Virtual Library

Start Your Search

Hironobu Hoshino



Author of

  • +

    P2.01 - Advanced NSCLC (ID 159)

    • Event: WCLC 2019
    • Type: Poster Viewing in the Exhibit Hall
    • Track: Advanced NSCLC
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall
    • +

      P2.01-28 - The Predictive Factor for Prolonged Air Leakage Over 48 Hours Using Log Data of Digital Drainage System (ID 2375)

      10:15 - 18:15  |  Presenting Author(s): Hironobu Hoshino

      • Abstract

      Background

      Air leakage is one of the most common complications after pulmonary resection and many risk factors of air leakage have been reported in the past. However, there are few studies about the objective predictive factors. Digital drainage system (DDS) has enabled us to measure the flow of air leakage after pulmonary resection objectively and quantitatively. We aimed to elucidate the predictive factors of prolonged air leakage (PAL: continuing air leakage over 48 hours) after surgery using the log data of DDS.

      Method

      The presence of air leakage was defined as being 20ml/min or more and/or spike in the flow on DDS. Chest tubes were removed at the time of continuing for 8 hours at less than 20ml/min and pleural effusion of less than 300ml/day. The 593 patients underwent pulmonary resection and monitored by DDS postoperatively from May 2016 to January 2018. The 92 patients had air leakage at the time of transferring to intensive care unit or recovery room (postoperative air leakage: POAL). The log data of these 89 patients were analyzed retrospectively and We examined their characteristics using univariate and multivariate manners in logistic regression analysis.

      Result

      The median age at the time of pulmonary resection in these 89 patients (72 men and 17 women) was 72 years (range, 40 to 86 years). The 75 patients (84%) had smoking history. The 17 patients (19%) had diabetes mellitus. The 49 patients (55%) had emphysema. Surgical procedures were a lobectomy in 71 patients, a segmentectomy in 4 patients, and a wedge resection in 14 patients. Fibrin glue was used during surgery in 45 patients (51%). The mean flow of POAL was 70.4 ml/min (range, 20.2-1267.9). The mean duration of air leakage was 60 hours (range, 9-257). In univariate analysis, diabetes mellitus (DM, p=0.0284, OR; 3.450), use of fibrin glue (p=0.0452, OR; 0.411), and POAL (p=0.0101) were statistically significant. In the final multivariate model, DM (p=0.0441) and flow of POAL (p=0.0228) were independently associated with PAL. In ROC curve, considering less than 20% false positive rate, the optimal cutoff in patients with DM was 100ml/min (sensitivity; 67%, specificity; 82%). The optimal cutoff in patients without DM was 150ml/min (sensitivity; 53%, specificity; 84%).

      Conclusion

      The POAL flow of 150ml/min or more in patients without DM and that of 100 ml/min or more in patients with DM will be likely to develop PAL with high probability. Pleurodesis may be considered earlier for this population.