We need to look critically at evidence for universal use

The article by Phillips et al is yet another so called evidence based report that fails to appreciate the clinical aspect of thromboembolism.¹ The report mentions only the incidence of the surrogate outcome measure of asymptomatic venous thromboembolism. Clinical outcomes of clinical deep vein thrombosis, pulmonary embolism, and mortality are not mentioned.

The authors have not used clinically relevant outcomes. They are incorrect to state without reference that asymptomatic venous thrombosis often gives rise to long term morbidity. This is contrary to published evidence in orthopaedics, according to which such asymptomatic thrombi resolve without any associated morbidity.²

In orthopaedics there is lamentable evidence about the efficacy of graduated compression stockings. The four studies on orthopaedic patients in the Cochrane review entailed a total of only 316 participants.³ A reduction in asymptomatic thrombosis was noted but no clinical benefits recorded. A later, more detailed study of 795 participants found no benefit for compression stockings.⁴

Phillips et al recommend the universal use of compression stockings and the incorporation of measures of their use into hospital quality assessment measures.¹ This conclusion is seriously flawed. The widespread use of such measures would cost the health service millions of pounds to supply the stockings and provide assistance in applying such stockings. Good quality large and independently conducted randomised trials are urgently needed to evaluate the effectiveness of such measures before we make dramatic and unsupported statements about their use.