Antibiotics for Delirium in Older Adults With No Clear Urinary Tract Infection: A Multicentre, Pragmatic, Randomized Trial-Vanguard Results.

BACKGROUND

It is unclear whether antibiotics improve outcomes for older adults with delirium who have pyuria or bacteriuria in the absence of other signs or symptoms of a urinary tract infection.

METHODS

Our study is a multicenter, parallel-group, single-blinded, non-inferiority pragmatic randomized trial. The trial included a vanguard phase of the first 30 participants, and herein we report the aggregate results from this phase. We included hospitalized adults aged 60 years and older who had delirium and pyuria and/or bacteriuria. We excluded patients with fever, upper or lower urinary tract symptoms, an indwelling foley for more than 72 h, more than 24 h of antibiotics at the time of eligibility, or another indication for antibiotics (e.g., pneumonia). The main outcomes of the vanguard phase were rates of participant recruitment, protocol adherence, and study withdrawal. Our initial goals were 4 patients randomized monthly per site, greater than 95% adherence to allocation, and less than 5% withdrawal rate.

RESULTS

The median age was 86 years (interquartile range [IQR] 81-92), 68% were female, and most (79%) had either minor or major neurocognitive impairment. Of the participating sites that recruited at least one participant, the recruitment rates ranged from 0.05 to 1.8 participants per month. Adherence to allocation was 100%. Two participants (6.7%) withdrew. For participants randomized to receive antibiotics, the median duration was 5 days (IQR 5-7). By Day 7, delirium had resolved in 39% of the participants who remained hospitalized; 25% of participants were discharged, 32% had ongoing delirium, and 4% did not have a delirium assessment available. By Day 7, 1 (3.6%) developed C. difficile , 1 (3.6%) died, and 1 (3.6%) developed bacteremia.

CONCLUSION

We completed our vanguard phase with 100% adherence to allocation and withdrawal was near the target of 5%. The recruitment rate varied widely and overall was lower than anticipated.