Outcomes after lobectomy using thoracoscopy vs thoracotomy: a comparative effectiveness analysis utilizing the Nationwide Inpatient Sample database

OBJECTIVES: We examined the Nationwide Inpatient Sample (NIS) database to compare short-term postoperative outcomes following open and thoracoscopic lobectomy. Thoracoscopic (video-assisted thoracic surgery) lobectomy has been demonstrated to be associated with fewer postoperative complications compared with open thoracotomy lobectomy in several large case series. However, as no randomized trial has been performed, there are many who question this. METHODS: We examined the NIS database for all patients undergoing lobectomy as their principal procedure either via thoracoscopic or open thoracotomy from 2007 to 08. We compared the postoperative outcomes of these two groups of patients after propensity matching these groups based on several preoperative variables. RESULTS: Over a 2-year-period, 68 350 patients underwent a lobectomy by either thoracoscopy [n = 10 554 (15%)] or thoracotomy [n = 57 796(85%)]. Thirty-two percent of thoracoscopic lobectomies (n = 3421) were performed in either rural or non-teaching urban centres. Although in propensity-matched cohorts there was no difference in operative mortality, thoracoscopic lobectomy was associated with a lower incidence of postoperative complications [n = 4146 (40.8%) vs n = 13 913 (45.1%), P < 0.001] and shorter length of stay (5.0 vs 7.0 days; P < 0.001) compared with open lobectomy. Specifically, the incidences of supraventricular arrhythmias, myocardial infarction, pulmonary embolism and empyema were lower. CONCLUSIONS: This large national database study demonstrates that thoracoscopic lobectomy is associated with fewer in-hospital postoperative complications compared with open lobectomy. Thoracoscopic lobectomy appears to be applicable to the wider general thoracic surgical community.