Pulmonary Embolism Prevention (PEP) Trial Collaborative Group.
Prevention of pulmonary embolism and deep vein thrombosis with low
dose aspirin: Pulmonary Embolism Prevention (PEP) Trial. Lancet. 2000
Apr 15; 355:1295-302.
Question: In patients having surgery for hip fracture
or those having elective arthroplasty, is low-dose aspirin effective
in preventing deep venous thrombosis (DVT) and pulmonary embolism
(PE)?
Design: Randomized (allocation concealed), blinded
(patients, clinicians, and outcome assessors), placebo-controlled
trial with 35-day follow-up.
Setting: 148 hospitals in Australia, New Zealand,
South Africa, Sweden, and the United Kingdom.
Patients: 13,356 patients (mean age, 79 years; 79%
women) who had a femoral neck fracture or other fracture of the
proximal femur. 4088 patients having elective hip or knee arthroplasty
(mean age, 67 years; 53% women) were also enrolled. Patients for whom
there was a clear indication for aspirin were excluded. Follow-up
for in-hospital morbidity and 35-day mortality was >99% for all
patients.
Intervention: Hip fracture patients were allocated
to low-dose enteric-coated aspirin, 160 mg daily for 35 days (n =
6679), or to placebo (n = 6677). 1332 hip arthroplasty patients
were allocated to aspirin and 1316, to placebo; 715 knee arthroplasty
patients were allocated to aspirin and 725, to placebo.
Main outcome measures: DVT, PE, other vascular events, mortality,
and bleeding.
Results: Analysis was by intention-to-treat. Among
hip fracture patients, aspirin reduced symptomatic DVT (P = 0.03*)
and PE (P = 0.002*), including fatal PE (P =
0.002*), compared with placebo (TableTable). The groups did not differ
for nonfatal myocardial infarction or fatal ischemic heart disease
(P = 0.05*), nonfatal or fatal stroke (P =
0.6*), or overall mortality (P = 0.6*). Aspirin
was associated with an increase in postoperative bleeding requiring
transfusion (2.9% vs 2.4%; relative risk increase, 24%; P =
0.04*), but fatal bleeds were rare in both groups (0.2% vs 0.2%).
Among patients having elective arthroplasty, DVT or PE was diagnosed
in 1.1% of patients receiving aspirin and in 1.4% of patients receiving
placebo (hazard ratio, 0.81; 95% CI, 0.47 to 1.42; P =
0.41*).
Conclusions: In patients having surgery for hip fracture,
low-dose aspirin compared with placebo reduced deep venous thrombosis
and pulmonary embolism, including fatal events. In patients having
elective hip or knee arthroplasty, no reduction was found.
Sources of funding: Health Research Council, National
Heart Foundation, Wishbone Trust, Auckland Orthopaedic Society (New
Zealand); National Health and Medical Research Council of Australia;
British Heart Foundation; Bayer AG; Sterling Winthrop; Aspirin Foundation; Reckitt
and Coleman; SmithKline Beecham.
For correspondence: Dr. A. Rodgers, Clinical Trials Research
Unit, Department of Medicine, University of Auckland, Private Bag
92019, Auckland, New Zealand.
*P value calculated from data in article.