In order to realize the full potential of total laboratory automation (TLA) in the clinical microbiology laboratory, workflows must be optimized to match each laboratory's capabilities, patient population, and staffing model. Using TLA-based digital photography to monitor urine cultures, we sought to improve culture result turn-around-time (TAT) by changing the time at which a culture is first photographed and thus available for analysis/work-up (Pre1) from 18 h (16,391 cultures) to 16 h (53,113 cultures) (with a total of 24-h culture incubation in both time periods); in both time periods, cultures were set up 24/7, and culture work-up occurred during the day shift only. With this change, we observed a significant decrease in time-to-final-result TAT for positive cultures (18 h-Pre1 median: 71.6 h; 16 h-Pre1 median: 61.0 h). This effect was most pronounced for Gram-negative organisms, with a median reduction in time-to-final-result for Escherichia coli cultures (51.8% of positive urine cultures) of 14.2 h (18 h-Pre1 median: 77.3 h; 16 h-Pre1 median: 63.1 h). This reduction in TAT was accompanied by a decrease in sensitivity at the Pre1 time point (18 h-Pre1 91.01%; 16 h-Pre1 88.06%), which we also found to vary by species: there was a reduction in sensitivity at the first culture reading of 1 to 2% for cultures with Gram-negative microorganisms, but for some Gram-positive microorganisms (e.g., Aerococcus urinae and non-aureus Staphylococcus species), there was a reduction in sensitivity at the Pre1 time-point of 3 to 7%. These results can guide workflow decisions for laboratories seeking to implement and/or optimize TLA systems, demonstrating a tradeoff between TAT and the sensitivity of preliminary urine culture results.
Keywords: Automation; Escherichia coli; Total laboratory automation; Urine.