Benefits and harms of direct oral anticoagulation and low molecular weight heparin for thromboprophylaxis in patients undergoing non-cardiac surgery: systematic review and network meta-analysis of randomised trials

Dear Editor

I would like to congratulate the authors of this excellent study for their diligence and clear interpretation of the available data. The restriction of the review to clinical outcomes (rather than asymptomatic thrombi) and to include potential harms as well as benefits is to be commended.

The findings of the study fit well with my ongoing audit of bleeding and thrombosis after hip fracture surgery. The occurrence of thrombotic and gastrointestinal bleeding complications for 11,873 consecutive patients treated at my hospital with a proximal femoral fracture from January 1989 to February 2021 have been recorded. All thrombotic complications were confirmed by radiographic or ultrasound investigation and the diagnosis of gastrointestinal bleed was by clinical observation of hematemesis. The policy of our hospital is that all patients without contraindications receive low molecular weight heparin for 28 days from admission. The unpublished audit results from this database show the overall 90-day mortality for all patients was 14.4%. Deep vein thrombosis was found for 1.5% of patients, for which 16.2% died by 90 days. Pulmonary embolism was found for 0.7% of patients, for which 44.6% died by 90 days. Gastrointestinal bleeding was noted for 1.5% of patients, for which 39.2% died by 90 days.

These data indicate that deep vein thrombosis is the most common of these complications, but not associated with any notable increase in mortality. Pulmonary embolism and gastrointestinal bleeding are both associated with a significant increase in mortality. These finding in conjunction with those from the review of Marcucci and colleagues demonstrate that it is a fine balance between reducing the occurrence of thromboembolic complications with pharmacological thromboembolic prophylaxis against increasing the risk of bleeding complications. It may be that we are doing more harm than good with the current heparin policies by causing more deaths from bleeding complications without any reduction in mortality from thromboembolic complications being reduced. We require large, independently conducted, randomised trials on this topic with full reporting of all relevant clinical outcomes to answer these important issues.