Limitations include the small patient cohort, the inherent retrospective nature of chart reviews that may predispose internal bias, and the fact that data were originally recorded for the purposes of clinical care as opposed to clinical research.