Objective To explore the influence of high-sensitivity cardiac troponin T (hs-cTnT) on the prognosis of frail and pre-frail elderly hospitalized patients without acute coronary syndrome (ACS). Methods The elderly patients who were admitted to our department and identified as frailty or pre-frailty by Fried Frailty Phenotype assessment from January 2017 to December 2019 were enrolled in this study. Their hs-cTnT level, general information, laboratory indicators and echocardiographic indicators were recorded. Follow-up was conducted each three months through phone call after discharge and all-cause deaths of the patients were observed. These patients were divided into low-, median-, and high-value groups according to hs-cTnT tertiles. The survival curves of the three groups were compared using Kaplan-Meier method (K-M curve). Cox proportional hazard-regression model was used to analyze the effect of 3 hs-cTnT groups on the risk of death. SPSS statistics 18.0 was used for statistical analysis. Data comparison between two groups was perfomed using t test, Fisher exact test, Wilcoxon test or χ² test depending on data type. Results A total of 450 frail and pre-frail elderly inpatients were subjected in this study, 72.4% (326/450) of them had hs-cTnT levels exceeding the cut-off value of 0.014 μg/L. There were significant differences in age [(83.67±5.72)vs (86.06±4.93) vs (87.67±5.23)years], male ratio [62 (41.3%) vs 94 (62.7%) vs 108 (72.0%)], chronic heart failure ratio [4 (2.7%) vs 5 (3.3%) vs 22 (14.7%)], hypertension ratio [116 (77.3%) vs 119 (79.3%) vs 133 (88.7%)], arrhythmia ratio [41 (27.3%) vs 38 (25.3%) vs 60 (40.0%)], chronic kidney disease ratio [27 (18.0%) vs 38 (25.3%) vs 61 (40.7%)], comorbid conditions [≥4,72 (48.0%) vs 73 (48.7%) vs 99 (66.0%)], hemoglobin level [(125.11±16.03)vs (121.50±18.08) vs (115.38±16.97)g/L], estimated glomerular filtration rate [(87.91±30.74)vs (76.73±23.08) vs (66.69±27.80)ml/(min·1.73m²)], albumin level [(38.15±3.81)vs (37.60±3.98) vs (36.04±4.41)g/L], total cholesterol level [(4.01±0.98)vs (3.62±0.88) vs (3.70±0.85) mmol/L], low-density lipoprotein cholesterol level [(2.42±0.88)vs (2.05±0.73) vs (2.19±0.77)mmol/L], N-terminal pro-brain natriuretic peptide level [163.5 (104.8,398.9) vs 314.7 (171.4,683.8) vs  547.3 (288.3,1568.3) pg/ml], interventricular septal thickness [(11.30±0.83)vs (11.42±0.79) vs (11.71±1.07)mm], left ventricular post-wall depth [(11.31±0.83)vs (11.46±0.75) vs (11.65±0.83)mm], left ventricular mass index [(107.38±13.32)vs (109.90±13.86) vs (112.72±19.29)g/m²] and left ventricular ejection fraction [(60.65±2.97)% vs (59.58±3.91)% vs (58.54±5.08)%] among 3 groups (all P<0.05). Cox regression model analysis showed that after adjustment for gender, age, comorbidities, laboratory indicators and echocardiographic indicators, the mortality risk was significantly higher in the patients of the high-value group than those in the low-value group (HR=3.133,95%CI 1.381-7.109, P<0.01). Survival curve analysis indicated that median survival time was estimated at 53,51 and 48 months in the low-, median- and high-value groups, respectively. And in 10-20 months after discharge, the survival rate of the high-value group was significantly lower than that of the other two groups, and even in 20-30 months after discharge, the rate of the high-value group and the median-value group was obviously lower than that of the low-value group. Log-Rank test showed that the survival rates of the three groups was significantly different (P<0.001). Conclusion For the frail and pre-frail elderly inpatients without ACS, higher hs-cTnT level, higher than the cut-off value, indicates higher risk of death. The increment exerts adverse effect on all-cause death, with the higher the level, the higher risk of death.