Flucloxacillin associated cholestatic hepatitis. An Australian and Swedish epidemic?

The clinico-pathological entity of flucloxacillin-associated cholestatic hepatitis is described and the recognition and documentation of cholestasis associated with flucloxacillin and with related isoxazolyl-penicillins (cloxacillin, dicloxacillin) is examined on an international basis, with particular reference to Australia. Data were obtained from the literature, from the Australian adverse drug reaction monitoring agency and from the Collaborative Centre for International Drug Monitoring (World Health Organisation) in Sweden. Approximately 600 cases of flucloxacillin-associated cholestatic hepatitis were collected, as well as 164 cases associated with other isoxazolyl penicillins. Jaundice and pruritus may first appear several weeks after administration of the drug has ceased and typically are severe and protracted. Liver tests may be abnormal for months after symptomatic recovery. Death is uncommon. Liver pathology shows centrizonal bile stasis with portal tract inflammation and variable loss of bile ducts. Approximately 1 in 15,000 users of flucloxacillin will develop the reaction. Increasing age (> 55 years) and prolonged intake (> 14 days) are particular risk factors. Cholestasis associated with cloxacillin/dicloxacillin appears to be similar in nature but is less well defined. Recognition and reporting of the reaction have been uncommon in the United Kingdom inter alia and high in Sweden and Australia, although estimates of risk have been similar. In Australia, the remarkably high rate of reports appears to be the result of sustained publicity for the reaction. There is only a trickle of reports of cholestatic hepatitis in association with the use of cloxacillin and dicloxacillin from the USA and Canada. The high level of awareness of the reaction and consequential regulatory action so far have not resulted in a diminution of its occurrence in Australia.

Post-marketing surveillance of drugs. The spontaneous reporting scheme: role of the Adverse Drug Reactions Advisory Committee.

Post-marketing surveillance of drugs in Australia operates predominantly through the spontaneous reporting of suspected adverse drug reactions (ADRs). Approximately 50% of reports are submitted by hospitals and the rest by individual doctors, pharmacists and dentists. Some 4500 reports ("blue cards") are now reviewed annually by the Adverse Drug Reactions Advisory Committee (ADRAC) and its Secretariat. The register of ADRs has now accumulated more than 65,000 reports. Collations and analyses of data derived from the review process are published to increase awareness by health professionals of drug associated morbidity. Continued educational efforts by professional bodies and regulatory agencies will play a key role in rationalising drug use and reducing drug induced disease.

Ascorbic acid deficiency affects the metabolism of cytosolic ferritin but not lipid-associated ferritin in livers of guinea pigs.

Two physicochemically and metabolically separate pools of ferritin, namely cytosolic ferritin and lipid-associated ferritin, are present in the livers of guinea pigs. In this paper we establish that the iron content of cytosolic ferritin is dependent on and linearly related to ascorbic acid concentration, whereas changes in concentration of this vitamin do not affect the iron content of lipid-associated ferritin. In livers of ascorbic acid-deficient guinea pigs both synthesis and degradation of cytosolic ferritin are diminished equally. Consequently cytosolic ferritin is metabolized more slowly without changes in its pool size. In contrast with cytosolic ferritin, the metabolism of lipid-associated ferritin is unaffected by ascorbic acid deficiency. The differential effects of ascorbic acid deficiency on the physicochemical characteristics as well as on the metabolism of cytosolic ferritin and lipid-associated ferritin suggest that the two forms of ferritin have different functional roles.

Simultaneous liquid-chromatographic determination of vitamin K1 and vitamin E in serum.

We describe a high-performance liquid chromatographic procedure for the simultaneous measurement of vitamins K1 and E in human serum. Delipidated human serum (free of vitamins K1 and E) was used to make standard solutions of these vitamins, and cetyl naphthoate and alpha-tocopheryl acetate were the internal standards for vitamin K1 and vitamin E, respectively. A simple, novel separation method utilizing liquid-liquid partition chromatography was used as a preparative "clean-up" procedure. Cetyl naphthoate and vitamin K1 (after post-column reduction) were detected by fluorescence, alpha-tocopheryl acetate and vitamin E by ultraviolet absorption. Sensitivity (detection limit) of the assay was 30 pg for vitamin K1 and 5 ng for vitamin E per injection. The method is specific, precise, and more rapid than previously described procedures. Within- and between-assay CVs were 8.1% and 12.9%, respectively, for vitamin K1; 3.5% and 6.0%, respectively, for vitamin E. Analytical recoveries of vitamins K1 and E were 80% and 93%, respectively, from serum and from delipidated serum (standards). The average neonatal serum concentration of vitamin K1 was 83 ng/L, 2.5 mg/L for vitamin E; for normolipidemic adults, the values were 343 ng/L and 7.9 mg/L, respectively, and for hyperlipidemic adults, 541 ng/L and 11.1 mg/L, respectively.

Cytosolic ferritin and lipid-associated ferritin are metabolically different in guinea-pig livers.

A distinct pool of liver ferritin has been described in man, guinea pigs and rats [Cham, Roeser, Nikles & Ridgway (1986) Clin. Chim. Acta 158, 71-79]. This ferritin accounts for approx. 30% of total intracellular ferritin. It differs from previously described cytosolic and 'microsomal-fraction' ferritin by its firm association with lipid and by the absence of heat-stability at 75 degrees C. The present study demonstrates that cytosolic ferritin and lipid-associated ferritin in guinea-pig livers have distinctly different rates of turnover. Cytosolic ferritin has a rate of turnover approx. 3.5 times as high as lipid-associated ferritin. The apparent metabolic heterogeneity suggests that the two forms of ferritin may have different functional roles.

Hypothyroidism and Graves' disease after mantle irradiation: a follow up study.

Fifty of 66 patients whose thyroid function had previously been assessed 7-139 months after irradiation for Hodgkin's disease were re-evaluated 35 +/- 3 months later. They could be divided into three groups: those whose thyroid function had been normal in the first study (N = 26), those who had had asymptomatic impaired thyroid reserve (N = 19), and those in whom evidence of Graves' disease had developed (N = 5). The 26 patients who had been euthyroid when first studied had developed significant increases in mean thyroid-stimulating hormone (TSH) levels (basal and following thyrotrophin releasing hormone) without changes in mean free thyroxine index (FTI). In three of these patients, each studied within six years of irradiation, basal TSH had risen to hypothyroid levels. There were no significant changes in mean FTI or basal and peak TSH in 19 patients who had demonstrated impaired thyroid reserve in the first study. The cumulative incidence of impaired thyroid reserve in the total cohort is now 30/66 (45%) but only one of these 30 has developed clinical hypothyroidism. Five patients developed evidence of Graves' disease. Two patients with thyrotoxicosis and one with euthyroid Graves' disease were found in the initial study. On re-evaluation, a third patient had developed frank thyrotoxicosis and another euthyroid Graves' disease, giving a cumulative incidence of Graves' disease of 5/66 (7%). Three of these five were HLA-DR3 and three had measurable thyrotrophin binding inhibiting immunoglobulins. We conclude that impaired thyroid reserve continues to develop within six years of mantle irradiation in adults but once established appears to remain stable.(ABSTRACT TRUNCATED AT 250 WORDS)

A procedure for the purification of ferritin from human liver by heating a methanol-treated homogenate.

A simple, rapid technique for purification of ferritin from human liver tissue is described. Methanol, at a final concentration of 40% (v/v) in liver homogenate, precipitates the majority of proteins but does not affect ferritin. Subsequent heating of this homogenate at 75 degrees C for 10 min results in a purified ferritin preparation as judged by immunoelectrophoresis and polyacrylamide gel electrophoresis. The resultant purified ferritin contained the same amount of iron as the original endogenous ferritin. There were no significant differences (paired t tests) in the amount of protein in the purified ferritin preparation when measured by rocket immunoelectrophoresis and by the Lowry procedure, suggesting that the antigenecity of ferritin was unaffected by the methanol and heat treatment. Both endogenous liver ferritin and radiolabeled human liver ferritin added to liver homogenates were recovered after methanol and heat treatment with similar yields (77 +/- 7% and 70 +/- 2%, respectively) when compared with the standard treatment of heating a homogenate at 75 degrees C. The overall ferritin yield with this rapid procedure was 40%.

Quantitation of reticuloendothelial iron kinetics in humans.

Reticuloendothelial iron kinetics were investigated in a simultaneous dual-isotope study in 10 healthy adult subjects in whom 55Fe-ferric hydroxide phosphate colloid was used to label the reticuloendothelial iron pools, and 59Fe-transferrin was used to define plasma iron kinetics. The simultaneous clearance of 55Fe and 59Fe from plasma and the uptake of each into red blood cells were measured over 14 days. The 55Fe-colloid was cleared almost immediately, and its iron was rapidly released to bind to plasma transferrin. Red cell incorporation of 55Fe was, however, much slower than that of 59Fe bound to transferrin in vitro. The data were analyzed by a new model of reticuloendothelial iron metabolism that contained two reticuloendothelial iron pools; one had a rapid turnover and donated iron to transferrin, and the other, a storage pool, had a slower turnover. The transit pool contained a mean of 164 mumol iron with little variation between subjects, whereas the storage pool was somewhat larger (mean 873 mumol iron) and showed more marked variation between subjects. In general an equal proportion of the iron leaving the transit pool went to transferrin and to the storage pool. The distribution between the two routes did not appear to be related either to plasma iron concentration, latent iron-binding capacity, or transferrin saturation.

Iron overload associated with congenital pyruvate kinase deficiency and high dose ascorbic acid ingestion.

Iron overload is an uncommon complication of untransfused chronic hemolytic anemias. This paper describes only the fourth reported case of erythrocyte pyruvate kinase deficiency and iron overload. Important aspects of this case are the presence of the HLA genotype commonly associated with genetic (idiopathic) hemochromatosis and a history of ingestion of large doses of ascorbic acid. Their roles in the development of iron loading and toxicity are discussed. A beneficial response to treatment with desferrioxamine was observed before significant iron removal. A mechanism for this action of desferrioxamine is proposed.

Tissue ferritin in scorbutic guinea-pigs.

Tissue ferritin metabolism was compared in control and ascorbic acid (AA) deficient guinea-pigs. Concentrations of ferritin protein in the liver (0.98 +/- 0.61 mg/g wet weight) and spleen (0.48 +/- 0.23 mg/g) of control animals did not change after tissue depletion of AA. Iron dextran (75 mg/kg weight, i.m.) caused a 4-5-fold increase in tissue ferritin concentrations in controls whereas no increase in tissue ferritin occurred in scorbutic animals. The rise in tissue total iron concentration was similar in the two groups. Liver ferritin synthesis was similar in control and scorbutic animals. After stimulation with iron (8 mg/kg iron dextran i.v.), ferritin synthesis rose in both groups of animals. However, the pattern of response differed. At 24 h after iron dextran, ferritin synthesis in controls was still significantly elevated (P less than 0.001) and liver ferritin protein continued to rise, whereas in scorbutic animals, ferritin synthesis had declined to pre-iron injection levels, and no rise in ferritin protein values occurred. It is concluded that the ferritin synthetic apparatus in AA deficient tissues remained intact and capable of responding to added iron. The absence of a sustained elevation in tissue ferritin protein after an iron load appeared to be due to inadequate stimulation of ferritin synthesis by intracellular iron. It is suggested that AA has a physiological role in the reduction of intracellular iron and that it is the reduced form of iron which stimulates ferritin synthesis. Abnormalities of iron metabolism occur in AA depleted tissues when the quantity of Fe3+ entering cells exceeds the residual reducing capacity of those cells.

Effect of a high energy, low carbohydrate diet on serum levels of lipids and lipoproteins.

The effect of a high energy, low carbohydrate (ketogenic) diet on serum levels of lipids and lipoproteins was studied in two normal individuals. A marked rise in serum levels of cholesterol and phospholipid occurred during the diet, but serum triglyceride levels remained within the normal range. High density lipoprotein cholesterol level did not change, and the rise in serum level of cholesterol levels with its distribution in lipoproteins, these observed changes pose a potential atherogenic risk. Serum lipid and lipoprotein levels reverted to normal on discontinuation of the diet. A modest weight loss observed during the diet was regained within 24 hours of the diet's cessation.

A comparison of the antifertility effects of alkylating agents and vinca alkaloids in male rats.

1 The anti-fertility effects of cyclophosphamide, nitrogen mustard, vincristine and vinblastine were studied and compared in male rats. 2 The effects of the drugs on body weight and haematological values were used to monitor the pharmacological actions of the drugs. 3 All four drugs impaired fertility, the severity of the impairment depending on dose and duration of treatment. 4 Testicular size and histological appearances remained mostly normal, even in infertile animals, but seminiferous tubules were fewer in number and maturation arrest at the spermatid level was evident in some sections. 5 Recovery of drug-induced infertility occurred in 64% of treated animals, 9 to 40 weeks after cessation of treatment. 6 Morbidity and mortality were much higher with alkylating agents than with vinca alkaloids for approximately similar degrees of impairment in fertility.

Testicular damage due to cytotoxic drugs and recovery after cessation of therapy.

Testicular function was assessed in 32 patients who received standard chemotherapy regimens for disseminated lymphomas. Thirty-one had evidence of germ cell damage, as assessed by the finding of azoospermia and/or high plasma levels of follicle-stimulating hormone (FSH). In addition, five patients had persistently low plasma testosterone levels associated in three with elevated plasma luteinising hormone (LH) levels. Reproductive function did not recover in any patient while chemotherapy continued. Cessation of therapy was possible in 16 patients with prolonged remissions of disease. Among these, recovery of germinal epithelium differed greatly between the cyclophosphamide / vincristine / prednisone treated group and the mustine/procarbazine/vincristine/prednisone treated group. Seventy per cent of the former patients had evidence of recovery after 34 months of follow-up while only one (17%) of the latter had begun to recover at 52 months post-therapy. Serial measurement of plasma FSH levels proved useful in predicting likely recovery of spermatogenesis.