Initiate antibacterial treatment early in patients with carbapenem-resistant or extensively drug-resistant Acinetobacter baumannii infection

Abstract

Survival is consistently improved with early initiation of appropriate antibacterial therapy in patients with carbapenem-resistant or extensively drug-resistant Acinetobacter baumannii infections. However, because of a general lack of data, there is no clear consensus for the optimum empirical antibacterial therapy in these patients. Regimens that include colistin, sulbactam or tigecycline have been the most widely evaluated.

Major cause of hospital infections

In recent decades, Gram-negative coccobacilli Acinetobacter spp. have become increasingly major pathogens in hospital-acquired and healthcare-associated infections, occurring most frequently with ventilator-associated pneumonia, bloodstream infection, wound infection and urinary tract infection [1]. Among the >30 known genomic species, A. baumannii is the most clinically relevant on the basis of virulence and multidrug resistance (MDR), causing invasive infections in patients with co-morbid conditions and/or pre-existing illness. Of note, A. baumannii is not distinguishable from three other genomic species (two of which are also clinically significant pathogens) by standard biochemical, non-genetic methods, and references to A. baumannii in the literature may mean either just A. baumannii or the A. baumannii complex (i.e. all four genomic species) [1].

Characterized by resistance to multiple agents …

A. baumannii is well known for its capacity to develop resistance to multiple classes of antibacterials [1]. MDR is technically defined as the non-susceptibility of an isolate to at least one agent in three or more antibacterial classes [2] (e.g. an isolate is defined as MDR if it is resistant to ceftriaxone, ciprofloxacin and trimethoprim–sulfamethoxazole), which means that many treatment options are still available [1]. However, the use of the term extensive drug resistance (XDR; defined as non-susceptibility of an isolate to one or more agents in all but two or fewer antibacterial categories [2]), is more clinically relevant when discussing the treatment of drug-resistant A. baumannii infections [1]. XDR isolates of A. baumannii (e.g. isolates susceptible to colistin and tigecycline, but resistant to all other categories tested) are increasingly encountered in clinical practice, with certain XDR strains being epidemic in hospitals [3]. Most clinical isolates of A. baumannii are resistant to cephalosporins; other antibacterial classes for which A. baumannii is resistant include carbapenems, sulbactam, rifampicin (rifampin), aminoglycosides, fluoroquinolones, colistin (polymixin E), penicillins and tetracyclines [1]. A. baumannii can survive for weeks on dry surfaces, which further facilitate its dissemination [1].

… in particular carbapenem

Carbapenem resistance is of particular importance, as second-line options for the treatment of A. baumannii have less defined efficacy and are often associated with various toxicities [1]. The incidence of carbapenem resistance is increasing [4]. Carbapenem-resistant A. baumannii is an independent risk factor for mortality, and is associated with higher mortality rates than other A. baumannii complex species [5]. In addition, carbapenem-resistant A. baumannii strains are often also XDR and vice versa (i.e. XDR A. baumannii strains are resistant to carbapenem) [1]. This article provides a summary of the treatment of carbapenem-resistant and XDR A. baumannii infections as reviewed by Viehman et al. [1].

Multiple acquisition and mortality risk factors

Risk factors for acquiring XDR [6] or carbapenem-resistant [7] A. baumannii include recent exposure to antibacterial (in particular carbapenem), the presence of central venous catheter or urinary catheter, more severe illness, relatively longer duration of hospital stay, treatment in an intensive care unit (ICU) or relatively large hospital, and recent surgery.

Mortality rates associated with invasive A. baumannii infections vary, but are generally high, especially for carbapenem-resistant strains (16–76 vs. 5–53 % for carbapenem-susceptible infections) [8]. The high mortality rates are generally attributed to more severe illness and early, but inappropriate, antibacterial therapy [8]. Independent risk factors for mortality in patients with bloodstream infections with carbapenem-resistant A. baumannii include illness severity, underlying malignancy, history of transplant, higher age, septic shock, concurrent pneumonia, inappropriate antibacterial therapy, prolonged ICU stay and renal failure [8, 9].

Treat appropriately, and early …

In order to reduce the mortality resulting from severe sepsis and septic shock in patients with resistant-Gram negative bacteria, early and appropriate antibacterial therapy is crucial [1]. If the causative A. baumannii strain is resistant to carbapenem, the risk of inappropriate therapy [10] or a delay in appropriate therapy [11] is increased, resulting in a significant increase in mortality rates. Currently, it is unclear whether appropriate antibacterial therapy should consist of monotherapy or combination therapy [1].

… but no clear consensus for drug selection

Some clinicians suggest a polymyxin-containing regimen (colistin or the less widely available polymyxin B) to treat infections with colistin-susceptible A. baumannii, and also favour such a regimen for infections with colistin-resistant strains [12]. However, other clinicians prefer regimens based on other agents (e.g. sulbactam, tigecycline), because of the toxicity profiles of polymyxins [1]. Table 1 outlines the most studied treatment options for patients with carbapenem-resistant or XDR A. baumannii. Of note, colistin is administered intravenously in the form of its less active prodrug colistimethate (colistin methanosulfonate); one million international units of colistimethate is equivalent to ≈30 mg of colistin base activity, which corresponds to ≈80 mg of chemical colistimethate.

Table 1 Treatment options for patients with carbapenem-resistant or extensively drug-resistant Acinetobacter baumannii infections, as reviewed by Viehman et al. [1]

Is combination therapy better than monotherapy?

Treatment with a combination of concomitant antibacterials has advantages over antibacterial monotherapy. Overall efficacy may be improved, as the use of more than one antibacterial may provide multiple mechanisms of antibacterial activity and have a faster onset of action than monotherapy (e.g. colistin takes time to reach therapeutic concentrations) [1]. The increasing incidence of carbapenem-resistant A. baumannii may mean that colistin-based combinations are used as empirical therapy in the future, instead of as salvage therapy as currently used [1].

For colistin [24, 25] and tigecycline [22, 26], there are concerns regarding the unpredictable and suboptimal pharmacokinetics and the emergence of drug resistance during their use; for sulbactam, there are concerns regarding unclear optimal dosing and clinical outcomes that are not always predicable based on its in vitro efficacy [27]. The available trial data (Table 2) suggest that patients with carbapenem-resistant A. baumannii may respond better to colistin-based combination therapy than to monotherapy, but the components of the optimal combination are as yet unclear [1]. In two retrospective studies (Table 2), the use of colistin and a carbapenem was an independent predictor of survival [28], and microbiological eradiation was significantly higher with combination therapy than with monotherapy (79.9 vs. 55.6 %; p = 0.001) [29]. However, in two other retrospective studies, clinical cure rates did not favour various colistin combinations over colistin monotherapy [55.2 vs. 67.9 % (no significant difference) [30] and 80.3 vs. 87 % (p-value not reported) [31]].

Table 2 Summary of studies comparing mortality rates with a combination of concomitant antibacterials vs. antibacterial monotherapy in patients with carbapenem-resistant or extensively drug-resistant Acinetobacter baumannii infections

Even less data for other agents

Fosfomycin has shown in vitro synergy with colistin or sulbactam [1], and a clinical trial has shown higher microbiological eradication rates with fosfomycin 4 g every 12 h plus colistin than with colistin alone (100 vs. 81.2 %; p < 0.01), with no significant difference in clinical outcomes or mortality rates [35]; the fosfomycin dosage used was lower than that used for other infections, and the study was underpowered to detect a difference in mortality rates. Given that fosfomycin is bactericidal and relatively well tolerated (with the possibility of using higher dosages), the combination of fosfomycin and colistin in treating A. baumannii infections warrants further investigation [1].

The use of minocycline has been superseded by the use of tigecycline (Table 2); however, micocycline may be of use in combination therapy or as step-down therapy, as it is available in intravenous and oral formulations [1].

Glycopeptides (e.g. vancomycin) do not generally penetrate the outer membrane of Gram-negative bacteria, although their addition to colistin showed strong in vitro synergy against such bacteria, including A. baumannii. However, conflicting results were reported in clinical studies [30, 34], and more studies are needed to evaluate this combination [1].

Despite good in vitro synergy and in vivo activity [1], clinical trials did not show a benefit with using rifampicin plus colistin compared with using colistin alone [32, 33]; given its association with hepatotoxicity and likelihood of cytochrome P450 3A4-related drug interactions, the use of rifampicin is not recommended to treat A. baumannii infections [1].