Optimum Accuracy of Massive Transfusion Protocol Activation: The Clinician’s View

Background Massive transfusion protocols (MTP) aid in the efficient delivery of blood components to rapidly exsanguinating patients. Unfortunately, clinical gestalt and currently available clinical scoring systems lack the optimal accuracy to prevent blood product wastage (through over-activation), as well as individual patient morbidity and mortality (through under-activation). In order to help refine the MTP activation criteria and protocols, we surveyed clinicians on acceptable over- and under-activation rates for massive transfusions. Methods We surveyed Canadian content experts in their respective fields, using a snowball survey technique. Respondents were categorized into two groups: Group 1 was comprised of trauma and acute care specialists (TACS), while Group 2 was comprised of clinical and laboratory medicine specialists (CLMS). Between-group differences were examined using Fisher’s exact test and the likelihood ratio. Statistical significance was set at p < 0.05. Results We received responses from 35 clinicians in the TACS group and 10 clinicians in the CLMS group. About half (45.7%) of respondents in the TACS group considered an MTP overactivation rate of 5% - 10% acceptable (vs. 60% of the CLMS group; not significant (NS)). Approximately one-third (34.2%) of the respondents in the TACS group considered an MTP under-activation rate of less than 5% acceptable, whereas the majority (60%) of respondents in the CLMS group considered an under-activation rate of less than 5% acceptable (NS). A significantly greater proportion of respondents in the TACS group felt that an anticipated need for > 20 units of packed red blood cells within the next 24 hours was an acceptable criterion for MTP activation. Respondents in the CLMS group were more likely to consider “poor communication” as a reason for blood component wastage. Conclusion Similarities in acceptable MTP over- and under-activation rates were noted across specialties. Collaboration between involved parties is necessary for MTP protocol development to improve patient outcomes and reduce blood wastage.


Introduction
While various definitions of massive transfusion exist, massive transfusion is most commonly defined as the transfusion of ≥ 10 units of packed red blood cells (pRBCs) in 24 hours [1]. Massive transfusion protocols (MTP) allow for the efficient delivery of large quantities of blood components to the rapidly exsanguinating patient [1][2][3]. Despite being somewhat algorithmic, MTPs are required in complex medical scenarios, and the considerations for MTP activation are often constrained by limited time and information [3][4].
An effective way to optimize patient outcomes is carefully selected and adhered to MTP activation criteria [5]. Critical Care Canada Guidelines advocate that these criteria are established in consultation with all involved medical specialties and with a comprehensive institutional plan to maximize patient outcomes, deliver a reasonable ratio of blood components, and minimize unnecessary wastage of blood components [6]. Unfortunately, established protocols and MTP activation criteria still require a fair amount of clinical gestalt which, at times, can be biased and potentially result in over-triage where patients who do not require activation of an MTP receive blood products through activation of an MTP [7][8][9]. This can cause harm as MTP "under-activation" places undue risk upon the individual patient, while "over-activation" can result in the overuse and/or wastage of blood components. That said, when needed, "every minute" of delay in receiving blood components increases patient mortality [4]. Further, currently available scoring systems, such as the "Assessment of Blood Consumption" (ABC) Score and "Shock Index", are lacking in varying degrees of sensitivity and specificity [9][10]. Therefore, clarifying clinically acceptable rates of over-and under-activation may sharpen MTP activation criteria and protocols, as well as set benchmarks for trauma and acute care systems.
To our knowledge, no study has examined the clinician's perspective on appropriate levels of over-and under-activation for massive transfusions. Thus, the purpose of this study was to survey physicians currently practicing in Canada commonly involved in MTP activation and administration to determine their perceptions of "ideal" MTP over-and under-activation rates.

Survey development and distribution
In consultation with the Social Sciences Research Laboratory (SSRL) at the University of Saskatchewan, we developed a 20-question survey to assess respondent demographics, opinions regarding MTP protocols, perceptions of appropriate MTP activation criteria, and acceptable rates of over-and under-activation. Respondents were also questioned about whether or not they would activate an MTP using case-based scenarios. A copy of the survey appears in Appendix 1.
We had initially planned to distribute the survey through Canadian specialist associations; however, not all respective associations had a survey mechanism in place. As such, we elected to distribute the survey to peer-identified content experts in the fields of Emergency Medicine, Anesthesia, Critical Care, General and Trauma Surgery, Hematology, Hematopathology, and Transfusion Medicine using a "snowball" survey technique. Initial survey respondents were content experts known to the study authors. These initial respondents subsequently provided names and contact information for additional Canadian content experts, ultimately "snowballing" into a growing list of participants. The survey was primarily distributed using the online survey platform SurveyMonkey® (http://www.surveymonkey.com). Prior to the survey, potential respondents were supplied with a participant information form that outlined the purpose of the study and the risks and benefits of participation. Responding to the survey implied consent. The survey was password protected for participant security, privacy, and anonymity and was accessible for six weeks. The University of Saskatchewan Research Ethics Board approved this study (REB #17-166).

Statistical analysis
For comparison purposes, physician respondents were categorized into one of two groups: Group 1 was comprised of trauma and acute care specialists (TACS) practicing in trauma care, emergency medicine, anesthesia, critical care, and surgery; Group 2 included clinical and laboratory medicine specialists (CLMS) practicing in hematology, hematopathology, and transfusion medicine. Statistical analysis was performed by the SSRL, and between-group differences were assessed using Fisher's exact test for 2 x 2 tables and the likelihood ratios for tables larger than 2 x 2. Statistical significance was set at p < 0.05.

Results
Our survey was distributed to 83 Canadian physicians, and we received a 54.2% response rate (45 respondents), including representative responses from the majority of the Canadian Provinces. Only one respondent worked in a hospital that did not have an MTP. Of the 45 respondents, 35 respondents were in the TACS group, and 10 were in the CLMS group. Just over half (53.3%) of the respondents practiced in an area whose population size was less than 500,000, 24.4% in a population between 500,000 and one million, and 22.2% in a population of more than one million people ( Table 1). When the number of years in practice was considered, 17.8% of respondents had been practicing for 0 -5 years, 35.6% had been practicing for 6 -10 years, 26.7% had been practicing for 11 -20 years, 13.3% had been practicing for 21 -30 years, and 6.7% had been practicing for more than 30 years at the time of completing our survey.  Fifty-percent of respondents in the CLMS group responded that they worked in a hospital whose MTP included an activation criterion for a peripartum woman with marked ongoing blood loss (vs. 8.6% of the TACS group, p < 0.008). In the TACS group, 34.2% of the physicians felt an under-activation rate of less than 5% was acceptable, while the majority (60%) of physicians in the CLMS group felt that an under-activation rate of less than 5% was acceptable (p = 0.120). In the TACS group, 45.7% of physicians felt an over-activation rate of 5% to 10% was acceptable, while the majority (60%) of physicians in the CLMS group felt that an overactivation rate of 5% to 10% was acceptable (p = 0.361).

Discussion
We examined the clinician's perspective on acceptable rates of under-and over-activation of MTPs. Interestingly, although not statistically significant, the majority of respondents in the CLMS group appeared to favor a lower under-activation rate for MTP activation. This approach would favor giving blood to the individual patient and suggests a confidence among CLMS in blood bank resources available to support bleeding patients while maintaining an adequate supply for routine use. This approach is consistent with the conclusions of a University of Pittsburgh study which showed that despite an over-activation rate of 53.8% in non-trauma patients, the benefits of MTP activation for an individual patient outweighed the potential detriment to system resources [11]. Given that MTP activation does not appear to be associated with increased waste, blood component management during an MTP appears to be more important than whether or not an MTP was activated. Furthermore, in the trauma population at least, "every minute" appears to count, as each minute of delay in receiving blood appears to be associated with increased patient morbidity and mortality [4,[12][13][14]. Ironically, a more liberal activation of an MTP may paradoxically reduce blood component wastage, as 25% of severely injured patients are coagulopathic [15], and early aggressive resuscitation may reduce the overall blood need [14].
Our results suggest that TACS are more likely to consider an "anticipated transfusion of > 20 units of pRBCs in the next 24 hours" appropriate for an MTP activation than CLMS (Figure 1). This criterion reflects a perceived degree of patient blood loss, rather than a measurable amount of blood loss, and the difference in responses likely reflects different clinical perceptions between bedside versus consultative care and the all too common predicament that the degree of hemorrhagic shock is difficult to predict [16][17]. This is likely why various institutional MTP protocols allow for a fair amount of clinical gestalt, which, on its own, has only moderate sensitivity and specificity but can be combined with an institutional protocol or clinical prediction tool to improve the overall accuracy of an MTP prediction [9,17].
Despite the best efforts of physicians, MTP activations result in a blood component waste which is a universal concern for all specialties [1,6]. In our study, both groups agreed that "over-anticipation of blood product need" was a common contributor to blood waste ( Figure 2). This is consistent with recent literature showing that over-anticipation of blood need, as well as limitations on safe storage times outside of the blood bank, result in blood component wastage rates of 0 -9% (for red blood cells) at their lowest and 0 -33% (for cryoprecipitate) at their highest [18]. About one-third (36.5%) of our respondents practiced in institutions where components could not be returned to the blood bank beyond one hour from the time of issue, suggesting that cooler times for refrigerated products had not been validated to allow for a period of extended component storage. Lastly, CLMS were statistically more likely to consider "poor communication" as a reason for blood product wastage. Poor communication about the patient's ongoing need for blood components has been shown to occur in up to one-third of cases [6], and difficulties in maintaining dynamic quantities of blood components in close proximity to patients that are moving through multiple hospital locations over a period of time can create potential issues. This suggests that there is room for improving closed loop communication between treating physicians and the lab, which may be enhanced through ongoing education and quality assurance initiatives [19][20][21][22]. Solutions to these problems could include regular meetings between treating services, encouraging physicians to collaborate via interprofessional grand rounds, and a multidisciplinary performance improvement group [20][21].
Finally, we examined differences in approach to MTP activation between the two studied groups of physicians. In our study, physicians in the TACS group was significantly more likely to activate an MTP in only one of the case scenarios, which involved a post-arrest patient found to have a massive GI bleed ( Figure 3). Differences in activation rates between the two specialty groups may not be surprising, as TACS may be more likely to rely on clinical gestalt than set activation criteria [16]. Further, MTP activation in non-trauma patients is highly variable, with estimations ranging from 8% -50% of all MTPs [22]. Given that there is a lack of clinical research that has established objective criteria for activation of an MTP in non-trauma patients, specialists whose primary patient population are not treated in a trauma setting may be at a significant disadvantage, which may explain the disparity in activation between specialist groups. The development of more precise guidelines with regards to activation of MTPs in nontrauma patients would likely benefit both trauma and non-trauma specialists alike.

Limitations
Our study had several limitations. Firstly, our data were collected using a "snowball" survey technique, which has the potential to introduce selection bias, as not all content experts may be invited to participate. Secondly, there is the potential that only content experts with similar views to prior survey respondents may be invited to participate. Lastly, the snowball technique is not random, and responses may not be fully representative of that population. This survey method was chosen for logistical reasons, as we had initially planned to distribute the survey through specialist associations nationally. Unfortunately, not all of these specialist groups have a survey mechanism in place. Furthermore, the specialist practice can be highly varied, and it is conceivable that, depending on practice patterns, the vast majority of respondents surveyed in this manner would not routinely be involved in MTP activation or administration. A final limitation of the study is that, despite a reasonable response rate and low-time commitment, our survey may have been biased by only attracting a relatively small number of respondents who potentially had strong opinions towards MTP activation. This, unfortunately, is inherent in any survey design, and given that there was a minimal disparity in survey responses within groups, we suggest that this is minimal.

Conclusions
Similarities in acceptable over-and under-activation rates of MTP highlight similar values with respect to MTPs across different specialties. Barriers to effective resuscitation include overanticipation of blood product need and poor communication between the resuscitation and laboratory teams. Collaboration between the resuscitation team and consultants in transfusion medicine is necessary for MTP protocol development to improve patient outcomes and reduce blood wastage.

Appendices
Appendix 1: Massive Transfusion Protocol Survey

What is your current medical specialty? (select all that apply)
ð Emergency Medicine All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.