Preditores da evolução da cardiopatia chagásica crônica em pacientes sem disfunção ventricular esquerda

Abstract

Introduction: Chronic chagasic cardiomyopathy (CCC) is primarily responsible for the high morbidity and mortality of Chagas disease, causing enormous social and medical and labor impact. The identification of markers of disease progression encourages early interventions that might slow the worsening of the lesions. Objectives: To study independent prognostic factors (or predictors) of the evolution of cardiomyopathy in patients with CCC with electrocardiogram (ECG) changes but without left ventricular dysfunction, coupled with socio-demographic, epidemiological, clinical and noninvasive cardiologic workup, to compare the G1 (patients with nonspecific ECG changes) and G2 (patients with ECG abnormalities suggestive of CCC) groups regarding the emergence of left ventricular dysfunction. Method: nonconcurrent cohort study of patients treated as outpatients in Chagas Disease Ambulatory in the period of 1985 to 2009. All patients had CCC either with ECG changes suggestive of disease or nonspecific ECG changes, and were followed-up for at least two years of the initial time (time 1). Explanatory variables assessed and included socio-demographic, epidemiological, clinical and results of cardiac tests, after patients having met inclusion and exclusion criteria. The outcomes were defined as worsening of CCC status and occurrence of left ventricular dysfunction on echocardiography (ECHO). The CCC was classified in stages 1-4, according to the progressive severity of ECG and ECHO abnormalities. McNemar's and Wilcoxon's tests were used for comparison of variables at the beginning and end of the study. In univariate analysis, we used the Kaplan-Meier survival curve and the univariate Cox model. Multivariate analysis was performed using the Cox regression model. For entry of predictive variables in the Cox regression model a p-value of 0.20 was used and for the variable to stay in the final model a 5% level of significance was adopted. The hazard ratio (HR) was estimated with confidence intervals of 95%. Results: We studied 165 patients. The minimal follow up was two and the maximum was 20 years, with an average of 8.2 and a median of 8.0 years. The mean age was 44.8 years (20- 77 years). There was a statistically significant increase in time 2 of PR interval 230 milliseconds, and QRS duration 120 milliseconds, besides the occurrence of heart rate below 60 beats per minute (McNemar p < 0.05 and Wilcoxon < 0,01). There was no statistically significant difference between groups G1(53%) and G2 (47%), with the power of the sample calculated at 12.23%. Univariate analysis, considering the outcome of worsening of CCC classification, revealed the following statistically significant variables: male gender (HR = 2.59), age > 50 years (HR = 2.18), current residence in rural areas (HR = 6.54) and in an endemic area (HR = 3.39); cardiothoracic index (CTI) > 0.50 (HR = 6.42), duration of use of loop diuretics (p = 0.036), digitalis (p = 0.000), spironolactone (p = 0.004) and angiotensin converting enzyme inhibitors/angiotensin II receptor blockers (p = 0.030), presence of ventricular arrhythmias in exercise test for grouping the classification of Lown 1-4 (HR = 2.36); ECG pauses 2 seconds (HR = 2.99), ventricular arrhythmia - pairs (HR = 3.09) and non-sustained ventricular tachycardia (HR = 1.47) all at Holter. In assessing the outcome left ventricular dysfunction at ECHO, there was a significant difference for the CTI 0.50 (HR = 12.73). In multivariate analysis, the variables that remained significant were male gender, current residence in a rural area, increased CTI by chest X-ray and time-use of digital. Conclusions: This study showed that male gender, current residence in a rural area, increasing CTI and time-use of digital were the variables that could predict CCC progression, pointing out growing need for rigorous follow up for these patients.