1709 potential toxicity of large doses of mesalazine given during pregnancy has been examined in animals.3 In rats, the no- observed-adverse-effect level (NOAEL) was in two studies 200 mg/kg bodyweight per day, and the lowest-observed- effect level for parental toxicity was 400 mg/kg per day. In rabbits, NOAEL was 800 mg/kg bodyweight per day.3 All these dose-regimens are far higher than those used in women. If we accept Colombel’s findings, two main issues are raised: (a) do we have to review design of protocols in pre-clinical studies and (b) should we conduct new studies in pregnant animals to mimic the case report in a woman? The potential renal prostaglandin inhibition during mesalazine treatment seems questionable. Indeed, although high concentrations of mesalazine (ie, 10-3 mol/L) inhibit prostaglandin synthesis in vitro, concentrations similar to those seen at the systemic level and in the cord at birth (ie, 1 0---- 10-6 mol/L) do not, and even stimulate prostaglandin synthesis.4 Mesalazine-delivering drugs differ in the systemic bioavailability of mesalazine, as indicated by the percentage of systemic passage and peak plasma concentrations. The latter depend also on the daily dose and administration schedule. Mesalazine-delivering drugs that have high systemic concentrations and in high doses should therefore be used with caution during pregnancy.’ Ph Marteau, C B Devaux Ferring SA, 7 rue Jean Baptiste Clément, 94250 Gentilly, France 1 Habal FM, Hui G, Greenberg GR. Oral 5-aminosalicyclic acid for inflammatory bowel disease in pregnancy: safety and clinical course. Gastroenterology 1993; 105: 1057-60. 2 Mogadam M, Dobbins WO, Korelitz BI, Ahmed S. Pregnancy in inflammatory bowel disease: effect of sulfasalazine and corticosteroids on fetal outcome. Gastroenterology 1981; 80: 72-76. 3 French Registration Dossier Part 2—Ref 10-11-13. 4 Greenfield SM, Punchard NA, Teare JP. Thompson RPH. Review article: the mode of action of the aminosalicylates in inflammatory bowel disease. Aliment Pharmacol Ther 1993; 7: 369-83. 5 Järnerot G. Prise en charge des maladies inflammatoires de l’intestin au cours de la grossesse. In: Management of inflammatory bowel disease during pregnancy. Res Clin Forum 1993; 15: 137-43. Modified APC resistance assay for patients on oral anticoagulants SIR-Jorquera and colleagues (Oct 22, p 1162) describe a modification of the test proposed by Dahlback to improve sensitivity and to use the test in patients on oral anticoagulants. We tried a similar approach after the report by Sun and co-workers’ was published. These workers used dilutions of plasma samples in factor V deficient plasma to demonstrate the role of factor V in activated protein C (APC) resistance. We have developed this test to overcome the limitation of the original test, which is that many patients with unexplained thrombosis are on oral anticoagulants. We report results for 64 patients: 43 without and 21 with oral anticoagulants in whom we have compared the original and modified methods. The original test was done with the Coatest APC resistance kit (Chromogenix, Sweden) on a KC 10 instrument (Amelung, Germany). Results obtained in citrated plasmas were expressed as APC-ratio (activated partial thromboplastin time [APTT] with APC, divided by APTT without APC). Normal values were determined in 50 healthy subjects taking no medication. A ratio lower than 2-2 (mean -2 SD) was regarded as showing APC resistance. Modified APC resistance assays were done with plasma diluted 1 in 10 and 1 in 5 in factor V deficient lyophilised plasma (Diagnostica Stago, France). Although there was a significant correlation (r=0-91) between the two dilutions, we present results for 1 in 5 dilutions only, because clotting times with 1 in 10 dilutions were long and 1 in 5 dilutions Vol 344 • December 17, 1994 Figure: APC-ratios in three groups of patients O=original assay results, A=modified assay results with dilution of 1 in 5. showed better reproducibility (data not shown). Molecular genetic investigation of the factor V gene was done as previously described2 with polymerase chain amplification from genomic DNA and digestion with MnlI of the 267 base-pair fragments. We investigated 3 groups of patients: 18 APC-sensitive (ie, normal), 25 APC-resistant according to the original test, and 21 patients on oral anticoagulants who could not be reliably classified by the first test. The figure shows the results. In the APC-sensitive group values with the original and modified assays did not differ significantly. These results allowed a cut-off of 2-0 (mean -2 SD) to be established for the modified test. All APC- resistant patients were abnormal with the modified test (ratio <2-0). In every patient in this group, the molecular study of factor V gene confirmed the presence of the mutation Arg506GIn. In this group and the group on oral anticoagulants, the ratios of the modified test were significantly lower than the original test (p<0-01 1 and p<0001, respectively). In the group on oral anticoagulants, 6 patients with a ratio below 2-0 were regarded as abnormal (mean ratio 1-53, SD 0-28): all were carriers of the factor V Leiden mutation (5 heterozygous and 1 homozygous). Our results confirm and extend those of Jorquera and colleagues, and show that reliable screening of APC-resistant patients is possible during oral anticoagulation with this modified test. Supported by the special trustees of Charing Cross and Westminster Hospital and Medical School. M Trossadrt, J Conard, M H Horellou, M M Samama, H Ireland, T A Bayston, D A Lane Service d’Hématologle Biologique, H6tel-Dieu, Paris 75181, France; and Department of Haematology, Charing Cross and Westminster Medical School, London, UK 1 Sun X, Evatt B, Griffin JH. Blood coagulation factor Va abnormality associated with resistance to activated protein C in venous thrombophilia. Blood 1994; 83: 3120-25. 2 Bertina RM, Koeleman BPC, Koster T, et al. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994; 369: 64-67. Resistance to activated protein C and risk of premature myocardial infarction SiR-Several workers have shown that resistance to activated protein C (APC), due to a single G—A mutation that converts arginine 506 of factor V to glutamine, is associated with a 5-10-fold increased risk of venous thrombosis.1-3 APC is a key component in a physiologically important anticoagulant system that cleaves and inactivates the