Despite chemical prophylaxis, final rate may be considered “high”

Advances in thromboprophylaxis and improved adherence to prophylactic guidelines should theoretically lead to lower rates of venous thromboembolism (VTE). However, a large-scale Danish study published in the July 6 issue of The Journal of Bone and Joint Surgery–American shows that the incidence of VTE among knee arthroplasty patients actually increased over the 10-year study period (1997–2007), from 0.82 percent in the first 3 years to 1.32 percent in the final 3 years, when more than half the surgeries were performed.

According to Alma B. Pedersen, MD, PhD, and coauthors, Denmark takes a relatively aggressive approach to chemical thromboprophylaxis; more than 80 percent of the study’s patient population were given some form of low-molecular-weight heparin, and all but 0.5 percent of knee replacement patients receive some form of pharmacologic thromboprophylaxis. This, combined with contemporary clinical methods such as early mobilization for knee arthroplasty patients, might have been expected to have lowered the risk of VTE; instead, the authors found VTE after knee surgery to be on the rise

The authors proposed the following possible explanations for the surprising results:

• a decline in length of hospitalization time for knee arthroplasty
• noncompliance with guidelines for postdischarge thromboprophylaxis
• improved diagnostic methods and tests to identify patients with suspected VTE

Patient characteristics
Selected characteristics of the study population, which was assembled from the Danish Knee Arthroplasty Registry, are shown in Table 1. Of the 441 knee procedures (1.2 percent) for patients who received thromboprophylaxis and were hospitalized with VTE within 90 days after surgery, 323 (0.9 percent) of patients had deep vein thrombosis (DVT) and 127 (0.3 percent) had pulmonary embolism; 9 patients had both. The length of hospital stay did not affect the overall risk for VTE, but patients who were hospitalized for more than 11 days were at higher risk than those who were hospitalized for less than 6 days.

Patients who had a primary knee arthroplasty for rheumatoid arthritis had a lower rate of VTE than those with primary osteoarthritis. By age group, the risk of hospitalization for VTE was greatest for patients older than 80 years. In addition, the following groups also had an elevated risk of VTE:

• Patients with a high Charlson comorbidity index (a predictor of 10-year mortality based on a range of clinical comorbidities)
• Patients who had previously been hospitalized with cardiovascular disease
• Patients who had previously been hospitalized with VTE

Type of anesthesia, operative time, and laterality were not associated with higher risk. Patients who underwent partial knee arthroscopy (medial or lateral) had a low overall rate of VTE.

Of the patients receiving low-molecular-weight heparin, 33 percent received dalteparin, 33 percent received enoxaparin, 15 percent received tinzaparin, and 2 percent received fondaparinux; the remainder received a mixed group of other drugs. The 0.5 percent of patients who did not receive pharmaceutical thromboprophylaxis were excluded from the study, as were those who had bilateral knee replacement in a one-stage procedure.

The authors wrote that the overall rate of 1.2 percent “could be considered high, given previous research showing that venous thromboembolism is associated with mortality and a new VTE event.” They urged caution in interpreting the data, noting the lack of individual data on the duration and type of thromboprophylaxis and on possible variation in VTE prevention strategies during the study period.

They also noted, “Although studies in other settings have suggested that age alone is a less important risk factor for venous thromboembolism than underlying comorbidity, this does not appear to apply to orthopaedic patients.”

The authors concluded that future investigations should focus on the improvement of prophylaxis following hospital discharge, particularly among elderly patients and those with a history of cardiovascular disease or a previous VTE.

Online commentary
In an online commentary, JBJS associate editor Seth S. Leopold, MD, professor in the department of orthopaedics and sports medicine at the University of Washington, praised the study as “well done,” while noting that it may be limited by “the usual suspects known to beset registry studies ... namely, a low, but non-zero, likelihood of administrative error and misclassification, and an underestimation of the frequency of comorbid conditions.”

Dr. Leopold also raised two questions about the study—first, whether the frequency of the use of chemical thromboprophylactic agents changed, and second, why the authors decided to attribute all study deaths to thromboembolism, “when nearly every other paper that has looked at mortality has found most deaths in this context to be from other causes.”

The authors had reported that the overall rate of VTE was 1.6 percent, based on the 441 patients reported to have a VTE event and “161 deaths due to all causes.” Dr. Pedersen responded that the deaths were attributed in this way “to calculate the incidence of VTE in the worst-case scenario, as a compensation for possible underestimation of VTE incidence.”

In comparing the chemical thromboprophylaxis protocol recommended by the American College of Chest (ACCP) physicians with AAOS clinical practice guidelines on prevention of pulmonary embolism, Dr. Leopold noted that the ACCP protocol may be “more aggressive than necessary for most patients.” He expressed a preference for the AAOS guidelines, which allow for the use of aspirin as an alternative to stronger anticoagulants in selected patients.

Although current approaches to treatment, most notably earlier, more aggressive mobilization, are frequently promoted by “the aspirin-for-DVT-prophylaxis” proponents, the Danish study provides fairly convincing evidence “that thromboembolic disease is not a smaller problem now than it was 10 years ago,” wrote Dr. Leopold.

He also noted that aggressive chemical thromboprophylaxis is viewed by some surgeons as “a ‘U.S. thing,’ driven by a medical malpractice system run amok.” But in investigating this viewpoint, he found that ACCP-compliant chemical prophylaxis for knee replacement is used slightly more often outside the United States than within the country. Furthermore, he noted that Denmark is a less litigious environment than America, so that Danish surgeons “are not prescribing these drugs for the lawyers,” he said. “They are prescribing these drugs because they believe the drugs are right for patients having total knee arthroplasty, and there is literature to support that point of view.”

“Venous Thromboembolism in Patients Having Knee Replacement and Receiving Thromboprophylaxis” is a prognostic Level II study. Dr. Pedersen’s coauthors are Frank Mehnert, MSc; Soren P. Johnsen, MD, PhD; Steen Husted, MD, DMSc; and Henrik T. Sorensen, MD, DMSc.

Disclosure information: Dr. Leopold—Journal of Bone and Joint Surgery–American; Clinical Orthopaedics and Related Research. The other authors reported no conflicts.

Terry Stanton is senior science writer for AAOS Now. He can be reached at tstanton@aaos.org

Bottom Line

• Symptomatic (VTE) events—with a previously reported frequency of 1.3 to 2.3 percent of patients undergoing knee arthroplasty receiving thromboprophylaxis—remain a serious problem.
• This large-scale Danish study of knee replacement patients, most of whom received low-molecular-weight heparin, detected an increase in thromboembolism over the 10-year study period, with a rate of 1.2 percent over the period.
• Factors associated with elevated risk included age (especially older than 80 years), comorbidities, and previous thromboembolism. Type of anesthesia was not associated with increased risk.
• The increase in VTE occurred despite evolving approaches in knee replacement that include more aggressive mobilization.

Additional Links:
ACCP chemical thromboprophylaxis protocol
AAOS clinical practice guidelines on preventing venous thromboembolic disease