This considerably reduced requirement for warfarin prompted a retrospective review of medical records at our hospital for the period 1981-6 to examine whether there might be an interaction between tamoxifen and warfarin. The records of women with breast cancer and a subsequent admission for a serious thromboembolism were examined. There were 18 such women: seven who had had deep venous thrombosis and 11 who had had a pulmonary embolus with probable deep venous thrombosis. Five of them were taking tamoxifen when they started taking warfarin. Two of them had complications after loading doses of warfarin: in one the prothrombin time after three daily doses of warfarin (10 mg, 10 mg, and 5 mg) rose to 50 seconds, and she developed a left subdural haematoma requiring warfarin to be withdrawn; in the other a similar loading dose regimen resulted in a prothrombin time of 49 seconds on day 7, the patient developed severe haematuria, and her anticoagulant treatment was changed to phenindione. The three other patients taking tamoxifen required daily doses of 2 mg, 2 mg, and 3 mg to maintain appropriate prothrombin times.The dose of warfarin that maintained appropriate prothrombin times in the 13 patients not taking tamoxifen varied from 4 to 10 mg (mean 6 25 mg) daily. None of these patients had complications in the first month of treatment with warfarin. CommentOur observations suggest that women with breast cancer requiring warfarin need a lower dose if they are taking tamoxifen. The mechanism of the interaction between the two drugs is unclear, but protein binding and competition for metabolic pathways may both play a part. Both warfarin and tamoxifen are metabolised by the microsomal enzyme systems in the liver. Warfarin is a racemic mixture and the (S) isomer is four to five times more physiologically active than the (R) isomer. This more active (S) isomer is converted to the 7-hydroxy metabolite, which is inactive, by the cytochrome P450 enzyme system.2 Three metabolites have been detected in the serum ofpatients taking tamoxifen. Normally tamoxifen accounts for 36% of the drug and metabolites present, desmethyltamoxifen for 58%, metabolite Y for 4%, and 4-hydroxytamoxifen for 1-5%.3 All of these metabolites have some affinity for the oestrogen receptor, but the affinity of the 4-hydroxy metabolite is 50-100 times greater than that of the parent drug.4 Accordingly, 4-hydroxytamoxifen is about 100 times more potent than tamoxifen in inhibiting the growth of MCF7 breast cancer cells in culture.5The hydroxylations are probably carried out by similar enzyme systems. In the case of warfarin competition for the enzymes may increase the concentration of the active parent drug and decrease the concentration of inactive metabolites. In the case of tamoxifen, however, the resultant changes in the amounts of metabolites present may have important implications for the activity of the drug. Altering the percentage of the 4-hydroxy metabolite present may have an effect on the response of the tumour.The hazards of an increased pha...

SummaryA patient is presented where routine venepuncture associated with anaesthetic induction resulted in bradycardia and asystole. The case highlights the need for special caution with, and ECG monitoring throughout induction for, patients with a history of syncope. It also demonstrates the need for caution when attributing cardiovascular events during induction to the effect of the induction agents used. Key words Anaesthesia; intravenous.Complications; asystole.Arrythmias associated with the effect of anaesthetics or other medication on the cardiac conducting tissues are often reported. It is unusual to find a patient where the mildly painful procedures associated with an intravenous induction result in a profound bradycardia and cardiac arrest. This report describes such a patient. Case HistoryA 20-year-old male patient was scheduled for a routine follow-up cardiac catheterisation. He had had a total correction of a Fallot's tetralogy in 1976, and the operation had achieved a good result, with slight residual pulmonary stenosis. His general health and exercise tolerance were good, and he had been receiving no medication since this operation. His pre-procedure ECG showed a high-normal P-R interval (0.2 second) with no other abnormality.The procedure was commenced in the usual manner. ECG monitoring was applied and lignocaine 1 % was infiltrated subcutaneously over the femoral area. The patient had a short syncopal episode when the introducing needle was inserted into this area, with a heart rate of 20/minute. This spontaneously recovered in 20 seconds and the procedure was continued. A similar sinus bradycardia with a heart rate of 20/minute occurred when the venous sheath was inserted, with recovery taking 40 seconds. It was decided to postpone the catheterisation and do it later under general anaesthesia. The patient had not complained of any painful sensation during this attempted procedure, and a postoperative ECG showed no change from that before the procedure.The patient was premedicated with oral diazepam 15 mg before the second catheterisation attempt and appeared adequately sedated. However, when a 22-gauge cannula was inserted into the dorsum of his hand he developed bradycardia. This rapidly progressed to asystole, which was associated with a tonic-clonic convulsion. Oxygen 100% was given and cardiac massage was performed. A bolus dose of intravenous atropine 1 mg resulted in a sinus rate of 80/minute. The procedure was again abandoned. A subsequent detailed family history revealed that two of the patient's immediate family, his mother and twin brother, had experienced similar syncopal attacks under stressful circumstances, but had not had a documented cardiac arrest, nor any history of hospital treatment. A repeat ECG was identical to the pre-arrest one. A 24-hour Holter recording showed occasional multifocal ventricular ectopics with a short period of nocturnal Wenckebach A-V heart block with a maximal pause of 2.9 seconds. Such changes alone would not explain these attacks and a neurogenic basi...