Drug PointsFatal hepatic necrosis associated with trazodone and neuroleptic drugsRhabdomyolysis induced by thioridazineBMJ 1994; 309 doi: https://doi.org/10.1136/bmj.309.6951.378 (Published 06 August 1994) Cite this as: BMJ 1994;309:378
Skin necrosis induced by streptokinase
Drs J Penswick and A L Wright (Bradford Royal Infirmary, Bradford BD9 6RJ) write:
We describe a case of skin necrosis induced by treatment with streptokinase. There have been several previous reports of similar cases associated with heparin and warfarin; more recently there has been a report of a purpuric eruption in three patients who received streptokinase.1 To our knowledge, there are no similar reports of such extensive skin necrosis.
A 70 year old man with a history of angina, hypertension, and peripheral vascular disease was admitted with chest pain. Changes in his electrocardigram suggested an inferior myocardial infarction, and he received 1.5 million units intravenous streptokinase. Twenty four hours later he complained of diffuse tenderness across the upper thighs. Nursing staff noticed areas of bruising on his lower trunk. Over the next 72 hours these areas spread to the upper parts of his buttocks and thighs. He remained haemodynamically stable, his haemoglobin concentration not changing and platelet concentrations and results of a clotting screen remaining normal. After four days no further bruising had developed and he was discharged.
Nine days later large areas of ulceration had developed in the distribution of the original bruises (figure). All haematological investigations including protein C and protein S concentrations were normal. A skin biopsy specimen showed fibrin deposition in dermal blood vessels with a patchy perivascular inflammatory infiltrate. The ulcers were debrided and dressed and have since healed with minimal scarring.
Skin necrosis induced by warfarin treatment has been clearly associated with deficiency of both protein C and protein S. The rapid fall in protein C concentrations on starting warfarin treatment is thought to produce a hypercoaguable state leading to thrombosis.2 Skin necrosis induced by heparin has been linked to a heparin induced platelet antibody that promotes platelet aggregation and the formation of thrombi.3 Information on adverse reactions to all the thrombolytic enzymes details a total of 40 peripheral vascular problems including cases of arterial embolisation, thrombophlebitis, and vasculitis.4 There are 28 reports of purpura and bruising. None of these apparently noted large areas of cutaneous necrosis, and there was no evidence for a vasculitis or embolisation in the skin biopsy specimen from our patient. We suggest that skin necrosis is the end result of a variety of disturbances of the clotting and thrombolytic cascade.
Fatal hepatic necrosis associated with trazodone and neuroleptic drugs
Drs M Hull, R Jones, and M Bendall (University Hospital, Nottingham NG7 2UH) write : We report a case of fatal hepatic necrosis associated with the antidepressant drug trazodone and neuroleptic drugs.
A 72 year old woman was admitted for treatment of acute psychotic depression. She was taking atenolol (100 mg once a day). Trifluoperazine (2 mg twice a day), trazodone (200 mg at night), and lithium carbonate (400 mg once a day) were added. On admission there was no evidence of previous liver disease, and the results of liver function tests were normal.
Ten weeks later her serum alanine aminotransferase concentration had risen to 107 U/I (normal range <50 U/I). Eleven weeks after admission trifluoperazine was replaced by thioridazine (25 mg three times a day). Nine weeks later she became jaundiced. Drug induced hepatitis was diagnosed, and all drug treatment was stopped. She developed hepatic encephalopathy and died 54 days after the onset of jaundice, despite active management. A postmortem liver biopsy specimen showed acute hepatic necrosis with cholestasis and a moderate chronic inflammatory infiltrate, mainly in portal tracts.
We believe that trazodone and concurrent neuroleptic treatment were responsible for the acute hepatic necrosis. However, the possibility that phenothiazine treatment was solely responsible cannot be excluded, especially in view of the known cross sensitivity between phenothiazines. Atenolol and lithium carbonate have not been associated with liver failure. Tri-fluoperazine and thioridazine cause cholestasis and jaundice but have not been implicated in an episode of hepatic necrosis.
Trazodone may cause a mixed hepatocellular-cholestatic reaction that is reversible.1,2 Trazodone has been associated with adverse effects on the liver in 14 reports to the Committee on Safety of Medicines up to August 1991. One of these was an episode of fatal hepatic necrosis (Committee on Safety of Medicines, personal communication). The datasheet for trazodone makes no mention of possible adverse hepatic effects. Doctors should be aware that liver failure may occur in association with the commonly used combination of trazodone with neuroleptic drugs.
Rhabdomyolysis induced by thioridazine
Drs B J Nankivell, and P K Bhandari, and Ms L J Koller (Westmead Hospital, Westmead NSW 2145, Australia) write: We report a case of non- traumatic rhabdomyolysis following an overdose of thioridazine.
A 22 year old man had been taking thioridazine (100 mg a night) for one month for schizophreniform psychosis. Twenty four hours after taking an overdose of 9.4 g thioridazine he presented with difficulty in walking, moving his arms, and speaking. He had not consumed alcohol or taken illegal drugs, and he had no history of trauma, coma, immobilisation, or other illnesses. On examination he had a tachycardia but no fever, and he had symmetrical swelling and tenderness over his upper arms, thighs, and calves. He had ataxia and a transient dysarthria, which was attributed to generalised muscle weakness. He also had mild intermittent confusion and a dry mouth, consistent with thioridazine's anticholinergic activity. Muscle tone and tendon reflexes were normal. He had no bruises or features of the neuroleptic malignant syndrome such as high fever, autonomic instability, muscle rigidity, or dystonia. Serum creatine kinase concentration was 32 620 IU/I (normal range 24-204 IU/I), serum asparatate aminotransferase concentration 532 IU/I (10-47 IU/I), and serum creatinine concentration 167 mumol/l (60-125 mumol/l). His urine contained myoglobin. Screening his urine for drugs showed phenothiazine metabolites but no other agents capable of inducing rhabdomyolysis such as cocaine or phencyclidine (“angel dust”). His electrocardiogram showed sinus tachycardia, a prolonged QT interval, and an intraventricular conduction defect, consistent with thioridazine poisoning. He was treated with a stomach washout and activated charcoal. He was rehydrated with intravenous fluids and given mannitol and sodium bicarbonate. Serum biochemistry returned to normal over one week, and the muscle tenderness and weakness disappeared.
This case illustrates rhabdomyolysis induced by thioridazine in the absence of other drugs or toxins, the neuroleptic malignant syndrome, or muscle damage secondary to immobilisation or overactivity. The possibility of a direct effect of the drug is supported by the generalised tenderness in all major muscle compartments. A raised serum creatine kinase concentration (1440 IU/I) has been previously reported in an agitated man receiving a therapeutic dose of thioridazine.1 Rhabdomyolysis has also occurred in overdoses of other phenothiazines.2,3