Abstract
Background Digital devices could help clinical practice measuring the air leak, but their role is still debated. Aim of this study is to test advantages using these devices.
Methods From June 2012 to May 2015, we enrolled 95 patients undergoing lobectomy or wedge resection in a prospective randomized trial. Patients were divided into two groups: group D (digital, 50 patients) evaluated with digital device and group E (empirical, 45 patients) evaluated with water seal. Logistic regression analysis and receiver operating characteristic (ROC) curve analysis were used to select and assess predictors of persistent air leak. In group D, chest drain was removed if the air leak was absent or < 0.5 L/h for 12 consecutive hours and in group E if clinical signs of air leak were absent.
Results Mean hospitalization and mean chest tube stay was 6.0 ± 3.3 and 4.4 ± 3.2 days, respectively, prolonged air leak occurred in nine (8%). Hospitalization was 5.8 ± 2.5 versus 6.2 ± 4.2 days (p = 0.5), while mean chest tube stay was 4.1 ± 2.0 versus 4.6 ± 3.5 days (p = 0.4) in group D and group E, respectively. Clamping test was needed in one patient in group D and in seven in group E (p = 0.019). At multivariate analysis, heart disease (p < 0.0001), lobectomy (p < 0.0001), fused fissure (p < 0.0001), and air leakage in first postoperative (p.o.) day were predictors of persistent air leak (AUC on the ROC curve of 69.7%, sensibility: 77.8%). In group D, an air leak value > 0.2 L/h with spikes over 0.5 L/h in third p.o. was predictive of persistent air leak, with chest tube duration of 7.73 ± 5.20 versus 4.32 ± 1.33 days (AUC: 83%, sensitivity: 80%, p < 0.0001).
Conclusion In our experience, digital devices reduced observer variability and mistakes in chest tube management, thus identifying patients at risk for prolonged air leak.
Keywords
air leak - digital device - pneumothorax - prolonged air leak
