Prevention and management of cytomegalovirus (CMV) disease after hematopoietic stem cell transplantation (HSCT) is critically important, but clinical practices have historically been heterogeneous. The Centers for Disease Control (CDC), as part of a larger clinical practice guideline initiative, has published evidence-based recommendations, but their effect on clinical practice has not been assessed. A survey was sent to all Canadian HSCT program directors to describe current practices. Current practices were subsequently compared with CDC guideline recommendations and with a Canadian survey published before the CDC guidelines. When current practices did not conform to guideline recommendations, a literature review was performed to determine whether current practices were supported by evidence published after the CDC guidelines. The survey response rate was 100%. Variability in practices was observed in several aspects of patient care, including (1) indications for CMV-negative blood products, (2) surveillance tests used to detect CMV infection, (3) duration of surveillance testing, (4) duration of maintenance treatment after preemptive therapy was initiated, and (5) treatment of CMV disease. Overall adherence to guideline recommendations was good, especially when they were supported by high-quality data. However, deviations from guideline recommendations were observed: (1) most Canadian allogeneic programs used shorter courses of preemptive therapy; (2) prevention measures aimed at late CMV disease were not systematically applied at most allogeneic programs; and (3) most autologous HSCT programs did not administer preemptive therapy even in high-risk recipients. When deviations occurred, recent evidence supported current practices in some instances (shorter courses of preemptive therapy) but not in others (absence of strategies to prevent late CMV disease). Compared with practices in Canada before publication of the CDC guidelines, a higher proportion of programs used CDC-recommended surveillance tests and treatment for CMV infection. Current practices in Canada remain heterogeneous. Discrepancies between current practices and CDC guideline recommendations occurred in situations either in which practices had changed in response to recently published data or in which evidence supporting a recommendation was poor. These results suggest an urgent need for the development of well-designed clinical trials. Incorporation of recent data into updated guidelines may be appropriate.