Trauma Patients With Cirrhosis Have a Similar Risk of Unplanned Operation and Decreased Associated Risk of Mortality With Venous Thromboembolism Chemoprophylaxis

Abstract

Introduction

Patients with cirrhosis have an increased risk of complications after trauma, including bleeding, unplanned operations, and death. The benefit of venous thromboembolism (VTE) chemoprophylaxis in trauma patients with cirrhosis (CTPs) is not clear, especially since cirrhotic patients are hypercoagulable. We hypothesized that CTPs receiving VTE chemoprophylaxis (vCP) have a lower risk of death with no increased risk for unplanned operations compared to patients with cirrhosis not receiving vCP.

Methods

The 2017-2019 TQIP database was queried for patients with cirrhosis. Patients on outpatient anticoagulant therapy or with a history of bleeding diathesis, interhospital transfers, severe head injury, deaths < 72 hours, and hospitalization < 2 days were excluded. A multivariable logistic regression analysis was performed.

Results

From 10,011 CTPs, 6,350 (63.4%) received vCP. Compared to patients without vCP, the vCP group had decreased mortality (4.5% vs. 5.5%, P = 0.03) but a similar rate of unplanned operations (1% vs. 0.6%, P = 0.07). This persisted on multivariable analysis, with a decreased associated risk of mortality (OR 0.54, CI 0.42-0.69, P < 0.001), and a similar risk of unplanned operation (P = 0.85).

Conclusion

CTPs received VTE chemoprophylaxis in under two-thirds of cases. On multivariable analysis, vCP was associated with a decreased risk of mortality and a similar risk of unplanned operations. These findings suggest that vCP appears safe. Further investigation is needed to confirm this finding.

Keywords

critical care trauma cirrhosis venous thromboembolism chemoprophylaxis

Introduction

Less than .3% of the United States adult population suffers from cirrhosis. In contrast, more than 1% of trauma patients present with cirrhosis, and the wide-ranging systemic abnormalities in these patients impact outcomes after trauma.¹ Historically, patients with cirrhosis were thought to have an increased risk of bleeding and reduced risk of venous thromboembolism (VTE).² However, recent data demonstrate that cirrhosis may lead to a hypercoagulable state by affecting the production of anticoagulant factors, such as antithrombin and protein C and S.^3-7 In support, Subhani. et al. found that patients with chronic liver disease (CLD) have double the risk of venous thromboembolism (VTE) compared to those without CLD.⁸

Traumatic injury is a major risk factor for VTE by completing Virchow’s triad with the addition of endothelial injury and/or venous stasis secondary to immobility.⁹ As such, the use of VTE chemoprophylaxis (vCP) in trauma patients is crucial.^10,11 However, there are no specific recommendations for vCP initiation for trauma patients with cirrhosis, and practice patterns are highly variable, with some centers utilizing laboratory values such as platelet count, international normalized ratio (INR), and partial thromboplastin time (PTT) to determine if and when to initiate vCP.^7,12,13 However, studies on non-trauma populations have suggested that the use of vCP does not increase the risk of major bleeding.^4,12-15 Although, there are no large high-quality studies examining this relationship in only trauma patients.

Thus, we sought to determine the incidence of vCP in trauma patients with cirrhosis and analyze the association between the use of vCP and the incidence of mortality and unplanned operations. We hypothesize that trauma patients with cirrhosis benefit from vCP by having a lower risk of mortality but similar risk of an unplanned return to the operating room (OR) compared to patients who do not receive vCP.

Methods

This study was deemed exempt by the institutional review board at our institution, and a waiver of informed consent was granted due to the use of a deidentified national database. A retrospective analysis of the Trauma Quality Improvement Program (TQIP) was performed between 2017 and 2019. All trauma patients ≥ 18 years of age with a diagnosis of cirrhosis were included. Those on outpatient anticoagulant therapy, with a history of bleeding diathesis, interhospital transfers, serious traumatic brain injury (TBI), deaths < 72 hours, and patients hospitalized less than two days were excluded. Serious TBI was defined by an abbreviate injury scale (AIS) grade 2 or higher for the head.

Demographic variables collected included age, sex, and comorbidities such as hypertension, diabetes, smoking status, previous cerebrovascular accident (CVA), chronic obstructive pulmonary disease (COPD), chronic kidney disease (CKD), and previous myocardial infarction (MI). The injury profile data collected included the mechanism of injury, injury severity score (ISS), and AIS for the thorax, abdomen, and spine.

The primary outcome was in-hospital mortality. The secondary outcomes were VTE and unplanned return to the OR. Other outcomes measured included packed red blood cells (PRBC) transfusion within the first four hours of arrival, total hospital length of stay (LOS), intensive care unit (ICU) LOS, ventilator days, and in-hospital complications, including CVA, MI, cardiac arrest, acute respiratory distress syndrome (ARDS), pneumonia, acute kidney injury (AKI), superficial and deep incisional surgical site infection (SSI), organ space SSI, and sepsis.

Descriptive statistics were performed for all variables and stratified by the use of vCP. Categorical data were compared using chi-square and reported as percentages. Continuous variables were reported as medians with an interquartile range.

A multivariable logistic regression model was also used to evaluate the risk of mortality and VTE after adjusting for sex, age, ISS, vital signs on admission (hypotension (systolic blood pressure <90 mmHg), tachypnea (respiratory rate >22 breaths per minute), and tachycardia (heart rate >120 beats per minute)), Glasgow coma scale (GCS) on admission, smoking status, obesity, pelvic fracture, long bone fracture, spine fracture, and requirement of an operative procedure. These were chosen after coauthors' discussion and a literature review.^12,16 The adjusted likelihood for mortality and complications were reported with an odds ratio (OR) and 95% confidence intervals (CI) after stratification by patients who received vCP and patients who did not receive vCP. A Hosmer–Lemeshow goodness-of-fit test was performed for the model for risk of death and found a chi-square value of 17.6 with a P-value of .20, which suggests that the model fits the data well. For the model on the risk of unplanned return to the OR a Hosmer–Lemeshow goodness-of-fit test was performed and found a chi-square value of 7.22 with a P-value of .512, which suggests that the model fits the data well.

To account for possible survival bias, the model analyzing risk of complication only included a subset of patients who survived the entire index hospitalization. We also performed a subset analysis on patients with severe TBI. All P-values were two-sided, with a statistical significance level of < .05. All statistical analyses were performed with IBM SPSS Statistics for Windows, Version 28. (Armonk, NY: IBM Corp).

Results

A total of 10,011 trauma patients with a diagnosis of cirrhosis were included. Of these patients, 6350 (63.4%) received vCP. The vCP group had a higher median ISS (9 vs 5, P < .001) and higher rates of severe injuries (AIS ≥ 3) to the thorax (2.1% vs 1.3%, P = .003), abdomen (1.9% vs 1.3%, P = .034), and spine (.8% vs .4%, P = .025), compared to the no vCP group (Table 1).

Table 1.

Demographics, Comorbidities, and Injury Profile for trauma patients with cirrhosis who received vs not received vCP.

	+vCP	-vCP	P-value
Characteristic	(n = 6350)	(n = 3661)	P-value
Age, year, median (IQR)	61 (54, 68)	59 (50, 66)	<.001
Male, n (%)	3942 (62.1%)	2419 (66.1%)	<.001
Injury severity score, median (IQR)	9 (4, 10)	5 (4, 9)	<.001
Mechanism of injury, n (%)
Blunt mechanisms of injury
Fall	4141 (65.2%)	2351 (64.2%)	.315
Motor vehicle collision	801 (12.6%)	391 (10.7%)	.004
Pedestrian struck	263 (4.1%)	103 (2.8%)	<.001
Bicycle collision	99 (1.6%)	58 (1.6%)	.922
Penetrating mechanisms of injury
Stab wound	185 (2.9%)	168 (4.6%)	<.001
Gunshot wound	65 (1.8%)	94 (1.5%)	.255
Comorbidities, n (%)
Hypertension	3175 (50%)	1563 (42.7%)	<.001
Diabetes	1867 (29.4%)	968 (26.4%)	.002
Smoking	1954 (30.8%)	988 (27%)	<.001
Cerebrovascular accident	228 (3.6%)	91 (2.5%)	.002
Chronic obstructive pulmonary disease	1090 (17.2%)	441 (12%)	<.001
Chronic kidney disease	313 (4.9%)	177 (4.8%)	.833
Myocardial infarction	63 (1%)	27 (.7%)	.194
Hypotension on arrival, n (%)	311 (5%)	174 (4.9%)	.775
Glasgow coma score category, n (%)			.039
14-15	5560 (94.2%)	3164 (92.9%)
9-13	201 (3.6%)	146 (4.3%)
3-8	133 (2.3%)	97 (2.8%)
Severe injury*, n (%)
Thorax	134 (2.1%)	47 (1.3%)	.003
Abdomen	119 (1.9%)	48 (1.3%)	.034
Spine	52 (.8%)	16 (.4%)	.025

vCP = venous thromboembolism chemoprophylaxis, IQR = interquartile range

Abbreviated Injury Scale = ≥ 3

The vCP cohort had a lower rate of mortality, compared to those who did not receive vCP. The rate of unplanned return to the OR (1% vs .6%, P = .07) and PRBC transfusion (6.3% vs 5.7%, P = .3) were similar between groups. However, the median LOS (7 vs 4 days, P < .001) and ICU LOS (4 vs 3 days, P < .001) were longer in the vCP group compared to the no vCP group (Table 2).

Table 2.

Clinical Outcomes For Trauma Patients With Cirrhosis Who Received Versus Not Received vCP.

	+vCP	-vCP	P-value
Outcome	(n = 6350)	(n = 3661)	P-value
Mortality, n (%)	285 (4.5%)	200 (5.5%)	.029
Complications, n (%)
Cerebrovascular accident	12 (.2%)	1 (.0%)	.031
Myocardial infarction	18 (.3%)	5 (.1%)	.139
Acute respiratory distress syndrome	61 (1%)	13 (.4%)	<.001
Pneumonia	54 (.9%)	4 (.1%)	<.001
Acute kidney injury	152 (2.4%)	54 (1.5%)	.002
Cardiac arrest	61 (1%)	28 (.8%)	.315
Superficial incisional surgical site infection	7 (.1%)	1 (.0%)	.157
Deep incisional surgical site infection	12 (.2%)	2 (.0%)	.083
Organ space surgical site infection	12 (.2%)	0 (0%)	.008
Sepsis	101 (1.6%)	30 (.8%)	.001
Packed red blood cells transfused, n (%)	400 (6.3%)	210 (5.7%)	.262
Unplanned return to the operating room	60 (1%)	21 (.6%)	.066
Unplanned intensive care unit admission	330 (5.2%)	84 (2.3%)	<.001
Length of stay, days, median (IQR)	7 (4, 11)	4 (3, 7)	<.001
Intensive care unit, days, median (IQR)	4 (2, 9)	3 (2, 5)	<.001
Ventilator, days, median (IQR)	5 (2, 11)	3 (2, 6)	<.001
Operative procedure, n (%)	1287 (20.3%)	463 (12.6%)	<.001

vCP = venous thromboembolism chemoprophylaxis, IQR = interquartile range

After adjusting for covariates in a multivariable logistic regression model, trauma patients with cirrhosis receiving vCP had a decreased associated risk of mortality (OR = .54, CI 0.42-.69, P < .001) compared to patients not receiving vCP (Table 3).

Table 3.

Multivariable Logistic Regression Analyses* for Risk of Mortality and Unplanned Return to the OR in Adult Trauma Patients With Cirrhosis Who Received Vcp Vs Did Not Receive vCP.

	Risk Factor	OR	CI	P-value
Complete sample	Mortality	.539	.424-.685	<.001
Complete sample	Unplanned return to the OR	1.061	.58-1.941	.847
Patients who survived index hospitalization	Unplanned return to the OR	1.34	.579-3.1	.494
Patients with severe traumatic brain injury	Mortality	.339	.258-.446	<.001

Each model controlled for sex, age ≥65-years-old, injury severity score ≥ 25, hypotension on admission, tachypnea on admission, tachycardia on admission, Glasgow Coma Score on admission, smoking status, obesity pelvic fracture, long bone fracture, spine fracture, and requirement of major surgery including orthopedic surgery.

vCP = venous thromboembolism chemoprophylaxis, OR = Operative room.

On subgroup analysis of patients who survived index hospitalization, there was a similar risk of unplanned return to the OR (OR = .968, CI 0.490-1.911, P = .925) between cohorts (Table 3).

In a subset analysis of trauma patients with cirrhosis and a serious head injury, patients receiving vCP still had a decreased associated risk of mortality (OR = .339, CI 0.258-.446, P < .001) compared to patients not receiving vCP (Table 3).

Discussion

The decision for the use of vCP in trauma patients with cirrhosis is difficult and lacks significant evidence. The results of this national analysis show that vCP is administered to less than two-thirds of trauma patients with cirrhosis staying in the hospital for more than two days. Confirming our hypothesis, vCP was associated with a decreased associated risk of mortality but a similar risk of unplanned return to the OR, suggesting vCP is safe at least in select patients determined by practicing trauma surgeons.

The use of vCP in trauma patients with cirrhosis is still controversial.^17,18 Available studies have reported vCP practices in cirrhotic patients hospitalized for medical conditions to be very low. A recent meta-analysis including 19 million patients reported that just over 25% of patients with cirrhosis hospitalized for medical conditions received vCP.⁸ This current study focused only on trauma patients and found that nearly 65% of patients with cirrhosis hospitalized for longer than two days received vCP. The significantly higher rate of vCP in trauma patients compared to medical patients can be explained by the fact that trauma patients are inherently at higher risk for VTE, and vCP is almost always part of an ICU and trauma checklist during rounds. Additionally, trauma surgeons may be more comfortable identifying and managing bleeding, whereas VTE and specifically a pulmonary embolism represents a potentially preventable cause of death.

Patients with cirrhosis represent a high-risk population for hemorrhage and, thus, unplanned return to the operating room or death. Therefore, the safety of vCP in patients with cirrhosis is of significant concern. A single-center retrospective cohort study of non-trauma patients with cirrhosis demonstrated that patients receiving vCP had a lower risk of bleeding compared to those receiving mechanical or no vCP.⁴ Yerke, J. et al reported a similar finding in a single-center retrospective study, where vCP administration to hospitalized non-trauma patients with cirrhosis was associated with decreased risk of major bleeding.¹⁴ In our study, vCP administration in trauma patients did not increase the risk of unplanned operations or transfusions and was associated with decreased risk of mortality. These results also remained true in a subset of patients with severe TBI. This suggests that vCP in trauma patients with cirrhosis is likely safe and beneficial. However, the decision to start vCP needs to be individualized, including assessment of the degree of cirrhosis as determined by the Child-Pugh classification or Model for End-Stage Liver Disease (MELD) scores and other laboratory abnormalities (e.g., thrombocytopenia and viscoelastic assays), which may impact decisions and outcomes.^19,20 Future research using a more granular dataset may help elucidate the relative impact of these variables on the overall risk of morbidity and mortality in trauma patients with cirrhosis and thereby inform guidelines on exactly when to initiate vCP.

This study has multiple limitations inherent to its retrospective database design, such as coding errors, missing information, and reporting bias. Given the retrospective design, the findings cannot be interpreted as causation, and there is likely some degree of selection bias between the cohorts despite a robust attempt to control for potential confounders. Also, this study lacks pertinent variables, including the dosage and timing of vCP and the rate of missed dosing, which have been demonstrated to be associated with VTE.¹¹ In addition, TQIP lacks data regarding therapeutic levels (e.g., anti-Xa) of vCP and other factors associated with VTE, such as the use of subcutaneous compression devices and ambulation. Furthermore, we did not account for differences in cirrhosis severity (e.g., Child-Pugh classification or MELD scores), including thrombocytopenia and international normalized ratio. Despite these limitations, this study is strengthened by its large national population.

Conclusion

Current guidelines for vCP in trauma lack specific recommendations for trauma patients with cirrhosis. Misinterpretation of hemostatic alterations can predispose clinicians to avoid the use of vCP for fear of increased bleeding risk, as seen by the fact that over one-third of cirrhotic trauma patients staying in the hospital for more than two days did not receive vCP. In addition, vCP was found to be associated with a decreased risk of mortality and a similar rate of unplanned return to the OR. In conjunction, these results suggest that the use of vCP may be safe in trauma patients with cirrhosis. However, further prospective research is needed to validate these findings and develop guidelines to help standardize care for this high-risk subset of trauma patients with cirrhosis.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Claudia Alvarez

References

Georgiou

Inaba

Teixeira

, et al. Cirrhosis and trauma are a lethal combination. World J Surg. 2009;33(5):1087-1092. doi:10.1007/s00268-009-9923-7

Paramo

Rocha

. Hemostasis in advanced liver disease. Semin Thromb Hemost. 1993;19(3):184-190. doi:10.1055/s-2007-994024

Buresi

Hull

Coffin

. Venous thromboembolism in cirrhosis: A review of the literature. Can J Gastroenterol . 2012;26(12):905-908. doi:10.1155/2012/175849

Alsheikh

Kamar

Kreidieh

, et al. The incidence of venous thromboembolism and practice of deep venous thrombosis prophylaxis among hospitalized cirrhotic patients. Gastroenterology Res. 2022;15(2):67-74. doi:10.14740/gr1493

Tripodi

Anstee

Sogaard

Primignani

Valla

. Hypercoagulability in cirrhosis: Causes and consequences. J Thromb Haemost. 2011;9(9):1713-1723. doi:10.1111/j.1538-7836.2011.04429.x

Northup

McMahon

Ruhl

, et al. Coagulopathy does not fully protect hospitalized cirrhosis patients from peripheral venous thromboembolism. Am J Gastroenterol. 2006;101(7):1524-1528. doi:10.1111/j.1572-0241.2006.00588.x

Dabbagh

Oza

Prakash

Sunna

Saettele

. Coagulopathy does not protect against venous thromboembolism in hospitalized patients with chronic liver disease. Chest. 2010;137(5):1145-1149. doi:10.1378/chest.09-2177

Subhani

Sheth

Ahmed

, et al. Incidence and prevalence of venous thromboembolism in chronic liver disease: A systematic review and meta-analysis. Thromb Res. 2022;215:19-29. doi:10.1016/j.thromres.2022.05.004

Ruskin

. Deep vein thrombosis and venous thromboembolism in trauma. Curr Opin Anaesthesiol. 2018;31(2):215-218. doi:10.1097/ACO.0000000000000567

10.

Yorkgitis

Berndtson

Cross

, et al. American association for the surgery of trauma/American college of surgeons-committee on trauma clinical protocol for inpatient venous thromboembolism prophylaxis after trauma. J Trauma Acute Care Surg. 2022;92(3):597-604. doi:10.1097/TA.0000000000003475

11.

Ley

Brown

CVR

Moore

, et al. Updated guidelines to reduce venous thromboembolism in trauma patients: A Western trauma association critical decisions algorithm. J Trauma Acute Care Surg. 2020;89(5):971-981. doi:10.1097/TA.0000000000002830

12.

Al-Dorzi

Tamim

Aldawood

Arabi

. Venous thromboembolism in critically ill cirrhotic patients: Practices of prophylaxis and incidence. Thrombosis. 2013;2013:807526. doi:10.1155/2013/807526

13.

Shatzel

Dulai

Harbin

, et al. Safety and efficacy of pharmacological thromboprophylaxis for hospitalized patients with cirrhosis: A single-center retrospective cohort study. J Thromb Haemost. 2015;13(7):1245-1253. doi:10.1111/jth.13000

14.

Yerke

Bauer

Bass

, et al. Effectiveness of venous thromboembolism prophylaxis in patients with liver disease. World J Hepatol. 2019;11(4):379-390. doi:10.4254/wjh.v11.i4.379

15.

Intagliata

Henry

Shah

Lisman

Caldwell

Northup

. Prophylactic anticoagulation for venous thromboembolism in hospitalized cirrhosis patients is not associated with high rates of gastrointestinal bleeding. Liver Int. 2014;34(1):26-32. doi:10.1111/liv.12211

16.

Serrano

Liu

Nwabuo

Langness

Juillard

. The effect of cirrhosis on trauma outcomes: A systematic review and meta-analysis. J Trauma Acute Care Surg. 2020;88(4):536-545. doi:10.1097/TA.0000000000002464

17.

Stevens

Woller

Kreuziger

, et al. Antithrombotic therapy for VTE disease: Second update of the CHEST guideline and expert panel report. Chest. 2021;160(6):e545-e608. doi:10.1016/j.chest.2021.07.055

18.

O'Leary

Greenberg

Patton

Caldwell

. AGA clinical practice update: Coagulation in cirrhosis. Gastroenterology. 2019;157(1):34-43. doi:10.1053/j.gastro.2019.03.070

19.

Talving

Lustenberger

Okoye

, et al. The impact of liver cirrhosis on outcomes in trauma patients: A prospective study. J Trauma Acute Care Surg. 2013;75(4):699-703. doi:10.1097/TA.0b013e31829a2c19

20.

Dangleben

Jazaeri

Wasser

Cipolle

Pasquale

. Impact of cirrhosis on outcomes in trauma. J Am Coll Surg. 2006;203(6):908-913. doi:10.1016/j.jamcollsurg.2006.08.001