Computed tomography improves the differentiation of infectious mediastinitis from normal postoperative changes after sternotomy in cardiac surgery
- 159 Downloads
To identify CT parameters independently associated with infectious mediastinitis after cardiac surgery and to improve the discrimination of patients with acute infection from those with normal postoperative changes.
In this single-center, retrospective, observational cohort study, we evaluated thoracic CT scans of poststernotomy cardiac surgery patients. Inclusion criteria were clinically suspected mediastinitis, unclear CT signs (e.g., retrosternal mass), and subsequent deep revision surgery. Revision surgery and microbiological samples determined the mediastinitis status. Overall, 22 qualitative and quantitative CT imaging parameters were assessed and associated with infectious mediastinitis in univariate and multivariate regression models. Discriminative capacity and incremental value of the CT features to available clinical parameters were determined by AUC and likelihood-ratio tests, respectively.
Overall 105 patients (82% men; 67.0 ± 10.3 years) underwent CT and deep revision surgery. Mediastinitis was confirmed in 83/105 (79%) patients. Among available clinical parameters, only C-reactive protein (CRP) was independently associated with infectious mediastinitis (multivariate odds ratio (OR) (per standard deviation) = 2.3; p < 0.001). In the CT, the presence of free gas, pleural effusions, and brachiocephalic lymph node size were independently associated with mediastinitis (multivariate ORs = 1.3–6.3; p < 0.001–0.039). Addition of these CT parameters to CRP increased the model fit significantly (X2 = 17.9; p < 0.001; AUC, 0.83 vs. 0.73).
The presence of free gas, pleural effusions, and brachiocephalic lymph node size in CT is independently associated with infectious mediastinitis in poststernotomy patients with retrosternal mass. These imaging features may help to differentiate mediastinitis from normal postoperative changes beyond traditional clinical parameters such as CRP.
• Presence of free gas, pleural effusions, and brachiocephalic lymph node size on CT are associated independently with infectious mediastinitis.
• Combination of these CT parameters increases the discriminatory capacity of clinical parameters such as CRP.
KeywordsMultidetector computed tomography Mediastinitis Sternotomy Cardiac surgery
Intra-class correlation coefficient
Region of interest
Volume of interest
BF received unrelated funding from the German Research Foundation (DFG) project 290004377 (FO 993/1).
The authors state that this work has not received any funding.
Compliance with ethical standards
The scientific guarantor of this publication is Prof. Dr. Matthias Gutberlet.
Conflict of interest
The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article.
Statistics and biometry
One of the authors has significant statistical expertise. No complex statistical methods were necessary for this paper.
Written informed consent was waived by the Institutional Review Board.
Institutional Review Board approval was obtained.
• performed at one institution
- 7.Chan M, Yusuf E, Giulieri S et al (2016) A retrospective study of deep sternal wound infections: clinical and microbiological characteristics, treatment, and risk factors for complications. Diagn Microbiol Infect Dis 84:261–265. https://doi.org/10.1016/j.diagmicrobio.2015.11.011 CrossRefGoogle Scholar
- 12.Jolles H, Henry DA, Roberson JP, Cole TJ, Spratt JA (1996) Mediastinitis following median sternotomy: CT findings. Radiology 201:463–466. https://doi.org/10.1148/radiology.201.2.8888241
- 14.Yamaguchi H, Yamauchi H, Yamada T, Ariyoshi T, Aikawa H, Kato Y(2001) Diagnostic validity of computed tomography for mediastinitis after cardiac surgery. Ann Thorac Cardiovasc Surg 7:94–98Google Scholar
- 17.Carrol CL, Jeffrey RB Jr, Federle MP, Vernacchia FS (1987) CT evaluation of mediastinal infections. J Comput Assist Tomogr 11:449–454Google Scholar
- 18.Elia S, End A, Canalis E, Lode H, Granai Vittoria A, Petrella G (2013) Mediastinitis: causes, management and outcomes. In: Rohde G, Subotic D (eds) Complex pleuropulmonary infections. European Respiratory Society, Norwich, UK, pp 122–140Google Scholar
- 20.Herrmann M, Kniehl E, Mauch H, Podbielski A (2007) MiQ: Qualitätsstandards in der mikrobiologisch-infektiologischen Diagnostik, MiQ Grundwerk Heft 1–29. Urban & Fischer, MunichGoogle Scholar
- 23.Kim SH, Chung JH, Kwon BJ, Song SW, Choi WS (2014) The associations of epicardial adipose tissue with coronary artery disease and coronary atherosclerosis. Int Heart J 55:197–203Google Scholar
- 24.McWilliams SR, O’Connor OJ, McGarrigle AM et al (2012) CT-based estimation of intracavitary gas volumes using threshold-based segmentation: in vitro study to determine the optimal threshold range: gas volume estimation with CT. J Med Imaging Radiat Oncol 56:289–294. https://doi.org/10.1111/j.1754-9485.2012.02375.x CrossRefGoogle Scholar
- 25.Riquet M, Le Pimpec-Barthes F, Hidden G (2001) Lymphatic drainage of the pericardium to the mediastinal lymph nodes. Surg Radiol Anat 23:317–319. https://doi.org/10.1007/s00276-001-0317-2
- 27.Klein AL, Abbara S, Agler DA et al (2013) American Society of Echocardiography clinical recommendations for multimodality cardiovascular imaging of patients with pericardial disease: endorsed by the Society for Cardiovascular Magnetic Resonance and Society of Cardiovascular Computed Tomography. J Am Soc Echocardiogr 26:965–1012.e15. https://doi.org/10.1016/j.echo.2013.06.023 CrossRefGoogle Scholar