Abstract

Introduction: Venous thromboembolism (VTE) is a serious and sometimes fatal condition, encompassing deep vein thrombosis (DVT) and pulmonary embolism (PE). Evidence suggests that major orthopaedic surgery, such as total hip replacement (THR) and total knee replacement (TKR), is associated with a high risk of postoperative VTE. Because of this risk, the American College of Chest Physicians (ACCP) guidelines recommend that these patients receive prophylaxis with anticoagulant therapy. However, the degree to which these guidelines are followed in routine practice is variable. In addition, few studies have documented the relationship between the variability in guideline alignment and VTE rates. Therefore, a retrospective database analysis was conducted to

  • determine the extent to which the ACCP guidelines for VTE prophylaxis are followed after THR/TKR and

  • evaluate the incidence of VTE of those who received ACCP recommended prophylaxis according to the guidelines (‘ACCP’) and those who did not (‘non-ACCP’).

Methods: In order to fully evaluate VTE prophylaxis patterns and outcomes, patients should be followed from the time of surgery through to hospital discharge and subsequently to care in the community. To our knowledge, a single database that captures this information for a specific patient in these three phases is not yet available. Therefore, a claims database associated with a large US health plan was linked to the Premier database, which provides details of inpatient medication use. Patients age ≥18 years undergoing TKR/THR between April 1, 2004 and December 31, 2007 and enrolled in the health plan 90 days prior to and 90 days following discharge from index hospitalization (or until death) were included in the analysis. Patients were considered to have received ACCP-guideline prophylaxis if they:

  • initiated prophylaxis in hospital with LMWH, fondaparinux, or VKA

  • initiated prophylaxis within one day of surgery (for THR patients) and

  • were prescribed prophylaxis for a minimum of ten days, or until the occurrence of a bleeding event, a VTE-related event, or death.

In addition, the number of DVTs and PEs occurring in ACCP and non-ACCP patients was recorded. Multivariate logistic regression was used to assess whether there was a significant relationship between ACCP guideline prophylaxis and the probability of a DVT or PE following THR/TKR.

Results: Of the 30,644 eligible patients from the health plan, 3,497 patients could be linked to the inpatient database. Except for geographic indicators, there were no significant differences in patient demographics or baseline co-morbidities between those included and excluded from the final study sample. Of the 3,497 linked patients, 1,395 received ACCP guideline prophylaxis (40%). The number of DVTs occurring in the ACCP and non-ACCP groups were 27 and 62, respectively. Thus, 2.01% of ACCP and 3.76% of non-ACCP patients have a DVT following surgery and up to the end of 90-day followup (p=0.0521), suggesting that non-ACCP patients were almost twice as likely as ACCP patients to have a DVT. The number of PEs occurring in the ACCP and non-ACCP groups were 2 and 25, respectively. Thus, 0.14% of ACCP and 1.19% of non-ACCP patients experienced a PE following the date of surgery and up to the end of 90-day follow-up (p<0.0001), suggesting that non-ACCP patients were more than eight times more likely than ACCP patients to experience a PE. Multivariate logistic regression indicated that the probability of DVT and PE were significantly lower in patients receiving ACCP guideline prophylaxis (p=0.008 and p=0.03, respectively)

Conclusions: By following patients from surgery through to care in the community, this study offers a unique perspective on ‘real-world’ prophylaxis patterns and clinical outcomes related to such patterns in THR and TKR patients. It suggests that:

  • only 40% of THR/TKR patients receive prophylaxis that is in alignment with ACCP guidelines and that

  • patients receiving prophylaxis according to the ACCP guidelines were less likely to have a DVT or PE than those not receiving ACCP guideline prophylaxis, the latter were almost twice as likely to have a DVT and more than eight times as likely to experience a PE.

Disclosures: McDonald:Bayer HealthCare: Employment. Crowther:Various Pharma companies: Consultancy, Honoraria, Membership on an entity’s Board of Directors or advisory committees, Research Funding, Speakers Bureau. Selby:Bayer Healthcare: Honoraria. Wells:Bayer Healthcare: Consultancy, Honoraria.

Author notes

Corresponding author