Achromobacter Infections and Treatment Options

Achromobacter is a genus of nonfermenting Gram-negative bacteria under order Burkholderiales. Although primarily isolated from respiratory tract of people with cystic fibrosis, Achromobacter spp. can cause a broad range of infections in hosts with other underlying conditions. Their rare occurrence and ever-changing taxonomy hinder defining their clinical features, risk factors for acquisition and adverse outcomes, and optimal treatment. Achromobacter spp.

ABSTRACT

Achromobacter is a genus of nonfermenting Gram-negative bacteria under order Burkholderiales. Although primarily isolated from respiratory tract of people with cystic fibrosis, Achromobacter spp. can cause a broad range of infections in hosts with other underlying conditions. Their rare occurrence and ever-changing taxonomy hinder defining their clinical features, risk factors for acquisition and adverse outcomes, and optimal treatment. Achromobacter spp. are intrinsically resistant to several antibiotics (e.g., most cephalosporins, aztreonam, and aminoglycosides), and are increasingly acquiring resistance to carbapenems. Carbapenem resistance is mainly caused by multidrug efflux pumps and metallo-β-lactamases, which are not expected to be overcome by new β-lactamase inhibitors. Among the other new antibiotics, cefiderocol, and eravacycline were used as salvage therapy for a limited number of patients with Achromobacter infections. In this article, we aim to give an overview of the antimicrobial resistance in Achromobacter species, highlighting the possible place of new antibiotics in their treatment.

INTRODUCTION

Genus Achromobacter was first established in 1923 by the Committee of the Society of American Bacteriologists (today the American Society for Microbiology) as “non-pigment-forming, motile or nonmotile Gram-negative bacteria occurring in water and soil” (1). Close resemblance of genus Achromobacter to genus Alcaligenes, both of which are members of the Alcaligenaceae family of the order Burkholderiales, prompted reassignment of several Achromobacter species to genus Alcaligenes and vice versa. Genus Achromobacter currently comprises 19 officially designated species, most of which were characterized within the last decade (2). Fifteen species to date have been isolated from clinical specimens, including Achromobacter xylosoxidans, Achromobacter denitrificans, Achromobacter ruhlandii, Achromobacter piechaudii (3), Achromobacter animicus, Achromobacter mucicolens, Achromobacter pulmonis (4), Achromobacter insolitus, Achromobacter spanius (5), Achromobacter deleyi (6), Achromobacter aegreficans, Achromobacter insuavis, Achromobacter anxifer, Achromobacter dolens (7), and Achromobacter marplatensis (8). Worldwide, A. xylosoxidans is the most common species recovered from clinical samples, including those derived from persons with cystic fibrosis (CF). Distribution of other species show geographical diversity. A. ruhlandii is the second most common species in the Americas (9–11), whereas A. dolens and A. insuavis are more prevalent in Europe (12–14). Clinical significance of species variation is not well characterized.

Identification.

Genus Achromobacter is an obligately aerobic, nonfermentative; oxidase- and catalase-positive; indole-, urease-, and DNase-negative bacterium (15). Achromobacter spp. are frequently misidentified as other common (i.e., Pseudomonas aeruginosa, Stenotrophomonas maltophilia, Burkholderia cepacia complex, Acinetobacter spp.), and rare (i.e., Pandoraea spp. and Ralstonia spp.) nonfermenting Gram-negative bacilli with conventional methods due to biochemical similarities (16–18). Furthermore, most Achromobacter species were referred as A. xylosoxidans with conventional methods. More accurate species determination became possible with the utilization of the genotypic methods such as nrdA gene sequencing and multilocus sequence typing (8, 9). However, for many routine clinical microbiology laboratories, sequence-based identification using these techniques is not feasible. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) was successful at distinguishing Achromobacter from other nonfermenters at the genus level (17, 19, 20). Identification to the species level has been hampered by the limited number of species included in the MALDI-TOF databases (e.g., two and six species for Vitek MS V3.0 and MALDI Biotyper IVD-CE, respectively) (21). MALDI-TOF was successful at identifying most species (i.e., ≥90%) accurately when its database was expanded using 18 and 9 different Achromobacter species in two different studies (21). Correct identification rates with the default MALDI Biotyper database in these studies were 51 and 66%, respectively, misidentification mainly stemming from referring the species not involved in the database as A. xylosoxidans (22). These results are promising, and their incorporation into the commercial databases will facilitate a more accurate identification at the species level. Until then, confirmation of the MALDI-TOF results with genotypic methods is warranted for correct species identification.

Host predisposition and clinical spectrum.

Achromobacter spp. are predominantly recovered from persons with CF as chronic respiratory pathogens and are common causes of CF after lung transplant infections with poor outcomes (23–25).

Outside the context of CF, data on the clinical spectrum of Achromobacter infections come from case reports and case series. Pneumonia and bacteremia are the two most common clinical presentations of Achromobacter infections in non-CF hosts (26). Infections of the skin and soft tissue, urinary tract, intraabdominal organs, central nervous system (CNS), eye, and ear are less frequently reported, endocarditis and bone infections being very rare (27–34). Most Achromobacter infections are either hospital acquired or health care associated and often develop in relation to foreign devices (35). Achromobacter infections do not solely occur in immunocompromised hosts, as previously thought. Patients with devices (e.g., catheters and endotracheal tubes), underlying conditions (e.g., diabetes mellitus, chronic renal failure, chronic heart diseases), and current or previous hospitalization or health care exposure are at risk (26, 36).

ANTIBIOTIC RESISTANCE MECHANISMS

The two main intrinsic resistance mechanisms of Achromobacter species comprise multidrug efflux pumps and chromosomal OXA-114-like β-lactamases (Table 1). Extended-spectrum β-lactamases (ESBLs), AmpC type β-lactamases, and metallo-β-lactamases (MBLs) have been detected in Achromobacter isolates and appear to contribute to resistance to β-lactams, including carbapenems.

TABLE 1.

Antibiotic resistance mechanisms

Multidrug efflux pump(s)	Spectrum
AxyABM	Cephalosporins (except cefuroxime and cefepime), and aztreonam
AxyXY-OprZ	Aminoglycosides, tetracyclines including tigecycline, fluoroquinolones, erythromycin, cefepime, carbapenems
β-Lactamases
OXA-114-like (chromosomal)	Piperacillin, ticarcillin, benzylpenicillin, cephalothin
ESBL (CTX-M, VEB-1) and AmpC (CMY-2, AmpC)	All β-lactams except carbapenems
Plasmidic (IMP and VIM) and chromosomal (TMB-1) carbapenemases	All β-lactams except aztreonam
Other
aac(6′)Ib-cr, qnrA, oqxA, oqxB	Fluoroquinolones, aminoglycosides
gyrA, parC	Fluoroquinolones

Multidrug efflux pumps.

Achromobacter species harbor two well characterized multidrug efflux pumps and several putative efflux pump genes (37). AxyABM efflux pump is found in all publicly available Achromobacter genomes and share common properties with the MexAB-OprM efflux pump of P. aeruginosa (38). AxyABM plays a major role in the extrusion of cephalosporins other than cefepime and cefuroxime and of aztreonam, but it does not appear to be the sole mechanism of resistance for these agents since aztreonam and cephalosporin susceptibilities were not restored after AxyABM inhibition in vitro (39). AxyABM inhibition resulted in decreases in cefotetan, cefoxitin, cefotaxime, ceftriaxone, and aztreonam MICs from >256 μg/ml to 32, 128, 12, 12, and 16 μg/ml, respectively (39). The ceftazidime MIC dropped from 4 to 1.5 μg/ml, whereas the cefuroxime, cefepime, amikacin, colistin, tigecycline, and carbapenem (i.e., meropenem and imipenem) MICs remained unchanged. Changes in fluoroquinolone MICs were not significant (drop from 0.75 μg/ml to 0.5 and 0.38 μg/ml for ciprofloxacin and levofloxacin, respectively).

The second efflux pump AxyXY-OprZ has a broader spectrum and is involved in the extrusion of aminoglycosides, cefepime, carbapenems, fluoroquinolones, tetracyclines, and erythromycin to various degrees (40). AxyXY-OprZ is the main resistance determinant that accounts for high-level intrinsic aminoglycoside resistance in Achromobacter spp. For other antibiotics (e.g., cefepime, ceftazidime, carbapenems, tetracyclines, and fluoroquinolones) AxyXY-OprZ appears to be a contributor to the resistance rather than being the main resistance mechanism. Inhibition of AxyXY-OprZ restored susceptibilities to all aminoglycosides and tigecycline (40). Imipenem and meropenem MICs dropped from 4 to 1 μg/ml and 12 to 2 μg/ml, respectively, for a carbapenem-resistant strain, whereas the drop was 1-fold for a carbapenem-susceptible strain. Fluoroquinolone MICs were reduced 2- to 4-fold for fluoroquinolone-resistant strains, but susceptibilities were not restored. AxyXY-OprZ inhibition resulted in 1-fold and 0- to 4-fold decreases in the ceftazidime and cefepime MICs, respectively, without restoring susceptibilities to these agents. A previous study found AxyXY-OprZ to be present only in certain strains with phenotypic aminoglycoside resistance (i.e., A. xylosoxidans, A. ruhlandii, A. dolens, A. insuavis, A. denitrificans, A. insolitus, and A. aegrifaciens) and absent in aminoglycoside-susceptible strains (i.e., A. mucilocens, A. animicus, A. piechaudii, and A. spanius) (41), whereas a pangenome analysis from 2016 demonstrated AxyXY-OprZ to be present in all publicly available genomes (38). The reason for the inconsistency between these studies is unclear.

Whether the production of efflux pumps is induced under antibiotic pressure is a question that remains to be answered. In vitro tobramycin exposure lead to mutations in the AxyXY-OprZ regulator resulting in MIC increases in its substrate antibiotics in laboratory strains (42). The same mutation was observed in a clinical strain from a CF patient after tobramycin exposure, but this patient had been exposed to several other antibiotics during her chronic colonization, and it is uncertain which of these, if any, induced the mutation (42). In contrast, successive strains with increased antibiotic MICs from four other patients did not have the same mutation, suggesting the involvement of other genetic mechanisms in AxyXY-OprZ regulation (42). Finally, no association could be demonstrated between antibiotic exposure and resistance development for meropenem, ticarcillin, clavulanic acid, and colistin in Achromobacter isolates from a single CF patient (43).

β-Lactamases.

Achromobacter species produce a constitutive chromosomal β-lactamase, namely, OXA-114, with activity against penicillin G, early cephalosporins, piperacillin, and ticarcillin. OXA-114-like enzymes are efficient piperacillin hydrolyzers in vitro. Ticarcillin is hydrolyzed to a lesser degree, whereas extended-spectrum cephalosporins, such as ceftazidime, cefotaxime, or cefepime, are not substrates of OXA-114-like enzymes (44). Hydrolytic activity of OXA-114 against imipenem is very poor.

The contribution of OXA-114 to phenotypic piperacillin resistance of Achromobacter isolates is unclear since piperacillin susceptibility is common among OXA-114-positive Achromobacter isolates (45). Piperacillin hydrolysis by OXA-114 does not appear to be inhibited by tazobactam. In their study in which OXA-114 was discovered, Doi et al. demonstrated all five OXA-114 positive Achromobacter strains to be susceptible to piperacillin (MICs 1 to 4 μg/ml), and the addition of tazobactam decreased the piperacillin MICs 1-fold, except for the strain with piperacillin MIC of 4 μg/ml, for which no MIC change was observed (44). Likewise, piperacillin and piperacillin-tazobactam MICs were similar (MIC_50/90 = 32/>128 μg/ml; range, 4 to >128 μg/ml for both antibiotics) in another collection of 94 CF isolates (16).

The presence of acquired β-lactamases such as ESBLs (i.e., bla_CTX-M and bla_VEB-1), AmpC type β-lactamases (i.e., bla_AmpC and bla_CMY−2), and carbapenemases (i.e., bla_IMP and bla_VIM) in Achromobacter isolates were reported from several countries, including Greece, Italy, France, Japan, and Korea (46–55). In addition to IMP and VIM, which are among the most common MBLs worldwide, another rarer MBL, namely, Tripoli MBL (TMB), was discovered in the chromosome of an A. xylosoxidans strain from Tripoli, Libya (56). Amino acid structure of TMB was similar to that of DIM and GIM MBLs of P. aeruginosa, with a lower hydrolytic activity against cephalosporins and carbapenems. The meropenem and imipenem MICs were 4 and 2 μg/ml, respectively, for the isolate harboring TMB (56).

The spread of acquired β-lactamases, particularly MBLs, is concerning since none of the currently available β-lactamase inhibitors can overcome the β-lactam resistance caused by MBLs. Aztreonam, the only β-lactam antibiotic resistant to MBL hydrolysis, is a substrate of the AxyABM efflux pump. The proportion of MBLs among carbapenem-resistant Achromobacter isolates is not well described despite increasing carbapenem resistance rates. Achromobacter strains with elevated carbapenem MICs should be screened for the presence of these MBLs, while keeping in mind that resistance to carbapenems may occur via other resistance mechanisms (e.g., AxyXY-OprZ, unidentified β-lactamases). Some in vitro studies demonstrated that mutations in transcription regulators of unnamed β-lactamases, or increased expression of a putative β-lactamase gene (i.e., bla_AXC), may contribute to carbapenem resistance in Achromobacter spp. (43, 57, 58). Comparative genomic analysis of carbapenem-susceptible and -resistant strains is essential to elucidate the genetic mechanisms behind carbapenem resistance of genus Achromobacter.

Other.

A. xylosoxidans contains at least 50 intrinsic resistance genes, including class A to D β-lactamases (five new β-lactamases), efflux pump systems (seven new RND-type efflux pumps), aminoglycoside acetyltransferase and phosphotransferases, dihydrofolate reductases, and others, more than half of which show significant similarities with that of P. aeruginosa (37). Furthermore, clinical isolates carry additional β-lactam, aminoglycoside, and sulfonamide resistance genes compared to the environmental isolates (37, 38).

The presence of plasmid-mediated acetyltransferases (i.e., aac-(6′)-Ib-cr) and other plasmidic (i.e., oqxA, qnrA, and oqxB) and chromosomal (i.e., gyrA and parC) fluoroquinolone resistance genes were demonstrated in clinical Achromobacter isolates from Serbia (59) and environmental Achromobacter isolates from Brazil (60). Their contribution to phenotypic fluoroquinolone resistance in Achromobacter strains awaits further evaluation.

ANTIBIOTIC SUSCEPTIBILITY

Optimal genus-specific susceptibility testing methods and widely accepted susceptibility breakpoints are not defined for Achromobacter spp. CLSI provides MIC breakpoints for Achromobacter spp. under the “other non-Enterobacterales” category, but most of the previous studies used various antimicrobial susceptibility testing methods and susceptibility breakpoints (Table 2) (61, 62). Despite these heterogeneities, some common phenotypic susceptibility patterns emerge among Achromobacter isolates. Wild-type strains appear to demonstrate phenotypic resistance to narrow-spectrum penicillins, first- and second-generation cephalosporins, ceftriaxone, cefotaxime, aztreonam, tetracycline, and aminoglycosides, whereas they may remain susceptible to ceftazidime, cefepime, piperacillin, carbapenems, sulfonamides, fluoroquinolones, doxycycline, tigecycline, and colistin. Contemporary isolates, particularly those from chronically colonized CF patients, are often resistant to most of these antibiotics, particularly to fluoroquinolones (63). Trimethoprim-sulfamethoxazole, ceftazidime, piperacillin, and carbapenems are the most active agents against Achromobacter isolates. However, acquired resistance to these are increasingly being reported (64). The proportion of ceftazidime-susceptible isolates was 71% in a U.S. collection (susceptibility breakpoint, ≤8 μg/ml; CLSI other non-Enterobacterales; MIC_50/90, 8/32 μg/ml) (65). In another collection from Europe, trimethoprim-sulfamethoxazole MIC_50/90 was 0.5/8 μg/ml (susceptibility breakpoint ≤ 2 μg/ml, CLSI other non-Enterobacterales), and it was one of the two most active agents against 59 Achromobacter isolates, the other being imipenem (MIC_50/90, 2/8 μg/ml) (66). Several studies showed imipenem to be more active than meropenem against Achromobacter isolates (63, 65, 66). This was also observed in P. aeruginosa isolates and was demonstrated to be caused by the overexpression of the mexAB efflux pumps (67). A similar mechanism may be responsible for the discordant carbapenem susceptibilities in Achromobacter spp. Conversely, some other studies report meropenem (MIC_50/90, 0.125/1 μg/ml) to be more active than imipenem (MIC_50/90, 1/4 μg/ml) (68) (Table 2), and the genetic determinants of this phenotype is not characterized.

TABLE 2.

Antibiotic susceptibility of Achromobacter isolates^a

Reference	Region, yr, and source (n)	AST method	Antibiotic
			TZP		CAZ		IPM		MEM		SXT		CIP
			S% (breakpoint)	MIC50/90 (range)	S% (breakpoint)	MIC50/90 (range)	S% (breakpoint)	MIC50/90 (range)	S% (breakpoint)	MIC50/90 (range)	S% (breakpoint)	MIC50/90 (range)	S% (breakpoint)	MIC50/90 (range)
65	USA, 2013−2018, CF respiratory (100)	BMD	13 (≤16)	≤2/128 (≤2–>128)	29 (≤8)	8/32 (1–>32)		2/8 (≤0.5–>32)	28 (≤4)	1/>32 (≤0.5–>32)		≤0.5/4 (≤0.5–>8)
77	Spain, 2007–2017, bloodstream (13)	Aut	77 (≤16)		69 (≤8)		77 (≤2 and ≤4)		92 (≤2)		54 (≤2)		36 (≤1 and ≤0.5)
96	Hungary, 2013–2016, various (171)	DD	90 (≤16)		38 (≤8)		50 (≤4)		86 (≤2)		83b		<25 (≤0.5)
26	West Indies, 2006–2016, HAI (79)	Aut	75a		79b		84b				91b		13b
66	Europe, 2003–2016, CF respiratory (59)	BMD				8/128		2/8		2/32		0.5/8		8/32
10	Brazil, NR, CF respiratory (94)	Etest			88 (≤8)	0.032–256	90 (≤4)	0.25–≥32			90 (≤2)	0.032–256	76 (≤1)	0.38–4
10	Brazil, NR, CF respiratory (28)	Etest			100 (≤8)	0.75–12	100 (≤4)	0.125–1			79 (≤2)	0.023–32	43 (≤1)	0.38–6
97	France, 2010–2015, Various nonrespiratory (63)	DD	100 (≤16)		92 (≤8)		83 (≤2)		97 (≤2)				24 (≤1)
13	UK, 2015, CF respiratory (15–81)	Various	89 (≤16)		44 (≤8)		88 (≤4)		54 (≤2)		67b		4 (≤0.5)
63	France, 2014, CF respiratory (109)		88 (≤16)				79 (≤4)		72 (≤2)				23 (≤0.5)
47	Serbia, 2012–2013, Mostly respiratory (69)	DD, BMD	84 (≤16)		97 (≤8)		77 (≤4)		94 (≤2)		54 (≤2)		88 (≤1)
11	Argentina, 1996-2013, CF respiratory (41)	AD		0.06–32	0.25–64			0.03–128		0.06–64		0.03–256		0.5–64
29	Spain, 2007–2012, Skin and soft tissue (12–14)	Aut	93b		100b		79b				92b		21b
98	France, 2011–2012, Hospital (66%), domestic (18%) and outdoor (16%) (50)	DD	100 (≤16)		98 (≤8)		70 (≤4)		100 (≤2)				2 (≤0.5)
46	Argentina, NR, CF respiratory (24)	AD		0.3–4		4–32		0.5–4		0.125–4		0.13–256		2–64
36	China, 2008–2011, Respiratory (HAI) (41)	Aut	90 (≤16)		71 (≤8)		44 (≤2)		56 (≤2)		85 (≤2)		34 (≤0.5)
68	Argentina, 1995–2008, Various clinical (92)	AD		0.5/4 (0.06–32)		8/16 (1–256)		1/4 (0.5–256)		0.125/1 (0.016–8)		0.25/32 (0.03–128)		0.5–>64 (4–16)
70	France, 2008, Various clinical (25)	Etest					72 (≤2)	2/≥32 (1–≥32)	76 (≤2)	0.5/≥32 (0.06–≥32)
99	Italy, 2003–2008, CF sputum (53)	DD	100 (<4)		81 (<8)		81 (<2)		81 (<1)		81 (<0.5)		81 (<2)
100	Italy, 2003–2007 CF respiratory (42)	Aut	64 (≤64)b		29 (≤16)		74 (≤8)				38 (≤2)		12 (≤2)
28	Spain, 1994–2006, Urinary tract (9)	Aut	100 (≤16)		89 (≤8)		100 (≤4)				78 (≤0.5)		22 (≤1)
101	Italy, 2002–2004, CF sputum (62)c	Aut	85 (≤16)		44 (≤8)		84 (≤4)		68 (≤2)		53 (≤0.5)		20 (≤1)
102	USA, 1989–2003, bloodstream (9–48)	BMD	100 (≤16)		92 (≤8)		87 (≤4)		100 (≤2)		94 (≤0.5)		23 (≤1)
103	Latin America, 1997–2002, SENTRY surveillance (25)	BMD	76 (≤16)	1/64	64 (≤8)	8/>16	84 (≤4)	2/8	88 (≤2)	0.25/8	68 (≤0.5)	≤0.5/>2	32 (≤1)	2/>2
16	USA, 2001, CF respiratory (94)	BMD	55 (≤16)	32/128 (4–>128)	45 (≤8)	64/128 (2–>64)	59 (≤4)	4/>16 (1–>16)	51 (≤2)	8/>16 (0.5–>16)	0 (≤0.5)	>16/>16	19 (≤1)	>8/>8 (0.5–>8)
35	Spain, 1991–2000, bloodstream (12–54)	BMD	95 (≤16)				100 (≤4)		95 (≤2)		10 (≤0.5)		10 (≤1)
76	USA, 1991−1996, various (11)	BMD	100 (≤16)	≤1/≤1	100 (≤8)	2/2 (1–8)	100 (≤4)	1/1 (0.5–1)			100 (≤0.5)	≤2.5/5	9 (≤1)	2/2 (1–4)
76	USA, 1991−1995, various (24)	BMD	100 (≤16)	≤1/≤1 (≤1–2)	83 (≤8)	4/16 (2–16)	83 (≤4)	1/8 (0.5–8)			83 (≤0.5)	≤2.5/>160	4 (≤1)	8/>8 (1–>8)
104	USA, 1990, bloodstream and sputum (14)	BMD				4/>64 (1–>64)		2/4 (0.25–4)				0.03/0.03 (0.0075–0.06)		2/4 (0.5–8)
80	USA, 1983−1988, bloodstream (10)	BMD				2/4 (2–>32)		2/4 (0.5–4)				0.03/0.06 (0.03–0.12)		2/4 (0.12–4)
105	Belgium, 1983−1988, various (33)	BMD		0.5/64c (0.5–>128)		4/16 (1–64)		0.5/2 (0.25–4)		0.5/2 (0.25–2)		0.12/16 (0.12–64)		4/16 (1–64)

^aAbbreviations: S%, percent susceptible; TZP, piperacillin-tazobactam; CAZ, ceftazidime; IPM, imipenem; MEM, meropenem; SXT, trimethoprim-sulfamethoxazole; CIP, ciprofloxacin; AST, antibiotic susceptibility testing; DD, disk diffusion; AD, agar dilution; BMD, broth microdilution; Aut, automated systems; CF, cystic fibrosis; HAI, hospital-acquired infection. MIC values are expressed as μg/ml.

^bSusceptibility breakpoints are not specified.

^cPiperacillin susceptibility.

Studies that present data on tigecycline susceptibility are limited. The MIC_50/90 values were 2–4/4–16 μg/ml in three different studies (69–71). In an in vitro study comparing the activities of four tetracyclines against Achromobacter isolates, tigecycline and minocycline demonstrated lower MICs (MIC_50/90, 2/4 and 2/8 μg/ml, respectively), compared to doxycycline and tetracycline (MIC_50/90, 16/64 and 256/256 μg/ml, respectively) (71). Colistin may remain active in vitro against Achromobacter isolates. According to EUCAST antimicrobial wild-type distributions, the colistin modal MIC was 4 μg/ml for 150 A. xylosoxidans isolates obtained from three undefined sources, and the MIC range was 0.5 to 16 μg/ml except for the three isolates with an MIC of 512 μg/ml (72).

Azithromycin is a macrolide agent commonly used in CF patients to improve lung function and reduce P. aeruginosa exacerbations, mainly provided by its immunomodulatory and anti-inflammatory properties (73, 74). Achromobacter species are intrinsically resistant to azithromycin, and whether CF patients colonized with Achromobacter spp. would benefit from indirect antimicrobial effects of azithromycin is unclear. An in vitro study demonstrated enhanced Achromobacter killing in the presence of azithromycin when tested in mammalian tissue culture media rather than the standard bacterial growth media, which needs to be confirmed on larger collections and in vivo studies (75).

Susceptibility data on different Achromobacter species is limited. A. ruhlandii strains may exhibit a more resistant phenotype as demonstrated during a pandrug-resistant A. ruhlandii epidemic in Denmark, and more recently in an Argentinian collection, where most imipenem-resistant strains belonged to A. ruhlandii (i.e., the imipenem MIC was >4 μg/ml for 3/7, 1/26, and 0/8 of A. ruhlandii, A. xylosoxidans and other Achromobacter spp. [A. dolens, A. insuavis, A. pulmonis, and A. spiritinus], respectively) (11). In an early study, where species were identified using biochemical methods, A. denitrificans strains (n = 11) demonstrated a more susceptible phenotype compared to A. xylosoxidans strains (n = 24) (i.e., resistance to ceftazidime, meropenem, and trimethoprim-sulfamethoxazole was 7, 17, and 27% among 24 A. xylosoxidans isolates, whereas all 11 A. denitrificans remained susceptible to these agents) (76). Antibiotic susceptibility testing of large collections of isolates identified using reference methods is essential for a clearer picture of the susceptibility variation among different Achromobacter species.

Achromobacter genomes appear to harbor various resistance genes that are involved in the resistance to several antibiotics to various degrees. However, genetic mechanisms and environmental factors influencing resistance gene expression are not well characterized. Despite common resistance phenotypes among Achromobacter strains, (e.g., resistance to aminoglycoside, aztreonam, and cephalosporins except ceftazidime), rare reports of strains with various degrees of susceptibility to these agents are also present (76, 77). These inconsistencies may be caused by many factors, including variation in gene expression level, the presence of other unidentified resistance mechanisms, inaccurate identification of the nonfermenting Gram-negative bacilli, and variation in antibiotic susceptibility testing methods.

TREATMENT

Treatment of Achromobacter infections in CF patients is challenged by the intrinsic and acquired resistance to several first-line antibiotics. Efficacy of different treatment regimens on the outcomes of Achromobacter infections is unknown. Its rare occurrence, co-presence with more pathogenic nonfermenters (e.g., P. aeruginosa), and poorly understood clinical features hamper the evaluation of treatment efficacy on patient outcomes. Despite being regarded as nonpathogenic colonizers of the CF respiratory tract in earlier studies, Achromobacter species may cause frequent exacerbations and devastating post-lung transplant infections in CF patients (78). There are no standard treatment protocols for CF Achromobacter infections, and treatment usually consists of systemic and/or inhaled antibiotics. Addition of inhaled antibiotics may provide some benefit compared to systemic therapy alone, but the evidence comes from observational studies with few patients. In a study from Denmark, the addition of inhaled antibiotics (i.e., ceftazidime, colistin, and tobramycin) to active systemic therapy resulted in Achromobacter clearance after 3 years in 10 of 17 (59%) patients compared to one of six (17%) patients without inhaled antibiotics (79). Similar results were observed among those who received inactive systemic antibiotics. The outcomes in this study may be influenced by factors other than the inhaled therapy per se, and small numbers do not allow to adjust for confounders. Controlled clinical studies are needed to evaluate the effect of inhaled antibiotics and of different systemic antibiotics for the treatment of Achromobacter infections (79). Until then, treatment of Achromobacter respiratory infections will need to be evaluated on a case-by-case basis, taking into account several factors, including infection severity, infection frequency, suppressive antibiotics received, and in vitro antibiotic susceptibility.

Treatment of Achromobacter infections outside the context of CF depends on the site of infection and patient factors in addition to disease severity. Favorable clinical outcomes were observed with β-lactams, including ceftazidime, piperacillin-tazobactam, carbapenems, and trimethoprim-sulfamethoxazole for bloodstream infections (80) (Table 3). In a series of 10 cancer patients with Achromobacter bacteremia, all had favorable outcomes with β-lactams (e.g., ceftazidime, piperacillin, and ticarcillin) and/or trimethoprim-sulfamethoxazole, including six patients who received monotherapy with one of these agents. It is important to note that four of six patients with monotherapy had their central venous catheter removed, which may have contributed to the favorable outcomes. Similarly, successful outcomes were observed with foreign device associated CNS infections treated with ceftazidime or doripenem in addition to foreign device removal (81, 82). In another case series, 15 elderly patients (median age, 89 years) with hospital-acquired Achromobacter pneumonia were successfully treated with piperacillin-tazobactam, meropenem, and imipenem monotherapies (n = 9) (45). Five patients died on day 30, and four of these were receiving combination therapies (e.g., meropenem or piperacillin-tazobactam plus ceftazidime or minocycline).

TABLE 3.

In vitro activities of new antibiotics against Achromobacter isolates^a

Reference	Yr collected	Region	No. of isolates	ERV MIC50/90 (range)	FDC MIC50/90 (range)	MEM		MVB		CAZ		CZA
						MIC50/90 (range)	R%	MIC50/90 (range)	R%	MIC50/90 (range)	R%	MIC50/90 (range)	R%
6	2013–2018	USA	100			1/>32 (≤0.5–>32)	72	≤0.5/8 (≤0.5–>32)	86	8/32 (1–>32)	71	8/32 (1–>32)	78
85		USA	15		<0.03/0.125 (<0.03–16)	1/16 (0.125–16)						4/32 (1–>64)
–b	2013–2014, 2016–2017	Global	19	1/8 (0.06–8)
87	2017–2019	USA	14	2c,d (1–2)	Se			64c (0.25–256)				8c (2–16)
92		France	11			32/128 (0.5–>128)	9			128/>128 (32–>128)		128/>128 (16–>128)
88		France	30							128/256	40	128/256	40

^aAbbreviations: R%, percent resistant; ERV, eravacycline; FDC, cefiderocol; MEM, meropenem; MVB, meropenem-vaborbactam; CAZ, ceftazidime; CZA, ceftazidime-avibactam; AST, antibiotic susceptibility testing. MIC values are expressed as μg/ml.

^b–, Personal communication, Tetraphase Pharmaceuticals.

^cMean value.

^dMIC for three isolates.

^eThe “S” interpretation is based on P. aeruginosa disk diffusion breakpoints provided by Shionogi (breakpoint not provided).

Achromobacter infections in non-CF hosts can be managed with appropriate source control and treatment with a single β-lactam antibiotic (e.g., ceftazidime and piperacillin-tazobactam) with in vitro activity against the infecting strain. Carbapenems should be spared where possible. Trimethoprim-sulfamethoxazole alone may be considered for less severe infections, such as urinary tract infections, but the data are very limited for its use in severe infections. For strains with elevated β-lactam MICs, combination with an active non-β-lactam antibiotic such as trimethoprim-sulfamethoxazole, tetracyclines, fluoroquinolones, or colistin may be considered, but the evidence behind this recommendation is very poor.

NEW THERAPEUTIC OPTIONS AGAINST ACHROMOBACTER INFECTIONS

Cefiderocol.

Cefiderocol is a new-generation cephalosporin with in vitro activity against carbapenem-resistant Gram-negative nonfermenters such as P. aeruginosa, A. baumannii, and Burkholderia spp. (83, 84). Although in vitro data on cefiderocol activity against Achromobacter spp. are limited (85), this agent was used to treat nine patients with multidrug-resistant (MDR) Achromobacter infections as part of the compassionate use program (Tables 3 and 4). Most were CF respiratory tract infections (four with a lung transplant) (86). One of the cases was chronically colonized with A. xylosoxidans and developed A. xylosoxidans bacteremia post-lung transplant (A. xylosoxidans susceptible to cefiderocol [MIC, 0.12 μg/ml], piperacillin-tazobactam, trimethoprim-sulfamethoxazole, imipenem, and colistin). This patient was put on extended-infusion piperacillin-tazobactam therapy, with no improvement. At this stage, cefiderocol was added to the treatment and continued for 28 days with a favorable clinical response. However, a relapse of pneumonia occurred 14 days after discharge, and he was put on cefiderocol plus imipenem this time, which was continued for 42 days. The patient was reported to be doing well at 8-month follow-up (24). The second case was chronically colonized with A. xylosoxidans pretransplant (intermediate susceptibility to piperacillin-tazobactam and resistant to all other drugs tested) and was treated with cefiderocol (MIC, 1 μg/ml) posttransplant (6 weeks) together with meropenem (5 weeks) as part of the planned peri-transplant treatment regimen. The patient was reported to be well at 4-months follow up, although still colonized with A. xylosoxidans (24). An implantable cardioverter-defibrillator-associated bacteremia case and postoperative wound infection complicated with osteomyelitis were two other complicated cases treated with cefiderocol with favorable outcomes (cefiderocol MICs were not reported) (86). Cefiderocol was always used in combination, in one case together with meropenem-vaborbactam and phage therapy. These results are promising, particularly for the CF patients infected with carbapenem-resistant Achromobacter infections. However, an FDA warning for increased all-cause mortality with cefiderocol in patients with carbapenem-resistant Gram-negative bacterial infections is in place due to the higher death rate in the cefiderocol arm compared to the best available therapy in a recent randomized controlled trial of cefiderocol for the treatment of serious carbapenem-resistant Gram-negative infections (86).

TABLE 4.

Achromobacter infections treated with cefiderocol^a

Case	Age (yr) and sex	Infection type	Underlying condition	Causative pathogen	Susceptibility	Concomitant antibiotics	Treatment duration (days)	Outcome
1	28, M	Bacteremia, pneumonia	CF, lung transplant	A. xylosoxidans	Susceptible to TZP and SXT	TZP, IPM, DOX	28+42 (with a 14-day break)	Patient well at 8-month follow-up
2	17, F	Respiratory tract colonization	CF, lung transplant	A. xylosoxidans	Intermediate to TZP, resistant to all other drugs tested	CAZ, CST (nebulized), DOX, IPM, SXT	42	Patient well at 4-month follow-up, asymptomatically colonized with A. xylosoxidans.
3	56, F	Pneumonia, empyema (post-op)	CF, lung transplant	Achromobacter spp.		TZP, CST	2+13 (with a 59-day break)	Discharged (rehabilitation facility)
4	28, F	Respiratory tract infection	CF, lung transplant	Achromobacter spp.		CST, SXT	15	Discharged (home) with clinical improvement
5	41, M	Respiratory tract infection	CF, chronic bronchiectasis	A. denitrificans	XDR	ERV, AZM, C/T, CST (nebulized)	67	Discharged (home) with clinical improvement
6	10, F	Respiratory tract infection	CF, asthma	Achromobacter spp.	Pan-resistant	MVB, bacteriophage therapy	21	Discharged (home) with clinical improvement
7	70, F	Respiratory tract infection	Chronic bronchiectasis	A. xylosoxidans		-	26	Discharged (rehabilitation facility)
8	66, M	ICD associated bacteremia	Severe left ventricle dysfunction, lung adenocarcinoma	A. xylosoxidans		TZP, TGC	11	Patient well with oral suppressive minocycline therapy
9	28, F	Osteomyelitis	Postoperative wound infection with osteomyelitis	A. xylosoxidans [with P. aeruginosa (VIM+), A. baumannii (OXA-23+), E. cloacae (KPC+)]		CST, TGC, CZA	20	Patient well with complete resolution of bone infection

^aAbbreviations: CF, cystic fibrosis; IPM, imipenem; TZP, piperacillin-tazobactam; SXT, trimethoprim-sulfamethoxazole; DOX, doxycycline; CAZ, ceftazidime; CST, colistin; ERV, eravacycline; AZM, azithromycin; C/T, ceftolozane-tazobactam; MVB, meropenem-vaborbactam; TGC, tigecycline; CZA ceftazidime-avibactam; ICD, intracardiac device; XDR, extensively drug resistant. The table modified from the FDA (86).

Eravacycline.

Eravacycline is a new tetracycline stable against most tetracycline efflux pumps of Enterobacteriaceae. However, it is extruded by AdeABC pump of A. baumannii, which is an RND type efflux pump similar to AxyXY-OprZ of Achromobacter. Whether eravacycline is a substrate of the AxyXY-OprZ pump or not remains unknown. Limited in vitro data demonstrate some activity of eravacycline against 19 Achromobacter isolates from IGNITE1 and IGNITE4 clinical trials (the MIC_50/90 was 1/8 μg/ml; range, 0.06 to 8 μg/ml) (87). Eravacycline was also used for the treatment of seven patients with MDR Achromobacter spp. respiratory tract infections (Table 5) (personal communication, Tetraphase Pharmaceuticals) (87). It is difficult to evaluate the impact of eravacycline on patient outcomes since it was often used in combination with other antibiotics, except two cases where the favorable outcome may have been provided exclusively by eravacycline. The first case was a 30-year-old CF patient, previously treated with meropenem-vaborbactam and colistin twice due to A. xylosoxidans growth in sputum (clinical syndrome not specified). In the current episode, she was treated with eravacycline for 7 days, which was replaced with tigecycline for the last 4 days of the treatment (the MICs were not reported). She had favorable outcomes (i.e., did not have in-hospital mortality or 30-day readmission due to infection; further details of the “favorable outcome” were not reported). The second case was a 22-year-old CF patient who had A. xylosoxidans growth in the bronchial wash and was treated with eravacycline (MIC 2 μg/ml) from days 2 to 11 and days 14 to 18 of hospitalization and had favorable outcomes. Among the other agents that were included in the treatment regimen of the second case, only colistin had MICs within susceptible range, but it was started at day 10.

TABLE 5.

Achromobacter infections treated with eravacycline and meropenem-vaborbactam^a

Caseb	Age and sex	Infection type	Underlying condition	Susceptibility (MIC)	Treatment regimen	ERV or MVB treatment duration (days)	Concomitant microorganism(s) in respiratory culture	Outcome
1	40, M	Sinusitis	CF	MDR	ERV single therapy	22	None	Favorablec
2	41, M	Respiratory tract	CF	FDC, 0.06 μg/ml; ERV, 2 μg/ml	ERV, D1-31; FDC, D10-31; C/T, D4-10; MEM, D5-10; DLX, D10-31	31	MDR P. aeruginosa	Favorable
3	63, M	Respiratory tract	COPD	TZP, 16 μg/ml; TOB, 8 μg/ml; MEM, ≥16 μg/ml; MIN, 4 μg/ml; CST, 2 μg/ml; CIP, 4 μg/ml; CAZ, 8 μg/ml; SXT, ≤20 μg/ml; ERV, not determined	TZP, D1-2; MIN, D2-9 and D24-34; CST, D6-9 and D35-37; CIP, D9-27; CZA, D22-35; SXT, D27-34; ERV, D29-34	5	MDR P. aeruginosa, MDR S. maltophilia, MRSA	Patient died in hospital
4	45, F	Respiratory tract	Chronic respiratory failure	CZA, 4 μg/ml; MVB, 8 μg/ml; TZP, 8–≥128 μg/mld	CZA, D1-15 and D23-27; MVB, D22-23 and D42; TZP, D42-57, D61-79, and D114-121; ERV, D106-108 and D112-114	4	Enterobacter cloacae complex in separate culture	Patient died in hospital
5	75, F	Respiratory tract	COPD, chronic bronchiectasis	MEM susceptiblee	MEM, D1-4; ERV, D4-13	9	None	Infection-related readmission in 30 days
6	30, F	Respiratory tract	CF	MDR	ERV, D1-7; TGC, D7-11	7	None	Favorable
7	22, M	Respiratory tract	CF	TOB, ≥16 μg/ml; FEP, ≥64 μg/ml; ERV, 2 μg/ml; SXT, 40 μg/ml; TZP, 256 μg/ml; CST, 0.5 μg/ml	TOB, D1-5; FEP, D1-5; ERV, D2-11 and D14-18; SXT, D5-25; TZP, D5-7; CST, D10-18	13	Group A Streptococcus	Favorable
8	80, F	Respiratory tract	Chronic bronchiectasis	MDR	MVB single therapy	20	None	Favorable
9	40, M	Respiratory tract	CF	CST, 0.12 μg/ml; MVB, 32 μg/ml	MVB+CST combination therapy	14	MDR P. aeruginosa	Favorable
10	29, F	Respiratory tract	CF	MDR	MVB+CST combination therapy	10	None	Favorable
11	26, M	Respiratory tract	CF	MEM 4 μg/ml	MEM, D1-2; MIN, D1-5; MVB, D3-5; CPT, D5-16	2	MRSA, E. coli	Favorable

^aAbbreviations: CF, cystic fibrosis; COPD, chronic obstructive pulmonary disease; MDR, multidrug resistant; D, day; MRSA, methicillin-resistant Staphylococcus aureus; FEP, cefepime; FDC, cefiderocol; CIP, ciprofloxacin; CZA, ceftazidime-avibactam; C/T, ceftolozane-tazobactam; CST, colistin; ERV, eravacycline; MEM, meropenem; MVB, meropenem-vaborbactam; MIN, minocycline; TZP, piperacillin-tazobactam; TGC, tigecycline; SXT, trimethoprim-sulfamethoxazole; TOB, tobramycin; DLX, delafloxacin; CPT, ceftaroline. The table was modified from Canonica et al. (87). A. xylosoxidans and A. denitrificans were the causative pathogens for cases 1 to 10 and case 11, respectively.

^bCases 1 and 2 are the same patient.

^c“Favorable” is defined as the patient alive until discharge and not readmitted within 30 days due to infection.

^dThe MIC was provided as a range for TZP in the original table, and the reason was not specified.

^eThe MIC was not provided.

Clinical experience with eravacycline for the treatment of Gram-negative bacterial infections in general is limited. Although eravacycline was not tested in randomized trials for the treatment of pneumonia, an FDA warning for higher all-cause mortality for the treatment of pneumonia was issued for tigecycline, predecessor of eravacycline, which warrants caution with eravacycline, as well.

β-Lactam/β-lactamase inhibitors.

Acquired carbapenem resistance occur via MBLs in Achromobacter species, limiting the utility of new β-lactam/β-lactamase inhibitors against carbapenem-resistant Achromobacter isolates. Despite vaborbactam being inactive against MBLs, meropenem-vaborbactam was used to treat Achromobacter infections in a number of cases (Table 5) (87). One of them was a 29-year-old CF patient with A. xylosoxidans growth in sputum (the clinical syndrome was not specified). Meropenem-vaborbactam was used in combination with colistin, yielding favorable clinical outcomes, which could have been provided by meropenem per se or colistin, rather than meropenem-vaborbactam (antibiotic MICs were not reported). This patient had a subsequent episode of A. xylosoxidans growth in sputum, which was treated with eravacycline, followed by tigecycline. Meropenem-vaborbactam monotherapy was also used to treat an 80-year-old patient with bronchiectasis and A. xylosoxidans growth in sputum, with favorable outcomes. Meropenem and meropenem-vaborbactam susceptibilities were not reported for this isolate, either. For the two other cases with meropenem-vaborbactam use, meropenem-vaborbactam the MIC values were 8 and 32 μg/ml, and the drug was used as part of combination therapies against polymicrobial infections (87). The in vitro activity of meropenem-vaborbactam was demonstrated to be slightly higher than that of meropenem against Achromobacter spp. Meropenem-vaborbactam was active against 86% of 100 CF-derived A. xylosoxidans, A. ruhlandii, and A. dolens isolates (MIC_50/90, ≤0.5/8 μg/ml; range, ≤0.5 to 32 μg/ml; susceptibility breakpoint, ≤4 μg/ml), whereas meropenem alone was active against 72% of the isolates (MIC_50/90, 1/>32 μg/ml; range, ≤0.5 to >32 μg/ml; susceptibility breakpoint, ≤4 μg/ml) (65). The reason behind this difference remains unknown since vaborbactam is not expected to enhance the activity of meropenem against Achromobacter species.

Among the other new β-lactam/β-lactamase inhibitors, the in vitro activity of imipenem-relebactam against 345 Achromobacter isolates was somewhat favorable, since 44% of the isolates were inhibited at a concentration of 1 μg/ml, and 95% were inhibited at 4 μg/ml (MIC range, 0.12 to >32 μg/ml; personal communication, Merck & Co.). The imipenem MICs for these isolates are not reported, and any additional benefit of relebactam to imipenem is unknown. Ceftazidime-avibactam and ceftazidime showed similar in vitro activities against Achromobacter isolates, the MIC_50/90 being 8/32 μg/ml for ceftazidime with or without avibactam (65). Higher MICs (MIC_50/90, 128/>128 μg/ml) and lower susceptibility rates with ceftazidime-avibactam were reported in other studies (88, 89). Avibactam combined with either meropenem or aztreonam had poor activities against Achromobacter spp. (89). Ceftolozane-tazobactam does not have any activity against Achromobacter species, as demonstrated in vitro (87, 90–92).

Plazomicin.

Plazomicin is a new-generation aminoglycoside that remains stable against most aminoglycoside modifying enzymes but is influenced by the efflux pumps of A. baumannii and P. aeruginosa (93). Although data are lacking regarding the activity of AxyAX-OprZ of Achromobacter against plazomicin, the susceptibility of plazomicin to the MexXY efflux pump of P. aeruginosa was demonstrated (94), making the inhibition of Achromobacter by plazomicin unlikely.

Phage therapy.

Achromobacter-specific phages with a spectrum of 24 different Achromobacter strains, including MDR strains, were used for the treatment of a 17-year-old CF patient chronically colonized with A. xylosoxidans (95). The patient was put on a 20-day oral and inhaler Achromobacter-specific phage therapy, following a recent P. aeruginosa infection episode and continuing Achromobacter colonization. Her symptoms and lung capacity improved after a 20-day phage therapy, which was repeated quarterly for a year. Achromobacter-specific phages may be a promising option for the treatment of chronic colonization and infection with Achromobacter in CF population.

CONCLUSION

Treatment of Achromobacter infections pose a clinical challenge. Intrinsic resistance to several antibiotic classes mediated mainly by multidrug efflux pumps and chromosomal β-lactamases accompanied by acquired carbapenem resistance caused by MBLs leave very few treatment options for their treatment. New β-lactam/β-lactamase inhibitor combinations with anti-carbapenemase activity do not provide much benefit since serine carbapenemase production is not the main mechanism of resistance in Achromobacter spp. Among the other new antibiotics, eravacycline and cefiderocol may have a role for the treatment of MDR Achromobacter infections. The need for more clinical, microbiological, and genomic data is obvious for the management of MDR Achromobacter infections, which can only be provided by multicenter studies due to the rarity of these infections. Furthermore, whole-genome sequencing of environmental isolates in addition to clinical collections would facilitate characterization of the population structure and identification of antimicrobial resistance mechanisms of Achromobacter species.