Denting hopes in one of the few treatments that had shown some initial promise in severe patients, a WHO Guideline Development Group (GDG) of international experts said there was "no evidence based on currently available data that it does improve patient-important outcomes".
The United States, the European Union and other countries have granted temporary approval for the use of remdesivir after initial research showed it may shorten recovery time in some coronavirus patients.
President Donald Trump was treated with remdesivir among other medicines after he tested positive for Covid-19 in October.
Friday's WHO recommendation was based on four international randomised trials among more than 7,000 patients hospitalised with the virus.
Publishing updated treatment guidance in the BMJ medical journal, the panel acknowledged that their recommendation does not mean that remdesivir has no benefit for patients.
But based on the latest figures, costs and delivery methods, it advised "against administering remdesivir in addition to usual care for the treatment of patients hospitalised with Covid-19, regardless of disease severity".
The panel recommended that remdesivir trials continued, however, adding that their advice didn't constitute proof that the treatment couldn't work in Covid-19 cases.
"That's why it's a conditional recommendation. (There may) still be potential small benefit, maybe in a subgroup (of patients)," Janet Diaz, the WHO's head of clinical care, told reporters on Friday.
"Is there a subgroup that may benefit more, such a severe patients versus critical patients"?
Manufacturer Gilead said that remdesivir was recognised as Covid-19 treatment by several official health agencies, including the US National Institutes of Health and Infectious Diseases of America, based on "robust data from several randomised control trials".
"We regret that the WHO recommendation do not take this data into account even as the number of new cases globally is increasing considerably," it said.
More about: WHO coronavirus
