Routine Duplex Ultrasonography Screening
We recommend against routine post-operative duplex ultrasonography screening of patients who undergo elective hip or knee arthroplasty.

We cannot recommend the routine use of ultrasound for the screening of patients after knee or hip arthroplasty for VTED. The best available evidence comes from two randomized controlled studies, both of high quality and moderate applicability (see Table 14 for a summary of the results of these studies, Table 15 through Table 18 for a detailed presentation of results, and Table 47 in Appendix XIII for our appraisal of their quality and applicability), that compared routine ultrasound screening to not screening. The control group was prolonged prophylaxis in one study, and a sham ultrasound in the other. In the ultrasound groups, treatment of asymptomatic DVTs was based on the ultrasound findings. Neither study found a statistically significant difference in symptomatic PE rates (Table 15) between the ultrasound-screened and unscreened patients, despite the fact that they had adequate statistical power.

Similar results are found when screening is accomplished using venography (Table 14 summarizes the results of the studies that evaluated the effects of ultrasound and venographic screening on patient outcomes). Two retrospective comparative studies of low quality and moderate applicability (see Table 47 in Appendix XIII) compared results of patients who were screened for DVT by venography against results of patients who were not screened (Table 16). Treatment of asymptomatic DVT varied according to venographic results. Rates of readmission for PE and DVT did not significantly differ between those who received screening venography and those who did not.

The available evidence also suggests that D-dimer is not a useful screening test for DVT after arthroplasty. Three studies, one of high quality and two of moderate quality and all of moderate applicability (Table 48 in Appendix XIII), evaluated the screening performance of D-dimer. Two used ultrasound as the reference standard, while one used venography.
One study of high quality and moderate applicability evaluated the screening performance of magnetic resonance (MR) venography as compared to standard venography. These data indicated that MR venography may be a good “rule in” test but not a good “rule out” test.

Given the lack of utility of ultrasound for diagnosis of unsuspected DVT’s and the lack of any commonly available alternative screening test with greater utility, we do not recommend routine screening for DVT in the hip and knee arthroplasty postoperative patient population.

The reasons we excluded some studies initially considered for this recommendation appears in Appendix XIV, Table 57.