Additional risk stratification may provide more aggressive and focalized preventive treatment to high-risk hypertensive patients according to the Japanese hypertension guidelines. We prospectively investigated the predictive value of high-sensitivity troponin I (hsTnI), both independently and in combination with N-terminal pro-B-type natriuretic peptide (NT-proBNP), for incident heart failure (HF) in high-risk hypertensive patients with preserved left ventricular ejection fraction (LVEF). Baseline hsTnI and NT-proBNP levels and echocardiography data were obtained for 493 Japanese hypertensive outpatients (mean age, 68.5 years) with LVEF ≥ 50%, no symptomatic HF, and at least one of the following comorbidities: stage 3–4 chronic kidney disease, diabetes mellitus, and stable coronary artery disease. During a mean follow-up period of 86.1 months, 44 HF admissions occurred, including 31 for HF with preserved ejection fraction (HFpEF) and 13 for HF with reduced ejection fraction (HFrEF; LVEF <50%). Both hsTnI (p < 0.01) and NT-proBNP (p < 0.005) levels were significant independent predictors of HF admission. Furthermore, when the patients were stratified into 4 groups according to increased hsTnI (≥highest tertile value of 10.6 pg/ml) and/or increased NT-proBNP (≥highest tertile value of 239.7 pg/ml), the adjusted relative risks for patients with increased levels of both biomarkers versus neither biomarker were 13.5 for HF admission (p < 0.0001), 9.45 for HFpEF (p = 0.0009), and 23.2 for HFrEF (p = 0.003). Finally, the combined use of hsTnI and NT-proBNP enhanced the C-index (p < 0.05), net reclassification improvement (p = 0.0001), and integrated discrimination improvement (p < 0.05) to a greater extent than that of any single biomarker. The combination of hsTnI and NT-proBNP, which are individually independently predictive of HF admission, could improve predictions of incident HF in high-risk hypertensive patients but could not predict future HF phenotypes.