Electrophysiological Adverse Effects of Direct-Acting Antivirals in Patients With Chronic Hepatitis C.

Abstract

Dear Editor, We read with great interest the article by Durante-Mangoni et al.1 Sofosbuvir (SOF) has recently been correlated with bradyarrhythmias, especially if coadministered with amiodarone. In this study, Durante-Mangoni et al assessed several electrocardiographic (ECG) parameters in 26 subjects who were consecutively treated with various SOF-based regimens. They did not observe significant changes in ECG parameters except a transient increase of QTc duration during the first week that returned to baseline values afterward. In contrast to what was reported by Renet et al2 and Fontaine et al,3 the authors concluded that SOF-based regimens could be considered safe. HIV/HCV coinfected patients are often treated with methadone and antiretrovirals, such as atazanavir and rilpivirine, which have an effect on QTc duration. Thus, we performed a similar evaluation of ECG parameters in 58 coinfected individuals treated with different SOF-based regimens from January 2015 to July 2016. ECG parameters were prospectively collected at baseline, after 1 and 4 weeks, and at the end of therapy. Compared to the study by Durante-Mangoni and colleagues, our population is larger and has several differences, especially in terms of liver fibrosis severity (Table 1). In the overall analysis, QTc duration did not change significantly over treatment (P = .31). Patients receiving atazanavir or rilpivirine had stable QTc intervals during treatment. No difference was noted for any direct-acting antiviral combination. At baseline, F3 (defined by a stiffness between 10 and 14 kPa) and cirrhotic subjects had comparable QTc values (408 milliseconds vs 422 milliseconds for F4 [defined by a stiffness >14 kPa] overall, 421 milliseconds for Child-Pugh-Turcotte class [CPT] A and 425 for CPT B/C), but such difference was not significant (P = .16). At week 4, F3 individuals maintained the same QTc value (407 milliseconds), but F4 patients had a slight increase to 424 milliseconds (P = 0.01). If we consider F3, CPT A (423 milliseconds), and CPT B/C (426 milliseconds) separately, the difference is at the limit of significance (P = .05). PR, QRS duration, and heart rate did not show any change. Our study confirms that SOF-based regimens cause no ECG changes except in cirrhotics at week 4. It is noteworthy to observe that liver cirrhosis was the only condition in common among all the published cases of arrhythmia. SOF caused the following pharmacokinetic changes in cirrhosis: area under the concentration-time curve increased by 126% and 143% in CPT class B and C patients, respectively.4 It might be speculated that electrophysiological alterations are the rare result of a multifactorial phenomenon that includes exposure to SOF and to several facilitating factors; our data suggest that cirrhosis might be included among them. Thus, ECG monitoring could be taken into consideration when treating subjects with more advanced disease. The authors declare no personal or funding interests. Preliminary results of this study have been presented at the 12th International Workshop on Coinfection, June 2-3, 2016, Berlin, Germany.