IS THE NUMBER OF PRBCS TRANSFUSED DURING A MASSIVE TRANSFUSION PROTOCOL A GOOD PREDICTOR OF OUTCOME FOR GI BLEEDING PATIENTS?
M'hamed Turki*1, Mohammed El-Dallal1, Avery Pellnat2, Ryan Hart2, Nicholas Anderson3, Evan Anderson3, Saif Bella1, Amine Hila4
1Gastroenterology, Marshall University, Huntington, WV; 2SUNY The State University of New York, Albany, NY; 3Villanova University, Villanova, PA; 4United Health Services Hospitals Inc, Binghamton, NY
Introduction
Various forms of Massive Transfusion Protocols (MTP) have been implemented in most hospitals. They have been thoroughly studied in trauma patients and have proven their efficacy. Some studies have shown a negative correlation between the number of PRBC transfused and prognosis. Herein, we would like to assess if there is such a correlation for GI bleeding patients.
Methods
In the United Health Services in NY, the MTP is administered in rounds. Each round is made of 4 PRBC with 4 FFP and 2 platelets units; All are given through a rapid infuser.
We retrospectively gathered data for all MTPs performed between January 2011 and December 2020. Patients were divided into 2 groups based on the number of MTP rounds, >2 MTP rounds versus ≤2 MTP rounds. Two-way ANOVA test was used to compare continuous variables and Chi-squared test for dichotomous variables. The two groups were compared based on hospital stay, ICU admissions, surgical requirement and 30-days mortality.
Results
Among the 89 MTPs that were administered over the studied period, 19 were for GI bleeding. Thirteen of our patients required ≤2 MTP.
The average age for all the subjects was 62.9 with 42.1% being female. The baseline characteristics for the 2 groups were statistically comparable (Table 1). The Hb level before initiation of the MTP was lower in the group requiring more than 2 rounds, (5.77, 2.48) vs (7.16, 2.78) (p=0.309).
There was no statistical difference between the 2 groups regarding the length of hospital stay, (9.33, 7.81) for the group with lesser rounds against (5.17, 3.54) in the higher transfusion group (p=0.236). All the patients in the former group required an ICU admission against 76.9% in the group with 2 rounds, this difference was not statistically significant (p=0.545).
Surgical interventions were more frequent (50%) in the group requiring 2 rounds versus (7.7%) in the group requiring 2 rounds, (p=0.134).
The mortality at 30 days following MTP was not statistically different between both groups, (33.3%) in the >2 MTP group versus (61.5%) (p=0.515) (Table 2).
Discussion
Our study shows that the overall mortality is extremely elevated in patients with GI bleeding requiring MTP; 10/19 subjects (52.6%). While our patient population remains small, running a Chi-square analysis is not showing any statistical difference in mortality (p=0.515), the length of hospital stay (p=0.236) or ICU admission (p=0.545) between the 2 groups.
Patients requiring more MTP rounds needed more surgical intervention. This suggests the need to involve a surgical team early on in GI bleeding patients requiring more than 2 rounds of MTP.
As expected, the >2 MTP rounds group had a lower initial Hb level.
In conclusion, this study is not showing any correlation between the number of PRBC transfused and the patients' outcomes but is showing an overall increased mortality when there is a need for MTP.
Table 1: Summary of baseline characteristics for the two groups
Table 2: Summary of results and outcomes
Back to 2023 Posters