Cefiderocol-Resistant Elizabethkingia anophelis Bacteremia Following WATCHMAN Implantation: A Case Report and Review of the Literature

TL;DR

This paper reports the first case of cefiderocol-resistant Elizabethkingia anophelis bacteremia after a WATCHMAN implantation, highlighting the challenges of treating this drug-resistant pathogen.

Contribution

The first reported case of cefiderocol resistance in Elizabethkingia spp., expanding the known resistance profile of this emerging pathogen.

Findings

A case of cefiderocol-resistant Elizabethkingia anophelis bacteremia was identified following WATCHMAN implantation.

This is the first documented instance of cefiderocol resistance in Elizabethkingia species.

The case underscores the growing challenge of antibiotic resistance in treating E. anophelis infections.

Abstract

Elizabethkingia anophelis is an emerging pathogen associated with nosocomial and community outbreaks. Treatment of infection caused by E. anophelis is not well-defined given its extensive drug resistance profile, and infection carries a poor prognosis. E. anophelis is intrinsically resistant to many classes of antibiotics including carbapenems and polymyxins due to multiple resistance genes. Resistance to novel β-lactam/β-lactamase inhibitors has also been reported. The activity of cefiderocol (FDC) is unknown. Here, we describe a case of FDC-resistant E. anophelis bacteremia following WATCHMAN implantation. To our knowledge, this is the first reported case of FDC resistance amongst Elizabethkingia spp.

Full text

1. Introduction

Elizabethkingia anophelis, an environmental, aerobic, glucose-nonfermenting, oxidase-positive, Gram-negative bacilli [1], was first identified in 2011 from the Anopheles gambiae mosquito [2]. Since the first episode of human disease was reported [3], numerous nosocomial and community outbreaks have been reported worldwide [4–10]. Many of these cases were originally misidentified as Elizabethkingia meningoseptica using commercial identification systems (i.e., VITEK 2, MALDI-TOF) [11].

E. anophelis infections occur predominantly in patients at extremes of life or with underlying comorbidities such as chronic obstructive pulmonary disease, diabetes, unhealed wounds, and ventilator-dependence. Common disease manifestations include pneumonia (PNA), catheter-related bacteremia, and neonatal meningitis, with mortality ranging from 18% to 70% [11–13].

E. anophelis is intrinsically resistant to many classes of antibiotics including carbapenems due to multiple chromosomally encoded β-lactamases: blaB and blaGOB, two metallo-β-lactamases (MBLs); blaCME, a Class D serine β-lactamase [14, 15]. Resistance to non-β-lactam agents including fluoroquinolones, tetracyclines, aminoglycosides, and newer β-lactam/β-lactamase inhibitors has also been reported [1]. The activity of cefiderocol (FDC), a novel siderophore cephalosporin, against Elizabethkingia spp. is relatively unknown given its recent approval and limited number of cases reported. To date, FDC resistance amongst Elizabethkingia spp. has not been reported.

The WATCHMAN device is a small, parachute-shaped device that is implanted in the left atrial appendage of the heart to seal it off. It is used as an alternative to anticoagulation in selected patients to prevent strokes. While rare, infections are a potential complication associated with the WATCHMAN device.

2. Case Presentation

An 82-year-old patient presented with acute fever (103°F), nausea, vomiting, leukocytosis (13.6 × 10^9^ cells/L), and elevated CRP (126 mg/L), immediately after elective cardioversion for atrial fibrillation. Blood cultures were obtained and empiric vancomycin (VAN) and ceftriaxone (CRO) were initiated. Other comorbidities included chronic kidney disease, psoriasis, and recent right-knee prosthetic joint infection with Pseudomonas aeruginosa treated with surgical washout and a 6-month course of antibiotics (oral levofloxacin [LVX] plus minocycline). Six weeks prior to admission, he underwent atrial ablation with WATCHMAN implantation.

Both pairs of blood cultures on two consecutive days revealed E. meningoseptica. Antimicrobial susceptibility obtained by automated broth microdilution (BMD) showed resistance to all cephalosporins, including FDC, aminoglycosides, carbapenems, piperacillin/tazobactam (TZP), aztreonam (ATM), and tetracyclines. It was susceptible to fluoroquinolones and cotrimoxazole (SXT) (Table 1), and LVX 750 mg daily was initiated. The patient had no known prior exposure to FDC. The isolate was submitted to a reference lab for confirmatory testing and subsequent whole genome sequencing (WGS). Further workup revealed bibasilar opacities on CT chest. Sputum culture grew Candida species, and blood cultures repeated 48 h after antibiotic treatment remained negative. Given the propensity of Elizabethkingia spp. to form biofilms, transesophageal echocardiogram was performed to rule out device-related infection, which did not show any vegetation. He was treated with LVX for 2 weeks. At 6-week follow up, the patient was asymptomatic, chest X-ray showed resolution of lung opacities, two sets of blood cultures were negative, and his CRP had normalized.

3. Discussion

WGS identified E. anophelis, resistant to multiple antimicrobials, including FDC, with an MIC of 8 mcg/mL (Table 1) determined via BMD. The discrepancies in several MIC values may be secondary to known limitations with automated testing panels, whereas classic BMD allows for more precise values to be obtained [16]. For FDC specifically, reproducibility of MIC values may be a challenge given historic testing difficulties when utilizing iron-depleted cation-adjusted media [17]. To date, there are no reported cases of FDC resistance amongst Elizabethkingia spp. Mechanisms leading to reduced susceptibility or resistance to FDC amongst Gram-negative organisms have been evaluated in surveillance studies and case reports.

Treatment of E. anophelis infection is not well-defined given its extensive drug resistance profile and lack of standardized breakpoints for Elizabethkingia spp. Interpretations are inferred using CLSI breakpoints for Enterobacterales and lactose nonfermenters. E. anophelis exhibits a wide variation in susceptibility to TZP (0%–92%), ciprofloxacin (CIP) (1%–100%), LVX (16%–96%), and SXT (4%–70.6%) [18, 19]. Although clinical outcomes with monotherapy appear mixed, fluoroquinolone use and improved survival have been observed [20]. Improved survival utilizing dual therapy including either SXT, minocycline, rifampin, or a fluoroquinolone has been observed [6, 21, 22]. The role of newer β-lactam/β-lactamase inhibitors such as ceftazidime-avibactam (CZA), imipenem-relebactam (I-R) and meropenem-vaborbactam (MVB) against E. anophelis is not well defined.

A recent review by Koh et al. [23] found that the most common causes of Elizabethkingia spp. bacteremia were catheter-related (71%), followed by PNA (14%) and skin and soft tissue (1%). All of these were potential etiologies in our patient, although no etiology could be proved with certainty.

Several extended-spectrum β-lactamases have been identified in bacterial isolates with elevated FDC minimum inhibitory concentrations (MIC). Takemura et al. isolated PER, NDM, and VEB in Acinetobacter baumannii complex isolates, while NDM and VEB were isolated amongst Enterobacterales [24]. Tiseo et al. demonstrated that a KPC-3 variant with a D19Y substitution in a Klebsiella pneumoniae isolate led to FDC resistance [25]. Amongst Enterobacter cloacae complex isolates, deletions in the R2 loop of the AmpC β-lactamase were associated with FDC resistance [26]. In a study of five FDC-resistant E. cloacae isolates from a patient following prolonged FDC exposure, OXA-48 and NDM β-lactamases, and cirA gene mutation, which encodes for a siderophore iron transporter, were identified [27].

Disruptions in TonB-dependent siderophore uptake receptor genes can lead to loss of FDC transport into the bacterial cell. Iron transporter mutations in pirA/D, cirA, and fiu genes are documented in the literature [28]. Mutations in ompR/envZ and exbD, which affect iron transporter genes, may be associated with FDC resistance and were isolated in FDC-resistant K. pneumoniae strains by Yamano et al. [29]. These isolates also exhibited mutations in baeS and yicM, of which the clinical effects are unknown. Increased FDC MICs were observed amongst P. aeruginosa isolates following drug exposure with associated mutations in fecA (iron transporter-related gene) and pvdS, a regulator of pyoverdine synthesis [30].

OmpK35, OmpK36, and OmpK37 porin mutations in K. pneumoniae and OmpC and OmpF porin mutations in Enterobacter spp. have been detected in FDC-resistant strains [28]. Efflux pumps including sugE and chrA in K. pneumoniae, MexAB-OprM in P. aeruginosa, smeDEF in Stenotrophomonas maltophilia, and AxyABM in Achromobacter xylosoxidans are associated with elevated FDC MICs and resistance [31–34]. Mutations in ftsI, which encodes for PBP-3, associated with FDC resistance have been documented in strains of Escherichia coli and A. baumannii. Increased FDC MICs were observed in strains of E. coli from a Turkish hospital following YRIN or YRIK insertion after Positions P333 and I532L [35]. YRIN insertion at P333, in combination with NDM and cirA mutations, has also been detected in FDC-resistant E. coli [36]. Additionally, mutations in PBP-3 (including H370Y) have been detected in FDC-resistant A. baumannii isolates, often in combination with various β-lactamases [28].

Studies have showed that multiple mechanisms of resistance, often β-lactamases in combination with permeability defect mutations, appear synergistic, leading to FDC resistance. Lan et al. demonstrated that in isolates of K. pneumoniae, the presence of NDM-5 or cirA mutations alone led to an eightfold (0.5–4 mcg/mL) and twofold (0.5–1 mcg/mL) increase in FDC MIC, respectfully [37]. When both NDM-5 and cirA mutations were introduced, the FDC MIC increased to greater than 256 mcg/mL.

Classified as a multidrug-resistant organism, E. anophelis has intrinsic resistance to several β-lactams. Elizabethkingia spp. possess two distinct MBLs, blaB and blaGOB, which can be divided into three subclasses: B1, B2, and B3 [14, 15]. blaB, a subclass B1 MBL, broadly hydrolyzes penicillins, cephalosporins, carbapenems, and novel β-lactam/β-lactamase inhibitors. blaGOB, a subclass B3 MBL, is hindered by elevated concentrations of blaB and is unlikely to contribute to additional β-lactam resistance. Novel MBL subclasses, including blaB−1, blaB−16, and blaGOB−19 have been reported in Elizabethkingia spp. [38, 39] blaCME, intrinsic to E. anophelis, confers broad resistance to penicillins and cephalosporins. E. anophelis may possess additional resistance mechanisms affecting other antibiotic classes. Fluoroquinolone resistance is primarily mediated by mutations in quinolone resistance determining regions; the most common mutations via amino acid alterations at Positions 83 and 95 in GyrA yielded high-level resistance [40]. Fluoroquinolone-related efflux pump AcrB upregulation has been documented in fluoroquinolone-resistant E. anophelis strains [41]. Tetracycline resistance may be mediated by mutations in ribosomal protection proteins tetB(P) and otrA, and efflux proteins otrB, tetA, and tetB [14]. Additionally, five E. anophelis isolates in China conferred pan-tetracycline resistance via tetX, a flavin-dependent monooxygenase that leads to enzymatic degradation of tetracyclines [42]. dfrA12, sul I, sul II, and folP mutations conferring resistance to SXT are found in E. anophelis, although sporadic documentation in SXT-resistant strains leaves their role uncertain [40]. Additionally, isolation of aminoglycoside-inactivating enzymes, aac3-i, ant-6, ant-6-i, aac6, and aac6-lad, and vanW mutations have removed aminoglycosides and VAN as reliable treatment options [14]. The macrolide, lincosamide, and streptogramin (MLS) resistance genes ermF, ereD, mefC, and mphG have been identified in Elizabethkingia spp [43].

A positive association between E. anophelis biofilm formation and antibiotic resistance has been observed. The analysis of 197 multidrug-resistant E. anophelis isolates identified 20 biofilm-positive, extensively drug-resistant isolates. Sputum samples showed the highest percentage of strong biofilm-forming strains (p < 0.05), whereas weak biofilms formed mainly in bloodstream infections (p < 0.05). Strong–moderate biofilm-forming strains were associated with a statistically significant higher average number of resistances (11.01 ± 0.1643, p=0.0006) compared with weak formers (9.657 ± 0.3447), especially for TZP, cefepime, amikacin, and CIP [44].

Yasmin et al. characterized the molecular mechanisms of antimicrobial resistance in an E. anophelis isolate resistant to all β-lactams and carbapenems and found it was resistant to CZA, I-R, MVB, and CZA plus ATM combination [1]. FDC, which is stable against hydrolysis by both serine and MBLs, was determined susceptible via BMD (MIC 2 mcg/mL). Additional literature regarding the activity of FDC versus Elizabethkingia spp. is extremely limited; two studies reported on four E. meningoseptica isolates, all were FDC sensitive with MICs ≤ 4 mcg/mL and ≤ 1 mcg/mL [31, 45].

Although blaB−3, blaGOB−4, blaCME−type, and two single nucleotide polymorphisms in ftsI were identified via WGS, the exact mechanism behind FDC resistance in our isolate is unknown, although likely multifactorial. Further WGS, including evaluation of iron transport and porin channels in E. anophelis, is necessary to elucidate FDC resistance mechanisms.

4. Conclusion

Despite early data suggesting that FDC may be an option for the treatment of Elizabethkingia spp. infections, clinicians should exercise caution and obtain susceptibility testing especially when considering FDC monotherapy, considering FDC resistance confirmation in our isolate.