STUDY OBJECTIVE: To compare the effectiveness and adverse effects of air-fluidized beds and conventional therapy for patients with pressure sores. DESIGN: Randomized trial with both masked and unmasked comparisons of outcome after a median follow-up of 13 days (range, 4 to 77 days). SETTING: Urban, academic referral, and primary care medical center. PATIENTS: Of 140 potentially eligible hospitalized patients with pressure sores, 72 consented to randomization; 65 (90%) completed the study. INTERVENTIONS: Thirty-one patients on air-fluidized beds (Clinitron Therapy, Support Systems International, Inc., Charleston, South Carolina) repositioned every 4 hours from 0700h to 2300h without use of other antipressure devices. Thirty-four patients on conventional therapy used an alternating air-mattress covered by a foam pad (Lapidus Air Float System, American Pharmaceal Company, Cincinnati, Ohio) on a regular hospital bed; were repositioned every 2 hours; and had elbow or heel pads as needed. Topical therapy was standardized for both groups. MEASUREMENTS AND MAIN RESULTS: Pressure sores showed a median decrease in total surface area (-1.2 cm2) on air-fluidized beds, but showed a median increase (+ 0.5 cm2) on conventional therapy; 95% confidence interval (CI) for the difference between medians, -9.2 to -0.6 cm2 (p = 0.01). Improvement, as assessed from serial color photographs by investigators masked to treatment group, occurred in 71% and 47%, respectively; 95% CI for the difference, 1% to 47% (p = 0.05). For pressure sores 7.8 cm2 or greater, outcome differences between air-fluidized beds and conventional therapy were greater: median total surface area change was -5.3 and +4.0 cm2, respectively; 95% CI for the difference, -42.2 to -3.2 cm2 (p = 0.01). Improvement rates were 62% and 29% respectively; 95% CI for difference, 1% to 65% (p = 0.05). After adjusting for other factors associated with sore outcome, the estimated relative odds of showing improvement with air-fluidized beds were 5.6-fold (95% CI, 1.4 to 21.7) greater than with conventional therapy (p = 0.01). No significant increase in adverse effects was seen with air-fluidized beds. CONCLUSIONS: Our findings suggest that air-fluidized beds are more effective than conventional therapy, particularly for large pressure sores. Studies are needed to determine the effectiveness of air-fluidized beds in long-term care settings.