Cardiovascular diseases are the leading cause of death in the world and were responsible for 32% of deaths on the planet by 2015.1 In Brazil, as reported by the Mortality Information System - SIM - ischemic heart disease is the leading cause of death,2 generating very high medical and socioeconomic expenditures to the public health system.3 In the United States of America, cardiovascular diseases have generated direct and indirect costs of more than 200 billion dollars to the health system with estimates of about US$ 220 billion in expenditures by 2020.4

It is estimated that about 5% of patients with chest pain and myocardial infarction are mistakenly discharged without a confirmed diagnosis in the emergency room.5 Early diagnosis and treatment of ischemic heart diseases are critical for the reduction of morbidity and mortality, hospitalization time and optimization of costs in emergency care.3 Proper management and follow-up of these patients has been a priority for the public health system.6

The 24-hour Emergency Care Units (UPA 24h) emerged as part of Brazil's Emergency Policy program, which structures and organizes emergency care in the country. The UPA 24h are structures of intermediate complexity between the primary health care units and the chain of hospitals. In the state of Rio de Janeiro, the UPA 24h were implemented in 2007. According to the state's demand profile, there was a need for cardiology support to general practitioners working at the UPA 24h units in the care of patients with chest pain. In 2009, the Cardiology Consultancy Nucleus (CCN) was set up for the purposes of providing specialized remote assistance in the medical care for patients with cardiac disorders at these units.

The objective of this study is to evaluate the support of telecardiology in the diagnostic qualification of chest pain in twenty-two 24-hour Emergency Care Units (UPA 24h) in the state of Rio de Janeiro.

The mean age found in the group of patients admitted with chest pain was 59.20 ± 11.33 years, with a predominance of the male gender (59.32%, n = 3,450).

The average length of stay in the UPA 24h was three days.

The main comorbidities reported were systemic arterial hypertension (49.87%; n = 2,900), diabetes mellitus (10.09%; n = 587), smoking (14.79%; n = 860), previous coronary artery disease (7.60%; n = 442), dyslipidemia (8.82%; n = 513), dilated cardiomyopathy (3.73%; n = 217), previous stroke (2.13%; n = 124), family history of coronary artery disease (2.13%; n = 124) and obesity (0.84%; n = 49).

Of the 5,816 patients admitted for chest pain supposedly due to cardiac disorders and evaluated by the telecardiologist, there was diagnostic agreement with the general practitioner from the UPA 24h in 72.61% of the cases (4,223 patients) and diagnostic disagreement in 27.39% of the cases (1,593 patients).

In the group in which there was diagnostic disagreement, the general practitioner classified patients as follows: 1,477 (92.72%) non-ST-segment elevation acute myocardial infarction (NSTEMI), 74 (4.64%) ST segment elevation acute myocardial infarction (STEMI), 40 (2.52%) acute pulmonary edema (APE) and 2 (0.12%) tachyarrhythmia. After evaluation by telecardiology, the diagnoses were: 385 (24.17%) unstable angina (UA), 289 (18.14%) congestive heart failure (CHF), 212 (13.31%) APE, 174 (10.92%) STEMI, 152 (9.54%) hypertensive emergency (HE), 113 (7.09%) acute chronic renal failure (ACRF), 89 (5.59%) non-STEMI, 89 (5.59%) pneumonia, 39 (2.45%) sepsis, 26 (1.63%) myopericarditis, 20 (1.26%) tachyarrhythmia, 5 (0.31%) orovalvar disease (OVD). (Table 1).

The general practitioner had requested intensive care admissions for the 1,593 patients identified by telecardiology as discordant diagnoses. After a careful evaluation by telecardiology, only 338 patients really needed to be transferred, and the great majority, 1,178 patients (73.94%), were discharged after clinical stabilization. Sixty-two (3.9%) patients died and 15 (0.95%) had an unknown outcome (Table 2).

The main points of diagnostic disagreement identified by telecardiology are described in the following table.

By analyzing the results found in this study, we observed the importance of telecardiology in remote support to the general practitioner in the differential diagnosis of chest pain at the emergency care, directly influencing proper diagnosis, treatment and referral of the patients.

The importance of cardiological remote support for diagnosis and clinical management has been described by other authors7^,8 and Molinari et al.,9 that evaluated 456 patients with chest pain and/or other symptoms suggestive of an acute cardiac event, comparing the initial diagnosis of the general practitioner with the final diagnosis of the cardiologist. In this study,9 two points stand out in our findings: the first one is the high percentage of diagnostic disagreement 31% (140 patients)9 compared to 27.39% (1,593 patients) in our study, demonstrating the need of the remote specialist; and the second one is the reduction of unnecessary admissions that was 63%9 compared to 73.9% (1,593 patients) in this study, thus representing effective savings in terms of unnecessary costly hospitalizations. Similar studies10^-13 related to cost reduction have also proved the effectiveness of telecardiology in primary care.

In our study, the main points of diagnostic disagreement in patients admitted with chest pain were abnormalities on the electrocardiogram, showing ST-segment elevation not recognized at admission, similar to those described by Brunetti et al.,14 high CPK and MBK values ​​with negative troponin defined as non-ST-elevation acute myocardial infarction and other causes of chest pain with positive troponin related and not related to cardiac disorders but not characterizing acute myocardial ischemia, which was also described by Chiu et al.15

Of all the inconsistent diagnoses, it is worth mentioning the patients arriving at the UPA 24h with ST-segment elevation acute myocardial infarction (STEMI) but who did not have this diagnosis recognized by the general practitioner. In this group, 80% of patients presented time from onset of chest pain to admission (delta t) shorter than 12 hours, but only 34% were evaluated in time by the telecardiology and received reperfusion treatment. This result was compatible with the finding in the TIET study,16 in which 44% of the STEMI patients who were within the 12-hour window of the onset of symptoms did not receive any thrombolytic therapy without any justification. The recognition of this diagnosis by the general practitioner, as well as the percentage of thrombolysis is still very low, as previously reported by other authors,15^,17^,18 with a significant impact on morbidity and mortality rates.19^,20 This demonstrates the urgent need to implement measures that allow the rapid diagnosis of STEMI and, consequently, the therapy in a timely manner.

Regarding the safety of the thrombolytics administered (alteplase and tenecteplase), there were no reports of hemorrhagic complications or side effects secondary to medication in the period evaluated.

A total of 1,255 admissions at tertiary hospital were avoided. As already described by other authors,10^,11 telecardiology plays an important role in avoiding unnecessary hospital admissions in patients with a supposed cardiac event in the first care by the general practitioner.

This study also shows the importance of a continued relationship established by telemedicine between the telecardiologist and the general practitioner on a daily basis to guide the management of patients admitted with suspected acute coronary syndrome until outcome, as recommended by the Brazilian Guidelines on Telecardiology in Acute Coronary Syndrome and other Cardiac Diseases.21 This new procedure of work can be considered a support in the country's health system.

Telecardiology has been shown to be an important and effective tool to support the differential diagnosis of chest pain in patients admitted at emergency care units, optimizing bed occupancy in the public health system.

The use of telecardiology as a remote support for general practitioners handling patients with chest pain symptoms is feasible, resulting on diagnostic support and identifying patients that really need to be transferred to tertiary referral hospital, thus avoiding unnecessary bed occupancy and expensive tests.