Risk of myocardial infarction persisting after discontinuation of non-steroidal anti-inflammatory drugs in the general population.
J Thromb Haemost. 2009 May;7(5):892-4.
ABSTRACT
We have recently read with great interest the analysis of the extended
observational follow up in the APPROVe trial [13]. One of the most
controversial findings of this study was that, according to the original report,
CV toxicity does not appear until 18 months after treatment initiation.
Although this statement was later corrected by the authors who acknowledged
an erroneous statistical testing of this hypothesis, it now becomes clear from
this extended follow up that there is at least some evidence (although not
statistically significant) that duration of use is an independent predictor of the
risk. Furthermore this is fully compatible with the proposed mechanism,
especially in those populations that do not present a high baseline CV risk [3].
But probably the most interesting finding of the extended follow up analysis is
that the risk associated to rofecoxib use lasts at least one year after
discontinuing the drug. Unfortunately the number of observed events is rather
small and it only refers to rofecoxib, which precludes a more conclusive
inference.
To that respect, we have recently carried out an observational study using
THIN database in which we have analyzed a total of 8,852 cases of non-fatal
MI and 20,000 controls.