Abstract 473 Sensitivity and specificity of diagnostic methods in patients with assumed infective endocarditis and histopathologically examined valves

Background:

Diagnostic accuracy and the time to final diagnosis of infective endocarditis (IE) is a major challenge for clinicians despite developments of diagnostic methods. Perioperative valve histopathology (HP) is the gold standard for diagnostic accuracy. In this study, we aimed to evaluate the prospective value of diagnostic tools for patients with IE.

Methods:

Between Jan 2015 and Dec 2019, 97 patients with preoperatively diagnosed IE, following the current guidelines, and conducted valve histopathology were enrolled in this retrospective study. Sensitivity and specificity of microbiological testings (valve culture and PCR, blood culture), the presence of vegetation (transesophageal echocardiography) and F-18 cardiac PET/CT imaging data as well as quantitative parameters for inflammation, cardiac stress and the echocardiographic measured size of vegetation were evaluated for HP positive (HP+) and negative (HP-) patients.

Results:

82 of 97 patients were HP+. Sensitivity/specificity for qualitative parameters were: valve culture 19/100% (CI 0.11-0.30/CI 0.73-1.00), valve PCR 56/93% (CI 0.44-0.68/CI 0.64-0.99), blood culture 88/40% (CI 0.78-0.93/0.17-0.67), presence of vegetation 96/0% (CI 0.89-0.99/CI 0.0-0.25), PET/CT 48/100% (CI 0.26-0.69/CI 0.39-1.00). Parameters for systemic inflammation and cardiac stress were significantly higher in HP+ group: WBC (11.8±2.4 vs. 9.1±1.3 /nL, mean±SD, p=0.006, Student’s t-test), CRP (39.5±19.5 vs. 95.8±56.1 mg/L, p=0.004), PCT (6.1±5.8 vs. 0.2±4.5 ng/mL, p<0.001) and NT-proBNP (13942±11866 pg/mL vs. 2076±2555 pg/mL, p<0.001). The mean size of the vegetation in HP+ group was significantly larger (14.6±3.7 vs. 10.9±1.3 mm, p=0.015).

Conclusion:

Due to manifold symptoms and various disciplines involved, the diagnosis of IE may be challenging. A single diagnostic method is not reliable in many patients. We found, that the echocardiographic detection of vegetation has a high sensitivity, whereas microbiological valve culture and PET/CT have a high specificity. Serum markers (particularly NT-proBNP and PCT) were strikingly higher in HP+ patients. We recommend using the mentioned markers in addition to the Duke criteria in order to facilitate and accelerate correct diagnosis of infective endocarditis. Valve culture and PCR should guide duration of postoperative antibiotic treatment.

Clin Res Cardiol (2021) DOI DOI https://doi.org/10.1007/s00392-021-01843-w
Sensitivity and specificity of diagnostic methods in patients with assumed infective endocarditis and histopathologically examined valves
C. Schach¹, J. Li², K. Menhart³, L. S. Maier¹, C. Schmid²
¹Klinik und Poliklinik für Innere Med. II, Kardiologie, Universitätsklinikum Regensburg, Regensburg; ²Herz-, Thorax- und herznahe Gefäßchirurgie, Universitätsklinikum Regensburg, Regensburg; ³Nuklearmedizin, Universitätsklinikum Regensburg, Regensburg;
Background: Diagnostic accuracy and the time to final diagnosis of infective endocarditis (IE) is a major challenge for clinicians despite developments of diagnostic methods. Perioperative valve histopathology (HP) is the gold standard for diagnostic accuracy. In this study, we aimed to evaluate the prospective value of diagnostic tools for patients with IE. Methods: Between Jan 2015 and Dec 2019, 97 patients with preoperatively diagnosed IE, following the current guidelines, and conducted valve histopathology were enrolled in this retrospective study. Sensitivity and specificity of microbiological testings (valve culture and PCR, blood culture), the presence of vegetation (transesophageal echocardiography) and F-18 cardiac PET/CT imaging data as well as quantitative parameters for inflammation, cardiac stress and the echocardiographic measured size of vegetation were evaluated for HP positive (HP+) and negative (HP-) patients. Results: 82 of 97 patients were HP+. Sensitivity/specificity for qualitative parameters were: valve culture 19/100% (CI 0.11-0.30/CI 0.73-1.00), valve PCR 56/93% (CI 0.44-0.68/CI 0.64-0.99), blood culture 88/40% (CI 0.78-0.93/0.17-0.67), presence of vegetation 96/0% (CI 0.89-0.99/CI 0.0-0.25), PET/CT 48/100% (CI 0.26-0.69/CI 0.39-1.00). Parameters for systemic inflammation and cardiac stress were significantly higher in HP+ group: WBC (11.8±2.4 vs. 9.1±1.3 /nL, mean±SD, p=0.006, Student’s t-test), CRP (39.5±19.5 vs. 95.8±56.1 mg/L, p=0.004), PCT (6.1±5.8 vs. 0.2±4.5 ng/mL, p<0.001) and NT-proBNP (13942±11866 pg/mL vs. 2076±2555 pg/mL, p<0.001). The mean size of the vegetation in HP+ group was significantly larger (14.6±3.7 vs. 10.9±1.3 mm, p=0.015). Conclusion: Due to manifold symptoms and various disciplines involved, the diagnosis of IE may be challenging. A single diagnostic method is not reliable in many patients. We found, that the echocardiographic detection of vegetation has a high sensitivity, whereas microbiological valve culture and PET/CT have a high specificity. Serum markers (particularly NT-proBNP and PCT) were strikingly higher in HP+ patients. We recommend using the mentioned markers in addition to the Duke criteria in order to facilitate and accelerate correct diagnosis of infective endocarditis. Valve culture and PCR should guide duration of postoperative antibiotic treatment.
https://dgk.org/kongress_programme/jt2021/aP1422.html