Background: Venous thromboembolism (VTE), both pulmonary embolism (PE) and deep vein thrombosis (DVT), causes morbidity and mortality in hospitalized patients. The duration of VTE risk in trauma, particularly after discharge, is not well understood, especially in the context of shortened hospital stays. Although guidelines provide recommendations for extended VTE prophylaxis after major orthopedic surgery, such guidance is lacking in trauma. We aimed to describe the time course and risks for symptomatic VTE in trauma patients at our institution as well as prevention strategies.

Methods: In our case control study, we identified cases and controls from our institutional trauma registry. We used an existing information technology tool (J Hosp Med 2014) to identify all VTE events diagnosed during hospitalization and up to 90 days after discharge in adult trauma patients between 1/1/2013 and 9/30/2014. Controls were a 2:1 random sample of adult trauma patients hospitalized at the same time without VTE. Those who died in the emergency department or operating room on the day of admission were excluded. VTE data were linked with demographic and clinical information abstracted from the Electronic Health Record including possible VTE risks, prophylaxis type, and activity restriction in hospital and at discharge. Injury characteristics including injury severity score (ISS) were obtained from the trauma registry. We used proportions and means to describe our sample and chi-square tests and t-tests to compare groups.

Results: Over 21 months, 9,185 adults were hospitalized with trauma. VTE was diagnosed in 161 (1.8%): 124 (77%) in hospital; 37 (23%) within 90 days of discharge. 42% had PEs; 62% DVTs. VTE was diagnosed an average of 15 days (range 1-112) after admit. Patients with and without VTE were similar in age (51 vs 52 years); 79% were men. There were no significant differences in personal VTE history, family VTE history, or malignancy between groups. Patients with VTE had lower ISS (mean 20 vs 24, p=0.002). When compared with surviving patients with no VTE who followed up, patients with VTE after discharge had similar rates of in-hospital guideline-directed prophylaxis (95% vs 92%, p=0.11) but were more likely to have restricted activity in hospital (57% vs 41%, p<0.001) and at discharge (62% vs 45%, p=0.05). Those with post-discharge VTE were significantly more likely to be discharged with pharmacologic prophylaxis than those with no VTE (73% vs 45%, p<0.001).

Conclusions: Only VTE diagnosed at our institution could be captured; results likely under-represent post-discharge VTE. While patients with and without VTE had high rates of in-hospital prophylaxis, those with post-discharge VTE were more likely to have restricted activity. We were surprised to find that they had higher rates of pharmacologic prophylaxis at discharge. Our results suggest that VTE risk remains significant after discharge despite prophylaxis. Additional study to guide optimal VTE prophylaxis at discharge is warranted.