Principles of antimicrobial therapy of Pseudomonas aeruginosa infections

INTRODUCTION — 

Pseudomonas aeruginosa is a key gram-negative aerobic bacillus in the differential diagnosis of a number of infections. This organism is important because it is often antibiotic resistant and can cause severe hospital-acquired infections associated with a high mortality rate, especially in immunocompromised hosts.

The principles of antimicrobial treatment of infections caused by P. aeruginosa will be reviewed here. Discussion on the epidemiology, pathogenesis, clinical manifestations, diagnosis, and treatment of specific types of pseudomonal infections can be found elsewhere:

●(See "Epidemiology, microbiology, and pathogenesis of Pseudomonas aeruginosa infection".)

●(See "Pseudomonas aeruginosa pneumonia".)

●(See "Pseudomonas aeruginosa bacteremia and endocarditis".)

●(See "Pseudomonas aeruginosa skin and soft tissue infections".)

●(See "Pseudomonas aeruginosa infections of the eye, ear, urinary tract, gastrointestinal tract, and central nervous system".)

Treatment issues related to P. aeruginosa lung infection in patients with cystic fibrosis are also discussed separately. (See "Cystic fibrosis: Antibiotic therapy for chronic pulmonary infection", section on 'Pseudomonas aeruginosa' and "Cystic fibrosis: Antibiotic therapy for pulmonary exacerbations".)

GENERAL PRINCIPLES OF TREATMENT — 

The following principles apply to the management of P. aeruginosa infections:

●The risk of antibiotic resistance, both intrinsic and acquired, is an important consideration when selecting empiric or directed therapy. (See 'Antimicrobial resistance' below.)

●Monotherapy is generally adequate, although combination therapy is indicated in certain high-risk patients and in severe infections. (See 'Role of combination antimicrobial therapy' below.)

●Prompt initiation of antimicrobial therapy is important, as delayed therapy correlates with increased mortality.

●Source control is important. For example, infected catheters and other implanted devices should be removed, abscesses should be drained, and obstructions should be relieved.

ANTIMICROBIAL RESISTANCE — 

P. aeruginosa is intrinsically resistant to numerous antibiotics and can acquire resistance to other agents during therapy. Some strains are multidrug resistant (MDR; ie, they are resistant to three or more classes of antibiotics) [1]. These features limit the choices of antibiotics for P. aeruginosa.

Definitions — A joint initiative by the European Centre for Disease Prevention and Control (ECDC) and the United States Centers for Disease Control and Prevention (CDC) in 2011 proposed specific definitions for characterizing drug resistance in organisms that cause many health care-associated infections [2]. The following definitions were established, based on the extent of resistance to antibiotic classes to which Pseudomonas does not have intrinsic resistance (table 1):

●Multidrug-resistant (MDR) – Isolate is not susceptible to at least one agent in three or more antibiotic classes.

●Extensively drug-resistant – Isolate is not susceptible to at least one agent in all but two or fewer antibiotic classes.

●Pandrug-resistant – Isolate is not susceptible to all agents.

●Difficult to treat (DTR) – Isolates test intermediate or resistant to all reported carbapenems, beta-lactams, and fluoroquinolones [3].

Epidemiology of resistance — The global burden of antimicrobial resistance in clinical isolates of P. aeruginosa is high. However, national surveillance data should be interpreted with caution because local resistance rates are highly variable and closely linked with local antimicrobial usage patterns and patient type.

Data collected from over 3000 hospitals in the United States' National Healthcare Safety Network from 2018 to 2021 revealed the following rates of MDR among P. aeruginosa isolates [4]:

●Ventilator-associated pneumonia – 15.4 percent

●Central line-associated blood stream infection – 14.2 percent in intensive care units (ICUs) and 11.9 percent in hospital wards

●Catheter-associated urinary tract infection – 8.7 percent in ICUs and 10.6 percent in hospital wards

●Surgical site infection – 3.9 percent

A subsequent study analyzing data from 890 hospitals in the United States reported that the rate of MDR P. aeruginosa infections among hospitalized patients decreased by 30 percent between 2012 and 2017 (from 13.10 to 9.43 cases per 10,000 hospitalizations) [5].

Rates of resistance are higher in resource-limited than in resource-abundant settings. In a study conducted from 2015 to 2020 on P. aeruginosa health care-associated infections, isolates were collected from over 600 ICUs in 45 resource-limited countries; rates of resistance to amikacin and imipenem were 31.5 and 50.7 percent, respectively [6]. In contrast, corresponding resistance rates in ICUs in the United States were 10 and 26 percent, respectively.

Antimicrobial resistance among P. aeruginosa is associated with increased length of hospitalization and increased mortality [7]. Risk factors for infection with resistant P. aeruginosa isolates include the following [8-12]:

●ICU stay

●Bedridden status

●Presence of invasive devices

●Prior use of certain antibiotics, including broad-spectrum cephalosporins, aminoglycosides, carbapenems, fluoroquinolones

●Diabetes mellitus

●Surgery

Nosocomial outbreaks of resistant pseudomonal infections have been reported in health care facilities around the world [13-16]. Outbreaks have also been associated with contaminated medical devices, such as endoscopes and transesophageal probes [12,17]. In 2023, a nation-wide outbreak in the United States of extensively drug-resistant P. aeruginosa, resistant to all antipseudomonal agents including ceftolozane-tazobactam and ceftazidime-avibactam as well as polymixins, was found to be associated with contaminated over-the-counter artificial tears [11].

Prior use of carbapenem, including ertapenem (which does not have antipseudomonal activity), has been repeatedly identified as a risk factor for carbapenem resistance [18-20]. Additional risk factors include male gender, ICU stay, and urinary bladder catheterization longer than seven days.

Emergence of antibiotic resistance during therapy for P. aeruginosa infections, resulting in increased rates of morbidity and mortality and higher costs, is a well-known problem. As an example, in a study involving 3393 P. aeruginosa isolates from ICU and non-ICU settings in a tertiary-care center, standard antibiograms were less able to predict susceptibility as the duration of hospitalization increased [21]. Similarly, in an observational study, resistant strains of P. aeruginosa emerged in 10.2 percent of 271 cases during treatment with four individual antipseudomonal agents [22]. Ceftazidime was associated with the lowest risk of emergent resistance, and imipenem with the highest risk. Emergence of resistance during therapy is a challenge with newer agents that have activity against Pseudomonas; this issue is discussed elsewhere. (See 'Emergence of resistance to novel agents' below.)

Acquired resistance is mediated by various mechanisms, including degrading enzymes, reduced permeability, and active efflux via mutation or acquisition of exogenous resistance determinants [23,24]. (See "Epidemiology, microbiology, and pathogenesis of Pseudomonas aeruginosa infection".)

ANTIBIOTICS WITH ANTIPSEUDOMONAL ACTIVITY — 

A number of antimicrobial agents have activity against P. aeruginosa isolates (table 2).

Intravenous antibiotics

Dosing strategies — Due to high levels of intrinsic and acquired resistance among pseudomonal isolates, higher doses of traditional antibiotics are often required to achieve sufficient levels.

Some beta-lactams are often administered as extended infusions (eg, over two to four hours) to maximize the duration that drug concentrations exceed the minimum inhibitory concentration (MIC) of the organism [25-30]. Although clinical data are limited, administration of prolonged infusions for certain beta-lactams and carbapenems is appropriate, especially for critically ill patients or for isolates with elevated MICs to the agent. (See "Prolonged infusions of beta-lactam antibiotics".)

Dosing for antipseudomonal antibiotics is listed in the table (table 2).

Antibiotic options — Multiple antibiotics from different classes are available to treat pseudomonal infections. Certain first-line agents are preferred due to extensive clinical experience, clinical trial results, and antibiotic stewardship considerations.

First-line agents — The following antibiotics, grouped by class, can usually be used alone as single agents if in vitro testing shows they are susceptible. The selection of individual agents from this group depends on factors such as the site of the infection, local resistance rates of P. aeruginosa, prior culture results, the patient's history of allergies, and the agents' availability in the local hospital formulary.

Doses for intermittent or prolonged infusion are outlined in the tables (table 2).

●Antipseudomonal penicillins in combination with a beta-lactamase inhibitor include:

•Piperacillin-tazobactam

Although piperacillin has antipseudomonal activity, the addition of tazobactam inhibits AmpC-type beta-lactamases and results in a lower MIC than piperacillin.

•Ticarcillin-clavulanate (not available in most countries)

●Cephalosporins with antipseudomonal activity include:

•Ceftazidime

•Cefoperazone (not available in the United States)

•Cefepime

●Monobactam:

•Aztreonam

●Fluoroquinolones:

•Ciprofloxacin

Levofloxacin has no advantage over ciprofloxacin for infections due to P. aeruginosa since its additional spectrum of coverage is usually unnecessary and potentially harmful. Levofloxacin is primarily indicated for treatment of respiratory tract infections when additional empiric P. aeruginosa coverage is warranted and in rare situations, such as a culture-positive polymicrobial infection that includes susceptible strains of streptococci and P. aeruginosa. A study published in 2023 showed that levofloxacin required fewer passages in P. aeruginosa isolates when compared with ciprofloxacin to reach their breakpoints. Thus, resistance against levofloxacin develops faster than resistance against ciprofloxacin [31]. We do not advise using other quinolone agents, such as moxifloxacin, for treatment of P. aeruginosa.

●Carbapenems:

Carbapenems with activity against Pseudomonas spp include meropenem and imipenem. In vitro studies have shown that carbapenem's MICs are lower with meropenem than with imipenem [32]. It is unknown whether this in vitro potency difference translates into any difference in clinical outcomes [22,33]. Ertapenem has no activity against Pseudomonas spp, and doripenem should be avoided for ventilator-associated pneumonia due to poorer outcomes in some studies.

All carbapenems have been associated with emergent resistance during therapy; thus, we reserve their use for the treatment of P. aeruginosa infections resistant to other agents or in polymicrobial infections [33].

Role of aminoglycosides — Aminoglycosides (tobramycin, gentamicin, amikacin) are active against P. aeruginosa but are generally not used as single agents because of inadequate clinical efficacy at most sites. As examples, aminoglycosides should not be used as monotherapy for pneumonia because they perform poorly in an acidic environment, and aminoglycosides used as monotherapy for bacteremia are associated with high mortality rates [34]. Instead, aminoglycosides are frequently used in combination with other antibiotics for empiric therapy, pending susceptibility results or for the treatment of select serious infections.

Aminoglycosides can be used as a single agent for the treatment of lower urinary tract infections (eg, cystitis).

When using an aminoglycoside as part of therapy for an infection with a high risk of P. aeruginosa, we favor tobramycin over gentamicin or amikacin as it has greater intrinsic antipseudomonal activity; however, it may not be widely available.

The dosing of aminoglycosides is discussed in detail elsewhere. (See "Dosing and administration of parenteral aminoglycosides".)

Antibiotics reserved for multidrug-resistant isolates — Several novel agents have activity against P. aeruginosa but are generally reserved for serious infections with isolates resistant to other agents because of cost and antibiotic stewardship concerns (and, for polymyxins, because of poor activity and toxicity, only to be used when no other options are available) (see 'Antibiotic selection' below). Doses are listed in the table (table 2).

●Novel beta-lactam-beta-lactamase inhibitor combinations:

•Ceftolozane-tazobactam [35,36]

•Ceftazidime-avibactam [37-40]

●Novel cephalosporin:

•Cefiderocol [41-43]

●Novel carbapenem-beta-lactamase combination:

•Imipenem-cilastatin-relebactam

●Polymyxins:

•Polymyxin B

•Colistin

Polymyxins are generally used as part of a combination regimen when treating Pseudomonas infection. Dosing of polymyxin B and colistin are discussed in detail elsewhere. (See "Polymyxins: An overview", section on 'Dosing and administration'.)

Other potential therapeutic options are still in development [44].

Oral antibiotics — Fluoroquinolones are the only antibiotic class available as an oral formulation that is reliably active against P. aeruginosa. As with other classes of antibiotics, higher doses of certain fluoroquinolones (eg, ciprofloxacin) are often used to treat Pseudomonas.

For infections due to P. aeruginosa alone, we suggest ciprofloxacin instead of levofloxacin because levofloxacin's additional spectrum of coverage is usually unnecessary; doses are found in the table (table 2). Prulifloxacin (600 mg once daily) is another oral fluoroquinolone available in some European countries.

We suggest not using other quinolone agents (eg, moxifloxacin) for treatment of P. aeruginosa. A new fluoroquinolone, delafloxacin, was approved in 2017 by the US Food and Drug Administration (FDA). This novel antibiotic has been reported to exhibit a broad-spectrum activity against resistant pathogens, including multidrug-resistant (MDR) P. aeruginosa. Delafloxacin has been shown to have excellent lung and biofilm penetration and enhanced activity at low pH environments, such as in cystic fibrosis. A study demonstrated favorable in vitro activity of delafloxacin compared with alternative fluoroquinolones against MDR P. aeruginosa. Thus, it may be considered as a treatment option for MDR strains [45].

We also suggest not using fosfomycin as a single agent due to an increased risk of baseline resistance and rapid emergence of resistance while on therapy.

Alternative formulations — Certain antipseudomonal antibiotics are also available in inhaled, topical, intravitreal, intraventricular, and impregnated cement formulations for specific indications.

Inhaled antibiotics with activity against P. aeruginosa are sometimes used as an adjunct to treatment of pneumonia due to drug-resistant P. aeruginosa and to prevent exacerbations of bronchiectasis (eg, in the setting of cystic fibrosis) [46]. Inhaled administration has the theoretic advantage of targeting drug levels to bronchial secretions in patients with pneumonia while reducing potential systemic side effects. (See "Pseudomonas aeruginosa pneumonia" and "Cystic fibrosis: Antibiotic therapy for chronic pulmonary infection", section on 'Inhaled antibiotics' and "Bronchiectasis in adults: Maintaining lung health", section on 'Inhaled antibiotics'.)

Other antibiotic formulations are discussed elsewhere. (See "Bacterial endophthalmitis", section on 'Intravitreal antibiotics' and "Prosthetic joint infection: Treatment", section on 'Resection arthroplasty with reimplantation' and "Health care-associated meningitis and ventriculitis in adults: Treatment and prognosis", section on 'Intrathecal and intraventricular therapy'.)

ROLE OF COMBINATION ANTIMICROBIAL THERAPY — 

One of the most controversial management questions involves the use of combination versus monotherapy for serious infections due to P. aeruginosa. High-quality data informing the decision are lacking.

Empiric combination therapy — Combination therapy is used by many clinicians for empiric coverage of known or suspected pseudomonal infections, and is usually discontinued once susceptibility results become available.

Indications — We suggest the use of two agents from different classes with in vitro activity against P. aeruginosa for empiric treatment of serious infections known or suspected to be caused by P. aeruginosa when there is high risk of antimicrobial resistance or in hosts for whom inappropriate antibiotic therapy would likely be associated with an especially high mortality. Such circumstances include the following:

●When signs of severe sepsis or septic shock are present

●Neutropenic patients with bacteremia

●Burn patients (who have a high incidence of multidrug-resistant [MDR] P. aeruginosa infections) with serious infections

●In other settings where the incidence of resistance to the chosen antibiotic class is high (eg, >10 to 15 percent)

In other circumstances, empiric treatment using only one antipseudomonal agent is appropriate.

Our rationale for using empiric combination therapy in the settings listed above is to increase the likelihood that at least one of the agents in the combination is active against the infecting strain of Pseudomonas. Further discussion of the rationale for combination therapy is discussed below. (See 'Rationale' below.)

Selection of agents — The choice of agents for empiric therapy depends on numerous factors, such as the site and severity of the infection, local resistance rates of P. aeruginosa, prior culture susceptibility results, allergies, and the availability of specific agents in the local hospital formulary.

If empiric combination therapy is used, two agents with different mechanisms of action should generally be chosen based on prior patient isolates and local resistance rates. In general, we suggest a beta-lactam as the first agent and an aminoglycoside as the second, provided there are no contraindications to the use of either. The role and selection of aminoglycosides is discussed above. (See 'Role of aminoglycosides' above.)

Aminoglycosides should be avoided in patients with renal insufficiency or in those hospitalized in institutions with a high percentage of P. aeruginosa resistant to aminoglycosides. In such cases, we use a fluoroquinolone as a second agent. One study suggested better outcomes when using a fluoroquinolone rather than an aminoglycoside as the second agent [47], but further data are warranted before routinely favoring a fluoroquinolone over an aminoglycoside in this setting.

Rationale — For empiric therapy, the rationale for using combination therapy is detailed below:

Ensure receipt of an active agent — The best rationale for the use of combination therapy is to provide initial broad spectrum of activity when there is risk for MDR P. aeruginosa so that if the P. aeruginosa is resistant to one agent, it may be susceptible to the other. Delays in active antimicrobial therapy of P. aeruginosa infections have been associated with increased mortality, and the role of a second agent may thus be to cover possible resistant pathogens when resistance rates to the primary agent are high [1,48-52]. Delay in treatment has been linked to increased mortality even when a patient is considered clinically stable at the time of initial evaluation [53].

This is a similar rationale to that used to support empiric combination therapy for gram-negative bacillary bacteremia in other settings. (See "Gram-negative bacillary bacteremia in adults", section on 'Indications and rationale for combination therapy'.)

In one retrospective study of patients with ventilator-associated pneumonia due to P. aeruginosa, empiric combination therapy was associated with a higher likelihood that appropriate antibiotic therapy was initially used, although an impact on mortality was not demonstrated [54]. A multinational, retrospective study of 294 episodes of P. aeruginosa bacteremic pneumonia in neutropenic individuals found that appropriate empiric combination therapy was associated with improved overall survival (adjusted hazard ratio 0.46, 95% CI 0.27-0.78) [55]. Similarly, a multicenter retrospective cohort study analyzed data from 14 public hospitals in Italy and demonstrated that adequate empirical combination therapy was associated with a lower 30-day all-cause mortality compared with monotherapy among patients with septic shock due to P. aeruginosa blood stream infections. However, the administration of definitive adequate monotherapy or combination therapy yielded similar outcomes, suggesting that once susceptibility is documented, switching to a single active in vitro drug is safe and feasible [56].

Possible synergistic activity — A commonly cited reason for use of combination therapy is the potential for synergistic activity against P. aeruginosa with two agents, which in turn may result in better outcomes than single-drug therapy. However, there is no compelling evidence that two agents offer improved survival outcomes for treating P. aeruginosa infections.

Data on the use of combination versus single-drug therapy for P. aeruginosa infections are mixed, but overall, the best current evidence suggests that there is no additional benefit of a second active agent [57-67].

●In a meta-analysis of 64 randomized trials that compared beta-lactam monotherapy with combination therapy with a beta-lactam and an aminoglycoside for over 7500 immunocompetent patients with sepsis of all-cause (bacteremia did not need to be present), there was no survival advantage with combination therapy [65]. A lack of additional benefit was also observed in the subset of 426 patients with P. aeruginosa infection.

●In a meta-analysis of two randomized trials and 15 observational studies that included patients with documented gram-negative bacteremia, many of whom were immunocompromised, combination therapy was not associated with a mortality benefit overall [66]. However, among the subgroup of patients with P. aeruginosa bacteremia, there was a mortality benefit associated with combination therapy (odds ratio 0.5, 95% CI 0.30-0.79). Of note, several of the studies in the meta-analysis used aminoglycosides as a single agent in the monotherapy group, which is considered inadequate therapy [68].

●Observational studies have also not supported an association between empiric or definitive combination therapy (generally with a beta-lactam plus an aminoglycoside or fluoroquinolone) and improved mortality [67,69]. In one study of patients with P. aeruginosa bacteremia, definitive combination therapy was associated with reduced 30-day mortality when ciprofloxacin was the second agent, but not when tobramycin was the second agent [47].

For the most part, these results are in contrast to a frequently quoted retrospective study analyzing data collected more than 20 years ago, which concluded that combination therapy was associated with a reduced mortality rate compared with monotherapy for P. aeruginosa bacteremia (27 versus 47 percent) [57]. However, questions about the validity of these results have arisen because of a number of methodologic problems, including the lack of randomization, the choice of therapy (which may have been influenced by the severity of illness), and the observation that monotherapy in the majority of patients consisted of an aminoglycoside, which is associated with a mortality as high as 70 to 80 percent when used for the treatment of P. aeruginosa bacteremia [58].

Although not all studies showed clear evidence for a survival benefit with the use of combination therapy, it is possible that it offers other clinical benefits. As an example, studies comparing combination therapy with monotherapy in pulmonary infections in patients with cystic fibrosis have shown no differences in outcome, but the density of bacteria was lower in the sputum, and the time to the next pulmonary infection requiring hospitalization was more prolonged in patients receiving combination therapy [70,71].

Prevent emergent resistance — There is limited clinical evidence evaluating the ability of combination therapy to prevent the emergence of antimicrobial resistance in patients with P. aeruginosa bacteremia.

Emergence of resistance during therapy for P. aeruginosa infections can occur, resulting in increased rates of morbidity and mortality and higher costs. However, most studies of human P. aeruginosa infections have been underpowered to evaluate whether combination therapy can prevent emergent resistance. An observational study reported that resistant P. aeruginosa emerged in 10.2 percent of 271 cases during treatment with four individual antipseudomonal agents [22]. Ceftazidime was associated with the lowest risk and imipenem with the highest risk. The addition of an aminoglycoside did not alter this risk. Similarly, a four-year cohort study in an intensive care unit (ICU) was performed on patients with P. aeruginosa isolates from clinical specimens. During the follow-up, resistant phenotypes emerged in at least one antibiotic in isolates from 38.1 percent of the patients during the ICU stay. Specifically, 31.6 percent of P. aeruginosa isolates became resistant to piperacillin-tazobactam, 27.5 percent to ceftazidime, 22 percent to meropenem, 13.3 percent to ciprofloxacin, and 12.9 percent to amikacin [72].

This is in contrast to animal and in vitro studies that have suggested that combination therapy may alter this risk. As an example, combination therapy with two antipseudomonal antimicrobial agents limited the risk of emergence of resistant pseudomonal strains compared with monotherapy in an animal model of pseudomonal peritonitis [73,74]. Additionally, an in vitro study suggested that levofloxacin and imipenem might be an effective combination for preventing the emergence of resistance during treatment of P. aeruginosa infections [75].

Drawbacks — The main drawbacks to combination therapy include additional cost and toxicity. As an example, in a meta-analysis of 64 randomized trials that compared beta-lactam monotherapy with combination therapy with a beta-lactam and an aminoglycoside for sepsis, combination therapy was associated with a significant increase in nephrotoxicity [65].

Directed combination therapy — Definitive therapy can be tailored to the results of susceptibility tests once they are available. Definitive therapy with a single active agent (other than an aminoglycoside) is appropriate for most infections as there are no convincing clinical data that demonstrate a mortality benefit to combination therapy. (See 'Possible synergistic activity' above and 'Prevent emergent resistance' above and 'Drawbacks' above.)

Nevertheless, combination therapy is often used in situations in which the risk of emergent resistance or significant morbidity or mortality is high. For example, combination therapy is often used to treat P. aeruginosa endocarditis and bacteremia in high-risk hosts. (See "Pseudomonas aeruginosa bacteremia and endocarditis".)

Initiation of a second antipseudomonal agent for directed therapy may also be reasonable in infections that initially fail or are slow to respond to a single active agent, although there are few clinical, in vitro, or experimental data to support this practice.

In light of these uncertainties, we recommend consultation with an expert in treating such infections (when possible) if combination therapy is used for a multidrug-resistant organism. (See 'Management of multidrug-resistant organisms' below.)

MANAGEMENT OF MULTIDRUG-RESISTANT ORGANISMS — 

Infections due to difficult to treat (DTR) P. aeruginosa are a growing clinical problem and should be managed with the assistance of an expert in the treatment of such infections.

Susceptibility testing — For infections due to P. aeruginosa resistant to first-line agents (including piperacillin-tazobactam, cephalosporins, and carbapenems), we check susceptibility to novel agents, including ceftazidime-avibactam, ceftolozane-tazobactam, cefiderocol, and imipenem-cilastatin-relebactam, as well as polymyxins (colistin or polymyxin B).

For patients previously infected with a highly resistant isolate who present with a sepsis-like picture, we suggest repeating antibiotic susceptibility testing for novel agents because emergence of resistance during therapy has been reported. (See 'Emergence of resistance to novel agents' below.)

Antibiotic selection — For infections due to P. aeruginosa resistant to first-line agents (including piperacillin-tazobactam, cephalosporins, and carbapenems), we suggest monotherapy with either ceftolozane-tazobactam or ceftazidime-avibactam if the isolate is susceptible to one of these. If neither of these can be used, cefiderocol or imipenem-cilastatin-relebactam are alternatives. Polymyxins are typically reserved for infections when there are no other options.

Our recommendations are largely consistent with the Infectious Diseases Society of America (IDSA) guidance on management of antimicrobial-resistant gram-negative infections, which also favor monotherapy if in vitro susceptibility to one of these agents is confirmed and list ceftolozane-tazobactam, ceftazidime-avibactam, and imipenem-cilastatin-relebactam as preferred treatment options for DTR P. aeruginosa infections [76]. Cefiderocol is another preferred agent for urinary tract infections in the IDSA guidance, but for infections outside the urinary tract, it is an alternative agent reserved for patients who cannot use other options. European guidelines generally match IDSA guidance, except ceftolozane-tazobactam is the only preferred agent in the European guidelines, which consider the other regimens to have insufficient evidence [77].

Preferred agents — For infections resistant to first-line agents, we suggest one of the following antibiotics, if available, guided by susceptibility data. Doses are listed in the table (table 2).

For patients who were recently treated with one of these agents and present with a sepsis-like syndrome, we suggest choosing a different novel agent, at least until culture and susceptibility data are available. If ceftolozane-tazobactam was used previously, we suggest also avoiding ceftazidime-avibactam due to cross-resistance between these agents. (See 'Emergence of resistance to novel agents' below.)

●Ceftolozane-tazobactam – Several observational studies have described successful use of ceftolozane-tazobactam [78-81]. In a retrospective study of 200 patients with multidrug-resistant (MDR) P. aeruginosa infections (52 percent with ventilator-associated pneumonia), receipt of ceftolozane-tazobactam was associated with higher rates of clinical cure (adjusted odds ratio [aOR] 2.6) and lower rates of acute kidney injury (aOR 0.8) compared with aminoglycoside or polymyxin-containing regimens; in-hospital mortality was no different [81]. In another study of 205 patients, ceftolozane-tazobactam was associated with clinical success in 74 percent; receiving ceftolozane-tazobactam within four days of culture results was associated with better outcomes [78].

●Ceftazidime-avibactam – In a pooled analysis of results from randomized trials of patients with ceftazidime-resistant infections, ceftazidime-avibactam resulted in comparable clinical cures compared with the best-available therapy (usually a carbapenem); among 56 patients with P. aeruginosa treated with ceftazidime-avibactam in this study, the microbiologic response rate was 57 percent [82]. Although real-world data are limited, a review showed favorable outcomes with ceftazidime-avibactam treatment in some patients with MDR and extensively drug-resistant P. aeruginosa infections. Thus, ceftazidime-avibactam may have a potentially important role in the management of serious and complicated P. aeruginosa infections, including those caused by MDR and extensively drug-resistant strains [83].

There are recent data comparing the two agents directly. Ceftolozane-tazobactam is generally regarded as having more potent activity [84], but the choice between them should depend on the susceptibility pattern of the infecting isolate as there are regional differences in susceptibility to the two agents. In a study of 42 isolates of non-carbapenemase-producing, carbapenem-resistant P. aeruginosa isolates from South Korean hospitals, 95 percent were susceptible to ceftolozane-tazobactam compared with 71 percent to ceftazidime-avibactam [85]. In contrast, in a study of isolates from the Middle East, where many of the isolates produced class A, C, and D beta-lactamases (which are inhibited by avibactam but not tazobactam), susceptibility rates were slightly higher to ceftazidime-avibactam [86]. New studies have compared the emergence of resistance during therapy with both agents and showed conflicting results. For example, a study published in 2024 involving patients with MDR P. aeruginosa bacteremia or pneumonia revealed that those treated with ceftazidime-avibactam were more likely to develop resistance (defined as more than fourfold increased minimal inhibitory concentration [MIC]) than those treated with ceftolozane-tazobactam (40 versus 10 percent, respectively; p = 0.002). Ceftazidime-avibactam resistance was associated with new mutations in AmpC and efflux regulatory pathways [87]. In contrast, another multicenter prospective study of patients with DTR P. aeruginosa infections from 2018 to 2023 found that the emergence of resistance was higher in the ceftolozane-tazobactam group compared with ceftazidime-avibactam (38 versus 25 percent) but did not achieve statistical significance. Thus, the emergence of resistance may be more concerning with the former [88].

Alternative agents — If one of the preferred agents is not an option, cefiderocol and imipenem-cilastatin-relebactam are novel agents that are acceptable alternatives. There is even less clinical experience with these agents. Doses are listed in the table (table 2).

●Cefiderocol – Most carbapenem-resistant isolates remain susceptible to cefiderocol, a novel siderophore cephalosporin [11,89,90]. In a trial of patients with complicated urinary tract infection, cefiderocol resulted in similar clinical and microbiologic cure rates for the few infections caused by P. aeruginosa compared with imipenem; however, the study did not include carbapenem-resistant organisms [41]. The IDSA does not list cefiderocol as a preferred agent for infections outside of the urinary tract because it was associated with higher all-cause mortality compared with best available therapy (typically a polymyxin-containing regimen) for non-urinary tract infections caused by carbapenem-resistant bacteria; only 28 P. aeruginosa infections were included in the trial [43,91]. However, with the emergence of ceftolozane-tazobactam and ceftazidime-avibactam resistance, this agent can be considered for use.

●Imipenem-cilastatin-relebactam – The addition of relebactam to imipenem-cilastatin restores activity against some P. aeruginosa isolates that are resistant to imipenem. In a trial of patients with serious infection caused by imipenem-resistant organisms (70 percent of which were P. aeruginosa), there was a trend toward lower 28-day mortality rates with imipenem-cilastatin-relebactam compared with imipenem-cilastatin plus colistin [92].

Other agents — If the above antibiotics are not available, effective options are limited. Doses are listed in the table (table 2).

●Polymyxins – We only use polymyxins (colistin or polymyxin B) when no other options exist because of their potential for toxicity. If a polymyxin is used for therapy of infections other than cystitis, we typically use it with a second susceptible agent. For cystitis, colistin can be used as monotherapy, but polymyxin B should be avoided due to suboptimal urinary levels. Care must be taken when dosing colistin and polymyxin B, as formulations differ between United States and European products, and each has distinct dosing recommendations. (See "Polymyxins: An overview", section on 'Dosing and administration'.)

Polymyxins have been used with some success in observational studies and case series [93-99]. As an example, in a study of 22 patients with metallo-beta-lactamase-producing Pseudomonas infections who were treated with intravenous colistin, 67 percent had a partial or complete response [97]. However, two-thirds developed at least mild nephrotoxicity.

●Aminoglycosides – We typically do not use these agents as monotherapy for infections other than cystitis. For uncomplicated blood stream infections in which source control has been achieved, we sometimes use aminoglycosides when no other options are possible due to resistance or lack of availability. Careful monitoring of dosing, drug levels, and side effects should be assured in these cases. More information regarding the role of aminoglycosides is discussed above. (See 'Role of aminoglycosides' above.)

●Delafloxacin – Delafloxacin is a novel fluoroquinolone with a broad spectrum of activity that includes MDR pathogens, such as methicillin-resistant Staphylococcus aureus and P. aeruginosa. It has been approved as a new option for the treatment of acute bacterial skin and skin-structure infections as well as potentially for complicated urinary tract infections and severe community-acquired bacterial pneumonia. Delafloxacin offers both oral and intravenous options, which allows ease of administration in both inpatient and outpatient settings. Delafloxacin is generally well tolerated and may potentially have an improved side effect profile to currently available fluoroquinolones with respect to QTc prolongation, tendinitis, and tendon rupture. Delafloxacin should be used judiciously to preserve its effectiveness and prevent collateral damage [100].

●Antibiotics that should not be used – Many novel antibiotics are not recommended for Pseudomonas infections due to high probability of resistance. These include fosfomycin, tigecycline, and eravacycline. Meropenem-vaborbactam does not provide additional activity beyond meropenem alone and is not active against meropenem-resistant isolates.

Adjunctive inhaled antibiotics for pneumonia — Although we do not routinely use inhaled antibiotics for the treatment of P. aeruginosa pneumonia susceptible for first-line agents, they may be effective as an adjunct to intravenous therapy in cases of infection due to MDR strains. This is discussed in detail elsewhere. (See "Pseudomonas aeruginosa pneumonia", section on 'Inhaled antibiotics for selected patients'.)

Emergence of resistance to novel agents — Reports of emergence of resistance during or following therapy with novel agents are accumulating. The frequency of emergent resistance with these agents is difficult to ascertain due to limited data, most of which are extrapolated from treatment trials.

Small observational studies have reported resistance to ceftolozane-tazobactam in 50 to 100 percent of patients who had positive cultures after being treated with the agent for at least 72 hours [101,102]. Of additional concern, cross-resistance between ceftolozane-tazobactam and ceftazidime-avibactam occurs because of similar mechanisms of resistance; in one study of 28 patients treated with ceftolozane-tazobactam, ceftazidime-avibactam resistance occurred in 86 percent of isolates previously susceptible to the agent [102]. (See 'Preferred agents' above.)

Reports of resistance after exposure to imipenem-relebactam or cefiderocol for pseudomonal infections are less common, but these agents are newer than ceftolozane-tazobactam and ceftazidime-avibactam. Concern is warranted for all novel agents until further data emerge.

Pandrug-resistant isolates — Isolates that are resistant to first-line agents, novel beta-lactam-beta-lactamase inhibitors, and cefiderocol present particular therapeutic challenges.

●Combination therapy – For extensively resistant organisms, we use combination therapy with the agent that has the lowest MIC combined with either a sensitive aminoglycoside or polymyxin B, if possible [25].

For isolates resistant to all available antibiotics, some clinicians have used combinations of drugs that separately have little or no activity against the isolate [103,104]. There are minimal clinical data to support such combination therapy. If combination therapy is used for treatment of organisms with extreme or unusual multidrug resistance patterns, it should be done in consultation with an expert in treating such infections whenever possible.

In a single clinical series of 64 patients with nosocomial pulmonary infections due to a highly resistant P. aeruginosa susceptible only to colistin, treatment with the combination of cefepime and amikacin was associated with survival in 44 (69 percent) [105]. These agents were the least inactive antibiotics by MIC determination and had demonstrated synergy in vitro. Dual beta-lactam therapy [106] and combinations with fosfomycin [107-109] have also been described with some success.

Aztreonam-avibactam emerged as a novel option for the treatment of bloodstream infections or lower respiratory tract infections caused by extensively drug-resistant (XDR) or pandrug-resistant (PDR) P. aeruginosa. A study published in 2023 found that XDR/PDR P. aeruginosa may carry some metallo-beta-lactamases and these isolates were inhibited by aztreonam-avibactam. The study also demonstrated that aztreonam-avibactam was more potent than ceftazidime-avibactam against these metallo-beta-lactamases-producing isolates, making it an optimal treatment option [110].

Combination regimens of other common antibiotics have been found to have enhanced activity against MDR P. aeruginosa in vitro, but overall clinical data are extremely limited [104,111-119]. The mechanisms for the enhanced activity are unknown for most combinations.

●Investigational approaches

•Phage therapy – Phage therapy has been evaluated for P. aeruginosa endovascular infections in case reports and animal studies [120,121]. The frequency and significance of acquired resistance to phage treatment are uncertain issues. (See "Pseudomonas aeruginosa bacteremia and endocarditis".)

•Antibiotics in development – Murepavadin is a promising antibacterial agent with potent activity against carbapenem-resistant and colistin-resistant P. aeruginosa [122]. It is being evaluated for P. aeruginosa ventilator-associated pneumonia [123].

TREATMENT IN CHILDREN — 

The general principles of antipseudomonal therapy in children are similar to those in adults, with the main exception that antimicrobial dosing is weight based. Additionally, the use of fluoroquinolones in children warrants specific consideration.

There have been reports of tendon inflammation and/or rupture with fluoroquinolone antibiotics in all ages; risk may be increased with concurrent steroids and in solid organ transplant recipients. In some pediatric studies, an increased risk of reversible adverse events involving joints or surrounding tissues, often due to reports of arthralgia, has been observed [124,125]. However, no compelling published evidence supports the occurrence of sustained injury to developing bones or joints in children treated with available fluoroquinolone agents [126,127]. Use of a fluoroquinolone is a consideration for gram-negative infections caused by isolates susceptible to these agents. Specifically, this includes treatment for infections when caused by P. aeruginosa, including chronic suppurative otitis media, malignant otitis externa, bony infections associated with nail wound puncture, and complicated urinary tract infection. The risks and benefits should be considered if a fluoroquinolone is prescribed in a child younger than 18 years of age. (See "Chronic suppurative otitis media (CSOM): Treatment, complications, and prevention", section on 'Treatment failure' and "Necrotizing (malignant) external otitis" and "Pseudomonas aeruginosa skin and soft tissue infections", section on 'Infection following nail puncture' and "Urinary tract infections in infants older than one month and children younger than two years: Acute management, imaging, and prognosis" and "Fluoroquinolones", section on 'Children'.)

SUMMARY AND RECOMMENDATIONS

●Antibiotic resistance – Pseudomonas aeruginosa is intrinsically resistant to a number of antibiotics and can acquire resistance during therapy. These features limit the choices of antibiotics for P. aeruginosa. (See 'Antimicrobial resistance' above.)

●Antibiotic dosing strategies – Certain agents require higher dosing when used for presumptive or known P. aeruginosa infections, and prolonged infusions optimizes pharmacokinetics for some agents (table 2). (See 'Dosing strategies' above and "Prolonged infusions of beta-lactam antibiotics".)

●First-line antibiotics – Multiple antibiotics from different classes are available to treat pseudomonal infections. (See 'Antibiotics with antipseudomonal activity' above.)

•Intravenous options – Intravenous options include piperacillin-tazobactam, ceftazidime, cefepime, aztreonam, fluoroquinolones, and carbapenems. Aminoglycosides are generally not used as single agents because of inadequate clinical efficacy at most sites. Doses are discussed in the table (table 2). (See 'Antibiotics with antipseudomonal activity' above.)

•Oral options – Fluoroquinolones are the only class of antibiotics with an oral formulation that is reliably active against P. aeruginosa. Doses are discussed in the table (table 2). (See 'Oral antibiotics' above.)

●Combination therapy – There is no convincing clinical evidence that using two active agents instead of one leads to improved outcomes. However, using two agents for empiric therapy may increase the likelihood that an active agent will be used for a potentially resistant organism, and this, in turn, is associated with better outcomes for serious infections. (See 'Role of combination antimicrobial therapy' above.)

For patients who have sepsis or septic shock, have neutropenia and bacteremia, have severe burns, or are in a setting where the incidence of resistance to the chosen antibiotic class is high (eg, >10 to 15 percent), we suggest empiric therapy with a combination of two antipseudomonal agents (Grade 2C). The two agents should be from different antibiotic classes (eg, a beta-lactam with an aminoglycoside or a quinolone). (See 'Indications' above and 'Selection of agents' above.)

For patients without any of these additional risk factors for mortality or resistant organisms, we suggest empiric treatment with a single antipseudomonal agent (Grade 2B).

●Directed therapy – Once antimicrobial susceptibility data are available, we typically use a single active antipseudomonal agent. The rare exceptions when we might continue combination include neutropenia and bacteremia, endocarditis, failure to improve on therapy, and multidrug resistance. (See 'Directed combination therapy' above.)

●Management of multidrug-resistant (MDR) organisms – For P. aeruginosa isolates resistant to first-line agents (including carbapenems), we suggest ceftolozane-tazobactam or ceftazidime-avibactam, as long as the isolate is susceptible (Grade 2C). MDR P. aeruginosa should be managed with the assistance of an expert in the treatment of such infections. (See 'Management of multidrug-resistant organisms' above.)

●Treatment of children – The general principles of antipseudomonal therapy in children are similar to those in adults, with the main exception that antimicrobial dosing is weight based. (See 'Treatment in children' above.)