A publication in the latest issue of the journal Science (15 February 2013), in which Tomas Brodin and colleagues at the University in Umeå, Sweden, report on the effects of traces of the benzodiazepine oxazepam on the behavior of perch, excited not only in the lay press quite a stir (Brodin et al, Science 2013, 339: 814-815). The authors report that the concentrations of oxazepam, which are usually found in river waters, lead to significant, potentially ecologically relevant behavioral changes in fish which is exposed to these waters. In a medium sized Swedish river (Fyris) the authors found an oxazepam concentration of 0.58 γg/L. According to the authors, this concentration is comparable with the values that other authors have found in American and European waters. They exposed perch for seven days two different concentrations of oxazepam: a low, environmentally relevant concentration of 1.8 γg/L and a high concentration of 910 γg/L. Not only the high, but also already the low concentration led to significant behavioral effects in fish. Their activity was increased, while their social activity was reduced. In addition, the fish ate more effective and faster, resulting in a more rapid depletion of the water on plankton. The authors speculate that this might have implications for the ecology of the affected waters. They also demonstrated an accumulation of oxazepam in the muscle of the perch: In case of exposure to the low concentration, they found an average of 3.6 γg oxazepam per kilogram of muscle.
When discussing the significance of these findings, especially the question arises how such low concentrations of benzodiazepines produce their effects. Assuming a specific pharmacological effect, this should be mediated – like in humans – through the modulation of the GABA receptor. Benzodiazepines are highly effective and very specific drugs. As agonist drugs, they develop their sedative, anxiolytic, hypnotic, anticonvulsant and muscle relaxant effects even if they occupy only a few percent of the benzodiazepine receptors. In humans, the therapeutic, behaviorally relevant dose is 10 mg/tablet. That is more than 5000 times more than the content in a liter of water, which is reported to lead to the effects described in the perch! The Task Force Therapeutic Drug Monitoring of the AGNP suggests a therapeutic plasma concentration reference range of 200-1500 ng/ml (200-1500 γg/L) in humans (Hiemke et al., 2011). This means that in the blood plasma (!) of humans for a therapeutic effect 100-1000-fold higher concentrations of the substance must be present than in a liter of water in question. The federal authorities have therefore – and rightly so (!) – soon the publication of the results hastened to deny a harmful effect on humans.
Brodin and colleagues report that after a week of exposure they found an accumulation of oxazepam in the muscle of perch up to a concentration of 3.6 γg/kg. A perch of the studied species usually has a lenghth of not more than 20 cm and only rarely reaches a weight of 1 kg. Even if such a large fish would consist to 100% of muscle, a person would have to eat at least 1000 perch at once in order to ingest a pharmacologically active amount of oxazepam. Even these accumulated concentrations are therefore still very small compared with the levels, which are measured at human administration of oxazepam.
Interestingly, there are numerous studies in animals, especially mice and rats, in which much higher doses were administered to achieve behavioral effects. Usually mice and rats are administered approximately 10 mg/kg, sometimes considerably more. This is approximately 50-fold higher than the doses administered to humans! Thus, are perch (or other fish) especially sensitive to benzodiazepines? Are the incredibly low concentrations measured high enough to induce specific pharmacological effects? It is helpful to look at the literature regarding the effects of benzodiazepines in fish, which quite exists. Facciolo et al. (2012) also reported in a recent paper (Facciolo et al., Behav Brain Res 2012) that diazepam and zolpidem lead to behavioral change in goldfish, especially a change in feeding behavior. These goldfish were administered 2-20 micrograms diazepam per gram of body weight (goldfish are about the size of perch), i.e. 2-20 mg/kg, i.e. about the dosages administered to rodents! The behavioral effects were strongest at a dosage of 5 mg/kg. Assuming that diazepam and oxazepam not significantly differ in their potency, 5 mg are almost 3,000 times more than is contained in a liter of water in the study by Brodin et al. Thus, some questions remain, in any case the observations require replication.
It should be finally noted that at least one claim that is placed in the paper, which seems to emphasize the significance of the findings, however, is not quite right. Benzodiazepines are by far not the most frequently ingested psychotropic drugs. Secondly, while the prescription of antidepressants is increasing, the intake of benzodiazepines – at least in Germany and other industrialized countries – has been declining for several years.
This post is also available in: German