User: Guest  Login
Document type:
journal article 
Author(s):
Mayr, N Patrick; Wiesner, Gunther; Hapfelmeier, Alexander; van der Starre, Pieter; Husser, Oliver; Bleiziffer, Sabine; Schunkert, Heribert; Lange, Rüdiger; Tassani-Prell, Peter; Martin, Klaus 
Title:
Periprocedural transfusion in patients undergoing transfemoral transcatheter aortic valve implantation. 
Abstract:
The aim of this investigation was to identify patient's characteristics and periprocedural variables related to periprocedural transfusion in transfemoral Transcatheter Aortic Valve Implantation (tf-TAVI).Transfusion of allogenic red-blood cells (RBC) in tf-TAVI and the number of transfused units has been linked to an increased 30-day mortality. In line with the trend of minimization and cost-effectiveness, transfusion should be avoided, wherever possible.Between 2007 and 2015, 1,734 procedures were analyzed from our prospective registry for RBC-transfusion. Multiple logistic regression analysis was used to identify the dependent variables.Transfusion was considered necessary in 14% (n = 243) of the patients. Female gender (OR [95% CI]) (1.680 [1.014-2.783]) and preprocedural moderate (7.594 [4.404-13.095]) and severe anemia (8.202 [0.900-74.752]) according to WHO were the most important preprocedural variables. Periprocedural, pericardial effusion (12.109 [3.753-39.063]), emergency extracorporeal circulation (54.5288 [6.178-481.259]) and major vascular injury (2.647 [1.412-4.962]) were related to transfusion. The same applies to moderate (4.255 [1.859-9.740]) and severe anemia (31.567 [8.560-116.416]) as well as periprocedural experience (0.072 [0.035-0.149] - 0.141[0.079-0.251], P< 0.001) CONCLUSION: Procedural experience, serious adverse events, low pre- and periprocedural Hb levels and female gender were the main variables relating to transfusion. Even in experienced high-volume centers, transfusion is still necessary in a considerable number of patients. 
Journal title abbreviation:
Catheter Cardiovasc Interv 
Year:
2017 
Language:
eng 
Fulltext / DOI:
Print-ISSN:
1522-1946 
TUM Institution:
Institut für Medizinische Informatik, Statistik und Epidemiologie