Assessment of the interobserver reliability in discharge decision-making for patients undergoing thoracic surgery based on chest ultrasound compared to chest X-ray findings - 23/03/26

Doi : 10.1016/j.resmer.2026.101265

Zaineb Guesmi ^a , Martin Faure ^b , Julia Ballouhey ^a , Lina Atlagh ^a , François Bertin ^c , Anaelle Chermat ^c , Caroline Rivera ^d,^e , François Vincent ^a , Benoit Aguado ^a , Jérémy Tricard ^c,^d,^e,^⁎

^a Department of Respiratory Medicine and Allergology, Limoges University Hospital, 16 rue Bernard Descottes, 87042 Limoges, France.

^b Department of Clinical and Research Data Center, Limoges University Hospital, 2 av. Martin-Luther King, 87042 Limoges, France.

^c Department of Cardiac and Thoracic Surgery, Limoges University Hospital, 16 rue Bernard Descottes, 87042 Limoges, France.

^d Department of Vascular and Thoracic Surgery, Côte Basque Hospital, 13 avenue de l'interne Jacques Loëb, 64109 Bayonne, France.

^e G-ECHO-Chir group of the French Society of Thoracic and Cardiovascular Surgery.

^⁎ Corresponding author: Jérémy Tricard, Cardiac and Thoracic Surgery Department, Dupuytren 2 Hospital, 16 rue Bernard Descottes, 87042 Limoges, France. tel: +33555056391; fax: +33555056384. Cardiac and Thoracic Surgery Department Dupuytren 2 Hospital, 16 rue Bernard Descottes Limoges 87042 France

Sous presse. Manuscrit accepté. Disponible en ligne depuis le Monday 23 March 2026
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Abstract

Background

Safe authorization of patient discharge after thoracic surgery must be ensured following chest tube removal. This decision is commonly based on chest X-ray findings. Chest ultrasound (US) has emerged as a potentially valuable alternative. The aim of this study was to assess the interobserver reliability of discharge authorization decisions based on chest X-ray compared with chest US findings.

Methods

We conducted a single-center prospective study including 60 consecutive patients undergoing thoracic surgery. A semi-quantitative evaluation of pneumothorax, pleural effusion, subcutaneous emphysema, and lung consolidation was performed in a blinded fashion using both chest X-rays (interpreted by surgeons) and chest US (performed by pulmonology residents) at three time points: day 0, day 1, and after chest tube removal. Discharge eligibility was determined using a score derived from imaging findings.

Results: Among the 60 patients (mean age

60.7 ± 14 years, 70% male), 42 (69%) underwent lung resection, and the remaining patients had mediastinal or pleural surgery. Both imaging modalities identified the same three patients who did not meet the theoretical discharge criteria (concordance rate: 100%, Po = 57+3/60, p < 0.0001). Agreement according to Fleiss’ kappa was excellent for discharge decisions, but poor to slight for the evaluation of postoperative complications using chest X-ray and US.

Conclusions

Chest US reliably identifies patients requiring continued monitoring or repeat drainage after chest tube removal in the postoperative thoracic surgery setting. However, the interpretation of mild to moderate complications differed substantially between chest X-ray and chest US.