The incidence of venous thromboembolism after total hip arthroplasty: a specific hypotensive epidural anesthesia protocol
G H Westrich,C Farrell,J V Bono,C S Ranawat,E A Salvati,T P Sculco
DOI: https://doi.org/10.1016/s0883-5403(99)90101-8
Abstract:We retrospectively reviewed all consecutive unilateral primary total hip arthroplasty (THA) procedures performed by 3 attending surgeons on the Arthroplasty Service at our institution from January 1, 1990, to December 31, 1993. All surgery was performed under a specific hypotensive epidural anesthesia protocol. Hypotensive epidural anesthesia at our institution provides a lower level of hypotension (mean arterial pressure of 50-60 mmHg) as compared to hypotensive anesthesia used more generally around the world (mean arterial pressure >70 mmHg). For each patient, hospital and postdischarge office records for a minimum of 3 months after surgery were reviewed for the type of postoperative screening test, the incidence of deep venous thrombosis (DVT), and the incidence of symptomatic pulmonary embolism (PE). Overall, 2,592 primary unilateral THAs were performed with 78.6% (2,037 of 2,592) of patients receiving a venogram. Our protocol for thromboembolic disease prophylaxis in these patients included aspirin postoperatively as well as antithromboembolic disease stockings and early ambulation (24-48 hours postoperatively). The 555 patients who did not receive venography were managed with a different protocol that included warfarin postoperatively as well as antithromboembolic disease stockings and early ambulation. This high-risk group consisted of patients who received warfarin preoperatively (ie, cardiac valve) or patients with a history of DVT who were to receive warfarin postoperatively, regardless of venography result. Overall, DVT was diagnosed in 10.3% (210 of 2,037) of patients who had a venogram. Of these patients who had venography, 2.3% (46 of 2,037) had an isolated proximal DVT; 6.0% (123 of 2,037), a distal DVT; and 2.0% (41 of 2,037), both a proximal and a distal DVT. Of the 87 cases of proximal DVT identified, 60.9% (53 of 87) were femoral DVT; 18.4% (16 of 87), popliteal DVT; and 20.7% (18 of 87), both femoral and popliteal DVT. Of the 164 distal DVT, 68.3% (112 of 164) were major calf DVT and 31.7% (52 of 164) were minor calf DVT. The overall incidence of major venous thrombosis (sum of proximal and major calf thrombi) was 9.8% (199 of 2,037) in patients who had venography. Ventilation-perfusion scanning was used selectively in patients symptomatic for PE. Overall, symptomatic PE was diagnosed by ventilation-perfusion scan in 1.0% (26 of 2,592) of patients, with 0.58% (15 of 2,592) of patients having an in-hospital PE. Of the 15 patients who had an in-hospital PE, 11 patients had a venogram, and only 3 of 11 were positive. Late symptomatic PE was defined from discharge (mean, 7 +/- 2 days) to 3 months after discharge from the hospital and occurred in 0.42% (11 of 2,592) of patients. One of the 11 late symptomatic PEs was fatal. In the overall study, this represents 0.04% (1 of 2,592) fatal PE. Of the 11 patients with a late symptomatic PE, 10 had venograms in the hospital, and all 10 were negative for DVT. Overall, in the patients with a positive venogram, the incidence of symptomatic PE was 1.4% (3 of 210), whereas in the patients with a negative venogram, the incidence of symptomatic PE was 0.44% (8 of 1,827). At our institution, patients who undergo primary THA performed with hypotensive epidural anesthesia, postoperative aspirin, antithromboembolic disease stockings, and early ambulation have a low risk for thromboembolic disease.