IJCS | Volume 31, Nº2, March / April 2018

111 These data are different from those reported by Franco et al., 7 who assessed 112 patients with STEMI, with baseline characteristics similar to ours and showed a weak to moderate correlation (r = 0.24) between ∆-T and level of schooling. The comparison of marital status by the same authors showed that married, divorced and widowed patients had a higher ∆-T than the single individuals, but also without statistical significance (p = 0.06). The MMSE median score found in our population was of 25, noting that it is considered a normal score when ≥ 27, and cognitive dysfunction (dementia) when≤ 24 or ≤ 17 for illiterate or time of schooling less than 4 years. 8 This result reflected the low educational levels in our sample, which were even lower in the group of women. The ∆-T medians found in our study were 4 hours 51 minutes for the general population and 3 hours and 49 minutes, considering only patients with STEMI, similar to that reported by other national studies, such as the one by Bastos et al. 4 in São José do Rio Preto (4 hours and 08 minutes) and by Franco et al. 7 in Porto Alegre (3 hours and 11 minutes). Our result is also similar to that found in an international study, which found a median ∆-T of 3 hours and 30 minutes in the United States, 4 hours and 24 minutes in South Korea, 4 hours and 30 minutes in Japan and 30 minutes in England. 9 However, our ∆-T is still much higher than that recommended by theAmerican Heart Association guidelines. 10 When available in a timely fashion (< 2 hours), primary angioplasty results in a benefit regardingmortalityof 25 to30%. Thebenefit of thrombolysis ismaximal when administeredwithin 2 hours of symptom onset, especiallywithin the first 70minutes, since resistance to thrombolysis is time-dependent. 11 The GRACE study found that individuals with a previous history of AMI had arrived earlier at the hospital when compared to those without a history of angina, diabetes, heart failure, or hypertension. 12 Our study did not include patients with prior AMI, but most individuals were aware of at least one risk factor for coronary heart disease and it was expected that they would seek care faster because they associated their symptoms with the heart; however, this was not confirmed in this study. Nevertheless, it was observed that the patients with arterial hypertension had a worse performance in the MMSE. This result may be simply related to the lower level of schooling observed in this group, or even suggest that hypertension is a factor associated with the identified cognitive decline. The role of systemic arterial hypertension in determining loss of cognitive function, in the absence of a previous cerebral vascular accident, is still not a consensus in the literature. Some clinical studies have shown that hypertensive individuals show a poor performance in neuropsychological tests. 13 Reports from the Framingham study observed an inverse association between blood pressure and cognition, concluding that elevated blood pressure might be related to the presence of cognitive decline. 14 Other investigations found the inverse, that is, systemic arterial hypertension was associated with better cognitive function in the elderly. 15 Finally, other studies did not demonstrate this association. 16 The results of our study suggest that factors such as level of schooling, cognitive performance and marital status were not determinant for the delay in hospital arrival, similar to the data reportedbyDracup andMoser. 17 It is known that other prehospital factors influence the patient’s time of arrival, such as the region of origin, the availability of their own means of transportation or ambulance availability for transportation to the hospital, and the search or transfer to the closest health service, which, in most cases is not a specialized one and does not have the recommended resources for the care of infarction victims. 18,19 Our study reflected the logistical and operational obstacles faced in the care of infarction in the public health system, since the factors that most often determined the delay in the specialized treatment of our population were the transfer from another health service and the region of origin. Among this study limitations, the fact that we exclusively evaluated patients treated by the Brazilian Unified Health System and mostly in a single institution (only one patient was treated in another public hospital in Florianópolis), may have disregarded a portion of the population that, supposedly, could have better socioeconomic conditions and higher levels of schooling. Additionally, the small sample size may have influenced the lack of statistically significant associations, although there was enough power to evaluate the suggested correlations. Also, according to Bonnet and Wright, 20 perhaps the data would be more accurate if the correlation had been calculated with a confidence interval, which was not used in the present study. However, these biases do not invalidate our study, because even if the sample were larger, it is possible that any found correlation would be weak and without clinical significance. Takagui et al. Infarction and delay in hospital care Int J Cardiovasc Sci. 2018;31(2)107-113 Original Article