Document resource
Introduction Cancer-associated thrombosis is a major cause of morbidity, and patients with luminal gastrointestinal (GI) cancers have both high thrombotic and GI bleed risks. Direct Oral Anticoagulants (DOACs) and warfarin pose a higher GI bleed risk than Low Molecular Weight Heparin (LMWH). NICE and regional guidelines recommend LMWH as the preferred therapeutic anticoagulant for this group. This 3-cycle quality improvement project (QIP) evaluated local practice and delivered interventions to improve compliance.Methods Data was collected for adult inpatients with luminal GI cancers (provided by IT department) receiving therapeutic anticoagulation prior to or started during admission. Data was collected between August 2023 and March 2024 (cycle 1). Targeted resident doctor teaching formed the first intervention, followed by a reaudit between February and April 2025 (cycle 2). Teaching for pharmacists was delivered as a second intervention, with a final reaudit between July and September 2025 (cycle 3).Results Cycle 182 admissions were identified; 83% (68/82) were admitted on anticoagulation - 66% (45/68) on a DOAC, 28% (19/68) on LMWH, 6% (4/68) on warfarin. 56% (25/45) were discharged on their original DOAC and 25% (1/4) on their original warfarin. Out of 13 admissions with a GI bleed, 40% (4/10) continued on a DOAC.17% (14/82) started anticoagulation during admission - 86% (12/14) started LMWH, 14% (2/14) started a DOAC.Cycle 228 admissions were identified; 75% (21/28) were admitted on anticoagulation - 43% (9/21) on a DOAC, 48% (10/21) on LMWH, 9% (2/21) on warfarin. 67% (6/9) were discharged on their original DOAC and 50% (1/2) on their original warfarin. Out of 3 admissions with a GI bleed, 1 was admitted on a DOAC which was continued.25% (7/28) started anticoagulation during admission. All received LMWH.Cycle 332 admissions were identified; 94% (30/32) were admitted on anticoagulation - 77% (23/30) on a DOAC, 6% (2/30) on LMWH, 17% (5/30) on warfarin. 57% (13/23) were discharged on their original DOAC and 60% (3/5) on their original warfarin but necessitated by an underlying thrombotic condition. 1 admission was with a GI bleed; admitted on a DOAC which was stopped.All patients who started anticoagulation (6%, 2/32) received LMWH.Conclusion Our QIP showed noncompliance with guidance recommending LMWH as the firstline therapeutic anticoagulant in patients with luminal GI cancers. Compliance for newly initiated anticoagulation improved; LMWH use increased from 86% (cycle 1) to 100% (cycles 2-3). However, there was little impact on rationalising preexisting regimes, with most patients discharged on their original DOACs. Clinical teams rotating may have reduced durability of teaching interventions. Systemlevel multidisciplinary prompts at admission and discharge are likely needed to ensure consistent anticoagulant review in this highrisk group.