The Emergency Medical Care of Patients With Acute Myocardial Infarction

Results from the First Medical Contact Study in the context of the Berlin Myocardial Infarction Registry

Background: Optimizing the emergency medical care chain might shorten the time to treatment of patients with ST-elevation myocardial infarction (STEMI). The initial care by a physician, and, in particular, correct ECG interpretation, are critically important factors.

Methods: From 1999 onward, data on the care of patients with myocardial infarction have been recorded and analyzed in the Berlin Myocardial Infarction Registry. In the First Medical Contact Study, data on initial emergency medical care were obtained on 1038 patients who had been initially treated by emergency physicians in 2012. Their pre-hospital ECGs were re-evaluated in a blinded fashion according to the criteria of the European Society of Cardiology.

Results: The retrospective re-evaluation of pre-hospital ECGs revealed that 756 of the 1038 patients had sustained a STEMI. The emergency physicians had correctly diagnosed STEMI in 472 patients (62.4%), and they had correctly diagnosed ventricular fibrillation in 85 patients (11.2%); in 199 patients (26.3%), the ECG interpretation was unclear. The pre-hospital ECG interpretation was significantly associated with the site of initial hospitalization and the ensuing times to treatment. In particular, the time from hospital admission to cardiac catheterization was longer in patients with an unclear initial ECG interpretation than in those with correctly diagnosed STEMI (121 [54; 705] vs. 36 [19; 60] minutes, p <0.001). After multivariate adjustment, this corresponded to a hazard ratio* of 2.67 [2.21; 3.24].

Conclusion: Pre-hospital ECG interpretation in patients with STEMI was a trigger factor with a major influence on the time to treatment in the hospital. The considerable percentage of pre-hospital ECGs whose interpretation was unclear implies that there is much room for improvement.

Coronary heart disease (CHD) is the leading cause of death world-wide, cited by the World Health Organization (WHO) as causing 13.2% of all deaths (1). The number of deaths due to CHD in Germany in 2013 is reported at 125 220, corresponding to 14% of all deaths—a similar percentage (2).

The heart muscle can tolerate ischemia only briefly. This is why prompt diagnosis and rapid reperfusion are central to limiting early infarction-related mortality (3). Measures aimed at rapid reperfusion (4) also contribute to long-term improvement of myocardial function and help to prevent systolic heart failure and significantly increased mortality in the longer term (5). The European guidelines allow <60 minutes (6) for the diagnosis and treatment of acute ST-elevation myocardial infarction (STEMI) and <120 minutes for non-STEMI (NSTEMI) (7) in patients with shock, less than 24 hours in high-risk patients, and less than 72 hours in intermediate-risk patients.

When patients with STEMI are attended by a physician^* accompanying the emergency medical services, acute percutaneous coronary intervention is initiated more quickly (8, 9). In addition to optimizing hospital procedures, therefore, it also seems valuable to study the prehospital processes and the triggering initial ECG carried out by emergency services responding to a myocardial infarction. The Berlin Myocardial Infarction Registry (BHIR, Berliner Herzinfarkt-Register) initiated the First Medical Contact Study (FMC) reported on here, which evaluated how the ECG diagnosis is established and also how the subsequent times to treatment are related to this diagnosis in myocardial infarction patients in Berlin. The study investigated how, in the day-to-day provision of care in a large city, the medical care received by patients with myocardial infarction from the emergency services influences their treatment in hospital. The study aimed to produce results that would allow the identification of potential improvements in care between the emergency services and the hospital.

Methods

Patient selection

Since 1999, data on the in-hospital care of patients with myocardial infarction have been collected at the Berlin Myocardial Infarction Registry (BHIR). The objective of the BHIR is quality assurance of the hospital care of patients in Berlin with myocardial infarction, based on the recommendations of international guidelines (10).

The present study included all patients who were hospitalized with a myocardial infarction in Berlin from 1 January to 31 December 2012, who were first attended by the emergency medical services accompanied by an emergency physician and who reached hospital within 24 hours of the onset of symptoms. During the study period, 17 (out of a total of 20) Berlin hospitals with and two hospitals without cardiac catheterization laboratories took part. In Berlin the emergency services are centrally coordinated by the Berlin Fire Department, and therefore all 17 emergency medical support units participated in the data collection.

Data collection (BHIR source data)

The BHIR collects demographic data and information on cardiovascular and clinical risk factors, procedural aspects of coronary interventions and concomitant medication, together with in-hospital mortality, all in de facto anonymized form. Time points recorded in the process of infarct treatment are the exact times (to the minute) of infarction onset, hospital admission, catheter insertion, and passage of the guidewire into the infarct vessel. The data are recorded in a standardized manner, on either electronic or paper-based versions of the BHIR data collection form, and sent to the BHIR central study office. Regular monitoring and peer review with on-site visits ensures that patients are enrolled consecutively. Data validation is also carried out, by taking a random 10% sample of the cases from each hospital and comparing the contents of the data collection forms with patient records. All input data are examined for errors and hospitals are asked for any missing information. The BHIR was approved by the data protection officer for Berlin and follows the ethical guidelines of the Helsinki Declaration.

Additional data collection for the FMC study

Data on emergency medical care that were not recorded on the basic BHIR form had to be collected separately for the study. These data were: the exact time (to the minute) at which the patient had called the emergency services, the findings of the initial ECG carried out by the physician accompanying the emergency services, the emergency physician’s diagnosis, and the hospital department to which the patient was admitted (emergency department, intensive care unit, or catheterization laboratory). The time at which the emergency physician reached the patient was not noted in the Berlin emergency services records during the study period, for reasons relating to data collection. In all cases, therefore, 13 minutes were added to the time of callout in order to define the time of first medical contact. These 13 minutes correspond to the average emergency physician response time in Berlin in 2012. Other relevant time intervals compared were (in minutes) “first medical contact to hospital admission,” “hospital admission to catheter insertion for percutaneous coronary intervention,” and “percutaneous coronary catheter insertion to wire passage into the infarct vessel.”

The additional data collected for the FMC study were linked on-site in the hospitals to the data from the BHIR basic form and then passed to the BHIR in de facto anonymized form.

Blinded validation of ECG findings

To evaluate the interpretation of ECG findings by emergency physicians, the initial emergency ECGs were independently reinterpreted by three experienced interventional cardiologists (hereafter referred to as the “CoreLab interpretation”) who were blinded to the original interpretation. This reinterpretation determined whether or not the case could be unequivocally diagnosed as STEMI based on the ECG. The determination was based on the age- and sex-specific criteria for the definition of STEMI laid down in the European Society of Cardiology guidelines for the management of acute STEMI (6). STEMI patients thus identified from the emergency ECGs formed the basis of the further analysis, during which the validated STEMI diagnoses were compared with the emergency physician ECG findings. The two findings were taken to be congruent if ST elevation was diagnosed by both the CoreLab and the emergency physician. Uncertain findings included those patients given a clear diagnosis of STEMI by CoreLab but diagnosed by the emergency physician as having no ST elevation but a variety of other ECG findings (e.g., ST depression, T-wave inversion, arrhythmias, AV block) or who had no documented diagnostic findings. A third group was made up of patients with ventricular fibrillation.

Hence, the emergency diagnoses were grouped into three categories:

• “Clear diagnosis of STEMI”
• Ventricular fibrillation
• “Uncertain diagnosis” with no mention of STEMI by the emergency physician

Statistics

Data are presented as percentages and times to treatment as medians and interquartile ranges. Nominally scaled variables were compared using the χ² test and non-normally distributed variables using the Mann–Whitney U test or Kruskal–Wallis test. Normal distribution was tested for using the Shapiro–Wilk test. For all times to treatment, there was a non-normal distribution pattern with p-values <0.001. The cumulative number of patients with different times to treatment was compared in dependence on the ECG diagnosis and on the admitting hospital department using Kaplan–Meier analysis and log rank testing. A p-value of <0.05 was regarded as significant. The Table and Figures 1 and 2 show 95% confidence intervals [95% CI] for percentages together with the Kaplan–Meier curves.

Table

Basic characteristics of patients classified by the ECG CoreLab as having STEMI.

Cox regression analysis was used for multivariate analysis of the association between age, sex, renal failure, heart failure, previous infarction, percutaneous coronary intervention or bypass operation, time from onset of symptoms to calling the emergency services, and prehospital ECG diagnosis.

Results

Patient population

This study was based on 1157 patients who experienced a myocardial infarction in Berlin in 2012, received their first medical care from an emergency physician, and were registered by the BHIR. Of these, 119 patients could not be included in the analysis because of lack of documentation by the emergency physician, or their initial ECGs were missing or no longer readable. Blinded ECG reinterpretation by the CoreLab was carried out for 1038 patients. The CoreLab diagnosis of these 1038 patients showed that 756 patients (72.8%) had a prehospital ECG showing significant ST elevation. Another 237 patients (22.8%) showed no significant elevation, and a small subgroup of 45 patients (4.3%) showed no ST elevation in their prehospital ECG, but did develop significant ST elevations during their stay in hospital.

Of the 756 patients diagnosed unequivocally by CoreLab as having identifiable STEMI on their prehospital ECGs, the emergency physicians had diagnosed ST elevation in 472 (62.4%), ventricular fibrillation in 85 (11.2%), and absence of clear ST elevation in 199 (26.3%), even though these patients did show ST elevation on the ECG. Instead of a STEMI diagnosis, possible left bundle branch block was diagnosed in 11 (5.5%), “no ST elevation” in 32 (16.1%), ST depression in 37 (18.6%), another diagnosis in 90 (45.2%), and no ECG diagnosis in 29 (15%). All 756 patients had been transferred for percutaneous coronary intervention.

The demographic data of the CoreLab-defined STEMI patients and the subgroups defined on the basis of the emergency physicians’ ECG findings are shown in the Table.

Relationship between the emergency physician’s ECG diagnosis and time to treatment

For patients with a clear and those with an uncertain emergency diagnosis of STEMI, there was no difference in time from first medical contact to hospital admission (34 [95% CI: 26; 42] versus 33 [28; 41] min, p = 0.869). Patients with prehospital ventricular fibrillation reached hospital after a longer delay (56 [43; 70] min; p <0.001). This difference remained even after multivariate adjustment (hazard ratio = 2.51 [1.93; 3.26]). Patients with clearly identified STEMI underwent cardiac catheterization significantly sooner after hospital admission (36 [19; 60] min) than did patients with an uncertain emergency diagnosis (121 [54; 705] min; p <0.001) or those with ventricular fibrillation (45 [25; 79] min; p = 0.027). The difference relating to uncertain versus clear prehospital ECG diagnosis remained even after multivariate adjustment (hazard ratio = 2.67; [2.21; 3.24]).

The time from insertion of the cardiac catheter to passage of the guidewire into the infarct vessel was slightly longer in patients with an uncertain STEMI diagnosis (21 [13; 32] min) than those with a clear STEMI diagnosis (17 [10; 23] min; p <0.001). Here, too, the difference between uncertain versus clear prehospital ECG diagnosis remained after multivariate adjustment (hazard ratio = 1.31; [1.09; 1.57]). Patients with ventricular fibrillation likewise had a short time to wire passage (16 [11; 23] min; p = 0.939). At 91 [69; 116] minutes, the total time from first medical contact to wire passage into the infarct vessel was shorter in patients with a clear prehospital STEMI diagnosis than it was in patients with an uncertain emergency diagnosis (186 [109; 757] minutes; p <0.001; hazard ratio after adjustment = 2.71; [2.24; 3.28]) and in those with ventricular fibrillation (132 [92; 167] minutes; p <0.001; hazard ratio after adjustment = 1.67; [1.29; 2.17]). The variations in time to treatment depending on prehospital ECG diagnosis can be seen in the Kaplan–Meier analysis shown in Figure 1.

Figure 1

Kaplan–Meier analysis

Relationship between the emergency physician’s ECG diagnosis and the admitting hospital department

Among patients with a clear prehospital STEMI diagnosis, 24% went directly to the catheterization lab, 27% were admitted straight to intensive care, and 48% received their initial treatment in the emergency department. Against this, 77% of patients whose STEMI was not identified by the emergency physician were initially treated in the emergency department. Patients with prehospital ventricular fibrillation most often (46%) received initial treatment in the intensive care unit, 20% in the catheterization lab, and 34% in the emergency admissions department (Figure 2).

Figure 2

Association between emergency ECG diagnosis

Relationship between admitting hospital department and time to treatment

Patients who were first admitted to the hospital’s emergency department reached the catheterization lab later (65 [38; 180] min) than those who were first admitted to the intensive care unit (45 [30; 75] min; p <0.001) or directly to the catheterization lab (14 [9; 24] min; p <0.001). The time from first medical contact to hospital admission differed little between admitting departments, being slightly shorter, at 34 minutes [27; 42], for patients admitted to the emergency department compared to those first admitted into intensive care (38 [29; 52] min; p <0.001) or the cardiac catheterization lab (38 [26; 48] min; p <0.064). At 19 [13; 27] minutes, the time from catheter insertion to wire passage was a few minutes longer for patients admitted to the emergency department than for those admitted straight to the intensive care unit (15 [9; 24] min; p <0.001) or the catheterization lab (16 [10; 23] min; p = 0.001). The Kaplan–Meier analysis of the cumulative number of patients per time interval in dependence on admitting hospital department can be seen in the eFigure.

eFigure

Kaplan–Meier analysis

Discussion

The results of this registry analysis of the prehospital treatment of patients with myocardial infarction in Berlin show that correct diagnosis of an acute STEMI by an emergency physician is associated with rapid treatment in hospital. STEMI patients with prehospital ventricular fibrillation, after a longer prehospital emergency treatment phase, also received rapid invasive care in hospital. An uncertain ECG diagnosis by the emergency physician, on the other hand, was associated with a longer time to treatment in the hospital. For patients with an uncertain prehospital STEMI diagnosis that was later clearly confirmed in the CoreLab, the time from hospital admission to percutaneous coronary intervention was more than three times that of patients with a clearly identified STEMI. Incorrectly interpreted ECGs led to treatment times that were way beyond the 90 minutes from first medical contact to percutaneous coronary intervention demanded by the guidelines (6), and also way beyond the limit of 60 min (6) allowed for the time from hospital admission to percutaneous coronary intervention. Another difference dependent on the ECG interpretation was that patients with an unidentified STEMI more often received their initial hospital treatment in the emergency department (and not in the intensive care unit or the catheterization lab). In turn, the emergency department as admitting department was associated with significantly delayed invasive diagnostic intervention. Data collected earlier in the USA suggest that the relative risk of infarct-related hospital mortality increases by 10% for every 30-minute increase in time to treatment (3), and that the advantage of percutaneous coronary intervention compared to thrombolysis disappears with longer times to treatment (11). Several studies have shown that prehospital ECG diagnosis reduces the time to reperfusion (12, 13). The importance of an early, correct ECG diagnosis is accordingly highlighted in a position paper by the American Heart Association (14).

In the prototypical individual case, STEMI can be diagnosed by a glance at the ECG. The results presented here, however, point up the fact that in some cases a STEMI is not identified despite the presence of clear morphological criteria. Uncertainty can arise because ST elevations are often small in the traces from limb leads. ST segment elevations in V2 and V3 can also cause confusion. Age- and sex-dependent normal variants in these leads can include J-point elevation up to 0.25 mV in men up to the age of 40 years, up to 0.2 mV in men over the age of 40, and up to 0.15 mV in women (15, 16). In the other leads, however, an ST elevation of just 0.1 mV must be regarded as pathological. Bundle branch block patterns and the constellation of findings representing early repolarization can also make it difficult to interpret the ECG.

Limitations

• The CoreLab cardiologists re-evaluating the ECGs did not know whether STEMI had been diagnosed at the time of the emergency ECG, but they did know that infarction had been diagnosed at some point, and this made interpretation easier for them than for the emergency physicians. It can not be ruled out that without this knowledge, the CoreLab cardiologists would also have been unable to give a clear diagnosis of STEMI in the case of one or two STEMI patients. However, according to the guidelines ST elevation is defined solely on the basis of of the ECG (6), making this less likely.
• Since only cases with a prehospital time <24 hours were recorded by the BHIR, cases where diagnosis was delayed for longer because the emergency services were alerted later are under-represented in this study.
• Information given by patients regarding when their symptoms began can be unreliable and inconsistent, and this can affect the recorded interval from onset of symptoms to calling the emergency services.
• The occurrence of false-positive ECG diagnoses could not be investigated in this study, which included only patients who did have an infarction.

Future perspectives

The data presented here allow the conclusion that the treatment of patients with STEMI in Berlin could be further optimized by increasing the level of certainty of prehospital diagnoses and improving the interpretation of ECGs by emergency physicians. Although regional differences in the organization of treatment for myocardial infarction in Germany mean that these data cannot be directly extrapolated to other regions, the conclusions should also prompt efforts to ensure accurate interpretation of prehospital ECGs elsewhere. Implications for action to improve the management of myocardial infarction in Berlin are already being discussed between the hospitals and the prehospital emergency physicians. So far, competence in interpreting emergency ECGs has only implicitly been included in the training content set out by the regulations of the Berlin Medical Association (Berliner Ärztekammer) for a specialist qualification in emergency medicine (17). A greater emphasis on ECG competence in the training for emergency medicine, and structured implementation of appropriate education and refresher programs by hospital cardiologists in conjunction with regional medical associations, could be steps on the way to improvement. Teletransmission of difficult-to-classify ECGs, for which the technology already exists, offers another possible way forward towards optimizing the prehospital interpretation of ECGs.

Conflict of interest statement
The authors declare that no conflict of interest exists.

Translated from the original German by Kersti Wagstaff, M.A.

Manuscript received on 10 October 2015, revised version accepted on
29 February 2016.

Corresponding author:
PD Dr. med. Martin Stockburger
Medizinische Klinik I
Ketziner Str. 21
14641 Nauen, Germany

@Supplementary material
eFigure:
www.aerzteblatt-international.de/16m0497

* In Germany, the emergency services are frequently accompanied by a qualified emergency physician. In this article, that term is used exlcusively referring to this person, not to a physician working in a hospital emergency department

References

1. WHO Fact sheet N8310, updated May 2014, www.who.int/mediacentre/factsheets/fs310/en/index.html (last accessed on 10 October 2016).

2. Statistisches Bundesamt: Zahlen und Fakten, www.destatis.de/DE/ZahlenFakten/GesellschaftStaat/Gesundheit/Todesursachen/Tabellen/HaeufigsteTodesursachen.html (last accessed on 10 October 2016).

3. De Luca G, Suryapranat H, Ottervanger JP, Antman EM: Time delay to treatment and mortality in primary angioplasty for acute myocardial infarction: every minute of delay counts. Circulation 2004; 109: 1223–5

4. Bradley EH, Herrin J, Wang Y, et al.: Strategies for reducing the door-to-balloon time in acute myocardial infarction. N Engl J Med 2006; 355: 2308–20

5. Global Group in Chronic Heart Failure (MAGGIC): The survival of patients with heart failure with preserved or reduced left ventricular ejection fraction: an individual patient data meta-analysis. Eur Heart J 2012; 33: 1750–7

6. Steg PG, James SK, Atar D, et al.: The Task Force on the management of ST-segment elevation acute myocardial infarction of the European Society of Cardiology (ESC). ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J 2012; 33: 2569–619

7. Roffi M, Patrono C, Collet JP, et al.: 015 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: Task Force for the Management of Acute Coronary Syndromes in Patients Presenting without Persistent ST-Segment Elevation of the European Society of Cardiology (ESC). Eur Heart J 2016; 37: 267–315

8. Welsh RC, Chang W, Goldstein P, et al.: Time to treatment and the impact of a physician on pre-hospital management of acute ST elevation myocardial infarction: insights from the ASSENT-3 PLUS trial. Heart 2005; 91: 1400–06

9. Maier B, Behrens S, Graf-Bothe C, et al.: Time of admission, quality of PCI care, and outcome of patients with ST-elevation myocardial infarction. Clin Res Cardiol 2010; 99: 565–72

10. Maier B, Thimme W, Schoeller R, Fried A, Behrens S, Theres H: Berlin myocardial infarction registry improved therapy and outcome for patients with acute myocardial infarction-data of the Berlin Myocardial Infarction Registry from 1999 to 2004. Int J Cardiol 2008; 130: 211–9

11. Pinto DS, Kirtane AJ, Nallamothu BK, et al.: Hospital delays in reperfusion for ST-elevation myocardial infarction: implications when selecting a reperfusion strategy. Circulation 2006; 114: 2019–25

12. Curtis JP, Portnay EL, Wang Y, et al.: The pre-hospital electrocardiogram and time to reperfusion in patients with acute myocardial infarction, 2000–2002: findings from the National Registry of Myocardial Infarction-4. J Am Coll Cardiol 2006; 47: 1544–52

13. Brown JP, Mahmud E, Dunford JV, Ben-Yehuda O: Effect of prehospital 12-lead electrocardiogram on activation of the cardiac catheterization laboratory and door-to-balloon time in ST-segment elevation acute myocardial infarction. Am J Cardiol 2008; 101: 158–61

14. Ting HH, Krumholz HM, Bradley EH, et al.: Implementation and integration of pre-hospital ECGs into systems of care for acute coronary syndrome: a scientific statement from the American Heart Association Interdisciplinary Council on Quality of Care and Outcomes Research, Emergency Cardiovascular Care Committee, Council on Cardiovascular Nursing, and Council on Clinical Cardiology. Circulation 2008; 118: 1066–79

15. Mcfarlane PW: Age, sex, and the ST amplitude in health and disease. J Electrocardiol 2001; 34: 35–41

16. Thygesen KT, Alpert JS, Jaffe AS, Simoons ML, Chaitman BR: White HD, the Writing Group on behalf of the Joint ESC/ACCF/AHA/WHF Task Force for the universal definition of myocardial infarction. Third universal definition of myocardial infarction. Eur Heart J 2012; 33: 2551–67

17. Weiterbildungsordnung der Ärztekammer Berlin: www.aerztekammer-berlin.de/10arzt/15_Weiterbildung/10Weiterbildungsordung/ 00_WbO_2004_inkl_1_bis_10_Nachtrag.pdf (last accessed on 10 October 2016).