Moll and Ortel report that the use of the international normalized ratio (INR) to standardize prothrombin times is invalid for some patients with lupus anticoagulants, who should be individually monitored with a test that is insensitive to lupus anticoagulants [1]. Their conclusions support our suggestion that monitoring of oral anticoagulation with prothrombin time in patients with lupus anticoagulants should be performed at high plasma dilutions to avoid unnecessary excessive anticoagulation for patients who may have major bleeding complications [2]. In a series of 10 patients with persistent lupus anticoagulants and venous thromboembolic disease who had been treated with acenocoumarol for a total of 40 patient-years, we monitored INRs with a recombinant thromboplastin reagent (Recombiplastin, Ortho Diagnostic Systems, Raritan, New Jersey) and with a combined thromboplastin (Pro-IL Complex, Instrumentation Laboratory; Milano, Italy), an international sensitivity index-certified reagent that permits reporting of INRs of prothrombin-proconvertin times [3]. Acenocoumarol dosage was always assigned on the basis of the results obtained with the combined reagent with the goal of an INR of 2.0 to 3.0. Single INR determinations with both reagents were obtained throughout the study period in 110 patients who were receiving stable oral anticoagulation but were free of lupus anticoagulants. Results with the recombinant reagent were significantly higher than results with the combined reagent in patients with lupus anticoagulants but were lower in patients without lupus anticoagulants. After INR values with recombinant thromboplastin were corrected by using local international sensitivity index calibration in patients without lupus anticoagulants, INRs of 3.1 and 4.6 with recombinant thromboplastin corresponded to INRs of 2.0 and 3.0 with combined thromboplastin in patients with lupus anticoagulants, an observation that clearly questions the validity of the INR system in the monitoring of oral anticoagulation in these patients [2].

Moll and Ortel state that because the prothrombin-proconvertin time is not available to many clinicians, other approaches to the management of anticoagulant therapy in patients with lupus anticoagulants must be identified. However, they do not provide a valid alternative [1]. Because lupus anticoagulants may be present in up to 14% of patients who present with venous thromboembolism for the first time [4] and because it has been recommended that these patients receive life-long oral anticoagulation [5], a high proportion of these patients are monitored by laboratories at any given time. Laboratory directors should be aware that choosing reagents on the basis of the prothrombin-proconvertin time is the only realistic approach to the correct monitoring of oral anticoagulant therapy in patients with lupus anticoagulants.