Both BNP and left ventricle diastolic diameter can predict all-cause mortality among Colombian asymptomatic, Trypanosoma cruzi-seropositive carriers after 22 years: results of the CHICAMOCHA cohort

Abstract Introduction Chronic Chagas’ cardiomyopathy (CCC) may take decades to unfold after a silent Trypanosoma cruzi chronic infection. EKG and other cardiac diagnostic findings have been proposed as markers of worsening prognosis in CCC-free infected populations from Brazil and Argentina. Prognostic data from northern endemic countries, where parasites and natural history of infection may differ, are scarce. Purpose To explore the association between selected cardiac function markers and all-cause mortality among CCC-free T. cruzi-seropositive carriers (TcS+) from Colombia. Methods A cohort of voluntary blood donors with confirmed T. cruzi serology status (consecutive TcS+ and a sample of TcS- bank-matched controls, 1:4 ratio) residents in Bucaramanga was first seen in 2000-2003. At baseline all participants had an EKG, and a subsample had a serology-blind assessment with cardiac autonomic function testing (CAFT), resting transthoracic echocardiography (ECHO) and BNP levels. Mortality data were obtained in 2024 by cross-linkage with government databases after validation. From 18 pre-selected candidate markers (4 EKG-based, 8 ECHO-based, 5 CAFT-based and BNP values), we chose the best predictors following statistical criteria (after comparing by serological status, and then by vital status in TcS+, using as threshold the 90 and 95th tiles among TcS-). Finally, we computed crude and covariate-adjusted Hazard ratios (aHRs) for the proposed predictors using time-to-event regression analyses. Results Of 2134 participants (at baseline 488 TcS+, median age 37, IQR 13 years, 64% males, 74.9% of low socioeconomic status), 4.3% TCs+ Vs 0.2% TcS- (p0.001) had two concurrent EKG rhythm/conduction abnormalities, and in the subsample (the first 400 positives and a random selection of 200 controls) 2.8% TcS+ Vs 1% TcS- had 10 or more ventricular premature beats in 20-minute recordings (p=0.162). The best markers were left-ventricle diastolic diameter (LVDD) for ECHO, pNN50 over 10 minutes at rest for CAFT, and BNP values. As of September30th, 2024 (median follow up 23.6, IQR 2.2 years), 137 deaths had been recorded (38, 7.8% among TcS+ Vs 99, 6.0% in controls, p=0.163). LVDD (32.3% Vs 5.6% deaths when 5.7 cm or greater, p=0.007) and BNP (17.8% Vs 6.5% deaths when 80.6 ng/ml or greater) were both positively associated with death among TcS+ (LVDD aHR=5.7, 95%CI 2.5-13.0; BNP aHR=3.8, 95%CI 1.7-8.9, Figures 1 and 2). When included in the same model both markers predicted death (LVDD aHR=4.6, 95%CI 2.1-10.3 and BNP aHR=2.5, 95%CI=1.1-5.9, p for interaction=0.856). Conclusion In this population, baseline BNP and LV diastolic diameter predicted all-cause mortality after 22 years. BNP may be an efficient, accessible tool upon, or along with, serological diagnosis in community screening. Further study with cardiac (fatal and non-fatal) events in this and other cohorts will enhance and validate these findings.Mortality by LV diastolic diameter Mortality by BNP levels