ABC | Volume 111, Nº2, August 2018

Original Article Duarte et al Iron impact in acute coronary syndrome Arq Bras Cardiol. 2018; 111(2):144-150 Methods Sample The sample was retrospectively evaluated and consisted of consecutive patients admitted to a Coronary Unit with an ACS diagnosis between June 2011 and June 2013. Patients whose iron profile was not determined during hospitalization were excluded. Variables The population was characterized according to their baseline characteristics (age and gender); clinical characteristics (personal history, type of ACS, Killip Class, LV ejection fraction) and laboratory test results (serum levels of creatinine, BNP, hemoglobin), and was grouped according to the tertiles of distribution of serum iron (1 st tertile ≤ 40; 2 nd tercile > 40 and ≤ 67; 3 rd tertile >67 mcg/dL) and ferritin (1 st tertile ≤ 110; 2 nd tertile < 110 and ≤ 219; 3 rd tertile >219 ng/mL). The cut-off values of serum iron and serum ferritin were, respectively, 60-180 mcg/dL and 10-120 ng / mL, according to the hospital laboratory. End-point The short- and long-term prognoses were assessed based on primary adverse events: in-hospital death and 1-year death; in-hospital heart failure (Killip Class ≥ 2 and BNP ≥ 400 pg/mL) and 1-year follow-up (ejection fraction < 50% and NYHA Class ≥2). Other secondary endpoints were reinfarction and ischemic cerebrovascular accident at 1 year of follow-up. Statistical analysis The program IBM SPSS Statistics, version 20 for Windows 8, was used to perform the statistical analysis. Continuous variables were shown as mean ± standard deviation and compared according to iron and ferritin tertiles by ANOVA. Categorical variables were shown as absolute values and/or percentages and compared using the chi-square test. The associations were considered statistically significant in the presence of a p-value < 0.05. Continuous variables were associated with primary adverse events (Death and HF) through receiver operating characteristic (ROC) curves. The predictive value of iron and ferritin levels over the risk of in‑hospital and 1-year adverse events was determined by the odds ratio, with a 95% confidence interval (95%CI). Results The baseline, clinical and laboratory characteristics of the total population and according to serum iron and ferritin tertiles are shown in Tables 1 and 2. A total of 280 patients were studied (73%males) with a mean age of 68 ± 13 years. The distribution of serum iron and ferritin levels is shown in figure 1. The main diagnosis at admission was ST-segment elevation myocardial infarction (STEMI) in 45% (n = 125) and non-ST- segment elevation myocardial infarction (NSTEMI) in 44% (n = 122) of the patients. Approximately 87% of the patients (n = 244) were admitted with Killip I Class, and only 2.5% (n = 7) were admitted in cardiogenic shock. In 11 (5%) patients, the transthoracic echocardiography showed severe left ventricular systolic dysfunction. Regarding the short-term prognosis: 1.1% (n = 3) of the patients died during hospitalization and 28% (n = 79) showed evidence of heart failure. Regarding the long-term impact, approximately 7% (n = 19) of the patients died in the first year of follow-up and 12% (n = 33) developed HF criteria during the clinical follow-up (Tables 3 and 4). The multivariate regression analysis showed that a ferritin value > 316 ng/mL is an independent risk predictor for 1-year death (adjustedOR:14; 95%CI: 2.6-75.9, p=0.0023). (Table 5) The survival curves for Death and HF according to iron and ferritin tertiles showed no statistical difference. (Figures 2 and 3) Discussion Iron deficiency is a common and clinically relevant heart failure comorbidity, being associated with a worse prognosis. Some studies (CONFIRM-HF) have shown the benefit of iron correction in terms of quality of life and exercise tolerance in patients with HF and decreased systolic function. 1,3 According to current European recommendations for HF treatment, intravenous iron is indicated with Class IIa in symptomatic patients with decreased systolic function and iron deficiency (serum ferritin levels <100ng / mL or ferritin level between 100-299 ng / mL and transferrin saturation <20%). 3 On the other hand, the role of iron and ferritin is uncertain in the context of atherosclerotic disease and ACS. 5-8 In this population, the type of ACS, as well as its clinical presentation regarding Killip class and LV systolic function impairment were not statistically influenced by iron or ferritin levels. Age had a statistical impact on serum iron and ferritin levels, which may be explained by an iron-poor diet, impaired intestinal absorption that increases with age, and the presence of more comorbidities that interfere with iron metabolism. Lower levels of iron and ferritin were statistically associated with lower Hb levels, as expected, with a mean value of 12 g/dL. Mean BNP values >450 pg/mL were statistically associated with the 1 st tertiles of iron and ferritin, which is consistent with several studies in the HF scenario, in which iron and ferritin deficiencies were found as a frequent comorbidity of HF. 1,3 Alberto Dominguez-Rodriguez demonstrated that low iron levels may be associated with major cardiovascular events (MACE) in patients with ACS. 5 In this population of ACS patients, alterations in iron metabolism were associated with a higher occurrence of adverse events. Iron levels ≤ 40 mcg/dL had a negative impact regarding mortality and in-hospital HF, with statistical significance; however, serum iron levels were not an independent risk factor for the occurrence of cardiovascular events. 145