in agreement with the evaluation of the neurologist in the hospital in 58 cases of the 97 reported by the EMD (59.8%); in the remaining 20 cases (2 cases are missing) classified by EMDs as unknown, 65.0% (n=13) were identified by the hospital as occurring within 4 hours.
The average time between the call and the arrival at the hospital was 31 minutes (25 – 43); when the suspected stroke was identified by the EMDs, the time was 31 minutes (25 – 42), while if the stroke was identified by the rescuers on field it was 33 (25 – 44). No significant difference was found in the interval from onset to first neurological contact if the suspicion of stroke was posed by EMDs or rescuers: with EMD stroke recognition it was 126.5 minutes (64 – 316), and with rescuer identification it was 120 minutes (64 – 360). A significant difference was found in the time of first neurological contact between EMS and self-presentation: 123.5 minutes (64 – 329) for the patients who called EMS versus 317.5 minutes (107 – 2033) for self-presentation (p < 0.000) (Fig. 3).
The study of the correlation between NIHSS and the SDxT did not find significant results (Table 2): the NIHSS at first aid for patients with PARTIAL evidence was 9.7 (Cl: 7.4 -12.0), while for STRONG or CLEAR evidence it was 10.9 (Cl: 7.3 – 14.4). Stroke cases confirmed by neurologists but unidentified during the emergency call (n = 78) have been replayed with two specific objectives: to determine whether the selected Chief Complaint was correct, and to determine whether during the call any spontaneous information related to the stroke was provided by the caller (Fig. 4). In 17 cases (21.8%), it was not possible to find the emergency call recording. Of 61 cases remaining, in 11 cases (18.0%) a Chief Complaint other than Stroke was selected. Selected Chief Complaints included Sick Person (n=6, 54.5%), Unconscious/ Fainting (n=3, 27.3%) and Unknown Problem (n=2, 18.2%). In 34 (68.0%) of the remaining 50 cases, during the playback it was possible to identify at least one piece of information spontaneously provided by the caller referring to the stroke symptomatology: one mention in 21 cases (42.0%), two in 12 cases (24.0%), and three in one case (2.0%) (no information n=16, 32.0%). Spontaneous information included mentions of difficulties in speaking (n = 17), problems with balance or coordination (n = 11), weakness or numbness (n = 5), headache (n = 4), and visual problems (n = 3).

Rapidly identify an acute stroke: discussion on the results

Good ability of EMDs is identified by the results. 58.1% of stroke cases transported to the hospital due to suspected stroke were identified by EMDs during the emergency calls. The results obtained demonstrate the key role of the additional “face to face” evaluation by rescuers, who were able to identify stroke in 37.9% of the cases studied.
This study identified Falls, Sick Person, and Unconscious/Fainting as the most common Chief Complaints in cases of patients with unidentified stroke during the emergency call. Even if the caller declares a Chief Complaint different from the stroke, the critical review of the emergency calls also showed that spontaneous information referring to the stroke symptomatology is sometimes present during the call.
In addition, access to the hospital via EMS generally guarantees an improvement of time to first neurological contact and, presumably, also access to definitive therapies.

Rapidly identify an acute stroke: what are the limitations?

This is a pilot study, which means it is limited in time and in the number of cases. Furthermore, results have been altered due to the high number of missing values. The review of cases was conducted according to IAED standards only for the selection of the Chief Complaint. In addition, the EMD-Q who performed the reviews was informed that those cases concerned patients with stroke: this may have influenced their determination of the correct Chief Complaint
and increased their likelihood of identifying spontaneous information related to the stroke during the playback. The information came from a center that is not an ACE and is limited to a specific geographical and cultural territory (Genova city).

Ideas in the conclusion of this study on the acute stroke

The MPDS showed an excellent ability to identify patients with stroke during the emergency call. In the cross-section analyzed, most patients with a suspected stroke called EMS (80.6%) to transport them to the hospital. Of the 205 cases identified by the EMDs, 75.1% were confirmed by neurologists at the hospital.
Further studies should examine in depth the stroke cases in which the time of onset was reported by the EMD as “unknown.” EMS considerably reduces the time between symptom onset and the first contact with the neurologist. The correlation between SDxT and NIHSS would seem to be useful for telephone screening of patients with NIHSSalO, but this study is inconclusive for this topic.