Elizabethkingia meningoseptica, formerly known as Chryseobacterium meningosepticum or Flavobacterium meningosepticum, is a gram-negative rod that is aerobic, non-motile, non-fermenting, and does not generate spores (1). The bacteria are distributed widely in nature, including in water, fish, soils, insects, and frogs. They are also present in hospital settings, where they may contaminate medical equipment and flushing solutions (1, 2). Infections with E. meningoseptica are usually involved with indwelling devices and often influence immunocompromised patients, as well as neonates with neonatal meningitis and sepsis (1). It is an opportunistic pathogen that forms biofilms and can survive for extended periods in moist environments or water sources, including tap water (3). Currently, the genus Elizabethkingia contains six species: Elizabethkingia anophelis, E. meningoseptica, Elizabethkingia bruuniana, Elizabethkingia miricola, Elizabethkingia occulta, and Elizabethkingia ursingii (4). E. meningoseptica is the most virulent of the six known Elizabethkingia bacterial species (5). Furthermore, E. meningoseptica is a hospital pathogen, and correct identification of pathogens is essential for clinical diagnosis and treatment (6). Because of the intrinsic multidrug resistance of E. meningoseptica to commonly used antibiotics including aminoglycosides and β-lactams, the infection caused by this bacteria is difficult to cure and has a high mortality rate (1, 7). According to a study in Australia, antimicrobial resistance (AMR) genes, bla_BlaB, bla_GOB, and bla_CME, were discovered in the genomes of all Elizabethkingia clinical isolates from Australia. Because of unique metallo-β-lactamases and unique extended-spectrum β-lactamases, Elizabethkingia species are considered to be resistant to most β-lactams (8).

In clinical microbiology laboratories, the matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) system with an extended spectral database is extensively employed for microbial identification. It successfully identifies E. meningoseptica and E. anophelis but is unable to differentiate between the remaining species. Accurate species identification requires molecular approaches, such as whole genome sequencing and housekeeping gene sequencing (2, 9). 16S rRNA gene sequencing is regarded as an accurate approach for identifying Elizabethkingia species, according to the published studies (10). Some studies used sequencing of the 16S rRNA gene as a standard to verify the accuracy of the species identification system. The results of recognition of E. meningoseptica and E. anophelis were almost the same (6, 8, 10–12).

However, the clinical importance of isolating E. meningoseptica has always been questioned due to its poor pathogenicity. According to estimates from two medical centers in Taiwan, the annual incidence of E. meningoseptica infection has increased over the last decade (7). Studies have reported that almost all cases infected with E. meningoseptica occurred in the hospital environment, and the data on the incidence of E. meningoseptica infection mainly came from Taiwan. It is a new pathogen in hospitals, which is related to the high mortality in hospitals (13, 14). Notably, E. meningoseptica has intrinsic resistance to many antibiotics commonly used in intensive care settings, and patients in intensive care units (ICU) are increasingly at risk for E. meningoseptica (3, 13). With continuing advances in healthcare, clinicians are becoming increasingly aware of its clinical importance (7). Despite this, E. meningoseptica infection can be easily overlooked and misdiagnosed, resulting in further challenges in clinical practice with regard to timely diagnosis.

The majority of the reported cases originated in Taiwan, with only a few instances reported from Australia, India, the United States, and Europe (15). Determining whether E. meningoseptica is an emerging pathogenic bacterium is critical for clinical diagnosis and therapy (16). Particularly, there have only been a few studies reporting E. meningoseptica in Hefei City, and knowledge of how it infects and causes the disease is extremely limited. The clinical manifestations of infection with E. meningoseptica are diverse. Due to their inherent carbapenem resistance, the literature has paid less attention to their clinical features (17). Therefore, this study reviewed the clinical manifestations of patients infected with E. meningoseptica in the First Affiliated Hospital of Anhui Medical University, a tertiary hospital in Hefei, China, and further evaluated the laboratory findings, treatment history, and prognosis related to the bacteria. It is anticipated that this investigation may provide useful data for enhancing the detection and diagnosis of E. meningoseptica infection.

Elizabethkingia meningoseptica is most commonly isolated from freshwater, saltwater, and soil, as well as from moist and dry clinical environments, intravenous lipid solutions, equipment surfaces, and municipal water supplies, including those that are adequately chlorinated. It is an opportunistic pathogen, despite being nearly widespread. E. meningoseptica primarily infects elderly and immunocompromised patients in intensive care settings (20). Elizabeth O. King, a bacteriologist at the Centers for Disease Control in Atlanta, Georgia, United States, was the first person to report the bacteria in 1959 (21). Infections have been reported all over the world, including in the Central African Republic, Mauritius, Singapore, Taiwan, and the United States (8). According to a study conducted in Taiwan, China, E. meningoseptica was the third most frequent respiratory pathogen, next to Pseudomonas aeruginosa and the Acinetobacter calcoaceticus–Acinetobacter baumannii (ACB) complex, in a medical Center in southern Taiwan. It was also the fourth most frequent pathogen of carbapenem-resistant bacteremia there (17). With the increasing incidence of E. meningoseptica infections, the importance of being a hospital infection bacterium is increasingly recognized by human beings. However, information regarding this disease's incidence and treatment in the city of Hefei was insufficient. Thus, clinical data for this study was retrospectively collected from the hospital's electronic medical records of 24 patients who had E. meningoseptica infections. In addition, demographic information, clinical and laboratory features, treatment history, and infection outcomes were reported.

Among the 24 E. meningoseptica clinical isolates collected from the First Affiliated Hospital of Anhui Medical University, the major sources were sputum (33.3%) and blood (25.0%), respectively. Due to the high mortality rate, we divided the patients into survival group and death group. In this retrospective analysis, we found that patients over 45 had a mortality rate of 90% compared to those under 45 who had a mortality rate of 10%. Mortality rates varied from 23% to 52% in different studies (22). In our study, 41.6% of patients died, and most of the deaths were elderly patients with impaired immune function. Infections with E. meningoseptica throughout the year were most prevalent in summer and autumn (70.8%). In previous studies, neonatal patients, especially premature infants, had a high risk of E. meningoseptica infection. In this investigation, it was discovered that all isolates, except for one, were from adults, demonstrating the low rate of infectivity rate of this bacterium in children in the province of Anhui (18, 23). Univariate analysis of a prior study revealed that individuals with liver cirrhosis (P = 0.032) had a higher mortality rate (4). The only risk factor for underlying diseases in this study was hepatic disease (P = 0.009). According to our data, hypertension (37.5%) and brain disease (37.5%) were the most prevalent underlying diseases. However, a prior study revealed that diabetes mellitus (25%) and malignancy (36%) were the most prevalent underlying diseases (18). A study found that congestive heart failure (42.1%), chronic lung disease (63.2%), and hypertension (57.9%) were the most prevalent underlying diseases (24). These inconsistent results may be attributable to geographical disparities, variations in the size of the study sample, and changes in the study design. Noteworthy, invasive catheters, such as intravascular catheters or endotracheal intubation, are the significant factors in E. meningoseptica-related infections (25). The outbreak of E. meningoseptica infection was previously reported to be related to environmental pollution. The sources of environmental pollution included contaminated syringes in refrigerators, respiratory equipment, bottles, sink taps, sink drains, tube feeding, arterial catheter flushing fluid, pressure sensors, and disinfectants. However, it has also been reported that inadequate disinfection of nipple storage tanks in baby nurseries led to infant infections (26).

Few investigations have characterized the clinical manifestations of E. meningoseptica infection. Fever (83.3%) was the most prevalent clinical symptom in Hefei City. It was comparable to the 83.8% fever rate recorded by a Taiwan medical Center previously (18). Most patients presented with fever, suggesting infection. Bacterial infection often causes fever, and the clinical microbial culture and identification are needed to distinguish the infections of E. meningoseptica and other bacteria (21, 27). Lack of fever may delay the detection of infection or just indicate a poor general state (18). According to a study conducted in Taiwan, individuals with E. meningoseptic bacteremia had a greater prevalence of primary bacteremia, used fewer antibiotics, and experienced less shock when the condition first manifested than patients who did not have the infection (24). The clinical symptoms of the two groups in this investigation were not significantly different. Pneumonia, sepsis, and meningitis are the three main types of infections caused by E. meningoseptica, but there are rare reports of abdominal infection, endocarditis, eye infection, osteomyelitis, keratitis, urinary tract infection, septic arthritis, and skin or soft tissue infection (28–30). Among the complications in this study, there were 12 cases of pneumonia (50.0%) and 11 cases of hydrothorax (45.8%). Pneumonia was a major complication in patients infected by E. meningoseptica. There were 70% patients with pneumonia in the death group, as well as 35.7% patients with pneumonia in the survival group.

There was very little information available in the literature about the laboratory characteristics of E. meningoseptica, and most previous research only included a limited number of patients. Hypoproteinemia was identified as a risk factor for mortality in patients with E. meningoseptica infection in a prior investigation (22). In univariate analysis, lymphocyte count (P = 0.033), anemia (P = 0.024), and hypoproteinemia (P = 0.024) were the risk factors for death in our study. Specifically, anemia (75%) and hypoproteinemia (75%) were the most significant clinical findings in our investigation, as they were observed in the majority of the patients. As a result, we postulated that measuring albumin (ALB), total protein (TP), and hemoglobin (Hb) levels could aid in the diagnosis of E. meningoseptica infection. Comparable to usual bacterial infections, peripheral blood WBC and neutrophils counts, as well as the biochemical index CRP rose in the majority of E. meningoseptica-infected patients. A study by Arbune et al. (31) showed an infant's laboratory results were leukocytosis with neutrophilia, and anemia. At least one increased liver enzyme, including AST, ALT, ALP, LDH, and GGT, was observed in 24 patients. Consequently, the effect of E. meningoseptica infection on liver function was also worthy of consideration during the progression of the disease.

The Clinical and Laboratory Standards Institute had not developed interpretive breakpoints for E. meningoseptica minimum inhibitory concentrations (MICs) until 2013 (13). Antibiotics, as the main factors influencing the prognosis, are an essential component of the treatment. Elizabethkingia species infections are difficult to cure and have a high case-fatality rate, most likely as a result of their intrinsic antibiotic resistance (32). The bacteria are susceptible to ciprofloxacin, rifampicin, trimethoprim-sulfamethoxazole, and vancomycin, but resistant to β-lactam agents, aminoglycosides, and chloramphenicol (31, 33). In a case report from Sichuan, four isolates were isolated from urine, which were resistant to aminoglycosides (amikacin, tobramycin, and gentamicin), cephalosporins (ceftazidime, cefotaxime), cefoxitin, and aztreonam, and susceptible to trimethoprim-sulfamethoxazole (30). Our observations were consistent with previous findings. In this investigation, all isolates showed resistance to cefazolin, and the majority exhibited resistance to amikacin, aztreonam, imipenem, tobramycin, ceftazidime, cefepime, ceftriaxone, and gentamycin. According to earlier research conducted in Taiwan, the isolates of E. meningoseptica were more susceptible to the antibiotics piperacillin (15%), levofloxacin (30%), and minocycline (60%) (4). Our study showed that the most active antibiotic was cotrimoxazole (87.5%), followed by levofloxacin (75.0%), piperacillin/tazobactam (75.0%), and ciprofloxacin (58.3%). In a hospital in Beijing, 15 of 26 E. meningoseptica strains were resistant to sulfamethoxazole/ trimethoprim. A polymerase chain reaction was utilized to detect the resistance-determining genes for trimethoprim/sulfamethoxazole. Six isolates held the sulI gene and four isolates possessed the sulII gene; however, just one isolate contained the dfrA12 gene (34). In the investigation conducted by Jian et al., all E. meningoseptica isolates were resistant to ciprofloxacin, whereas 44% of them were resistant to levofloxacin. This resistance was mostly mediated by a single nucleotide mutation in the gyrA gene QRDR (35). The difference in drug susceptibility in different studies may be due to the regional drug resistance distribution and the size of the sample base. Chiu et al. (23) reported that vancomycin was successfully used for the treatment of E. meningoseptica infection. And genes related to tetracycline and vancomycin resistance were detected (36). However, in other studies, vancomycin was not suggested to be effective against this microorganism (37). As this was a retrospective analysis, data on vancomycin and minocycline were not available. An important aspect of infection management is the use of antibiotics in a timely manner. In a study, there was a contradiction between the genotype and phenotype of tetracyclines, sulfonamides, and quinolones. There were no quinolone resistance genes found, and 90% of the bacteria were ciprofloxacin-resistant (38). Nevertheless, in our study, 58.3% of strains were susceptible to ciprofloxacin. Twenty-one strains of Elizabethkingia isolated from five hospitals in Hong Kong were susceptible to ciprofloxacin, vancomycin, and cefoperazone-sulbactam (1). The main finding of Chen et al. (39) was that the microbial cure rate of piperacillin/tazobactam combined with methoxyprolin/sulfasalazine or fluoroquinolone seemed to be high, and the mortality rate was relatively low. Fifty-one percent of 100 E. meningoseptica isolates in a hospital in Shanghai had moderate antibacterial activity against sitafloxacin, suggesting that sitafloxacin may be a promising treatment for E. meningoseptica infection (40). In terms of our drug susceptibility results, the medication to consider first should be cotrimoxazole, followed by quinolones and piperacillin/tazobactam.

It is difficult to correctly identify E. meningoseptica by traditional microbiological methods, and it is common to misidentify E. anophelis and E. meningoseptica (6). However, delays in biological identification may lead to treatment failure. Different AST methods (viz., disc diffusion, broth-micro dilution, and E-test) have different drug susceptibility results, which further complicate the treatment. Literature shows that the paper diffusion method is not reliable, the broth-micro dilution method is the preferred method. While the automatic drug susceptibility testing system is more commonly used in clinical laboratories, there are still differences between the two (11, 41, 42). As a pathogen causing nosocomial infections, there is little information about its genome composition and related characteristics in the literature. A document indicated that the genomes of E. meningoseptica isolated from patients carried three β-lactamases. These strains were resistant to β-lactams and cephalosporins, which may be attributed to bla_CME; resistance to carbapenems and penicillin–β-lactamase inhibitor combinations may be attributed to bla_BlaB and bla_GOB, which explains the molecular mechanism of drug resistance (8). A study in Hainan showed that clinical strains of E. meningoseptica had β-lactam, macrolide, tetracycline, quinolone, glycopeptide, and multidrug-resistance efflux pump genes. The clinical isolates contained antibiotic efflux pump genes, cmeB, adeF, and vanB, and glycopeptide resistance gene, vanW, in which vanB glycopeptide resistance gene vanW mutation was observed to be involved in the regulation of teicoplanin resistance (38). A literature in Wuhan showed that 23 and 32 new Bla_BlaB and Bla_GOB variants were found in Elizabethkingia spp., respectively. Some variations did not aggregate according to the species-specific branching, indicating that the MBL gene may be transmitted across species in Elizabethkingia species (43). In the study of Girdhar et al. (44), they identified 18 unique proteins related to the metabolic pathway of E. meningoseptica from 3,391 annotated proteins, which may be the starting point for drug design and development.

In addition, we used the MALDI-TOF MS database software and constructed the development tree for homology analysis of eight strains of the total 24 E. meningoseptica. The tree diagram based on correlation obtained by MALDI-TOF MS clustered the eight strains in the same group, showing that they had a strong genetic relationship. We found that three strains were highly similar in E. meningoseptica (~87%), suggesting common hospital sources. Eight strains were identified as E. meningoseptica by 16S rRNA gene sequencing, which was the same as the result of MALDI-TOF MS. Since there are few cases of patients collected at present and E. meningoseptica that have not accumulated large data, we will continue to collect cases for analysis in the later stage to provide help for clinical diagnosis and identification.

Various studies have shown that mortality is associated with inappropriate empirical antimicrobial therapy (4, 7, 15, 18). In addition, the tendency to form biofilms poses further challenges to the treatment of patients, especially when organisms have been established on indwelling devices, such as endotracheal tubes (7). In a study from Singapore, the identification of the gene for capsule biosynthesis, capD, and the gene for the AdeFGH efflux pump, adeG, in all Elizabethkingia species leads to the possibility of biofilm formation, which gives the bacteria the ability to remain on varied surfaces (6). The potential of E. meningoseptica to produce biofilm indicates that it has the ability to impede treatment. Numerous investigations involving E. meningoseptica infection or outbreaks caused by flume or faucet drainage pollution have demonstrated that once biofilm is developed, it is difficult to eliminate from the environment (5, 20, 45).

This study has a number of limitations. Because this is a retrospective single-center study, some data may be lost, incomplete, or improperly reported. Furthermore, because of the study's limited sample size, the explanation for our results may not be applicable in larger populations. Despite these limitations, this work offers an important potential for the investigation of E. meningoseptica infection in Hefei. It is also expected that this research would offer guidance and reference for clinicians to timely diagnose and treat E. meningoseptica infection.

In summary, the most common symptom in E. meningoseptica patients was fever, which was followed by altered consciousness, edema, cough, sputum, abdominal distension, chest discomfort, and chilly. The only significant factor of underlying diseases was that hepatic disease (P = 0.009) often occurred in death groups. A majority of patients experienced anemia and hypoproteinemia in the laboratory result. Our findings suggest that hypoproteinemia, unexplained fever, anemia, and lymphocyte count should be considered as E. meningoseptica selected diagnosis references for early treatment intervention. The susceptibility of some quinolones, piperacillin-tazobactam and cotrimoxazole was relatively high, suggesting that they may be the preferred drugs for the treatment of E. meningoseptica infection. Since such bacteria can form biofilms in a humid environment and spread with water or equipment relating to water as a source of transmission, it is necessary to strengthen the hand hygiene compliance of medical staff and control nosocomial infections to avoid the spread and spread of E. meningoseptica infection. This study analyzed 5 years of patient data from Anhui Province and could be used to improve strategies for preventing, diagnosing, and treating E. meningoseptica infections in other provinces and cities.

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

This study was reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Anhui Medical University in accordance with the Declaration of Helsinki (ethical approval number: Quick-PJ 2022-10-31). Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin. Written informed consent was obtained from the individual(s), and minor(s)' legal guardian/next of kin, for the publication of any potentially identifiable images or data included in this article.

YL and TLiu analyzed data and wrote original draft. CS, BW, TLi, YH, YX, and LT reviewed and edited the manuscript. YX and YL provided funding. All authors have read and agreed to publish the final manuscript.

This work was financially supported by Anhui Natural Science Foundation (grant number: 9021138201) and Scientific Research Project of Universities in Anhui Province (grant number: KJ2020A0170).

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.