Colditz GA, Tuden RL, Oster G.
Rates of venous thrombosis after general surgery: combined results of randomised
clinical trials.
Lancet 1986 Jul 19;2(8499):143-6

Despite evidence that prophylaxis against deep-vein thrombosis (DVT) is
effective, a large proportion of general surgical patients receive no
prophylaxis. To determine the extent to which various prophylactic methods reduce
the incidence of postoperative DVT, data from randomised controlled trials of
these methods were combined by means of meta-analysis. The rates of DVT diagnosed
by the fibrinogen uptake test were: control/no therapy 27.0% (95% confidence
limits 21.9%, 32.1%); heparin 9.6% (7.2%, 11.9%); gradient compression stockings
11.1% (5.3%, 16.8%); intermittent pneumatic compression 17.6% (6.1%, 29.1%);
heparin plus stockings 6.3% (0%, 17.6%); heparin plus dihydroergotamine 9.9%
(6.2%, 13.6%); and stockings plus intermittent pneumatic compression 4.5% (1.1%,
8.0%). These results confirm the value of prophylaxis to reduce the incidence of
DVT and suggest that combined treatments may be most effective.