Enhancing the detection sensitivity of mNGS in Bronchoalveolar Lavage Fluid through cell counting: An empirical study

Lower respiratory tract infections pose significant clinical challenges due to their high morbidity and mortality rates. While metagenomic next-generation sequencing (mNGS) has emerged as a promising diagnostic tool, its sensitivity is often compromised by host DNA contamination that overwhelms microbial signals. Selective host DNA depletion through cell lysis effectively reduces host DNA; however, it has an impact on microorganisms with relatively thin cell walls, and samples with low host content may introduce more environment or reagent-derived microbial contamination, interfering the detection results. Methods for determining host DNA depletion based on sample type, sample characteristics or using spike-in controls to monitor sensitivity do not fully consider the potential limitations of host depletion technology on microbial detection, nor do they evaluate the possible significant impact on detection efficiency. This study aimed to develop a pre-analytical method for accurate host DNA content assessment.

We established a cell-counting-based method for precise cellular content measurement in clinical bronchoalveolar lavage fluid (BALF) samples. The protocol involved: (1) evaluating the linearity and robustness of cell-counting dyes in BALF samples with varying characteristics, (2) assessing the correlation between cell counts and extracted nucleic acid mass, (3) investigating cellular counting thresholds for host depletion in clinical BALF analysis, and (4) implementing the optimized cell-counting method in clinical mNGS testing to guide selective-lysis treatment.

Acridine orange/propidium iodide (AO/PI) staining demonstrated superior performance compared to trypan blue and 4',6-diamidino-2-phenylindole (DAPI), particularly in turbid and bloody BALF samples. Implementing a host depletion threshold at 1 × 10⁶ cell counts significantly improves pathogen detection rates in high host background samples, while effectively preserving the detection sensitivity for pathogens in moderate and low host background samples.

Our findings demonstrate that cell counting serves as a reliable pre-analytical tool for determining optimal selective-lysis treatment in BALF mNGS testing, enhancing diagnostic accuracy while preserving pathogen integrity.