Determining the impact of culture on venous thromboembolism prevention in trauma patients: A Southwestern Surgical Congress Multicenter trial.

INTRODUCTION

Venous thromboembolism (VTE) remains one of the principal causes of morbidity and death in trauma patients that survive the first 24 h. Recent literature on VTE prevention focuses on choice of chemoprophylaxis, specifically unfractionated heparin (UFH) versus low molecular weight heparin (LMWH). This singular focus on a multifactorial process may be inadequate to fully understand the optimal approach to VTE prevention. We hypothesized that variations in care between trauma centers could be used to identify key components of VTE prevention associated with better outcomes.

METHODS

A 50 question survey of VTE management for years 2014-2016 was sent to 15 trauma centers. The survey included: demographics of the trauma centers, type and timing of chemoprophylaxis, ambulation expectations, and complementary services (geriatric trauma service (GTS), mobility teams, physical and occupational therapy (PT/OT)). Each center submitted their American College of Surgeons Trauma Quality Improvement Program (TQIP) Benchmark Report for Spring 2017. TQIP data included: mortality, observed rates of deep vein thrombosis (DVT) and pulmonary embolus (PE), and time to VTE prophylaxis. The survey and TQIP reports were blinded for analysis; descriptive statistics were utilized. The top DVT & PE TQIP performers were used to identify potential aspects of better care on the survey responses. The institutions' DVT and PE rates were then compared for these responses using Wilcoxon-Rank-Sum test.

RESULTS

Fifteen trauma centers (13 Level-1, 2 Level-2) completed the survey; the centers admitted 1050-7200 trauma patients per year (median 3000). The majority of centers were University-affiliated (11 of 15) with general surgery residencies (14 of 15), Acute Care Surgery or Surgical Critical Care Fellowships, (9 of 15) and critical care boarded-surgeons only on-call (9 of 15). Few have geriatric trauma services (3 of 15) or mobility teams (1 of 15). Half the trauma centers have dedicated PT/OT teams for trauma or weekend coverage. With a total of 20,878 TQIP patients analyzed, the average observed DVT and PE rates were 1.27% (range 0.1-5.2%) and 0.68% (range 0-1.6%), respectively. Weekly lower extremity surveillance duplex (2 of 15) increased DVT detection (4.15% vs 0.80%, p = 0.034) but did not decrease PE rates (1.05% vs 0.62%, p = 0.229). Great variance was seen in choice, dosing and timing of chemoprophylaxis: UFH, LMWH daily, LMWH twice-daily, LMWH weight-based dosing, and LMWH anti-Xa dosing. The top 3 performers for DVT and PE all used different types of chemoprophylaxis. These top performers had a prominent culture of mobility: dedicated PT/OT teams for trauma or weekends and an expectation to ambulate 3-times per day. Weekend PT/OT teams were associated with lower DVT rates (median 0.40%, range 0.10-1.10% vs 1.30%, 0.60-5.20%, p = 0.018), and ambulation 3-times per day was associated with lower PE rates (median 0.20%, range 0.00-0.20% vs 0.80%, 0.40-1.60%, p < 0.005).

CONCLUSIONS

Considerable variation in VTE chemoprophylaxis exists among trauma centers. "Best practices" in this area requires further investigation. An expectation of mobility and investment in mobility resources may serve to decrease VTE rates in trauma patients compared to a singular focus on type of chemoprophylaxis administered.