Bleeding complications following intramuscular injections among hospitalized anticoagulated patients: a retrospective observational comparative study.

Abstract

Background: Bleeding complications following intramuscular (IM) injections are generally considered rare with reported incidence of 0.06%. However, evidence on safety of IM injections among anticoagulated hospitalized patients is lacking. The objective of the current study was to examine the incidence of injection site bleeding complications following IM injection among anticoagulated hospitalized patients.

Methods: A retrospective comparative study comprised of all hospitalized patients ≥ 18 years old that were treated with ≥ 1 IM injection between 2009 and 2019 in a large tertiary medical center. Bleeding complications were defined as focal hematoma, local bleeding, intramuscular bleeding or compartment syndrome. Each case with IM injection was searched for ICD9 codes (e.g., hematoma, hemorrhage or compartment syndrome) and for indirect evidence suggestive of potential bleeding: hemoglobin drop ≥ 2 g/dl, AST or CPK increase, packed red blood cell transfusion, or abrupt cessation of the anticoagulation. These case were then verified for true injection-site bleeding by natural language processing model and manual review of the electronic medical record.

Results: A total of 71,710 patients were treated with 236,406 IM injections. Mean age 53 (± 22) and 63% were females. Concomitant anticoagulation (Heparins: 90.3%, warfarin: 6.8% and DOACs: 4.7%) occurred in 40,819 IM injections (8189 patients). Suspected bleeding complications at the IM injection site were identified among 7,111 patients following 23,089 IM injections, the majority were unrelated to the IM injection-site (e.g., gastrointestinal bleeding, retroperitoneal, etc.). Two cases were verified as true injection site bleeding complication, both in the anticoagulated group (2/8189, 0.02%).

Conclusion: Bleeding complications at site of IM injections among anticoagulated hospitalized patients are rare, and their risk is probably not higher compared to patients without anticoagulation.