While fierce arguments have raged round the place of anticoagulants in the management of ischaemic heart disease, evidence has been brought forward that, such treatment having been started, cessation may be hazardous (B.M.7., 1962). A number of reviews suggesting that the stopping of anticoagulant therapy may be followed by an increased thrombotic tendency have been published (Dinon and Vander Veer, 1960;Kuhn et al., 1961 ;Sise et al., 1961) and a possible theoretical basis has been suggested by Poller and Thomson (1964). Dinon and Vander Veer (1960) reported a series of 20 patients who sustained a myocardial infarction (the first infarction in 18 cases) and reinfarcted within one month of stopping anticoagulant therapy. Anticoagulants had been given for an average period of 40 days (range 24 to 82 days) and in most cases, therefore, the new infarction occurred at a time when the chances of such a relapse were still high. Kuhn et al. (1961), in a study of long-term anticoagulants in patients who had suffered myocardial infarction, noted a reinfarction rate of 5 % in the month following cessation of anticoagulants-this was higher than in any other month. However, the numbers involved-five reinfarctions (four of which were fatal)-were rather small and could be regarded only as suggestive.Sise et al. (1961), in a study of a large series of patients who had stopped anticoagulants, noted that among 45 instances where anticoagulants were stopped because of bleeding (and the patient was given vitamin K or a blood transfusion) there were 20 " relapses " (10 fatal and 10 non-fatal) in the following year, the largest proportion occurring in the first month. By contrast, in 53 instances when anticoagulants were given up because of difficulties in cooperating, because of the necessity for operation and for other non-urgent reasons, there were 15 " relapses " (seven fatal and eight non-fatal) in the subsequent year, and these appeared to be fairly evenly spread.The difference in the mortality rate in the first month was at the 5% level of significance, while if all relapses were considered the difference was at the 1 % level.These observations show that stopping of anticoagulants is associated with an increased thrombotic tendency only when cessation is because of haemorrhage.
ObservationsThe present data were obtained during the course of a study of the effects of a low-fat diet compared with an ordinary diet on the progress of ischaemic heart disease. A consecutive series of men were used for the trial. They were all under the age of 65, had sustained one episode of cardiac infarction, and were subsequently free from serious complications such as persistent heart failure. Cases where the cardiac infarction was thought to be a complication of an existing disease-for example, diabetes or severe hypertension-were likewise excluded. Many of the men received up to 14 weeks' anticoagulant treatment but none were on long-term anticoagulants.