Current use of antipsychotics significantly increases the risk of
venous thromboembolism, according to US researchers.
The findings were drawn from an analysis of data from the UK-based General Practice
Research Database by Dr Gwen Zornberg and Dr Hershel Jick (Boston collaborative
drug surveillance programme, Boston university school of medicine).
Current exposure to conventional antipsychotics was associated with a “significantly
increased risk of idiopathic venous thromboembolism” compared with non-use,
with an odds ratio of 7.1, they say. The risk was highest during the first three
months of drug use and the relationship between antipsychotics and venous thromboembolism
was strongest with short-term use.
Odds ratios did not differ between types of antipsychotics; no difference was
found between phenothiazines, thioxanthenes or other conventional antipsychotics.
However, a stronger association was found for low-potency antipsychotic drugs,
eg, chlorpromazine and thioridazine, than for high-potency antipsychotics, such
as haloperidol. There was no suggestion of higher risk with higher doses.
The authors say that possible explanations for the reaction include an association
between conventional antipsychotics and enhanced platelet aggregation or an
increase in anticardiolipin antibodies (associated with an increased risk of
thromobosis). Alternatively, venous stasis could be exacerbated by sedation,
a side effect commonly found with low-potency antipsychotic drugs.
The study included 29,952 patients under 60 years of age, who had used at least
one antipsychotic medicine between 1990 and 1998. From this group, the researchers
identified 42 patients with first-time venous thromboembolism who had no clinical
risk factors and compared them with 168 matched controls (Lancet 2000;356:1219).
In an accompanying editorial (ibid, p1206), Dr Victor Tapson (division of pulmonary
and critical care medicine, Duke university medical centre, Durham, US) says
that a relationship between antipsychotic medicines and venous thrombo-
embolism was first suggested about four decades ago but that evidence to prove
such a link was inconclusive. This study indicates that the association should
be re-examined. He adds that on the basis of the study’s findings, it would
be inappropriate for patients on antipyschotic medicines to receive anticoagulation
prophylaxis against venous thromboembolism. However, advice to avoid bedrest
and to increase the amount of ambulation (factors which the study did not address)
would be prudent.