incidence of deep vein thrombosis after hip and knee arthroplasties in Japanese patients: A prospective study, The

Journal of Orthopaedic Surgery, Dec 2003 by Sudo, A, Sano, T, Horikawa, K, Yamakawa, T, Et al

ABSTRACT

Purpose. To document the incidence of proximal deep vein thrombosis and pulmonary embolism in 58 consecutive Japanese patients undergoing total hip arthroplasty or total knee arthroplasty.

Methods. Patients were routinely examined for proximal deep vein thrombosis by B-mode ultrasonography before and after surgery. Those patients who had ultrasonographic findings of deep vein thrombosis were also investigated for pulmonary embolism by ventilation-perfusion lung scan.

Results. The incidence of deep vein thrombosis after total hip arthroplasty and total knee arthroplasty were 9.1% and 4.0% respectively, and the incidence of pulmonary embolism were 3.0% and 0%, respectively. There were no cases of fatal pulmonary embolism.

Conclusion. The incidence of deep vein thrombosis and pulmonary embolism in Japanese patients may have increased over the last few decades.

Key words: deep vein thrombosis; hip replacement arthroplasty; knee replacement arthroplasty

INTRODUCTION

The reported incidence of deep vein thrombosis (DVT) in Europe and North America ranges from 12% to 23% following total hip arthroplasty (THA), and from 17% to 57% following total knee arthroplasty (TKA) when using an accepted form of prophylaxis,1"3 and without prophylaxis ranges from 48% to 70% after THA, and from 58% to 84% after TKA.2~4 As the incidence of subsequent fatal pulmonary embolism (PE) is high (1.7%-3.4%), some form of prophylaxis is commonly used after THA and TKA in those countries.3,5,6 In contrast, as the incidence of postoperative DVT in Japan has always been considerably lower, there has not been a clear indication for prophylaxis. A report written in 1964 revealed that the incidence of DVT found in an autopsy study in Japan was 3.9%.7 However, more recently the number of cases of DVT and PE reported in Japan has increased, and there have been some fatalities.

In order to clarify the incidence of DVT and PE in Japan, we prospectively studied consecutive patients before and after THA and TKA. B-mode ultrasonography was used for detecting major proximal thrombi, for which there was a risk of detachment from the venous wall, proximal migration and subsequent fatal PE.

MATERIALS AND METHODS

Between October 1996 and May 1998, we performed THA for 33 patients and TKA for 25 patients at Mie University Hospital. All patients were included in this study. Preoperative diagnoses for primary and revision surgeries at each site are shown in Table 1. Postoperatively, the lower extremity involved was compressed by an elastic bandage as the only form of prophylaxis. No antiplatelet or anticoagulant medication were used. Patients were advised to move both ankle joints immediately after operation. After bed rest for 2 to 3 weeks, wheelchair use and partial weightbearing were allowed for 6 weeks. After 6 weeks, activities progressed towards full weight-bearing. If bone grafting had been performed, more prolonged support with crutches was advised.

Preoperatively, ultrasound scanning to detect deep vein thrombi was completed for all patients, and was repeated between day 8 and day 14 after surgery. Contrast venography to confirm the diagnosis was performed in all patients in whom DVT was suspected. All patients with evidence of DVT had ventilation-perfusion scintigraphy to detect pulmonary emboli. Ultrasonography involved imaging from the common femoral vein to the popliteal veins by axial B-mode sensing, using a 7.5 MHz ultrasonography probe (EUB-165A; Hitachi Medical Corporation, Tokyo, Japan). Veins below the knee were not surveyed, because it was unlikely that these sites contained large thrombi that could cause pulmonary artery obstruction. DVT was diagnosed by decreased compressibility of veins during ultrasonography. Normal veins collapse when, compressed by probing whereas veins containing thrombi do not. PE was diagnosed on identification of a discrepancy between the ventilation scintigram and the perfusion scintigram.

The frequency of DVT, PE, and fatal PE were calculated, and risk factors including age, sex, body mass index (BMI), and number of days of bed rest were compared between patients with and without evidence of DVT. Results were analysed using the Chi squared test or Student's t test, with p

RESULTS

None of the patients had DVT prior to surgery. Four patients had a postoperative DVT detected by ultrasonography, and confirmed by venography (Table 2). There was one case of non-fatal PE among these 4 cases. The incidence of DVT was 9.1% in patients following THA and 4.0% following TKA, while that of PE was 3.0% following THA and 0% following TKA. The incidence was 6.4% in patients undergoing primary surgery, and 9.1% in those having revision surgery. All cases with DVT were females. For the patients with DVT, age at surgery was slightly older, BMI was slightly higher, and the number of days of bed rest was slightly longer, but none of these differences were shown to be statistically significant (Table 3).

DISCUSSION

DVT and PE are major complications after hip and knee arthroplasties in Europe and North America. In Japan, there have been sporadic case reports of DVT in the past. Due to the presumed low incidence of DVT in Japan, there has been a low level of clinical suspicion and poor recognition of postoperative DVT and PE, and lack of systematic approach to diagnosis, treatment, or prevention. In order to investigate whether a different therapeutic paradigm should be considered, we assessed the frequency of DVT and PE in our group of surgical patients, and compared our findings with reports from Europe and the North America, where prophylactic regimens have been in use for at least 2 decades.