73 Two cases of culture-negative endocarditits with positive Karius results

Abstract Karius (plasma metagenomics next-generation sequencing) has clinical utility under select circumstances, but there is variation in selecting patients for Karius testing. At our institution, Karius orders require the approval of the microbiology director. We present two cases of positive Karius results in culture-negative endocarditis. In the first case, a previously healthy 19-year-old Ecuadorian man was admitted for embolic stroke secondary to bicuspid aortic valve endocarditis. Serology was positive for Bartonella henselae IgG but negative for IgM. Blood PCR for B. henselae and quintana was negative. However, he had Karius results of B. henselae and Helicobacter pylori. He later seroconverted to positive B. henselae IgM and a higher IgG titer. After treatment, the patient’s IgG and IgM titers decreased. His course was complicated by stroke recrudescence, fever, and chills. The differential diagnosis included relapse, but he was ultimately diagnosed with influenza A. In the second case, a previously healthy 48-year-old man was admitted for embolic stroke secondary to aortic valve endocarditis and underwent valve repair. Blood cultures, valve tissue culture, 16S broad-range PCR, and histology were negative. Two months later, he was re-admitted for prosthetic aortic valve endocarditis. Blood cultures were negative. Karius was positive for Prevotella melaninogenica and Corynebacterium striatum. However, neither was considered the cause of the patient’s endocarditis and death by the infectious diseases consultants. The cases illustrate two appropriately ordered Karius tests with different clinical utility. In addition, we are performing a utilization review of the Karius tests sent over 5.5 years. So far, we have found that 46 out of 82 Karius tests were positive (56%). Out of a total of 71 organisms identified, 52 were bacteria. Of the bacterial results, 32 (62%) were Staphylococcus spp, Streptococcus spp, Enterococcus spp, Klebsiella spp, Pseudomonas aeruginosa, and Escherichia coli, which are non-fastidious. In addition, 11 bacterial results were species considered fastidious or difficult to culture; however, only 5 were considered pathogenic. In conclusion, our institution’s Karius positivity rates are similar to that of peer institutions published in the literature. Based on similar studies, we hypothesize that very few Karius results change clinical treatment. However, some clinicians may be more likely to use Karius in the correct clinical context than others.