ABC | Volume 111, Nº4, Octuber 2018

Original Article Rodrigues et al Predictors of late presentation in ST-elevation myocardial infarction Arq Bras Cardiol. 2018; 111(4):587-593 This finding is compatible with data from the CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation) registry, which showed that Caucasians arrive earlier than Blacks while studying a population of more than 100,000 patients ( Odds Ratio – OR -2.2; 95% Confidence Interval: 95%CI -4.2 - -0.3; p = 0.03). 14 On the other hand, a large study involving more than 43,000 consecutive patients with STEMI from the ACTION‑GWTG database showed that there is no significant difference regarding time of arrival between Black and Caucasian patients. 15 Racial differences could be explained by genetic or socioenvironmental characteristics, and we are unaware of studies that have showed differences in pain threshold according to ethnicity. On the other hand, individuals of Black ethnicity in Brazil show unfavorable socioeconomic and cultural status when compared to those of White ethnicity, which could explain our findings. From this perspective, low-wage income was also identified as an independent predictor of late presentation in our study. Nguyen et al. 8 performed a systematic review that also showed that patients with low socioeconomic status seek medical attention later. 8 Low income may be associated with the patient's recognition of their symptoms and pathology, inferring that people with better level of schooling seek emergency services earlier. 8 On the other hand, Qian et al. 1 analyzed 100 patients with STEMI in China, with no association of low wage income with late presentation. 1 In our study, the diagnosis of DM was also an independent predictor of late presentation, which is compatible with the evidence available in the literature. 16-21 Patients with DM more frequently have silent ischemia, which may be explained by the presence of diabetic neuropathy and a higher pain threshold. Previous heart disease was considered a protective factor for late presentation, and the association between this characteristic and the time of presentation varied according to the studies. Kuno et al. 22 demonstrated that patients who had been previously submitted to a percutaneous coronary intervention procedure had a shorter time of presentation. 22 In a cross-sectional study that included 335 patients and considered late presentation arriving at the hospital within 12 hours of pain onset, previous STEMI and revascularization did not show a statistically significant association with time of presentation. 23 Our study did not include analyses of the associated mechanisms between the presence of predictors and the occurrence of late presentation, but it could be speculated that patients who had previous cardiac events or were submitted to myocardial revascularization procedures would be more familiarized and conscious about the disease and the need to seek medical attention quickly. Women showed significantly longer time until hospital arrival than men, but female gender did not remain an independent predictor of late presentation in the multivariate analysis. The association between female gender and hospital arrival delay after chest pain onset has also been reported in other studies, and it has been found that women more often have atypical symptoms than men. 24-26 Limitations In this study, we did not have available information regarding the distance between the patients and the hospital when they had the chest pain onset, a fact that may have an influence on hospital arrival delay. However, most of the patients who come spontaneously to our institution are city residents. Because it is located downtown, travel time does not exceed 30 minutes in most cases. It is important to emphasize that patients transferred from other hospitals and health institutions were excluded from our study, since the objective was to analyze the factors that influence spontaneous delay in search for medical care of patients with infarction, and not to analyze factors that have an impact on medical transfer time. We considered analyzing the association between the distance from the patients' home to our institution, but many patients were not at home at the time of pain onset, but at work or another location, and therefore this analysis was not included in the present report. We did not have available ventricular function information from all patients, because left ventriculography is not routinely performed during catheterization and primary percutaneous coronary intervention (pPCI) to minimize contrast volume. However, the percentage of patients with previous CHF who presented with Killip III/IV class at the time of STEMI was similar, suggesting that left ventricular function in both groups was not significantly different. This was a single-center study in a large tertiary cardiology hospital, and the results shown hereinmay not be valid for populations that are significantly different from ours. Conclusions The independent predictors for late presentation to the hospital in patients with acute ST-segment elevation myocardial infarction were Black ethnicity, low-income and DM, whereas a history of previous heart disease was a protective factor. Approximately one-fourth of the patients in this sample were late arriving at the hospital, and their mortality rate was significantly higher than those who arrived early. Patients who had all of the characteristics associated with late presentation showed a two-fold delay related to hospital arrival when compared to those without these characteristics, which illustrates the potential opportunity to decrease the mean time of arrival if public health interventions focused on these high-risk subgroups are carried out. Author contributions Conception and design of the research: Rodrigues JA, Quadros AS; Acquisition of data: Melleu K; Analysis and interpretation of the data: Rodrigues JA, Schmidt MM, Quadros AS; Statistical analysis: Rodrigues JA, Schmidt MM; Writing of the manuscript: Rodrigues JA; Critical revision of the manuscript for intellectual content: Gottschall CAM, Moraes MAP, Quadros AS. Potential Conflict of Interest No potential conflict of interest relevant to this article was reported. 591