Identification and Antibiotic Susceptibility of Eggerthella lenta in Bloodstream Infections

Eggerthella lenta (formerly Eubacterium lentum) is a non-sporulating, anaerobic, Gram-positive bacillus. It was renamed in 1999 based on the 16S rRNA gene sequencing (Kageyama et al. 1999; Wade et al. 1999). The bacterium grows optimally at 37°C with arginine, and it is a component of the human gut microbiome. Genome sequencing showed that it was closely related to Paraeggerthella hongkongensis and Eggerthella sinensis (Saunders et al. 2009). Unlike the anaerobes commonly identified in blood cultures, Bacteroidetes fragile and Clostridium difficile, E. lenta is relatively rare (Deng et al. 2018; Silva-García et al. 2019; Fatima et al. 2020). As a pathogenic agent, it may cause appendicitis, liver and skin abscesses, and on rare occasions may enter the blood after the primary disease, resulting in bacteremia.

In addition, it has been found that Cgr2, a cardiac glycoside reductase from the gut microbe E. lenta, is responsible for reducing and inactivating the widely used cardiac drug digoxin. It was observed that the Cgr2-digoxin interaction primarily involved negatively charged polar amino acids and several hydrophobic residues. The bioactivity of plant-derived lignans is also dependent on E. lenta, which is critical for enterolignan production (Kumar et al. 2018; Bess et al. 2020). Few reports describe this bacterium in China (Lee et al. 2014; Woerther et al. 2017; Ugarte-Torres et al. 2018). In addition, over the past four years, only two cases involving E. lenta have been identified in blood cultures in our hospital.

Nevertheless, the mortality rate of bacteremia caused by E. lenta is high, reaching 22–43%, suggesting that this bacterium should receive more attention (Haiser et al. 2014). In this study, we describe two bacteremia samples positive for E. lenta. The samples were cultured, and their characteristics and antibiotic sensitivity were investigated.

Colonies appeared gray, translucent, smooth, and slightly raised after 72 h in culture. The diameters were between 0.25 and 0.75 mm, and hemolytic rings were

not apparent (Fig. 1). The bacteria were Gram-positive (Fig. 2). The MALDI-TOF MS results indicated 99.8% homology to E. lenta (Fig. 3). The 16S rRNA gene sequence was consistent with that from E. lenta. Phylogenetic analysis showed that the isolate was present on the same branch as E. lenta (Fig. 4). The isolates were thus identified as E. lenta with the GenBank accession number ON 197174 for the 16S rRNA gene sequence. Both strains were sensitive to piperacillin-tazobactam, ampicillin-sulbactam, imipenem, meropenem, metronidazole, clindamycin, and vancomycin. They were resistant to amoxicillin, penicillin, cefotaxime, and ceftriaxone. Table I shows the minimum inhibitory concentrations (MICs) of the antibiotics tested.

Fig.1

Fig. 2

Fig. 3

Fig. 4

It is challenging to identify anaerobic Gram-positive bacilli because many are small, slow-growing, fastidious, and biochemically inert. The isolation of E. lenta is traditionally challenging, presenting obstacles to an accurate diagnosis. However, the application of mass spectrometric techniques allows for the rapid and accurate identification of bacterial pathogens (Ugarte-Torres et al. 2018). Here, two isolates were identified as E. lenta using the MALDI-TOF MS identification and the 16S rRNA gene sequencing, with phylogenetic analysis confirming the molecular findings. The MALDI-TOF MS and sequencing results were consistent and highly accurate.

Currently, the confirmation of E. lenta as the cause of infection presents significant difficulty. The bacillus is found as part of the microbiome in the gastrointestinal tracts of healthy individuals. It has also been identified in other sites such as the frontal sinus, skin, vagina, and liver. Overall, the risk factors for E. lenta infection can be classified into three categories: a compromised immune system, digestive disorders, and potential predisposing conditions. Infections are usually associated with clinical conditions, such as pressure sores, malignancies, diabetes mellitus, trauma, surgery, gastrointestinal pathology, and hepatobiliary diseases. The E. lenta infections were associated with pressure sores and malignancy in our study. The first patient had severe pressure sores, and his skin barrier was likely broken, allowing the bacteria to access the bloodstream. The patient in case 2 had a malignant gastric tumor, and the bacteria might access the bloodstream through damaged areas of the stomach. Bacteremia due to E. lenta should be managed cautiously as it has been found to have a poor prognosis. Unfortunately, few studies on the efficacy of antimicrobial treatment and outcomes could guide the management of these infections. In terms of drug resistance, Stinear et al. (2001) identified a vanB locus in the E. lenta genome, suggesting that the bacillus could develop resistance to vancomycin similarly to enterococci with vancomycin resistance (Venugopal et al. 2012; Stinear et al. 2018).

However, another study by Gardiner et al. (2015) concluded that while E. lenta was susceptible to various antibiotics, including amoxicillin-clavulanate and meropenem; it did not appear to be resistant to vancomycin. The present study found that both isolates were sensitive to vancomycin, piperacillin-tazobactam, imipenem, meropenem, metronidazole, and clindamycin hydrochloride. Studies have demonstrated that treating bacteremia caused by E. lenta empirically with piperacillin-tazobactam or ceftriaxone monotherapy may increase mortality. Ceftriaxone resistance is likely due to the presence of the β-lactamase gene identified in most E. lenta genomes (Buttimer et al. 2022). Antibacterial treatment with amoxicillin-clavulanate, cefoxitin, or the carbapenems has been recommended as a first-line treatment for E. lenta bacteremia (Gardiner et al. 2015; Liderot et al. 2016). In the first patient, meropenem and vancomycin were chosen. In contrast, imipenem-cilastatin sodium and vancomycin were given immediately to cure the infection in the second patient, and both treatments resulted in an excellent clinical response.

In conclusion, E. lenta was isolated from two clinical samples and was identified by Gram staining, MALDI-TOF MS, and the 16S rRNA gene sequencing. These are rapid and accurate detection methods and may be completed within 24 h. The isolates were sensitive to piperacillin-tazobactam, ampicillin-sulbactam, imipenem, meropenem, metronidazole, clindamycin, and vancomycin. The above demonstration of rapid identification and antibiotic sensitivity may increase awareness of the rare infection with E. lenta.