P1803 - Non-Vitamin K Antagonist Oral Anticoagulation vs Standard of Care for the Treatment of Portal Venous Thrombosis in Cirrhotic Patients; A Systematic Review with Meta-Analysis

Monday, October 25, 2021

10:30 AM – 4:15 PM PT

Location: Shoreline Exhibit Hall

Has Audio

Award: Presidential Poster Award

Presenting Author(s)

FA

Faisal Ali, MD

University of Texas Health Science Center at Houston
Houston, TX, United States

Award: Presidential Poster Award

Faisal Ali, MD¹, Anna Laskova, MD², Sudha Misra, MD², Ismail Mohammed, MD³, Humaed Mohammed Abdul, MD², Zain I. Siddiqui, MD², Mohammed Gandam, MD⁴, Maryam R. Hussain, MD, MPH⁵, Dimpal Bhakta, MD⁶, Nirav Thosani, MD, MHA⁷, Sushovan Guha, MD, PhD⁶, Srinivas Ramireddy, MD⁶, Tomas DaVee, MD, MSCI⁶, Moises Nevah, MD⁶
¹University of Texas Health Science Center at Houston, Chicago, IL; ²AMITA Health Saint Joseph Hospital, Chicago, IL; ³Saint Joseph Hospital, Chicago, IL; ⁴Presence Saint Joseph Hospital, Chicago, IL; ⁵Icahn School of Medicine at Mount Sinai, New York, NY; ⁶University of Texas Health Science Center at Houston, Houston, TX; ⁷University of Texas Health Science Center, Houston, TX

Introduction: There remains hesitancy among practitioners with using non-vitamin K antagonist oral anticoagulants (NOACs) to treat portal venous thrombosis (PVT) in the setting of cirrhosis due to their unclear safety profile in this patient population.

Methods: A systematic review of Medline and Embase was conducted from inception-March 2021. Studies comparing NOACs to the current standard of care (SOC; vitamin-K antagonist oral anticoagulant or no anticoagulation) for the treatment of PVT in cirrhosis were included. A meta-analysis yielding risk ratios (RR) was performed using a random effects model for safety and efficacy outcomes of interest. Influence analysis was also conducted to assess robustness of any significant signals identified herein. Findings are reported as pooled effect sizes with their corresponding 95% confidence intervals.

Results: We found 7 non-randomized and one randomized study. Most patients had HCV-associated cirrhosis and Child Pugh Class B disease (Table 1). Esophageal varices were reported in 74% and 66% of patients in the NOAC and SOC arm, respectively. Rivaroxaban and Apixaban were the most commonly used NOACs.

Complete vessel recanalization was reported in 54% and 12% of patients in the NOAC and SOC arms, respectively; NOACs were associated with a higher complete recanalization rate (RR 2.75; 95% CI 1.75-4.34; p= 0.00; Table 2); this finding remained significant throughout the influence analysis (Table 3). NOACs were associated with a decreased risk of clot progression (RR 0.23; 0.08-0.66; p= 0.01); this finding remained significant upon removal of three of the four studies in the influence analysis.

There was no significant difference in the risk of overall adverse events (AEs), bleeding events, need for hospitalization, surgery, or endoscopy due to AEs the NOAC and SOC arms (Table 2). There was no significant difference in the need for anticoagulation discontinuation in the two groups.

Discussion: NOAC are associated with a higher rate of complete recanalization in the setting of PVT in cirrhosis, with no significant increase in bleeding risk compared to SOC. Although some patients may require discontinuation of NOAC due to AEs, the difference in the need for discontinuation compared to SOC is not significant. These findings do not apply to patients with Child Pugh Class C disease or patients with CKD. The current body of evidence, although promising, is very limited. The relative safety documented herein may set the premise for future randomized clinical trials.

Figure: Comparative meta-analysis of NOAC vs Standard of Care for the treatment of PVT in cirrhosis

Table 1. Pooled Baseline Characteristics – NOACs vs Standard of Care for Portal Venous Thrombosis in Patients with Cirrhosis
Variable	NOAC	Control
Variable	Effect Size (95% CI; No. of Studies)	Effect Size (95% CI; No. of Studies)
HBV	7% (3-16%; 2 studies)	10% (5-20%; 2 studies)
HCV	80% (8-99%; 3studies)	80% (69-88%; 2 studies)
Alcoholic Cirrhosis	8% (1-42%; 3 studies)	3% (1-13%; 2 studies)
NASH	16% (4-47%; 4 studies)	12% (6-23%; 2 studies)
Child Pugh Class A	62% (48-76%; 2 studies)	49% (35-63%; 2 studies)
Child Pugh Class B	84% (16-99%; 3 studies)	84% (18-99%; 3 studies)
Child Pugh Class C	NA	7% (3-16%; 2 studies)
HCC	10% (5-21%; 2 studies)	16% (9-26%; 2 studies)
Esophageal Varices	74% (53-94%; 4 studies)	66% (37-95%; 4 studies)
PVT	98% (78-100%; 7 studies)	99% (74-100%; 5 studies)
Splenic Vein Thrombosis	4% (1-22%; 3 studies)	1% (0-9%; 2 studies)
Mesenteric Vein Thrombosis	2% (0-11%; 2 studies)	1% (0-9%; 2 studies)
Rivaroxaban	92% (54-99%; 5 studies*)	NA
Apixaban	91% (12-100%; 3 studies*)	NA
Dabigatran	12% (5-18%; 2 studies)	NA
Complete Recanalization	54% (26-82%; 5 studies)	12% (3-36%; 5 studies)
Partial Recanalization	16% (8-23%; 3 studies)	7% (1-32%; 3 studies)
Stable Disease	7% (1-43%; 4 studies)	41% (0-85%; 3 studies)
Clot Progression	4% (2-10%; 4 studies)	25% (9-41%; 4 studies)
Overall Adverse Events	23% (11-41%; 7 studies)	11% (4-27%; 5 studies)
Overall Bleeding Events	17% (7-36%; 8 studies)	11% (4-27%%; 5 studies)
Major Bleeding Events	7% (3-14%; 7 studies)	12% (0-24%; 3 studies)
CRNMB	7% (3-17%; 7 studies)	4% (2-11%; 3 studies)
Fatal Bleeding Events	1% (0-4%; 3 studies)	10% (2-17%; 2 studies)
Variceal Bleeding	1% (0-6%; 4 studies)	16% (8-23%; 2 studies)
Hospitalization due to Adverse Event	4% (1-20%; 3 studies)	6% (0-13%; 2 studies)
Surgical Intervention due to Adverse Event	4% (1-20%;3 studies)	NA
Endoscopic Intervention due to Adverse Event	2% (2-14%; 2 studies)	16% (8-23%; 2 studies)
Bleeding Requiring Blood Transfusion	2% (0-11%)	NA
Major Bleeding – GIT	8% (2-13%; 5 studies)	12% (0-24%; 3 studies)
Major Bleeding – RT	3% (0-7%; 2 studies)	NA
CRNMB – GIT	7% (0-15%; 3 studies)	4% (0-8%; 3 studies)
CRNMB – GU Tract	3% (0-7%; 2 studies)	NA
Anticoagulation Dose Reduction due to Adverse Event	4% (1-17%; 3 studies)	NA
Anticoagulation Stopped due to Adverse Event	6% (2-15%; 5 studies)	21% (0-43%; 3 studies)

Table: Table 1. Pooled Baseline Characteristics – NOACs vs Standard of Care for Portal Venous Thrombosis in Patients with Cirrhosis

Disclosures:

Faisal Ali indicated no relevant financial relationships.

Anna Laskova indicated no relevant financial relationships.

Sudha Misra indicated no relevant financial relationships.

Ismail Mohammed indicated no relevant financial relationships.

Humaed Mohammed Abdul indicated no relevant financial relationships.

Zain Siddiqui indicated no relevant financial relationships.

Mohammed Gandam indicated no relevant financial relationships.

Maryam Hussain indicated no relevant financial relationships.

Dimpal Bhakta indicated no relevant financial relationships.

Nirav Thosani indicated no relevant financial relationships.

Sushovan Guha indicated no relevant financial relationships.

Srinivas Ramireddy indicated no relevant financial relationships.

Tomas DaVee indicated no relevant financial relationships.

Moises Nevah indicated no relevant financial relationships.

Faisal Ali, MD¹, Anna Laskova, MD², Sudha Misra, MD², Ismail Mohammed, MD³, Humaed Mohammed Abdul, MD², Zain I. Siddiqui, MD², Mohammed Gandam, MD⁴, Maryam R. Hussain, MD, MPH⁵, Dimpal Bhakta, MD⁶, Nirav Thosani, MD, MHA⁷, Sushovan Guha, MD, PhD⁶, Srinivas Ramireddy, MD⁶, Tomas DaVee, MD, MSCI⁶, Moises Nevah, MD⁶. P1803 - Non-Vitamin K Antagonist Oral Anticoagulation vs Standard of Care for the Treatment of Portal Venous Thrombosis in Cirrhotic Patients; A Systematic Review with Meta-Analysis, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.