The association of ABO blood groups and trauma outcomes: A retrospective analysis of 3779 patients.

BACKGROUND: There is currently a lack of understanding regarding the link between ABO blood types with outcomes of traumatically injured patients. The purpose of this study was to determine the association of ABO blood types with outcomes in traumatically injured patients separated by injury type.
METHODS: This retrospective study evaluated trauma patients at an urban, Level 1 trauma center from January 1, 2017, through December 31, 2017. Patients were excluded if they were pregnant or <16 years old. Recorded outcomes included: ABO blood group, mortality, Injury Severity Score (ISS), race, injury type, mechanism of injury, and complications. Data analysis was performed using descriptive statistics including Chi-squared, Kruskal-Wallis, and F-test calculations.
RESULTS: A total of 3779 patients were included in this study. No significant differences were present in mean age or ISS between blood types. In patients with penetrating injuries, blood type O was associated with a significant increase in mortality (P = 0.017), red blood cell transfusion (P = 0.027), and massive transfusion protocol (MTP) (P = 0.026) compared to non-O blood types. In patients with blunt injuries, blood type AB was associated with a significant increase in mortality rate compared to non-AB blood types (P = 0.03).
CONCLUSION: ABO blood type is connected with an underlying process which affects trauma outcomes, including mortality. Blood type O is associated with increased blood transfusion, MTP, and mortality during the initial hospitalization following a traumatic penetrating injury, while blood type AB is associated with increased mortality during the initial hospitalization following a blunt traumatic injury. Copyright:

INTRODUCTION

ABO blood types have been shown to have significant ties to race and are known to play important roles in the susceptibility and outcome of numerous disease processes.[1] Several large studies, some including over 100,000 individuals with over 20 years of follow-up, have suggested that while blood type O individuals have significantly higher rates of bleeding-related complications, non-O individuals are associated with significantly higher rates of vascular and thrombotic disorders.[2345] Recent studies have identified certain ABO blood types as potential risk factors for a variety of disease processes including acute respiratory distress syndrome, acute kidney injury, myocardial infarction, diabetes mellitus, thyroid gland disease, abnormal gut microbiota composition, and certain patterns of musculoskeletal injuries.[678910111213] In addition to their roles as risk factors, certain ABO blood types have also been linked to favorable characteristics including increased levels of defense against various infectious microbes as well as protective mechanisms against certain forms of neoplastic conditions.[141516171819202122] When paired with global environmental conditions, such epidemiologic effects have been proposed as contributing to the uneven presence of available blood types throughout the world.[23] Currently, a global distribution exists in which populations divided both racially and geographically exhibit unique trends in the presence of blood type frequencies.[24]

While an extensive amount of literature has been published regarding various clinical applications of blood types, a relative paucity exists regarding implications of blood type in the setting of traumatic injury. The currently available literature regarding blood type and trauma mortality consists of two studies which display conflicting results.[2526] While one of these studies suggests a correlation between blood type O and increased mortality in traumatically injured patients in Japan,[25] the other suggests no association between blood type and mortality among severely injured trauma patients in the United States.[26]

At present, there are no published studies analyzing the relationship between blood type and trauma outcomes in the context of specific injury types. As the description of a consistent correlation of certain ABO blood types with adverse sequelae in the setting of traumatic injury could have the potential to enhance patient outcomes through improved best practice guidelines, our study aimed to investigate the implications of ABO blood type on outcomes of patients suffering from blunt versus penetrating traumatic injuries in the United States. Based on the results of a Japanese study,[25] we hypothesized that blood type O would be associated with a higher all-cause mortality in traumatically injured patients regardless of injury type.

METHODS

Study design

A retrospective study was undertaken at an urban, Level 1 trauma center. It is the institution's standard practice to routinely perform a blood type and screen for all patients seen in the trauma bay. The study was approved by the Institutional Review Board prior to the commencement of data collection (IRB #: 1212092-1). This manuscript was written in adherence with all applicable STROBE guidelines. As this was a retrospective analysis of de-identified data, patient consent was not required for this study.

Case selection

Inclusion criteria consisted of all highest and moderate-level trauma activations [Appendix A] from January 1, 2017, to December 31, 2017. In order to minimize confounding variables, patients were excluded if they were pregnant, <16 years old, or if blood type data were unavailable.

Variables and outcomes

Demographic information such as age, sex, race, and ABO blood group were collected. Injury type, mechanism of injury, and Injury Severity Score (ISS) were recorded for each patient. Race was recorded based on what was documented in each patient's medical record. Mortality was the primary outcome. The secondary outcomes were blood product transfusions, massive transfusion protocol (MTP), number of ventilator days, intensive care unit (ICU) length of stay, hospital length of stay, time from admission to patient death, and inpatient complications. Measured inpatient complications included: deep vein thrombosis, sepsis, ventilator-assisted pneumonia, unplanned intubation, and acute kidney injury. For the purposes of this study, patients with ISS > 15 at the time of trauma were considered to be severely injured.

Statistical analysis

Descriptive statistics were used to characterize numerical data; mean and standard deviation were reported for continuous variables, while frequency and percentage were presented for dichotomous or categorical variables. Bivariate data analysis was performed using Chi-square tests (for categorical variables) or two-sample t-tests (for continuous variables), as appropriate. Continuous outcomes were modeled using Kruskal–Wallis and F-test calculations. Ninety-five percent confidence intervals were constructed where informative for the sample as a whole and for subgroups of interest. Differences were considered statistically significant for P < 0.05. The primary relationship evaluated was mortality with respect to blood type.

RESULTS

A total of 3866 patients met the inclusion criteria. Eighty-seven total patients were excluded from the analysis. Of these, 65 patients did not have blood type recorded on their medical record, 16 were pregnant, and 6 were under age 16 at the time of injury. After exclusion criteria, 3779 patients were eligible for analysis. Overall blood type distribution was 38% A, 12% B, 47% O, and 3% AB. The proportions of ABO blood types present in our sample of 3779 patients were similar to published national averages when considered as a whole as well as when separated by race [Figure 1].[27] As shown in Table 1, the mean age in our sample was 47.8, 63% were male, mean ISS was 9.9, and 92% suffered from blunt injuries. The two most common mechanisms of injury were motor vehicle crash and fall, together accounting for 67% of the sampled patient population.

Figure 1

Measured ABO blood type frequency compared with published national averages[27]

Table 1

Measured demographics of sampled trauma patients with respect to blood type

	Total	A	B	O	AB	P
n (%)	3779	1428 (37.8)	433 (11.5)	1789 (47.3)	129 (3.4)
Mean age	47.8±21.13	48.8±21.07	46.3±21.54	47.4±20.98	48.4±22.38	0.12
Male, n (%)	2360	851 (36.1)	265 (11.2)	1155 (48.9)	89 (3.8)	0.005*
Female, n (%)	1418	577 (40.7)	167 (11.8)	634 (44.7)	40 (2.8)
Caucasian, n (%)	2827	1187 (42.0)	264 (9.3)	1291 (45.7)	85 (3.0)	<0.0001*
African American, n (%)	833	208 (25.0)	157 (18.8)	432 (51.9)	36 (4.3)
Other, n (%)	119	18 (27.7)	8 (10.1)	32 (55.5)	7 (6.7)
Mean ISS	9.9±9.00	9.9±8.97	9.1±8.59	10±9.06	10.3±9.91	0.15
Injury type, n (%)
Blunt	3485	1331 (38.2)	385 (11.0)	1649 (47.3)	120 (3.4)	0.19
Penetrating	280	90 (32.1)	48 (17.1)	133 (47.5)	9 (3.2)
Other	13	7 (53.8)	0 (0)	6 (46.2)	0 (0)
Mechanism of injury, n (%)
Motor vehicle crash	1389	512 (36.9)	170 (12.2)	662 (47.7)	45 (3.2)	0.09
Fall	1144	459 (40.1)	116 (10.1)	528 (46.2)	41 (3.6)
Assault	278	99 (35.6)	30 (10.8)	142 (51.1)	7 (2.5)
Gunshot wound	197	60 (30.5)	38 (19.3)	93 (47.2)	6 (3.0)
Motorcycle crash	188	74 (39.4)	16 (8.5)	90 (47.9)	8 (4.3)
Other	583	224 (38.4)	63 (10.8)	274 (47.0)	22 (3.8)

ISS: Injury Severity Score

As shown in Table 2, the overall population of sampled patients did not exhibit any statistically significant differences between ABO blood types regarding trauma alert category (P = 0.23), hospital length of stay (P = 0.36), ICU length of stay (P = 0.29), time to expiration (P = 0.24), ventilator usage, or transfusion requirement.

Table 2

Measured outcomes with respect to blood type

	Total	A	B	O	AB	P
Trauma alert, n (%)
Category 1	702	253 (36)	85 (12.1)	332 (47.3)	32 (4.6)	0.23
Category 2	3077	1175 (38.2)	348 (11.3)	1457 (47.4)	97 (3.2)
Mean hospital LOS	4.6±7.52	4.7±6.15	4.1±5.15	4.7±7.00	5.6±21.73	0.62
Mean ICU LOS	3.9	4.1	3.1	3.9	3.3	0.29
Mean LOS for expired patients	5.2	5.7	2.5	5.3	4.6	0.24
Number of ventilated patients, n (%)	544	198 (36.4)	53 (9.7)	271 (49.8)	22 (4.0)	0.62
Mean ventilator days >0	4.2±6.23	4.7±6.84	3±5.13	4.2±6.14	2.5±2.74	0.29
Number of transfused patients, n (%)	219	87 (39.7)	19 (8.7)	105 (47.9)	8 (3.7)	0.60
Number of patients who received MTP, n (%)	75	24 (32.0)	9 (12.0)	36 (48.0)	6 (8.0)	0.14
Number of expired patients	149	57 (38.3)	12 (8.1)	72 (48.3)	8 (5.4)	0.34
Number of patients who expired on day 1	75	27 (36.0)	7 (9.3)	37 (49.3)	4 (5.3)	0.74
Number of patients who expired after day 1	74	30 (40.5)	5 (6.8)	35 (47.3)	4 (5.4)	0.49

*P<0.05. LOS: Length of stay, ICU: Intensive care unit, MTP: Massive transfusion protocol

In patients suffering from penetrating injuries, blood type O was associated with a statistically significant increase in mortality (P = 0.017), red blood cell transfusion rate (P = 0.027), and MTP activation (P = 0.026) compared to non-O blood types [Table 3]. Inversely, in patients suffering from blunt injuries, blood type AB was associated with a statistically significant increase in mortality rate compared to non-AB blood types (P = 0.03) [Table 4]. In our sample, no statistically significant differences were present in blood transfusion requirements between ABO blood groups in patients suffering from blunt injuries (P = 0.11) [Table 4].

Table 3

Select outcomes with respect to blood type in patients suffering from penetrating injury

	Total (%)	A (%)	B (%)	O (%)	AB (%)	P
Mortality	31/280 (11.1)	8/90 (8.9)	2/48 (4.2)	21/133 (15.8)	0/9 (0)	0.017*
RBC transfusion	56/280 (20.0)	13/90 (14.4)	8/48 (16.7)	34/133 (25.6)	1/9 (11.1)	0.027**
MTP	30/280 (10.7)	5/90 (5.6)	5/48 (10.4)	20/133 (15.0)	0/9 (0)	0.026***

*O (15.8%) versus others combined (6.8%): P=0.017; OR=2.6 (95% CI=1.2–5.7), **O (25.6%) versus others combined (15.0%): P=0.027; OR=2.0 (95% CI=1.1–3.5), ***O (15.0%) versus others combined (6.8%): P=0.026; OR=2.4 (95% CI=1.1–5.4). OR: Odds ratio, CI: Confidence interval, MTP: Massive transfusion protocol, RBC: Red blood cell

Table 4

Select outcomes with respect to blood type in patients suffering from blunt injury

	Total	A	B	O	AB	P
Mortality	114/3485 (3.3)	47/1331 (3.5)	10/385 (2.6)	49/1649 (3.0)	8/120 (6.7)	0.03*
RBC transfusion	163/3485 (4.7)	74/1331 (5.6)	11/385 (2.9)	71/1649 (4.3)	7/120 (5.8)	0.11

*AB (6.7%) versus others combined (3.2%): P=0.03. RBC: Red blood cell

No statistically significant differences existed between the proportions of patients who received any blood product transfusion when separated by ABO blood type (P = 0.19) [Table 2]. Furthermore, no statistically significant differences were found between mortality rates of transfused patients when separated by ABO blood type (P = 0.07).

No statistically significant differences were present in the incidence rate of any measured inpatient complications among ABO blood groups (deep vein thrombosis (P = 0.55), sepsis (P = 0.96), ventilator assisted pneumonia (P = 0.72), unplanned intubation (P = 0.5), and acute kidney injury (P = 0.16).

In the overall sampled population, differences in mortality were not statistically significant between blood groups whether the patient population was considered as a whole (P = 0.34) [Table 2] or subdivided by race (P = 0.17 for Caucasians and P = 0.18 for African Americans).

No statistically significant differences were present among ABO blood groups regarding the proportions of patients who expired on the day of trauma occurrence. This was true for the overall sampled population (P = 0.74) [Table 2], as well as for severely injured patients (P = 0.75). Likewise, no statistically significant differences were present among ABO blood groups regarding the proportions of patients who expired after the day of trauma occurrence in either the overall sampled population (P = 0.49) [Table 2] or in the subset of severely injured patients (P = 0.25).

DISCUSSION

In this retrospective study, we evaluated 3779 traumatically injured patients. In our sample, we found that blood type O was linked to increased blood transfusion, MTP requirement, and mortality during the initial hospitalization following a traumatic penetrating injury, while patients with blood type AB were more likely to expire during the initial hospitalization following a blunt traumatic injury. To the best of our knowledge, this is the first study to investigate the association between ABO blood type and mortality in groups of traumatically injured patients separated by injury type.

Clinical investigations have previously demonstrated that blood type O is linked with a higher risk for clinically relevant hemorrhages.[2829] This relationship is hypothesized to be mediated by decreased levels of circulating procoagulatory glycoproteins including Factor VIII and von Willebrand factor as well as the presence of certain platelet surface glycoproteins found in blood taken from type O individuals.[3303132] In their 2001 analysis, Asensio et al. found that penetrating injuries were present in 82% of patients suffering from traumatic exsanguination, suggesting a strong link between penetrating injuries and exsanguination.[33] Therefore, a possible explanation for our finding that blood type O is associated with increased transfusion and mortality following penetrating trauma is that the profile of circulating coagulation factors and platelet glycans associated with blood type O lends itself to increased susceptibility to the bleeding-related complications which are often present in penetrating traumatic injuries. If this was to be the case, it is possible that the use of recombinant von Willebrand factor or desmopressin could be investigated as potential adjunct treatments for patients with blood type O who are suffering from acute penetrating trauma. However, as the routine use of such agents would require further investigation, it is outside the scope of this study to make any recommendation regarding their usage at this point.

In contrast to blood type O, blood type AB has been associated with higher levels of circulating procoagulatory proteins including Factor VIII and von Willebrand factor, leading to increased risk for vascular and thrombotic complications.[343435] While acute traumatic coagulopathy is a well-known phenomenon which frequently occurs within minutes of injury in severe trauma patients due to consequences of shock, hypoperfusion, and vascular damage,[363738] patients sustaining blunt abdominal injuries have been observed to undergo rapid transition to a hypercoagulable state often within 12 h of injury.[3940] It is possible that the inherent hypercoagulable profile linked with blood type AB when combined with the induced hypercoagulable state following blunt traumatic tissue damage may lead to increased morbidity and mortality in the acute window. Further investigation utilizing point-of-care coagulation testing could help to shed light on this idea. Additionally, some data have been published which link the genetic loci controlling ABO antigens to various immunologic regulatory proteins including cytokines and cellular adhesion molecules.[414243] As nonspecific systemic inflammatory reactions are common in patients following an initial phase of trauma,[44] it is possible that the immunomodulatory effects of ABO blood types may play a role in outcomes following the period of acute resuscitation as well.

Several limitations exist in our study. First, due to the nature of this retrospective investigation of a limited sample size, the possibility of residual confounding variables is unavoidable. This was also a single site study which may limit the generalizability of the results. Second, our study only analyzed data regarding the initial hospitalization following a traumatic injury and did not account for long-term patient outcomes. Third, this study did not consider whether patients were receiving anticoagulant and/or antiplatelet therapy or whether patients had been previously diagnosed with conditions affecting hemostasis. In theory, patients with blood type O have a lower risk of venous thromboembolism and may be less likely to be on anticoagulant medications. Fourth, this study did not consider the age of transfused blood products. While this information is utilized by the institutional blood bank, it is not routinely recorded in the electronic medical record and was therefore unable to be used as a reliable data point for this study. Finally, our study did not include data from point-of-care viscoelastic assays such as thromboelastography (TEG). Although TEG is utilized at our site, this information was not consistently present throughout the electronic medical records at the time this study was conducted. Therefore, it was not included in our original query to the institutional trauma database.

In summary, the results of this study provide evidence that ABO blood type is related to outcomes in traumatically injured patients. Specifically, certain blood types may serve as indicators of underlying physiologic mechanisms which may affect mortality based on injury type. Potentially, through the development of a future multi-site Trauma Quality Improvement Program study, the idiosyncratic effects of ABO blood groups could be utilized in order to establish best practice guidelines of traumatic resuscitation and early care once further understood. While blood type is not a variable which can be adjusted, further elucidation of the association of blood types with outcomes of traumatically injured patients may form the basis upon which further research can be conducted, ultimately resulting in improved patient survivability.

CONCLUSION

Blood type O is linked to increased blood transfusion, MTP requirement, and mortality during the initial hospitalization following a traumatic penetrating injury, while blood type AB is associated with increased mortality during the initial hospitalization following a blunt traumatic injury. This finding potentiates the concept that ABO blood type is connected with underlying processes which affect trauma outcomes. A future study which investigates the implications of ABO blood type on trauma patient outcomes while incorporating point-of-care TEG data could add clarity to the mechanism underlying this trend.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Research quality and ethics statement

The authors of this manuscript declare that this scientific work complies with reporting quality, formatting, and reproducibility guidelines set forth by the EQUATOR Network. The authors also attest that this clinical investigation was determined to require the Institutional Review Board/Ethics Committee review, and the corresponding protocol/approval number is 1212092-1. We also certify that we have not plagiarized the contents in this submission and have done a Plagiarism Check.