Spontaneous Pleural, Pericardial, and Intracranial Hemorrhages in a Patient with End-Stage Kidney Disease Receiving Apixaban: A Case Report

Abstract

Patient: Male, 73-year-old

Final Diagnosis: Apixaban induced hemorrhage

Symptoms: Altered mental status • breathlessness

Clinical Procedure: Thoracocentesis

Specialty: General and Internal Medicine

Objective: Unusual clinical course

Background

Apixaban, the direct oral anticoagulant (DOAC) with the lowest renal clearance (~27%), is often preferred for stroke prevention in non-valvular atrial fibrillation (NVAF) among patients with kidney impairment. According to U.S. prescribing information, a dose of 5 mg twice daily is permitted in patients with severe kidney disease, unless they are aged ≥80 years or weigh ≤60 kg. However, in severe chronic kidney disease (CKD) or end-stage kidney disease (ESKD), even limited renal excretion can result in elevated drug levels and bleeding. Apixaban-related bleeding most commonly occurs in the gastrointestinal tract, but rare events such as pleural or pericardial effusions can also develop and may precede intracranial hemorrhage.

Case Report

We describe a 73-year-old man with NVAF and stage 3b CKD who presented with progressive dyspnea and chest pain. Renal function had declined significantly from baseline. Imaging revealed large bilateral hemorrhagic pleural effusions and a concurrent pericardial effusion; both improved following apixaban discontinuation and therapeutic thoracentesis. During hospitalization, he was diagnosed with ESKD (CKD G5D), defined by a glomerular filtration rate <15 mL/min/1.73 m² and the initiation of dialysis. Despite this, apixaban was restarted in a nursing facility, after which he developed an intracranial hemorrhage. His condition deteriorated despite aggressive management, and he subsequently died.

Conclusions

This report emphasizes vigilant monitoring for rare but serious hemorrhagic complications of apixaban. Such bleeding may involve uncommon sites like the pleura, pericardium, or intracranial space, especially in patients with severe kidney disease, despite guideline-based dosing.

Introduction

Anticoagulant therapy is widely used for the prevention and treatment of thromboembolic disorders, particularly in patients with non-valvular atrial fibrillation (NVAF) or venous thromboembolism [1]. With 33.5 million atrial fibrillation (AF) patients worldwide, stroke risk is high, and anticoagulants are recommended for nearly 90% of cases. Warfarin, the only option for decades, has numerous drug and food interactions, requires frequent monitoring and dose adjustments, and is particularly challenging for patients with impaired kidney function [2]. Apixaban, a direct oral anticoagulant (DOAC), is a novel oral anticoagulant (NOAC) that was FDA-approved in 2012 for lowering the risk of cardioembolic stroke in NVAF.

Apixaban is often preferred due to its efficacy and safety profile compared to traditional anticoagulants like warfarin [3,4]. Apixaban appears to have similar or better comparative effectiveness and safety across the range of kidney function compared with warfarin [2,5]. In a retrospective cohort study of patients with end-stage kidney disease and atrial fibrillation undergoing dialysis, apixaban use was associated with a lower risk of major bleeding compared to warfarin, and the standard dose of 5 mg twice daily was additionally linked to reduced risks of thromboembolism and all-cause mortality [6,7].

Furthermore, the estimated renal excretion of direct oral anticoagulants (DOACs) varies significantly, with approximately 80% to 85% for dabigatran, 35% for edoxaban and rivaroxaban, and 25% for apixaban. The lower renal clearance of apixaban supports its preferential use in patients with impaired kidney function [8]. However, despite its favorable risk-benefit balance, bleeding remains a serious complication associated with apixaban use. Major bleeding events, such as gastrointestinal and urogenital hemorrhages, are well-documented, but reports of pleural, pericardial effusions and ICH related to apixaban are uncommon, with only a few cases reported in the literature [9–11]. Here, we present a rare case of apixaban-induced hemorrhagic bilateral pleural and pericardial effusions in a patient with NVAF and severe kidney disease, which preceded the development of an intracranial hemorrhage.

Case Report

A 73-year-old white man, transferred from a nursing home, presented with generalized weakness, retrosternal chest pain, and shortness of breath for 3 days. He had a past medical history of NVAF, with CHA₂DS₂-VASc of 5 points, heart failure with preserved ejection fraction (HFpEF), chronic obstructive pulmonary disease (COPD), diabetes mellitus type 2, hypertension, chronic kidney disease (CKD) stage 3b per CKD/EPI (baseline creatinine 2.6 mg/dl, eGFR 35ml/min/1.73m²), spinal stenosis, neurogenic bladder, obesity, and metabolic dysfunction-associated steatotic liver disease (MASLD). He had no history of tobacco use. He was taking apixaban (Eliquis) 5 mg twice daily for stroke prevention. His other home medications were hydralazine, metoprolol, amlodipine, furosemide, and insulin. His vitals on arrival were: blood pressure 116/64 mmHg, heart rate 74 beats/min, respiratory rate 22 cycles/min, temperature 36.5°C (97.7°F), saturating at 88–89% on room air, weight 107 kg, BMI 36.94 kg/m². On physical exam, he was in respiratory distress and was lethargic but oriented to person, place, and time. He had reduced breath sounds bilaterally, irregular heart rhythm without murmurs or rubs, bilateral lower-extremity pitting edema up to the shins, chronic gluteal ulcers, and a long-term urinary indwelling catheter due to neurogenic bladder.

His hemoglobin level was 9.5 gm/dl (13.9–16.3 gm/dl) (baseline Hgb 11.7 gm/dl), platelet count 321 k/ul (150–400 k/ul) (baseline platelet count 229 k/ul), and white blood cell (WBC) count 15.63 k/ul (3.8–10.5 k/ul) with 88% neutrophils. Sodium was 135 mEq/l (135–145 mEq/l), potassium 5.9 (3.5–5.3 mEq/l), bicarbonate 17 mEq/l (22–31 mEq/l), serum creatinine (SCr) 3.7 mg/dl (0.5–1.3 mg/dl), blood urea nitrogen (BUN) 190 mg/dl (7–23 mg/dl), eGFR 15 ml/min/1.73 m² (>60 ml/min/1.73 m²), creatinine clearance (CrCl) 23 ml/min, troponin T 0.063 ng/ml (<0.010ng/ml) and, pro-BNP 2430 pg/ml (<125pg/ml). Urinalysis showed many bacteria and large leukocyte esterase, suspicious for urinary tract infection (UTI), but a liver function test was within normal limits. A computed tomography (CT) scan of the head was unremarkable. A portable chest X-ray showed findings suspicious for congestive heart failure. EKG on admission showed atrial fibrillation (Figure 1). The patient was started on oxygen supplement and intravenous furosemide and his Foley catheter was changed. He also received aztreonam 2 g twice daily for sepsis likely secondary to complicated UTI.

Figure 1.

Electrocardiogram (EKG) showing atrial fibrillation with preserved ventricular rate.

Transthoracic echocardiogram showed a moderate pericardial effusion (Video 1, Figure 2), left ventricular ejection fraction (LVEF) 55–60%, with no evidence of pericardial tamponade. CT angiography of the chest showed ascending thoracic aortic aneurysm measuring 4.3 cm without dissection or rupture, a moderate-to-large pericardial effusion (likely hemorrhagic), and moderate-to-large bilateral pleural effusions (left greater than right, likely hemorrhagic) (Figure 3). There was no history of chest trauma prior to presentation. The differential diagnoses included apixaban-associated hemorrhage, uremic pleuritis and pericarditis, malignancy, and infectious etiologies. Results of the infectious disease work-up, PT/INR and PTT, TSH, C3, C4 levels, antinuclear antibody, anti-dsDNA antibody, histone antibody, rheumatoid factor, and anti-neutrophilic cytoplasmic antibody were unremarkable. Ultrasound of the kidneys and bladder did not show obstructive uropathy. Apixaban was withheld because of the likelihood of hemorrhagic pleural and pericardial effusion on imaging.

Video 1.

Transthoracic apical four-chamber view echocardiogram demonstrating a moderate-to-large circumferential pericardial effusion without evidence of tamponade physiology. Notably, there is no late systolic right atrial collapse or early diastolic right ventricular collapse.

Figure 2.

Transthoracic echocardiogram shows a moderate-to-large circumferential pericardial effusion measuring approximately 1.87 cm in diastole (blue caliper) and 1.87 cm posteriorly (yellow caliper). The effusion appears echo-dense with heterogeneous internal echoes, consistent with hemorrhagic or complex fluid rather than a simple serous effusion, as illustrated in the parasternal long-axis view (A), apical four-chamber view (B), and subcostal view (C), with the effusion highlighted by red arrows in each image.

Figure 3.

Contrast-enhanced axial chest CT shows bilateral pleural effusions (orange arrows) with high attenuation, consistent with hemorrhagic effusions (A). The density (Hounsfield units) is higher than simple fluid, supporting the presence of blood within the pleural spaces. Contrast-enhanced sagittal chest CT demonstrates a large pericardial effusion (blue arrow) with increased attenuation, indicating a hemorrhagic pericardial effusion. This effusion surrounds the heart and is of higher density than simple serous fluid, consistent with blood (B).

On the first day of admission, the patient underwent right-sided thoracentesis with chest tube placement, resulting in the drainage of 2.8 liters of hemorrhagic pleural fluid, which transitioned to serosanguineous in character within 36 hours (Figure 4). Pleural fluid analysis was bloody red in appearance with WBC 1474/ul (neutrophils 18%, lymphocytes 18%, monocytes 64%), red blood cells (RBC) 569 000/ul (>100 000 RBC denotes frankly hemorrhagic effusion), moderate mesothelial cells, few macrophages, glucose 224 mg/dl, LDH 202U/l (serum LDH 178 U/l), protein 2.5 g/dl (serum protein 5.9 g/dl), adenosine deaminase 18 U/l (0–30 U/l), cholesterol 25 mg/dl and, and triglycerides 23 mg/dl. Pleural fluid analysis showed negative Gram stain, no acid-fast bacilli or fungal growth, and no malignant cells on cytology. The patient had worsening leukocytosis to 19 000 k/ul and was started on cefepime. A septic work-up found no infection source, but WBC levels gradually declined over 7–10 days. The patient also underwent 3 hemodialysis sessions for acute kidney injury (AKI) on CKD within the first 3 hospital days without significant improvement in BUN and SCr.

Figure 4.

Hemorrhagic pleural fluid obtained during thoracentesis. Left panel shows samples collected in sterile containers, while the right panel demonstrates the chest tube collection system containing large-volume bloody effusion.

Left-sided thoracentesis was performed 5 days after discontinuing apixaban, with 1400 ml of serosanguinous fluid drained. The shift from frankly hemorrhagic to serosanguinous pleural fluid suggests reduced intrapleural bleeding and resolution of active hemorrhage, supporting apixaban as the likely cause (Naranjo score of 6, indicating a probable relationship). His shortness of breath improved significantly after bilateral chest tube drainage. Analysis of fluid showed findings similar to the previous diagnostic thoracentesis but with fewer red blood cells (34 000/ul). Pericardial effusion improved without pericardiocentesis (Video 2, Figure 5). A chest CT scan taken after chest tube drainage showed near complete resolution of pleural effusion (Figure 6). The patient was discharged without anticoagulation back to a nursing home after 37 days in the hospital to continue regular hemodialysis for end-stage kidney disease. Figure 7 shows SCr and Cr Cl trends during initial admission. However, the patient was restarted on apixaban 5 mg twice daily while in the nursing home at the discretion of the covering physician. Eight weeks after discharge, he developed altered mental status with quadriparesis and dysphagia following a session of hemodialysis. There was no history of fall or head trauma while in the nursing home. He was admitted and managed at another medical facility. A CT scan of the head revealed multifocal intracranial hemorrhage with numerous acute hemorrhages across multiple compartments. The largest intraparenchymal hemorrhage, measuring 2.3×1.5 cm, was in the left frontal lobe, also involving the bilateral subinsular region and left basal ganglia. A mild mass effect was present without midline shift, along with subarachnoid hemorrhage (Figure 8). These findings were in keeping with coagulopathy. A CT angiogram of the head was negative for aneurysm. Blood pressure on admission was 103/70 mmHg, PTT was 25.40 s (26.60–33.68 s), and Hgb was 9.4 gm/dl (13.9–16.3 gm/dl). Apixaban was promptly discontinued, and the patient received Kcentra (4-factor prothrombin complex concentrate, human) for reversal along with a platelet transfusion. The hospital course was complicated by seizures and acute-on-chronic respiratory failure. He became ventilator-dependent and underwent tracheostomy. Unfortunately, the hospital course was complicated by sepsis and the patient died about 3 months later.

Video 2.

Repeat transthoracic apical four-chamber view echocardiogram demonstrating reduction of the pericardial effusion following conservative management without any interventional procedure.

Figure 5.

Repeat transthoracic echocardiogram shows a reduction in pericardial effusion, as seen in the apical four-chamber view (A), subcostal view (B), and parasternal long-axis view (C), with the effusion indicated by brown arrows in each image.

Figure 6.

Axial contrast-enhanced computed tomography (CT) images of the chest after thoracocentesis. A cross-sectional view demonstrating near complete resolution of right-sided pleural effusion.

Figure 7.

Serial monitoring of renal function, including serum creatinine (reference range: 0.5–1.4 mg/dL), estimated glomerular filtration rate (eGFR; normal: >60 ml/min/1.73 m²), and creatinine clearance (CrCl; normal: 80–160 ml/min/BSA for men). The values illustrate a trend over time, with persistent elevations in serum creatinine and a decline in eGFR.

Figure 8.

Axial computed tomography (CT) images of the head. CT scan of the head revealed multifocal intracranial hemorrhage with numerous acute hemorrhages across multiple compartments. The largest intraparenchymal hemorrhage, measuring 2.3×1.5 cm, was in the left frontal lobe, also involving the bilateral subinsular region and left basal ganglia. A mild mass effect was present without midline shift, along with subarachnoid hemorrhage.

Discussion

Apixaban, a factor Xa inhibitor, is widely used for stroke prevention in atrial fibrillation and the treatment and prevention of venous thromboembolism due to its predictable pharmacokinetics, reduced need for monitoring, and reduced risk of serious bleeding, including intracranial hemorrhage, compared to traditional anticoagulants like warfarin [1,3]. However, as with all anticoagulants, the risk of bleeding remains a significant concern, and this case underscores the potential for major bleeding events with apixaban use [12–15]. According to the American College of Cardiology (ACC) Expert Consensus Decision Pathway, a major bleed is defined as one involving a critical site, causing hemodynamic instability, a hemoglobin drop of ≥2 g/dL, or necessitating transfusion of ≥2 units of red blood cells; such events typically require prompt cessation of anticoagulation and active bleeding control [15].

Hemorrhagic pleural and pericardial effusions in patients taking anticoagulants are rare but potentially life-threatening [16–19]. Hemothorax and hemopericardium are usually trauma-related, but common non-traumatic causes include iatrogenic injuries, vascular anomalies, malignancy, coagulopathies, and infections such as tuberculosis [20,21]. In this case, the development of bilateral hemorrhagic pleural and pericardial effusion, with transition of an overtly hemorrhagic pleural fluid to serosanguineous fluid following apixaban discontinuation, which has a half-life of approximately 12 hours, suggests anticoagulant-related bleeding with a temporal and likely causal relationship [1]. Although uremia was initially considered a potential cause of the pleural and pericardial effusions, the decrease in bleeding in the absence of significant improvement BUN/Cr levels argues against uremia as the primary etiology. Furthermore, uremia is an uncommon cause of frankly hemorrhagic pleural effusions. While uremic pleuritis can result in exudative pleural effusions, the fluid is typically serous or serosanguinous rather than overtly hemorrhagic, except in patients concurrently receiving dual antiplatelet therapy or anticoagulation [22,23]. Lung cancer and infections like tuberculosis are differential diagnoses of the etiology of hemorrhagic pleural effusion; however, these typically present unilaterally, and pleural fluid analysis in this case was negative for malignant cells and infection [24]. While heart failure can cause bilateral pleural effusion, it is usually transudative, and pericardial effusion is an uncommon feature [25,26]. Hemorrhagic pleural effusions in the context of anticoagulation therapy are uncommon, with reported cases typically involving warfarin, especially in patients receiving supratherapeutic INR or heparin [19]. However, bleeding into unusual spaces such as pleural, pericardial, and alveolar spaces is increasingly recognized as a potential complication of DOACs [12,14].

Intracranial hemorrhage (ICH) is a known complication of anticoagulation therapy, with DOACs such as apixaban generally demonstrating a lower risk of ICH compared to warfarin [2]. However, the risk is not eliminated. Risk factors for DOAC-induced intracranial bleeding include advanced age, uncontrolled hypertension, prior cerebrovascular events, renal impairment, concomitant use of antiplatelets or NSAIDs, and underlying cerebral vulnerabilities such as microbleeds or amyloid angiopathy [27]. Cerebral amyloid angiopathy (CAA) and hypertensive microangiopathy were differentials of ICH in this patient. The multifocal intracranial hemorrhages involving both lobar and deep structures (eg, basal ganglia) with subarachnoid extension, as seen on CT, are more suggestive of DOAC-associated bleeding rather than classical CAA, which typically presents with isolated lobar hemorrhages. The presence of hemorrhage in deep brain regions, particularly in the context of anticoagulant use, supports a coagulopathy-related etiology over CAA [27,28]. While the basal ganglia involvement is consistent with hypertensive microangiopathy, the multifocal and compartmental distribution, especially the lobar and subarachnoid involvement, suggests a broader etiology, likely DOAC-associated hemorrhage on a background of chronic small vessel disease [27,29].

Renal function, concomitant medications, and the presence of comorbidities such as hypertension or a history of prior bleeding episodes are all known to increase the risk of major bleeding in patients on DOACs. In the presented case, impaired renal function was likely the primary factor contributing to the increased bleeding risk. The estimated renal excretion rates of direct oral anticoagulants (DOACs) are approximately 80–85% for dabigatran, 35% for both edoxaban and rivaroxaban, and 25% for apixaban [8]. While apixaban is generally preferred in patients with compromised renal function due to its reduced renal excretion compared to other DOACs, even modest renal impairment can increase drug levels and bleeding risk. A population pharmacokinetic analysis in patients with NVAF showed that severe renal impairment increased apixaban exposure by 55% [30]. In the United States, the standard dosing recommendation for apixaban in patients with NVAF is 5 mg taken orally twice daily for CrCl >50 mL/min or SCr <1.5 mg/dL, with a reduced dose of 2.5 mg twice daily for patients aged ≥80 years and weighing ≤60 kg [8,31]. For patients with a SCr level ≥1.5 mg/dL, a reduced apixaban dose of 2.5 mg twice daily is recommended if they are also aged ≥80 years or have a body weight ≤60 kg [1,31,32].

There have been concerns about the use of DOACs in individuals with CrCl less than 30 mL/minute. For those with CrCl <25 mL/minute, manufacturers recommend the same doses as recommended for patients with SCr ≥1.5 mg/dL, while some experts suggest apixaban 2.5 mg twice daily [30,31]. Canadian and European guidelines recommend against using apixaban in individuals with CrCl <15 mL/min [8,30]. In the United States, phase I clinical trials and a retrospective cohort study suggest no renal dose adjustment for apixaban in hemodialysis patients, with standard dosing of 5 mg twice daily reducing thromboembolic and mortality risks, although some guidelines recommend against its use in this population [7,30,33,34].The RENAL-AF trial, which evaluated apixaban versus warfarin in patients with atrial fibrillation undergoing hemodialysis, was inconclusive due to early termination from slow enrollment; however, it indicated comparable rates of major bleeding between the 2 agents, underscoring the need for further randomized studies to assess the risk-benefit profile of anticoagulation in this high-risk population [6,35]. Due to its high plasma protein binding, apixaban is unlikely to be removed by dialysis, which can increase bleeding risk [36].

Before starting apixaban, laboratory tests should include platelet count, coagulation panel, SCr, CrCl, and liver function to assess bleeding risk and to establish baseline organ function for dose adjustments [37]. According to current guidelines, routine coagulation monitoring is not required due to apixaban’s predictable pharmacokinetics and the absence of an established therapeutic range [37]. However, some evidence suggests monitoring can improve safety and efficacy, especially in patients with baseline CKD. A reduced apixaban dose of 2.5 mg twice daily may be appropriate for patients with CrCl <25 mL/min, with discontinuation advised when CrCl drops below 15 mL/min, consistent with select clinical guidelines, irrespective of age or weight [30]. A study on apixaban dosing in atrial fibrillation patients with CKD stages 4 or 5 found higher bleeding risk with 5-mg compared to 2.5-mg doses, with no significant differences in stroke, embolism, or mortality rates [2]. These findings validate dose adjustments based on kidney function and emphasize the need for further research to refine dosing strategies for patients with severe CKD, including those with acute decline and those on hemodialysis [2,38,39]. Calibrated anti-factor Xa monitoring could be used for patients with kidney disease, but it is not widely accessible [39].

The concurrent use of apixaban and heparin during hemodialysis can elevate bleeding risk, particularly in patients with advanced chronic kidney disease or a recent history of hemorrhagic events [40]. In clinical practice, low-dose heparin is frequently employed, with dosing tailored to the patient’s bleeding risk and institutional protocols [41]. In select high-risk individuals or those with recent bleeding, some centers use heparin-free dialysis strategies utilizing saline flushes to minimize hemorrhagic complications [42]. In our patient, the specific use or dosage of heparin administered during hemodialysis at the nursing facility remains uncertain. Nevertheless, the patient’s normal activated partial thromboplastin time (aPTT) at the time of ICH presentation suggests that heparin was not a significant contributor to the bleeding event.

Although elimination of apixaban occurs via multiple pathways, including hepatic metabolism and biliary excretion, no dose adjustment is needed for mild or moderate hepatic impairment, and apixaban pharmacokinetics are unaffected by MASLD [30]. However, apixaban use is not recommended in patients with severe hepatic impairment (Child-Pugh Class C) due to limited data, potential coagulation abnormalities, and lack of clinical experience [30]. A dose reduction of apixaban is advised for patients concurrently receiving strong dual inhibitors of CYP3A4 and P-glycoprotein [8]. The risk of bleeding with apixaban also increases when used alongside drugs affecting hemostasis, such as NSAIDs, platelet inhibitors or other anticoagulants, SSRIs, and SNRIs [37]. Our patient was not receiving any medications known to increase bleeding risk, and although imaging revealed MASLD, liver function test results remained within normal limits.

Early diagnosis and management of apixaban-induced hemorrhage is essential to improving outcomes. Apixaban-induced bleeding is a clinical diagnosis based on recent or ongoing use of apixaban, absence of alternative causes, a temporal association, and resolution of bleeding upon discontinuation, particularly in high-risk patients such as those with renal impairment [20,43]. Management includes immediate discontinuation of apixaban, with additional interventions determined by the location and severity of bleeding. Non-traumatic bilateral hemorrhagic pleural effusion is often diagnosed through imaging, such as chest X-ray or CT scan, and confirmed with thoracentesis. Management involves treating the underlying cause, such as coagulopathy, and can include pleural drainage to relieve symptoms and prevent complications.

Echocardiography is the primary diagnostic tool for pericardial effusion, but it has limitations [44]. In emergency settings, thoracic multidetector CT is increasingly utilized, offering a reliable, complementary method for cardiac imaging [44], and enabling quick and precise identification of pericardial diseases. Multidetector CT can help diagnose hemopericardium and guide medical or surgical management without requiring pericardiocentesis, in conjunction with history and physical exam findings [26]. On CT imaging, it appears as hyperdense fluid in the pericardial space, indicating blood. It also helps identify cases of bleeding pericardial effusion where a “watch and wait” strategy or deferred treatment is appropriate, enabling a personalized and timely management approach [44]. The “watch and wait” strategy for hemopericardium is suitable when the patient has stable hemodynamics, a small or non-progressive pericardial effusion, and a low-risk, controlled, or resolving etiology [13,45]. Although the echocardiogram revealed significant pericardial effusion, our patient remained hemodynamically stable, suggesting a gradual accumulation of pericardial fluid over an extended period [45]. Following the discontinuation of apixaban, the pericardial effusion significantly decreased within a few days.

Further interventions, such as reversal of anticoagulation, are tailored to the patient’s specific condition. Apixaban’s anticoagulant effect lasts at least 24 hours after the final dose, corresponding to approximately 2 half-lives. However, in the case of life-threatening bleeding such as intracranial hemorrhage, andexanet alfa, a reversal agent for apixaban anti-factor Xa activity, is available for clinical use [37,46]. The effectiveness of andexanet alfa in managing apixaban-related bleeding remains to be fully established, and while 4-factor prothrombin complex concentrate (4F-PCC) is also recommended for urgent reversal, its benefit in improving clinical outcomes is not yet well-defined [1,47]. In phase III trials (ANNEXA-A) and real-world bleeding cases (ANNEXA-4), andexanet alfa rapidly and significantly reduced anti-factor Xa activity by over 90% within minutes of administration, restoring normal thrombin generation and achieving excellent or good hemostasis in approximately 80% to 83% of patients [48]. Observational comparisons indicate that andexanet alfa may offer higher hemostatic efficacy and lower thrombotic complication rates than 4F-PCC in intracranial hemorrhage associated with apixaban. A study reported 64.7% effective hemostasis with andexanet alfa versus 54.8% with 4F-PCC, with fewer thrombotic events [49].

Neurosurgical evaluation may be necessary for hematoma evacuation in severe cases. Notably, 4F-PCC was not administered upon initial presentation with hemorrhagic pleural and pericardial effusions. Although the bleeding was classified as major, clinical improvement followed apixaban discontinuation and thoracentesis. In patients unable to tolerate anticoagulation or with high bleeding risk, such as the case presented, left atrial appendage (LAA) closure with a Watchman device may offer a safer alternative to reduce stroke risk and prevent life-threatening bleeding [50]. If our patient had survived, LAA occlusion would have been a suitable strategy following recurrent major bleeding, contingent on shared decision-making with the patient.

Conclusions

This case serves as a reminder of the potential for severe hemorrhagic complications associated with apixaban. Although DOACs such as apixaban offer significant advantages over warfarin, including reduced monitoring and lower risk of hemorrhage, they are not without risk. Clinicians must remain vigilant for signs of bleeding, particularly in multiple body compartments, as demonstrated in this case. Current guidelines vary in their recommendations, and real-world practice must weigh the theoretical benefits of standard dosing against potential harms. Further research is essential to clarify the mechanisms and risk factors contributing to multicompartmental hemorrhage in patients treated with DOACs, particularly those with chronic kidney disease. These insights are necessary to refine current clinical guidelines and develop more effective strategies to prevent such severe adverse events. This case illustrates that even when guideline-based dosing is followed, outcomes can be unfavorable, reinforcing the need for further randomized studies and highly individualized clinical decision-making in this population.