Key result
Hepatic cirrhosis in patients undergoing left atrial appendage occlusion was associated with higher rates of cardiac tamponade and 30-day readmission.
Why the study?
Does left atrial appendage occlusion (LAAO) carry a higher risk of acute procedural complications and readmission in patients with cirrhosis compared to those without?
Does left atrial appendage occlusion (LAAO) carry a higher risk of acute procedural complications and readmission in patients with cirrhosis compared to those without?
Answers are grounded in this paper and cite their sources.
LAAO in patients with liver cirrhosis is associated with significantly higher upfront procedural risks and 30-day readmission rates, highlighting the critical need for careful patient selection and shared decision-making.
This editorial refers to ‘Left Atrial Appendage Closure Device Outcomes Among Cirrhosis Patients with Atrial Fibrillation; A United States National Cohort Study’, by T. Mir et al., https://doi.org/10.1093/eurheartj/euad004. Percutaneous left atrial appendage occlusion (LAAO) is increasingly recognized as both a safe and effective measure for ischaemic stroke prevention in patients with atrial fibrillation (AF).1 Nevertheless, data exploring its role in certain populations at high bleeding risk—such as those with hepatic cirrhosis—are lacking. In this edition of Europace, Mir and colleagues present a retrospective analysis of LAAO procedural outcomes in patients with cirrhosis compared to those without.2 Using the National Readmissions Database, which covers almost half of all hospitalizations in the USA, the authors identified over 54 000 LAAO procedures performed between 2016 and 2019, of which 905 (1.65%) had cirrhosis. Patients with cirrhosis were younger, but had higher rates of comorbidities including heart failure, diabetes, obesity, chronic renal failure, cancer, and alcohol use as compared to those without. They found that LAAO in cirrhotic patients was associated with a higher rate of adverse outcomes during the index hospitalization as well as a much higher rate of readmission within 30 days. Although ischaemic strokes and intracranial haemorrhages were similar between groups, and mortality data were not provided, cardiac tamponade was higher in the cirrhotic group with pericardiocentesis required more frequently (2.3% vs. 1.0%; P = 0.04). In the cirrhotic group, post-procedural acute kidney injury was significantly more common (5.7% vs. 2.3%; P = 0.001), the rate of gastrointestinal (GI) bleeding was higher (4.9% vs. 2.9%; P = 0.05), as was the need for blood transfusion. Furthermore, as many as 18.5% of patients in the cirrhotic group needed readmission within 30 days, which was twice as high as in the non-cirrhotic group (9.0%). The most common causes for rehospitalization in both groups were cardiovascular, especially hypertensive heart disease and heart failure. To interpret these findings, we must consider two important questions. Firstly, how high is the bleeding risk in cirrhosis and should anticoagulation be avoided? Secondly, is LAAO beneficial in the longer term for these patients? Anticoagulation is a well-established pillar of AF management. Although the pathophysiologic mechanisms of cirrhosis promote development of AF,3 cirrhotic patients have largely been excluded from the seminal anticoagulation trials. Nonetheless, subsequent studies have demonstrated prognostic benefits to anticoagulation with direct oral anticoagulants without a significantly increased bleeding risk in compensated cirrhotic patients.4,5 There was previously a belief that these patients were ‘auto-anticoagulated’—based largely upon abnormal laboratory markers of haemostasis, such as the international normalized ratio (INR). However, as liver dysfunction affects intrinsic production of both pro- and anticoagulant factors, as well as platelet function, a raised INR is not necessarily predictive of bleeding risk. The current thinking is that the haemostatic system ‘rebalances’ in the setting of cirrhosis.6 Indeed, these patients may actually be at increased thromboembolic risk7 and meta-analytical data suggest ischaemic stroke risk is reduced with anticoagulation, without significantly increasing bleeding risk.8 Furthermore, there is some evidence that anticoagulation may play a role in the treatment of cirrhosis, with animal models showing reductions in inflammation and fibrosis, and improvements in hepatic function.9 The evidence in humans is limited, but a randomized controlled trial of 70 patients with cirrhosis demonstrated a reduction in portal vein thrombosis (0% vs. 16.6%; P = 0.025) and hepatic decompensation (11.7% vs. 59.4%; P < 0.0001).10 There is evidence suggesting that prothrombosis contributes to progression of hepatic fibrosis and cirrhosis.11 Thus, maintaining systemic anticoagulation, especially in the setting of AF, is desirable, though more research is needed. Efforts to minimize bleeding risk—such as optimizing cirrhotic therapies and considering prophylactic variceal banding—should be undertaken, but despite this, there will be clinical situations in which a cirrhotic patient cannot take anticoagulation safely. In this setting, is there a benefit from LAAO? As mentioned above, Mir and colleagues show LAAO in cirrhotic patients is associated with higher procedural risk. These data are congruent with those of a recent, similar study that analysed outcomes of LAAO in over 62 000 patients, of whom 1.87% had cirrhosis.12 After multivariable regression, cirrhosis was associated with a significantly higher odds of mortality (OR 8.6, 95% CI 4.1–17.9; P < 0.001) as well as intracranial haemorrhage, retroperitoneal bleeding, transient ischaemic attack, and venous thromboembolism. These are important data, even accounting for the obvious limitations of database-driven research, and that of propensity matching in such unevenly matched populations. For an intervention such as LAAO, that is in most cases performed for preventative rather than therapeutic reasons, to be associated with such a high rate of acute adverse complications and rehospitalizations is sobering. The present study does not include follow-up data, and so we are none the wiser whether this high upfront risk is recouped down the line in terms of low rates of stroke, bleeding, or mortality. It is well recognized that the net benefits of LAAO take time to accrue13 in order to offset the upfront procedural risk. Having established that the upfront risk is higher in cirrhotic patients, we must now consider their longer-term prognosis. Exact prognostication is challenging and varies depending upon severity of liver disease. For example, the survival time of patients with compensated cirrhosis may be over 12 years, whilst for decompensated disease, it may be closer to 2 years.14 Furthermore, as discussed earlier, discontinuation of anticoagulation may adversely affect the progression of cirrhosis. Additionally, LAAO will not prevent other thromboembolic phenomena which are common in cirrhosis, such as portal vein thrombosis or venous thromboembolism. This further complicates the decision to offer LAAO. It is worth pointing out that cirrhosis is not a single entity, but an endpoint of multiple different liver diseases. The present study does not differentiate the underlying liver disease nor the severity of cirrhosis, both of which are important in predicting clinical outcomes. The evidence discussed here emphasizes the critical importance of patient selection. A small minority of cirrhotic patients may benefit from LAAO; however, this decision must be made carefully and in close discussion with the patient. Where LAAO is to be offered to these patients, the following recommendations should be considered. Before the procedure, close liaison should be maintained with the patient’s hepatic specialist team. This allows optimization of medical therapy, such as non-selective beta-blockade, and minimization of GI bleeding risk, for example with prophylactic variceal banding if appropriate. Peri-procedure, the operator should be aware of, and plan ahead, for a higher risk of acute procedural complications such as cardiac tamponade. Following discharge, one should consider early follow-up and community support in light of the increased risk of early readmission. Left atrial appendage occlusion remains an option for patients in whom lifelong anticoagulation is contraindicated. However, its benefits often take years to manifest and the prognosis of competing morbidities must be considered carefully. Liver cirrhosis is a complex pathology, increasing both bleeding and thromboembolic risk. It is worth bearing in mind that these patients are at higher risk of thromboembolism unrelated to the left atrial appendage, and that they suffer increased complications and mortality relating to the LAAO procedure. For these reasons, many cirrhosis patients will not benefit from LAAO. Where anticoagulation cannot be maintained, careful patient selection and shared decision-making are critical if LAAO is to be considered. Pre-procedural optimization, planning ahead for complications, and close post-procedural follow-up are key in ensuring good outcomes. None received for this article.
No takes yet. Share an insight, caveat, or question.
Calvert et al. (2023) conducted an editorial in Atrial fibrillation with hepatic cirrhosis (n=54,000). Hepatic cirrhosis vs. No cirrhosis was evaluated on Cardiac tamponade (p=0.04). Hepatic cirrhosis in patients undergoing left atrial appendage occlusion was associated with higher rates of cardiac tamponade and 30-day readmission.
Synapse has enriched 10 related papers. Consider them for comparative context: