Revised international surveillance case definition of transfusion-associated circulatory overload: a classification agreement validation study

Background Transfusion-associated circulatory overload (TACO) is a major cause of transfusion-related morbidity and mortality in countries with well developed transfusion services. The International Society of Blood Transfusion, the International Haemovigilance Network, and AABB (formerly American Association of Blood Banks), have developed and validated a revised definition of TACO. Methods International Haemovigilance Network-member haemovigilance systems (Australia, Austria, Denmark, Finland, Greece, India, Ireland, Italy, Japan, Malta, Netherlands, New Zealand, Norway, Slovenia, United Kingdom and United States) provided cases of respiratory complications categorised by their systems, including clinical parameters listed in the 2017 draft definition (part 1). Individual transfusion professionals were then invited to assess 24 case descriptions according to the draft definition (part 2). Positive and negative agreement and inter-rater agreement (kappa) were calculated. Based on validation results, cases were reanalysed and slight adjustments made to yield the final 2018 TACO definition. Findings In part 1, 16 (44%) of 36 haemovigilance systems provided 178 cases, induding 126 TACO cases. By use of the 2018 definition, 96 (76%) of 126 cases of TACO were in positive agreement. 19 (37%) of 52 cases were recognised as non-TACO respiratory complications. In part 2 (47 experts from 20 countries), moderate all-case agreement (kappa=0.43) and TACO-specific agreement (kappa=0.54) were observed. Excluding cases missing some clinical information (eg, N terminal pro-brain natriuretic peptide, distinctive chest x-ray findings, and relationship with existing respiratory co-morbidities like pneumonia and chronic obstructive pulmonary disease) improved all-case agreement to kappa=0.50 (moderate) and kappa=0- 65 (good) for TACO cases. Interpretation The two-part validation exercise showed that the revised 2018 TACO surveillance case definition captures 76% of cases endorsed as TACO by participating haemovigilance systems. This definition can become the basis for internationally consistent surveillance reporting and contribute towards increased awareness and mitigation of TACO. Further research will require reporting more complete clinical information to haemovigilance systems and should focus on improved distinction between TACO and other transfusion respiratory complications. Copyright (C) 2019 Elsevier Ltd. All rights reserved.