Symptomatic pulmonary embolism (PE), a significant and life-threatening complication following total knee arthroplasty (TKA), has been described as a “never event.” Despite a number of advancements in care, PE continues to occur following TKA. This study evaluates symptomatic PE rates over time in TKA patients enrolled in multicenter randomized clinical trials assessing the efficacy of venous thromboembolism prophylaxis regimens.
The MEDLINE and Cochrane Central Register of Controlled Trials were searched to identify clinical trials assessing prophylactic anticoagulation in patients undergoing TKA between January 1995 and December 2016. A random effect model was used to combine PE rates across studies. The pooled proportion of symptomatic PEs was calculated and heterogeneity was quantified with the I2 statistic. A 95% prediction interval was constructed to examine what the expected range in the proportion of symptomatic PEs would be in future studies. Meta-regression was used to explore the effect of time on the rate of symptomatic PEs.
A total of 18 studies representing 27,073 patients were included in the meta-analysis. The symptomatic PE rate was 0.37% (95% confidence interval, 0.24%-0.52%). There was significant heterogeneity across studies, I2 = 66%. Between 1996 and 2010, the proportion of PEs did not change in the regression analysis. The 95% prediction interval was 0.0002 to 0.0106, indicating that in similar future studies, the true proportion of symptomatic PEs would range from 0.02% to 1.06%.
Over a 14-year period, the symptomatic PE rate after TKA was relatively constant even when patients received potent anticoagulation. These results suggest that some patients may have a genetic predisposition to develop a PE and more effective risk stratification protocols need to be developed to make sure patients receive appropriate anticoagulation.