Postgrad Med J: first published as 10.1136/pgmj.41.479.563 on 1 September 1965. Downloaded from September, 1965 HEATH: Insulin Resistant Diabetes Mellitus 563 MICHON, P., LARCAN, A., HURLET, C., and VERT, P. Hemochromatose et Insulino4R6sistance, Diabe'te (1961): Hemochromatose et Insulino Resistance. i(Le Raincy), 6, 54. Efficacite d'un Traitment par A.C.T.H., Diabete VALLANCE OWEN, J. (1960): Insulin Antagonists, Brit. (Le Raincy), 9, 128. med. Bull., 16, 214. OAKLEY, W., FIELD, J. B., SOWTON, G. E., RIGBY, B., VALLANCE OWEN, J., DENNES, E., and CAMPBELL, P. N. and CUNLIFFE, A. C. (1959): Action of Prednisone (1958): Insulin Antagonism in Plasma of Diabetic in Insulin Resistant Diabetes, Brit. med. J., i, 1601. Patients and Normal Subjects, Lancet, ii, 336. WILLIAMS, R., SCHEUER, P. J., and SHERLOCK, S. OLSEN, N. S., and NEUTZEL, J. A. (1950): Resistance (1962): The Inheritance of Idiopathic Haemo- to Small Doses of Insulin in Various Clinical chromatosis, Quart. J. Med., 31 NS., 249. Conditions, J. clin. Invest., 29, 862. WOOD, F. C., and FITZHUGH, T. (1929): Haemochro- ROOT, H. F. (1929): Insulin Resistance and Bronze matosis in a Metal Worker. Case Report with Diabetes, New Engl. J. Med., 201, 201. Autopsy and a Brief Review of the Literature, SHELDON, J. H. (1935): 'Haemochromatosis', London: Arch. intern. Med., 44, 882. Oxford University Press. WRIGHT, P.IH. (1960): Insulin Antibodies, Brit. med. UHRY, P., MARCEL, G., and COHEN, A. (1958): Bull., 16, 219.

POTENTIATION OF ANTICOAGULANT THERAPY BY OXYPHENYLBUTAZONE

(A Probable Case) Protected by copyright. C. B. HOBBS, M.A., M.B. (Cantab), M.C. Path. A. L. MILLER, M.B. (Lond.), M.C. Path. J. H. THORNLEY, M.A., M.B. (Cantab.), D.R.C.O.G. The Courtauld Institute of Biochemistry, The Middlesex Hospital Medical School, W.].

THE DANGERS of giving aspirin or antibiotics to an anticoagulant action (Mandel, 1962) and could patients on anticoagulant therapy are widely therefore be expected to exert a synergistic effect recognised. Aspirin and other salicylates which on anticoagulant therapy. themselves produce a mild hypoprothromibinaemia Kindermann (1961) reported two cases of are also gastric irritants. Gastrointestinal bleed- spontaneous bleeding after combined therapy with ing may follow the use of these drugs. The phenindione and phenylbutazone (butazolidin) in possible causes of such bleeding were reviewed by the management of superficial thrombophlebitis; Watson and Pierson (1961). Antibiotic therapy and earlier Humble (1953) had reported anomalies http://pmj.bmj.com/ can affect anticoagulant control by interfering in blood clotting in patients on butazolidin. The with the source of Vitamin K from the intestinal purpose of this communication is to report a flora and thus indirectly affecting prothrombin patient on anticoagulant therapy who showed production in the liver. marked prolongation of the prothrombin time In addition to these there is a growing list of and spontaneous bleeding some two weeks after other drugs which potentiate anticoagulant beginning a course of the reliated compound therapy. Attention has recently been drawn to oxyphenylbutazone (Tanderil). the reduced anticoagulant requirements of certain on September 26, 2021 by guest. patients treated with Atromid, a preparation used Case Report to reduce serum triglyceride and cholesterol levels The patient, a printer, with known rheumatic heart (Oliver, Roberts, Hayes, Partridge, Suzman and disease and a history of angina for nine years, was Bersohn, 1963). A similar effect on anticoagulant aged 50 years when, in May 1960, he was admitted therapy was noted by Winters and Soloff (1962) to hospital with a posterior myocardial infarct. He in a patient in whom hypercholesterolaemia was was discharged from hospital to continue on long treated with D-thyroxine. term anticoagulant therapy with phenindione. Other drugs reported to potentiate anticoagulant A year later in May 1961 he developed multiple small spontaneous bruises and a small subconjunctival therapy were reviewed by Hellemans (1962) and haemorrhage, during the course of an upper res- include dinitrophenol, chlorpromazine, p-amino piratory tract infection with a productive cough. salicylic acid, testosterone, alcohol and probenecid. The Quick one-stage prothrombin time was 3.2 times Quinine hydrochloride, chloroquine and hydroxy- that of the control plasma (therapeutic range 2.0-2.5). chloroquine have also been reported as having Further slight spontaneous bruising occurred in May Postgrad Med J: first published as 10.1136/pgmj.41.479.563 on 1 September 1965. Downloaded from 564 POSTGRADUATE MEDICAL JOURNAL September, 1965 1962. The Thrombotest result at this time was 7.4%, the accepted therapeutic range being 10-15% (Miller, Farrer-Brown and Pether, 1964). No cause was found /00 THROMBOTEST PERCENTAGE for this escape from control; the dose of phenindione was slightly reduced and no further bruising occurred. WARFARIN 975 MGPER DAY In August 1962 the patient developed diarrhoea, 50 THROUGHOUT PERIOD with 5-6 loose but normal coloured motions each day. Stool culture was negative and the diarrhoea was attributed to phenindione therapy, of which it 30 is a well recognised complication. Accordingly, the 25 drug was discontinued and treatment with warfarin 20 sodium substituted, and within a short time the diarrhoea ceased. After this the patient continued on anticoagulant /0 therapy without further trouble until the end of April 1964, when he presented with a moderate 5 haemorrhagic effusion into the right olecranon bursa <5 and extensive bruising of the forearm extending from above the elbow to the wrist. The elbow and superior A M J J A S O N D J F M A M MONTHS radio-ulnar joints showed a full range of movements /963 and there was no evidence of hamarthrosis. There /964 was no history of trauma, and the patient was other- FIG. 1.-Record of Thrombotest results over the wise well. The Thrombotest result was now less the year than 5% of normal, the patient having been reason- preceding haemorrhagic episode. ably well controlled (see accompanying figure) on a constant dose of warfarin sodium, mean daily dose report of any interaction between Stemetil and 9.75 mg., throughout the preceding twelve months. anticoagulants in man or in experiments carried On further questioning it was found that a three out on there have been week course of 100 three animals, though reports oxyphenylbutazone, mg. that another Protected by copyright. times daily, had been prescribed elsewhere for treat- phenothiazine derivative, chlorproma- ment of some "rheumatic" pains of which the patient zine, may increase coagulation time in both had complained. The patient had been taking this animals and man. However, these studies are drug for eighteen days prior to the appearance of not conclusive and Hrdina and Kovalcik (1961) the spontaneous bruising. In addition to the oxy- found that chlorpromazine did not significantly phenylbutazone and warfarin sodium, he was taking affect the action of the anticoagulants ethylbis- digitalis and quinidine, both of which had been coumacetate and phenindione. It would be unwise exhibited since 1960, and also prochlorperazine in any case, to to an effect of a (Stemetil) prescribed for control of attacks of attempt predict since November 1963. vertigo particular phenothiazine derivative on blood clot- Investigations. Hb. 88% (12.8 g./100 ml.), WBC ting on the basis of experimental findings with 5,500/cu.mm., platelets 190,000/cu.mm. Liver another. function tests normal-bilirubin less than 0.4 mg./ On the other hand, phenylbutazone, to which 100 ml., thymol turbidity 2, zinc sulphate 6, alkaline Tanderil is closely related and of which it is a phosphatase 7.4 units/100 ml., total protein 6.7 g./ metabolite, is known to disturb the blood clotting 100 ml., paper electrophoresis normal, SGOT 20 mechanism and to produce haemorrhagic com- S.F. units/ml., SGPT 9 units/ml. Humble on 44 Both oxyphenylbutazone and warfarin were dis- plications. (1953) reported patients continued, and the right arm treated by simple on phenylbutazone of whom about one thirdhttp://pmj.bmj.com/ support in a sling. After 48 hours, the Thrombotest showed prolongation of the one stage prothrombin was still less than 5% of normal. The effusion into time which was sometimes alarming. As a result of the olecranon bursa was aspirated. After five days studies on these patients he suggested that the off anticoagulants the Thrombotest result was 33% clotting defect was due to a deficiency of pro- of normal and warfarin sodium was recommenced thrombin itself and supported this statement by cautiously. The bruising slowly subsided and the showing that while plasma from these patients arm returned to normal. could restore to normal the prothrom,bin time of plasmas deficient in factors V and it failed Discussion VII, on September 26, 2021 by guest. to do so It is when added to the plasma of patients clearly impossible to say with certainty that with liver disease who are known to have a the escape of this patient from control was of attributable to deficiency prothrombin. potentiation of anticoagulant This known action of the closely related com- therapy by oxyphenylbutazone. However, its oc- pound, together with the time relationship be- currence shortly after the exhibition of this drug, tween the exhibition of oxyphenylbutazone and the patient having been stable for many months the onset of escape from control and bleeding is highly suggestive. None of the other drugs lead us to conclude in this which the patient was have been that, case, the to taking reported oxyphenylbutazone potentiated anticoagulant produce this effect. The administration of therapy, presumably by a mechanism similar to digitalis to patients in failure will, in fact, often that increase the dose of produced by phenylbutazone; it must be anticoagulant required, pre- admitted, however, that Strobel (1963) found no sumably by reducing hepatic congestion and so alteration in one stage prothrombin time, fibri- improving liver function. There has been no nogen or platelet count after oxyphenylbutazone. Postgrad Med J: first published as 10.1136/pgmj.41.479.563 on 1 September 1965. Downloaded from September, 1965 HOBBS, MILLER and THORNLEY: Anticoagulant Therapy 565 It seemed to us important to draw attention to given by Pharmaceutical Specialities (May and the possible haemorrhagic hazards of combined Baker) Ltd. who searched the literature and their therapy with anticoagulants and phenylbutazone personal reports for information on the effect of or its derivatives. There are, of side Stemetil on blood coagulation. We are also grateful course, other to Evans Medical Ltd. and the Geigy Pharmaceutical effects of phenylbutazone which make such com- Co. Ltd. for information supplied; and to those who bined therapy a matter for caution. Mauer (1955) helped reviewed the literature on the toxicity of us with problems in translation. phenylbutazone and summarised reports from 26 papers covering a total of 3934 patients. In this REFERENCES series he found 415 patients with upper gastro- ANNOTATION. (1962): Phenylbutazone and Leukaemia. intestinal symptoms with 23 cases of gastro- Brit. med. J., i, 459. intestinal bleeding and 40 cases of peptic ulcer. COUNCIL ON DRUGS. (1963): New Drugs and Develop- Other toxic manifestations included 32 cases of ments in Therapeutics. J. Amer. med. Ass., 184, as 577. thrombocytopenia, as well diarrhoea, toxic HELLEMANS, J. (1962): Factoren die de Coumarine- hepatitis, rashes, blood dyscrasias, water reten- Werking Kunnen Beinvloeden. Belg. T. Geneesk., tion and cardiac decompensation. Oxyphenylbuta- 18, 361. zone is generally accepted as having the same HRDINA, P., KOVALCIK, Y. (1961): K Otazke adverse reactions as the parent compound, Antagonismu Niektorych Centralnych Depresancii though possibly giving rise to fewer gastro- A Nepriamych Antikoagulancii, Cs. Fysiol, 10, 335. intestinal disturbances. (Council on Drugs 1963). HUMBLE, J. G. (1953): Empire Rheumatism Council An annotation in the British Medical Journal Symposium on Butazolidin. Anomalies of Blood (1962) estimated that some 100,000 persons were Clotting in Patients on Butazolidin. KINDERMANN, A. (1961): Vaskulares Allergid nach probably receiving phenylbutazone each week. Butalidon und Gefahren Kombinierter Anwendung The widespread use of this drug and related com- mit Athrombon (Phenylindandion) Derm. Wschr., pounds makes it important that the hazards of 143, 172. such should be It would therapy widely known. MANDEL, E. H. (1962): The Anticoagulant Proper- Protected by copyright. seem wise for reasons discussed above to avoid ties of Chloroquine Dihydrochloride, Hydroxy- using these drugs in the treatment of patients who chloroquine Sulphate and Quinine Dihydrochloride. are already receiving anticoagulants. If, how- J. Mt. Sinai Hosp., 29, 71. ever, in a particular patient, this combination of MAUER, E. F. (1955): The Toxic Effects of Phenylbutazone (Butazolidin). New Engl. J. Med., drugs seems necessary, extremely careful control 253, 404. of anticoagulant therapy is advisable. MILLER, A. L. FARRER-BROWN, G., PETHER, J. (1964): Initial Experience with Thrombotest. J. clin. Path., Summary 17, 182. Attention is drawn to the growing list of OLIVER, M. F., ROBERTS, S. D., HAYES, D., PANT- drugs reported to potentiate anticoagulant therapy. RIDGE, J. F., SUZMAN, M. M., BERSOHN, I. (1963): A case is reported of a patient, well controlled Effect of Atromid and Ethyl Chlorophenoxyisobuty- rate on anticoagulant requirements. Lancet, i, 143. for many months on a constant dose of warfarin STROBEL, E. (1963): Die Therapie Superfizieller sodium, in whom a dramatic fall in prothrombin Thrombophlebitiden mit p-Hydroxyphenyl-butazon time followed the administration of Tanderil. und sein Einfluss auf das Bluttgerinnungssystem. Great caution is advised in undertaking com- Arznemittel-Forsch, 10, 497. bined therapy with this drug, or its parent com- WATSON, R. M., PIERSON, R. N. (1961): Effect of http://pmj.bmj.com/ pound phenylbutazone, and anticoagulants. Anticoagulant Therapy upon Aspirin-Induced Gastrointestinal Bleeding. Circulation, 24, 613. WINTERS, W. L., SOLOFF, L. A. (1962): Observations We wish to thank Professor Sir Charles Dodds on Sodium D-thyroxine as a Hypocholesteremic for his interest and Dr. J. D. N. Nabarro for his Agent in Persons with Hypercholesteremia with permission to publish clinical details of this patient. and without Ischaemic Heart Disease. Amer. J. Also we acknowledge wi,th thanks the great help med. Sci., 243, 458. on September 26, 2021 by guest.