In Vitro Susceptibility of Achromobacter Species Isolated from Cystic Fibrosis Patients: a 6-Year Survey

ABSTRACT

We conducted in vitro antimicrobial susceptibility testing of 267 Achromobacter isolates for 16 antibiotics from 2017 to 2022. The highest susceptibility was found for piperacillin-tazobactam (70%) and ceftazidime-avibactam (62%). Between 30% and 49% of strains were susceptible to tigecycline, ceftazidime, and meropenem. We applied species-specific Achromobacter xylosoxidans breakpoints for piperacillin-tazobactam, meropenem, and trimethoprim-sulfamethoxazole and EUCAST pharmacokinetic/pharmacodynamic (PK/PD) breakpoints for the others. A. xylosoxidans was the most frequently isolated species, followed by Achromobacter insuavis and Achromobacter ruhlandii.

TEXT

Achromobacter spp. are human opportunistic pathogens causing infections in immunocompromised hosts. These nonfermenting Gram-negative bacilli are widely distributed in nature, especially in aquatic environments, and are part of the normal gastrointestinal microbiota (1). They are commonly isolated from respiratory samples of patients with cystic fibrosis (CF) (1, 2). Achromobacter species have characteristics that intrinsically favor persistent lung colonization (3). They express specific secretion systems and several virulence factors that significantly impair respiratory cell survival (4, 5). Even though it is not possible to definitely establish a causal relationship between these pathogens and deterioration of lung structure and function, infection with Achromobacter spp. is considered a marker of CF severity (3). When these pathogens are detected in samples from CF patients, there is general agreement on the need for antibiotic treatment (6, 7).

Unfortunately, eradicating Achromobacter infections in CF patients is challenging, and there are no standard treatment protocols recommended by scientific societies. Therefore, a case-by-case approach is required (7, 8). The severity of infection, frequency of infection, prior antibiotic usage, and in vitro antibiotic susceptibility are factors that must be considered. Treatments with systemic and/or inhaled antibiotics are typically prescribed (8, 9). Trimethoprim-sulfamethoxazole, ceftazidime, piperacillin, and carbapenems have been the most frequently used antimicrobial agents against Achromobacter isolates (10, 11). However, resistance to these agents is increasing (7, 12).

Achromobacter spp. are intrinsically resistant to several antibiotics, including most cephalosporins, aztreonam, and aminoglycosides. These multidrug-resistant pathogens are increasingly acquiring resistance to carbapenems (7, 13, 14). The two major intrinsic resistance mechanisms of Achromobacter spp. are multidrug efflux pumps and chromosomal OXA-114-like beta-lactamases. Efflux pump-associated genes are abundant in Achromobacter strains (15). Two multidrug efflux pumps, AxyABM and AxyXY-OprZ, alongside some putative other efflux pump genes, have been identified in Achromobacter species (7, 15). All publicly available Achromobacter genomes have AxyABM, which is associated with the efflux of beta-lactams, fluoroquinolones, and chloramphenicol but not aminoglycosides (16, 17). The MexAB-OprM efflux pump of Pseudomonas aeruginosa and the AxyABM of the Achromobacter efflux pump share some properties; however, MexAB-OprM exports cefepime and meropenem, while AxyABM does not (17–19). The second efflux pump, AxyXY-OprZ, has a wider spectrum and is involved to various degrees in the extrusion of aminoglycosides, cefepime, carbapenems, fluoroquinolones, tetracyclines, and erythromycin (17, 18). Other mechanisms that contribute to increased beta-lactam and carbapenem resistance include extended-spectrum beta-lactamases (ESBLs), AmpC-type beta-lactamases, and metallo-beta-lactamases (MBLs) (7).

Newly developed drugs, such as ceftazidime-avibactam, offer potential new treatment options. However, further studies are urgently needed to determine the proper dosage, treatment duration, efficacy, and safety of this new antibiotic for CF patients (2). The ceftazidime-avibactam combination, which is a combination of a cephalosporin and a non-beta-lactam beta-lactamase inhibitor, can inhibit class A, class C, and some class D beta-lactamases (20–22). This new antimicrobial combination is indicated for complicated urinary tract infections, intra-abdominal infections, pneumonia, and targeted therapy of severe multidrug-resistant infections caused by Enterobacterales and Pseudomonas spp. (20, 23–25).

Currently, MIC breakpoints for Achromobacter spp. have not been definitively established due to different resistance mechanisms. Recently, the European Committee on Antimicrobial Susceptibility Testing (EUCAST) proposed new clinical Achromobacter xylosoxidans MIC breakpoints for piperacillin-tazobactam, meropenem, and trimethoprim-sulfamethoxazole (26, 27). The Clinical and Laboratory Standards Institute (CLSI) reports MIC breakpoints for Achromobacter spp. under the category of “other non-Enterobacterales” (28).

At the Belgian National Reference Center for Burkholderia cepacia complex and other Gram-negative nonfermenters (except P. aeruginosa and Acinetobacter spp.), we conducted in vitro antimicrobial susceptibility testing of 267 Achromobacter isolates. These isolates were mainly from CF patients and were tested against 16 antibiotics by broth microdilution between 2017 and 2022. Samples were collected from our own hospital (Universitair Ziekenhuis Brussel) as well as from other Belgian hospitals, including Algemeen Ziekenhuis Voorkempen Malle, Centre Hospitalier Chrétien Clinique MontLégia Liège, Cliniques Universitaires Saint-Luc Brussel, Vitaz Ziekenhuizen Campus Sint-Gillis-Waas, and Universitair Ziekenhuis Leuven.

To identify Achromobacter isolates to species level, sequence analysis of a 765-bp nrdA gene fragment (nrdA_765) was performed as described by Spilker et al. (2, 29). DNA was extracted using an automated Maxwell DNA preparation instrument (Promega, USA). DNA extracts were treated with RNase (2 mg/mL, 5 μL per 100 μL extract) and incubated at 37°C for 1 h. DNA quantification was performed using the QuantiFluor ONE double-stranded DNA (dsDNA) system and the Quantus fluorometer (Promega, USA). Amplification and sequencing were performed using the forward primer 5′-GAACTGGATTCCCGACCTGTTC-3′ and reverse primer 5′-TTCGATTTGACGTACAAGTTCTGG-3′ as described by Spilker et al. (2, 29). The resulting products were purified using a NucleoFast 96 PCR cleanup kit (Macherey-Nagel). Purified products were sequenced by a commercial company (Eurofins, Germany). Sequences were assembled using BioNumerics 7.6 (Applied Maths, Belgium) with default settings. All nrdA gene sequences were compared with those of Achromobacter type and reference strains in the Achromobacter pubMLST database (https://pubmlst.org/organisms/achromobacter-spp) (30).

MIC values were determined using broth microdilution according to ISO standard 20776-1 in microtiter plates and were read on a Sensititre Vizion system (Thermo Scientific) (27). To determine in vitro susceptibilities of piperacillin-tazobactam, meropenem, and trimethoprim-sulfamethoxazole, we used the recently released species-specific EUCAST MIC breakpoints for A. xylosoxidans (26, 27). For temocillin, we used the breakpoints described by Fuchs et al. (31). As species-specific EUCAST breakpoints were not available for the other 12 antibiotics, we used the EUCAST pharmacokinetic/pharmacodynamics (PK/PD) breakpoints (26, 27). We compared these in vitro susceptibilities with those derived from CLSI breakpoints for “other non-Enterobacterales” (Table 1) (28).

TABLE 1.

MIC distributions and in vitro susceptibilities of all Achromobacter strains (n = 267) for 16 tested antibiotics^f

Antibiotic	No. of isolates with MIC (mg/L):										EUCAST breakpoint (mg/L)c			CLSI other non-Enterobacterales breakpoint (μg/mL)d
	0.125a	0.25a	0.5a	1a	2a	4a	8b	16 b	32b	64b	S≤	R>	Susceptibility (%)	S≤	R>	Susceptibility (%)
Amikacin						9	4	4	163	86	1	1	0	16	64	6
Amoxicillin						3	22	144			2	8	0
Amoxicillin-clavulanic acid					1	9	49	33	78		2	8	1	16	128	54
Aztreonam			2			1		2	261	1	4	8	1	8	32	1
Cefepime					1	7	24	164	70		4	8	3	8	32	12
Ceftazidime				1	19	69	68	79	30		4	8	33	8	32	59
Ceftazidime-avibactam				1	14	49	40	65			8	8	62
Ceftolozane-tazobactam						4	12	153			4	4	2
Ciprofloxacin		2	1	15	47	55	102	44	1		0.25	0.5	1	1	4	7
Colistin			7	19	59	61	23	65	32		IEe	IE
Meropenem		33	9	89	25	28	19	42	22		1	4	49	4	16	69
Piperacillin-tazobactam			27	24	109	26	18	8	38	16	4	4	70	16	128	80
Temocillin				1	1		1	1	165	97	8	8	1
Tigecycline		25	55	78	55	29	20	5			0.5	0.5	30
Tobramycin				5	8	5	10	237		1	0.5	0.5	0	4	16	7
Trimethoprim-sulfamethoxazole	8	1	47	97	11	20	25	45	13		0.125	0.125	3	2	4	61

^aBelow or equal to the respective MIC.

^bAbove or equal to the respective MIC.

^cTemocillin breakpoints are from the work of Fuchs et al. (31). Meropenem, piperacillin-tazobactam, and trimethoprim-sulfamethoxazole breakpoints are from the Achromobacter xylosoxidans EUCAST breakpoints (26). Other breakpoints are from the PK/PD nonspecies EUCAST breakpoints (27).

^dAmoxicillin-clavulanic acid breakpoints are derived from those for ticarcillin-clavulanic acid (28).

^e“IE” indicates that there is insufficient evidence that the organism or group is a good target for therapy with the agent. A MIC with a comment but without an accompanying S, I, or R categorization may be reported (27).

^fS, susceptible; R, resistant; light gray shading, susceptible based on EUCAST breakpoints; medium gray shading, intermediate based on EUCAST breakpoints; boldface, resistant based on EUCAST breakpoints; italic, median of all samples of the corresponding antibiotic (the MIC above the italic number corresponds to the MIC₅₀).

The most effective agents against Achromobacter isolates were piperacillin-tazobactam and ceftazidime-avibactam, with 70% and 62% of isolates susceptible, respectively, using EUCAST breakpoints. Based on CLSI breakpoints, piperacillin-tazobactam was the most effective antibiotic at 80%, followed by meropenem (69%), trimethoprim-sulfamethoxazole (61%), ceftazidime (59%), and amoxicillin-clavulanic acid (54%). As expected, we found no in vitro antimicrobial activity for amikacin, amoxicillin, amoxicillin-clavulanic acid, aztreonam, cefepime, ceftolozane-tazobactam, ciprofloxacin, temocillin, and tobramycin using the EUCAST breakpoints. Using CLSI breakpoints, amikacin, aztreonam, cefepime, ciprofloxacin, and tobramycin displayed a poor sensitivity (6%, 1%, 12%, 7%, and 7%, respectively). Contrary to general expectations, only 3% of the isolates were susceptible to trimethoprim-sulfamethoxazole using the most recent species-specific EUCAST breakpoints of A. xylosoxidans. Moderate susceptibility was observed for ceftazidime (33%), meropenem (49%), and tigecycline (30%) using EUCAST breakpoints. The largest differences in susceptibility using EUCAST and CLSI breakpoints were observed for amoxicillin-clavulanic acid, ceftazidime, meropenem, and trimethoprim-sulfamethoxazole (Table 1). A median value below the EUCAST-S breakpoint may indicate an appropriate therapy selection, as observed for ceftazidime-avibactam and piperacillin-tazobactam (Table 1).

Among the 267 Achromobacter isolates identified to species level, A. xylosoxidans was the most frequently isolated species (65.5%), followed by A. insuavis (17.6%), A. ruhlandii (5.6%), and A. aegrifaciens (3.4%). A. mucicolens (1.9%), A. marplatensis (1.9%), A. spanius (1.9%), A. pulmonis (0.4%), A. animicus (0.7%), and A. dolens (0.4%) occurred less frequently (Table 2). The distributions of Achromobacter species appear to be different among countries (Table 2) (32–36).

TABLE 2.

Prevalence of isolates of Achromobacter species in different countries with available data^a

Achromobacter species	% of isolates by country (no. in study)
	Belgium (267)	Denmark (42)	France (63)	UK (147)	Spain (26)	Argentina (41)
A. xylosoxidans	65.5	35.7	44.4	60.5	57.7	63.4
A. insuavis	17.6	23.8	14.3	11.6		4.9
A. ruhlandii	5.6	19.0		2.0		17.1
A. aegrifaciens	3.4	4.8	11.1
A. mucicolens	1.9	2.4	1.6	0.7
A. marplatensis	1.9	2.4	0.0	1.4		2.4
A. spanius	1.9	2.4	1.6
A. pulmonis	0.4				3.8	2.4
A. animicus	0.7		7.9
A. dolens	0.4	2.4		11.6		9.8

^aValues in table are taken from references 32 to 36.

The present study investigated the susceptibility of 267 Achromobacter isolates, in contrast to previous studies that analyzed a limited number of isolates (32–35). Notable differences in in vitro susceptibilities among Achromobacter spp. were observed (Table 3). All 15 A. ruhlandii strains were resistant to tigecycline, while A. xylosoxidans (n = 175), A. insuavis (n = 47), and A. aegrifaciens (n = 9) had sensitivities of 23%, 49%, and 22%, respectively. The highest sensitivity for A. xylosoxidans and A. insuavis was achieved with piperacillin-tazobactam, at 68% and 74%, respectively. Although piperacillin-tazobactam also achieved good sensitivity for A. aegrifaciens (67%), it had a much poorer result for A. ruhlandii (47%). As such, ceftazidime-avibactam would be the first-choice agent, since it had good sensitivity for all species studied, especially for A. aegrifaciens with 83%. For the latter species, meropenem also stands out with a sensitivity of 78%, making it a valuable alternative. However, for the other species, meropenem is clearly less effective, with only 13% of A. ruhlandii isolates being susceptible (Table 3).

TABLE 3.

In vitro susceptibility by Achromobacter species

Antibiotic	EUCAST breakpoint (mg/L)a		Susceptibility by species (%)
	S≤	R>	All (n = 267)	A. xylosoxidans (n = 175)	A. insuavis (n = 47)	A. ruhlandii (n = 15)	A. aegrifaciens (n = 9)
Amikacin	1	1	0	0	0	0	0
Amoxicillin	2	8	0	0	0	0	0
Amoxicillin-clavulanic acid	2	8	1	1	0	0	0
Aztreonam	4	8	1	1	2	0	0
Cefepime	4	8	3	2	0	20	0
Ceftazidime	4	8	33	26	45	40	44
Ceftazidime-avibactam	8	8	62	57	66	60	83
Ceftolozane-tazobactam	4	4	2	2	0	0	0
Ciprofloxacin	0.25	0.5	1	0	2	0	0
Colistin	IE	IE
Meropenem	1	4	49	47	40	13	78
Piperacillin-tazobactam	4	4	70	68	74	47	67
Temocillin	8	8	1	1	2	0	11
Tigecycline	0.5	0.5	30	23	49	0	22
Tobramycin	0.5	0.5	0	0	0	0	0
Trimethoprim-sulfamethoxazole	0.125	0.125	3	2	0	7	0

^aTemocillin breakpoints are from the work of Fuchs et al. (31). Meropenem, piperacillin-tazobactam, and trimethoprim-sulfamethoxazole breakpoints are from the Achromobacter xylosoxidans EUCAST breakpoints (26). Other breakpoints are from the PK/PD nonspecies EUCAST breakpoints (27). “IE” indicates that there is insufficient evidence that the organism or group is a good target for therapy with the agent. A MIC with a comment but without an accompanying S, I, or R categorization may be reported (27).

Spencer et al. reported piperacillin-tazobactam, trimethoprim-sulfamethoxazole, meropenem, and ceftazidime as recommended first-line empirical treatments for infection (37). Isler et al. also identified trimethoprim-sulfamethoxazole, ceftazidime, piperacillin, and carbapenems as effective agents against Achromobacter isolates (7). For our isolates, piperacillin-tazobactam, along with ceftazidime-avibactam, were the most active antibiotics. Using the new species-specific breakpoints for A. xylosoxidans, we found reduced sensitivity (3%) to trimethoprim-sulfamethoxazole. This is in contrast to the sensitivity (69%) found with the previously used EUCAST breakpoints for nonfermenters and the sensitivity (61%) found with the CLSI breakpoints. Therefore, we recommend caution with the use of trimethoprim-sulfamethoxazole as first-line therapy for Achromobacter infections. For meropenem, moderate in vitro susceptibility (49%) was found using the EUCAST species-specific breakpoints for A. xylosoxidans. Isolates were more susceptible (69%) to meropenem when CLSI breakpoints were used. However, increasing resistance is observed with carbapenems due to multidrug efflux pumps and metallo-beta-lactamases (15, 38). Only 33% of isolates were sensitive to ceftazidime. Combining the non-beta-lactam beta-lactamase inhibitor avibactam with ceftazidime increased sensitivity to 62%. As a result, ceftazidime-avibactam was the second most active antibiotic among our samples.

In general, piperacillin-tazobactam and ceftazidime-avibactam were the most effective antibiotic agents with a susceptibility of 70% and 62%, respectively, based on in vitro testing using EUCAST breakpoints. Meropenem had a susceptibility of 49%, while ceftazidime and tigecycline had 33% and 30% susceptibility, respectively. When using the new species-specific EUCAST breakpoint for A. xylosoxidans, little to no susceptibility was observed for trimethoprim-sulfamethoxazole. Among the isolates, A. xylosoxidans was the most frequently isolated species. In vitro susceptibility varied among different Achromobacter species. These findings provide new insights into therapeutic options for the treatment of Achromobacter infections in CF patients.