During the past one to two decades, the clinical relevance of various catalase-negative, Gram-positive cocci has been increasingly pointed out. Among these, Aerococcus urinae and Aerococcus sanguinicola are relatively newcomers and emerging organisms in clinical and microbiological practice [1-3]. Both species have worldwide been associated with urinary tract infections especially in elderly predisposed patients. A prevalence of A. urinae isolates in up to 4% of urine specimens examined has been reported [4]. Both species have also been isolated from blood from patients suffering from urogenic bacteremia/septicemia, in few cases with complicating infective endocarditis. Isolates of both species have furthermore casuistically been isolated from other foci too [1].

Isolates from both species are characterized by Staphylococcus-like morphology in Gram stains with growth characteristics resembling α-hemolytic streptococci on blood containing agar plates, a negative catalase reaction, a consistent antibiogram demonstrating susceptibility to β-lactams, and for A. urinae resistance to sulfonamides and aminoglycosides [5]. Recognition of the species may be harmed by their fastidious growth, which not seldom require supplementation with CO₂ for optimal growth [5]. Diagnosing infections caused by both organisms are still missed in many laboratories around the world. However, with the increasing use of MALDI-TOF mass spectrometry for species identification, both species are diagnosed with increasing frequency [6].

Data on in vitro susceptibility of A. urinae isolates are accumulating [3, 4, 7-10]. The number of susceptibility reports on A. sanguinicola are relatively sparse compared to A. urinae [7, 11-14]. A lack of standardized susceptibility test methods and interpretive criteria for Aerococcus spp. have been problematic for clinical laboratories and clinicians [9]. Most publications have used the interpretive criteria for viridans group streptococci when evaluating the susceptibility. In general, both species exhibit susceptibility to penicillin and vancomycin and low level resistance towards aminoglycosides [10]. A. urinae is susceptible to nitrofurantoin [13].

A variety of test methods have been reported, and interpretive criteria for streptococci, staphylococci, and even enterococci have been applied [9]. In order to obtain comparable results it is desirable to develop a standardized methodology and interpretation when performing susceptibility testing on especially these two Aerococcus species. In the last revision from August 2016 of the US guideline from the Clinical and Laboratory Standards Institute (CLSI) on “Methods for antimicrobial dilution and disk susceptibility testing of infrequently isolated or fastidious bacteria” (M45) [15] a table on aerococci has been included.

In this study, 120 aerococcal isolates, A. urinae (n = 81) and A. sanguinicola (n = 39), of European origin were tested for susceptibility to six antimicrobial agents using the methodology recommended by The European Committee on Antimicrobial Susceptibility Testing (EUCAST). Data obtained in this study acted as part of the decision background for creating European clinical breakpoints for A. urinae and A. sanguinicola [16].

On MH-F agar, confluent lawn of growth was observed after 16-20 h incubation for A. urinae and A. sanguinicola isolates. By disk diffusion, Etest, and BMD determinations, 14 of the A. urinae isolates had weak growth after 16-20 h incubation and needed prolonged incubation (40-44 h).

Antimicrobial susceptibility test results (disk diffusion zone diameters, MIC₅₀, MIC₉₀, and MIC ranges) with the three used methods are given in (Tables 1 and 2), for A. urinae and A. sanguinicola isolates, respectively. Zone histograms of disk diffusion against corresponding BMD MICs are visualized using different colors for each MIC in the supplementary material. Normal MIC distributions were obtained for both methods (data not shown). Using the EUCAST recommendations, Etest and BMD MIC values were obtained for 76-81 A. urinae strains and for 38-39 A. sanguinicola strains.

For the wild-type population of A. urinae isolates, the range of zone diameters differed among examined antimicrobial agents from 6 mm (vancomycin) to 14 mm (cefotaxime). For A. sanguinicola, wild-type populations ranged from 5 mm (vancomycin) to 11 mm (meropenem).

CLSI (August 2016) and EUCAST (January 2017) have currently established species-specific breakpoints for A. urinae and A. sanguinicola and before the availability of such data, MICs needed to be interpreted using breakpoints established for related bacteria as viridans group streptococci.

The EUCAST standardized disk diffusion method for fastidious microorganisms uses Mueller-Hinton agar supplemented with 5% defibrinated horse blood and 20 mg/L β-NAD (MH-F), and a standard inoculum of McFarland 0.5 for obtaining sufficient demarcating growth following 16-20 h incubation in ambient air supplemented with 5% CO₂. In a study by Humphries & Hindler [9] they demonstrated good growth of 128 A. urinae isolates when using cation-adjusted Mueller-Hinton (MH) broth (CAMHB) supplemented with 2.5% to 5% lysed horse blood (LHB), which is the CLSI recommended standard medium (CAMHB-LHB). In the present study, growth of A. urinae isolates were weaker as compared to A. sanguinicola isolates and prolonged incubation were needed for 14 of the A. urinae isolates to obtain readable growth.

In Table 3 the recommended MIC breakpoints by EUCAST and CLSI in addition to EUCAST recommended zone diameter breakpoints are given for all of the examined antibiotics in this study. The EUCAST recommendations cover A. urinae and A. sanguinicola and the CLSI recommendations additionally include Aerococcus viridans. The inhibition zone diameters and MICs obtained were within CLSI and EUCAST recommendations for being susceptible, except for one A. urinae strain using cefotaxime [15, 16].

Broth microdilution and Etest MICs from this study corresponded to MIC values using BMD, agar dilution, and gradient tests as reported in a review from Rasmussen M [1]. Rasmussen M [1] summarized published antibiotic susceptibility testing results of A. urinae and A. sanguinicola isolates using EUCAST breakpoints for viridans group streptococci for 13 antibiotics for between 58 to 342 A. urinae isolates and between 8 to 65 A. sanguinicola isolates. The antibiotic susceptibilities varied slightly between the two species. Both Aerococcus species were susceptible to β-lactam antibiotics with modal MICs for penicillin based on 308 A. urinae and 65 A. sanguinicola isolates in the MIC range of 0.03-0.06 mg/L [1]. In the present study, penicillin susceptibilities were in agreement with these findings. Penicillin MICs for A. sanguinicola isolates were in general one dilution higher than MICs for A. urinae. When comparing MICs and zone diameters, all isolates were susceptible using the EUCAST breakpoints. The MICs for cephalosporins, such as cefotaxime and ceftriaxone, have been reported close to the upper limit for susceptibility, but still often within the range of being susceptible, using the CLSI breakpoints for streptococci [1]. In the study of Humphries & Hindler [9] which was based on 128 A. urinae isolates, all but one isolate tested were susceptible to meropenem (MIC ≤ 0.5 mg/L), with a modal MIC of ≤ 0.015 mg/L. The sole meropenem-nonsusceptible isolate was also reproducibly reduced susceptible to ceftriaxone (4 mg/L) and cefotaxime (2 mg/L), but susceptible to penicillin (0.06 mg/L). One of the A. urinae isolates in this study demonstrated an Etest MIC value one dilution above the CLSI breakpoint for cefotaxime, thereby being intermediate susceptible, but zone diameter and BMD (Sensititre) MIC determinations indicated susceptibility. For meropenem, MIC ranges were observed as ≤ 0.25 mg/L for both Aerococcus species.

Both species being capable of causing urinary tract infections, septicemia, and though seldom infective endocarditis makes it favorable that isolates exhibited no resistance to penicillin. Two earlier studies have additionally indicated that there is a synergistic killing effect when combining penicillin and an aminoglycoside [10, 20], though a new larger study could only demonstrate such synergy in a minority of isolates [2]. This favorable susceptibility pattern is important as both organisms essentially are uropathogens, that may be resistant to more of the otherwise empirically used antibiotics for treating urinary tract infections (nalidixic acid, trimethoprim, sulfonamides, and mecillinam), though, the degree of resistance and methods for their detection are being debated [5, 7, 9].

Aerococci, as other Gram-positive cocci, except the inborn resistant pediococci, have modal MICs for vancomycin of 1 mg/L or below. This has been reported for at least 308 isolates of A. urinae and 23 isolates of A. sanguinicola [1] and makes vancomycin treatment of invasive infections as septicemia and infective endocarditis an attractive alternative in the penicillin allergic patient, eventually in combination with gentamicin. Except for two A. urinae isolates (Etest MIC 2 mg/L), results from the current study demonstrated vancomycin MICs for A. urinae and A. sanguinicola isolates ≤ 1 mg/L, which were in line with vancomycin MICs as reviewed by Rasmussen M [1] and susceptible according to EUCAST and CLSI breakpoints.

Our isolates of both species were susceptible to linezolid using CLSI breakpoint and susceptible to rifampicin (Etest testing only) using EUCAST breakpoints. Humphries & Hindler [9] also found isolates of both species linezolid susceptible, except two isolates with linezolid MICs of 4 and 8 mg/L. Sierra-Hoffman et al. [21, 22] examined 56 A. urinae isolates and Facklam et al. [11] 15 isolates of A. sanguinicola that were found susceptible to rifampicin, an antibiotic with a possible role in combination antibiotic therapy in severe infective endocarditis cases.

Some limitations of the study are the number of strains that were included in the study and the limited geographic distribution by the clinical strains only originating from Denmark, Switzerland, Spain, and the CCUG culture collection of Gothenburg (Sweden). Not all of the clinical strains were having sufficient growth by following the EUCAST and CLSI recommended incubation conditions, whereof no MIC values were stated for some of the strains. Another limitation by using a commercial BMD system is that only manufacturer selected antibiotic concentrations were tested and a MIC reading at the lowest MIC concentration can only identify the lowest value without identification of an exact MIC value. In a daily usage, the commercial BMD system is easy to use, reduces the hands-on time during preparation of the plates and less time consuming.

In conclusion, A. urinae and A. sanguinicola are emerging pathogens, primarily uropathogens, but have also been etiologic agents of bacteremia/septicemia and infective endocarditis; the latter though seldom well documented for A. urinae. Suggestions on breakpoints for most relevant antibiotics were just recently released by CLSI and EUCAST. The data obtained in this study has been integrated as part of the background data for the EUCAST clinical breakpoint recommendations.

Data for six antimicrobial agents using disk diffusion, Etest, and BMD were obtained for 120 A. urinae and A. sanguinicola isolates. MIC values obtained by Etest and BMD were in accordance within ± one dilution. All isolates were found susceptible to the six antimicrobial agents of penicillin, cefotaxime, meropenem, vancomycin, linezolid, and rifampicin.