Thorac Cardiovasc Surg 2022; 70(S 01): S1-S61
DOI: 10.1055/s-0042-1742897
Oral and Short Presentations
Tuesday, February 22
Aorta and Lungs

New Smartphone Scoring App Precisely Predicts the Dignity of Pulmonary Nodules

F. Dörr
1   University Hospital of Cologne, Köln, Deutschland
,
A. Giese
2   University Hospital of Cologne, Cologne, Deutschland
,
H. Menghesha
1   University Hospital of Cologne, Köln, Deutschland
,
G. Schlachtenberger
1   University Hospital of Cologne, Köln, Deutschland
,
M. Heldwein
2   University Hospital of Cologne, Cologne, Deutschland
,
T. Wahlers
2   University Hospital of Cologne, Cologne, Deutschland
,
K. Hekmat
2   University Hospital of Cologne, Cologne, Deutschland
› Author Affiliations

Background: The classification of pulmonary lesions in terms of malignancy is at times controversial and MDT meetings tend to delay because of long debates. To remedy this, score systems have been developed predicting the likelihood of malignancy. Due to unfavorable variables, the existing systems such as “Brock model” lack precision. We developed the “LIONS PREY” (Lung lesION Score PREdicts malignancY), a new logistic scoring system, which precisely predicts malignancy of pulmonary nodules in percent.

Method: We prospectively evaluated all patients that were presented to the MDT of our university hospital between January 2013 and December 2020. Patients that presented with a pulmonary mass (> 40 mm) were excluded. After resection or CT-/EBUS-guided sampling, tissue was histologically analyzed. Availability of pathological result was mandatory upon study inclusion. Patients were grouped according to the dignity of their pulmonary nodule. Two groups were formed: Group A: Malignant pulmonary nodule (N = 238). Group B: Benign pulmonary nodule (N = 148). Initially 21 potential score parameters were derived from the patients’ medical history. In case of missing data, patients or relatives were contacted.

Results: After uni- and multivariate analysis we identified eight relevant score parameters. Age (years): Group A: 64.5 ± 10.2 versus Group B: 61.6 ± 13.8; (multivariate) p-value: 0.024. Nodule size (mm): 21.8 ± 7.5 versus 18.3 ± 7.9; 0.003. Nodule spiculation (%): 73.1 versus 41.9; <0.0001. Nodule solidity (%): 84.9 versus 62.8; <0.0001. Size dynamics (mm/3 months): 6.4 ± 7.7 versus 0.2 ± 0.9; <0.0001. Positive history of smoking (%): 92.0 versus 43.9; <0.0001. Pack-years: 35.1 ± 19.1 versus 21.3 ± 18.8; 0.041. Positive history of cancer (%): 34.9 versus 24.3; 0.023. The new system achieved excellent precision in predicting the malignancy of pulmonary nodules. Overall correct classification: 96.0%; Calibration (observed/expected ratio): 1.1. Discrimination (ROC analysis): AUC (95% CI) 0.943 (0.922–0.965).

Conclusion: The eight LIONS PREY variables are simple, routinely and reproducibly measurable and readily available. All MDT members could use the new score to reinforce and facilitate decision making. Especially inexperienced physicians have a tool that helps them focusing on essential parameters. Patients may find it easier to consent to surgery knowing the likelihood of pulmonary malignancy. The LIONS PREY App is available for free for iOS and Android devices.



Publication History

Article published online:
03 February 2022

© 2022. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany