Letters

Differences in mortality after fracture of hip

EDITOR,--Anticoagulant prophylaxis may reduce the incidence of venous thromboembolism in hospital, but C J Todd and colleagues' study does not dispel doubts about its overall effect on mortality.¹ There was no significant difference between the survival rate at 90 days (82.2% at the four hospitals that routinely gave such prophylaxis (to 79.5% of patients)) and 82.0% at the other four hospitals (prophylaxis given to only 16.1%) despite the fact that 62.3% of patients in the former hospitals had surgery within 24 hours (favourable to outcome) compared with only 50% in the latter (P<0.01, ² test). Hospital 4, which gave prophylaxis to the highest proportion of patients (91%), showed joint lowest 90 day survival (76%). In hospitals 5 and 8, in which only 10.5% of patients were given prophylaxis, 83% survived despite 4% dying of pulmonary embolism. None of the 13 patients diagnosed at necropsy as having fatal embolism was taking anticoagulants, but we wonder how many of these patients had recognised contraindications,² especially other conditions rendering them particularly prone to thromboembolism. Haematoma with subsequent infection is a worry with anticoagulants, and wound infection occurred in 9.8% (25/256) of patients in the hospitals that routinely gave anticoagulants compared with 4.9% (15/304) in the others (P<0.01).

There is a danger of relying on anticoagulants in hospital and neglecting to teach patients and carers about mechanical prophylaxis, including postures to increase the velocity of femoral vein blood,³ the avoidance of bad postures, and the value of leg movements, since the risk of thrombosis continues after discharge from hospital.² Either infection or venous stasis would predispose to thromboembolism after discharge: did all 82 patients who died undergo necropsy?

Another factor that can affect outcome is the packed cell volume: lower values indicate lower viscosity (a value </=0.33 discourages thromboembolism⁴) and better tissue oxygenation than at higher values. Thus it would be interesting to know whether thromboembolism and survival correlated with differences in policy concerning transfusion of red cells or haemodilution. The report by the Thromboembolic Risk Factors Consensus Group can be criticised for omitting to mention posture or packed cell volume.⁵ The decision whether to give anticoagulant treatment to a patient should take into account many factors, including the risk of haemorrhage as well as the risk of thrombosis. For many patients, mechanical methods to increase the peak velocity of femoral vein blood combined with haemodilution (for example, not replacing red cells lost) could be safer overall, especially if monitored by routine postoperative venography or colour-duplex scans and if clinically important thromboses are treated. Raising the legs moderately can enhance the blood velocity³ and is easily continued after discharge. We agree that more research is needed, particularly large scale trials of alternatives to anticoagulant treatment.

Senior house officer in surgery Consultant general surgeon St Richard's Hospital, Chichester PO 19 4SE

J Calder, E C Ashby 

1. Todd CJ, Freeman CJ, Camilleri-Ferrante C, Palmer CR, Hyder A, Laxton CE, et al. Differences in mortality after fracture of hip: the East Anglian audit. BMJ 1995;310:904-8. (8 April.) [Abstract/Free Full Text]
2. Kakkar VV, Cohen AT, Edmondson RA, Phillips MJ, Cooper DJ, Das SK, et al. Low molecular weight versus standard heparin for prevention of venous thromboembolism after major abdominal surgery. Lancet 1993;341:259-64. [Medline]
3. Ashby EC, Ashford NS, Campbell MJ. Posture, blood velocity in common femoral vein, and prophylaxis of venous thromboembolism. Lancet 1995;345:319-21.
4. Duruble M. Prophylaxie des thromboses veinouses profondes. Phlebologie 1982;35:143-72. [Medline]
5. Thromboembolic Risk Factors (THRIFT) Consensus Group. Risk of and prophylaxis for venous thromboembolism in hospital patients. BMJ 1992;305:567-74.