Skip to main content

Metagenomic next-generation sequencing for rapid detection of pulmonary infection in patients with acquired immunodeficiency syndrome



Acquired immunodeficiency syndrome (AIDS) is associated with a high rate of pulmonary infections (bacteria, fungi, and viruses). To overcome the low sensitivity and long turnaround time of traditional laboratory-based diagnostic strategies, we adopted metagenomic next-generation sequencing (mNGS) technology to identify and classify pathogens.


This study enrolled 75 patients with AIDS and suspected pulmonary infections who were admitted to Nanning Fourth People’s Hospital. Specimens were collected for traditional microbiological testing and mNGS-based diagnosis. The diagnostic yields of the two methods were compared to evaluate the diagnostic value (detection rate and turn around time) of mNGS for infections with unknown causative agent. Accordingly, 22 cases (29.3%) had a positive culture and 70 (93.3%) had positive valve mNGS results (P value < 0.0001, Chi-square test). Meanwhile, 15 patients with AIDS showed concordant results between the culture and mNGS, whereas only one 1 patient showed concordant results between Giemsa-stained smear screening and mNGS. In addition, mNGS identified multiple microbial infections (at least three pathogens) in almost 60.0% of patients with AIDS. More importantly, mNGS was able to detect a large variety of pathogens from patient tissue displaying potential infection and colonization, while culture results remained negative. There were 18 members of pathogens which were consistently detected in patients with and without AIDS.


In conclusion, mNGS analysis provides fast and precise pathogen detection and identification, contributing substantially to the accurate diagnosis, real-time monitoring, and treatment appropriateness of pulmonary infection in patients with AIDS.


Human immunodeficiency virus (HIV) infection causes acquired immunodeficiency syndrome (AIDS), which is a serious threat to public health. This pathophysiological state reduces and destroys cell-mediated immunity in humans, resulting in a wide range of opportunistic infections by viruses, bacteria, fungi, and parasites. Moreover, AIDS leads to the emergence of complex ecosystems of multiple pathogens, which makes identifying pathogenic mechanisms difficult. For example, of all body’s organs, the lungs are most frequently affected by AIDS-related microbial infections [1, 2]; therefore, respiratory system failure is one of the leading causes of morbidity in patients with AIDS worldwide [3]. Thus, pathologists should proactively consider AIDS-related microbial infections via rapid and accurate microbiological diagnoses to identify more pathogenic microorganisms, explore pathogenic mechanisms, and enable the optimal management and treatment of patients with AIDS.

The traditional growth and isolation of microorganisms from various cultures are limited by the time required, low detection sensitivity, and difficulty of growing pathogens such as Helicobacter canis [4], Helicobacter pylori [5], Neisseria meningitidis [6], and Bartonella spp. [7]. In addition, invasive procedures are required when a pathogenic infection is confined to an anatomical location. There has been a rapid development of nucleic acid-based polymerase chain reaction (PCR) technologies, including real-time quantitative PCR and digital PCR, which can detect targeted pathogenic microorganisms at lower concentrations in less time and demonstrate a high degree of sensitivity and precision. However, PCR relies on highly specific primers and fails to detect untargeted genes, which is not conducive for precise treatment. In contrast, metagenomic next-generation sequencing (mNGS, Ion Torrent PGM, Roche’s 454 pyrosequencing technology, Illumina Miseq and Illumina Hiseq) has the potential to overcome the many limitations of conventional diagnostic approaches, and it has many advantages. Theoretically, mNGS-based diagnostic strategies allow for the detection, classification, serology, and antimicrobial resistance characterization of all classes of infectious agents [8]. Technically, mNGS is more precise, accurate, and sensitive, as has been shown and validated in a few clinical studies [9, 10]. More importantly, it greatly benefits immunocompromised patients in terms of a fast diagnosis, preventing their condition from worsening, shortening their hospital stay, and even improving survival rates [8]. Over the last two decades, mNGS technology has been extensively applied in research on the pathogenesis of infectious diseases. For example, accumulating evidence demonstrates the role of pathogenic microorganisms in periodontal disease [11], chronic/acute inflammation [12,13,14], cancer [15], metabolic disorders, and neurological diseases [16]. Furthermore, mNGS-based molecular diagnosis is more suitable when a causative etiological agent is suspected but no pathogen is detected through traditional laboratory-based diagnostic strategies, especially culture-based diagnostic methods.

Few studies have comprehensively evaluated the overall diagnostic performance of mNGS in AIDS-related pulmonary infections. Research on the implementation of mNGS is required to compare it with traditional because it must be compared with culture-based approaches and to determine the feasibility of transforming it into a clinical diagnostic test should be discussed. Here, mNGS diagnosis was used to evaluate its value in the efficient and accurate identification of pathogenic microorganisms in patients with AIDS and pulmonary infections. The results mNGS, including precise classification, positive detection rate, turnaround time, and diagnostic positivity rate, were compared with those from conventional culture-based screening. We also revealed the characteristics of the pathogen spectrum in 75 patients with AIDS and 32 patients without AIDS using mNGS. Further exploration of the value of mNGS in detecting co-infections in patients is necessary. Our results demonstrate that mNGS provides a new opportunity to investigate the pathogens associated with AIDS-associated pulmonary infections.

Materials and methods

Patients and culture-based assays

A total of 107 individuals with pulmonary infection, aged 19–97 years, participated in this study. Of these, 75 patients with AIDS were recruited from Nanning Fourth People’s Hospital, and 32 participants without AIDS were recruited from Nanning First People’s Hospital, China. AIDS was diagnosed on the basis of the diagnostic criteria and principles of HIV/AIDS (National Center for AIDS/STD Control and Prevention, China Centers for CDC and Prevention). AIDS with a CD4 + T lymphocyte count < 200 /µL may be accompanied by an unexplained irregular fever at 38 °C lasting for more than 1 month, diarrhea for 1 month (i.e., bowel movement more than three times per day), weight loss of more than 10% within 6 months, and typical bacterial, fungal, and viral infections. Prior to the study, all eligible participants and their legal guardians received a full explanation of the study and provided written informed consent. Bronchoscopy-guided puncture sputum and bronchoalveolar lavage fluid samples with suspected pulmonary infections were collected at Nanning Fourth People’s Hospital between January 2020 and May 2021. The samples were promptly stored in sterile containers, sent to the microbiology laboratory, and placed at -80 °C before performing culture-based assays and mNGS-related analysis.

Further bacterial and fungal culture-based assays were performed according to the standard procedure of microbial identification using culture-based assays v4.0 (JYK -SOP-WSW-014, Nanning Fourth People’s Hospital, Guangxi, China). Bronchoalveolar lavage fluid and sputum samples were added to an appropriate volume of culture broth and incubated at 30 °C for 12 h. Subsequently, 100 µL of each dilution (100 and 1,000 colony forming units [CFU]/mL) of each culture was plated on MacConkey agar (Oxoid Ltd., Basingstoke, United Kingdom), sheep blood agar (Oxoid Ltd., Basingstoke, United Kingdom), and chocolate agar (Difco, BD, Le Pont de Claix, France). After incubation at 35 °C for 16 h, the number of CFUs in each plate was counted and recorded using a Synbiosis Protocol 3 automatic colony counter (Synbiosis, Frederick, MD, USA). Individual colonies were purified thrice using the corresponding selective plates. All preliminary detected individual colonies were identified and classified using gram stain, catalase tests, and oxidase tests (Oxoid Ltd., Basingstoke, United Kingdom) and the VITEK-2 COMPACT system (BioMérieux, Marcy-l’Etoile, France) according to the manufacturer’s instructions. Mycobacterium spp. were detected using the BacT/ALERT 3D system (BioMérieux, Marcy-l’Etoile, France) and Xpert MTB (Cepheid, Sunnyvale, CA, USA), according to the manufacturers’ instructions. Bronchoalveolar lavage fluid and sputum samples were added to an appropriate volume of Sabouraud agar (pH 4.0–6.0, BioMérieux, Marcy-l’Etoile, France) and incubated at 28 °C for 7 days. If a typical colony grew, suspected colonies of the pathogens were subcultured on CHROMagar Candida agar (CHROMagar, France). Presumptive identification of clinically relevant Candida spp. was performed according to colony color and morphological differences. Individual white colonies were further classified using the ID 32 Fungus test kits (BioMérieux, Marcy-l’Etoile, France). Suspected cases of Pneumocystis spp. were diagnosed using Giemsa-stained smears, methylene blue staining, and Gomori methenamine silver staining.

Preparation of genomic DNA

Prior to DNA extraction, an equal volume of sodium hydroxide (NaOH) (40 mg/mL) was added to the sputum, thoroughly mixed, and incubated at 25 °C for 30 min. The mixture was then centrifuged at 25 °C for 10 min at 13,000 rpm, and the pellet was washed twice with physiological saline solution. In contrast, we mixed 0.5 mL of bronchoalveolar lavage fluid with 1 g of 0.5 mm glass beads (TUOTAINUO, Shenzhen, China) in a 1.5 mL centrifuge tube (KIRGEN, Shanghai, China) that was placed on the horizontal plate of the vortex (Kylin-Bell Vortex-5, Haimen, China) at 2,800–3,200 rpm for 30 min. After that, we separated 0.5 mL of bronchoalveolar lavage fluid into a new 1.5 mL centrifugal tube. DNA was extracted from the pellet and tissue using a TIANamp Micro DNA Kit (TIANGEN BIOTECH, China) according to the manufacturer’s instructions. Finally, the DNA was stored at -20 °C before sequencing.

Identification of bacterial pathogens using the mNGS approach

Whole mNGS was performed using an Illumina NextSeq 550 DX sequencer (Illumina, San Diego, CA, USA) according to the manufacturer’s protocol. Briefly, fragmentation of the extracted DNA was performed to generate 300 bp fragments using Covaris LE220 (Covaris, Inc., Woburn, MA, USA). Subsequently, the DNA fragments were subjected to end-repair, phosphorylation, A-tailing reactions, and Illumina adapter ligation. Subsequently, DNA libraries were constructed using the Nextera XT Library Construction Kit (Illumina, San Diego, CA, USA) and cleaned using a Magnetic Beads Kit (MAGEN Guangzhou, CHINA). The library size was analyzed using the Agilent 2100 Bioanalyzer system (Agilent Technologies, Santa Clara, CA, USA), and accurate quantification was performed by qualitative PCR using a Bio-Rad CFX96 PCR System (Bio-Rad Laboratories, Hercules, CA, USA). Finally, the qualified libraries were sequenced using Nextseq 550 platform (Illumina, San Diego, CA, USA) for 75 bp single-end reads, and raw sequencing files were uploaded for further analysis.

Upon generation of the sequencing results, FastQC ( was used for quality control of the raw data from sequencing by trimming low-quality reads and Illumina adapters. Subsequently, host reads were excluded after alignment of the human genome (Homo sapiens version: GRCh38) using Burrows-Wheeler Alignment ( [17]. The remaining non-human sequence data were mapped and classified into the unique pathogenic microorganism database of Guangzhou Sagene Biology for bacterial, viral, fungal, and parasitic species. This database, carrying medical microbiological information, obtains references from the NCBI database (fp://ncbi, the Ensemble database (, Virus Resource database (Virus get Variation Resource), JGI Fungi Porta ( and other authoritative microorganism databases. A microbial standard was used for verification and optimization. The depth and coverage of each species were calculated using Soap Coverage software ( The base number was calculated using the formula “read number * read length,” after which a clinical test report was generated.

Statistical analyses

Comparisons on clinical characteristics between two groups were conducted according to one-way analysis of variance (ANOVA). A k-mer-based metagenomic taxonomic classifier named centrifuge was used to identify the pathogens in each sample [18]. The mNGS results were considered as microorganisms with pulmonary pathogenicity if one of the following criteria were met [45]. A detected single species of pathogenic bacteria, virus or fungi were considered positive if at least 10 unique reads were aligned to the reference genome. mNGS reported a bacterium, virus and parasite whose unique reads at species level scored at least 10-fold greater than that of any other microbes, and reported a fungi whose unique reads at species level scored at least 5-fold greater than that of any other fungus. Although only one unique read was aligned to the reference genome of Mycobacterium tuberculosis using mNGS, the results of clinical nucleic acid tests were positive [6, 19, 20]. Mycobacterium avium, e and Mycobacterium abscessus were defined as positive when the number of unique reads (at the genus or species level) was in the top 10 of the reported bacterial list. Microbes identified using mNGS were classified as confirmed pathogens if they were positive in any clinical test (including culture, smear, or Xpert). Microbes were considered potential pathogens if they were identified as positive infectious pathogens using only mNGS but not any other clinical test. All other microbes identified using mNGS were considered uncertain pathogens, but did not meet the criteria for a positive mNGS result. The detection rates of all the pathogens were verified and reported.


Comparison of microbiota detected from culture-based assays and mNGS

Based on the inclusion and exclusion criteria, 75 patients with AIDS and 32 patients without AIDS were recruited. The demographic and clinical characteristics of both the groups are presented in Table 1. There were no significant differences in gender proportion (P value = 0.069, Chi-square test). However, the level of lymphocyte, neutrophil, white blood cell, C-reactive protein, and procalcitonin in the group of patients without AIDS were significant higher than those of the AIDS group (P value < 0.05, one-way ANOVA).

Table 1 Clinical characteristics (age, gender, Lymphocyte, Neutrophil, White blood cell, C-reactive protein and Procalcitonin) of enrolled AIDS patients and non-AIDS participants

The bronchoalveolar lavage fluids of all participants were clinically tested using culture and Giemsa-stained smear screening. Table 2 includes the data for the detection of bacteria, fungi, and viruses from the samples. Among them, the positive detection rates of the culture-based diagnostic approach and Giemsa-stained smear screening were obtained from 22 (29.3%) and 6 (8.0%) patients, respectively. According to the bacteriological results, Mycobacterium spp. (n = 10, 13.3%), Haemophilus influenzae (n = 2, 2.7%), Escherichia coli (n = 1, 1.3%), Klebsiella pneumoniae (n = 1, 1.3%), Rhodococcus equi (n = 1, 1.3%), and gram-negative bacteria (n = 3, 4.0%) were culture-positive on the bacterial growth plates mentioned above. One of them (sample ID: B2102820_AA) was simultaneously infected with both E. coli and Mycobacterium tuberculosis. In contrast, two (2.7%) cases of Talaromyces marneffei, two (2.7%) Aspergillus spp., and one (1.3%) of each of Candida albicans, Candida glabrata, and Candida krusei were detected on CHROMagar Candida agar. One of these samples (sample ID: B2102121_AA) was co-infected with both C. glabrata and C. krusei. In addition, six cases of Pneumocystis spp. were reported to be positive by Giemsa-stained smear screening.

Table 2 Distribution of diagnoses based on culture, Xpert and mNGS confirmation of species identity of isolates

We evaluated the detection performance of the traditional culture-independent screening and mNGS. Notably, 15 patients with AIDS showed concordant results between culture and mNGS, whereas 5 showed discordant results. Concordance between mNGS and culture for the detection of Mycobacterium spp. (5 cases) was found, including 3 cases (B2106316_AA, B2100621_AA and B2006273_AA) of Mycobacterium tuberculosis, 1 case (B2101349_AA) of Mycobacterium kansasii and 1 case (B2102822_AA) of Mycobacterium avium. However, discordant results between culture-positive and mNGS-negative samples (sample ID: BH211505_AA, B2100619_AA, B2102820_AA, B2006130_AA, and B2006023_AA) of Mycobacterium tuberculosis. were strongly associated with limitations in sequencing coverage. Moreover, consistent results for the identification of Haemophilus haemophilus (sample ID: B2103830_AA and B2101782_AA), E. coli (sample ID: B2102820_AA), K. pneumoniae (sample ID: B2102821_AA), and R. equi (sample ID: B2100396_AA) between the two detection methods were also found in our study. Furthermore, the positive results for T. marneffei (samples ID: BH211504_AA and B2102995_AA), Candida spp. (samples - ID: B2103362_AA and B2102121_AA), and Aspergillus fumigatus (sample ID: B2100395_AA) were validated by both culture and mNGS. Regarding the detection of P. jirovecii, only one case showed concordant results between Giemsa-stained smear screening and mNGS. The other five Giemsa-stained smears that screened positive remained negative on mNGS. Pathogens (Mycobacterium tuberculosis, Haemophilus influenzae, e, Escherichia coli, eeMycobacterium avium, Helicobacter pylori, Candida glabrata, Mycobacterium kansasii, Aspergillus fumigatus and Rhodococcus equi) eboth mNGS and clinical tests (including culture, smear, or Xpert) were considered as confirmed microorganisms with pulmonary pathogenicity (Table 2).

In our investigation, both gram-negative and gram-positive bacteria took 3–5 days to culture and classify before report generation. Furthermore, identification of Mycobacterium spp. and fungi took more than 14 days. In contrast, the total process of mNGS-based detection, including nucleic acid extraction, library construction, high-throughput sequencing, bioinformatics analysis, and reporting, required only 16 h (Fig. 1).

Fig. 1
figure 1

Technical roadmap and flow diagram for pathogenic detection based on mNGS

Notably, many other potential pathogens of lung infections in 75 patients with AIDS were identified using mNGS, which provided a wide range of profiles of pathogenic microorganisms. Based on annotations obtained from the pathogenic microorganism database, the percentage of mNGS-positive results was 70/75 (93.3%). There were 47 (62.7%) bacterial infections, 34 (45.3%) fungal infections, and 57 (76.0%) viral infections. The positive detection rates for gram-negative and gram-positive bacteria were 23/75 (30.7%) and 38/75 (50.7%), respectively. In addition, the detection rate of multipathogenic infections (at least three types of pathogenic microorganisms) was 45/75 (60.0%). When analyzed using mNGS, Human gammaherpesvirus 4 (n = 39, 52.0%) was the most prevalent pathogen found in patients with AIDS and lung infection, followed by Streptococcus pneumoniae (n = 27, 36.0%), Human cytomegalovirus (n = 25, 33.3%), P. jirovecii (n = 22, 29.3%), Penicillium marneffei (n = 12, 16.0%), H. influenzae (n = 9, 12.0%), Streptococcus mitis (n = 8, 10.7%), K. pneumoniae (n = 7, 9.3%), Tropheryma whipplei (n = 6, 8.0%), and Mycobacterium spp. (n = 5, 6.7%). For AIDS patients, Human gammaherpesvirus 4, Streptococcus pneumoniae, Penicillium marneffei and Human cytomegalovirus were typical potential pathogens. The presence of Streptococcus angina, Cryptococcus neoformans, human parvovirus B19, human herpesvirus 8, Fusobacterium nucleatum, Ureaplasma urealyticum, Actinomyces grivenis, and Mycoplasma pneumoniae detected using mNGS in only a single patient was considered false positives because of the low level of reads and lack of validation.

Characteristics of pathogens spectrum in patients with AIDS revealed by mNGS

We compared the differences in the profiles of pathogenic microorganisms between 75 patients with AIDS and 32 patients without AIDS. As shown in Fig. 2, we identified 46 types of pathogens in patients with AIDS, whereas 37 were identified in those without AIDS. When further analysis was performed, 18 common pathogen members were consistently detected in both AIDS and non-AIDS groups. We also compared positive detection rates between the patients with AIDS and the 32 without AIDS. As shown in Fig. 3, Stenotrophomonas maltophilia (n = 9, 28.1%) was the most prevalent pathogen in patients without AIDS, followed by Enterococcus faecium (n = 8, 25.0%), Pseudomonas aeruginosa (n = 6, 18.8%), Helicobacter gammaherpesvirus 4 (n = 5, 15.6%), Acinetobacter baumannii (n = 4, 12.5%), Corynebacterium striatum (n = 4, 12.5%), and Aspergillus fumigatus (n = 4, 12.5%). When compared with the mNGS-positive results in patients with AIDS, the detection rates of H. gammaherpesvirus 4, S. pneumoniae, Human cytomegalovirus, and P. jirovecii were approximately decreased 36.4%, 29.8%, 27.1%, and 26.2%, respectively.

Fig. 2
figure 2

Venn diagram of shared and specific genera. N means without AIDS group. P means AIDS group

Fig. 3
figure 3

Differences in the profiles of pathogenic microorganisms between 75 patients with AIDS and 32 without AIDS. N means without AIDS group. P means AIDS group. The detection rate of each pathogenic microorganism with pulmonary pathogenicity was calculated


Among the microorganisms that cause AIDS-related opportunistic infections, R. equi [21,22,23], S. pneumoniae [24,25,26], E. coli [27], Mycobacterium spp. [28], Candida spp. [29, 30], P. marneffei [31, 32], Aspergillus spp. [33], H. gammaherpesvirus 4 [34], hepatitis B [35], and hepatitis C [35] are frequently identified. When AIDS-related pulmonary infection occurs, the delay in diagnosis may impede precise therapy and block management, leading to poor prognosis and increased mortality and morbidity. Thus, infectious diseases should be accurately diagnosed to guide proper follow-up therapy and prevent cross-infections. Currently, the most common microbial diagnoses for pathogens in specimens are through culture (a gold standard in microbiology for the analysis of microbiota), histopathology, and smear microscopy [36, 37]. In this study, we found that only 22 patients with AIDS were infected with at least one species via culture-based identification, such as Mycobacterium spp., H. influenza, E. coli, K. pneumoniae, and R. equi. We also validated that the spectrum of pulmonary infection-related pathogens in patients with AIDS was wider than that in patients without AIDS (Fig. 2). Generally, there is a progressive decline in CD4+ levels in AIDS, which causes a decrease in both humoral and cell-mediated immunity. CD4+ cells release cytokines that contribute to the activation of antigen-presenting, phagocytic, natural killer, and cytotoxic T cells against pathogenic infections. They also support the conversion of B lymphocytes into long-lived plasma and memory B cells. Ultimately, the decline in CD4 + levels predisposes patients with AIDS to a higher risk of opportunistic infections due to pathogens [38,39,40,41]. Therefore, the differences between patients with AIDS and the control group were reasonable, indicating that mNGS detection may be more helpful than traditional methods in detecting a wide range of AIDS-related pathogens or even multiple infections. Furthermore, it was shown that mNGS carried a higher diagnostic positivity rate than bacterial (P value < 0.0001, Chi-square test) and fungal (P value < 0.0001, Chi-square test) culturing methods. Because mNGS was able to detect pathogens from patients’ tissue displaying potential infection and colonization, while culture results remained negative. Coincidentally, Wang et al. reported that mNGS-based detection of mixed pulmonary infections increased the sensitivity rate by approximately 83.3% compared with that of culturing (97.2% vs. 13.9%; P < 0.01) [42].

Theoretically, the high sensitivity of mNGS, which is strongly affected by the sequencing depth, could overcome the dilemma of low levels of microbial sequences. Therefore, more potential pathogens were detected and classified, which made an excellent contribution to the diagnosis of multiple infections and development of appropriate treatment strategies. An increasing number of case reports have revealed that mNGS has been successfully used as a diagnostic tool for infectious diseases caused by Streptococcus suis [43], P. jirovecii [44, 45], cytomegalovirus [46], and T. whipplei [45] in patients with AIDS.

However, mNGS may not be able to identify some specific culturable pathogens. In our study, mNGS missed Mycobacterium (four cases) and Aspergillus spp. (one case) and performed poorly in the detection of Mycoplasma spp. (two cases), U. urealyticum (two cases), and M. kansasii (one case) (Table 1). Among the six P. jirovecii-positive samples, five were not reported using mNGS. This is possibly due to the relatively low abundance of these pathogens within the samples or hindrance of the commensal microbiome in the respiratory tract. Because of the use of PCR-enriched sequences in library construction (Fig. 1), some sequences with less content may fail to be amplified, resulting in the loss of some information and missed detection of pathogens. The introduction of third-generation sequencing platforms such as Pacific BioSciences (PacBio) and Oxford Nanopore Technology (ONT) may provide the full structure of the microbial genome and reduce the sequencing bias caused by PCR amplification [47]. However, the high cost of third-generation sequencing has restricted its widespread application in clinical diagnosis.

Another limitation of mNGS is that while it identifies, it cannot discriminate the pathogenicity status. It could be confusing and even misleading for doctors reporting diagnoses and deciding clinical therapeutic strategies. Herein, 16 cases of AIDS patients carried a specie of confirmed pathogen with pulmonary pathogenicity diagnosed by mNGS and re-identified by the gold standard method. In addition, it lacks a standard criterion for explaining and reporting the number of reads, coverage of bases, sequencing depth, mNGS-positive or -negative results, the concentration of organisms, and functional differences, which strongly affect the clinical application of mNGS-based diagnosis [48]. Respiratory colonizing microorganisms (e.g. Staphylococcus aureus, e. aeruginosa, A. baumannii, Herpes simplex virus type 1, Human gammaherpesvirus 4 or Human cytomegalovirus) existed in equilibrium with the host by evading clearance of the immune system [49]. However, people acquired immunodeficiencies increased susceptibility to both invasive disease and higher density of colonizers, which were further transformed into pathogens and induced pulmonary disease development [50]. We must clarify that these 6 species in AIDS patients should be considered as potential pathogens and given priority attention even in cases with relatively low level of unique reads detected by mNGS. In contrast, cases with less than 50 unique reads in non-AIDS patients (without gold standard validation or clinical characterization) might be colonizers. Actually, oral streptococci, e, salivary streptococci, Prevotella spp., e and Rhodobacter spp. were common colonizing microorganisms of the human oral cavity, and confirmation of the causative agents of pulmonary infections was subject to exclusion of contamination during sampling and adequate clinical diagnosis [51]. Thus, the clinical application of mNGS does not mean that doctors do not need a conventional forensic pathological diagnosis in forensic medicine; but it could improve diagnostic efficiency and serve as a supplementary method to culture. However, the explanation and reporting of mNGS results can be confusing and unsubstantiated.

Furthermore, the relatively small sample size in the present study may have affected the accuracy of the mNGS assessment. For example, only 75 patients with AIDS from the same area participated in this investigation, and one microbiome sample was tested per individual. Generally, pathogenic diversity and composition fluctuate according to sex, area, time, and other factors. In addition, some of the detected pathogens were characterized by low-level reads of mNGS data. A set of probable false-positive results that stem from experimental errors should be avoided by future PCR or digital PCR validations. Moreover, analyses of serology, drug resistance, phylogenetic evolution, and other biological characteristics are lacking. This limits the extent to which we can statistically identify actual pathogens based on the differences in different detection technologies and the power of the correlational analyses between microbiota and functional differences.


In conclusion, mNGS has emerged as a promising tool for collecting the sequence information of various pathogenic microorganisms in an unbiased manner. Subsequent bioinformatics’ analyses of the microbiome can provide a reliable basis for accurate diagnosis, real-time monitoring, and effective therapy. The results of our study strongly support carried a less turnaround time and higher diagnostic positivity rate in the routine clinical diagnosis of patients with AIDS and unknown pulmonary pathogenic infection, especially when the results of conventional laboratory-based diagnostic screening are negative or the patients have multi-pathogenic infections. To achieve early diagnosis, standard operating procedures and management decisions must be established according to well-designed trials to eliminate or reduce biases between samples.

Data Availability

The datasets used and/or analysed during the current study are available in the NCBI Bioproject repository, [PRJNA891782].



Acquired immunodeficiency syndrome


metagenomic next-generation sequencing


Human immunodeficiency virus


Polymerase Chain Reaction


Pacific BioSciences


Oxford Nanopore Technology


  1. Klatt EC, Nichols L, Noguchi TT. Evolving trends revealed by autopsies of patients with the acquired immunodeficiency syndrome. 565 autopsies in adults with the acquired immunodeficiency syndrome, Los Angeles, Calif, 1982–1993 [corrected]. Arch Pathol Lab Med. 1994;118:884–90.

    CAS  PubMed  Google Scholar 

  2. Moore RD, Chaisson RE. Natural history of opportunistic disease in an HIV-infected urban clinical cohort. Ann Intern Med. 1996;124:633–42.

    Article  CAS  PubMed  Google Scholar 

  3. Afessa B, et al. Autopsy findings in HIV-infected inner-city patients. J Acquir Immune Defic Syndr. 1992;5:132–6.

    CAS  PubMed  Google Scholar 

  4. Lardinois B, Belkhir L, Verroken A. Helicobacter canis: a review of Microbiological and Clinical features. Front Microbiol. 2021;12:814944.

    Article  PubMed  Google Scholar 

  5. Testerman TL, McGee DJ, Mobley HL. Helicobacter pylori growth and urease detection in the chemically defined medium Ham’s F-12 nutrient mixture. J Clin Microbiol. 2001;39:3842–50.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Li H, et al. Detection of pulmonary infectious pathogens from lung biopsy tissues by Metagenomic Next-Generation sequencing. Front Cell Infect Microbiol. 2018;8:205.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Riess T, et al. Analysis of a novel insect cell culture medium-based growth medium for Bartonella species. Appl Environ Microbiol. 2008;74:5224–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Afshinnekoo E, et al. Precision Metagenomics: Rapid Metagenomic analyses for infectious Disease Diagnostics and Public Health Surveillance. J Biomol techniques: JBT. 2017;28:40–5.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Huang J, et al. Metagenomic next-generation sequencing versus Traditional Pathogen detection in the diagnosis of Peripheral Pulmonary Infectious Lesions. Infect drug Resist. 2020;13:567–76.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Wang Q, et al. Optimal specimen type for accurate diagnosis of infectious peripheral pulmonary lesions by mNGS. BMC Pulm Med. 2020;20:268.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Zambon JJ. Periodontal diseases: microbial factors. Annals of periodontology. 1996;1:879–925.

    Article  CAS  PubMed  Google Scholar 

  12. Akira S, Uematsu S, Takeuchi O. Pathogen recognition and innate immunity. Cell. 2006;124:783–801.

    Article  CAS  PubMed  Google Scholar 

  13. Takeuchi O, Akira S. Pattern recognition receptors and inflammation. Cell. 2010;140:805–20.

    Article  CAS  PubMed  Google Scholar 

  14. Fernandez-Real JM, et al. Burden of infection and insulin resistance in healthy middle-aged men. Diabetes Care. 2006;29:1058–64.

    Article  CAS  PubMed  Google Scholar 

  15. Iizasa H, Ishihara S, Richardo T, Kanehiro Y, Yoshiyama H. Dysbiotic infection in the stomach. World J Gastroenterol. 2015;21:11450–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Peace O, Rachakonda K, Kress M, Villalta F, Rachakonda G. Respiratory and neurological disease across different ethnic groups is influenced by the Microbiome. Microorganisms. 2021; 9.

  17. Abuin JM, Pichel JC, Pena TF, Amigo J. BigBWA: approaching the Burrows-Wheeler aligner to Big Data technologies. Bioinformatics. 2015;31:4003–5.

    Article  CAS  PubMed  Google Scholar 

  18. Watts GS, et al. Identification and quantitation of clinically relevant microbes in patient samples: comparison of three k-mer based classifiers for speed, accuracy, and sensitivity. PLoS Comput Biol. 2019;15:e1006863.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Chen Y et al. Application of Metagenomic Next-Generation sequencing in the diagnosis of pulmonary infectious pathogens from Bronchoalveolar Lavage samples. Frontiers in cellular and infection microbiology. 2021; 11:541092.

  20. Su SS, et al. Diagnostic performance of the metagenomic next-generation sequencing in lung biopsy tissues in patients suspected of having a local pulmonary infection. BMC Pulm Med. 2022;22:112.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Fiaccadori F, Elia GF, Calzetti C, Degli Antoni A, Magnani G. Rhodococcus equi infection in HIV-positive patients: report of 5 cases and literature overview. Monaldi archives for chest disease = Archivio Monaldi per le malattie del torace. 1994; 49: 380–8.

  22. Kwon KY, Colby TV. Rhodococcus equi pneumonia and pulmonary malakoplakia in acquired immunodeficiency syndrome. Pathologic features Archives of pathology & laboratory medicine. 1994;118:744–8.

    CAS  Google Scholar 

  23. Arlotti M, et al. Rhodococcus equi infection in HIV-positive subjects: a retrospective analysis of 24 cases. Scand J Infect Dis. 1996;28:463–7.

    Article  CAS  PubMed  Google Scholar 

  24. Feldman C, Anderson R. HIV-associated bacterial pneumonia. Clin Chest Med. 2013;34:205–16.

    Article  PubMed  Google Scholar 

  25. Cilloniz C, et al. Community-acquired lung respiratory infections in HIV-infected patients: microbial aetiology and outcome. Eur Respir J. 2014;43:1698–708.

    Article  CAS  PubMed  Google Scholar 

  26. Cilloniz C, et al. Community-Acquired Pneumococcal Pneumonia in Virologically suppressed HIV-Infected adult patients: a matched case-control study. Chest. 2017;152:295–303.

    Article  PubMed  Google Scholar 

  27. Rameshkumar MR, et al. Occurrence of extended-spectrum beta-lactamase, AmpC, and carbapenemase-producing genes in gram-negative bacterial isolates from human immunodeficiency virus infected patients. J Infect Public Health. 2021;14:1881–6.

    Article  PubMed  Google Scholar 

  28. Nakatani SM, et al. Efficient method for mycobacterial DNA extraction in blood cultures aids rapid PCR identification of Mycobacterium tuberculosis and Mycobacterium avium. Eur J Clin Microbiol Infect diseases: official publication Eur Soc Clin Microbiol. 2004;23:851–4.

    Article  CAS  Google Scholar 

  29. Odds FC. Candida infections in AIDS patients. Int J STD AIDS. 1992;3:157–60.

    Article  CAS  PubMed  Google Scholar 

  30. Clark-Ordonez I, Callejas-Negrete OA, Arechiga-Carvajal ET, Mourino-Perez RR. Candida species diversity and antifungal susceptibility patterns in oral samples of HIV/AIDS patients in Baja California, Mexico. Med Mycol. 2017;55:285–94.

    CAS  PubMed  Google Scholar 

  31. Le T, et al. AIDS-associated Penicillium marneffei infection of the central nervous system. Clin Infect diseases: official publication Infect Dis Soc Am. 2010;51:1458–62.

    Article  Google Scholar 

  32. Wong SY, Wong KF. Penicillium marneffei infection in AIDS. Pathol Res Int. 2011; 764293.

  33. Shi J, et al. Pathogenic microorganism detection in AIDS patients using bronchoalveolar lavage fluid. Int J Clin Exp Pathol. 2020;13:1727–32.

    PubMed  PubMed Central  Google Scholar 

  34. Freigassner P, Ardjomand N, Radner H, El-Shabrawi Y. Coinfection of the retina by Epstein-Barr virus and cytomegalovirus in an AIDS patient. Am J Ophthalmol. 2002;134:275–7.

    Article  PubMed  Google Scholar 

  35. Lodenyo H, Schoub B, Ally R, Kairu S, Segal I. Hepatitis B and C virus infections and liver function in AIDS patients at Chris Hani Baragwanath Hospital, Johannesburg. East Afr Med J. 2000;77:13–5.

    CAS  PubMed  Google Scholar 

  36. Menon LR, et al. Utility of bronchoalveolar lavage in the diagnosis of pulmonary infections in immunosuppressed patients. J Assoc Phys India. 2002;50:1110–4.

    CAS  Google Scholar 

  37. Miller SA, van Zante A, Schwartz BS. Cytologic evaluation can predict microbial culture results for infectious causes of pulmonary nodules in patients undergoing fine needle aspiration biopsy. Diagn Microbiol Infect Dis. 2010;68:330–3.

    Article  PubMed  Google Scholar 

  38. Alimonti JB, et al. Mechanisms of CD4 + T lymphocyte cell death in human immunodeficiency virus infection and AIDS. J Gen Virol. 2003;84(7):1649–61.

    Article  CAS  PubMed  Google Scholar 

  39. Achhra AC, et al. Relationship between CD4 cell count and serious long-term complications among HIV-positive individuals. Curr Opin HIV AIDS. 2014;9(1):63–71.

    Article  PubMed  Google Scholar 

  40. Ai JW, et al. Updates on the risk factors for latent tuberculosis reactivation and their managements. Emerg microbes infections. 2016;5:e10.

    CAS  Google Scholar 

  41. Ford N et al. The evolving role of CD4 cell counts in HIV care. Current opinion in HIV and AIDS. 2017; 12(2):123–8.

  42. Wang J, Han Y, Feng J. Metagenomic next-generation sequencing for mixed pulmonary infection diagnosis. BMC Pulm Med. 2019;19:252.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Xie Y, Ruan B, Lang G, Zhu B. Case Report: Streptococcus Suis Meningitis diagnosed in a HIV-Infected patient with cryptococcal meningitis using next-generation sequencing. Front Med. 2021;8:736064.

    Article  Google Scholar 

  44. Xie Y, Ruan B, Jin L, Zhu B. Case Report: next-generation sequencing in diagnosis of Pneumonia due to Pneumocystis jirovecii and Cytomegalovirus in a patient with HIV infection. Front Med. 2021;8:653294.

    Article  Google Scholar 

  45. Yan J, et al. Case Report: Tropheryma whipplei hide in an AIDS patient with Pneumocystis Pneumonia. Front public health. 2021;9:663093.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Ameur A, et al. Single-molecule sequencing: towards clinical applications. Trends Biotechnol. 2019;37(1):72–85.

    Article  CAS  PubMed  Google Scholar 

  47. Jacob JJ, et al. Metagenomic next-generation sequencing in clinical microbiology. Ind J Med Microbiol. 2019;37(2):133–40.

    Article  Google Scholar 

  48. Peng JM, et al. Metagenomic next-generation sequencing for the diagnosis of suspected pneumonia in immunocompromised patients. J Infect. 2021;82(4):22–7.

    Article  CAS  PubMed  Google Scholar 

  49. Hu B. Expert consensus on clinical application and interpretation of metagenomic next-generation sequencing in respiratory infections. Chinaeournal Clin Infectiouseease. 2022;15(2):90–102.

    Google Scholar 

  50. Siegel SJ, Weiser JN. Mechanisms of bacterial colonization of the respiratory tract. Annu Rev Microbiol. 2015;69:425–44.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Ptasiewicz M, et al. Armed to the Teeth-The oral mucosa immunity system and microbiota. Int J Mol Sci. 2022;23(2):e.

    Article  Google Scholar 

Download references


We would like to thank Jian Xie, and Zhenyi He for their suggestions.


This research was supported by the Science and Technology Planning Project of Nanning City (No. 20203052), and the Youth Science and Technology Talents Cultivation Project of Nanning City (No. RC20180103).

Author information

Authors and Affiliations



Writing-original draft preparation, Juan Zhong, Yanfen Liu and Na Luo; writing-review and editing, Hongmian Li and Jianyan Lin; methodology, Qiu Wei, Qisi Su and Jun Zou; resources, Xiaozhong Wu and Xianzhen Huang; the raw material Providing and validation, Juan Zhong and Yuting Jiang; data analysis, Lijuan Liang, Yanfen Liu and Na Luo.

Corresponding authors

Correspondence to Juan Zhong, Hongmian Li or Jianyan Lin.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhong, J., Liu, Y., Luo, N. et al. Metagenomic next-generation sequencing for rapid detection of pulmonary infection in patients with acquired immunodeficiency syndrome. Ann Clin Microbiol Antimicrob 22, 57 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: