For a study, researchers sought to evaluate and understand the discrepant Droplet digital PCR (ddPCR) results for detecting bloodstream infections (BSI) in intensive care unit (ICU) practice; prospective clinical trials were lacking. Between May 21 and December 22, 2021, a multiplex ddPCR panel prospective diagnostic examination was carried out in a general ICU. The 12 most prevalent BSI pathogens and 3 antimicrobial resistance (AMR) genes were discovered using paired blood cultures (BCs) and ddPCRs (2.5 h). First, the efficacy of different BSI was compared to that of ddPCR. Second, a comparison between composite clinical diagnosis and clinical validation of ddPCR was made. Calculations were made for sensitivity, specificity, and positive and negative predictive values. The AMR gene positive rate was studied in the third section. There were 150 critically ill patients who had a total of 438 likely BSI incidents included. For the targeted bacteria, BC and ddPCR were positive in 40 (9.1%) and 180 (41.1%) occurrences, respectively. There were 158 discordant and 280 concordant replies. As opposed to BCs, ddPCR’s specificity ranged between species from 73.5 to 92.2% with an average of 63.1%, while its sensitivity varied between species from 58.8 to 86.7% with an average of 72.5%. Additionally, 20 (n=20) out of the 128 (out of 147%) individuals with ddPCR+/BC findings, or 10.8% (n=108) had probable or suspected BSIs. The overall sensitivity and specificity of the ddPCR increased to 84.9% and 92.5% when clinically diagnosed BSI was utilized as a true positive. About blaKPC, 3blaNDM and 38 mecA genes were also found, with 90.5% of them having positive blaKPC confirmations. Comprehensive microbiological tests projected that Staphylococcus species would be mecA-positive in 65.8% of the samples. Multiplexed ddPCR was a versatile and all-purpose platform that can be employed in addition to traditional BC. ddPCR exhibited potential advantages for quickly detecting suspected BSIs and AMR genes in ICU practice when combined with clinical infection evidence.

Source: ccforum.biomedcentral.com/articles/10.1186/s13054-022-04116-8