Potential role of left atrial appendage closure in trauma patients with atrial fibrillation on anticoagulation

Abstract

Introduction

Patients with atrial fibrillation (AF) on anticoagulation (AC) have increased mortality after trauma. Left atrial appendage closure (LAAC) procedures may reduce the need for AC in patients with AF. The purpose of this study is to evaluate the outcomes of trauma patients with AF on AC compared with patients not taking AC, and assess the frequency in which trauma patients are candidates for LAAC procedures.

Methods

A retrospective review of adult trauma patients after a fall was performed. AF patients on AC (+AC) were compared with patients not on AC (−AC). A CHA2DS2-VASc score is a stroke-risk stratification tool incorporating congestive heart failure, hypertension, age >75 (2 points), prior stroke (2 points), vascular disease, age 65-74, and female sex. This score was calculated for +AC patients to analyze their candidacy for LAAC.

Results

+AC patients (3452 total) were found to have increased red cell (7.3% vs 6.1%, p=0.02) and plasma (2.1% vs 1.2%, p<0.001) requirements. +AC patients had increased mortality (4.6% vs 3.5%, p=0.02). 54.8% of +AC patients had a CHA2DS2VASc score of >3 and met criteria for LAAC.

Conclusion

A substantial proportion of AF patients on AC meet existing criteria for LAAC. Further exploration of referral pathways may be warranted to analyze whether offering information on such procedures could reduce the risk of AC-related complications in high-risk trauma populations.

Evidence level

III.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Older trauma patients on therapeutic anticoagulation have higher mortality after injury. Left atrial appendage closure (LAAC) can reduce stroke risk in atrial fibrillation (AF) patients while avoiding long-term anticoagulation. However, trauma-specific LAAC candidacy has not been quantified.

WHAT THIS STUDY ADDS

• Over half (54.8%) of our geriatric fall patients on anticoagulation for AF had a CHA2DS2VASc score of 3 or more, and 11.2% of those patients had a HAS-BLED score (which evaluates bleeding risk based on hypertension, abnormal renal or hepatic function, prior stroke, history of major bleeding, labile International Normalized Ratio, age >65, and concomitant drugs or alcohol use) of 2 or 3, delineating a clinically relevant subgroup for potential LAAC consideration.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY

• These data identify a sizeable, high-risk trauma cohort that meets established criteria for LAAC in the general AF population. Further research is needed to establish candidacy in this specific population to consider LAAC as part of a multidisciplinary approach to injury prevention.

Introduction

Atrial fibrillation (AF) is the most common sustained arrhythmia with significant morbidity and mortality.^{1 2} Approximately 65% of patients with AF take anticoagulation (AC) to lower stroke risk, and the number on AC has only increased over the years.³ However, bleeding complications are not uncommon in patients taking AC.⁴ It is well established that trauma patients have worse outcomes when they take AC, with studies consistently showing increased mortality in these patients.5,7

During the past 20 years, significant research has focused on developing AC alternatives. Clot formation in AF typically occurs due to stagnation of blood flow in the left atrial appendage (LAA).⁸ This has led to the development of minimally invasive procedures to ablate or occlude the LAA. One example of a left atrial appendage closure (LAAC) procedure involves catheter-based insertion of a plug into the opening of the LAA that epithelializes over time and occludes the LAA. Another device acts as a lasso and is placed around the opening of the LAA.⁹ The first device was implanted in a human in 2002.¹⁰ Prospective studies have demonstrated the safety of these procedures, mitigating the risk of ischemic stroke incidence and avoiding the risks of chronic AC.^{11 12}

LAAC has demonstrated safety and efficacy in the general AF population and has been used to lower cardioembolic stroke risk and avoid long-term AC.¹² The CHA2DS2VASc score is a validated score for predicting stroke risk in patients with AF and includes variables such as congestive heart failure, hypertension, age >75 (2 points), diabetes, prior stroke (2 points), vascular disease, age 65-74, and female sex.¹³ Patients with a score of 3 or more are considered candidates for LAAC; however, some providers suggest patients with a score of 2 should also be considered if they have a high risk of bleeding or active lifestyle.¹⁴ The HAS-BLED score is another scoring system that has been validated in predicting a patient’s bleeding risk on AC.¹⁴ One point is given for hypertension, abnormal renal or liver function, stroke, history of major bleeding, labile International Normalized Ratio, age >65, and drug or alcohol use. Typically, patients with a HAS-BLED score of 2 or 3 are moderate-risk to high-risk and considered candidates for LAAC. LAAC candidacy has not been quantified in trauma cohorts. Up to 35% of patients undergoing LAAC have a reported procedural indication of high risk for falls.¹² Trauma patients are at particularly increased risk of falls, as multiple studies demonstrate that among older patients who fall, one-third to two-thirds will have a recurrent fall within a year.^{15 16}

The purpose of this article is to evaluate the mortality of trauma patients on AC for AF and to assess the frequency at which these trauma patients are candidates for LAAC procedures based on the candidacy guidelines for the general AF population. These data may help inform discussions about whether certain high-risk trauma patients might benefit from multidisciplinary evaluation for interventions, including LAAC, as part of broader injury prevention strategies.

Methods

The trauma registry database at our level 1 trauma center was used in this study. As a deidentified database, this study was deemed exempt by our institutional review board and a waiver of consent was granted. The database was queried from January 2018 to December 2024 for adult trauma patients 18 years and older who were evaluated after a fall mechanism. Patients were divided into two groups. The first group was patients with pre-existing AF on therapeutic AC (+AC). The second group consisted of all remaining fall patients not on therapeutic AC (−AC). Patients on antiplatelet therapy were not excluded from either group. Patients on AC for valvular disease or venous thromboembolism were excluded because this type of AC use cannot be intervened on with LAAC procedures. Similarly, antiplatelet use did not exclude patients in either group, as LAAC would not alter their use. AC reversal during this study period was not protocolized, and reversal decisions were at the discretion of the consulted neurosurgeon. The primary outcome was mortality, and the secondary outcome was development of any in-hospital complication including unplanned intubation, unplanned return to the operating room, pneumonia, acute respiratory distress syndrome, catheter-associated urinary tract infection, central line-associated bloodstream infection, cardiac arrest, cerebrovascular accident (CVA), deep venous thrombosis, pulmonary embolism, myocardial infarction (MI), and acute kidney injury. We performed a subgroup analysis of the +AC group to evaluate the frequency at which these patients were candidates for LAAC. To analyze candidacy for LAAC, a CHA2DS2-VASc Score and a HAS-BLED score were calculated for each patient in the +AC group. We used the US coverage criteria for percutaneous LAAC, defined as CHA2DS2-VASc >3.¹⁴

Demographic data points that were collected included age, sex, and comorbidities including congestive heart failure, chronic obstructive pulmonary disease (COPD), MI, CVA, chronic kidney disease, and alcohol/substance abuse. The injury data collected included traumatic brain injury, thoracic injury, solid organ and hollow viscus injuries, as well as extremity and spine fractures, and the Injury Severity Score (ISS). Additional outcomes evaluated were intensive care unit (ICU) admission, ICU length of stay (LOS), hospital LOS, and blood product administration.

All bivariate analyses were performed with IBM SPSS Statistics for Windows (V.29, IBM, Armonk, New York, USA). A Mann-Whitney U test was used to compare continuous variables, and a χ² was used to compare categorical variables in the bivariate analysis. Categorical data were presented as percentages, whereas continuous data were presented as a mean with SD. We then performed a multivariable logistic regression analysis to analyze the risk of mortality after ground level fall for patients on AC compared with those not on AC. We adjusted for potential confounders, which were selected based on discussion among coauthors, review of the literature, and identifying univariate statistically significant differences between proposed confounding variables. These included age, ISS, severe Abbreviated Injury Scale (AIS >3) for the head, and having two or more comorbidities. P values were defined as statistically significant if <0.05. This study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.

Results

A total of 9590 patients were evaluated at our trauma center between January 2018 and December 2024 after a fall. Of these patients 3452 (36.0%) had a pre-existing diagnosis of AF and were taking therapeutic AC. The +AC group was older than the −AC group. There was no difference in sex between the two groups.+AC patients had a lower mean ISS. The +AC group had a higher incidence of hypertension, heart failure, chronic kidney disease, end-stage renal disease, and COPD. There was no difference in the incidence of diabetes or cirrhosis. More patients in the −AC group were on antiplatelet medication (table 1)

Table 1. Demographics and comorbidities of fall patients with atrial fibrillation on therapeutic AC and fall patients without atrial fibrillation or therapeutic AC.

Characteristic	Fall patients +AC (n=3452)	Fall patients −AC (n=6138)	P value
Age, years, mean (SD)	80.6 (9.9)	67.9 (21.5)	<0.001
Male, n (%)	1879 (54.4%)	3241 (52.8%)	0.13
ISS, mean (SD)	4.9 (5.2)	6.3 (5.8)	<0.001
Comorbidities, n (%)
Hypertension	974 (28.2%)	1386 (22.6%)	<0.001
Diabetes mellitus	418 (12.1%)	718 (11.7%)	0.550
Heart failure	394 (11.4%)	219 (3.6%)	<0.001
Chronic kidney disease	208 (6.0%)	186 (3.0%)	<0.001
ESRD	47 (1.4%)	48 (0.9%)	0.01
COPD	210 (6.1%)	210 (3.4%)	<0.001
Cirrhosis	17 (0.5%)	37 (0.6%)	0.582
On antiplatelet	607 (17.6%)	1184 (19.3%)	0.041

AC, anticoagulation; COPD, chronic obstructive pulmonary disease; ESRD, end stage renal disease; ISS, Injury Severity Score.

The −AC group had a higher incidence of brain injury, rib fractures, lung injury, liver injury, pelvis fractures, and upper extremity injury. The +AC group had a higher incidence of lower extremity injury. There was no difference in occurrence of injury to the heart, spleen, stomach, colon, kidney, or bladder (table 2).

Table 2. Injury patterns for fall patients with atrial fibrillation on therapeutic AC and fall patients without atrial fibrillation or therapeutic AC.

	Fall patients +AC (n=3452)	Fall patients −AC (n=6138)	P value
Injury
Brain	376 (10.9%)	1020 (16.6%)	<0.001
Heart	2 (0.1%)	3 (0.1%)	0.99
Rib fracture	190 (5.5%)	529 (8.6%)	<0.001
Lung	56 (1.6%)	241 (3.9%)	<0.001
Diaphragm	0 (0%)	0 (0%)	N/A
Esophagus	0 (0%)	0 (0%)	N/A
Spleen	14 (0.4%)	38 (0.6%)	0.22
Liver	3 (0.1%)	35 (0.6%)	<0.001
Stomach	0 (0%)	1 (0.02%)	0.99
Small intestine	0 (0%)	0 (0%)	N/A
Colon	2 (0.1%)	1 (0.02%)	0.61
Rectum	0 (0%)	0 (0%)	N/A
Kidney	3 (0.1%)	16 (0.3%)	0.11
Bladder	2 (0.1%)	2 (0.03%)	0.95
Pelvic fracture	107 (3.1%)	254 (4.1%)	0.01
Upper extremity fracture	185 (5.4%)	456 (7.4%)	<0.001
Lower extremity fracture	487 (14.1%)	673 (10.9%)	<0.001

AC, anticoagulation.

There was no difference in LOS between the two groups. More patients in the −AC group required admission to the ICU, but of the patients admitted to the ICU, there was no difference in ICU LOS. A larger percentage of +AC patients received packed red blood cells and plasma. There was no significant difference in platelet transfusion percentages. More −AC patients received cryoprecipitate. The abbreviated injury score for head injuries was higher in the +AC group. There were no differences in in-hospital complications between the two groups aside from a higher incidence of acute renal failure in the +AC group. The mortality rate in the +AC group was higher (table 3)

Table 3. Clinical outcomes in fall patients with atrial fibrillation on therapeutic AC and fall patients without atrial fibrillation or therapeutic AC.

Outcome	Fall patients +AC (n=3452)	Fall patients −AC (n=6138)	P value
LOS, days, mean (SD)	4.1 (6.1)	4.3 (6.0)	0.33
ICU admission, n (%)	501 (14.5%)	1189 (19.4%)	<0.001
ICU LOS, days, mean (SD)	3.3 (3.6)	3.6 (4.9)	0.23
Received PRBC, n (%)	251 (7.3%)	372 (6.1%)	0.02
Received plasma, n (%)	73 (2.1%)	73 (1.2%)	<0.001
Received platelets, n (%)	88 (2.6%)	137 (2.2%)	0.36
Received cryoprecipitate, n (%)	1 (0.03%)	14 (0.2%)	0.04
AIS head, mean (SD)	2.8 (0.7)	2.6 (0.7)	0.04
Complications, n (%)
Stroke/CVA	4 (0.1%)	9 (0.2%)	0.92
Cardiac arrest	21 (0.4%)	49 (0.8%)	0.36
Myocardial infarction	13 (0.4%)	14 (0.2%)	0.26
Pneumonia/VAP	0 (0%)	7 (0.1%)	0.11
ARDS	0 (0%)	0 (0%)	N/A
Unplanned intubation	0 (0%)	0 (0%)	N/A
Unplanned return to OR	0 (0%)	0 (0%)	N/A
Unplanned ICU admission	0 (0%)	0 (0%)	N/A
CAUTI	0 (0%)	0 (0%)	N/A
CLABSI	0 (0%)	0 (0%)	N/A
Acute kidney injury	37 (1.1%)	38 (0.6%)	0.02
DVT	0 (0%)	0 (0%)	N/A
Pulmonary embolus	75 (2.2%)	113 (1.8%)	0.30
Mortality, n (%)	157 (4.6%)	218 (3.5%)	0.02

AC, anticoagulation; AIS, Abbreviated Injury Scale; ARDS, acute respiratory distress syndrome; CAUTI, catheter associated urinary tract infection; CLABSI, central line associated blood stream infection; CVA, cerebrovascular accident; DVT, deep vein thrombosis; ICU, intensive care unit; LOS, length of stay; OR, operating room; PRBC, packed red blood cells; VAP, ventilator associated pneumonia.

After controlling for potential confounders, the risk of mortality was higher for patients on AC. Age, ISS, severe head injury, and multiple comorbidities were all independently associated with increased mortality (table 4).

Table 4. Multivariable logistic regression analysis for risk of mortality after fall.

Risk factor	OR	CI	P value
Anticoagulation use	1.73	1.57 to 1.93	0.012
Age (years)	1.07	1.04 to 1.07	<0.001
Female sex	0.48	0.37 to 0.61	<0.001
Injury Severity Score	1.13	1.10 to 1.16	<0.001
Severe head injury (AIS 3+)	2.03	1.24 to 3.32	0.005
Multiple comorbidities (2+)	1.62	1.23 to 2.13	<0.001

AIS, Abbreviated Injury Scale.

Of the +AC patients, 1892 (54.8%) had a CHA2DS2VASc score of 3 or more, and 355 (10.3%) are considered moderate to high risk for a major bleeding episode based on their HAS-BLED score. Of those with a CHA2DS2VASc score of 3 or more (1892), 212 (11.2%) also had a HAS-BLED score of 2 or 3 (tables5 6).

Table 5. CHA2DS2VASc score for fall patients with atrial fibrillation on therapeutic AC.

CHA2DS2-VASc Score	Fall patients +AC (n=3452)
0	105 (3.0%)
1	333 (9.6%)
2	1122 (32.5%)
3	1315 (38.1%)
4+	577 (16.7%)

AC, anticoagulation.

Table 6. HAS-BLED score for fall patients with atrial fibrillation on therapeutic AC.

HAS-BLED score	Fall patients +AC (n=3452)
0	1781 (51.6%)
1	1316 (38.1%)
2 (moderate risk)	331 (9.6%)
3 (high risk)	24 (0.7%)

AC, anticoagulation.

Discussion

Therapeutic AC among trauma patients presents a significant problem. As the leading cause of therapeutic AC use, AF should be addressed in all geriatric patients. As evident in multiple prior studies, this study shows trauma patients with AF on AC have a higher mortality.5,7 Our study also shows there are a large number of trauma patients that qualify for minimally invasive procedures to reduce the need for AC. Given the elevated risk of recurrent falls after an initial event, it may be reasonable to consider whether some patients could benefit from evaluation for interventions such as LAAC, alongside established fall prevention programs.

Even after a low-impact mechanism such as a fall, more trauma patients with AF on therapeutic AC require blood product transfusion. Siletz et al found that transfusions were more common in patients on therapeutic AC; however, they included all trauma mechanisms.¹⁷ The increased use of packed red cells and plasma in +AC patients that fall is seen even in studies that look at specific anticoagulants.¹⁸ Others have found that the need for red cell transfusion is associated with an increase in all-cause mortality among trauma patients.¹⁹

This study found a higher mortality among fall patients with AF on AC. This could be related to the higher incidence of comorbidities compared with other patients that fell; however, both AC use and multiple comorbidities were independently associated with increased mortality on multivariate analysis. Given the higher incidence of blood transfusions and the higher AIS head scores in the +AC group, it is likely that AC plays a role in their increased mortality. Although patients without AF had a higher incidence of intracranial hemorrhage, the severity of head injury was higher in +AC patients. Sammy et al found that many factors contribute to mortality risk in older trauma patients, and that the relationship among these factors is quite complex.²⁰ Multiple studies have found that AC use is an independent risk factor for mortality in geriatric trauma patients, and although there are many reasons for a patient to be on AC, AF represents a disease process that can be intervened on to reduce the need for AC with LAAC.^{21 22}

Although trauma surgeons cannot modify all risk factors for mortality among elderly trauma patients, one way we can intervene is through injury prevention. A typical candidate for LAAC is a patient with AF who does not have a contraindication to short-term AC and has a CHA2DS2VASc score of 3 or more. Over half of our +AC patients had a CHA2DS2VASc score of 3 or more. Importantly, these trauma patients have already experienced a fall, and as recurrent falls are common, trauma patients may already be considered high bleeding risk regardless of their HAS-BLED score. Based on this increased fall risk, one could argue for the expanded use of LAAC in patients with a CHA2DS2VASc score of 2 plus a bleeding risk.¹⁴ Although intracranial hemorrhage may temporarily make AC contraindicated, ongoing randomized control trials are investigating the optimal timing for therapeutic AC after intracranial hemorrhage.²³ Given the large percentage of patients who qualify for LAAC based on general AF population guidelines, and given the high incidence of repeat falls within a year after the first fall, this represents a significant opportunity for injury prevention. Further studies are needed to analyze if these general indications can be safely applied to trauma patients who fall. A multidisciplinary protocol, including assistance from interventional cardiologists, electrophysiologists, and emergency room nurses/physicians, could allow for referral of a high-risk population for a minimally invasive procedure that might reduce their risk of hemorrhage and mortality in the future. However, this should be considered in conjunction with physical therapy and monitored exercise, which has been shown to independently reduce recurrence of falls in geriatric patients.²⁴

This study has several limitations. It is limited by the nature of a trauma registry study with only retrospective data. AC reversal practices at our institution were provider-dependent and not standardized. Variability in reversal choice and timing may influence bleeding severity, transfusion needs, and outcomes. Additionally, we do not have accurate data on patients who returned to a trauma center after a second fall, as they may have been taken to a different trauma center, or they may have been entered under a different medical record number without proper chart merging. Accurate data on the frequency of repeat falls would potentially add additional support for LAAC for patients on AC. Lastly, because we do not have information on the patient’s home activity level or lifestyle information, more patients could qualify for LAAC than initially identified in this study.

Conclusion

Patients with AF on AC represent a clinically significant number of trauma patients and have been shown to have a higher mortality, even after low-impact mechanisms such as a fall. Ideally, patients on AC for AF who experience a fall should be referred for physical therapy to prevent fall recurrence. However, some patients may also benefit from minimally invasive interventions that could reduce their need for AC before any traumatic event. Large trauma centers with high geriatric populations may wish to explore whether incorporating candidacy assessment for interventions such as LAAC into multidisciplinary care pathways could help reduce AC-related complications in select patients.

Data availability statement

Data are available upon reasonable request.