Cefiderocol Heteroresistance in NDM-Producing Enterobacterales: A Complex Polymicrobial Infection Successfully Salvaged With Ceftazidime-Avibactam Plus Aztreonam

Introduction

Carbapenem-resistant Enterobacterales (CRE) producing metallo-β-lactamases (MBLs) such as New Delhi metallo-β-lactamase (NDM) have emerged as formidable nosocomial pathogens associated with limited treatment options and high mortality [1]. NDM enzymes hydrolyze nearly all β-lactams, including carbapenems, and are often co-expressed with serine β-lactamases, which inactivate aztreonam, making conventional β-lactam therapy ineffective [2]. The Infectious Diseases Society of America (IDSA) currently recommends either ceftazidime-avibactam plus aztreonam or cefiderocol monotherapy as first-line treatments for NDM-producing Enterobacterales infections, although direct clinical comparisons are limited [3]. Cefiderocol, a siderophore cephalosporin, functions as a “Trojan horse” antibiotic by utilizing bacterial iron transport systems to improve entry and overcome multiple resistance mechanisms [4]. However, emerging evidence indicates that heteroresistance – a phenomenon in which a small bacterial subpopulation exhibits higher resistance levels than the dominant population – can lead to rapid development of cefiderocol resistance during treatment, even in isolates initially deemed susceptible [3–5]. Studies show that up to one-quarter of carbapenem-resistant E. coli isolates display cefiderocol heteroresistance, which is linked to treatment failure in animal models and clinical relapse [3,4,6].

Importantly, conventional antimicrobial susceptibility testing methods – such as disk diffusion and broth microdilution – evaluate the dominant bacterial population and may fail to detect minority resistant subpopulations, leading to false susceptibility interpretation and delayed recognition of therapeutic failure. Advanced techniques, including population analysis profiling and time-kill assays, are often required to identify these subpopulations but are not routinely performed in clinical laboratories [3–6]

We present a complex 5-month case of polymicrobial prosthetic joint infection (PJI) complicated by NDM-producing Klebsiella pneumoniae and Enterobacter cloacae in a patient with a prolonged postoperative course. This case highlights the diagnostic and therapeutic challenges associated with heteroresistant CRE, emphasizing the clinical implications of discordant cefiderocol susceptibilities and the successful salvage with ceftazidime-avibactam plus aztreonam combination therapy.

Case Report

A 64-year-old woman with a history of chronic kidney disease, previous deep venous thrombosis on anticoagulation, and hypoxic respiratory failure arrived on July 10, 2025, from a skilled nursing facility with sudden mental status changes and respiratory distress. Her orthopedic history included a left total knee arthroplasty (TKA) on March 27, 2025, complicated by recurrent prosthetic joint infections (Figure 1). She was readmitted on April 30 with knee pain and effusion, with cultures growing Candida albicans and Acinetobacter baumannii complex. She underwent multiple debridements: May 7 (articulating spacer with + Candida albicans), May 22 (static spacer with + Candida albicans), and May 29 (dermal regeneration template with + Stenotrophomonas maltophilia). A Hickman catheter was placed on June 2 for 6 weeks of intravenous ampicillin-sulbactam and micafungin, ending therapy on June 28. On June 13, she was readmitted with septic shock from an infected catheter and necrotic knee wound, followed by another incision and drainage procedure. Blood and catheter cultures were negative. On June 26, intraoperative cultures from her knee grew Klebsiella pneumoniae susceptible to ceftriaxone and cefepime (NDM-negative). She was discharged on oral minocycline and fluconazole, with treatment ending August 11. Upon presentation on July 10, 2025, she was intubated and placed on vasopressors for septic shock. CT (computed tomography) of the chest showed multifocal airspace disease with air bronchograms, and abdominal CT revealed colonic pneumatosis with possible contained perforation or mesenteric venous gas. She was started on vancomycin, cefepime, and micafungin empirically. Sputum cultures grew Klebsiella pneumoniae, and her wound cultures from the left knee grew Enterobacter cloacae and Klebsiella pneumoniae (susceptibilities at that time were pending). Antimicrobial therapy was empirically expanded to meropenem and minocycline; however, meropenem was stopped due to seizure concerns, and she was transferred to a Level III hospital on July 11 (Table 1). Upon transfer, she remained in septic shock with leukocytosis, and repeat imaging confirmed colonic pneumatosis without perforation. Antimicrobials were adjusted to empiric cefiderocol. She underwent exploratory laparotomy on July 12, which showed viable bowel. During this admission, repeat cultures on July 10 showed NDM-positive Enterobacter cloacae and Klebsiella pneumoniae from the wound, and NDM-positive Klebsiella pneumoniae from respiratory samples, indicating extensive drug resistance. Due to the polymicrobial infection and septic shock, the limb was considered unsalvageable, leading to a left above-knee amputation (AKA) on July 16. Her cultures had eventually shown discordant susceptibilities to cefiderocol: respiratory isolates were resistant, while wound isolates remained susceptible – reflecting heteroresistance within the same species. The microbiology lab confirmed positive synergy between ceftazidime-avibactam and aztreonam, and she was switched to this combination on July 16 for 14 days. She later underwent revision and primary closure of the left AKA site on July 19, and her condition stabilized.

Following transition to ceftazidime-avibactam plus aztreonam and surgical source control with left above-knee amputation, the patient demonstrated gradual clinical improvement with resolution of septic shock and stabilization of respiratory status. Due to prolonged ventilator dependence in the setting of critical illness and severe deconditioning, she subsequently underwent tracheostomy and percutaneous endoscopic gastrostomy (PEG) tube placement for long-term respiratory and nutritional support. Over the following weeks, her hemodynamic status remained stable, inflammatory markers improved, and no further microbiologic evidence of persistent CRE infection was identified. After completion of antimicrobial therapy and multidisciplinary rehabilitation planning, she was discharged to a long-term acute care (LTAC) facility in improved clinical condition for continued ventilatory weaning, nutritional optimization, and physical rehabilitation.

Discussion

Carbapenem-resistant Enterobacterales (CRE) are defined by resistance to at least 1 carbapenem or production of a carbapenemase enzyme, with diverse mechanisms, including carbapenemase production and non-carbapenemase pathways such as ESBL overexpression and porin loss [3]. In the United States, Klebsiella pneumoniae carbapenemase (KPC) remains the most common enzyme, although metallo-β-lactamases (MBLs) such as NDM, VIM, and IMP are increasing, reflecting evolving epidemiology [5–7]. Because treatment depends on resistance mechanisms, rapid molecular identification is essential [9]. No single β-lactam/β-lactamase inhibitor reliably covers MBL producers; therefore, recommended therapies include ceftazidime-avibactam plus aztreonam or cefiderocol monotherapy [3]. Aztreonam remains stable against MBL hydrolysis, while avibactam inhibits co-produced serine β-lactamases, restoring activity; in vitro susceptibility approaches 90% [10]. Clinical data demonstrate lower mortality with ceftazidime-avibactam/aztreonam (≈19%) compared with polymyxin- or tigecycline-based regimens (≈44%) [11]. Cefiderocol also shows broad activity against MBL-producing isolates, with susceptibility rates near 92% and clinical cure rates around 80%, although emerging resistance – particularly among NDM producers – has been reported [12,13]. While no head-to-head trials exist, ceftazidime-avibactam/aztreonam may offer advantages in polymicrobial infections or when resistance evolution is a concern, whereas cefiderocol provides convenient monotherapy with favorable pharmacodynamics. Heteroresistance remains clinically important, especially for cefiderocol, where TonB transport mutations, AmpC alterations, and increased NDM expression can drive treatment failure despite initial susceptibility [15,16]. Current IDSA guidance recommends ceftazidime-avibactam plus aztreonam or cefiderocol as first-line therapy for NDM-producing CRE, emphasizing early source control, repeat susceptibility testing, and individualized treatment decisions in severe infections [3,17–19].

Earlier initiation of combination therapy with ceftazidime-avibactam plus aztreonam may theoretically have reduced selective antibiotic pressure and limited the emergence of cefiderocol heteroresistance in this high-inoculum, polymicrobial infection [19]. However, initial susceptibility results supported cefiderocol use according to contemporary guidelines, and the transition to targeted combination therapy was made promptly once discordant susceptibilities and clinical deterioration were recognized. This case therefore highlights the importance of dynamic antimicrobial reassessment rather than suggesting that upfront combination therapy is universally required.

Conclusions

This case highlights the clinical consequences of prolonged empiric antimicrobial exposure and the emergence of NDM-producing CRE with cefiderocol heteroresistance. While both cefiderocol and ceftazidime-avibactam plus aztreonam remain key therapeutic options for MBL-producing Enterobacterales, treatment selection should be individualized based on infection burden, prior antimicrobial exposure, and evolving susceptibility patterns. Discordant susceptibility results should prompt repeat testing and consideration of combination therapy. Early recognition of heteroresistance and timely addition of targeted therapy, combined with definitive source control, are critical to improving outcomes in severe multidrug-resistant infections.