Abstract
Abstract
Background
Detection of cardiac abscess in patients with infective endocarditis (IE) is important for surgical planning and decision-making. While transesophageal echocardiography (TEE) is the method of choice for abscess detection, in many cases uncertainty remains. Since IE abscesses often affect ventricular muscle, biomarker release may be a helpful adjunct for abscess detection.
Purpose
We assessed the accuracy of TEE and the value of combining it with preoperative measurement of cardiac Troponin (cTn) in the diagnosis of abscess, using surgical inspection as a reference.
Methods
Data collected in the electronic clinical report form for patients included in the prospective, multicenter, randomized, controlled trial (REMOVE Trial) were analyzed. Continuous variables are presented as median (25th–75th percentile) and categorical data as absolute and relative frequencies. The diagnostic performance of TEE was identified by comparing preoperative TEE findings with intraoperative findings. The concordance rate is expressed as a percentage. Receiver operating characteristic (ROC) curve was used to assess the ability of cTn to predict the presence of abscess intraoperatively. Area under curve (AUC) with 95% confidence interval as well as sensitivity and specificity are provided as measures of accuracy. Youden index was used to estimate an optimal cut-off value for cTn.
Results
Among the 282 patients included, 84 had abscesses detected intraoperatively. The sensitivity and specificity of TEE to detect abscess were 58.3% and 86.4%, respectively. cTn-I and cTn-T were measured preoperatively in 113 and 96 patients, respectively. Median cTn-I in patients with abscess was 132 pg/ml (interquartile range (IQR): 54–1528) vs 53 pg/ml (IQR: 17–250) in patients without abscess, p=0.002. Median cTn-T in patients with abscess was 941 pg/ml (IQR: 387–1527) vs 697 pg/ml (IQR: 282–1423) in patients without abscess, p=0.353. The incidence of recent preoperative myocardial infarction was not different between patients with or without abscess (7.1% vs 5.6%, respectively, p=0.593). The AUC for predicting abscess was 0.69 (95% CI: 0.57 to 0.80, p=0.002) and 0.56 (95% CI: 0.43 to 0.69, p=0.353) for preoperative cTn-I and cTn-T, respectively. Optimal cut-offs according to the Youden index are 38 pg/ml for cTn-I and 1632 pg/ml for cTn-T. By adding preoperative cTn-I or cTn-T to the TEE findings, the AUC increased to 0.82 (95% CI: 0.73 to 0.91, p<0.001) and 0.72 (95% CI: 0.59 to 00.84, p=0.001), respectively.
Conclusion
The results suggest that using TEE alone is a poor method for abscess detection in IE. Adding preoperative troponin values to TEE findings significantly improved IE abscess detection.
Funding Acknowledgement
Type of funding sources: Public grant(s) – National budget only. Main funding source(s): The Federal Ministry of Education and Research (BMBF), Germany