2204. Microbiology of Pneumonia Due to Co-Infection in the ICU: Impact of Host Immune Status

BACKGROUND: Pneumonia epidemiology is increasingly showing the presence of co-infection due to the utilization of emerging diagnostic testing modalities such as multiplex polymerase chain reaction (PCR) panels. However, the prevalence and clinical significance of co-infection with respect to host immune status remain unclear. METHODS: A single-center retrospective analysis of mechanically ventilated adult patients treated in critical care units from January to October 2018 was performed on those with positive microbiological analysis of a bronchoalveolar lavage (BAL) sample. Host immune status and microbiological analyses were obtained including PCR and culture testing. Categorical variables and co-infection or immunocompetent status were assessed using Chi-Square, Fisher exact tests, or t-tests. REDCap was utilized for data abstraction and SAS software version 9.4 was used to perform all analysis. RESULTS: Of the 139 BAL samples that met inclusion criteria, 107 and 32 were obtained from immunocompetent and immunocompromised hosts, respectively. There was no statistical difference found between the frequency of co-infection detected by BAL culture with respect to host immune status. Immunocompetent patients had a higher proportion of positive bacterial cultures compared with immunocompromised (76.7% vs. 43.8% respectively, P = 0.0004). There was no significant difference seen with frequency of fungal or acid fast bacilli cultures between the two groups. Analysis of the microbiologic data obtained (figures) revealed different pathogens according to host immune status. CONCLUSION: Pneumonia due to co-infection in critically ill, mechanically ventilated immunocompromised hosts occurs at a similar frequency regardless of host immune status, however different microbiological patterns emerge. Interestingly, patients who were not immunocompromised had a higher proportion of positive bacterial cultures compared with those who were immunocompromised. Comparative analysis of the other pathogen types may also reveal differences in detection rates if sample size is increased. Clinically, this may help guide efficient use of microbiological testing among patients based on immune status. [Image: see text] [Image: see text] DISCLOSURES: All authors: No reported disclosures.