Time, money and precious hospital
space could be freed up at no cost in lives if patients
with venous thromboembolism (VTE) were treated with
older, cheaper heparins, according to Canadian research
published in the Journal of the American Medical
Association.
This is a treatment that has already
seen considerable cost savings in recent years, despite
the adoption of more expensive drugs. Until this decade,
the standard approach to venous thromboembolism was
warfarin therapy combined with infusion of unfractionated
heparin. To ensure that coagulation remained within
safe bounds, the patient required costly hospitalization.
HOMEWARD
BOUND
The arrival of low molecular weight (LMW) heparins freed
the patient from the need for close monitoring. Administered
subcutaneously, it is judged safe enough to let the
patient go home. So even though it costs about 20 times
as much as unfractionated heparin, LMW heparin has delivered
major savings to the health system.
But a team of Canadian and New
Zealand researchers say that equal results are achievable
using the older heparins on an outpatient basis. Given
subcutaneously, they argue, the bleeding risk is minimal
and there's no need for constant testing of coagulation.
"The current situation," says Dr
Clive Kearon of McMaster University and the Henderson
Research Centre in Hamilton, Ontario, "is that both
treatments are being used in this country, but low molecular
weight heparins are gaining ground and slowly becoming
the gold standard."
NONINFERIORITY
COMPLEX
Dr Kearon and colleagues at six centres in Canada and
New Zealand set out to prove the "noninferiority" of
unfractionated heparins. They recruited 708 adults with
acute venous thromboembolism and randomized them to
receive either the old or the new heparins.
All patients had the heparins administered
subcutaneously, and about three-quarters in both groups
were treated largely or entirely as outpatients.
Coagulation monitoring would be
expected in patients receiving infused unfractionated
heparins, but the researchers doubted its usefulness.
So they applied the standard activated partial thromboplastin
time (aPTT) test, not to adjust dosing, but to measure
its predictive value in outcomes. The usual warfarin
therapy was applied in all patients.
Tx
TIGHTROPE
Treating VTE is a high-wire act in which the physician
must guide the patient safely between the risk of further
clotting on the one hand, and excessive blood thinning
on the other. The study's endpoints reflected this,
measuring both the rate of recurrence of VTE and also
incidents of major bleeding.
The bleeding rates were similar.
During the first 10 days, major bleeding occurred in
1.1% of patients in the unfractionated heparin group
and 1.4% of the low molecular weight heparin group.
In three months of follow-up, it occurred in 1.7% of
the unfractionated heparin group and 3.4% of the low
molecular weight heparin group. There was one case of
fatal bleeding in each group.
As for the treatment's efficacy,
3.8% of patients using the older heparins experienced
recurrent venous thromboembolism, versus 3.4% of those
using the newer heparins. This proves the unfractionated
heparins "noninferiority" to a statistical p value of
0.02. There were six pulmonary embolisms, four of which
occurred in the low molecular weight heparin group.
NOT
aPtT TO HELP
When the aPTT test was applied to 197 of the patients
taking unfractionated heparins, it proved to have little
predictive value. There were five cases of recurrent
VTE in this group, but none of them occurred in the
patients with aPTT times of less than 60 seconds. And
since none of these 197 patients suffered major bleeding,
the test had no value in predicting that outcome either.
"I suspect our conclusions on the
aPTT test may raise more eyebrows than our findings
on heparins themselves," says Dr Kearon. "We're not
suggesting that the test isn't vital in all sorts of
settings, but we don't think it's necessary with subcutaneous
administration of heparins because you're not using
it to adjust dosing, since the dose is fixed. And the
rate of events is very low, which limits its predictive
power."
|
