Effects of epidural analgesia on cancer recurrence and long-term mortality in patients after non-small-cell lung cancer resection: a propensity score-matched study.

OBJECTIVES

Previous studies showed reductions in recurrence and mortality rate of several cancer types in patients receiving perioperative epidural analgesia. This study aimed to investigate the effects of thoracic epidural analgesia on oncological outcomes after resection for lung cancer.

DESIGN

Retrospective study using propensity score matching methodology.

SETTING

Single medical centre in Taiwan.

PARTICIPANTS

Patients with stages I-III non-small-cell lung cancer undergoing primary tumour resection between January 2005 and December 2015 and had either epidural analgesia, placed preoperatively and used intra- and postoperatively, or intravenous analgesia were evaluated through May 2017.

PRIMARY AND SECONDARY OUTCOME MEASURES

Primary endpoint was postoperative recurrence-free survival and secondary endpoint was overall survival.

RESULTS

The 3-year recurrence-free and overall survival rates were 69.8% (95% CI 67.4% to 72.2%) and 92.4% (95% CI 91% to 93.8%) in the epidural group and 67.4% (95% CI 62.3% to 72.5%) and 89.6% (95% CI 86.3% to 92.9%) in the non-epidural group, respectively. Multivariable Cox regression analysis before matching demonstrated no significant difference in recurrence or mortality between groups (adjusted HR: 0.93, 95% CI 0.76 to 1.14 for recurrence; 0.81, 95% CI 0.58 to 1.13 for mortality), similar to the results after matching (HR: 0.97, 95% CI 0.71 to 1.31; 0.94, 95% CI 0.57 to 1.54). Independent risk factors for both recurrence and mortality were male, higher pretreatment carcinoembryonic antigen level, advanced cancer stage, poor differentiation, lymphovascular invasion, microscopic necrosis and postoperative radiotherapy.

CONCLUSIONS

Thoracic epidural analgesia was not associated with better recurrence-free or overall survival in patients receiving surgical resection for stages I-III non-small-cell lung cancer.