Print PDF version
Whats New in the
Patient Safety World
September 2024
New Tool to Predict
Surgical Transfusion Need
Blood transfusions can be lifesaving, but they also have downsides. Our many prior columns related to transfusions have chronicled the movement over the last several decades to be more conservative and optimize the use of transfusions.
Historically, most decisions about planning for transfusions during surgery were based primarily on the type of surgery being performed. You would type and screen for those procedures with a high likelihood of transfusion need. On the other hand, if the percentage of patients needing transfusion for a specific procedure has been low, you could omit typing and screening. Researchers at Washington University School of Medicine (Lou 2022) developed a personalized surgical transfusion risk prediction tool using machine learning to guide preoperative type and screen orders. That tool used both surgery- and patient-specific variables to guide preoperative type and screen orders. That machine learning model was consistently more efficient than the conventional maximum surgical blood ordering schedule (procedure-centric) approach when it comes to identifying patients who need preoperative blood typing and antibody screening.
Those researchers have now assessed the generalizability of that machine learning tool (now called S-PATH) across a diverse group of institutions (Vlessides 2024). They used 2020-2021 data from the Multicenter Perioperative Outcomes Group (MPOG), which included 47 institutions and 3,455,295 surgical cases. Their abstract was presented at the 18th World Congress of Anaesthesiologists. The incidence of red cell transfusion at those institutions ranged from 0.0% to 6.5% (median, 1.6%). They found that to achieve a predetermined benchmark of 96% sensitivity, the machine learning tool recommended blood typing and antibody screening in a median of 32.4% of patients compared to 53.4% using the old procedure-centric approach.
Lead researcher Sunny Lou told Anesthesiology News The personalized model was very consistent in terms of reducing the number of recommended screens by about a third and the overall performance of the model was consistently excellent across all hospitals.
The potential use of this tool is quite exciting. While its application would be expected to lead to considerable cost savings by avoiding unnecessary screening, well need to see whether that translates to better patient outcomes.
Prior columns on potential detrimental effects related to red blood cell transfusions:
· March 2011 Downside of Transfusions in Surgery
·
August
2011 CPOE
Alerts Reduce Blood Transfusions in Children
·
January
2012 Need
for New Transfusion Criteria?
·
April
2012 New
Transfusion Guidelines from the AABB
·
February
2013 More
Evidence Favoring Restriction of Transfusions
·
June
2013 Hopkins
Blood Ordering Initiative
·
May 2014
Blood
Transfusion and Infection Risk
·
June
2015 Economics
and Blood Transfusion
·
November
2016 AABB Updates Transfusion
Guidelines Again
·
December
2017 Study Confirms Safety of
Restrictive Transfusion Policy
·
January
2018 Transfusion in Cardiac
Revision Surgery
·
August
2018 Thromboembolism: Another
Downside of Transfusions
·
August
2021 New Blood Management Guidelines
for Cardiac Surgery
· January 9, 2024 Hopkins Blood Management Program 10 Years Later
References:
Lou SS, Liu H, Lu C, Wildes TS, Hall BL, Kannampallil T. Personalized Surgical Transfusion Risk Prediction Using Machine Learning to Guide Preoperative Type and Screen Orders. Anesthesiology 2022; 137(1): 55-66
Vlessides M. Surgical Transfusion Risk Prediction Model Works Across Institutions. Anesthesiology News 2024; July 10, 2024
Print PDF version
http://www.patientsafetysolutions.com/
Whats New in
the Patient Safety World Archive