Ribavirin for Crimean-Congo hemorrhagic fever: systematic review and meta-analysis

Crimean-Congo hemorrhagic fever epidemics often occur in areas where health services are limited, and result in high case fatality rates. Besides intensive care, ribavirin is often recommended. A solid evidence base for the use of this drug will help justify assuring access to the drug in areas where epidemics are common.

We carried out a systematic review of observational and experimental studies of people with suspected or confirmed Crimean-Congo hemorrhagic fever that included comparisons between patients given ribavirin and those not. We extracted data on mortality, hospital stay, and adverse events. Risk of bias was assessed using a standard checklist, and data were presented in meta-analytical graphs, stratified by study design, and GRADE tables presented. The risk of bias was summarised using the GRADE method.

21 unique studies, including one randomised controlled trial of ribavirin, were included. Quality of the evidence was very low, with a Down and Black median score of 4 (maximum possible 33). Ribavirin treatment was not shown to be superior to no ribavirin treatment for mortality rate in a single RCT (RR: 1.13, 95%CI: 0.29 to 4.32, 136 participants, GRADE=low quality evidence); but ribavirin was associated with reduced mortality by 44% when compared to no ribavirin treatment in the pooled observational studies (RR: 0.56, 95%CI: 0.35 to 0.90, 955 participants; GRADE=very low quality evidence). Adverse events were more common with the ribavirin patients, but no severe adverse events were reported. No difference in length of hospital stay was reported.

No clear message of benefit is available from the current data on ribavirin as observational data are heavily confounded, and the one trial carried out has limited power. However, ribavirin could potentially have benefits in this condition and these results clearly indicate a pragmatic, randomised controlled trial in the context of good quality supportive care, is urgently needed and ethically justified.