Massive Transfusion Protocols: A Survey of Academic Medical Centers in the United States

Abstract

BACKGROUND: 

Massive transfusion protocols (MTPs) have been adopted in many hospitals, and they may improve outcomes, as well as decrease the number of blood products transfused. However, there are no specific guidelines regarding the number and types of products that should be included in these protocols. MTPs may vary from hospital to hospital.

METHODS: 

A short, web-based survey was sent to blood bank medical directors at academic institutions to learn details about MTPs.

RESULTS: 

A total of 107 survey requests were sent, and 56 were completed (52% response rate). All who responded had an MTP in place. Nearly all (n = 55, 98.2% [95% CI, 90.6%–99.7%]) base their protocol on delivery of fixed amounts and ratios of blood products, with only a minority incorporating any elements of laboratory-directed therapy. The most common target, red blood cell (RBC):plasma ratio, is 1:1 (n = 39, 69.9% [95% CI, 56.7%–80.1%] of respondents). The majority (n = 36, 64.3% [95% CI, 51.2%–75.6%]) provide 6 or more units of red blood cells in the first MTP packet.

CONCLUSIONS: 

One-hundred percent of survey respondents had an MTP in place. Despite a lack of published guidelines regarding MTPs, the survey results demonstrated substantial uniformity in numbers of products and target transfusion ratios.

In Brief

Published ahead of print October 3, 2016.

Historically, massive transfusion has been defined as the transfusion of ≥10 units of RBCs in <24 hours.^1–3 Recent data from military populations have suggested that early administration of plasma and platelets may improve morbidity and mortality.^4,5 Multiple retrospective analyses demonstrate better survival, with higher plasma:RBC ratios.^4–7 In addition, adoption of protocol-based transfusion practices may improve outcomes, improve efficiency of blood product delivery, and decrease costs.^8–11 Based on these reports, many hospitals have adopted MTPs to ensure that massively transfused patients receive blood products in a timely fashion and in the optimal RBC:plasma ratio.^9–11 The earliest adopters of MTPs were trauma centers; however, many hospitals have expanded the MTP to nontrauma users, such as intensive care units and labor and delivery units.

Although numerous studies have shown the benefits of having a massive transfusion protocol (MTP),^8–11 to our knowledge, no specific guidelines exist regarding what products should be included in an MTP, what ratios should be used, nor how products should be delivered to the clinical environment. The purpose of our study was to determine what percentage of academic hospitals across the United States employ an MTP and to learn the details of commonly used MTPs.

METHODS

This study was deemed to be exempt from review by the institutional review board of the University of Chicago. We designed an electronic web-based survey using SurveyMonkey, Inc.¹² The survey included questions concerning background information of the respondent and the hospital, whether the hospital has an MTP, and if there is an MTP, specific questions concerning component ratios, who can activate/deactivate it, and delivery of products (Table 1). Cognitive testing of the survey was completed by 4 nationally recognized faculty members from different institutions and specialties whose feedback was incorporated into the survey.

Table 1.: Survey Questions

We identified all accredited anatomical and clinical pathology residency programs in the United States and the principal hospital associated with each through the American Association of Medical Colleges website.¹³ We then identified the medical director of the blood bank associated with each of those hospitals as listed on the hospital Internet web page or by telephoning the hospital if the information was not apparent on the website. For cases in which the medical director could not be identified with these methods, we searched the Directory of Pathology Training Programs¹⁴ for the residency program and sent the survey to the transfusion medicine program director. An email with a link to the survey was sent to the identified physicians. For cases in which there was no reply to the survey within 4 weeks, a second request was sent; after an additional 2 weeks, a third request was issued. The SurveyMonkey program automatically tracked respondents and delivered additional appeals for response. The study survey period ended 2 weeks after the third reminder.

Results were tabulated electronically using the SurveyMonkey program and then placed into an Excel spreadsheet (Microsoft). Descriptive statistics were used to analyze the data that were expressed as a number and a percentage.

RESULTS

A total of 107 hospitals and physicians were identified and sent the survey; 56 completed the survey (52% response rate). On the first attempt, 38 responses were received. On the second attempt, 9 more responses were received, for a total of 47 completed surveys. On the third and final attempt, we received 9 additional responses, bringing our completed survey number to 56. See Table 2 for characteristics of the respondents and the hospitals. The majority of surveys were completed by the blood bank medical directors. The majority (74%) of hospitals had more than 500 beds, the majority of which were level I trauma centers (77%) and/or obstetric centers (91%), and most of the obstetric centers reported performing more than 1000 deliveries per year.

Table 2.: Respondent and Hospital Characteristics

All who responded had an MTP in place. Characteristics of MTPs are listed in Table 3. Approximately half of programs instituted MTPs within the past 5 years. The majority of programs (67.9% [95% CI, 54.8%–78.6%]) made modifications to the protocol as a result of situational experience. Thirty-eight of the respondents modified their MTP since implementation, and the most common alteration reported was a decrease in the number of units delivered (without a change in ratio; Table 4).

Table 3.: Massive Transfusion Protocol Characteristics

Table 4.: Most Common Modifications to Massive Transfusion Protocol Since Implementation

Nearly all (n = 55, 98.2% [95% CI: 90.6%–99.7%]) base their protocol on delivery of fixed amounts and ratios of blood products, with only a minority incorporating any elements of laboratory-directed therapy. The most commonly employed target RBC:plasma ratio is 1:1 (n = 39, 69.9% [95% CI: 56.7%–80.1%] of respondents). The majority (n = 36, 64.3% [95% CI, 51.2%–75.6%]) provide ≥6 units of RBC in the first MTP packet. Five respondents (8.9% [95% CI, 3.9%–19.3%]) reported including no plasma at all in the first MTP pack, but most add 4 (n = 17, 30.4% [95% CI, 19.9%–43.3%]), 5 (n = 7, 12.5 [95% CI, 6.2%–23.6%]), or ≥6 (n = 21, 37.5% [95% CI, 26.0%–50.6%]) units. Thirty-six protocols (64.3% [95% CI, 51.2%–75.6%]) call for inclusion of apheresis platelets with the first pack. Only one respondent (1.8% [95% CI, 0.3%–9.5%]) reported providing cryoprecipitate in the first MTP pack. Most continue with the same number and ratio of products when a second pack is requested. Those that modify subsequent packs do so by increasing the amount of plasma, platelets, and/or cryoprecipitate. Twenty-three survey responders (41.0% [95% CI, 29.2%–54.1%]) use ≥2 different protocols for different clinical settings. Of those with multiple protocols, a different one is followed in trauma situations by 14 (60.9% [95% CI, 40.8%–77.8%]), in obstetric hemorrhage by 10 (43.5% [95% CI, 25.6%–63.2%]), for pediatrics by 7 (30.4% [95% CI, 15.6%–50.9%]), and in the operating rooms by 1 (4.4% [95% CI, 0.8%–21.0%]). The MTPs specific to trauma tend toward a lower RBC:plasma ratio than that in nontrauma; obstetric hemorrhage MTPs often include an earlier emphasis on cryoprecipitation. MTP activations are formally reviewed by blood bank leadership, and feedback is provided to practitioners in most centers. The majority (82.1% [95% CI, 70.2%–90.0%]) found their MTPs mostly or extremely effective.

Nearly all establishments require activation by the bedside physician (Table 5). Most blood banks deliver products utilizing a runner from the unit. One center utilizes robots to deliver blood products during MTP activations. MTP packs are most often delivered in a cooler. Although a minority of institutions allow clinicians to check the whole pack just once, most require them to check each unit within the pack individually.

Table 5.: Activation Characteristics

Eight centers (14.3% [95% CI, 7.4%–25.7%]) require physicians to check laboratory values at set intervals; most frequently, complete blood count, prothrombin time/international normalized ratio, and fibrinogen. Calcium, recombinant factor VII, and antifibrinolytic therapy are addressed in only 5 (8.9% [95% CI, 3.9%–19.3%]), 10 (17.9% [95% CI, 10.0%–29.8%]), and 12 (21.4% [95% CI, 12.7%–33.8%]) of protocols, respectively. Seventeen (30.4% [95% CI, 19.9%–43.3%]) utilize thromboelastography (TEG) to help guide therapy. Thawed plasma is routinely available at 48 hospitals (85.7% [95% CI, 74.3%–92.6%]). Twenty-nine blood banks (51.8% [95% CI, 39.0%–64.3%]) store un-crossmatched type O RBCs outside of the blood bank, most frequently the emergency department, the operating rooms, and/or the labor and delivery unit. The median number of RBCs stored in those areas is 4.

DISCUSSION

In this survey of academic medical centers, 100% of survey respondents reported having an MTP in place. Despite the sample including varying numbers of inpatient beds, and all trauma levels, all had an MTP, and >90% had had one for >1 year. In 2008, Hoyt et al¹⁵ conducted an international survey that showed that 45% of trauma centers had an MTP in place. Our results, just a few years after that study, demonstrate usage of MTPs to be 100% within all trauma centers that responded to our survey.¹⁵ Furthermore, the majority considered their systems to be highly effective. MTPs improve turnaround time, decrease costs, and reduce the number of RBCs transfused as well.^9–11

Although much literature discussing MTPs exists, formal guidance as to what products and what ratios should be used, especially in nontrauma settings, has been lacking. In spite of this, the survey results demonstrated substantial uniformity in numbers of products and target transfusion ratios. The majority included 6 units of RBCs per pack and targeted a 1:1 RBC:plasma ratio. Several studies have demonstrated increased survival when plasma is introduced early in resuscitation of hemorrhaging patients; however, the optimal RBC:plasma transfusion ratio remains a topic of debate. A retrospective study in combat victims in Iraq between 2003 and 2005 demonstrated increased survival with a low RBC:plasma ratio during massive transfusion.⁴ However, many experts, including the authors of the previously mentioned investigation, point out that retrospective data often include unidentified confounders; for instance, high-ratio deaths occurred earlier than low-ratio deaths, possibly because these patients had more severe injuries and died before frozen plasma could be thawed and administered, a phenomenon known as survivor bias.^4,16,17 Indeed, Snyder et al¹⁸ adjusted for survival bias, and the survival benefit of low RBC:plasma ratios disappeared. Also, a study in civilian trauma victims identified 2:1 to 3:1 RBC:plasma ratios as optimal.¹⁹ Another retrospective study showed that plasma deficits in some populations may be a more sensitive marker than the ratio of RBC:plasma.²⁰ A recent randomized clinical trial comparing 1:1:1 RBC:plasma:platelet ratios to 2:1:1 ratios during resuscitation after severe trauma revealed that although the 1:1:1 ratio group subjects experienced better hemostasis, and fewer exsanguination deaths during the first 24 hours, overall mortality and complication rates did not differ between the groups.²¹ This trial set out with 23 complication end points, none of which had any statistically different outcomes.^21,22 Amid these apparent conflicts, the similarity in recommended ratios across institutions was surprising. In addition, despite little information on the role of platelets in massive transfusion, more than half of the protocols included one unit of apheresis platelets in the first pack.

The majority of blood banks have thawed plasma ready for use. Although we did not specifically ask about thawed plasma stored outside the blood bank, one respondent did report storing 2 units of thawed plasma in the emergency department. We suspect there are other institutions that are doing this, but we do not have those data from this survey. Four institutions reported that one modification to their MTP was switching to A plasma from using the universal AB donor plasma.

Some institutions employ specific MTPs for trauma, lowering RBC:plasma ratios. This is not surprising because the above-mentioned studies that underscore the importance of low ratios involved trauma patients. Some obstetric-specific MTPs provide cryoprecipitate earlier, consistent with modern thought on obstetric hemorrhage, which notes early fibrinogen consumption and emphasizes replacement with cryoprecipitate.^23–25

Only a small number of institutions (22%) incorporate antifibrinolytics into their protocol. The incorporation of antifibrinolytics varied in use from being given in all trauma cases, at the start of the MTP, dosed based on TEG results, dosed on Clinical Randomisation of an Antifibrinolytic in Significant Hemorrhage (CRASH)-2 study, and dependent on clinical team. The CRASH-2 trial showed that tranexamic acid significantly reduced the risk of death resulting from bleeding, and the authors recommended that consideration be given to tranexamic acid administration in bleeding trauma patients.²⁶ The results from this survey show that few institutions have adopted use of antifibrinolytics despite past and ongoing (World Maternal Antifibrinolytic [WOMAN] trial) studies that seem to suggest that antifibrinolytics are indicated as part of MTPs.

There are several limitations to this study. First, we garnered a response rate of 52%, so it is possible that representatives from programs without an MTP did not answer our survey. Second, we surveyed only academic centers; nonacademic community hospitals may not exhibit the same patterns we report here. The smaller academic hospitals that responded to the survey reported similar ratios as larger-bed hospitals. We recognize that smaller hospitals may not have the same resources or products, such as apheresis platelets or thawed plasma, available that a larger institution does. Community hospitals could consider applying similar ratios with a smaller number of products in the pack to avoid wastage of products and having a plan in place during an often chaotic time. In addition, a large proportion of our responders practice at level I trauma centers or those with obstetrics, and hospitals that do not serve these patient populations may be less apt to have MTPs in place. Finally, many of our respondents based their answers on estimates rather than institutional databases, and so some of these results may not accurately reflect true practice. However, it seems unlikely that the presence or absence of an MTP or target transfusion ratios were not accurately reported.

In summary, we surveyed academic centers across the United States, and all survey participants reported having a MTP. Many of them target a 1:1 RBC:plasma ratio.

DISCLOSURES

Name: Angela B. Treml, MD.

Contribution: This author helped design and conduct the study, analyze the data, and prepare the manuscript.

Name: Jed B. Gorlin, MD.

Contribution: This author helped analyze the data and prepare the manuscript.

Name: Richard P. Dutton, MD, MBA.

Contribution: This author helped analyze the data and prepare the manuscript.

Name: Barbara M. Scavone, MD.

Contribution: This author helped design and conduct the study, analyze the data, and prepare the manuscript.

This manuscript was handled by: Marisa Bicca Marques, MD.