Adverse events after protamine administration in patients undergoing cardiopulmonary bypass: risks and predictors of under-reporting.

A retrospective cohort study of patients undergoing cardiopulmonary bypass over 2 years at a single hospital was performed to determine the rate and predictors of attributing events to protamine and reporting them to a well-developed, hospital-based adverse drug reaction (ADR) program. Overall, 123 (12.9%) of 952 procedures were associated with an adverse event. Thirteen percent of these events were attributed to protamine in the medical record and 2.4% were reported to the ADR program. Only 19% of events attributed to protamine were reported. Even after excluding events with other measurable etiologies, the incidence of adverse events (2.6%) was still higher than that of attributed (0.8%) or reported (0.3%) events. Events that manifest as pulmonary hypertension occurred sooner after protamine, or were severe were significantly more likely to be attributed to protamine. These predictors demonstrated similar relationships with the probability of reporting events. Reliance on reporting of adverse events could not only underestimate the risk of these events, but might bias studies by identifying non-representative events.