Table 1.

Analytical considerations for plasma mNGS in diagnosis of sepsis and other system infections.

Potential clinical indications	Preanalytical	Analytical	Postanalytical
• Sepsisbreak • Endocarditis • Severe pneumonia • Invasive fungal infection • Disseminated mycobacterial infection • Local infection with systemic spread • Fever of unkown origin • Other system infections (non-bloodstream) but no optimal specimen • Local and plasma samples sent simultaneously	• 0·3 mL plasma required •  Mild/moderate infection not detected by conventional microbiological testing • Severe infection, recurrent infection or mixed pathogen infection are suspected • Suspected infection with atypical or difficult-to-cultivate pathogens • Infections in immunocompromised patients • Select patients with long-term turnaround time that still affects management • mNGS may have advantages over traditional testing turnaround time • Inconsistent or suboptimal antimicrobial treatment response • Tissues or body fluids are preferred for local infections, try plasma mNGS testing when access is not available due to invasive procedures or patient instability • mNGS detection is possible before or after antimicrobial treatment	• Detect cell-free DNA • RNA viruses not detected • Report pathogen sequencing reads, genomic coverage, and relative abundance • No antimicrobial susceptibility or resistance evaluated	• Adapter sequences and low-quality sequences are removed • Set conventional reads threshold: bacteria, viruses, fungi ≥3, parasites ≥100, Mycobacterium tuberculosis, Brucella are set to ≥1 • Interpretation of results by a doctoral-level clinical microbiologist with expertise in mNGS • Organisms frequently occurring interfering background sequences in negative controls are excluded or placed in a separate list of suspected background organisms (informal report content), note that this step may cause mistaken deletion and need to be comprehensively interpreted • Negative mNGS results do not rule out infection • Detection of residual nucleic acid or transient bacteria is usually not judged clinically significant • Distinguish normal human microbiota • Highest abundance organism does not equate to clinical causation • Antibacterial selection consult an infectious disease physician or conduct a multidisciplinary consultation, combining mNGS with other pathogen detection tests

mNGS, metagenomic next-generation sequencing