Striking differences in proconvulsant-induced alterations of seizure threshold in two rat models

During drug development, seizure threshold tests are widely used to identify potential proconvulsant activity of investigational drugs. The most commonly used tests in this respect are the timed intravenous pentylenetetrazole (PTZ) infusion seizure test and the maximal electroshock seizure threshold (MEST) test in mice or rats. To our knowledge, no study is available in which proconvulsant drug activities in these models are directly compared, which prompted us to perform such experiments in male Wistar rats. Five drugs with reported proconvulsant activity were tested in the two models: d-amphetamine, chlorpromazine, caffeine, theophylline, and tramadol. Furthermore, the anticonvulsant drug phenobarbital was included in the experiments. While phenobarbital exerted anticonvulsant activity in both models, the five proconvulsant drugs markedly differed in their effects. In the dose range tested, d-amphetamine significantly lowered the PTZ seizure threshold but increased the MEST, caffeine and theophylline did not alter the PTZ seizure threshold but decreased the MEST, and tramadol reduced the PTZ threshold but increased the MEST. These marked differences between seizure threshold tests are most likely a consequence of the mechanisms underlying seizure induction in these tests. Our data indicate that using only one seizure threshold model during preclinical drug development may pose the risk that potential proconvulsant activity of an investigational drug is overseen. However, the label “proconvulsant” may be misleading if such activity only occurs at doses high above the therapeutic range, but the drug is not proconvulsant or even exerts anticonvulsant effects at lower, therapeutically relevant doses.

► Important to identify proconvulsant activity of drugs prior to human clinical trials.
► The pentylenetetrazole (PTZ) seizure threshold test is commonly used in this respect.
► Comparison of drug effects in the PTZ and electroconvulsive seizure threshold models.
► Striking differences with known proconvulsant drugs in the two models.
► A battery of models should be used for correctly determining proconvulsant activity.