Ganapathi Asvin M, Speicher Paul J, Castleberry Anthony W, Englum Brian R, Osho Asishana A, Davis R Duane, Hartwig Matthew G
Department of Surgery, Duke University Medical Center, Durham, North Carolina.
Department of Surgery, Duke University Medical Center, Durham, North Carolina.
Ann Thorac Surg. 2014 Dec;98(6):1922-8; discussion 1928-9. doi: 10.1016/j.athoracsur.2014.06.042. Epub 2014 Oct 17.
Lung transplantation in patients with prior lobectomy or pneumonectomy is not well understood. Using the United Network for Organ Sharing (UNOS) database, we address the impact of prior major lung resection on lung transplantation outcomes.
Retrospective review of adult lung transplants from October 1999 to December 2011 in the UNOS database identified 15,300 lung transplants; 102 patients had undergone major lung resection, defined as prior pneumonectomy (n = 22) or lobectomy (n = 80). Propensity match with nonparametric 3:1 nearest-neighbor matching algorithm adjusted for treatment-level differences. After matching, the primary outcome (90-day mortality) and secondary outcome (airway dehiscence, need for dialysis, length of stay more than 25 days) were assessed with univariable and multivariable methods. Subanalysis of pneumonectomy and lobectomy individually compared with matched nonresection patients was done in a similar manner. The Kaplan-Meier method estimated long-term survival.
After matching, no significant differences were noted between groups for recipient, donor, or operative characteristics. There were 10 double lung and 12 single lung transplants after pneumonectomy and 51 double lung and 29 single lung transplants after lobectomy. Mortality at 90 days was 13.9% (n = 14) for the resection group and 8.6% (n = 1,247) for the nonresection group (p = 0.09). After matching, a significant increase was noted in 90-day mortality (p = 0.017) and perioperative dialysis (p = 0.039) for the resection versus nonresection patients. Dialysis was significantly higher among pneumonectomy patients (p = 0.03). No long-term survival difference was observed (p = 0.514).
After propensity-matching, resection was associated with increased 90-day mortality and dialysis. Careful patient selection is necessary with patients who have undergone prior major lung resection, given their increased risk of perioperative mortality and dialysis.
对于曾接受过肺叶切除术或全肺切除术的患者进行肺移植的情况，目前尚未完全了解。我们利用器官共享联合网络（UNOS）数据库，探讨既往大肺切除术对肺移植结局的影响。
对1999年10月至2011年12月在UNOS数据库中的成人肺移植进行回顾性研究，共识别出15300例肺移植；102例患者曾接受过大肺切除术，定义为既往全肺切除术（n = 22）或肺叶切除术（n = 80）。采用非参数3:1最近邻匹配算法进行倾向匹配，以调整治疗水平差异。匹配后，采用单变量和多变量方法评估主要结局（90天死亡率）和次要结局（气道裂开、需要透析、住院时间超过25天）。对全肺切除术和肺叶切除术分别与匹配的未行切除术患者进行亚组分析，方法类似。采用Kaplan-Meier法估计长期生存率。
匹配后，两组在受者、供者或手术特征方面未发现显著差异。全肺切除术后有10例双肺移植和12例单肺移植，肺叶切除术后有51例双肺移植和29例单肺移植。切除组90天死亡率为13.9%（n = 14），非切除组为8.6%（n = 1247）（p = 0.09）。匹配后，切除组与非切除组相比，90天死亡率（p = 0.017）和围手术期透析（p = 0.039）显著增加。全肺切除术患者的透析率显著更高（p = 0.03）。未观察到长期生存差异（p = 0.514）。
倾向匹配后，切除术与90天死亡率和透析增加相关。鉴于既往接受过大肺切除术的患者围手术期死亡和透析风险增加，对这类患者进行仔细的患者选择是必要的。
