In the Airborne Infection Research (AIR) facility near Pretoria we have conducted the first of a series of trials to determine the potential of UV222nm to disinfect air in settings where transmission of airborne infection is considered a significant hazard. Through rapid clinic review, including facemask sampling to assess infectiousness, we recruited 7 individuals newly diagnosed with pulmonary TB and prior to starting treatment (as agreed by local ethics) to reside at AIR where all ward extract air is passed to one of two guinea pig (GP) housing rooms. The patient rooms were illuminated with UV-C222nm (3uW/cm2 average fluence rate) on alternate days providing 90 GPs exposed to UV-treated air and 90 to control untreated air (one animal in the UV group died due to an unrelated cause). A total of 1040.54 person hours of GP exposure to extract air was recorded, with a slight excess (6.9%) for the control animals. Both control and UV-protected animals were skin tested for TB by standard tuberculin testing 6 weeks after study initiation. There was a striking and significant (p<0.0001) difference between the frequency of TST reactions with almost complete absence of reactivity (indicating infection) in the group exposed to UV-treated air. Although no participants either reported or were observed to show adverse effects attributable to UV-exposure, we now plan to determine the efficacy of lower UV irradiation levels in similar studies to minimise human exposure and optimise cost-benefit aspects of this approach to air disinfection.