OBJECTIVES:  To conduct a systematic review of the economic impact on glaucoma-associated services, comparing topical medications ("drop") vs. laser trabeculoplasty (LT), or any glaucoma surgery ("non-drop") therapies. METHODS:  We used a structured search of Pubmed to retrieve randomized controlled trials, cohort or case-control studies, and economic models published between January 1996 and August 2016. Inclusion criteria required a direct comparison of a drop vs. non-drop treatment. We extracted data on healthcare resource use and direct costs of glaucoma associated services (ophthalmologist visits, surgical procedures, medications, exams and other services); and indirect costs (production loss and absenteeism) of topical medications drop vs. non-drop therapies. RESULTS:  Twelve studies met the inclusion criteria (eight economic models, three retrospective cohort studies and one prospective case series). Analyses were from eight countries across four continents (North America, South America, Europe and Africa). Of the non-drop therapies included in the review, LT was the comparator in six studies, traditional surgical procedures were comparators in four, minimally invasive surgeries in one study; and one study assessed either surgery or laser procedures compared with drops. In four studies, the comparison between drops and LT were made in a primary glaucoma diagnosis population, with the other eight in populations with more advanced or refractory disease. All 12 publications reported direct costs, only two of the twelve included indirect costs or reported health care resource utilization. Potential cost-savings with non-drop treatments was seen in eight of the twelve studies, including all four studies comparing a primary laser procedure to initiation of drop therapy. CONCLUSIONS:  Open-angle glaucoma treatment using non-drop therapies may have the potential to be cost-effective when compared to drop therapy for both initial- and drop-refractory treatment for open-angle glaucoma. However, variability between studies and numerous potential intra-study limitations does not allow for a consensus recommendation and supports the need for further evaluation.