Metal-supported versus conventional ultra-fine chest drains after pulmonary resection: a historical cohort study of drainage outcomes and placement learning curves
Yaxuan Liu, Kexin Ma, Ruijiang Lin, Ning Wei, Songla Bai, Liangliang Yang, Jiacheng Su, Diego-Gonzalez Rivas, Minjie Ma, Biao HanAbstract
Objectives
Ultra-fine chest drains are susceptible to kinking and obstruction. We compared drainage outcomes and device-placement learning curves for a 10-F metal-supported, multi-side-port drain and a conventional 12-F drain after pulmonary resection.
Methods
This single-centre historical cohort comprised 5,235 patients treated between 2006 and 2023. A conventional drain was used in 2,614 patients during 2006-2017, whereas 2,621 patients received the metal-supported drain during 2018-2023. The co-primary outcomes were additional catheter drainage and encapsulated pleural effusion. Firth logistic regression adjusted for age, gender, BMI, and ASA score. For the learning-curve analysis, a designated observer timed the first 261 consecutive procedures after adoption of each device, from the initial skin puncture to fixation of the final drain.
Results
Additional catheter drainage was required in 7/2,621 (0.3%) patients in the metal-supported group and 30/2,614 (1.1%) in the conventional group (adjusted odds ratio [aOR] 0.24, 95% confidence interval [CI] 0.10-0.51; P < 0.001). Encapsulated pleural effusion occurred in 15/2,621 (0.6%) versus 34/2,614 (1.3%) patients, respectively (aOR 0.44, 95% CI 0.24-0.79; P = 0.006). Mean total placement time was 24.1±7.8 min with the metal-supported drain and 44.3±7.7 min with the conventional drain. The CUSUM curves reached their peaks at cases 90 and 71, respectively, and then declined. Although drainage duration, length of stay, and pain differed statistically, the absolute differences were small.
Conclusions
The metal-supported drain was associated with fewer drainage-related reinterventions, fewer encapsulated pleural effusions, and shorter placement times. Because device use was determined by treatment era, residual temporal confounding precludes firm causal inference.