Facebook
Twitter
LinkedIn
Email
 |
Checking for direct PDF access through Ovid | |
Excerpt
Empson et al1 performed the Herculean task of reviewing 575 studies of therapy for antiphospholipid antibody–associated pregnancy loss. Only ten of those studies fulfilled their strict inclusion criteria for performing a meta-analysis of the effectiveness of various treatment modalities. One conclusion of that study may need reexamination. The authors determined that prednisone was totally ineffective in preventing pregnancy loss. That result was based on analysis of only three studies. In the study by Cowchock et al,2 aspirin plus prednisone was as effective in preventing pregnancy loss as was aspirin plus heparin (a regimen that Empson et al found to be highly efficacious). I am surprised that the study by Laskin et al3 passed the rigid criteria for inclusion in the analysis. Only 42 of 202 treated patients had antiphospholipid antibodies, and those determinations were questionable. The antiphospholipid antibody tests were repeated 7 to 10 days apart to confirm positive reactions. Confirmation of a positive result should be performed at least 6 weeks after the initial test.4 Repeat testing performed before this time does not adequately distinguish false or transient positives from true positives. Additionally, lupus anticoagulant was detected by prolongation of coagulation, but no confirmatory tests were described. This is also unacceptable because lupus anticoagulant must be confirmed by mixing studies or neutralization of the anticoagulant by the addition of excess phospholipid.4 In the third study, by Silver et al,5 12 patients received prednisone plus aspirin and 22 received only aspirin. None of the pregnancies resulted in a perinatal loss. The total lack of perinatal loss is highly unusual and raises concerns. One necessarily makes the assumption that aspirin alone, which Empson et al claim is ineffective, completely prevented pregnancy loss in this study. One potential explanation is an apparent unusual distribution of antiphospholipid antibodies: only two of 36 patients had lupus anticoagulant, whereas 34 had anticardiolipin antibody alone. Therefore, Empson et al made their recommendation about prednisone use based on an analysis of three studies, two of which appear to be either flawed or have extremely atypical results. I would suggest the analysis be closely reexamined on this particular point before denying prednisone as an alternative to heparin therapy.
