Sir,
We thank Dr Morgan for his interest in our paper, however his
comments suggest a superficial reading of our paper and a lack of
understanding of statistical methodology. Most of his concerns are
addressed in our paper.
Our paper was an audit and not a randomised controlled trial. This
fact is acknowledged in the title and in the methods section of the paper.
As a result, patients were not stratified for risk factors. This accounts
for the chance occurrence of significantly more patients with known risk
factors for thromboembolic disease in the control group. We used multiple
logistic regression analysis (a commonly used statistical method that is
particularly appropriate in this circumstance) to determine the
independent quantitative effect of each identified factor (i.e. the
quantitative effect of each factor when all other factors were taken into
account) on the risk of developing post-operative venous thromboembolism
(VTE). As would be expected, our results showed that in our sample of
patients several established risk factors for thromboembolic disease
influenced the risk of post-operative VTE, but by far the greatest
independent influence on the risk of post-operative VTE was whether or not
patients were in the early mobilisation group and whether patients walked
early or not.
Dr Morgan states that deep vein thrombosis (DVT) usually occurs at two
weeks post-operatively, but does not provide any evidence for this.
Establishing exactly when deep vein thromboses occur after arthroplasty
remains the subject of controversy. In a review of 991 patients who had
undergone primary total knee replacement, Warwick and Whitehouse1
reported that the sixth post-operative day was the median day of the onset
of symptoms for those patients who developed DVT. It is reasonable to
assume that the thrombus is established before this. Sikorski et al2
found that the peak prevalence of DVT was on the fourth post-operative day.
On the other hand, Kakkar et al3 found that 29% of thrombi developed
between the 1st and 12th post-operative days, 48% on the 13th or
14th day, and 23% between the 15th and 24th days. We
did not set out to prove that early mobilisation eradicated
post-operative VTE but that it might reduce the risk.
Consequently, what matters is not at what time point the scans were done
but that timing was consistent across both groups. In fact, it could be
argued that an earlier time point is more likely to demonstrate the effect
of early mobilisation.
We believe that the evidence presented in the paper is convincing but
acknowledge that it is not conclusive. This is implicit in the title of
the paper with the use of the conditional “may”.
E.O. Pearse, Orthopaedic Specialist Registrar,
B.F. Caldwell,
R.J. Lockwood,
J. Hollard,
Toronto Private Hospital,
Toronto, Australia.
1. Warwick DJ, Whitehouse S. Symptomatic venous thromboembolism after
total knee replacement. J Bone Joint Surg [Br] 1997;79-B:780-6.
2. Sikorski JM, Hampson WG, Staddon GE. The natural history and
aetiology of deep vein thrombosis after hip replacement. J Bone Joint Surg
[Br] 1981;63-B:171-7.
3. Kakkar VV, Fok PJ, Murray WJ, et al. Heparin and
dihydroergotamine prophylaxis against thrombo-embolism after hip
arthroplasty. J Bone Joint Surg [Br] 1985;67-B:538-42.
eLetters: ![[Read eLetter]](/icons/misc/sectionDOWN.gif){kind=icon}
