IJCS | Volume 32, Nº2, March/April 2019

127 Machado et al. Care changes in primary PCI Int J Cardiovasc Sci. 2019;32(2)125-133 Original Article Statistical analysis Continuous variables were expressed as mean (± standard deviation) or median (interquartile range [IQR]) based on the presence of symmetrical and asymmetrical distribution, respectively. The normality of the distribution of each variable was assessed by the Shapiro-Wilk test. Categorical variables were represented by their relative and absolute frequencies. Temporal trends were tested using the chi-square test for binary variables and we performed either the Kruskal-Wallis one-way analysis of variance by rank for non-normally distributed values or one-way analysis of variance for normally-distributed values. All hypothesis tests had a two-sided significance level of 0.05. All statistical analyses were conducted using IBM Statistical Package for the Social Sciences (SPSS) Statistics, version 21.0. Results Between March 2011 and February 2017, 541 consecutive patients presenting with STEMI were enrolled in our registry. Baseline characteristics are shown in Table 1. The incidence of diabetes increased significantly during the study period (from20.0% in 2011 to 32.4% in 2016 (p trend = 0.005). Incidences of hypertension (63.3%), previous history of MI (9.6%), history of stroke (5.9%), chronic obstructive pulmonary disease (COPD, 4.1%), current smoking (50.9%) and obesity (23.6%) remained stable. There was no significant change over the 6-year period regarding the proportion of Killip III- IV patients. There was an important increase in the use of radial access in patients, from 20.0% in 2011 to 62.7% in 2016 (p trend < 0.0001). Additionally, thrombus aspiration decreased significantly, from 66.7% in 2011 to less than 3.0% in 2016 (p trend < 0.0001), as seen in Figure 1. Furthermore, the use of bare metal stents (BMS) decreased from 97.6% (in 2011) to 75.0% (in 2016) due to the increment of use of drug-eluting stents (DES, p trend < 0.0001). Another important decrease was observed regarding the dose of radiation and contrast volume during procedures (p trend < 0.0001). Other procedural characteristics and initial and final TIMI flow grades are detailed in Table 2. In-hospital short and long-term mortalities remained reasonably stable from 2011 to 2016 (p trend > 0.05) as the incidence of in-hospital MACCE and 30-day MACCE (p trend > 0.05). These data are detailed in Figure 2 and Table 3. Patients with femoral access had significantly higher rates of in-hospital mortality (19.8%) and in- hospital MACCE (20.7%) and long-term mortality (24.9%), when compared to patients treated via radial access (5.3%, 8.84%, 11.4% respectively; p < 0.0001 for both comparisons). However, patients treated through femoral access more often had Killip class III or IV (24.9% vs 5.0%), cardiac arrest (20.5% vs 6.5%), hypotension at admission (26.6% vs 4.1%) intra-aortic balloon use (8.5% vs 0.9%; p < 0.0001 for all comparisons). There was a significant reduction in the door-to- balloon median time from 80.0 minutes in 2011, to 67.0 minutes in 2016 (p trend = 0.002). In addition, there was an increase in the use of public emergency medical services (SAMU) from 13.3% in 2011, to 33.1% in 2016 (p trend = 0.01), with a significant reduction in hospital admissions through the emergency department from 40.0% in 2012, to 23.0% in 2016 (p trend = 0.004). There was no significant reduction in hospital length of stay during the study period (p = 0.56). Discussion During the period of analysis of our cohort of STEMI patients who underwent primary PCI, a rise in STEMI patients assisted in our service was observed, with changes in patient’s profile, reflected by a higher incidence of diabetes, and the increase in the number of patients admitted via SAMU. Nevertheless, there was a significant reduction in door-to-balloon time. Procedural characteristics also underwent modifications, such as the preference for radial access and the disuse of aspiration thrombectomy, in conformation with the changes in guidelines. The increased incidence of diabetes worldwide has been associated to changes in lifestyle and the environment brought on by industrialization. These changes have led to obesity, lack of physical activity and consumption of a diet rich in fats and calories. 7 The prevalence of diabetes in patients with AMI in the present study (24.9%) seems to be similar to that observed in the Brazilian population in general, which in this age group over 60 years is around 25%. 8,9 However, its increase from 20.0% in 2011, to 33.1% in 2016, agrees with an expected increase in the incidence found in other national prevalence studies. VIGITEL data – a national survey of chronic disease carried out through telephone contact – also detected a national increase from 6.3% to 8.9% from 2011 to 2016 in adults over 18