Multimodal thromboprophylaxis for total hip and knee arthroplasty based on risk assessment.

BACKGROUND: Orthopaedic surgeons are increasingly challenged to find a prophylaxis regimen that protects patients from thromboembolism while minimizing adverse clinical outcomes such as bleeding. We used a multimodal approach in which the treatment regimen is selected according to patient risk factors.
METHODS: We retrospectively reviewed the records on 1179 consecutive total joint arthroplasties in 970 patients who had undergone primary and revision total hip and total knee replacement. Preoperatively, patients were assigned to one of two deep venous thrombosis prophylactic regimens on the basis of an assessment of their risk factors. Eight hundred and fifty-six patients (1046 operations) were considered to be low risk and were managed with aspirin, dipyridamole, or clopidogrel bisulfate as well as intermittent pneumatic calf compression devices. One hundred and fourteen patients (133 operations) were considered to be high risk and were managed with low-molecular-weight heparin or warfarin and intermittent calf compression. All patients were mobilized from bed within twenty-four hours after surgery, and all underwent Doppler ultrasonography within the twenty-four hours before hospital discharge. All of the patients were followed for six months postoperatively. The prevalence of asymptomatic and symptomatic distal and proximal deep venous thrombosis, symptomatic and fatal pulmonary emboli, overall mortality, and bleeding complications was determined. Thrombotic events were expressed as a percentage of 1179 operations because some patients had two or more operations.
RESULTS: Overall, there were no fatal pulmonary emboli, three symptomatic pulmonary emboli (prevalence, 0.25%), and five clinically symptomatic deep venous thrombi (0.4%). Sixty-one asymptomatic deep venous thrombi (5.2%) were found with use of routine postoperative Doppler ultrasound scans. There were three deaths (prevalence, 0.25%) that were unrelated to thromboembolism, and there were two nonfatal gastrointestinal bleeding events (prevalence, 0.17%). Wound hematomas occurred in association with five (0.4%) of the 1179 operations. Three nonfatal pulmonary emboli (prevalence, 0.3%) were detected in association with the 1046 procedures in the low-risk group, and none were detected in association with the 133 operations in the high-risk group (p = 0.767). Clinically symptomatic deep venous thrombosis was detected in association with four (0.38%) of the 1046 operations in the low-risk group and one (0.75%) of the 133 operations in the high-risk group (p = 0.93). Asymptomatic distal deep venous thrombosis was detected in association with thirty-seven (3.5%) of the 1046 procedures in the low-risk group and four (3.0%) of the 133 operations in the high-risk group. Asymptomatic proximal thrombosis was detected in association with fourteen (1.3%) of the 1046 procedures in the low-risk group and six (4.5%) of the 133 procedures in the high-risk group (p = 0.03). Wound hematomas occurred only in patients being managed with warfarin or low-modular-weight heparin (p = 0.0001).
CONCLUSIONS: A multimodal thromboembolic prophylactic regimen is consistent with protecting patients while limiting adverse clinical outcomes secondary to thromboembolic, vascular, and bleeding complications.