Background:
To examine the association of use of different antivirals with hospital acquired acute kidney injury (AKI) among Chinese adults with herpes zoster.
Methods:
The study selected 3273 adult patients who received antiviral therapy for herpes zoster during hospitalization from the China Renal Data System (CRDS). We identified and staged AKI using patient-level serum creatinine data according to the Kidney Disease Improving Global Outcomes (KDIGO) criteria. We compared the relative risks of hospital acquired AKI among patients treated with different antivirals using Cox proportional hazards models.
Results:
Among 3273 patients, 1480 (45%), 681 (21%), 489 (15%) and 623 (19%) were treated with acyclovir/valacyclovir, ganciclovir, penciclovir/famciclovir, and foscarnet, respectively. During the follow-up period, a total of 111 cases of hospital acquired AKI occurred, predominantly classified as AKI stage 1. The cumulative incidences of hospital acquired AKI were 5%, 3%, 3% and 1%, in the patients receiving acyclovir/valacyclovir, ganciclovir, penciclovir/famciclovir, and foscarnet, respectively. Compared with acyclovir/valacyclovir, penciclovir/famciclovir/ganciclovir and foscarnet were associated with a lower risk of hospital acquired AKI, with an adjusted hazard ratio [aHR] of 0.59 (95% CI, 0.37-0.94) and 0.27 (95% CI, 0.11-0.63), respectively. Compared with intravenous acyclovir, intravenous penciclovir/ganciclovir and foscarnet were associated with a lower risk of hospital acquired AKI with an aHR of 0.53 (95% CI, 0.29-0.98) and 0.31 (95% CI, 0.12-0.76), respectively. The associations were consistent across various subgroups and sensitivity analyses.
Conclusions:
Among antiviral therapies for herpes zoster, we found different risks of hospital acquired AKI among the patients receiving different antivirals, in particular, those administered intravenously. Among intravenous antivirals, acyclovir was associated with highest risk of hospital acquired AKI, followed by penciclovir/ganciclovir and foscarnet. Confirmation studies with large samples from other populations are warranted.