Key Learning Points

What defines a complicated gram-negative bloodstream infection—and why it matters: a complicated GN–BSI is one in which infection control or clinical recovery is delayed due to factors such as impaired host immunity, inadequate source control, persistent bacteremia, or infection with an MDR organism. These cases carry a higher risk of relapse and mortality. Unlike uncomplicated infections, they require more prolonged IV therapy (at least 7 days, with duration individualized to clinical response and source control), definitive source control (e.g., catheter removal or drainage), and close microbiologic follow-up with repeat blood cultures to confirm clearance. Recognizing a case as complicated ensures the clinician prioritizes both the pathogen- and host-related challenges that demand intensified and individualized management (Mermel LA, et al. 2009; Evans L, et al. 2021).

Rapid diagnostics— rapid ID/AST platforms can shorten time to active therapy and de-escalation and support stewardship; however, high-quality evidence for mortality benefit is limited, and impact depends on local feasibility and epidemiology (Ledeboer NA, et al. 2015; Evans L, et al. 2021).

Quick Sequential Organ Failure Assessment (qSOFA) is best viewed as a risk stratification tool at presentation, not a longitudinal response metric; ongoing assessment should rely on clinical trajectory, lactate, and organ dysfunction scoring (e.g., SOFA). (Rudd KE, et al. 2018; Evans L, et al. 2021). Early, definitive source control—such as removal of infected hardware, drainage of collections, or surgical debridement—should occur within hours of diagnosis whenever feasible. It is a key determinant of outcome in sepsis and MDR-GN bloodstream infections (Evans L, et al. 2021).

Frequent creatinine monitoring safeguards both antibiotic effectiveness and patient safety—it enables timely renal dose adjustments, prevents nephrotoxicity, and provides early warning of sepsis-associated kidney injury (Evans L, et al. 2021).

In patients with sepsis, especially with evidence of organ dysfunction, routine coagulation monitoring provides valuable insight into disease severity and helps guide both supportive management and prognosis. Persistent or worsening abnormalities despite therapy signal severe systemic inflammation and warrant urgent reassessment (Evans L, et al. 2021).

Preventive Strategies

Recurrent CRBSIs constitute a significant complication in patients requiring long-term central venous access. Recent evidence supports the use of antimicrobial or taurolidine lock solutions to reduce catheter colonization and infection risk. A systematic review and individual-patient data meta-analysis demonstrated that the use of lock solutions—particularly those containing taurolidine—significantly reduced CRBSI incidence in patients receiving home parenteral nutrition, without increasing adverse events (AEs) or catheter occlusions. These findings support considering lock prophylaxis in high-risk patients or when catheter replacement is not feasible (Wouters Y, et al. 2020).

The global rise of MDR-GN bacteria, particularly those producing MBLs, has become a critical challenge for clinicians and health systems worldwide (Tamma PD, et al. 2024). MBL–producing Enterobacterales—including K. pneumoniae, E. coli, and Enterobacter cloacae—are of special concern because they hydrolyze almost all β-lactam antibiotics, including carbapenems, which have traditionally been the mainstay of therapy for serious gram-negative infections (Paul M, et al. 2022). The rapid dissemination of enzymes such as NDM, VIM, and IMP has led to increasing reports of infections that are refractory to most available agents and associated with high mortality (Tamma PD, et al. 2024). Although several new BL/BLI combinations—such as ceftazidime-avibactam, meropenem-vaborbactam, and imipenem–relebactam—have been introduced in the past decade, these agents inhibit only serine β-lactamases (Ambler class A, C, and some D) and are ineffective against class B enzymes, the MBLs (Ehmann DE, et al. 2012; Paul M, et al. 2022). Consequently, infections caused by MBL–producing organisms remain challenging to treat, often necessitating the use of nephrotoxic or suboptimal therapies such as colistin-based combinations (Paul M, et al. 2022).

Aztreonam–avibactam combines two agents with complementary mechanisms of action: aztreonam, a monobactam that is inherently stable to MBL hydrolysis but susceptible to inactivation by other β-lactamases, and avibactam, a non-BL/BLI that neutralizes serine β-lactamases and thereby protects aztreonam from degradation (Ehmann DE, et al. 2012). This synergy restores activity against organisms that co-produce MBLs and other serine β-lactamases, a phenotype increasingly observed in clinical isolates (Paul M, et al. 2022). Supported by evidence from the phase 3 REVISIT and ASSEMBLE trials, aztreonam–avibactam has now received regulatory approval in both Europe and the United States (2025) for the treatment of complicated intra-abdominal infection (cIAI), hospital-acquired and ventilator-associated pneumonia (HAP/VAP), and complicated UTI (cUTI) in adults with limited or no alternative treatment options (Carmeli Y, et al. 2025; Daikos GL, et al. 2025; EMBLAVEO FDA, 2025).

The REVISIT Trial

The REVISIT trial, published in Lancet Infectious Diseases in 2025, was the pivotal phase 3 study evaluating the efficacy and safety of aztreonam–avibactam compared with meropenem ± colistin for the treatment of serious gram-negative infections. Conducted at 81 centers across 20 countries, this prospective, multinational, open-label, central-assessor–masked, randomized trial enrolled hospitalized adults with either cIAI or HAP/VAP due to, or suspected to be due to, gram–negative pathogens. Patients were randomized in a 2:1 ratio to receive aztreonam–avibactam (with metronidazole for cIAI) or meropenem (with or without adjunctive colistin, at the investigator’s discretion). The study design incorporated stratification by infection type and disease severity—based on Acute Physiology and Chronic Health Evaluation II (APACHE II) score for cIAI, and ventilation status for HAP/VAP—to ensure balanced comparison across clinically relevant subgroups (Carmeli Y, et al. 2025).

Key inclusion criteria comprised adults aged ≥ 18 years with microbiologically confirmed or strongly suspected cIAI or HAP/VAP caused by gram-negative organisms and an anticipated requirement for IV antibiotics for 5–14 days (for cIAI) or 7–14 days (for HAP/VAP). Exclusion criteria included infections caused by intrinsically resistant pathogens such as Acinetobacter baumannii, infection limited to gram-positive species, or receipt of effective systemic antibiotics for more than 24 hours before randomization, unless prior therapy had clearly failed (Carmeli Y, et al. 2025).

The primary endpoint was clinical cure at the test-of-cure (TOC) visit, defined as resolution of baseline signs and symptoms such that no further antibiotic therapy was needed, assessed within 3 days before or after day 28. An independent, masked committee adjudicated cure rates to minimize bias inherent to the open-label design. In the intention-to-treat (ITT) population, 68.4% of patients in the aztreonam–avibactam arm achieved clinical cure, compared with 65.7% in the meropenem group (difference 2.7%, 95% CI –6.6 to 12.4). Similar results were observed in the clinically evaluable (CE) population, confirming noninferiority despite the descriptive design. Among infection types, cure rates were higher in cIAI (76.4% vs 74.0%) than in HAP/VAP (45.9% vs 41.7%), consistent with disease severity differences (Carmeli Y, et al. 2025).

Microbiological response rates closely mirrored clinical outcomes: favorable per-patient eradication or presumed eradication was achieved in 76% of aztreonam–avibactam–treated patients and 74% of meropenem-treated patients in the microbiological ITT set. All-cause 28-day mortality was low and numerically lower with aztreonam–avibactam (4%) than with meropenem (7%). In patients with MBL–positive baseline pathogens—a small subset of 10 patients—clinical cure was achieved in 2 of 7 (28.6%) in the aztreonam–avibactam group and 2 of 3 (66.7%) in the meropenem ± colistin group; interpretation of these results was limited by the small numbers and confounding prior treatment failures (Carmeli Y, et al. 2025).

Safety findings were consistent with the known profile of aztreonam monotherapy. Treatment-emergent adverse events (AEs) occurred in 64% of patients in both arms, most commonly gastrointestinal symptoms, liver enzyme elevations, and mild hypersensitivity reactions. Serious AEs were observed in 19% (aztreonam–avibactam) and 18% (meropenem), with no treatment-related serious events in the aztreonam–avibactam arm. Liver disorder–related AEs occurred in 18% and 15% of patients respectively, with no Hy’s Law cases reported, and no new safety signals identified (Carmeli Y, et al. 2025).

Interpretation and limitations: the REVISIT trial demonstrated that aztreonam–avibactam provides comparable efficacy and tolerability to meropenem for serious infections caused by gram-negative bacteria, including in patients with previous antibiotic failure. The study’s descriptive, non-inferential design was intended to meet regulatory requirements for pathogens with limited treatment options, but it precluded formal statistical hypothesis testing. Additionally, despite targeted recruitment in regions with endemic carbapenem resistance, the number of patients with confirmed MBL–producing organisms was low, limiting conclusions about clinical efficacy in this specific subgroup. Nevertheless, the findings, together with the complementary ASSEMBLE trial, formed the foundation for regulatory approval in 2025 and established aztreonam–avibactam as a promising therapeutic option for complex, MDR-GN infections (Carmeli Y, et al. 2025; Daikos GL, et al. 2025).

The ASSEMBLE Trial

The ASSEMBLE trial, published in JAC–Antimicrobial Resistance in 2025, was a pivotal phase 3, prospective, multicenter study specifically designed to evaluate aztreonam–avibactam for the treatment of serious infections caused by confirmed MBL–producing gram–negative pathogens. Unlike the broader REVISIT trial, which focused on cIAI and HAP/VAP of mixed resistance mechanisms, ASSEMBLE directly targeted the patient population with the most limited therapeutic options—those infected with MBL–positive, MDR Enterobacterales or non-fermenters (Daikos GL, et al. 2025; Carmeli Y, et al. 2025).

This was a randomized, open-label, parallel-group, assessor-blinded phase 3 trial that compared aztreonam–avibactam (plus metronidazole for cIAI) with best available therapy (BAT), the latter typically consisting of colistin-based or combination regimens. Hospitalized adults with microbiologically confirmed cIAI, cUTI, HAP/VAP, or BSI due to MBL–producing pathogens were eligible. Importantly, inclusion required genotypic confirmation of at least one MBL–positive gram-negative organism—such as NDM, VIM, or L1–producing strains—within 7 days of screening. Randomization followed a 2:1 allocation ratio favoring aztreonam–avibactam, stratified by infection type to ensure balanced representation across infection syndromes (Daikos GL, et al. 2025).

The primary endpoint—clinical cure at TOC on day 28 ± 3 days, adjudicated by a blinded central committee—was achieved in 5 of 12 (42%) patients receiving aztreonam–avibactam vs 0 of 3 (0%) in the BAT group. Similarly, per-patient microbiological response at TOC was favorable in 6 of 12 (50%) aztreonam–avibactam–treated patients and none in the BAT arm, with consistency between clinical and microbiological outcomes. Twenty-eight–day all-cause mortality was 8% (1/12) for aztreonam–avibactam compared with 33% (1/3) for BAT, although the very small comparator group limited interpretation. The most common baseline pathogen was K. pneumoniae, and MBL subtypes identified included NDM–1, NDM–5, VIM–2, and L1, reflecting global diversity in resistance mechanisms. All baseline Enterobacterales isolates in the aztreonam–avibactam group demonstrated MIC ≤ 4 mg/L, consistent with in vitro susceptibility per EUCAST criteria (Daikos GL, et al. 2025; EUCAST, 2025).

The safety profile was favorable and consistent with prior studies. AEs occurred in 11 of 12 (92%) patients in the aztreonam–avibactam group, mostly mild or moderate in intensity. Treatment-related AEs were infrequent (17%) and included transient gastrointestinal symptoms and reversible liver enzyme elevations. Notably, no treatment-related serious AEs occurred in the aztreonam–avibactam group, compared with one (acute kidney injury) in the BAT group. The absence of Hy’s law cases or treatment discontinuations due to adverse effects further reinforced the regimen’s tolerability (Daikos GL, et al. 2025).

The study provided the first prospective, randomized evidence focused exclusively on MBL–producing pathogens—an area previously supported only by in vitro and observational data. Despite its small size, its findings are significant. ASSEMBLE demonstrated both clinical benefit and microbiological efficacy of aztreonam–avibactam compared with existing salvage therapies, supporting its regulatory approval and its positioning as the preferred treatment option for infections caused by MBL–producing gram-negative bacteria. The results also underscored the formidable challenges in conducting interventional trials within this ultra-resistant patient population, highlighting the need for global surveillance-linked enrollment strategies and adaptive designs in future antimicrobial development programs (Daikos GL, et al. 2025).

Pharmacokinetics/pharmacodynamics (PK/PD) Considerations

Optimal dosing of antibiotics for MDR-GN infections requires understanding PK/PD principles to maximize efficacy while minimizing toxicity. For β-lactam antibiotics, including aztreonam and the newer BL/BLI combinations, the critical PK/PD parameter is the percentage of the dosing interval during which the free drug concentration remains above the MIC (fT > MIC) (Tamma PD, et al. 2024; Paul M, et al. 2022). Target attainment of 40%–70% fT > MIC is generally associated with bacteriostatic effect, while approaching 100% fT > MIC maximizes bactericidal activity and clinical outcomes in serious infections (Tamma PD, et al. 2024).

For aztreonam–avibactam, the approved dosing regimen was optimized using population PK modeling from the phase 3 REVISIT and ASSEMBLE trials to achieve joint PK/PD targets: aztreonam 60% fT > MIC at 8 mg/L and avibactam 50% fT > threshold concentration (CT) at 2.5 mg/L across varying degrees of renal function (EMBLAVEO FDA, 2025). The fixed 3:1 ratio (aztreonam 2 g plus avibactam 0.67 g) administered as a 3-hour infusion achieves a probability of target attainment exceeding 89%–99% at steady state for patients with creatinine clearance ranging from normal to moderately impaired (> 30 mL/min) (EMBLAVEO FDA, 2025).

Renal dose adjustment is essential for renally cleared agents to prevent both subtherapeutic exposure—which may lead to treatment failure and resistance emergence—and supratherapeutic accumulation resulting in toxicity (Paul M, et al. 2022; EMBLAVEO FDA, 2025). In critically ill patients with sepsis, altered PKs due to augmented renal clearance, increased volume of distribution, or acute kidney injury may necessitate therapeutic drug monitoring or extended infusions to optimize fT > MIC and improve clinical outcomes (Evans L, et al. 2021; Tamma PD, et al. 2024). For immunocompromised patients or those with high bacterial inocula, achieving higher PK/PD targets (approaching 100% fT > MIC) may be necessary to ensure adequate bacterial killing and prevent treatment failure (Paul M, et al. 2022).

Clinical Positioning and Guideline Context

Within today’s guidance landscape, aztreonam–avibactam slots squarely into the niche of infections due to MBL–producing Enterobacterales, especially when isolates co-produce a serine β-lactamase (e.g., ESBL, AmpC, KPC, OXA-48) that would otherwise inactivate aztreonam. The IDSA 2024 AMR guidance explicitly operationalizes this by listing the co-administration strategy of ceftazidime-avibactam plus aztreonam (simultaneous Y-site dosing) for MBL producers. It provides dosing to support bedside use—functionally positioning aztreonam protected from MBLs while avibactam shuts down accompanying serine enzymes (the same pharmacologic logic now delivered in a single vial with aztreonam–avibactam) (Tamma PD, et al. 2024). In parallel, contemporary overviews summarizing ESCMID 2022 and IDSA recommendations depict MBL producer pathways that favor either ceftazidime-avibactam plus aztreonam or cefiderocol, underscoring where aztreonam–avibactam naturally sits once available: Directed therapy for proven or strongly suspected MBL Enterobacterales across serious infections, rather than broad empiric use (Paul M, et al. 2022; Tamma PD, et al. 2024).