Impact of Pre-Analytical Time on the Recovery of Pathogens from Blood Cultures: Results from a Large Retrospective Survey

Background Prompt identification of bloodstream pathogens is essential for optimal management of patients. Significant changes in analytical methods have improved the turnaround time for laboratory diagnosis. Less attention has been paid to the time elapsing from blood collection to incubation and to its potential effect on recovery of pathogens. We evaluated the performance of blood cultures collected under typical hospital conditions in relation to the length of their pre-analytical time. Methods We carried out a large retrospective study including 50,955 blood cultures collected, over a 30-month period, from 7,035 adult septic patients. Cultures were accepted by the laboratory only during opening time (Mon-Fri: 8am-4pm; Sat: 8am-2pm). Samples collected outside laboratory hours were stored at room temperature at clinical wards. All cultures were processed by automated culture systems. Day and time of blood collection and of culture incubation were known for all samples. Results A maximum pre-analytical interval of 2 hours is recommended by guidelines. When the laboratory was open, 57% of cultures were processed within 2 h. When the laboratory was closed, 4.9% of cultures were processed within 2 h (P<0.001). Samples collected when the laboratory was closed showed pre-analytical times significantly longer than those collected when laboratory was open (median time: 13 h and 1 h, respectively, P<0.001). The prevalence of positive cultures was significantly lower for samples collected when the laboratory was closed compared to open (11% vs 13%, P<0.001). The probability of a positive result decreased of 16% when the laboratory was closed (OR:0.84; 95%CI:0.80-0.89, P<0.001). Further, each hour elapsed from blood collection to incubation resulted associated with a decrease of 0.3% (OR:0.997; 95%C1:0.994-0.999, P<0.001) in the probability of a positive result. Discussion Delayed insertions of cultures into automated systems was associated with lower detection rates, with potentially important consequences for patients. In each hospital setting the logistic factors able to shorten pre-analytical time should be carefully investigated and specifically targeted.