Comment on: Hepatotoxicity rates do not differ in patients with rheumatoid arthritis and psoriasis treated with methotrexate: reply

Abstract

SIR, We thank our colleague for the well-written and interesting comments [1] regarding our study [2]. The differences between the views of dermatologists and rheumatologists regarding prevention of MTX hepatotoxicity are well known; policies and approaches differ between these two medical communities. We believe that the major drawback of the current attitudes and guidelines held by many dermatologists are that they are not based on satisfactory evidence. The current recommendations are based on studies, some quoted by Dr Lynch, which were conducted in psoriatic patients more than two decades ago; however, these studies encompassed a limited number of patients. In the study by O’Connor et al. [3] that was also mentioned in Dr Lynch’s comment, liver enzyme monitoring was shown to be a good predictive model for positive findings from liver biopsy specimens of psoriatic patients; however, this model was based on only 78 patients, so far no consensus has been reached on the predictive value of liver function tests regarding liver histology in psoriasis patients. Our study has weaknesses; as we mentioned, clinical data were retrieved from the electronic files of all psoriatic and rheumatoid patients treated by the Maccabi health maintenance organization that insures more than 1.7 million members in Israel. However, some parameters that we wanted to analyse, such as consumption of alcoholic beverages, other hepatic comorbidities and liver biopsy results, were not coded and therefore not available for analysis. Although a possible confounder, one should know that consumption of alcoholic beverages and the rate of alcoholism in Israel is extremely low compared with European countries; therefore, this parameter probably had no major impact. Furthermore, we believe that these non-differential misclassification biases are unrelated to disease type and status and, therefore, are not potential confounders. Interestingly, although our data showed that rheumatoid patients had higher rates of abnormal liver enzymes than did psoriatic patients, this statistic did not reach significance. Our study has significant strengths. The computerized data enabled access to a ‘real-life population’ of several hundred patients who were managed routinely. Given the potential hazards of liver biopsies, we believe that the current recommendations should be revisited.