Elizabethkingia meningoseptica: Case report on an emerging pathogen and its associated treatment challenges

Abstract

Elizabethkingia meningoseptica is a Gram-negative bacillus that was previously thought to rarely infect humans but recently has been identified as an emerging pathogen in both community and nosocomial settings. Typically found in the environment, this organism has been shown to infect predisposed hosts with an immunocompromised status and/or a prolonged exposure to healthcare settings. Herein, we report a case of a 78-year-old male with newly diagnosed myelodysplastic syndrome initially admitted to the hospital with pneumonia and then re-admitted after discharge with septic shock and evidence of E. meningoseptica bacteremia isolated from blood cultures. Treatment with piperacillin-tazobactam was initiated and later escalated to dual-therapy with the addition of levofloxacin. After hemodynamic stability was achieved, the patient was discharged on oral levofloxacin to complete a 21-day course of therapy. E. meningoseptica displays a unique multi-drug resistant profile that often makes initial antimicrobial selection challenging. This case illustrates the importance of early detection and use of in vitro susceptibility testing to guide therapeutic decision-making in E. meningoseptica infections; a pathogen known to have both high morbidity and mortality particularly in the immunocompromised.

Introduction

Elizabethkingia meningoseptica is a non-fermenting, oxidase-positive, Gram-negative bacillus that has emerged as a pathogen of interest in nosocomial settings after first being described by Elizabeth King in the late 1950s in association with meningitis in neonates [1]. It is commonly found in the environment including soil, river waters, and reservoirs though rarely infects humans with only about 5–10 cases per state per year reported across the United States [2]. However, there have been increasing reports of this pathogen causing invasive infections such as meningitis, endocarditis, and bacteremia in hosts with risk factors including an immunocompromised status or a prolonged exposure to healthcare settings [3].

The most unique aspect in the management of this pathogen is the frequently encountered multi-drug-resistant phenotype that often results in a susceptibility pattern that can contradict classic dogmas in antimicrobial selection. Drug resistance is frequently conferred by a variety of mechanisms including, but not limited to, extended-spectrum beta-lactamase production as well as chromosomally mediated metallo-beta-lactamases. Thus, empiric treatment with beta-lactams may result in treatment failure and/or relapsing disease [4]. Previous studies have failed to unanimously identify a drug of choice in the treatment of E. meningoseptica infection though piperacillin-tazobactam, trimethoprim-sulfamethoxazole (TMP-SMX), and newer fluoroquinolones appear to have the most consistently low minimum inhibitory concentrations (MICs) upon in vitro testing [5]. This becomes especially problematic as the pathogen is associated with significant patient morbidity and mortality [6]. We report a case of E. meningoseptica bacteremia as a complication of pneumonia in an elderly man with a newly diagnosed immunocompromising condition.

Case

A 78-year-old Caucasian male presented to the hospital with complaints of shortness of breath and productive cough evolving over 3–4 weeks. He denied fever or chills at home but did report an increase in overall fatigue. Cough was productive for green-tinged sputum and small quantities of blood. The patient had a medical history of atrial fibrillation, hypertension, hyperlipidemia, pre-diabetes, chronic obstructive pulmonary disease, coronary artery disease, and prior knee replacement with complication of prosthetic joint infection for which he was maintained on cefadroxil 500 mg twice daily. His other medications included apixaban, lisinopril, metoprolol, budesonide-formoterol, tiotropium, atorvastatin, and metformin. His allergies included heparin and amiodarone. The patient worked as a law enforcement agent for the Society for the Prevention of Cruelty to Animals (SPCA) and reported frequent exposure to farms and various livestock. He also reported owning about a dozen head of cattle at his home farm. He otherwise denied recent hospitalization or domestic/international travel.

In the emergency room, he was afebrile with a heart rate of 75 beats/min, a respiratory rate of 18 breaths/min, blood pressure of 138/73 mmHg, and saturating 90 % on room air. Physical exam was notable for some mild wheezing and scattered rhonchi on lung auscultation, otherwise no significant exam findings. A complete blood count (CBC) was performed with a hemoglobin level of 8.0 g/dL (range 13–18 g/dL; noted to be 13.0 g/dL two years prior), along with concurrent elevation of mean corpuscular volume (MCV) at 123.9 fL (range 80–100 fL), leukopenia of 3.3 × 10³/uL (range 3.8 × 10³/uL-10.6 × 10³/uL) with an ANC of 2.0 × 10³/uL (range 1.4 × 10³/uL-6.3 × 10³/uL), and a normal platelet count of 421 × 10³/uL (range 150 × 10³/uL-440 × 10³/uL). Comprehensive metabolic panel (CMP) revealed an elevated creatinine to 1.4 mg/dL (range 0.6 mg/dL-1.2 mg/dL) as well as an elevated alkaline phosphatase to 181 U/L (range 39 U/L-113 U/L). A chest x-ray was performed showing diffuse patchy airspace disease throughout both lungs with some focal areas of small consolidation. On admission or Hospital Day (HD) 1, empiric antibiotics were withheld due to a low suspicion for a bacterial pneumonia given a low procalcitonin to 0.08 ng/mL (range 0.0–0.1 mg/mL). Blood cultures were never taken during this first admission. He was evaluated by the Pulmonology service for reported hemoptysis and then by the Hematology service for newly diagnosed anemia and leukopenia. A subsequent bone marrow biopsy on HD3 revealed a hypercellular marrow with increased myeloid cells as well as additional morphologic and genotypic findings consistent with myelodysplastic syndrome. He was then started on an oral prednisone taper (60 mg daily for one week followed by 40 mg daily until he was to be seen by the outpatient Pulmonary clinic at which time the steroid taper duration would be finalized) as well as atovaquone 1500 mg oral daily for pneumocystis prophylaxis. Respiratory cultures were obtained from Bronchoalveolar Lavage (BAL) on HD6 but did not reveal any growth within the first twenty-four hours. Having significantly improved, the patient was discharged home with follow-up in place. One day following discharge, respiratory culture was reported to contain 1+ growth of a Gram-negative bacilli which was eventually identified as Elizabethkingia meningoseptica by biochemical panel testing and by MALDI-TOF. These results were communicated to the patient, but, given his significant improvement, no further therapy was prescribed.

One week after discharge, the patient re-presented with relapsed respiratory failure and evidence of hypoxemia. His vital signs revealed a heart rate of 110 beats/min, a respiratory rate of 22 breaths/min, blood pressure of 94/65 mmHg, and an SpO2 of 84 % on room air. He was found to have an elevated WBC count to 15.8 × 10³/µL (range 3.8 × 10³/uL-10.6 × 10³/uL; 2.0 × 10³/µL one week prior upon discharge). A new chest x-ray showed persistent multifocal airspace opacities bilaterally most notably involving the mid to lower lung zones. Given the recent BAL respiratory culture data from the previous admission, Infectious Diseases was consulted for guidance regarding empiric antimicrobial selection. Based on in vitro antimicrobial susceptibility testing results (MicroScan NUC90, Beckman Coulter) performed on the BAL respiratory isolate, piperacillin-tazobactam 3.375 g IV every 6 hours was chosen given an MIC of ≤ 16/4 µg/mL (interpreted as susceptible by the CLSI breakpoint for non-Enterobacterales). This respiratory isolate was also susceptible to ciprofloxacin (MIC ≤0.5 µg/mL), levofloxacin (MIC ≤0.5 µg/mL), and trimethoprim-sulfamethoxazole (MIC ≤2/38 µg/mL), yet was notably resistant to cefepime (MIC ≥32 µg/mL), ceftazidime (MIC ≥32 µg/mL), meropenem (MIC ≥16 µg/mL), and tobramycin (MIC ≥16 µg/mL). The patient was admitted to the medical ICU for significant respiratory distress. Two sets of blood cultures drawn from separate peripheral sites on HD1 later turned positive with Gram-negative bacilli, each confirmed to be Elizabethkingia meningoseptica by MALDI-TOF. The in vitro antimicrobial susceptibility testing for the HD1 bloodstream isolate revealed an identical susceptibility pattern and MICs to the BAL isolate with the exception that piperacillin-tazobactam had increased to intermediate range with a new MIC of 32/4–64/4 µg/mL (concentrations tested by NUC90 are 16/4 µg/mL and 64/4 µg/mL exclusively). Given the patient’s critical illness and evidence of Gram-negative bacteremia, the decision was made to add levofloxacin 750 mg IV daily as a second agent in a dual-therapy approach for suspected disseminated Elizabethkingia infection. The patient ultimately did not require intubation; he improved with antibiotics and supplemental oxygen via high-flow nasal cannula. Repeat blood cultures were performed on HD2 with evidence of clearance. piperacillin-tazobactam was discontinued on the day of discharge (HD5) at which point he was sent on oral levofloxacin 750 mg daily with a plan to complete 21 days of therapy. At a 4-week follow-up, he was clinically well and with no reported symptoms.

Discussion

There are increasing reports of human infection caused by Elizabethkingia meningoseptica, a rarely encountered Gram-negative microbe with a complex drug-resistance profile. As previously stated, this bacterium was initially associated with infections involving the central nervous system in infants. However, more recent data demonstrates it as an emerging nosocomial threat in adults with potential links to contaminated water sources in hospitalized settings [7]. The formation of biofilm is a key factor in its ability to survive in tap water, disinfection fluid, and within various hospital equipment such as ventilators or hemodialysis machines [8]. This places individuals with frequent exposure to healthcare settings, such as the immunocompromised, at the highest risk of infection. This raises further concern for adverse outcomes given this pathogen’s potential to cause significant morbidity with previously reported mortality rates as high as 50 % [9].

As a rising nosocomial threat, early identification and introduction of effective therapy is critical in managing infections caused by E. meningoseptica. The usual resistance profile of this organism further contributes to difficulties achieving complete eradication and the subsequent high rates of initial treatment failure. Elizabethkingia species often express chromosomal metallo-beta-lactamases (MBLs) and extended-spectrum beta-lactamases (ESBLs) which confer resistance to most beta-lactams, including carbapenems, with varying susceptibility to piperacillin-tazobactam as well as colistin, aminoglycosides and fluoroquinolones [10]. Without prior in vitro susceptibility testing readily available, it can be exceedingly challenging to devise an appropriate antimicrobial regiment to combat this hardy organism.

Perhaps the most unique feature of this pathogen is the retained susceptibility to antibiotics more frequently used in the treatment of Gram-positive infections, notably vancomycin and rifampin [11]. The clinical efficacy of using vancomycin to treat E. meningoseptica infections remains controversial. Previous studies first demonstrating high rates of vancomycin susceptibility against Elizabethkingia were mostly based on disk diffusion testing which carries a propensity to overestimate susceptibility for this organism [12]. Thus, newer studies utilizing CLSI standard methods of agar dilution and broth microdilution have demonstrated that many species of the Elizabethkingia genus exhibit increased MICs to vancomycin that renders the drug ineffective for clinical use [13], [14]. In a review of patients who received vancomycin-only or in combination with other susceptible agents, individuals receiving vancomycin-only appeared to have done worse (as defined by mortality) indicating that vancomycin is not suitable for monotherapy in routine clinical use [4]. It should be noted that the patients involved in these studies are generally at elevated risk for adverse outcomes due to their underlying comorbidities, an immunocompromised status, or a prolonged hospital stay. Nonetheless, the degree of growth inhibition in vitro by vancomycin demonstrated by this organism makes it unique among Gram-negative pathogens.

This case report presents a 78-year-old male with newly diagnosed myelodysplastic syndrome who was admitted for relapsed, acute hypoxemic respiratory failure and sepsis in the setting of Elizabethkingia meningoseptica bacteremia likely stemming from an incompletely treated pneumonia. He was ultimately treated successfully with initial dual intravenous antibiotic therapy which was then de-escalated to oral step-down with a fluoroquinolone. Initial dissemination of the organism from a primary lung infection to the bloodstream causing disease relapse was likely a product of the host’s immunocompromised status. The etiology of bacteremia is predicated on the host’s ability to engage, maintain, and execute normal immunologic function; preventing transient bacterial contamination in the bloodstream from evolving into a true bloodstream infection [15]. The patient within the case report did not appear to have evidence of empyema or abscess within the lung leading to higher exposure. He did, however, have Myelodysplastic Syndrome which predisposes to bacteremia from reduced bacterial clearance.

Adult human infections caused by E. meningoseptica appear to be on the rise; especially in healthcare settings involving immunocompromised hosts. Treatment is challenging due to the ubiquitous nature and unique drug-resistant profile of the organism, which likely lends to its significant associated morbidity and mortality. Additional studies are needed to better characterize the pathogen to aid in clinical decision making and early initiation of targeted and thus effective antimicrobial therapy.

Author Statement

All authors have made substantial contributions to the following: the conception and design of the study, or acquisition of data, or analysis and interpretation of data, frafting the article or revising it critically for important intellectual content, and final approval of the version to be submitted.

CRediT authorship contribution statement

P. Rocco LaSala: Writing – review & editing, Writing – original draft. Jeffrey Aeschlimann: Writing – review & editing, Writing – original draft. Sonia Magano: Writing – review & editing, Writing – original draft. David Fraulino: Writing – review & editing, Writing – original draft. Fabrizio Tropea: Writing – review & editing, Writing – original draft.

Consent for publication

Verbal informed consent was obtained from the patient for the publication of this case report with anonymization.

Ethical approval

Obtained.

Funding

No funding was used in the making of this case report.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.