Summary
Background
The early recognition of acute coronary syndromes is a priority in health care systems, to reduce revascularization delays. In France, patients are encouraged to call emergency numbers (15, 112), which are routed to a Medical Dispatch Centre where physicians conduct an interview and decide on the appropriate response. However, the effectiveness of this system has not yet been assessed.
Aim
To describe and analyse the response of emergency physicians receiving calls for chest pain in the French Emergency Medical System.
Methods
From 16 November to 13 December 2009, calls to the Medical Dispatch Centre for non-traumatic chest pain were included prospectively in a multicentre observational study. Clinical characteristics and triage decisions were collected.
Results
A total of 1647 patients were included in the study. An interview was conducted with the patient in only 30.5% of cases, and with relatives, bystanders or physicians in the other cases. A Mobile Intensive Care Unit was dispatched to 854 patients (51.9%) presenting with typical angina chest pains and a high risk of cardiovascular disease. Paramedics were sent to 516 patients (31.3%) and a general practitioner was sent to 169 patients (10.3%). Patients were given medical advice only by telephone in 108 cases (6.6%).
Conclusions
Emergency physicians in the Medical Dispatch Centre sent an effecter to the majority of patients who called the Emergency Medical System for chest pain. The response level was based on the characteristics of the chest pain and the patient’s risk profile.
Résumé
Contexte
Le diagnostic précoce des syndromes coronaires aigus est la priorité des systèmes de soins afin de réduire les délais de revascularization. En France, il est vivement recommandé aux patients d’appeler les numéros d’urgences (15, 112) qui conduisent à des centres de régulation où des médecins, en fonction de leur interrogatoire, décident de la prise en charge la plus appropriée. Cependant, l’efficacité de ce système n’a à ce jour jamais été évaluée.
Objectif
Décrire et analyser les choix de réponse des médecins lors des appels pour douleurs thoracique par le service d’aide médicale urgente (SAMU).
Méthodes
Du 16 novembre au 13 décembre 2009, l’ensemble des appels aux centres de régulation pour douleur thoracique non traumatique ont été inclus prospectivement dans une étude observationnelle multicentrique. Les caractéristiques cliniques et les décisions d’orientation ont été recueillies.
Résultats
Un total de 1647 patients a été inclus dans l’étude. L’entretien téléphonique n’a été réalisé avec le patient que dans seulement 30,5 % des cas, avec des parents, des témoins ou des médecins dans les autres cas. Un service mobile d’urgence et de réanimation a été envoyé pour 854 patients (51,9 %) qui souffraient de douleurs thoraciques typiques avec un risque élevé de maladie cardiovasculaire. Une ambulance a été envoyée pour 516 patients (31,3 %), et un médecin généraliste pour 169 (10,3 %). Les patients ne recevaient qu’un avis médical par téléphone dans 108 cas (6,6 %).
Conclusion
Les médecins d’urgence dans les centres de régulation médicale envoient un effecteur pour la majorité des appels pour douleur thoracique. Le type d’effecteur est défini par les caractéristiques de la douleur et le profil de risque du patient.
Background
The incidence of acute coronary syndromes (ACS) in France is 280 per 100,000 in men and 60 per 100,000 in women . Myocardial infarction accounts for 10–12% of the global annual mortality rate. Prompt intervention with appropriate care can significantly improve mortality, with associated cost savings. Data from French registries show a huge decrease in the 30-day mortality rate from 1995 to 2010 (11.3–4.4%), mainly due to decreased delays and enhanced access to reperfusion strategies . Prehospital management of chest pain remains challenging. International guidelines highlight the need for shorter delays to improve prognosis, particularly in the acute setting of ST-segment elevation myocardial infarction (STEMI) . Dedicated regional protocols are recommended to accelerate and improve ACS diagnosis in the emergency department (ED) or by the prehospital ambulance service, with or without medical staff on board . Accurate prehospital patient orientation allows the initiation of effective therapy, such as fibrinolysis, transfer for primary angioplasty or admission to a coronary care unit for early coronary angiography . Avoiding admissions to such units for patients with chest pain not due to ACS increases the unit’s performance, limiting unnecessary and expensive hospital stays.
In France, patients with chest pain call a Medical Dispatch Centre (service d’aide médicale urgente [SAMU]) and emergency physicians assess the probability of an ACS; if an ACS is suspected, they dispatch a Mobile Intensive Care Unit (MICU) with a physician on board. In the other cases, paramedics or a general practitioner can be sent on site, or the physician can simply advise the patient by telephone .
The emergency physician in charge of the telephone triage can only rely on the clinical data gathered during the telephone call. Calls for ACS represent about 15% of all calls for chest pain at the Medical Dispatch Centre, which is similar to the incidence of patients with ACS presenting to EDs . Until now, no decision-making algorithm has been validated. Effective identification of ACS has failed using Advanced Medical Priority Dispatch call prioritization . To date, no study has been published analysing the variables that influence the decision to send a medical team to the patient at the time of telephone triage.
We performed a multicentre observational study in various Medical Dispatch Centres in France, to assess emergency call triage for chest pain by describing the population according to the type of strategy chosen. In particular we analysed the characteristics of patients to whom an MICU was sent.
Methods
Study
We conducted a multicentre observational study, supported by the French Society of Cardiology (SFC) and the French Society of Emergency Medicine (SFMU), with an educational grant from Eli Lilly. Between 16 November and 13 December 2009, all emergency calls for non-traumatic chest pain received by the Medical Dispatch Centres in Bayonne, Lille, Melun, Paris and Toulouse were included prospectively in the study. Exclusion criteria were age < 18 years and traumatic chest pain.
In accordance with French law, our local ethics committee considered that patient consent could be waived for participation in this observational study. Data file collection and storage were approved by the ‘Comité consultatif sur le traitement de l’information en matière de recherche’ (CCTIRS) and the ‘Commission nationale informatique et liberté’ (CNIL).
Patients
The emergency system in France is based on the early intervention of physicians. Emergency call numbers (15 or 112) are routed to the closest Medical Dispatch Centre. These calls are first received by the auxiliary medical triage staff members, whose role is to open a file containing the telephone number of the caller and the geographical location of the patient, and to assess the degree of emergency; the call is then transmitted to an emergency physician.
After analysing the situation, the emergency physician chooses the type of response: medical advice by telephone; consultation by a general practitioner; dispatch of a paramedic team for hospitalization without a medical evaluation; or dispatch of an ambulance (MICU) staffed with at least one emergency physician.
All patients included in the study were identified by a study number. Five files were planned and completed according to the care process. For each patient there was a file for the telephone triage and a file for the 30-day follow-up; there were also files for the MICU medical staff, the ED and the cardiology department. Patient demographics and clinical data were recorded, as well as the estimation of the probability of ACS by the clinician on site and the decisions taken at each step.
Statistical analysis
Statistical analyses were conducted using Stata Statistical Software, release 10 (StataCorp LP, College Station, TX, USA). Statistics are reported as means with standard deviations, and medians with interquartile ranges for delays. Means were compared using Student’s t test for normally distributed data or the non-parametric two-sample Mann-Whitney rank-sum test for data not fitting the assumption of parametric testing. Percentages were compared using Pearson’s Chi 2 test and Fisher’s exact test. Univariate analyses were performed to identify factors associated with the dispatch of the MICU.
Results
During the study, a total of 1647 emergency calls for non-traumatic chest pain were regulated by the Medical Dispatch Centres of Bayonne ( n = 94; 5.7%), Lille ( n = 588; 35.7%), Melun ( n = 68; 4.1%), Paris ( n = 521; 31.6%) and Toulouse ( n = 376; 22.8%).
First decision by regulation triage
An MICU was sent to the patient in 51.9% of cases ( Fig. 1 ), paramedics were sent in 31.3% of cases, and a general practitioner was sent in 10.3% of cases; patients were only given medical advice by telephone, with no dispatch, in 6.6% of cases.
The patient was admitted directly to a cardiology ward or a coronary care unit in 33.3% of cases handled by the MICU emergency physician and in fewer than 3% of cases when a paramedic or general practitioner was dispatched.
A total of 942 patients (61% of total calls) were admitted to an ED, while 303 patients (20%) were admitted to a cardiology department. In 5% of all calls that resulted in a staff carer (general practitioner, paramedics or MICU) being sent, the final decision was to leave the patient at home.
Call characteristics
During the study period, 68.9% of calls occurred on a weekday; an MICU was more frequently sent as a result of these calls. A total of 43.1% of calls were received during office hours (between 08.00 and 20.00); an MICU was less frequently dispatched as a result of these calls ( Table 1 ).
| All calls | Calls followed by MICU dispatch | Calls not followed by MICU dispatch | P value | |
|---|---|---|---|---|
| Person calling (n = 1612) | ||||
| Patient | 491 (30.5) | 218 (26.2) | 273 (35.0) | |
| Family or bystander | 692 (42.9) | 342 (41.1) | 350 (44.9) | |
| Paramedics | 206 (12.8) | 27 (15.2) | 79 (10.1) | <0.00001 |
| General practitioner | 140 (8.7) | 102 (12.2) | 38 (4.9) | |
| Other | 83 (5.1) | 44 (5.3) | 39 (5.0) | |
| Interview (n = 1534) | ||||
| Patient interviewed directly | 609 (39.7) | 324 (40.6) | 285 (38.8) | 0.48 |
| Day of call (n = 1645) | ||||
| Weekday | 1133 (68.9) | 609 (71.8) | 524 (65.7) | <0.01 |
| Time of call (n = 1624) | ||||
| 08:00 to 20:00 | 700 (43.1) | 341 (40.8) | 359 (45.6) | 0.05 |
The patient called the Medical Dispatch Centre directly in only 30.5% of all cases. In the other cases, the telephone call was made by a paramedic, a relative, a bystander or a general practitioner. An MICU was sent more frequently when the call came from a paramedic or a general practitioner rather than from the patient, a relative or a bystander ( P < 0.00001).
The physician in charge of the triage was able to question the patient in fewer than 40% of cases. Interestingly, speaking directly with the patient was not associated with a statistical difference in terms of MICU dispatch.
Characteristics of the patients
The mean age was 56.1 years; patients who were sent an MICU were older (61.2 vs 50.5 years; P < 0.00001) and were more frequently men (61.8% vs 50.2% for women; P < 0.00001). Patients with previous coronary artery disease (CAD) were sent an MICU more frequently (66.8%; P < 0.00001), especially if they had a history of myocardial infarction (77.5%; P < 0.00001) ( Table 2 ).
| Total | MICU dispatch | No MICU dispatch | P value | |
|---|---|---|---|---|
| Age (years; n = 1608) | 56.1 ± 19.6 | 61.2 ± 16.7 | 50.5 ± 20.9 | <0.00001 |
| Male (n = 1603) | 900 (56.1) | 507 (56.3) | 393 (43.7) | <0.00001 |
| Female (n = 1603) | 703 (43.9) | 313 (44.5) | 390 (55.5) | |
| Delay between call and symptom onset (minutes; n = 670) | 60 [0–5787] | 56 [0–4662] | 62 [0–5787] | 0.06 |
| Previous CAD (n = 1400) | 566 (40.4) | 378 (66.8) | 188 (33.2) | <0.00001 |
| Previous MI (n = 1400) | 204 (14.6) | 158 (77.5) | 46 (22.5) | <0.00001 |
| CAD risk factors | <0.00001 | |||
| No CAD risk factor | 192 (15.3) | 54 (28.1) | 138 (71.9) | 0.0005 |
| One or more | 718 (57.3) | 425 (59.2) | 293 (40.8) | |
| Unknown | 343 (24.4) | 183 (53.4) | 160 (46.6) | |
| Diabetes ( n = 1105) | <0.00001 | |||
| Yes | 99 (9.0) | 66 (66.7) | 33 (33.3) | |
| No | 516 (46.7) | 237 (45.9) | 279 (54.1) | |
| Unknown | 490 (44.3) | 256 (52.2) | 234 (47.8) | |
| Hyperlipidaemia ( n = 1121) | 0.0001 | |||
| Yes | 158 (14.1) | 105 (66.5) | 53 (33.5) | |
| No | 472 (42.1) | 217 (45.9) | 255 (54.1) | |
| Unknown | 491 (43.8) | 256 (52.1) | 235 (47.9) | |
| Smoker ( n = 1126) | <0.00001 | |||
| Yes | 288 (25.6) | 170 (59.0) | 118 (41.0) | |
| No | 419 (37.2) | 180 (42.9) | 239 (57.1) | |
| Unknown | 419 (37.2) | 223 (53.2) | 196 (46.8) | |
| Hypertension ( n = 1140) | <0.00001 | |||
| Yes | 243 (21.3) | 170 (69.9) | 73 (30.1) | |
| No | 432 (37.9) | 180 (41.7) | 252 (58.3) | |
| Unknown | 465 (40.8) | 239 (51.4) | 226 (48.6) | |
| Family history of CAD ( n = 1048) | <0.00001 | |||
| Yes | 90 (8.6) | 58 (64.4) | 32 (35.6) | |
| No | 416 (39.7) | 166 (39.9) | 250 (60.1) | |
| Unknown | 542 (51.7) | 297 (54.8) | 245 (45.2) | |
| Treatment | ||||
| Aspirin ( n = 1063) | <0.00001 | |||
| Yes | 129 (12.1) | 97 (75.2) | 32 (24.8) | |
| No | 459 (43.2) | 189 (41.2) | 270 (58.8) | |
| Unknown | 475 (44.7) | 245 (51.6) | 230 (48.4) | |
| Clopidogrel ( n = 1042) | <0.00001 | |||
| Yes | 77 (7.4) | 63 (81.8) | 14 (18.2) | |
| No | 485 (46.5) | 207 (42.7) | 278 (57.3) | |
| Unknown | 480 (46.1) | 249 (51.8) | 231 (48.2) | |
| Statin ( n = 1037) | <0.00001 | |||
| Yes | 104 (10.0) | 78 (75.0) | 26 (25.0) | |
| No | 449 (43.3) | 181 (40.3) | 268 (59.7) | |
| Unknown | 484 (46.7) | 253 (52.3) | 231 (447.7) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
