A case for reduced frequency of CD4 count monitoring for children on combination antiretroviral therapy with consistently undetectable HIV viral load