PurposeThis study evaluated the performance of donor corneoscleral rim cultures for predicting infection after corneal transplantation, and determines if there is a correlation between positive corneoscleral rim cultures and postkeratoplasty infection.Design and data sourcesThis was a systematic review, prognostic accuracy analysis, and cost-effectiveness analysis. Databases searched were: Medline (Ovid), Embase (Ovid), CINAHL, Cochrane Library, Web of Science, and BioSis Previews. Grey literature was also explored.Materials and methodsA systematic review was conducted to locate published and unpublished studies. All studies examining corneal button contamination and its association with endophthalmitis and keratitis posttransplantation were included. Extracted data were used to calculate sensitivity, specificity, positive predictive value, and negative predictive value. Cost data from the London Laboratory Services Group in London, ON were used to calculate the cost-effectiveness of culturing donor rim cultures.ResultsOf 7,870 grafts, 954 had a positive rim culture (12.1%), with 12 patients going on to develop keratitis or endophthalmitis (1.3%). The prevalence of keratitis and endophthalmitis in this study was 0.15%, and the positive predictive value 1.5%. Of the 12 infections, nine were fungal and three bacterial. The estimated cost of a positive and negative test result was CAD$45.99 and $14.15, respectively. The cost to run all 7,870 tests was estimated to be $141,735.86, with an incremental cost-effectiveness ratio of $40,215.70.ConclusionThere was a significant divergence between bacterial and fungal rim-culture results. Bacterial cultures predicted clinical infection poorly, did not change management, and were expensive. Fungal cultures predicted clinical infection in over 10% of patients, had the potential to change management, and were 40% less expensive than full rim culturing (bacterial and fungal tests). Fungal rim cultures may be considered in areas where fungal infection rates are high.