During August 2015–July 2017, we recruited cardiac patients for Chagas disease surveillance from Harris Health System–Ben Taub Hospital, a large county-funded tertiary care facility in Houston. Patients with known nonischemic cardiomyopathy who sought treatment at the outpatient cardiac clinic or who were admitted to a cardiac inpatient unit were invited to participate in our study. Inclusion criteria required a recorded ejection fraction <50% within the past year and a recent negative ischemic work-up based on stress echocardiography or invasive coronary angiography. We excluded patients of non-Latinx ethnicity and those who were currently incarcerated, had prior T. cruzi serologic testing, had evidence of acute coronary syndrome suspected to be of Takotsubo origin, or had documentation of an alternative etiology for their nonischemic cardiomyopathy (e.g., peripartum, genetic, or alcoholic cardiomyopathy). Consent forms were available in English and Spanish, and licensed translators ensured that all potentially eligible participants were invited to participate. This protocol was reviewed and approved by the Baylor College of Medicine Institutional Review Board (protocol no. H-36761).

After consent, participating patients provided a blood sample for T. cruzi diagnostic testing and completed a risk factor questionnaire. The 5-page questionnaire was administered by a study team member and included sections on residential and travel histories, potential triatomine exposures and sources, current health symptoms and health behaviors, clinical family history, and knowledge, attitudes, and practices regarding Chagas disease. Initial T. cruzi diagnostic testing included T. cruzi–specific antibody testing using Chagas STAT-PAK Assay (Chembio Diagnostic Systems, Inc., https://chembio.com) and Hemagen Chagas Kit (Hemagen Diagnostics, Inc., https://www.hemagen.com). Confirmation of positive and discordant results were then performed by using Chagatest ELISA Recombinante 3.0 (Wiener Laboratorios S.A.I.C., https://www.wiener-lab.com) and TESA blot by the Centers for Disease Control and Prevention.

During the 2-year study period, 97 patients with nonischemic cardiomyopathy were enrolled out of 132 eligible patients; 35 refused to participate because of lack of interest. The average age of participants was 52 years (range 28–91 years); 38% of participants were female and 62% male. Birth countries for the cohort were Mexico (53%), United States (14%), El Salvador (12%), Honduras (9%), Guatemala (4%), and other Latin America Spanish-speaking countries (8%). Patients born in the United States originated from Texas (n = 9), New York (n = 2), Indiana (n = 1), and Oregon (n = 1). Of the cohort, 43% reported having previously seen the triatomine vector; 20/42 (48%) reported sightings in Texas, compared with 31/42 (74%) in a Chagas-endemic Latin American country. Furthermore, 12% of the cohort reported a history of triatomine bites. Despite high triatomine recognition, only 8% of the patient cohort had ever heard of Chagas disease, and only half of these patients could correctly state how Chagas disease is acquired.

Overall, 7% of Latinx nonischemic cardiomyopathy patients seeking treatment for heart failure management were confirmed positive for T. cruzi infection by Centers for Disease Control and Prevention Wiener EIA and TESA blot confirmation testing. Discordant test results were common (Table), complicating the clinical decision-making process. All 7 patients who had laboratory-confirmed Chagas cardiomyopathy were born in a Latin America country: El Salvador (n = 4), Honduras (n = 1), Mexico (n = 1), and Venezuela (n = 1). All 7 confirmed positive patients had mothers who were born in or had lived in a Latin America country. Three had lived in a house with a dirt floor and 2 with a palm leaf thatched roof, which are both known risks for triatomine infestations (8,9). One participant had received a blood transfusion in their home country. Two were polyparous mothers, and none of their children had been tested for Chagas disease. Only 2 of the 7 patients with Chagas cardiomyopathy had ever heard of Chagas disease, and only 1 of these patients knew how Chagas disease was acquired.

Our study adds to the growing body of evidence supporting T. cruzi surveillance of Latinx patients with nonischemic cardiomyopathy or other risk factors for T. cruzi infection in the United States. T. cruzi infection accounts for a considerable proportion of nonischemic cardiomyopathy in foreign-born Latinx patients (7%–19%) (4,5), and the timely diagnosis of their infection is imperative.

Our investigation has a few limitations, including the inability to perform additional cardiac imaging and diagnostic studies or follow patients long-term to evaluate prospective identification of underlying etiology. As highlighted by our discordant results, further work is needed to develop a highly specific diagnostic test to prevent clinical confusion regarding accurate disease status. Determining the underlying etiology has a benefit for Chagas cardiomyopathy patients despite the limited efficacy of treatment with antiparasitics (benznidazole and nifurtimox). Patients with Chagas cardiomyopathy might be recommended for heart transplant (10) and can positively respond to implantable cardioverter-defibrillator placement (11) and amiodarone (12). Awareness of infection could lead to testing of at-risk family members who might respond favorably to early treatment.