Risk-Stratified Venous Thromboembolism Chemoprophylaxis After Total Joint Arthroplasty: Evaluation of an Institutional Approach

Figures

• 

  Figure 1. The GRANT prescribing tool (The Grant Medical Center Risk Assessment for Negating Thromboses). Legend: this summarizes our current institutional risk-stratified recommendations for VTE chemoprophylaxis after elective total hip or knee arthroplasty. The orthopedic service pharmacist provides patient-specific recommendations for discharge using this evidence-based institutional guidance, though the ultimate prescription is at the surgeon’s discretion. Note: all patients are recommended apixaban during hospitalization at least until adequate postoperative mobility is achieved, barring significant bleeding concerns. *: Diabetes, hypertension, and hyperlipidemia are not counted as risk factors individually or in combination. Abbreviations: BID—twice daily; BMI—body mass index; EC—enteric coated; Hx—personal history; LOS—length of stay; VTE—venous thromboembolism.

• 

  Figure 2. Population determination. Legend: PTA—prior to admission; TJA—total joint arthroplasty (of the hip or knee).

• 

  Figure 3. Case–control analysis of composite complication primary outcome by VTE risk strata subgroups and in total population. Legend: the population that did not experience a complication (“No Adverse Event” group) is represented by blue bars and the complication population (“Adverse Event” group) is represented by yellow bars. The median duration of anticoagulant did not differ between Adverse Event and No Adverse Events groups at the population level (5 days vs. 5 days, p = 0.2625) or in the Routine VTE Risk subgroup (0.75 day vs. 1 day, p = 0.5966, n = 1084). Non-significant differences were seen in the Moderate VTE Risk subgroup (6 days vs. 8.75 days, p = 0.4924, n = 3993) and High VTE Risk subgroup (6 days vs. 15.5 days, p = 0.1239, n = 327). The population was truncated to those with anticoagulant durations of ≤30 days for these subgroup analyses to align with the complication assessment timeframe and avoid immortal time bias. Abbreviations: AC—anticoagulant; md—median.

• 

  Figure 4. Multivariable logistic regression analysis for predictors of antithrombotic-related complications’ composite primary outcome. Footnote: after adjusting for multicollinearity, the final multivariable logistic regression model retained seven variables as being significant predictors of the composite complication primary outcome. Variables are ordered by odds ratio. Bolded p-values indicate statistically significant findings at prespecified alpha = 0.05. Since a personal history of VTE was not found to be a significant predictor of the composite complication primary outcome in our model but is known to be a strong predictor of postoperative VTE in the broader literature, we repeated the multivariable analysis with this variable forced into the model and found similar results for the seven inception variables and p = 0.0540 for history of any VTE (OR 1.527, 95%CI 0.993–2.348), confirming stability of the model. Abbreviations: COPD—chronic obstructive pulmonary disease; CVA—cerebrovascular accident; Hx—personal history; LOS—length of stay; PTA—prior to admission; PVD—peripheral vascular disease; THA—total hip arthroplasty; TJA—total joint arthroplasty (of the hip or knee); TKA—total…

• 

  Figure 5. Composite complication primary outcome rate as a function of anticoagulant duration by VTE risk subgroup. Routine Risk (A, green bars, total n = 1074), Moderate Risk (B, orange bars, total n = 3744), and High Risk (C, red bars, total n = 295). Legend: anticoagulant duration of 0 days indicates an aspirin monotherapy chemoprophylaxis regimen. Arrows indicate expected nadir of antithrombotic-related complication risk for each subgroup. p-values for within-group chi-square tests overlaid on each panel, and Cochran–Mantel–Haenszel test p = 0.0006 for overall effect of VTE risk strata on the complication rate vs. anticoagulant chemoprophylaxis duration relationship. Only patients receiving 0–28-day anticoagulant chemoprophylaxis durations were included in these subgroup analyses to align with typical prescribing pattern timeframes, for conciseness. Abbreviations: AC—anticoagulant; AE—adverse event (e.g., ≥1 of the antithrombotic-related complications included in the primary outcome).