Recurrent hyperkalemia in patients with chronic kidney disease and hepatitis C treated with direct antiviral agents

Hepatitis C virus (HCV) infection is a leading cause of chronic liver disease, cirrhosis, and hepatocellular carcinoma. Patients with chronic kidney disease (CKD), especially those with end-stage renal disease (ESRD) on hemodialysis, present high prevalence of HCV serum positivity. Currently, since the high efficacy and well tolerance to direct antiviral agents (DAAs), antiviral treatment is recommended for all patients infected with HCV, including those with stage 4–5 CKD on hemodialysis and waiting for renal transplantation [1].

Studies showed that sofosbuvir based regimens were highly efficacy and well tolerance in patients with HCV and stage 4–5 CKD [2, 3]. However, the safety of sofosbuvir, mainly sofosbuvir-derived metabolites in patients with CKD remains an opening question. In the safety pharmacology studies of sofosbuvir, high dose of GS-9851 (the parent of sofosbuvir) inhibited potassium current by approximately 13% at 159 μg/mL [4]. Patients with CKD are more prone to imbalance of homeostasis, such as metabolic acidosis, hyperkalemia. Based on our knowledge, up to now, serum potassium in patients with CKD treated with sofosbuvir has not been reported.

Previously, we reported a nosocomial HCV outbreak in Zhen’an County Hospital, Shaanxi Province, China, in January 2016 [5]. Thirty-three patients with hepatitis C and stage 5 CKD on hemodialysis were treated with half dose of sofosbuvir (200 mg) and full dose of daclatasvir (60 mg). All patients completed 24-week treatment and additional 12-week follow-up. This special population provided a unique opportunity for us to observe the serum potassium concentration during sofosbuvir treatment. Serum potassium concentrations were regular tested every 4 week or on demand during the treatment course and post-treatment up to 12 weeks. The levels of serum potassium 3 months prior to treatment were retrieved from the medical record database of the hospitals. Hyperkalemia was defined as serum potassium concentration higher than 5.5 mmol/L. Ten out of 33 were reported recurrence of hyperkalemia. We report hyperkalemia occurrences of these 10 patients with CKD and HCV infection who received sofosbuvir-based regimens.

This is the first time to report recurrent hyperkalemia in patients with CKD and HCV infection treated with half-dose of sofosbuvir based regimen, although bradycardia, changed intracellular Ca²⁺ and acute interstitial nephritis have been reported potentially associated with sofosbuvir.

The recurrence of hyperkalemia was highly related to the administration of sofosbuvir. Patients with CKD especially in uremia period are at high risk of hyperkalemia, while regular dialysis aims to replace the function of kidney and could maintain the balance of electrolyte. In our case serials, we cannot attribute recurrence of hyperkalemia to insufficiently dialysis, for the dialysis scheme was unchanged before, during and after antiviral treatment. The dialysis therapy was sufficient for these 10 patients; because the serum creatinine levels remained stable and majority (8/10) of the patients were not observed hyperkalemia prior to sofosbuvir based treatment. Furthermore, other electrolytes, such as serum sodium or serum calcium disorder was not observed, and acid-base kept in homeostasis, supporting the adequacy of hemodialysis. Moreover, after the end of treatment, 8 out of the 10 patients did not present hyperkalemia, and along with the time extension of drug cessation, no hyperkalemia was observed. The temporal consistency between the DAAs usage and hyperkalemia occurrence highly supported the association between DAAs and hyperkalemia. Based on the HEP Drug Interactions database, no clinically significant interaction was showed between the co-medicines. Collectively, the raising levels of serum potassium were highly suspected in association to the usage of sofosbuvir.

The mechanism of the observed hyperkalemia in patient with CKD and HCV infection treated with sofosbuvir based regimens was unclear. As mentioned above, the recurrence of hyperkalemia could not be attributed to the insufficiency dialysis. Previously, case serials of bradycardia resulted from the co-administration of sofosbuvir and amiodarone were reported [6]. Regan CP et al. and Millard DC et al. demonstrated that co-administration of sofosbuvir and amiodarone produced dysfunction of the sinoatrial node automaticity and atrioventricular node conduction, decrease in cardiomyocyte mechanical activity and intracellular Ca²⁺ transients [7, 8]. Acute interstitial nephritis revealed by renal biopsy was reported potentially related to sofosbuvir, we assume that potassium exchange between inside and outside of cells may be an important factor of hyperkalemia [9, 10]. Actually, although no literature reports the impact on potassium channel by sofosbuvir, high dose of GS-9851 (the parent of sofosbuvir) did inhibit hERG-mediated potassium current in HEK293 cells expressing cloned hERG channels [4]. Integrating these evidences together, sofosbuvir do induce the dysfunction of ion channel, and such impact is more prominent in those cases with CKD and co-administered with drugs acting on an ion channel. In current report, calcium-channel blockers (CCBs) were administered in 7 patients. The drug-drug interaction could not be neglected in patients with stage 5 CKD co-administered with sofosbuvir and CCBs, since the metabolic pathway of CCBs have an intersection with that of sofosbuvir.

This is a retrospective report, the occurrence of hyperkalemia may be underestimated especially in patients with stage 5 CKD. However, the observation of this report could still provide a clue for clinicians to strengthen the monitoring for serum potassium when treating HCV infection with sofosbuvir, even its dose is halved, based regimens in patients with CKD.