Abstract
Background Prolonged air leak (PAL) is a major cause of morbidities and increased treatment costs following lung resection. The Goddard score (GS) quantifies pulmonary emphysema on computed tomography, a risk factor for PAL, from 0 to 24.
Methods We evaluated the GS as a predictor of PAL in pulmonary segmentectomies by retrospectively analyzing 131 patients with anatomical segmentectomy. We identified predictors of PAL and performed logistic regression.
Results Eighty-three percent of patients had a history of smoking. The mean air leak duration was 4.2 days; 16% had PAL (>7 days). Median GS was 1. The optimal cutoff was GS ≥ 6 (area under the curve AUC = 0.625). GS did not correlate with air leak duration. In combination, body mass index, cumulative pack-years, immunosuppression, adhesiolysis, operation time, and GS ≥ 6 best predicted PAL with AUC = 0.914. Omitting GS still leads to an AUC = 0.864.
Conclusion The GS does not add enough information to be of practical value in anatomic segmentectomies, but it is of academic significance to quantify and compare lung emphysema in the context of clinical studies.
Keywords
computed tomography - prolonged air leak - emphysema - thoracic surgery - video-assisted thoracoscopic surgery
