Response to “Tricyclic antidepressants intoxication in Tehran, Iran: epidemiology and associated factors”

Abstract

I read with interest the study entitled ‘‘Tricyclic antidepressants intoxication in Tehran, Iran: epidemiology and associated factors.’’ I commend the authors for their study on identifying the individuals prone to tricyclic antidepressant (TCA) intoxication. Since it seems that their method for control selection is not correct, I have a few questions regarding their methodology as well as how to interpret the results. They have mentioned that they have divided the patients into two groups of TCA intoxication (case group) and the patients who had been poisoned by some other drug classes (control group). Can they provide more details regarding the type of the other drug classes in ‘Non-TCA intoxicated group?’ They have not presented the frequency of intoxication with each type of drug in this group, while they have even mentioned the number of the patients intoxicated by different types of TCAs in their ‘Results.’ Is it not necessary to present such a data in such type of study? Did they compare the level of consciousness of the patients intoxicated with the medications that do not usually decrease the level of the consciousness, for instance, those intoxicated by calcium channel blockers, digoxin, antimicrobials, or anticoagulants to the ones with TCA intoxication? If they did, is such a comparison correct? Another concern I had about the study is that the authors have compared a combination of the accidental and intentional intoxications in the two groups of the study. In this case, what is the value of the comparison of for example the marital status or the status of employment between the patients in the two study groups (as illustrated in Table 2)? Did not they have to compare the patients’ characteristics between the separated accidental and intentional poisoning groups? It seems that if we would like to identify the determinants of the TCA intoxication, we should compare the TCA-poisoned patients with a homogenously drug-poisoned group of the patients (for example, the intoxications induced by other types of antidepressants such as serotonin reuptake inhibitors, monoamine oxidase inhibitors, or antipsychotics, etc.). Did the authors perform such a comparison? It seems then that their results and conclusion is not correct due to their biased control selection.