Increased cTnI Predicts Early Death in Patients with Severe Fever with Thrombocytopenia: A Multicenter Study in North China.
Infect Drug Resist
; 17: 2579-2590, 2024.
Article
in En
| MEDLINE
| ID: mdl-38919833
ABSTRACT
Background:
Myocardial injury is common in severe fever with thrombocytopenia syndrome (SFTS) patients. Currently, research on the prognostic value of cardiac troponin I (cTnI) for predicting the mortality of SFTS patients, especially death within 7 days is limited.Methods:
Between May 2011 and October 2022, clinical and laboratory data on admission of consecutive SFTS cases were collected from six medical centres in China. The clinical endpoint was in-hospital all-cause death within seven days. Risk factors of myocardial injury and death were analysed using multivariable regression models. Prognostic models were established using Cox regression and performance of indicators was evaluated in terms of calibration, discrimination.Results:
A total of 1379 laboratory-confirmed patients were enrolled, in which 686 subjects were included for analysis. The median age was 66 years, with 48.1% of male. Eighty-seven patients died within seven days and 396 patients diagnosed with myocardial injury during hospitalization. Non-survivors had significant higher levels of cardiac indices than survivors, including cTnI, aspartic transaminase (AST) and lactate dehydrogenase (LDH). Elevated levels of cTnI (HR = 1.058, 95% CI1.032-1.085), AST (HR = 1.191, 95% CI1.150-1.234) and LDH (HR = 1.019, 95% CI1.009-1.029) predicted risk of early in-hospital mortality. cTnI model performed best, with area under curve of 0.850 (0.774-0.926) and concordance index of 0.842, respectively. Statistical differences were found between high and low levels of cTnI for mortality (P<0.001) using 0.35 ng/mL as the optimal cut-off.Conclusion:
The risk of early in-hospital death can be predicted by cTnI. Clinical doctors should remind vigilant concerning the elevation of cardiac enzyme as soon as possible.
Full text:
1
Collection:
01-internacional
Database:
MEDLINE
Language:
En
Journal:
Infect Drug Resist
Year:
2024
Document type:
Article
Publication country:
NEW ZEALAND
/
NOVA ZELÂNDIA
/
NUEVA ZELANDA
/
NZ