OBJECTIVE To compare the causes of failure between external and endoscopic dacryocystorhinostomy (DCR) techniques for the treatment of lacrimal obstruction. METHODS A retrospective cohort study. METHODS The study population consisted of 53 consecutive patients who underwent revision endoscopic DCR from 2002 to 2013 for lacrimal duct obstruction. Identified causes of previous DCR failure were compared between patients whose initial surgery was performed through an external versus an endoscopic approach. RESULTS Reasons for surgical failure after external (n = 21) versus endoscopic (n = 32) DCR included cicatricial closure of the internal lacrimal ostium (52.4 versus 53.1%; p = 0.96), inadequate removal of bone overlying the lacrimal sac (23.8 versus 9.4%; p = 0.15), sump syndrome (9.5 versus 9.4%; p = 0.99), and intranasal adhesions (65 versus 37.5%; p = 0.05). Adhesions that involved the middle turbinate were a particularly impactful cause of failure when the DCR was performed through an external versus the endoscopic approach (57.1 versus 28.1%; p = 0.04). Septoplasty was more likely to be needed at the time of revision surgery if the initial procedure was performed externally (71.1 versus 15.6%; p = 0.02). Surgical success rates for revision DCR were comparable between the groups (75.0% external versus 73.3% endoscopic; p = 0.90), with a mean follow-up of 12.7 months. CONCLUSIONS DCR failure associated with intranasal adhesions was more likely to occur when the surgery was performed through an external rather than an endoscopic approach. Endoscopic instrumentation allowed for identification and correction of intranasal pathology at the time of DCR, including an enlarged middle turbinate or a deviated septum, which may improve surgical outcome.