Thomas1 has proposed in the November 12, 2001, issue of the ARCHIVES a practical review of hypercoagulability syndromes. We would like to comment on several points. Indications of thrombophilia screening remain a matter of debate and we regret that this point was not discussed in the article. Limiting this debate to its financial aspect, as it could be interpreted from Table 2, which gives the cost of different assays, is not correct in our opinion. For instance, detecting an increase of factor VIII ($90) today has no recognized therapeutic implications, whereas elevated homocysteine concentration ($324) can lead to safe, inexpensive, and probably effective vitamin supplementation. Which screening to use and for which patients is a very difficult problem that has been discussed elsewhere and that must not be overlooked.2 In the summary, Thomas stated that hypercoagulability syndrome can be "correctly diagnosed" in approximately 80% to 90% of patients. This appears to be a rather presumptuous affirmation. If this means that extensive workup will lead to identification of coagulation abnormalities in this percentage of patients, this is probably true, even if perhaps a little optimistic. But as emphasized by Thomas himself, patients with antithrombin deficiency (which is the genetic thrombophilia associated with the higher relative risk of thrombosis) may have no problem until they undergo surgery or any other risk-associated situation and, in fact, the majority of carriers of antithrombin deficiency will never have thrombosis. We should probably in most cases leave the concept of thrombophilia/hypercoagulability as an explicative factor, to think in terms of risk factor. We do not "explain" a myocardial infarction by hypercholesterolemia; it is "just" an important risk factor. This way of thinking will perhaps facilitate therapeutic decisions. Thus, in the Figure that summarizes the recommended approach to treatment of hypercoagulability, Thomas suggests that people having more than 1 coagulation abnormality should be candidates for "lifelong" (another presumptuous expression in our opinion) anticoagulation after 1 episode of thrombosis, whereas people with recurrent thrombosis but no defect identified should be treated with "standard" therapy. We think this doesn't really reflect the view of the author but the Figure can be misleading and should be corrected. For instance, a 20-year-old man who has severe pulmonary embolism in the absence of favoring conditions (eg, surgery, prolonged immobilization) has indeed thrombophilia, whatever the result of hypercoagulability workup, and should receive long-term anticoagulation. Last, the Figure suggests that family studies may be prothrombin promoter gene: this attitude is extremely questionable because of the absence of therapeutic consequences in the family.