Achromobacter xylosoxidans bacteremia: clinical and microbiological features in a 10-year case series.

OBJECTIVE: The treatment of Achromobacter xylosoxidans bacteremia is challenged by antimicrobial resistance and the paucity of data. We aimed at offering a contemporary description of this uncommon entity.
METHODS: Retrospective case series of 13 episodes of A. xylosoxidans bacteremia diagnosed over a 10-year period (November 2007 to May 2017) in our tertiary care center.
RESULTS: Solid organ cancer and heart failure were the most common comorbidities (4/13 [30.7%]). All but one episodes were hospital-acquired. Most patients had received previous antibiotic therapy (7/13 [53.8%]) and had a central venous catheter in place (6/13 [46.1%]). Primary and intravascular catheter were the most common sources (4/13 [30.7%] each). Meropenem was the agent with best in vitro activity (92.3% [12/13] of susceptible isolates). All-cause 30-day mortality (overall 23.1%) was higher in patients with primary bacteremia (50.0% vs. 11.1%; P-value=0.203) and prior chemotherapy (66.7% vs. 10.0%; P-value=0.108).
CONCLUSIONS: Bacteremia due to A. xylosoxidans constitutes a serious infection among immunocompromised hosts. Carbapenem-based therapy may be appropriate in most cases. ©The Author 2018. Published by Sociedad Española de Quimioterapia. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)(https://creativecommons.org/licenses/by-nc/4.0/).

INTRODUCTION

Achromobacter xylosoxidans is an aerobic, motile, oxidase-positive, non-fermenting Gram-negative rod widely distributed in the environment that mainly causes healthcare-associated infection [1,2]. Previously termed Alcaligenes xylosoxidans, the name A. xylosoxidans (which had been originally proposed in 1971 by Yabuuchi and Ohyama) was formalized since 1998 and now encompasses two different subspecies (A. xylosoxidans subsp. xylosoxidans and A. xylosoxidans subsp. denitrificans) [3,4]. Most cases in the literature are described in patients with some form of immunosuppression, usually haematological malignancies. Primary bacteraemia represents the most common clinical presentation [5-8]. Other reported infection sites of infection include meningitis, urinary tract infection, abscesses, osteomyelitis, corneal ulcers, prosthetic valve endocarditis, peritonitis and pneumonia [2,9-12].

The treatment of patients with A. xylosoxidans bacteraemia is challenging due to the fact that this microorganism carries both intrinsic and acquired mechanisms of resistance, often conferring a phenotype of multidrug resistance (MDR) [13,14]. In addition, current data on this uncommon entity is mostly limited to very small series or single case reports. With the aim of gaining further insight into the clinical characteristics and optimal therapeutic approach of A. xylosoxidans bacteraemia, we present a retrospective review of 13 cases occurring in our institution over a 10-year period.

PATIENTS AND METHODS

We retrospectively analysed all consecutive episodes of bacteraemia due to A. xylosoxidans diagnosed between November 2007 and May 2017 at the University Hospital “12 de Octubre” (Madrid, Spain), a 1,300-bed tertiary care centre with a reference population of 431,325 inhabitants in 2015. It has five different adult intensive care units (medical, trauma, coronary, general surgery, and cardiac surgery) and an active program of solid organ and hematopoietic stem-cell transplantation. No age restriction was applied. Study protocol was approved by the local Clinical Ethics Committee (number 17014). Informed consent was waived due to the retrospective nature of the research.

Cases were identified through a search of the computerized database of the Department of Microbiology and defined by the isolation of A. xylosoxidans in one or more blood cultures in the presence of clinical signs compatible with systemic infection. Blood samples were inoculated in the BacT/ALERT® FA aerobic and FN anaerobic bottles and incubated in an automated system (BacT/ALERT 3D, Biomerieux, France) for 5 days. Bacterial identification and antimicrobial susceptibility testing were performed by the Wider® system (Soria Melguizo, Madrid, Spain), MicroScan WalkAway® system (Siemens Healthcare Diagnostics, Deerfield, IL, USA) and, since 2014, MALDI-TOF mass spectrometry (Bruker Daltonics, Bremen, Germany). Until 2014 isolates were categorized according to the Clinical Laboratory Standards Institute (CLSI) susceptibility breakpoints for non-fermenting gram-negative bacteria and, thereafter, the European Committee on Antimicrobial Susceptibility Testing (EUCAST) clinical breakpoints for Pseudomonas aeruginosa. The MDR phenotype was defined by demonstrated non-susceptibility to one agent belonging to ≥3 different antimicrobial categories [15].

Clinical charts were reviewed by means of a standardized case report form to obtain data on demographics, comorbidities, predisposing factors (i.e., surgical intervention within the previous 30 days, stay in the intensive care unit [ICU] or presence of intravascular catheters), use of systemic antibiotics within the previous 30 days, source of bacteraemia, severity of infection, antibiotic susceptibility of the isolate, antibiotics employed to treat the episode of bacteraemia, and 30-day all-cause mortality.

For the purposes of the present study, the diagnosis of intravascular catheter-related bacteraemia was considered when the criteria proposed by the Infectious Diseases Society of America were met [16] or in the presence of local signs of infection (erythema, swelling, tenderness or purulent discharge) at the site of insertion or along the subcutaneous tunnel tract and no alternative source. Other sources of bacteraemia such as pneumonia, urinary tract infection or skin and soft-tissue infection (SSTI) were considered in the presence of consistent clinical, radiological and laboratory data. The diagnosis of primary bacteraemia was reserved for cases with no identifiable source of infection. Severity of illness at time of bacteraemia onset was assessed according to the Pitt score [17]. Adequate therapy required the administration within the first 48 hours after the sampling of index blood cultures of the recommended dose of an antibiotic agent with in vitro activity according to the reported susceptibility testing.

Quantitative data were shown as the mean ± standard deviation or the median with interquartile ranges. Qualitative variables were expressed as absolute and relative frequencies. Categorical variables were compared using the Fisher’s exact test. All the significance tests were two-tailed. Statistical analysis was performed using SPSS, version 15.0 (Statistical Package for Social Sciences Inc., Chicago, IL).

RESULTS

We identified 13 patients with A. xylosoxidans bacteraemia whose demographics and clinical characteristics are detailed in table 1. The incidence rates across the years encompassed by the study period were overall very low (<0.2 cases per 1,000 hospital admissions or per 10,000 hospital patient-days) and no clear trend could be identified (figure 1). Solid organ malignancy and heart failure were the most common comorbidities (4 cases each [30.7%]). Most patients (7 [53.8%]) had received antibiotic therapy (mainly beta-lactams) within the previous 30 days and had a central venous catheter (CVC) in place at the onset of bacteraemia (6 [46.1%]). Recent surgery and chemotherapy were also common. Primary and catheter-related sources were equally represented (4 cases each [30.7%]). One patient developed bacteraemia during the first hours of hospital admission due to community-acquired SSTI, with no evidence of an alternative source. The remaining episodes occurred after 72 hours of hospitalization.

Table 1

Demographics and clinical characteristics of included patients (n = 13).

Variable
Age, years [mean ± SD]	52.1 ± 32
Gender (male) [n (%)]	5 (38.5)
Pitt score [median (IQR)]	2 (0-5.5)
Comorbidities [n (%)]
Solid organ malignancy	4 (30.7)
Heart failure	4 (30.7)
Hematologic malignancy	2 (15.3)
Diabetes mellitus	2 (15.3)
Chronic liver disease	3 (23.1)
Solid organ transplantation	1 (7.7)
Predisposing conditions [n (%)]
Prior systemic antibiotic therapy[a]	7 (53.8)
Recent surgery[a]	4 (30.7)
Chemotherapy	3 (23.1)
Recent ICU admission[a]	1 (7.7)
CVC in place at diagnosis[b]	6 (46.1)
Duration of CVC placement [median (IQR)]	17 (1-17)
Source of bacteraemia [n (%)]
Primary	4 (30.7)
Catheter-related	4 (30.7)
Intraabdominal infection	2 (15.3)
Pneumonia	1 (7.7)
Urinary tract infection	1 (7.7)
SSTI	1 (7.7)
Polimicrobial bacteraemia [n (%)][c]	2 (15.3)
Empirical antibiotic therapy [n (%)]
Monotherapyd	10 (76.9)
Combination therapy[d]	3 (23.1)
Regimen containing
Carbapenem	5 (38.5)
Third-generation cephalosporin	3 (23.1)
Beta-lactam/beta-lactamase inhibitor	2 (15.3)
Aminoglycoside	2 (15.3)
Vancomycin	2 (15.3)
Others	2 (15.3)
30-day all-cause mortality [n (%)]	3 (23.1)

CVC: central venous catheter; ICU: intensive care unit; SSTI: skin and soft-tissue infection.

Within the previous 30 days.

Includes non-tunneled temporary catheters (n = 3), skin-tunneled catheters (n = 2) and peripherally inserted central catheters (n = 1).

Concomitant bacteraemia by Bacillus spp. and Streptococcus sanguinis.

Combination therapy included meropenem plus vancomycin, ceftriaxone plus amikacin, and gentamicin plus aztreonam.

Figure 1

Incidence rates of Achromobacter xylosoxidans at our centre across the 10 years encompassed by the study period.

The susceptibility profile of A. xylosoxidans isolates is shown in table 2. Meropenem was the agent with the best in vitro activity (92.3% [12/13] of isolates were susceptible), followed by imipenem and piperacillin/tazobactam (76.9% [10/13] of isolates). On the other hand, resistance to aminoglycosides was common (69.2% [9/13] for gentamicin and 38.4% [5/13] for amikacin). The criteria for MDR phenotype [15] were fulfilled in 38.4% (5/13) of isolates.

Table 2

Results of in vitro susceptibility testing of A. xylosoxidans isolates (n = 13) and clinical breakpoints used for categorization.

				Clinical breakpoints (mg/L)[a]
Agent	Susceptible (%)	Intermediate (%)	Resistant (%)	Until 2014[b]	2014 onwards[b]
Amikacin	61.5	0.0	38.4	S: ≤16, R: >32	S: ≤8, R: >16
Aztreonam	0.0	0.0	100.0	S: ≤8, R: >16	S: ≤1, R: >16
Cefepime	46.1	0.0	53.8	S: ≤8, R: >16	S: ≤8, R: >8
Ceftazidime	69.2	0.0	30.7	S: ≤8, R: >16	S: ≤8, R: >8
Ceftriaxone	28.5	0.0	71.4	S: ≤8, R: >16	S: ≤8, R: >8
Ciprofloxacin	36.3	27.2	36.3	S: ≤1, R: >2	S: ≤0.5, R: >0.5
Fosfomycin	0.0	20.0	80.0	NA	S: ≤32, R: >32
Gentamicin	30.7	0.0	69.2	S: ≤4, R: >8	S: ≤4, R: >4
Imipenem	76.9	7.6	15.3	S: ≤2, R: >4	S: ≤4, R: >8
Meropenem	92.3	0.0	7.6	S: ≤2, R: >4	S: ≤2, R: >8
Piperacillin/tazobactam	76.9	7.6	15.3	S: ≤16/4, R: >64/4	S: ≤16/4, R: >16/4
Trimethoprim/sulfamethoxazole	53.8	0.0	43.1	S: ≤2/38, R: >4/76	NA

NA: not available; R: resistant; S: susceptible.

The intermediate should be interpreted as values between the S and the R breakpoints.

Clinical Laboratory Standards Institute (CLSI) susceptibility breakpoints for non-fermenting gram-negative bacteria and European Committee on Antimicrobial Susceptibility Testing (EUCAST) clinical breakpoints for Pseudomonas aeruginosa were applied until 2014 and thereafter, respectively.

Empirical antibiotic therapy was mostly based on carbapenems (5 cases [38.5%]), whereas combination therapy was used in 3 patients (23.1%). In most cases (8 [61.5%]) empirical therapy was adequate according to in vitro susceptibility tests and did not require further modification. None of the patients had persistent bacteraemia. The all-cause 30-day mortality rate was 23.1%, with the 3 observed deaths deemed attributable to infection due to multiple organ failure (table 1). Patients with primary bacteraemia had higher 30-day mortality than those with other sources (2/4 [50.0%] versus 1/9 [11.1%]; P-value = 0.203), as well as those who had received prior chemotherapy (2/3 [66.7%] versus 1/10 [10.0%]; P-value = 0.108), although such differences were not statistically significant. There were no apparent differences in mortality according to the appropriateness or not of empirical therapy (2/8 [25.0%] versus 1/4 [25.0%], respectively; P-value = 1.000).

DISCUSSION

In our single-centre experience, approximately half of the cases of A. xylosoxidans bacteraemia occurred in patients with underlying malignancies that had received previous antibiotic therapy. Catheter-related bacteraemia was predominant among those episodes with a documented source of infection, whereas patients with primary bacteraemia (which accounted for one third of the overall cohort) had higher 30-day mortality, as well as those who had received prior chemotherapy. For both associations, nevertheless, it is likely that the low sample size precluded statistical significance.

In the few existing reports on this entity, A. xylosoxidans bacteraemia has been mainly observed in patients with malignant diseases, especially haematological. In one of the largest series published to date, Aisenberg et al. included 52 episodes diagnosed during a 14-year period in 46 patients with cancer [5]. In a previous Spanish single-centre study, 39% of episodes of A. xylosoxidans bacteraemia occurred in patients with cancer, whereas prior chemotherapy and neutropenia were also common (35% and 30% of cases, respectively) [6]. In addition, we found that more than half of the patients had received antibiotic therapy within the previous 30 days. This predisposing factor has been also described by Ozden et al [18] in the setting of a neonatal outbreak, in which 100% of patients had been treated with broad-spectrum antibiotics.

In line with some previous reports [5-7], catheter-related infection was the main source of bacteraemia in the present cohort. Infection due to A. xylosoxidans is usually regarded as healthcare-associated, as supported by our experience (in which 92.3% of cases were hospital-acquired). However, we also observed a case of bacteraemia associated to a community-acquired SSTI, which is very infrequent. There is only one previous report in the literature of community-acquired SSTI due to A. xylosoxidans [19], although the microorganism was not isolated in blood samples in any of the 9 cases.

Most of our isolates showed in vitro sensitivity to carbapenems and piperacillin/tazobactam, in agreement with other authors [18,20]. However, we also found a notable rate of resistance to aminoglycosides, ciprofloxacin, ceftazidime or trimethoprim/sulfamethoxazole, to which A. xylosoxidans has been previously reported to remain sensitive [5,8]. Such information is important when making therapeutic decisions, especially among immunocompromised and critically ill patients, since the emergence of MDR strains has been reported [21]. In that sense, it should be highlighted that almost 40% of the isolates in the present study exhibited MDR phenotypes.

All cases of death at 30 days from diagnosis in our series were ultimately related to infection. This figure yielded a 30-day attributable mortality (23.1%) slightly higher than that previously reported, which ranged from 0.0% to about 15.0% [5-7]. In a case series restricted to patients with cancer, the presence of sepsis and high APACHE II scores at infection onset were independent predictors of 30-day mortality [5]. We were unable to identify significant differences in outcome across different subgroups of patients, although those who had primary bacteraemia and previous chemotherapy suffered from higher mortality. This possible association has not been previously described, although the number of patients analysed was too small to draw definitive conclusions.

In conclusion, bacteraemia due to A. xylosoxidans can be a serious complication among hospitalized patients, mainly those with cancer-related immunosuppression. Empirical antibiotic therapy including a carbapenem agent appears to be appropriate in most cases, although further adjustment should be performed according to the result of in vitro susceptibility testing. The potential impact on the outcome of different sources of infection, underlying conditions and appropriateness of initial therapy should be evaluated in future studies on this uncommon entity.