Central nervous system (CNS) infections are inflammatory conditions of the brain and spinal cord caused by a variety of pathogenic microorganisms, including meningitis, encephalitis, myelitis, and abscesses. CNS infectious diseases are the main cause of morbidity and mortality worldwide (Tunkel et al., 2008; Venkatesan et al., 2013). For CNS infections, empiric antimicrobial therapy is usually suboptimal. Early detection of the causative organism in severely infected patients is essential to informing clinical intervention and the use of appropriate target antibiotics (Gu et al., 2021). Without appropriate treatment, it will progress rapidly and lead to secondary neurologic emergencies, particularly stroke, epilepsy, hydrocephalus, and intracranial hypertension (McEntire et al., 2021). Moreover, it can prolong hospital stays, increasing medical costs and the risk of death (Kanjilal et al., 2019). In addition, patients who remain undiagnosed will always require empiric broad-spectrum antibiotic therapy, which will increase the risk of adverse side effects and antimicrobial resistance (Llor and Bjerrum, 2014). Fortunately, infectious diseases are one of the few critical neurological conditions where complete or near-complete recovery can be achieved after treatment, as appropriate antimicrobial therapy can often significantly improve the prognosis (McEntire et al., 2021). Therefore, timely diagnosis, accurate differentiation of the underlying causative agent, and appropriate treatment are vitally important (Kanjilal et al., 2019).

CNS infections can be induced by a considerable number of infectious agents, mainly viruses and bacteria; cases of fungi and parasites are much less common (Vetter et al., 2020). To identify viable pathogens, clinicians must rely on the microbiological examination of clinical specimens, particularly cerebrospinal fluid (CSF) (Kanjilal et al., 2019). Non-specific CSF testing methods include CSF chemistry and cell counting, as well as Gram staining and culture. Pathogen-specific detection techniques include 1) pathogen-specific antibody reactions, which provide indirect evidence of infection; 2) direct detection of a pathogen by microscopic observation or direct detection of microbial antigens; and 3) direct detection of pathogen nucleic acids from prespecified targets by PCR. These tests are known as traditional diagnostic techniques, and they have their advantages and limitations (Kanjilal et al., 2019; Vetter et al., 2020). Isolation and culture of microorganisms indicate the presence of viable microorganisms but are less sensitive and time-consuming. Both PCR and serology require the clinician to select the pathogen being interrogated. These complexities often lead clinicians to utilize a patchwork of tests, which increases time consumption and financial costs (Ramachandran and Wilson, 2018; Kanjilal et al., 2019; Bharucha et al., 2019; Houlihan et al., 2019). There is therefore an urgent need to explore a more comprehensive, rapid, and accurate diagnostic approach to improve the efficiency of diagnosis of CNS infections.

Bibliometrics allows for the recognition of emerging trends and knowledge structures in the research field through a quantitative analysis of patterns in the scientific literature (Chen et al., 2012; Chen et al., 2020; Deng et al., 2020; Zhang et al., 2020). To provide an accurate, validated, and systematic overview of developments in pathogen detection technologies for CNS infections, to explore the latest detection methods and their effectiveness, and to identify thematic trends, research frontiers, and leading collaborations among researchers at various institutions, we conducted this bibliometric study by using CiteSpace software.

In central nervous system infections, timely identification of the pathogen is key to effective treatment, but this remains a formidable challenge. Current data show that even in centers with the best resources, up to two-thirds of CNS infection cases remain undiagnosed (Granerod et al., 2010; McGill et al., 2018). Because most infectious syndromes have indistinguishable clinical presentations, there is an urgent need for broad-based multiplex diagnostic tests, which are currently unavailable for the vast majority of potential pathogens. Some microorganisms are difficult or impossible to culture, while others take weeks to culture and speciate (Rea et al., 2019). Accurate molecular tests by PCR and serological tests that detect pathogen-specific antibodies offer diagnostic alternatives to culture, but they are hypothesis-driven and require a priori suspicion of pathogenic pathogens (Kanjilal et al., 2019). Large multiplex test panels represent a paradigm shift in medical microbiology and clinical infectious diseases. The main benefit of these panels is the potential for more rapid results. However, because of the potential for contamination and the ability to detect latent or reactivated viruses, the results of the panels must be scrutinized and the laboratory must closely monitor the positivity rate (Hanson, 2016). Therefore, the development of more rapid, comprehensive, hypothesis-free pathogen detection methods with high sensitivity and specificity to improve diagnosis is a priority in neurological infection research.

In this study, we used CiteSpace software to explore trends and developments in the field of CNS infection pathogen detection research from 2000 to 2021. Trends in the number of annual publications and total citations of annual publications in the field reflect the interest and focus on the field over the years. The research area covers a wide range of disciplines and topics, including microbiology, infectious diseases, immunology, and neuroscience, reflecting the combined efforts of experts and scholars from various disciplines in the field. Collaborative network analysis can provide detailed information for assessing research collaborations and identifying key collaborators. Among countries/regions and institutions, the United States has an advantage in this area of research, given that it has published the most articles in this field and that three-fifths of the top ten institutions by the number of publications are located in the United States. Despite the close international scientific collaboration between scientists in this area of research, we found from the collaborative authorship network map that collaboration between top researchers is not sufficient.

Combining reference co-citation analysis and keyword co-occurrence analysis, this study found that research in the field of pathogenic diagnosis of CNS infections is focused on exploring new etiologies (autoimmune and rare microbial pathogens) and more advanced laboratory diagnostic methods. Based on the analysis of co-citation clusters, most cited articles, citation bursts, keyword co-occurrences, and burst keywords, metagenomic next-generation sequencing has become the most popular research hotspot in recent years.

Molecular diagnostic techniques have been continuously improved and developed in recent decades. The advent of high-throughput sequencing has made it possible to discover new microorganisms, rapidly address the causes of infectious disease outbreaks, and develop metagenomics. Metagenomics next-generation sequencing (mNGS) was initially limited to dedicated laboratories because of the high cost of instrumentation and consumables, as well as the need for specially trained personnel (Lipkin, 2013). With many published case reports and clinical study results confirming the validity of mNGS, and the continued decline in sequencing cost and time, mNGS is increasingly being used for clinical diagnoses, providing clinicians with a powerful diagnostic tool that has greatly improved our ability to detect infectious disease pathogens in clinical samples (Goldberg et al., 2015; Allcock et al., 2017).

Metagenomic next-generation sequencing is a comprehensive and rapid diagnostic tool. It allows unbiased and detailed testing of the total DNA or RNA content of all currently known pathogenic microorganisms (Fei et al., 2020). It offers a powerful advantage in infectious disease diagnosis as it does not require special probes and targeting primers and facilitates rapid detection of pathogens, including non-culturable organisms and novel organisms (Liu et al., 2015; Perlejewski et al., 2016; Bukowska-Ośko et al., 2017). mNGS also offers a diagnostic advantage over conventional methods for patients who have received empirical antimicrobial therapy before sample collection. The empirical use of antibiotics significantly reduces the detection rate of conventional methods by approximately 20%, whereas the detection rate of mNGS is unaffected (Zhang et al., 2020). There is growing evidence that mNGS plays an essential role in the diagnostic workup of CNS infections. Potentially treatable pathogens, such as Leptospira, Brucella, and Balamuthia mandrillaris, and non-treatable pathogens, such as astroviruses and novel viruses, which were not clinically suspected, have been identified using this technique (Wilson et al., 2014; Naccache et al., 2015; Greninger et al., 2015; Mongkolrattanothai et al., 2017). In 2017, the French “Guidelines on the management of infectious encephalitis in adults” recommended mNGS as level 1 evidence to assist clinical decision-making in CNS infections (Stahl et al., 2017).

Despite the appeal of mNGS for infectious disease diagnosis, there are many challenges before the technology becomes mainstream and part of the standard of clinical care. 1) Economic costs: given the cost of analysis and the required informatics infrastructure (e.g., sequencing platforms, computing resources, data storage), the use of mNGS is currently still limited to a small number of diagnostic laboratories (Zanella et al., 2019). 2) Standardization: standardized operational procedures for NGS diagnostics, including clinical specimen collection, sequencing parameters, data analysis algorithms, reference databases, quality control, and data reporting, need to be further defined (Goldberg et al., 2015). 3) Technical challenge: there is still a lack of effective methods to reduce host contamination and improve the sensitivity of mNGS to clinical CSF samples. Human DNA contamination is a major challenge for mNGS in detecting pathogens in most clinical specimens. Elevated white blood cell counts are one of the main manifestations of infection and a major source of host contamination. This is because the high background of host DNA greatly reduces the sequence coverage of pathogens, thereby reducing the sensitivity of mNGS to detect pathogens (Hasan et al., 2016). Several attempts have been made to reduce host DNA contamination. Selective degradation of host DNA in blood samples with methylation-dependent endonucleases, based on the different methylation patterns between host DNA and pathogen genomes, results in an enrichment of pathogen DNA (Auburn et al., 2011; Oyola et al., 2013). However, this approach does not apply to all pathogens, as many types of microorganisms also have methylated CpGs (e.g., fungi). Another approach to selectively isolate host DNA is to hybridize to pathogen DNA using specific probes complementary to the pathogen genome and capture it for amplification and sequencing. Although these targeted enrichment methods can partially decontaminate the human genome, they are insufficient to provide comprehensive, unbiased detection of infections of unknown etiology (Melnikov et al., 2011; Bright et al., 2012; Smith et al., 2012). 4) Interpret test results with caution: although one of the advantages of mNGS analysis is the ability to detect DNA or RNA of many microorganisms directly from patient samples, a positive test result does not establish that it is from a live microorganism, whereas a positive culture result indicates the presence of a live organism. In addition, mNGS has particular limitations in distinguishing which organisms are colonized, conditionally pathogenic, or active pathogens (Goldberg et al., 2015). 5) Higher demands on clinicians: molecular diagnostics presents challenges in terms of data interpretation and reporting, and the shift to molecular diagnostics will require a change in clinician thinking. Upcoming generations of physicians will need to adapt to the strengths and limitations of these new tools (Goldberg et al., 2015).

Therefore, we need to take additional efforts to address these technical challenges and to further reduce the economic costs. At the same time, we should also recognize that, in addition to identifying pathogens, treatment is vital. Therefore, the identification of virulence genes and sequence variants of causative pathogens, assessment of antibiotic resistance, and evaluation of the effectiveness of antibiotic therapy should also be given high priority (Goldberg et al., 2015). More clinical studies are needed to explore the capabilities of mNGS in these applications (Goldberg et al., 2015).

In conclusion, the analysis and citation-based expansion of the literature on pathogen detection in CNS infections have outlined the trajectory of the evolution of collective knowledge over the past two decades and highlighted areas of active pursuit. The exploration of more advanced laboratory tests to identify pathogenic pathogens has always been the core of research priorities in neurological infections. As a comprehensive, rapid, and unbiased test, mNGS offers diagnostic advantages over traditional methods, but at the same time, mNGS still has several technical and economic limitations that need to be improved upon. The results of our review and analysis of the hotspots and research trends may promote the development of this field.

There are some limitations to this study that need to be noted and addressed. Firstly, as our study was based on references searched in the WoS Core Collection database, this means that we may have omitted some important studies in other medical databases. However, because different databases have different methods of calculating citation frequencies, it is not appropriate to merge data from different databases, and it is difficult for current bibliometric software to merge two and more databases. Therefore, previous studies have usually selected one database as the primary search database (Zhong et al., 2020; Chen et al., 2020; Wu et al., 2021). Among these, WoS Core Collection is a commonly used reference database in bibliometric research. Secondly, our study only analyzed the English language literature, which may have led to the incompleteness of the data and the deviation of the research results. Thirdly, the deadline for the research publications in this study was August 12, 2021. All data in this work do not fully reflect the reality of 2021 and could be a reference.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

HL and YG designed the study. YG and JdZ performed the search. YG, JdZ, and MX analyzed the data and performed the software analysis. YG drafted the first vision. GS, JxZ, and HLL revised the manuscript critically. All authors contributed to the article and approved the submitted version.

This work was supported by the Beijing Tiantan Hospital Youth Fund (2018-YQN-15) and the National Natural Science Foundation of China (82000510).

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.