Ralf Stahlmann

Toxicity of Quinolones

Reactions of the gastrointestinal tract, the CNS and the skin are the most often observed adverse effects during therapy with fluoroquinolones. At least for some of the newer fluoroquinolones a steep dose-response relationship of adverse effects seems to exist. Pathogenesis of the neurotoxic effects of fluoroquinolones is still unknown. Among the newer drugs, trovafloxacin caused mild CNS reactions such as dizziness and lightheadedness in a considerable proportion of patients. Young females seem to be especially sensitive to this effect, which diminishes during treatment or if taken together with food. Cardiotoxic potentials of sparfloxacin and grepafloxacin are higher than those of other fluoroquinolones, but during therapy no clearcut drug-related serious reactions have been reported, apart from a slight prolongation of the QT interval. However, to avoid risks these drugs should not be prescribed to patients with known prolongation of the QT interval (e.g. patients on antiarrhythmics).Phototoxicity has been described for all quinolones, but derivatives with a halogen atom at position 8 show the highest potential for such reactions. Fleroxacin, sparfloxacin, clinafloxacin and lomefloxacin belong to this group of fluoroquinolones. The phototoxic potential of the other new fluoroquinolones is considerably lower, but extensive exposure to UV light should generally be avoided during therapy with all quinolones.Chondrotoxicity of quinolones, as observed in immature animals, can affect articular cartilage and/or the epiphyseal growth plate, depending on the developmental stage. Pathogenesis of chondrotoxicity can probably be explained by the magnesium-chelating properties of these drugs. As juveniles are especially sensitive, use of these drugs in paediatrics should be restricted to carefully selected indications (such as the use of ciprofloxacin in cystic fibrosis).Another manifestation of the toxic effects of quinolones on connective tissue structures are tendopathies. Tendinitis and tendon ruptures have occurred as late as several months after quinolone treatment.Overall, quinolones are well tolerated drugs. Their specific toxic potentials have to be considered when they are chosen for treatment of bacterial infections.

Clinical toxicological aspects of fluoroquinolones

Reactions of the gastrointestinal tract and the central nervous system are the most often observed adverse effects during therapy with fluoroquinolones. Pathogenesis of the neurotoxic effects of fluoroquinolones could be related to the activation of the NMDA receptor. Animal experiments as well as clinical experience show that the cardiotoxic potentials of sparfloxacin and grepafloxacin are higher than those of the other fluoroquinolones: they cause QT prolongation at rather low doses thus increasing the risk for severe arrhythmia (torsades de pointes). Phototoxicity has been described for all quinolones, but derivatives with a halogen atom at position 8 show the highest potential for such reactions (e.g. clinafloxacin). Chondrotoxicity of quinolones can affect the articular cartilage and the epiphyseal growth plate in immature animals; the use of these drugs in pediatrics should be restricted to carefully selected indications (such as the use of ciprofloxacin in cystic fibrosis). Tendinitis and tendon ruptures can also be induced by quinolones. Overall, quinolones are as well tolerated as most other anti-microbial agents. However, their specific toxic potentials have to be considered when they are chosen for treatment of bacterial infections.

Ofloxacin in juvenile non-human primates and rats. Arthropathia and drug plasma concentrations

Arthropathia in juvenile animals is the most important toxic effect induced by quinolones. We conducted pharmacokinetic and morphological studies with ofloxacin on non-human primates (Callithrix jacchus, Marmosets) and rats. In the marmoset, electron microscopy and the application of immuno-morphological methods proved to be suitable for the detection of specific alterations in cartilage (e. g. loss of proteoglycanes and altered chondrocytes). Subsequently performed electron microscopic examinations in rats showed similar specific alterations of the femur cartilage surface after multiple oral applications of 600 mg ofloxacin/kg body wt. These results were correlated with pharmacokinetic data obtained for the same species. After single oral application of 100, 300 or 600 mg ofloxacin/kg body wt to 5 week-old rats peak plasma levels were achieved 15–45 min after administration indicating a rapid absorption of the drug. The following peak concentrations were measured for the three doses applied (mean±SD): 8.9±2.1, 22.6±7.5 mg/l and 33.5±9.8 mg/l, respectively. After 360 min the concentrations were 1.1±0.4, 5.9±2.5 and 15.9±5.1 mg/l, respectively. After subcutaneous injection of 100 mg ofloxacin/kg body wt the mean peak concentration was 27.7±2.6 mg/l after 45 min (0.5±0.2 mg/l after 360 min). In the marmoset higher plasma concentrations were measured with comparable doses. One, 3, and 6 h after the last of nine administrations of 200 mg ofloxacin/kg body wt, the mean (±SD) plasma concentrations were: 42.7±16.7, 40.6±9.5, and 26.5±3.6 mg ofloxacin/l plasma. Typical alterations of the joint cartilage of juvenile rats (e. g. opened chondrocyte cavities, swelling of rough endoplasmic reticulum and mitochondrial swelling in the chondrocytes) were induced by oral administration of ofloxacin at doses that were approximately 100 times higher than therapeutic ones, but led to peak plasma concentrations which were only approximately 10 times above the therapeutic level. Since we found corresponding cartilage alterations in marmosets and rats, this species provides a convenient model for additional studies on chondrotoxic effects of quinolones.

Classification terms in developmental toxicology: need for harmonisation. Report of the Second Workshop on the Terminology in Developmental Toxicology Berlin, 27-28 August 1998.

Abstract The existence of ambiguities and inconsistencies in the use of terms for structural anomalies is a major problem in developmental toxicology and causes great difficulties for administrative decision makers involved in public health evaluation of chemical substances. The absence of harmonisation of terminology is no longer acceptable for regulatory purposes. The debate is unending, however, refinement and consensus are indispensable. This article is a report of the Second Workshop on Terminology in Developmental Toxicology. Experts from research institutions, regulatory agencies, and industries took part in this workshop, which has started a process of discussion that eventually will lead to a harmonisation of terminology used for classification of structural anomalies. The participants put forward a scheme of classification for foetal abnormalities that consists of only two categories: “malformation and variation.” Finally, consensus was achieved in defining the terms malformation and variation. Malformation is defined as a permanent structural change that is likely to adversely affect the survival or health of the species under investigation. The term variation is defined as a change that occurs within the normal population under investigation and is unlikely to adversely affect survival or health. This change might include a delay in growth or morphogenesis that has otherwise followed a normal pattern of development.

Magnesium deficiency induces joint cartilage lesions in juvenile rats which are identical to quinolone-induced arthropathy.

Quinolones accumulate in cartilage, and because they form chelate complexes with divalent cations, they possess the potential to induce a deficiency of functionally available magnesium. To test the hypothesis that quinolone-induced arthropathy is caused (or aggravated) by magnesium deficiency in cartilage, we induced magnesium deficiency by feeding juvenile rats a magnesium-deficient diet for 9 days and treated the rats with single oral doses of ofloxacin (0, 100, 300, 600, or 1,200 mg/kg of body weight) during this period. Additional groups of juvenile rats on a normal diet were treated with ofloxacin correspondingly. Typical cartilage lesions (e.g., swollen matrix, cleft formation) were found in knee joints of all magnesium-deficient rats, including those without ofloxacin treatment. Lesions in these groups were not distinguishable from lesions induced by a single dose of 600 mg of ofloxacin per kg of body weight or higher in rats on a normal diet. Ofloxacin levels in plasma after 600 mg/kg of body weight were approximately 10-fold higher than those in humans during therapy with this quinolone. Lesions in rats treated with ofloxacin plus magnesium deficiency were more pronounced than those in rats with normal magnesium concentrations. After intake of a magnesium-deficient diet for 9 days, the magnesium concentration in serum (mean +/- standard deviation) was 0.18 +/- 0.05 mmol/liter (control on normal diet, 0.82 +/- 0.10 mmol/liter). Magnesium concentrations in bone (femur) and cartilage (processus xiphoideus) samples were 64.7 +/- 10.5 and 14.3 +/- 3.9 mmol/kg of dry weight, respectively, which corresponded to approximately 50% of the concentrations measured in controls on a normal diet. It was concluded that quinolone-induced arthropathy is probably caused by a deficit of available magnesium in joint cartilage due to the formation of quinolone-magnesium chelate complexes. If juvenile patients must be treated with quinolones for serious infections, it seems prudent to ensure that these patients do not have a disturbed magnesium balance.

Polyhalogenated dibenzo-p-dioxins and dibenzofurans and the immune system. 1. Effects on peripheral lymphocyte subpopulations of a non-human primate (Callithrix jacchus) after treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)

Monoclonal antibodies were used to analyse the effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on peripheral lymphocytes from marmosets (Callithrix jacchus) following injections of single low doses of TCDD. A reduction of the percentage and the total number of lymphocytes among the white blood cells was seen 3 weeks after a single administration of 300 ng TCDD/kg body wt, but not following 167 ng TCDD/kg or less. After treatment of marmosets with the single subcutaneous dose of 10 ng TCDD/kg body wt, we observed an average decrease of about 20% in the percentage of CD4+ cells in the venous blood of treated animals. This change was not seen at the time of maximum absorption (2 weeks after the injection), but was demonstrable after 4 weeks and reached its maximum 15 weeks after the injection. The time course of the changes suggests an indirect effect (possibly via the thymus). Due to the considerable inter- and intra-individual variability in the number of peripheral lymphocytes the effect was less convincing on the basis of the total CD4+ cells per microliter blood. There was a significant concomitant increase in the percentage of cells with the CD8+ marker. A closer analysis of the lymphocyte subpopulation involved revealed a predominant effect on the cells with the CD4CDw29 (“leu 3a+4B4+”) surface marker (“helperinducer cells”). Subsequent to a dose of 10 ng TCDD/kg body wt the percentage of these cells was reduced by 50% or more when compared with the (24) controls. On an absolute basis (number of cells/μl blood) these CD4+CDw29+ cells were clearly reduced in only two out of four marmosets. There was no significant effect on the percentage of the CD4+CD45R+ (“leu3a+2H4+”) subpopulation (“suppressor-inducer cells”). The ratios: CD4+CDw29+/CD4+CD45R+, or CD4+CDw29+/CD8+ appear to be convenient measures for monitoring the effect described. Since the difference between the percentage of CD2+ cells minus the sum of CD4+ plus CD8+ cells was found to be increased following rather high doses (167 ng TCDD/kg body wt or more), this suggests that immature cells (CD2+ CD4-CD8-) are released into the periphery as a result of the exposure to TCDD. Furthermore, a decrease in the percentage of CD20+ (B1+) cells (about 50% when compared with controls) was observed in the treated animals following a single dose of 10 ng TCDD/kg body wt or higher. Subsequent to a dose of 10 ng TCDD/kg body wt the percentage of CD56+ (NKH1+) cells seemed to be slightly increased. The dose-response for the effects observed was rather poor. At present it cannot be decided whether the changes have to be considered as adverse health effects. All of the animals involved were apparently healthy and did not exhibit, e.g. any infections. However, the deviations described certainly represent biologicaleffects induced at very low dose levels. From the data available up till now, a single dose of 10 ng TCDD/kg body wt seems to be the “lowest-observed-effect-level” (LOEL) in the marmoset for all the effects studied. More extensive experiments within the dose range between 1 and 10 ng TCDD/kg body wt are under way in our laboratory. When peripheral lymphocytes of TCDD-treated animals were incubated in vitro with a mitogen (PWM = poke weed mitogen) an exaggeration of thedecrease in the percent of cells with the CD4 surface marker and a concomitant increase in the percentage of CD8+ cells was observed. This seems to be an especially sensitive experimental approach for demonstrating biological effects (but not necessarily adverse health effects) of this class of substances. Such an effect was demonstrated and found to be statistically significant already 2 weeks after a single injection of 10 ng TCDD/kg body wt in comparison to vehicle-treated controls.

Safety Overview: Toxicity, Adverse Effects, and Drug Interactions

Publisher Summary nThis chapter reviews the toxicity, adverse effects, and drug interactions of the quinolones. The fluoroquinolones are usually considered as relatively safe and well tolerated drugs. Under therapeutic conditions, no uniform life-threatening organ toxicity is known that would restrict their clinical use. However, in certain cases, unexpected adverse effects occurred after treatment with the fluoroquinolones as observed before with antimicrobials from other classes. Stained teeth in children due to tetracyclines, effects of certain cephalosporins on blood clotting, and severe reactions seen with the fluoroquinolone temafloxacin were noticed only several months or even years after launch of the drug. The fluoroquinolones temafloxacin, trovafloxacin, and grepafloxacin were taken off from the market due to rarely occurring severe adverse reactions. The use of temafloxacin was associated with a syndrome of hemolysis, renal failure, and thrombocytopenia. Due to this toxicity, the drug was withdrawn from the market shortly after approval for its clinical use in 1992. The use of trovafloxacin was associated with severe hepatic reactions in some rare cases. A total of 140 patients with severe hepatic reactions came into surface after the drug was prescribed worldwide to approximately 2.5 million patients. Another quinolone that was taken off the market in 1999 was grepafloxacin. Due to its effect on cardiac repolarization, manifested as QT interval prolongation on the electrocardiogram, some patients were considered at risk of torsade de pointes when treated with grepafloxacin.

Fluoroquinolones in the elderly: safety considerations.

Fluoroquinolones such as ciprofloxacin, levofloxacin, moxifloxacin and gatifloxacin are widely used for the treatment of bacterial infections. Fluoroquinolone-induced adverse effects have not been reported to occur with increased frequency in the elderly, but large trials comparing the tolerability in aged and young individuals are not available. Renal function declines consistently with age and recommendations for dosage changes of renally eliminated fluoroquinolones (ofloxacin, levofloxacin, gatifloxacin) are related to changes in kidney function rather than to age per se. However, during routine clinical work, creatinine clearance data are usually not available; thus it seems more practical to recommend dosage adjustment for elderly individuals in whom low creatinine clearance values can be expected.Reactions of the gastrointestinal tract are the most often observed adverse effects during therapy with fluoroquinolones; however, compared with many other antibacterials, fluoroquinolones are less frequently associated with diarrhoea. Similarly, hypersensitivity reactions, as observed during therapy with penicillins and other β-lactam agents, occur more rarely during fluoroquinolone therapy. Adverse reactions of the CNS are of particular concern for the elderly population. Elderly patients with impairments of the CNS (e.g. epilepsy, pronounced arteriosclerosis) should be treated with fluoroquinolones only under close supervision. Probably, many signs of possible adverse reactions such as confusion, weakness, loss of appetite, tremor or depression are often mistakenly attributed to old age and remain unreported.Fluoroquinolones can cause QT interval prolongation. Therefore, they should be avoided in patients with known prolongation of the QT interval, patients with uncorrected hypokalaemia or hypomagnesaemia and patients receiving class IA (e.g. quinidine, procainamide) or class III (e.g. amiodarone, sotalol) antiarrhythmic agents.Chondrotoxicity of fluoroquinolones, as observed in immature animals, has led to restricted use in paediatric patients, but there is no indication that similar effects could occur in joint cartilage of adults. Tendinitis and tendon ruptures have occurred in rare cases as late as several months after treatment with some fluoroquinolones. Chronic renal diseases, concomitant use of corticosteroids and age over 60 years have been recognised as risk factors for fluoroquinolone-induced tendon disorders.Overall, the widely used fluoroquinolones discussed in this review are generally well tolerated. Nevertheless, as with all drugs, their specific adverse effect profiles must be considered when they are chosen for treatment of bacterial infections. Because of physiological changes in renal function and in case of certain comorbidities, some special considerations are necessary when fluoroquinolones are used to treat elderly patients.

Ultrastructure of Achilles tendon from rats after treatment with fleroxacin

Abstract. Quinolone therapy can be associated with tendon disorders (tendinitis, ruptures), but little is known about possible ultrastructural changes in tendons after exposure to these antimicrobials. We studied the Achilles tendons from fleroxacin-treated adult rats by electron microscopy. Wistar rats were treated orally with single oral doses of 0, 30, 100, 300 or 600xa0mg fleroxacin/kg body weight (n=6 per group). The animals were killed 4 weeks after treatment, Achilles tendon samples were collected and tangential sections were made from the distal part of the tendon. Subsequently, tendons were cut crosswise for preparation of ultrathin sections. Samples were fixed by using glutaraldehyde, osmium tetroxide, tannic acid and finally contrasted with uranyl acetate/lead citrate before they were examined by transmission electron microscopy. The rats did not show any general effects such as behavioural changes or body weight changes which could be attributed to the treatment. However, we were able to detect pathological changes even at the lowest dose level (30xa0mg/kg), which increased in incidence and severity with increasing doses. Tenocytes exhibited degenerative changes such as multiple vacuoles and large vesicles in the cytoplasm that resulted from swelling and dilatation of cell organelles (mitochondria, endoplasmic reticulum). The nucleus became dense and the chromatin had clumped to form rough plaques. The cells detached from the extracellular matrix. Other important findings were a general decrease of the fibril diameter and an increase in the distance between the collagenous fibrils. The finding that these rather low single doses of a fluoroquinolone induce ultrastructural changes in Achilles tendons from rats, which were not associated with clinical symptoms and which were still present 4 weeks after treatment, is of concern. Further toxicological as well as clinical studies are needed to characterize the conditions under which quinolone-induced tendon lesions develop.

Integrins on joint cartilage chondrocytes and alterations by ofloxacin or magnesium deficiency in immature rats

Abstractu2002Recently, we showed that magnesium deficiency induces lesions in knee joint cartilage from 5-week-old rats that are very similar to ofloxacin-induced cartilage defects. We concluded that quinolone-induced arthropathy is probably due to chelation of magnesium and thus a deficit in functionally available magnesium in joint cartilage (Stahlmann et al. 1995). As magnesium deficiency in joint cartilage could impair chondrocyte-matrix interaction which is mediated by cation-dependent integrin receptors of the β1-subfamily, we investigated integrin expression in joint cartilage from untreated, ofloxacin-treated and magnesium-deficient Wistar rats. With immunohistochemical methods using monoclonal and polyclonal antibodies, we showed that the integrin pattern in joint cartilage from rats corresponded largely to integrin expression described for human cartilage tissue: β1, α1, α3 and αν subunits and the α5β1 and ανβ3 heterodimers were consistently expressed. Joint cartilage lesions were detected in ofloxacin-treated and magnesium-deficient rats. Lesions were more pronounced in the quinolone-treated group. Expression of several integrins was reduced in the vicinity of lesions after oral treatment with 2 × 600 mg ofloxacin/kg for 1 day. Gross-structural lesions (e.u200ag., cleft formation, unmasked collagen fibres) in magnesium-deficient rats were very similar but changes in integrin expression were less pronounced. On the other hand, changes in cartilage matrix composition showed similar alterations in ofloxacin-treated and magnesium-deficient rats: fibronectin deposition in the cartilage matrix increased in both groups while glycosaminoglycan content decreased. In summary, similar defects occur in ofloxacin-treated and magnesium-deficient rats and with immunohistochemical methods subtle differences are demonstrable.