Efficacy and Safety of Whole Blood Transfusion in Non-Trauma Patients

Introduction

For patients with significant hemorrhage, the associated morbidity and mortality remains high despite the widespread implementation of aggressive interventions aimed at reversing shock physiology. Early initiation of blood transfusion is associated with improved outcomes for severely injured trauma patients in hemorrhagic shock. In recent years, whole blood (WB) transfusion has made a significant re-emergence for civilian trauma patients in hemorrhagic shock. ¹ Despite the emergence of WB for trauma patients, the use of WB for non-trauma patients has not been as well described.^2,3 The importance and efficacy of pre-hospital blood transfusions for non-trauma patients in hemorrhagic shock has been demonstrated. Furthermore, it is being increasingly recognized that in-hospital massive transfusion protocols (MTP) are important for the survival of non-trauma patients as well. Potential non-trauma patients who could benefit include high-risk obstetric patients, gastrointestinal (GI) bleeding, oncologic patients, and ruptured abdominal aortic aneurysms. The objective of this study was to determine if WB transfusion could be performed in the non-trauma patient presenting with significant hemorrhage. We hypothesized that WB could be safely administered to non-trauma patients with minimal side effects similar to the experience to date with WB transfusions.

Methods

Consecutive adult patients receiving WB transfusion at a single tertiary academic medical center were reviewed from February 2018 to January 2020. Inclusion criteria were patients aged 18 years and older who underwent WB transfusion. Exclusion criteria were patients less than age 18 years old and members of vulnerable populations (pregnant women and prisoners). Institutional Review Board approval was obtained from the University of Texas Health Science Center at San Antonio prior to initiation of the research. A Health Insurance Portability and Accountability Act (HIPAA) waiver of informed consent was obtained.

Patients were stratified into two groups depending on the etiology of blood loss, traumatic injury vs non-traumatic injury. The primary outcome of the study assessed was in-hospital mortality. Secondary outcomes included the units of component therapy transfused, and the incidence and severity of transfusion-related reactions.

Results

A total of 237 patients who received WB during the study period were identified with 55 (23.2%) patients having a non-traumatic source of hemorrhage. The average age for the population was 54.3 +/− 2.2 years and 56.4% male (n = 31/55).

The most common etiology of non-traumatic hemorrhage was GI bleeding (43.6%, n = 24/55). Additional sources of non-traumatic bleeding included obstetrics (10.9%, n = 6/55); ruptured abdominal aortic aneurysm (9.1%, n = 5/55); gynecologic (9.1%, n = 5/55); emergency general surgery (7.3%, n = 4/55); advanced malignancy (7.3%, n = 4/55); aortic dissection (5.5%, n = 3/55); bleeding from dialysis access (5.5%, n = 3/55); and acute cardiovascular disease (1.8%, n = 1/55) (Table 1).

OPEN IN VIEWER

Table 1.

Etiologies of Hemorrhagic Shock for 55 Patients in the Non-Trauma Population who Received Whole Blood Transfusions at a Single Academic Institution.

Etiology of hemorrhage	n (%)
Gastrointestinal bleeding	23 (41.8)
Rupture abdominal aortic aneurysm	8 (14.5)
Obstetrics	6 (10.9)
Gynecologic	5 (9.1)
Emergency general surgery	4 (7.3)
Advanced malignancy	4 (7.3)
Aortic dissection	3 (5.5)
Dialysis access bleeding	3 (5.5)
Acute cardiovascular disease	1 (1.8)

Eight patients (14.5%) received pre-hospital WB. Approximately half of the non-trauma patients received component therapy (50.9%, n = 28/55). Compared to a group of trauma patients, the non-trauma patients received significantly fewer units of PRBCs (2.1 +/− .7 vs 3.5 +/− .2, P<.01). There was no difference between the two groups in the amount of FFP (1.3 +/− .7 vs 2.0 +/− .5, P = .5) and platelets transfused (.7 +/− .2 vs .3 +/− .8, P =.8).

The overall in-hospital mortality rate was 30.9% (n = 17/55). Nine patients in the non-trauma group died within the first 24 h compared to 8 patients who died 24 h after presentation. Transfusion-related events were rare, only occurring in 3 non-trauma patients (5.5%, n = 3/55). One patient had a mild allergic reaction. Two patients had severe allergic reactions. None of these patients had any major clinical consequences from these reactions.

Based upon the pre-existing published criteria utilized for transfusion in the trauma population, ⁴ a new algorithm was proposed for transfusion criteria of WB for the non-trauma patient. This decision-making pathway is demonstrated in Figure 1. This algorithm could be employed in the pre-hospital setting or early in the clinical evaluation of unstable patients with suspected non-traumatic hemorrhage as to not delay WB transfusion.

OPEN IN VIEWER

Figure 1.

Proposed clinical decision-making algorithm for the pre-hospital or early transfusion of whole blood in non-trauma patients.

Discussion

Whole blood transfusion has seen a resurgence of interest, specifically for the acute resuscitation of trauma patients with severe hemorrhage. This study demonstrated that non-trauma patients could safely receive WB transfusions with infrequent transfusion-related events. In addition, a new algorithm for clinical triggers of WB transfusion in the non-trauma population was proposed. These criteria for novel pre-hospital transfusion triggers in the non-trauma population may refine the use of WB for the non-trauma population.

Non-trauma patients with massive hemorrhage represent a unique demographic population compared to trauma patients. Non-trauma patients with massive hemorrhage tend to represent an older subset of the population with a greater likelihood of having pre-existing medical comorbidities or medications that can influence outcomes. This population also represents a more heterogenous group as compared to trauma patients.

The current indications for pre-hospital WB transfusion are being extrapolated from the trauma population, despite significant physiologic and mechanistic differences in the non-trauma population. From the results of this study, a new algorithm for clinical triggers of WB transfusion in the non-trauma population was proposed. These criteria for novel pre-hospital transfusion triggers in the non-trauma population may refine the use of WB for the non-trauma population. The translation of transfusion triggers and resuscitation practices from the trauma population data are less than ideal. For example, it is likely that a higher systolic blood pressure threshold of ≤100 mmHg might be more appropriate in this population compared to systolic blood pressure ≤90 mmHg and heart rate >120 beats per minute for younger, healthier trauma patients. Base deficit and lactic acid may not indicate bleeding in the non-trauma population, and therefore may not be valuable data point. Coagulopathy may be an important trigger for WB transfusion in the non-trauma population.

This study has several limitations that merit further discussion. First, the retrospective nature of this study introduces inherent bias. Second, these results are from a single institution with a significant volume of WB transfusions. The careful selection of an appropriate control group is another important step in further defining the role of WB transfusion in the non-trauma patient. Finally, the small sample size also limits the conclusions which can be drawn from this study. Appropriately powered prospective multi-center studies are needed to address these limitations.

In conclusion, this study is one of the first to demonstrate that WB can be safely and efficiently transfused to non-trauma patients with different etiologies of bleeding. Minimal transfusion-related side effects were observed. Future studies should focus on determining if outcomes are improved in non-trauma patients who receive WB transfusions, including strategic use of pre-hospital WB and early transfusion upon arrival to the hospital. A novel algorithm to help refine transfusion criteria in the pre-hospital setting was proposed from the results of this study and should be further analyzed in additional studies.