Skip to main content
NCBI home page
BMC Infectious Diseases logoLink to BMC Infectious Diseases
. 2024 Oct 28;24:1210. doi: 10.1186/s12879-024-10009-8

Blood microbiota in HIV-infected and HIV-uninfected patients with suspected sepsis detected by metagenomic next-generation sequencing

Qianhui Chen 1,2,#, Zhong Chen 3,4,#, Yuting Tan 1,2,#, Songjie Wu 2,5, Shi Zou 1,2, Jie Liu 1,2, Shihui Song 1,2, Qian Du 1,2, Min Wang 3,4,✉, Ke Liang 1,2,5,6,✉
PMCID: PMC11520051  PMID: 39468445

Abstract

Background

Information on the comparison of blood microbiota between human immunodeficiency virus (HIV)-infected and HIV-uninfected patients with suspected sepsis by metagenomic next-generation sequencing (mNGS) is limited.

Methods

Retrospectively analysis was conducted in HIV-infected and HIV-uninfected patients with suspected sepsis at Changsha First Hospital (China) from March 2019 to August 2022. Patients who underwent blood mNGS testing were enrolled. The blood microbiota detected by mNGS were analyzed.

Results

A total of 233 patients with suspected sepsis who performed blood mNGS were recruited in this study, including 79 HIV-infected and 154 HIV-uninfected patients. Compared with HIV-uninfected patients, the proportions of mycobacterium (p = 0.001), fungus (p < 0.001) and viruses (p < 0.001) were significantly higher, while the proportion of bacteria (p = 0.001) was significantly lower in HIV-infected patients. The higher positive rates of non-tuberculous mycobacteriosis (NTM, p = 0.022), Pneumocystis jirovecii (P. jirovecii) (p = 0.014), Talaromyces marneffei (T. marneffei) (p < 0.001) and cytomegalovirus (CMV) (p < 0.001) were observed in HIV-infected patients, compared with HIV-uninfected patients. In addition, compared with HIV-uninfected patients, the constituent ratio of T. marneffei (p < 0.001) in the fungus spectrum were significantly higher, while the constituent ratios of Candida (p < 0.001) and Aspergillus (p = 0.001) were significantly lower in HIV-infected patients.

Conclusions

Significant differences in the blood microbiota profiles exist between HIV-infected and HIV-uninfected patients with suspected sepsis.

Keywords: Metagenomic next-generation sequencing, Human immunodeficiency virus, Suspected sepsis, Blood microbiota

Background

Bloodstream infection is a severe complication which can lead to sepsis especially in immunocompromised patients. It is critical for diagnosis and treatment as early as possible. Due to immunologic deficiency, human immunodeficiency virus (HIV) patients suffered mainly from various infections, including bacteria, fungus, and viruses [1]. At present, invasive bloodstream infection remains an important cause of morbidity and mortality in HIV-infected patients [1–4]. Two main factors influence the epidemiology of bloodstream infections in HIV-infected patients. One is the availability of antiretroviral therapy (ART) treatment, which determines the reduced incidence and the change of clinical characteristics of bloodstream infections; the other is the geographic distribution of pathogens [3, 5]. Therefore, the disease spectrum of HIV-infected patients differs in many ways from that of HIV-uninfected individuals. A better understanding of the causative organisms could help guide clinicians to choose appropriate antimicrobial treatments and avoid using unnecessary antibiotics. However, the information on the comparison of blood microbiota between HIV-infected and HIV-uninfected patients with bloodstream infections is limited.

A comprehensive diagnostic evaluation for infected pathogens is costly and complex, and many potential bloodstream infections often could not be identified using conventional microbiological tests, including bacterial cultures, serologies, and antigen testing [6]. Metagenomic next-generation sequencing (mNGS), a high-throughput sequencing technique, is suitable for the comprehensive analysis of nucleic acids within the pathogenic specimens [7]. In the field of clinical microbiology, mNGS is considered to have the potential to obtain entire genomic information of all microorganisms, including bacteria, viruses, fungi, and parasites from a clinical sample [8, 9]. Recently, mNGS has been reported as a clinical application for diagnosing various diseases, such as the causative pathogens of bloodstream infections [10], acute encephalitis/encephalopathy [11], acute liver failure [12], severe respiratory failure [13] and acute myocarditis [14]. Therefore, in this study, we aimed to use mNGS analysis for comparison of potential microbiota in blood samples from HIV-infected and HIV-uninfected patients with suspected sepsis.

Methods

Study participants

Patients with suspected sepsis who performed mNGS test of blood samples were retrospectively reviewed from the first hospital of Changsha between March 2019 to August 2022. The exclusion criteria were as follows: (1) unknown HIV infection status; (2) age < 18 years old; (3) non-infectious causes of fever. The non-infectious causes of fever included non-infectious inflammatory diseases, malignancies and miscellaneous causes. The inclusion criterium was patients diagnosed with SEPSIS-3 criteria with reference to the international sepsis diagnostic criteria [15]. The diagnosis was determined by the consensus of two experienced senior clinicians.

Data collection

Clinical data were retrospectively obtained from medical records. These following data were extracted: age, sex, HIV infection status, date of ART initiation, date of blood sample collection, antibiotic use within 3 months, immunosuppressive therapy (glucocorticoids and immunosuppressants) within 3 months, HIV viral load, lymphocyte count, CD4+ T lymphocyte count (CD4 count). In addition, the blood mNGS was performed for all HIV-infected and non-HIV-infected septic patients at the time of hospitalization. We collected results of mNGS for analysis.

Blood sample processing and DNA extraction

For each participant enrolled, 3–4 ml whole blood samples were centrifuged at 1600 × g for 15 min to harvest the plasma. Then, using a 0.2 ml aliquot of the harvested plasma sample, we extracted microbial cell-free DNA using the TIANamp Micro DNA Kit (DP316, TIANGEN BIOTECH) according to the manufacturer’s instructions.

Sequencing and data processing

The protocol was performed according to the report as previously described [16]. The MGIEasy Cell-free DNA Library Prep Kit (MGI Technology, Shenzhen, China) was used to construct the library according to the manufacturer’s instructions, including DNA-fragmentation, end-repair, adapter- ligation and PCR amplification. Library quality was determined using an Agilent 2100 TapeStation (Agilent, Santa Clara, CA, USA). High-throughput sequencing was performed using BGISEQ-50 (MGI Technology, Shenzhen, China), according to the manufacturer’s manual. High-quality sequencing data were generated by removing the host and low-quality sequences from the metagenomic data. Then Burrows-Wheeler Alignment software was used to align the high-quality reads to the human reference genome (version hg19) to remove contamination by human sequences. The remaining clean reads was blasted to Pathogens metagenomics Database, consisting of bacteria, fungi, viruses and parasites. The classification reference databases were downloaded from NCBI (ftp://ftp.ncbi.nlm.nih.gov/genomes/). RefSeq contains 4945 whole genome sequence of viral taxa, 6350 bacteral genomes or scaffolds, 1064 fungi related to human infection, and 234 parasites associated with human diseases.

Criteria for positive mNGS results

The principle of positive mNGS results were defined as previously described [16]: (1) The number of strictly aligned sequences for bacteria (except for Mycobacterium tuberculosis), fungi, and viruses is greater than 3; (2) The number of strictly aligned sequences for parasites is greater than 100; (3) For Mycobacterium tuberculosis (MTB), at least 1 read mapped to the species or genus level; for Non-tuberculous mycobacteria (NTM), the reads mapped to the species or genus level rank in the top 10 in the bacteria list. The microbial abundance results were interpreted as viral, bacterial, fungal, or protozoan pathogens.

Statistical analysis

Statistical analyses and plotting were performed by SPSS 21.0 and Graphpad Prism 5.0. Continuous variables were denoted as medians with the 25th to 75th interquartile range (IQR) and categorical variables were denoted as proportion (%). The normal distribution of the measurement data was examined by Kolmogorov-Smirnov test. Continuous variables meeting the normal distribution were analyzed by group t-test, and otherwise by the non-parametric rank sum test. Chi-square test or Kruskal-Wallis rank sum test was used to compare the count data between the groups. p values < 0.05 were considered statistically significant.

Results

Patients’ characteristics

In total, 249 patients with suspected sepsis and performed blood mNGS test were recruited in our study. Of them, 12 had unknown HIV infection status; 4 were under 18 years old. Finally, a total of 233 patients were enrolled in this study.

Of 233 enrolled patients, 79 were HIV-infected and 154 were HIV-uninfected patients. The characteristics of HIV-infected and HIV-uninfected patients with suspected bloodstream infection were showed in Table 1. The ages in HIV-infected group were younger than in HIV-uninfected group (p < 0.001). The proportion of male and lymphocyte counts among HIV-infected patients were significantly lower (p < 0.001; p = 0.002) than that among HIV-uninfected patients. In addition, in HIV-infected group, 48.1% of the patients received ART treatment. And the medians of HIV viral load and CD4 count were 190,000 copies/ml and 28 cells/µl respectively.

Table 1.

Characteristics of HIV-infected and uninfected patients with suspected sepsis

HIV-infected patients
(n = 79)		 HIV-uninfected patients
(n = 154)		 p valune

Age [years, median (IQR)]

Male, n(%)

On ART, n(%)

38 (31 ~ 50)

70 (88.6)

38 (48.1)

67 (56 ~ 79)

98 (63.6)

/

< 0.001

< 0.001

/
NRTIS, n(%) 38 (48.1) / /
NNRTIS, n(%) 10 (12.7) / /
PIs, n(%) 4 (5.1) / /
INSTIs, n(%) 26 (32.9) / /
HIV-infection stage
III/IV, n(%) 79 (100) / /
hospitalizations within 1 month 0 / /
Antibiotic using within 3 months, n(%) 78 (98.7) 152 (98.7) 0.983

Immunosuppressive therapy use within

3 months, n(%)

 36 (45.5) 66 (42.8) 0.693
HIV viral load [copies/ml, median (IQR)] 190,000 (602 ~ 562,000) / /
Lymphocyte counts [/ul, median(IQR)] 550 (305 ~ 1002) 880 (462 ~ 1417) 0.002
CD4 count [/ul, median (IQR)] 28 (14 ~ 92) / /
Open in a new tab

nucleoside analogue reverse transcriptase inhibitors, NRTIs; non-nucleoside analogue reverse transcriptase inhibitors, NNRTIs; protease inhibitors, PIs; integrase strand transfer inhibitors, INSTIs

Bloodstream microbiota diversity detected by mNGS

As shown in Fig. 1A, viruses (69.62%) and fungus (41.77%) were the most common bloodstream microbiomes detected by blood mNGS among 79 HIV-infected patients with suspected sepsis, followed by mycobacterium (13.92%). While bacteria (29.22%), viruses (22.08%) were the most common bloodstream microbiomes detected by blood mNGS among 154 HIV-uninfected patients with suspected sepsis, followed by fungus (12.99%). Compared with HIV-uninfected patients, the proportions of fungus (p < 0.001), virus (p < 0.001) and mycobacterium (p = 0.001) were significantly higher in HIV-infected patients, while the proportion of bacteria was significantly lower in HIV-infected patients (p = 0.001). In addition, the proportion of atypical microorganism (including Rickett’s organism and Chlamydia) had no significant difference between the two groups.

Fig. 1.

Fig. 1

Open in a new tab

(A) Different types of microbiomes detected by mNGS of blood among HIV-infected and uninfected patients with suspected sepsis. (B) Comparison of blood microbiomes detected by mNGS between HIV-infected and HIV-uninfected patients with suspected sepsis; * P < 0.05; ** P < 0.01

Further comparison of bloodstream microbiomes spectrum detected by blood mNGS was showed in Fig. 1B. The proportions of NTM (p = 0.022), Pneumocystis jirovecii (P. jirovecii) (p = 0.014), Talaromyces marneffei (T. marneffei) (p < 0.001) and cytomegalovirus (CMV) (p < 0.001) among HIV-infected patients were significantly higher than that among HIV-uninfected patients. However, no significant difference in the positive proportion of various bacteria was found between HIV-uninfected and HIV-infected groups (all p > 0.05).

Bloodstream microbiota diversity detected by mNGS in HIV-infected patients without ART, HIV-infected patients with ART and HIV-uninfected patients

We further compared the bloodstream microbiota diversity in HIV-infected patients without ART and HIV-infected patients with ART. As shown in the Fig. 2A, viruses (68.29% versus 71.05%) and fungus (56.1% versus 26.32%) were the most common bloodstream microbiomes among both 41 HIV-infected patients without ART and 38 HIV-infected patients with ART. Compared with HIV-infected patients with ART, the proportions of fungus were significantly higher in HIV-infected patients without ART (p = 0.007) (Fig. 2A). While the proportion of bacteria, mycobacterium, virus and atypical microorganism had no significant difference between the two groups (all p > 0.05) (Fig. 2A). Moreover, no significant difference in the HIV positive proportion of various bacteria, mycobacterium, virus, fungus and atypical microorganism was found between HIV-infected patients with ART and HIV-infected patients without ART (all p > 0.05) (Fig. 2B).

Fig. 2.

Fig. 2

Open in a new tab

(A) Different types of microbiomes in blood among HIV-infected patients with ART and without ART. (B) Comparison of microbiomes in blood between HIV-infected patients with ART and without ART. (C) Different types of microbiomes in blood among HIV-infected patients without ART and HIV-uninfected patients. (D) Comparison of microbiomes in blood between HIV-infected patients without ART and HIV-uninfected patients. * P < 0.05; ** P < 0.01; *** P < 0.001

In addition, we compared microbiomes diversity of blood among HIV-infected patients without ART and HIV-uninfected patients. Compared with HIV-uninfected patients, the proportions of mycobacterium (p = 0.001), fungus (p < 0.001) and virus (p < 0.001) were significantly higher in HIV-infected patients without ART, while the proportion of bacteria was significantly lower in HIV-infected patients without ART (p = 0.011) (Fig. 2C). The proportion of atypical microorganism had no significant difference between the two groups (p > 0.05) (Fig. 2C). Furthermore, the proportions of NTM (p = 0.007), P. jirovecii (p = 0.003), T. marneffei (p < 0.001), CMV (p < 0.001) and EBV (p = 0.024) among HIV-infected patients without ART were significantly higher than that among HIV-uninfected patients (Fig. 2D). However, no significant difference in the positive proportion of various bacteria was found between HIV-uninfected patients and HIV-infected patients without ART (all p > 0.05) (Fig. 2D).

Constituent ratio of potential pathogenic bacteria and mycobacterium spectrum in bloodstream

As shown in Fig. 3A, Klebsiella pneumoniae (K. pneumoniae, 27.27%) and E. coli (27.27%) were more common in the bacteria spectrum detected by mNGS among HIV-infected patients with suspected sepsis. However, no significant difference in the constituent ratios of various detected bacteria was found between HIV-uninfected and HIV-infected patients (all p > 0.05).

Fig. 3.

Fig. 3

Open in a new tab

(A) Comparison of constituent ratios of bacteria detected by mNGS of blood between HIV-infected and HIV-uninfected patients; (B) Comparison of constituent ratios of Mycobacterium detected by mNGS of blood between HIV-infected and HIV-uninfected patients. MTB, Mycobacterium tuberculosis; NTM, Non-tuberculous mycobacteria

As shown in Fig. 3B, in the mycobacterium spectrum, MTB and Mycobacterium avium complex (Mac), one of NTM were detected by blood mNGS in HIV-infected patients. While only MTB was detected in HIV-uninfected patients. However, no significant difference of the constituent ratio detected by blood mNGS was found in the mycobacterium spectrum between HIV-infected and uninfected patients.

Constituent ratio of potential pathogenic fungus and virus spectrum in bloodstream

In addition, compared with HIV-uninfected patients, the constituent ratio of T. marneffei (p < 0.001) in the fungus spectrum was significantly higher, while the constituent ratios of Candida (p < 0.001) and Aspergillus (p = 0.001) were significantly lower in HIV-infected patients with suspected sepsis (Fig. 4A). Canidia albicans (C. albicans) is the major Candida isolates in both HIV-infected and uninfected patients with suspected sepsis. Aspergillus fumigatus is the major Aspergillus isolates in HIV uninfected patients with suspected sepsis. However, Aspergillus bloodstream infection was not detected in HIV-infected patients.

Fig. 4.

Fig. 4

Open in a new tab

(A) Comparison of constituent ratios of fungi detected by mNGS of blood between HIV-infected and HIV-uninfected patients; (B) Comparison of constituent ratios of virus detected by mNGS of blood between HIV-infected and HIV-uninfected patients. * P < 0.05; ** P < 0.01

As shown in Fig. 4B, the constituent ratio of CMV (p = 0.012) in the virus spectrum was significantly higher, while the constituent ratio of Human herpes virus 4 (EBV) (p = 0.007) was significantly lower in HIV-infected patients, compared with HIV-uninfected patients.

Discussion

Bloodstream infections are associated with the increased mortality rate, length of hospital stay and intensive care unit (ICU) admission rate [17]. HIV infection remains a cause of increased risk of bloodstream infections [1–4]. Understanding the blood microbiome of patients with bloodstream infections will be beneficial for early diagnosis and treatment. To our knowledge, this is the first study to provide comprehensive information on the blood microbiomes of HIV-infected patients with suspected sepsis by blood mNGS, and to explore the differences between the groups of HIV-infected and HIV-uninfected patients. We found significant differences of blood microbiomes profile between HIV-infected and uninfected patients with suspected sepsis. In addition, the higher positive rates of NTM, P. jirovecii, T. marneffei and CMV were observed in HIV-infected patients, compared with HIV-uninfected patients. Previous studies have found that compared with HIV-uninfected patients, opportunistic infections are more common in HIV-infected adult patients due to severe cellular immune dysfunction. Therefore, for attending physicians, multiple opportunistic infections should be actively sought during each septic episode in HIV-infected adult patients especially in patients with combined immunodeficiency.

Currently, bloodstream infection is a more frequent cause of ICU admission than P. jirovecii pneumonia in HIV-infected patients [18, 19]. However, few studies have systematically described the spectrum of bloodstream infections in HIV-infected patients. Studies have shown that HIV infection is associated with a higher prevalence of bloodstream infections, including mycobacteremia, bacteremia, and fungemia [20, 21]. A retrospective multicentre study aiming at analysing the characteristics of bloodstream infections in patients living with HIV between 2008 and 2015 showed that Gram-positive pathogens caused 44.5% of bloodstream infections, followed by Gram-negative, 40.3%, fungi, 10.9%, and mycobacteria, 4.2% [22]. But the diagnosis method was blood cultures for pathogen in this study. In our study, the results showed that virus (mainly CMV) and fungus (mainly T. marneffei and P. jirovecii) were the most common blood microbiomes among HIV-infected patients with suspected sepsis. It has been suggested that the sensitivity of mNGS to diagnose fungal infection was significantly higher than culture in blood and bronchoalveolar lavage fluid samples, and the specificity was similar to culture in HIV-infected patients [23]. In addition, the positive rates of mNGS for identifying mixed infections were significantly higher than that of routine microbiological examination. Aside from bacteria, multiple types of pathogens were detected by mNGS, especially for viruses (such as CMV and EBV) and fungi [23]. Therefore, the reason for the differences of pathogen spectrum between our study and previous research may be that mNGS greatly improves the detection rates of viruses and fungus among HIV-infected patients with suspected sepsis. Surprisingly, in our study, only 10.13% of bacteria was detected in HIV-infected group by blood mNGS, which was much lower than the ratio of bacteria among HIV-uninfected patients. This phenomenon may be due to the greater and stronger antimicrobial treatment of advanced HIV-infected patients with high-risk of mixed infections compared to HIV-uninfected patients. But further clinical cohort studies are still needed to confirm this possibility.

HIV-infected individuals with isolated different fungal species had higher mortality than HIV-uninfected individuals [24]. In our study, there was a significant proportion of fungal bloodstream infection. Interestingly, we observed that the higher positive rates of T. marneffei and P. jirovecii contributed to the significantly increased proportion of fungus, and the constituent ratio of T. marneffei in the fungus spectrum were significantly increased among HIV-infected patients, compared with HIV-uninfected patients. The proportion of T. marneffei (29.11%) in bloodstream infection among HIV-infected patients was much higher than that reported in southeastern Asia (2.7%) [25]. Notably, the blood culture was the main diagnostic method for bloodstream infection in the previous studies. However, the culture results may be negative because of the low fungal loads during the early stage of infection [23]. Given this, T. marneffei bloodstream infection may have been underestimated in previous practice and reports, and the increased proportion of T. marneffei in our study may be largely attributed to the high sensitivity of mNGS. In addition, study have shown that T. marneffei is an emerging pathogenic fungus in mainland China, especially the southern provinces [26]. We also found that T. marneffei is one of the prominent fungus of bloodstream infection among our study population. Accordingly, T. marneffei should be valued when treating HIV-infected patients with bloodstream infection in China, not just in the southern provinces.

Currently, few studies reported the epidemiology of Candida and Aspergillus in bloodstream infection. Surprisingly, in our study, the constituent ratios of Candida and Aspergillus in fungus spectrum among HIV-infected patients were significantly lower than that among HIV-uninfected patients. We also previously reported the similar results of fungus spectrum in BALF from HIV-infected and uninfected patients with pulmonary infection by BALF mNGS [16]. T cell immunity was severely defective in HIV-infected patients in our study. Therefore, the incidence of Candida infection is not completely correlated with CD4 count. Studies have shown that neutrophils can clear C. albicans infection through phagocytosis, degranulation, the production of reactive oxygen species, and the formation of neutrophil extracellular traps [27]. Thus, the activities of innate immune cells may be the important factors for the lower constituent ratio of Candida in HIV-infected patients. Aspergillus, also one of common fungus causing opportunistic infections, was not detected in HIV-infected patients in our study. One retrospective cross-sectional study also found that Aspergillus was not detected among the 229 cases of Chinese HIV/AIDS patients with bloodstream infection, which was consistent with our result [28]. These results suggested that Aspergillus infection was less common in HIV-infected patients with bloodstream infection, compared with HIV-uninfected patients. Previous studies have also shown that aspergillosis was an unusual reported opportunistic infectious disease in people living with HIV [29]. Of note, the median of CD4 count was 28 cells/µl in our study. Therefore, we considered no significant positive correlation between the incidence of Aspergillus in bloodstream infection and CD4 count. It is well documented that invasive aspergillosis is more commonly associated with the decrease in neutrophils and macrophages, rather than the depletion of the CD4 T-cell population [30, 31]. However, the pathogenesis of Candida and Aspergillus in bloodstream infection remains to be thoroughly investigated.

It is important to distinguish NTM from MTB, in order to provide appropriate treatment in time. The majority of mycobacteria infections are caused by MTB, but in patients with very low CD4 cell count (< 50 cells/ mL), NTM are also possible, especially in organisms of the Mac [32]. The challenge of differential diagnosis between MTB and NTB can be overcome by mNGS. Our results showed that the positive detection rate of NTM (5.06%), Mac, was significantly higher in HIV-infected patients than that in HIV-uninfected patients. NTM was not detected among HIV-uninfected patients in our study. In study performed in Uganda, nearly one-quarter of HIV-infected patients hospitalized with severe sepsis had MTB bacteremia. In this study, MTB was the commonest etiology of bloodstream infection (23.4% of cases), and NTM constituted 4% of isolates [33]. In addition, the proportion of MTB was higher than that from the USA and lower than in South Africa [28, 34]. Thus, the differences in ratio may be related to the sensitivity of detection methods and different levels of Mycobacterium epidemics among countries and regions.

Advances in mNGS have enabled the rapid identification of non-cultured pathogens [35]. In our study, viruse (mainly EBV and CMV) was one of the most common bloodstream microbiomes among both HIV-infected and HIV-uninfected patients with suspected sepsis. EBV and CMV are usually mild or completely asymptomatic in patients with normal immune function; however, when immunity decreases, both of them can escape the suppression of the immune system and lead to diseases [35]. Therefore, the clinical significance of positive blood mNGS detection for CMV and EBV needs to be further confirmed in combination with the clinical manifestations of patients.

One of a limitation of the study was that the data were collected retrospectively from a single hospital and the number of enrolled samples was small. Studies reported that microbiome can be altered by different types of sexual practices. But in our study, there were 70 males and 9 females in HIV-infected individuals. The number of men who have sex with men was 67. Because of limited sample size, it is difficult to assess the differences in blood microbiome composition of different of sexual practice. In addition, the detected microbiota by mNGS may not necessarily be the cause of sepsis. It is difficult to distinguish detected microbiota colonization from infection by mNGS alone, especially for opportunistic infectious pathogens. Because mNGS has no widely accepted quantitative cut-off or threshold values. Therefore, the cause of sepsis must be based on the comprehensive analysis of clinical features, laboratory abnormalities, radiologic findings and traditional microbiologic proofs rather than mNGS alone.

Nevertheless, we think that our study has some strengths. Firstly, the hospital in our study treats all HIV patients in the Changsha, as well as patients transferred from the surrounding area. Given that national data are not available yet, this may be the most reliable data to date. Moreover, it is the first study to provide comprehensive information on the blood microbiomes of HIV-infected and HIV-uninfected patients with suspected sepsis by blood mNGS.

Conclusion

We were able to use blood mNGS to detect pathogens in patients with suspected sepsis. Significant differences in the blood microbiota profiles including bacteria, mycobacterium, fungus and viruses exist between HIV-infected and HIV-uninfected patients, which could better guide clinicians in empirically choosing appropriate antimicrobial therapy.

Abbreviations

HIV

Human immunodeficiency virus

NTM

non-tuberculous mycobacteriosis; P. jirovecii: Pneumocystis jirovecii; T. marneffe: Talaromyces marneffei;

mNGS

Metagenomic next-generation sequencing

CMV

Cytomegalovirus

MTB

Mycobacterium tuberculosis

ART

Antiretroviral therapy

IQR

Interquartile range

Mac

Mycobacterium avium complex; K. pneumoniae: Klebsiella pneumoniae;

EBV

Human herpes virus 4

ICU

Intensive care unit

Author contributions

MW and KL conceived and designed this investigation. ZC, QD, JL and ZS collected the original data. QC, YT and SW analyzed the data. QC, ZC, YT, SS and KL contributed to the writing of the paper. All authors read and approved the final manuscript.

Funding

This work was supported by Medical Science and Technology Innovation Platform Support Project of Zhongnan Hospital, Wuhan University (PTXM2020008), Science and Technology Innovation Cultivation Fund of Zhongnan Hospital, Wuhan University (cxpy2017043). Medical Science Advancement Program (Basic Medical Sciences) of Wuhan University (TFJC2018004). Non-profit Central Research Institute Fund of Chinese Academy of Medical Sciences (2020-PT320-004).

Data availability

Data can’t be shared publicly due to ethical restrictions. Data analyzed in the manuscript will be made available by all the authors.

Declarations

Ethics approval and consent to participate

This study was approved by the institutional ethics committee of the first hospital of Changsha (202128). All subjects have voluntarily participated in the study and signed the informed consent form. The study was performed in accordance with the guidelines of the Declaration of Helsinki and relevant regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

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

Qianhui Chen, Zhong Chen, Yuting Tan these authors have contributed equally to this work and share first authorship.

Contributor Information

Min Wang, Email: wangmin2828@163.com.

Ke Liang, Email: keliang@whu.edu.cn.

References

  • 1.Paul S, Gilbert HM, Ziecheck W, Jacobs J, Sepkowitz KA. The impact of potent antiretroviral therapy on the characteristics of hospitalized patients with HIV infection. AIDS. 1999;13(3):415–8. 10.1097/00002030-199902250-00015. [DOI] [PubMed] [Google Scholar]
  • 2.Palella FJ Jr, Baker RK, Moorman AC, Chmiel JS, Wood KC, Brooks JT, Holmberg SD, HIV Outpatient Study Investigators. Mortality in the highly active antiretroviral therapy era: changing causes of death and disease in the HIV outpatient study. J Acquir Immune Defic Syndr. 2006;43(1):27–34. 10.1097/01.qai.0000233310.90484.16. [DOI] [PubMed] [Google Scholar]
  • 3.Huson MA, Stolp SM, van der Poll T, Grobusch MP. Community-acquired bacterial bloodstream infections in HIV-infected patients: a systematic review. Clin Infect Dis. 2014;58(1):79–92. 10.1093/cid/cit596. [DOI] [PubMed] [Google Scholar]
  • 4.Mayanja BN, Todd J, Hughes P, Van der Paal L, Mugisha JO, Atuhumuza E, Tabuga P, Maher D, Grosskurth H. Septicaemia in a population-based HIV clinical cohort in rural Uganda, 1996–2007: incidence, aetiology, antimicrobial drug resistance and impact of antiretroviral therapy. Trop Med Int Health. 2010;15(6):697–705. 10.1111/j.1365-3156.2010.02528.x. [DOI] [PubMed] [Google Scholar]
  • 5.Tumbarello M, Tacconelli E, Donati KG, Citton R, Leone F, Spanu T, Cauda R. HIV-associated bacteremia: how it has changed in the highly active antiretroviral therapy (HAART) era. J Acquir Immune Defic Syndr. 2000;23(2):145–51. 10.1097/00126334-200002010-00006. [DOI] [PubMed] [Google Scholar]
  • 6.Grundy BS, Parikh H, Jacob S, Banura P, Moore CC, Liu J, Houpt ER. Pathogen detection using metagenomic next-generation sequencing of plasma samples from patients with sepsis in Uganda. Microbiol Spectr. 2023;11(1):e0431222. 10.1128/spectrum.04312-22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chiu CY, Miller SA. Clinical metagenomics. Nat Rev Genet. 2019;20(6):341–55. 10.1038/s41576-019-0113-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Laudadio I, Fulci V, Palone F, Stronati L, Cucchiara S, Carissimi C. Quantitative Assessment of Shotgun Metagenomics and 16S rDNA amplicon sequencing in the study of human gut Microbiome. OMICS. 2018;22(4):248–54. 10.1089/omi.2018.0013. [DOI] [PubMed] [Google Scholar]
  • 9.Rausch P, Rühlemann M, Hermes BM, Doms S, Dagan T, Dierking K, Domin H, Fraune S, von Frieling J, Hentschel U, Heinsen FA, Höppner M, Jahn MT, Jaspers C, Kissoyan KAB, Langfeldt D, Rehman A, Reusch TBH, Roeder T, Schmitz RA, Schulenburg H, Soluch R, Sommer F, Stukenbrock E, Weiland-Bräuer N, Rosenstiel P, Franke A, Bosch T, Baines JF. Comparative analysis of amplicon and metagenomic sequencing methods reveals key features in the evolution of animal metaorganisms. Microbiome. 2019;7(1):133. 10.1186/s40168-019-0743-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Horiba K, Kawada JI, Okuno Y, Tetsuka N, Suzuki T, Ando S, Kamiya Y, Torii Y, Yagi T, Takahashi Y, Ito Y. Comprehensive detection of pathogens in immunocompromised children with bloodstream infections by next-generation sequencing. Sci Rep. 2018;8(1):3784. 10.1038/s41598-018-22133-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Kawada J, Okuno Y, Torii Y, Okada R, Hayano S, Ando S, Kamiya Y, Kojima S, Ito Y. Identification of viruses in cases of Pediatric Acute Encephalitis and Encephalopathy using next-generation sequencing. Sci Rep. 2016;6:33452. 10.1038/srep33452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Suzuki T, Kawada JI, Okuno Y, Hayano S, Horiba K, Torii Y, Takahashi Y, Umetsu S, Sogo T, Inui A, Ito Y. Comprehensive detection of viruses in pediatric patients with acute liver failure using next-generation sequencing. J Clin Virol. 2017;96:67–72. 10.1016/j.jcv.2017.10.001. [DOI] [PubMed] [Google Scholar]
  • 13.Takeuchi S, Kawada JI, Horiba K, Okuno Y, Okumura T, Suzuki T, Torii Y, Kawabe S, Wada S, Ikeyama T, Ito Y. Metagenomic analysis using next-generation sequencing of pathogens in bronchoalveolar lavage fluid from pediatric patients with respiratory failure. Sci Rep. 2019;9(1):12909. 10.1038/s41598-019-49372-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Takeuchi S, Kawada JI, Okuno Y, Horiba K, Suzuki T, Torii Y, Yasuda K, Numaguchi A, Kato T, Takahashi Y, Ito Y. Identification of potential pathogenic viruses in patients with acute myocarditis using next-generation sequencing. J Med Virol. 2018;90(12):1814–21. 10.1002/jmv.25263. [DOI] [PubMed] [Google Scholar]
  • 15.Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, Bellomo R, Bernard GR, Chiche JD, Coopersmith CM, Hotchkiss RS, Levy MM, Marshall JC, Martin GS, Opal SM, Rubenfeld GD, van der Poll T, Vincent JL, Angus DC. The Third International Consensus definitions for Sepsis and septic shock (Sepsis-3). JAMA. 2016;315(8):801–10. 10.1001/jama.2016.0287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Tan Y, Chen Z, Zeng Z, Wu S, Liu J, Zou S, Wang M, Liang K. Microbiomes detected by Bronchoalveolar Lavage Fluid Metagenomic Next-Generation sequencing among HIV-Infected and uninfected patients with pulmonary infection. Microbiol Spectr 2023 Jul 12:e0000523. 10.1128/spectrum.00005-23 [DOI] [PMC free article] [PubMed]
  • 17.Afessa B, Morales I, Weaver B. Bacteremia in hospitalized patients with human immunodeficiency virus: a prospective, cohort study. BMC Infect Dis. 2001;1:13. 10.1186/1471-2334-1-13. Epub 2001 Sep 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Chiang HH, Hung CC, Lee CM, Chen HY, Chen MY, Sheng WH, Hsieh SM, Sun HY, Ho CC, Yu CJ. Admissions to intensive care unit of HIV-infected patients in the era of highly active antiretroviral therapy: etiology and prognostic factors. Crit Care. 2011;15(4):R202. 10.1186/cc10419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Rosenberg AL, Seneff MG, Atiyeh L, Wagner R, Bojanowski L, Zimmerman JE. The importance of bacterial sepsis in intensive care unit patients with acquired immunodeficiency syndrome: implications for future care in the age of increasing antiretroviral resistance. Crit Care Med. 2001;29(3):548–56. 10.1097/00003246-200103000-00013. [DOI] [PubMed] [Google Scholar]
  • 20.Gordon MA, Walsh AL, Chaponda M, Soko D, Mbvwinji M, Molyneux ME, Gordon SB. Bacteraemia and mortality among adult medical admissions in Malawi–predominance of non-typhi salmonellae and Streptococcus pneumoniae. J Infect. 2001;42(1):44–9. 10.1053/jinf.2000.0779. [DOI] [PubMed] [Google Scholar]
  • 21.Reddy EA, Shaw AV, Crump JA. Community-acquired bloodstream infections in Africa: a systematic review and meta-analysis. Lancet Infect Dis. 2010;10(6):417–32. 10.1016/S1473-3099(10)70072-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Taramasso L, Liggieri F, Cenderello G, Bovis F, Giannini B, Mesini A, Giacomini M, Cassola G, Viscoli C, Di Biagio A. Bloodstream infections in patients living with HIV in the modern cART era. Sci Rep. 2019;9(1):5418. 10.1038/s41598-019-41829-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Mao Y, Shen H, Yang C, Jia Q, Li J, Chen Y, Hu J, Huang W. Clinical performance of metagenomic next-generation sequencing for the rapid diagnosis of talaromycosis in HIV-infected patients. Front Cell Infect Microbiol. 2022;12:962441. 10.3389/fcimb.2022.962441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Kiertiburanakul S, Watcharatipagorn S, Chongtrakool P, Santanirand P. Epidemiology of bloodstream infections and predictive factors of mortality among HIV-infected adult patients in Thailand in the era of highly active antiretroviral therapy. Jpn J Infect Dis. 2012;65(1):28–32. [PubMed] [Google Scholar]
  • 25.Varma JK, McCarthy KD, Tasaneeyapan T, Monkongdee P, Kimerling ME, Buntheoun E, Sculier D, Keo C, Phanuphak P, Teeratakulpisarn N, Udomsantisuk N, Dung NH, Lan NT, Yen NT, Cain KP. Bloodstream infections among HIV-infected outpatients, Southeast Asia. Emerg Infect Dis. 2010;16(10):1569–75. 10.3201/eid1610.091686. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Hu Y, Zhang J, Li X, Yang Y, Zhang Y, Ma J, Xi L. Penicillium marneffei infection: an emerging disease in mainland China. Mycopathologia. 2013;175(1–2):57–67. 10.1007/s11046-012-9577-0. [DOI] [PubMed] [Google Scholar]
  • 27.He Y, Liu J, Chen Y, Yan L, Wu J. Neutrophil Extracellular traps in Candida albicans infection. Front Immunol. 2022;13:913028. 10.3389/fimmu.2022.913028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Qi T, Zhang R, Shen Y, Liu L, Lowrie D, Song W, Chen J, Wang Z, Shen J, Cai R, Guan L, Luo B, Tang Y, Lu H. Etiology and clinical features of 229 cases of bloodstream infection among Chinese HIV/AIDS patients: a retrospective cross-sectional study. Eur J Clin Microbiol Infect Dis. 2016;35(11):1767–70. 10.1007/s10096-016-2724-7. [DOI] [PubMed] [Google Scholar]
  • 29.Holding KJ, Dworkin MS, Wan PC, Hanson DL, Klevens RM, Jones JL, Sullivan PS. Aspergillosis among people infected with human immunodeficiency virus: incidence and survival. Adult and adolescent spectrum of HIV Disease Project. Clin Infect Dis. 2000;31(5):1253–7. 10.1086/317452. [DOI] [PubMed] [Google Scholar]
  • 30.Kaur R, Mehra B, Dhakad MS, Goyal R, Dewan R. Pulmonary aspergillosis as opportunistic mycoses in a cohort of human immunodeficiency virus-infected patients: report from a tertiary care hospital in North India. Int J Health Sci (Qassim). 2017 Apr-Jun;11(2):45–50. [PMC free article] [PubMed]
  • 31.Denning DW, Morgan EF. Quantifying deaths from aspergillosis in HIV positive people. J Fungi (Basel). 2022;8(11):1131. 10.3390/jof8111131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Kaplan JE, Benson C, Holmes KK, Brooks JT, Pau A, Masur H. Centers for Disease Control and Prevention (CDC); National Institutes of Health; HIV Medicine Association of the Infectious Diseases Society of America. Guidelines for prevention and treatment of opportunistic infections in HIV-infected adults and adolescents: recommendations from CDC, the National Institutes of Health, and the HIV Medicine Association of the Infectious Diseases Society of America. MMWR Recomm Rep. 2009;58(RR–4):1–207. quiz CE1-4. [PubMed] [Google Scholar]
  • 33.Jacob ST, Pavlinac PB, Nakiyingi L, Banura P, Baeten JM, Morgan K, Magaret A, Manabe Y, Reynolds SJ, Liles WC, Wald A, Joloba ML, Mayanja-Kizza H, Scheld WM. Mycobacterium tuberculosis bacteremia in a cohort of hiv-infected patients hospitalized with severe sepsis in Uganda–high frequency, low clinical suspicion [corrected] and derivation of a clinical prediction score. PLoS ONE. 2013;8(8):e70305. 10.1371/journal.pone.0070305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.von Gottberg A, Sacks L, Machala S, Blumberg L. Utility of blood cultures and incidence of mycobacteremia in patients with suspected tuberculosis in a South African infectious disease referral hospital. Int J Tuberc Lung Dis. 2001;5(1):80–6. [PubMed] [Google Scholar]
  • 35.Yan G, Liu J, Chen W, Chen Y, Cheng Y, Tao J, Cai X, Zhou Y, Wang Y, Wang M, Lu G. Metagenomic next-generation sequencing of Bloodstream Microbial Cell-Free Nucleic Acid in Children with suspected Sepsis in Pediatric Intensive Care Unit. Front Cell Infect Microbiol. 2021;11:665226. 10.3389/fcimb.2021.665226. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data can’t be shared publicly due to ethical restrictions. Data analyzed in the manuscript will be made available by all the authors.

Articles from BMC Infectious Diseases are provided here courtesy of BMC

RESOURCES