Which pneumothorax volume/size measurement method can best predict surgical indication in primary spontaneous pneumothorax patients? A comparison of six different methods.

OBJECTIVES

We determined whether patients requiring surgery could be identified using pneumothorax volume/size in the first episode of primary spontaneous pneumothorax (PSP) and confirmed which of six different measurement methods of pneumothorax volume/size best predicted surgical indication (prolonged air leak, PAL, or ipsilateral recurrence).

METHODS

Between 2013 and 2017, 282 patients with PSP were analysed. The volume of the pneumothorax (%) was calculated using volumetric methods (Collins, Rhea, Kircher, Light), while its size (large or small) was determined using the British Thoracic Society (BTS) and American College of Chest Physicians (ACCP) guidelines for the first PSP episode. The optimal cut-off points of pneumothorax volume in patients requiring surgery were based on the most sensitive and specific scores, calculated using receiver operating characteristic analysis.

RESULTS

Surgery was indicated in 162 patients (57.4%) because of PAL (n = 92) or ipsilateral recurrence (n = 70). Kircher's method had a better area under the curve (AUC) value than the others (AUC = 0.731, 95% confidence interval [CI] 0.676-0.782). Comparing the calculated cut-off points for each volumetric method and the known thresholds from the BTS/ACCP, the best predictor for surgical indications (for both PAL and ipsilateral recurrence) was Kircher's method (odds ratio = 5.636, 95% CI 3.338-9.516, p < 0.001). The four volumetric methods were significantly correlated with pneumothorax volume (rho ≥ 0.8, p < 0.001, for all comparisons).

CONCLUSIONS

Pneumothorax volume/size calculated in the first PSP episode may help predict surgical candidates, with Kircher's method being the best predictor. There was a correlation between all the volumetric methods of the pneumothorax volume.