Immunosuppressive therapy is extensively used in patients with nephropathy and connective tissue disease, but severe side effects such as immunodeficiency, severe infection, and steroid-induced diabetes can occur (Doria et al., 2006; Rimes-Stigare et al., 2015). When immunodeficiency patients develop a severe pulmonary infection, the timely diagnosis of pathogenic microorganisms is difficult (Thong and Chan, 2019). Sepsis, hypoxia, and other factors can lead to multiple organ dysfunction and finally result in acute respiratory distress syndrome (ARDS) (McGregor et al., 2015).

Veno-venous extracorporeal membrane oxygenation (vv-ECMO) is increasingly used as a life-support technique in critically ill patients with reversible refractory respiratory failure and some other situations, such as organ transplantation (Professional Committee for Extracorporeal Life Support of Chinese Medical Doctor Association, 2018; Guglin et al., 2019). In recent years, with a deeper understanding of ECMO and the rapid improvement of ECMO accessories and operation methods, the survival rate of severely ill patients has also increased compared with before (Oh Tak et al., 2021; Kochanek et al., 2022). However, new issues and side effects have emerged. Immune suppression, long-term bed rest, physical weakness, and many invasive procedures, including continuous renal replacement therapy (CRRT) and tracheal intubation, could increase the rate of patients’ infection (Kühn et al., 2020). Infections are common complications in vv-ECMO patients and are associated with an increase in mortality (Burket et al., 1999).

Due to the high infection rate of vv-ECMO patients with low immunity (Dioverti et al., 2016; Nureki et al., 2020), the complex composition of pathogens, and the emergence of new, unknown, and drug-resistant pathogens, the difficulty of anti-infective treatment is increasing, resulting in huge pressure on doctors and patients. Timely and correct anti-infection treatment is the key to the successful treatment of vv-ECMO-supported immunocompromised patients with acute respiratory failure. In the case that the empirical treatment is ineffective, the causative pathogen is unknown, and the conventional microbial detection method has no positive indication, it is urgent for clinicians to find the causative pathogen quickly and accurately, give sensitive antibiotics for precise treatment, and improve the poor prognosis. At present, the commonly used pathogenic detection methods of pneumonia, such as morphological examination (smear and transmission electron microscopy), isolation and culture, and immunological methods, have their own advantages (Mazloomirad et al., 2021). However, for severe infection cases, the traditional methods have limitations in terms of detection time, positive rate, and specificity. Next-generation sequencing (NGS) technology, also known as high-throughput sequencing technology, is marked by the ability to sequence hundreds of thousands to millions of DNA molecules in parallel and generally shorter read lengths (Zhang et al., 2019). Metagenomic NGS (mNGS) is not based on culture, but the DNA of all microorganisms is directly extracted from clinical specimens (Yue et al., 2021). mNGS technology is widely used to explore the pathogens of difficult infectious diseases (Zhang et al., 2020).

Our present study reviewed 46 immunocompromised patients with acute respiratory failure who were treated with vv-ECMO in the First Affiliated Hospital of Zhengzhou University and analyzed the value of mNGS technology in the etiological diagnosis of patients and its guiding value for treatment plans.

In the present study, we retrospectively analyzed the pathogen detection effect and clinical personalized antibiotic therapy value of mNGS technologies in a Chinese cohort of immunocompromised patients receiving vv-ECMO for acute respiratory failure.

Previous studies have found that infection is an important factor affecting the prognosis of ECMO patients (Steiner et al., 2001; Bif et al., 2017). The clinical diagnosis of infection in ECMO patients is a challenging topic because they always have signs of systemic inflammation, which may be triggered by the ECMO itself, and because the body temperature is controlled by the ECMO heat exchanger, fever is usually not obvious. Immunocompromised patients are often co-infected with multiple pathogens, and it is not easy to accurately detect mixed pathogens by routine clinical tests (Crotty et al., 2015). The immunocompromised patients assisted by vv-ECMO were often characterized as having high infection rates, mixed pathogen infection, poor prognosis, and even overuse of antibiotics. At present, the commonly used clinical methods to diagnose pathogens include G/GM experiments, bacterial culture, and quantitative PCR (qPCR) (Burket et al., 1999; Haneke et al., 2016). However, these methods usually have limitations. Conventional pathogen culture is currently the universal and necessary “gold standard” in clinical practice, but the culture time is long, and the positive rate is low. The test results of the G/GM experiment will be affected by many factors, its specificity and sensitivity are not high, and the incidence of false-positive and false-negative results is not low and can only be used for preliminary approximate discrimination of fungi. Another commonly used technique is qPCR, which is widely used but has low sensitivity. However, when the copy number of the virus is too low, it cannot be detected, early infection cannot be detected in time, and only specific pathogens can be detected. The common defect of these testing methods is that the number of specimens required for testing is large, and sufficient blood volume must be obtained. Therefore, there is a need for a clinical detection method that is different from the traditional one, which can detect pathogens early and accurately and avoid uncontrolled infections that exacerbate disease progression and seriously affect the prognosis of critically ill patients during ECMO.

NGS technology has the advantages of high detection sensitivity, short detection cycles, and a wide application range. This technology does not require the isolation and culture of pathogens, reduces the false-negative rate, and can analyze multiple pathogens at the same time, which greatly advances the treatment of specific pathogens. In summary, NGS technology not only solves many problems in traditional detection methods but also the pathogen detection rate of this technology is significantly better than that of traditional detection methods such as qPCR (Brown et al., 2018; Carpenter et al., 2019). A recent study found that in immunocompromised children, half of BALF samples previously detected as negative by routine test have potential pathogens detected by mNGS test (Zinter et al., 2019). Theoretically, NGS technology can detect the presence of pathogens, even all nucleotide sequences, which broadens the diagnostic scope of infections. That is, NGS technology has an absolute advantage in the diagnosis of mixed pathogen infection, especially in immunocompromised patients (Wang et al., 2019). Consistent with the above findings, our findings also showed that the mNGS test is more effective than a routine test and can detect multiple pathogenic microorganisms in mixed infections. In addition, K. pneumoniae and A. baumannii were the most popular pathogens detected in immunocompromised patients assisted by vv-ECMO in our present study. K. pneumoniae is a common opportunistic pathogen in ICUs. In the present study, K. pneumoniae was believed to be a pathogenic bacteria, for the following reasons: firstly, the microorganisms we detected are not common in the laboratory, and laboratory contamination bacteria are excluded; secondly, the sequenced microorganisms can cover different regions in the genome; finally, the microorganism is potentially pathogenic in a given clinical setting in a specific patient.

Immunocompromised patients assisted by vv-ECMO often have long hospital stays, heavy financial burdens, and high mortality rates. In addition to the difficult diagnosis of pathogens in immunocompromised patients with mixed infections, another problem is the frequent overuse of antibiotics, especially broad-spectrum antibiotics. Fortunately, mNGS may offer the probability for optimized antibiotic strategy and improved clinical excessive antibiotic use in immunocompromised patients assisted by vv-ECMO to alleviate antibiotic resistance and misuse of medical resources. In 2020, clinicians paid attention to the application value of mNGS in immunosuppressed patients with ARDS and found that mNGS detection can shorten the ICU stay time and the ventilation time and reduce the ICU cost significantly (Zhang et al., 2020). In our present study, mNGS has exhibited the effect of comprehensively evaluating and even effectively adjSusting the antimicrobial therapy in immunocompromised patients supported by vv-ECMO.

Our present study has some limitations. This study is a single-center retrospective study with a small sample size. Although we selected other more comprehensive measures such as the SOFA score, CURB-65 Score, and APACHE II Score to assess the severity of the disease in the two groups, these measures were not significantly different between the two groups; the older age of the Group D may affect patient outcomes. Then, most patients had experienced mechanical ventilation, which may have affected the pathogens’ results. Moreover, ECMO itself is an invasive operation, which may affect the treatment after being ECMO assisted.

In immunocompromised patients assisted by vv-ECMO, the pathogens detected by mNGS were consistent with the traditional culture (BALF and blood), but the pathogen species and strains detected by mNGS were richer. Some patients had changed their antibiotic treatment scheme. mNGS had an advantage in detecting mixed pathogens and personalized antibiotic treatment in immunocompromised patients assisted by vv-ECMO.

The data that support the findings of this study have been deposited into EMBL database with accession number is PRJEB55072.

The studies involving human participants were reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

Y-CZ and Y-ZD designed the study and wrote the first draft of the manuscript. XZ, G-WF, M-JH, S-LW, Y-BK, and JL verified data extraction and data analysis and reviewed the manuscript. W-TM, Q-FX, Q-LL, H-BL, X-XL, and Q-QS supervised the data acquisition, data analysis, and interpretation. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.