ABC | Volume 111, Nº2, August 2018

Original Article Marino et al 123 I-MIBG Scintigraphy in Heart Failure Arq Bras Cardiol. 2018; 111(2):182-190 Table 1 – Demographic, clinical and echocardiographic characteristics of the patients CCC non-CCC HT p Male sex † 68.0 56.0 77.0 0.281 a Age (years)* 53.3 ± 9.2 43.3 ± 12 47.3 ± 13.1 0.016 c HR > 80 bpm † 30.8 33.3 69.2 0.072 a NYHA II-IV † 62.5 92.0 0.0 < 0.001 a LVEF % (Echo)* 30.6 ± 7.8 25.9 ± 8.0 66.6 ± 8.3 < 0.001 c CCC = non-CCC < HT ACEI † 91.3 88 77.3 0.394 b Beta-blockers † 91.3 100 18.2 < 0.001 a CCC = non-CCC > HT CCC: chronic Chagas cardiomyopathy; non-CCC: cardiomyopathy other than Chagas disease; HT: heart transplant; HR: heart rate (bpm); NYHA: New York Heart Association (heart failure functional classification); ACEI: angiotensin-converting-enzyme inhibitors; Echo: echocardiography; (*): expressed as mean and standard deviation; (†): expressed as percentage. Note: The probability of statistical significance for the comparison of the groups refers to (a) chi-square test, (b) Fisher exact test, and (c) analysis of variance. To characterize the sample, descriptive analysis of the following variables was initially performed: sex, age, heart rate (HR), LVEF, early HMR, late HMR and WO% of 123 I-MIBG expressed according to the distribution of frequency or measures of central tendency and variability. This analysis was stratified per group (CCC, non-CCC and HT). When comparing the three groups, for the categorical variables (sex, HR, NYHA functional class, use of beta‑blockers), Pearson chi-square test was performed; for the continuous variables (age, LVEF, WO%), one-factor analysis of variance (ANOVA) was used; and for multiple comparisons, the least significant difference (LSD) test was used. In addition, ANOVA was used to assess the variables (early and late HMR), based, however, on repeated-measures analysis and LSD test for multiple comparisons. It is worth noting that the assumptions to use ANOVA were verified and accepted, that is, normally distributed residuals (Kolmogorov-Smirnov test) and constant variances (Levene’s test). To analyze the correlation between the measures of early HMR, late HMR or WO% and LVEF, Pearson correlation test and its respective p value were used. In all analyses, a 5% significance level was considered, and the statistical software SPSS, version 17.0 (SPSS Inc., Illinois, USA), was used. 31,32 Results Table 1 shows the demographic, clinical and echocardiographic data of the patients studied. Those with HF on angiotensin-converting-enzyme inhibitors (ACEI) and beta‑blockers maintained their medications. In approximately 70% of the HT patients, HR was maintained over 80 bpm (mean of 90.7 bpm), while in individuals with CCC and non-CCC, the mean HR values were 72.7 bpm and 75.6 bpm, respectively (p = 0.03). No patient was on tricyclic antidepressants. The early and late HMR values were greater than those reported for HT patients (p < 0.001) (Table 2), but did not differ in CCC or non-CCC patients with HF, even when adjusted for age and age group (early HMR: p = 0.251; and late HMR: p = 0.011). The early HMR values of CCC patients were 8.6% higher than those found in non-CCC patients, and 39.7% higher than those found in HT patients. The late HMR values of CCC patients were 9.7% higher than those of non-CCC patients, and 31.7% higher than those of HT patients (Figure 2). TheWO% values showed no statistically significant difference between individuals with HF, and between individuals with HF and those submitted to HT (p = 0.577) (Figure 3). A weak positive correlation was observed between late HMR values and LVEF in CCC patients (r = 0.42; p = 0.045). Regarding the non-CCC patients, a positive correlation was observed both between LVEF and early HMR (r = 0.46; p = 0.023) and between LVEF and late HMR (r = 0.49; p = 0.015). However, none of the groups showed a correlation between LVEF and WO% (Figures 4, 5 and 6). Discussion This study investigated the presence and magnitude of cardiac dysautonomia in patients with HF and LVEF ≤ 45% by use of 123 I-MIBG scintigraphy. Patients were divided into three groups, CCC, non-CCC and HT, the latter, by representing the denervated heart model, served as the abnormality pattern. 23 There was scintigraphic evidence of sympathetic hyperactivity, based on the findings of low 123 I-MIBG uptake (early and late HMR) by the presynaptic endings in the three groups studied, which is aligned with the literature. 13,23,24,28 The low 123 I-MIBG uptake indicates dysfunction of the receptors and integrity loss of the presynaptic sympathetic fibers, reinforcing the theory of sympathetic hyperactivity in the pathogenesis of HF. 8-10,12,24 Scintigraphy is the only noninvasive and safe method, sufficiently sensitive to assess the autonomic sympathetic nervous system, 12,24 that can provide parameters known for their accuracy and reproducibility to estimate the efficacy of clinical treatment 13 and the prognosis of patients with HF. 13,24,25,33 However, the lack of standardization in the scintigraphic imaging acquisition and processing hinders the incorporation of the method into clinical practice, because there is no well-defined reference value. 28,29 In a meta‑analysis of seven studies with 96 healthy individuals, Patel and Iskandrian have reported HMR of 2.13 ± 0.3 and WO% of 20 ± 10% (ranging from 10 ± 6% to 37 ± 5%) for healthy individuals. 29 184