Acute myocardial infarction (AMI), including both ST elevation myocardial infarction (STEMI) and non-STEMI, is a common cardiac emergency condition and is associated with a high risk of serious morbidities and mortality [1]. Although the incidence of AMI has decreased significantly in recent decades [2], AMI is still a major global health burden, with AMI associated with an increased number of years with disability [3]. In Korea, AMI is the second leading cause of death, with 60 deaths due to AMI among every 100,000 people [4]. AMI has also been demonstrated to have a high economic cost [5].

Shortening the time delay in diagnosis and treatment of AMI is crucial for achieving optimal clinical outcomes [6,7]. The American College of Cardiology and American Heart Association proposed an ideal time window for intervention in STEMI patients [8], and recommended that suspected AMI patients receive emergency medical service (EMS) to shorten the time delay [9]. The length of stay (LOS) in the emergency department (ED), as an index of time delay, was found to be associated with clinical outcomes in time-sensitive conditions, including AMI [10,11]. ED routine issues, including diagnostic difficulties, have been demonstrated to contribute to in-hospital delays [12].

EMS management of AMI is important because early assessment, treatment, and expedited communication with the receiving hospital are associated with improved patient outcomes [13]. The ED bypass program based on early recognition of AMI by EMS was demonstrated to have a significant association with survival outcome in a STEMI cohort [14]. Additionally, Bright et al. [15] reported high accuracy among EMS personnel in the prehospital diagnosis of AMI. Prehospital 12-lead electrocardiography (ECG) by EMS personnel is strongly recommended and implemented worldwide [16]. Hutchison et al. [17] reported that the time to intervention was greatly decreased with clinical pathway activation based on prehospital 12-lead ECG findings. Prehospital 12-lead ECG was also found to help the ED achieve the ideal time window for intervention and was therefore associated with a lower mortality rate in AMI patients [18].

The effect of prehospital recognition on the ED process has not been well evaluated for patients with AMI. We hypothesized that prehospital recognition of AMI would reduce the ED LOS and that prehospital 12-lead ECG would strengthen this association. The purpose of this study was therefore to evaluate the association between prehospital recognition with or without 12-lead ECG and ED LOS in EMS-transported AMI patients.

In EMS-transported AMI patients admitted to the ED, prehospital recognition was found to be associated with a short ED LOS. There was no statistically significant difference according to prehospital 12-lead ECG. IP LOS and survival to discharge had no significant association with prehospital recognition with or without 12-lead ECG. The same trends were sustained and statistically significant in patients admitted to the ICU and in non-STEMI patients. Careful prehospital evaluation for AMI by EMS personnel should be emphasized to improve the clinical course of these patients.

Prehospital recognition of AMI is associated with a higher probability of notification of the receiving hospital. This should ideally result in activation of the AMI practice process in the hospital, which could shorten ED LOS. We found that there were more AMI patients with a short ED LOS (under 4 hours) when the hospital received notification of an AMI during the prehospital phase, even after excluding STEMI cases, then when no prenotification was received (42.1% vs 35.0%, respectively).

In the prehospital phase of AMI, patients present with more pronounced symptoms, and recognition of these symptoms can optimize patient assessment and triage [25]. With sufficient experience and focused education, EMS personnel can identify highrisk features that indicate the possibility of acute cardiovascular disease [26]. A detailed history taken at the scene can provide meaningful information to ED staff, who can use this information to direct the assessments performed, thus accelerating the clinical process. However, the delivery of misleading information may interfere with the direction of the assessments performed and thus the clinical course of the patient. Considering that communication barriers extend ED LOS, the first impression of EMS personnel can influence ED staff [27].

Additionally, the presence or absence of a sublingual nitroglycerin effect on chest pain is an important diagnostic predictor [28]. EMS personnel can evaluate this effect more clearly at the scene when chest pain has not yet subsided, as chest pain often improves spontaneously and changes to ambiguous symptoms at ED arrival. In such situations, ED LOS can be prolonged, as the application of proper diagnostic algorithms to reach the right diagnosis is limited [29]. In the prehospital recognition group, the number of sublingual nitroglycerin administrations was seven to eight times higher than in the other groups (Table 1).

In this study, prehospital recognition of AMI by EMS personnel was documented in the official registry. There is no detailed protocol about when EMS personnel should consider the patient to have suspected acute cardiovascular disease, but typical questions should be asked to complete the cardiovascular registry questions, such as the location and intensity of pain, whether the pain is radiating, and aggravating and relieving factors. The National Fire Agency provides a routine continuing education course for EMS personnel, including information on the presentations of acute critical illnesses such as AMI. EMS personnel are likely able to differentiate the possibility of AMI on the basis of complaints, past medical history, and vital signs. Sixty percent of the total patients had chest pain, but 40% of them were not recognized as having acute cardiovascular disease (Table 1), so we hypothesize that unmeasured factors affected EMS personnel assessments. Recognition of the overlooked AMI population is crucial to improve the clinical process.

ED LOS has been considered an indicator of the ED care process in previous studies [30-32]. Unlike cardiac arrest or trauma, in which prehospital resuscitation determines the status of patients at the ED entrance level, the role of EMS personnel in STEMI usually focuses on 12-lead electrocardiogram interpretation and ED bypass [8]. Considering that overcrowding and limited resources are major challenges to ED AMI diagnostic performance [33], detailed and accurate prehospital assessments are helpful from the beginning of ED care. Additionally, prenotification of receiving hospitals affects the ED care process. In the study group with prehospital recognition, a higher proportion of the patients’ hospitals were notified before ED arrival, even when no 12-lead ECG results were available.

Condition severity has been demonstrated to be associated with ED LOS [34]. Sensitivity analysis of patients admitted to the ICU was performed to correct for unmeasured biases, such as a delay in admission based on nonmedical issues. Similar trends with statistical significance were found regarding the outcomes of patients admitted to the ICU and non-STEMI patients. Therefore, we highly recommend that EMS personnel assess the probability of AMI at the scene, especially in patients with indications of high condition severity, even without evidence of STEMI.

Our results indicate that prehospital EMS assessment could be beneficial for the ED care process, but the effect of prehospital recognition on shortening ED LOS is not clear. The effects of implementation of more sensitive EMS tools to screen patients with suspected AMI and analyses of the effects of these tools on the ED care process are needed. Considering that over half of AMIs were not recognized at the prehospital level, the creation and use of advanced patient-assessment protocols at the scene could have clinical benefits.

This study had several limitations. First, prehospital recognition was dependent on individual experience or competency. There was no standardized EMS protocol to evaluate patients with suspected AMI based on the patient’s history, which is a significant limitation. Second, we could not confirm whether prehospital information was actually delivered to ED physicians. Although ED triage nurses write a triage note containing information provided by EMS personnel, it is entirely up to the physician as to whether they refer to this note or not. Next, information on the time of diagnosis, time to first 12-lead ECG at the ED, or time to consult cardiologist would provide more accurate information on the effect of prehospital recognition, but this information was not available in the databases. As ED LOS is a surrogate marker, we adjusted for factors affecting the ED LOS as much as possible. However, there were also unadjusted confounders such as administrative delays and preparation time in the ward or ICU. We included time and season of ED visit in the main analysis to adjust for variation, but variation remains a significant limitation of the study. However, when further sensitivity analysis excluding extremely delayed cases (10, 12, and 15 hours) was conducted, we observed the same trends (Supplementary Table 1.). Fourth, even though chief complaints and vital signs were adjusted for, detailed condition severity or characteristics were not analyzed. Generally, the ED LOS of patients with minor AMI or late presenters is prolonged due to ambiguous histories. Furthermore, we could not evaluate this association in the STEMI cohort as the majority of these patients stayed in the ED for less than 4 hours. We hypothesized that shortening of ED LOS would affect the treatment outcomes of AMI, which would shorten IP LOS and increase survival. There was no significant association between prehospital recognition and IP LOS and survival, suggesting that a delay in the ED might be tolerated in non-severe patients. Fifth, individual variations in practice could affect the results, such as physicians who always check for cardiovascular disease, cardiologists who respond quickly, and physicians who tend to skip performing laboratory tests in patients without certain needs. Compliance with hospitalization may also have affected our findings. Since the design of this study was observational, unmeasured biases may have been present. Last, the results of this study cannot be generalized because this study was conducted using data from three academic tertiary emergency departments and a basic-to-intermediate service level EMS system.

In summary, prehospital recognition of AMI by EMS personnel was found to be associated with a short ED LOS in EMS-transported AMI patients. Focused assessment at the scene can improve the clinical outcomes of AMI patients by decreasing their ED LOS.