Overview of Changes to the Clinical and Laboratory Standards Institute Performance Standards for Antimicrobial Susceptibility Testing, M100, 31st Edition

ABSTRACT

The Clinical and Laboratory Standards Institute (CLSI) Subcommittee on Antimicrobial Susceptibility Testing (AST) develops and publishes standards and guidelines for AST methods and results interpretation in an annual update to the Performance Standards for Antimicrobial Susceptibility Testing (M100). This minireview will discuss changes to M100 for the 31st edition, including new and revised breakpoints and testing recommendations. New MIC and disk diffusion breakpoints are described for azithromycin (Shigella spp.), imipenem-relebactam (Enterobacterales, Pseudomonas aeruginosa, and anaerobes), and lefamulin (Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae), and disk breakpoints are described for azithromycin and Neisseria gonorrhoeae. The rationale behind revised oxacillin MIC breakpoints for select staphylococci is discussed. Updates to test methods include a method for disk diffusion using positive blood culture broth and use of linezolid to predict tedizolid susceptibility. There is clarification on which drugs to suppress on bacteria isolated from the cerebrospinal fluid and clarification on the use of a caret symbol attached to the intermediate category (“I^”) to indicate those antimicrobials that concentrate in the urine.

INTRODUCTION

The Clinical and Laboratory Standards Institute (CLSI) Subcommittee on Antimicrobial Susceptibility Testing (AST) is a volunteer-led, multidisciplinary consensus body that develops and publishes standards and guidelines for AST methods and results interpretation. A key activity of the CLSI AST subcommittee is educational activities, coordinated by the Outreach Working Group (ORWG). This article is one such activity and was created on behalf of the ORWG. Weinstein and Lewis recently reviewed in detail the process that guides preparation, acceptance, and publication of CLSI AST standards (1). In an open forum, volunteers from the health care professions, government, and industry present suggestions, supported by data, for ways to improve the standards. When sufficient information has been obtained, a vote is taken to determine if the suggestion will be incorporated into the standard, based on the CLSI consensus process (1). The work of the AST Subcommittee is published in an annual update to the Performance Standards for Antimicrobial Susceptibility Testing (also known as M100). A no-fee read-only version of the 31st edition of M100 is available online at http://em100.edaptivedocs.net/dashboard.aspx, and a digital version for download or hard copy version of M100 and other CLSI documents can be ordered from clsi.org. This minireview will address the major changes in the 31st edition of the M100 document, published in March 2021 (2). At the beginning of each M100 is a listing of the “Overview of Changes” that occurred since the last edition of the document, including minor changes (e.g., formatting) that will not be covered in this review.

M100 is a companion document to the following CLSI standards that describe AST methods: M02, Performance Standards for Antimicrobial Disk Susceptibility Tests (3); M07, Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically (4); and M11, Methods for Antimicrobial Susceptibility Testing of Anaerobic Bacteria (5). M100 contains breakpoints to interpret results of disk diffusion (DD) and MIC tests and quality control (QC) ranges for these tests, along with guidance on which antimicrobial agents to test and report for specific organisms, methods by which to identify resistance mechanisms, how to confirm unusual results, and more (2).

BREAKPOINTS

The M100 31st edition includes several new or revised breakpoints, which are presented in Table 1. Three primary data elements are analyzed (6), as available, for each breakpoint decision: MIC distribution data, pharmacokinetic and pharmacodynamic (PK/PD) data, and clinical outcome data.

TABLE 1.

New and updated breakpoints in the CLSI M100 31st edition

Antimicrobial agent and organism/organism group	Disk content (μg)	Interpretive category and breakpoint bya:						FDA breakpoint(s)b
		Zone diam (mm)			MIC (μg/ml)
		S	I	R	S	I	R
Azithromycin
Shigella spp.	15	≥16	11–15	≤10	≤8	16	≥32	No breakpoints recognized
Neisseria gonorrhoeae	15	≥30			≤1c			No breakpoints recognized
Ceftolozane-tazobactam
Haemophilus influenzae					≤0.5/4			CLSI
Imipenem-relebactam
Enterobacterales	10/25	≥25	21–24^	≤20	≤1/4	2/4^	≥4/4	CLSI
Pseudomonas aeruginosa	10/25	≥23	20–22^	≤19	≤2/4	4/4^	≥8/4	CLSI
Acinetobacter calcoaceticus-A. baumannii complex								≤2/4 for S, 4/4 for I, and ≥8/4 μg/ml for R
Anaerobes	10/25				≤4/4	8/4	≥16/4	CLSI
Lefamulin
Staphylococcus aureus	20	≥23			≤0.25			CLSI
Haemophilus influenzae	20	≥17			≤2			CLSI
Streptococcus pneumoniae	20	≥17			≤0.5			CLSI
Oxacillind
Staphylococcus spp., except S. aureus  and S. lugdunensis					≤0.5		≥1	Not recognized

^aS, susceptible; I, intermediate; R, resistant; ^, concentrates in urine.

^bhttps://www.fda.gov/drugs/development-resources/antibacterial-susceptibility-test-interpretive-criteria (accessed 16 July 2021).

^cThis breakpoint was previously established and is not new for 2021.

^dOnly updated breakpoints are listed; see the M100 31st edition for oxacillin and cefoxitin disk diffusion breakpoints that were unchanged.

Azithromycin and Shigella spp.

Azithromycin is a first-line agent for the treatment of shigellosis (7, 8). Prior to 2021, clinical breakpoints for Shigella had not been established, but epidemiological cutoff values (ECVs), which define isolates without acquired or mutational resistance mechanisms, were available for Shigella flexneri and Shigella sonnei (≤8 μg/ml and ≤16 μg/ml, respectively). During 2011 to 2018, the overall percentages of Shigella isolates with azithromycin MICs above the ECV recorded by the U.S. National Antimicrobial Resistance Monitoring System (NARMS) increased from 12% to 53% among S. flexneri isolates and from 0.9% to 25% among S. sonnei isolates (wwwn.cdc.gov/narmsnow). NARMS data show concurrent increases in the percentages of Shigella spp. resistant to ampicillin, trimethoprim-sulfamethoxazole, and fluoroquinolones, limiting treatment options and highlighting the need for azithromycin clinical breakpoints for Shigella spp.

Azithromycin is an azolide antimicrobial agent that binds to the 50S subunit of the bacterial ribosome, leading to inhibition of bacterial protein synthesis. In general, Enterobacterales are intrinsically resistant to the majority of macrolides, as these antimicrobials poorly penetrate the Enterobacterales membrane. However, azithromycin harbors heightened activity compared to other members in the class, due to an increased basic character, favoring higher intracellular uptake. For this reason, unlike for Gram-positive bacteria, azithromycin results cannot be used to infer susceptibility to other macrolides or clindamycin. In Shigella, resistance mechanisms include acquisition of ermB, which encodes a methylase that modifies the target site, or mphA, a phosphotransferase that inactivates azithromycin. Azithromycin MICs of Shigella isolates harboring these resistance mechanisms typically measure at >16 μg/ml. Studies conducted by the U.S. Centers for Disease Control and Prevention (CDC) showed good correlation between MICs measured by broth microdilution (BMD) and the presence or absence of these resistance mechanisms.

New azithromycin breakpoints for Shigella were published in the M100 31st edition. These were based predominantly on new data from an observational cohort study of patients in Bangladesh with invasive diarrhea treated with 5 days of azithromycin (500 mg per day for adults) (9). A total of 149 patients were culture positive for Shigella and completed azithromycin therapy, including 84 infected with S. flexneri, 56 with S. sonnei, 4 with S. boydii, 3 with S. dysenteriae, and 2 whose infections were not identified to the species level. Azithromycin MICs of ≥32 μg/ml for S. sonnei and ≥16 μg/ml for S. flexneri and other Shigella spp. were associated with persistent diarrhea at day 5, stool culture positivity at day 5 or 6, hospitalization, and longer duration of diarrhea. Based on these data, a susceptible breakpoint of ≤8 μg/ml, an intermediate breakpoint of 16 μg/ml, and a resistant breakpoint of ≥32 μg/ml were selected for azithromycin. The intermediate category was established to allow for a buffer between susceptible and resistant categories, as difficult to interpret trailing MIC endpoints were noted, particularly with S. sonnei, when tested by the reference broth microdilution method (BMD). Although the ECVs for S. flexneri and S. sonnei differ by 1 doubling dilution, as noted above, a genus-level breakpoint was recommended by CLSI, as many clinical laboratories do not identify Shigella to the species level. Additionally, while the vast majority of infections were caused by S. flexneri and S. sonnei, available data for S. boydii and S. dysenteriae align with the breakpoints. Data provided by the CDC were used to establish a DD correlate to the MIC breakpoints (Table 1). Similar to BMD, azithromycin DD zones are difficult to read for Shigella, with hazy zones and double zones of inhibition surrounding the azithromycin disk noted on some brands of Mueller-Hinton agar. If an isolate has a zone of inhibition that is difficult to measure, it is recommended to perform an MIC test. To date, no commercial AST methods other than disk diffusion have been evaluated for Shigella spp.

Azithromycin DD breakpoints for Neisseria gonorrhoeae.

In 2019, a susceptible-only azithromycin MIC breakpoint of ≤1 μg/ml was published in the M100 29th edition for Neisseria gonorrhoeae (10). At that time, when disk-to-MIC correlate data were reviewed, 3/31 (9.7%) nonsusceptible isolates harbored disk zones of ≥29 mm, the proposed susceptible disk breakpoint. All three isolates had measured azithromycin MICs of 2 μg/ml. As this very major error rate (VME [i.e., susceptible by disk but resistant by MIC]) was above the CLSI’s threshold, the disk breakpoint was deferred while awaiting more data. Disk diffusion data for an additional 23 isolates with MICs in the 0.5- to 2.0-μg/ml range were evaluated, which showed good correlation with the MIC, with no VMEs and 1 (<1%) major error (ME [i.e., resistant by disk but susceptible by MIC]). A susceptible-only disk diffusion breakpoint of ≥30 mm was thus published in the M100 31st edition. Of note, while the current CDC guidelines no longer recommend use of azithromycin combination therapy for uncomplicated gonococcal infections, azithromycin remains a treatment option, in combination with gentamicin, for patients with cephalosporin allergies, and AST would be indicated for these cases (11).

Imipenem-relebactam.

Imipenem-relebactam (IMR [Recarbrio; Merck]) is a β-lactam combination agent consisting of the carbapenem imipenem-cilastatin (imipenem) and the β-lactamase inhibitor relebactam. It is given as an intravenous (i.v.) dose of 1.25 g (500 mg imipenem, 500 mg cilastatin, and 250 mg relebactam) every 6 h, as a 30-min infusion, for 4 to 14 days. Relebactam inhibits class A and class C β-lactamases but not class B or D β-lactamases. IMR has shown good activity against carbapenem-resistant Enterobacterales, the primary mechanism for which is Klebsiella pneumoniae carbapenemase (KPC), and carbapenem-resistant Pseudomonas aeruginosa, the primary mechanism for which is AmpC β-lactamase plus OprD porin loss (12). IMR provides anaerobic coverage due to the activity of imipenem (13). IMR is not more active than imipenem against Acinetobacter spp. or Stenotrophomonas maltophilia, as carbapenem resistance in these species is generally due to class D and class B β-lactamases, respectively. Furthermore, the target penicillin-binding proteins (PBPs) of the Morganellaceae (i.e., Proteus, Providencia, and Morganella spp.) have poor affinity for imipenem, resulting in intrinsic, low-level resistance to imipenem among these genera. This PBP resistance mechanism is not overcome by the addition of relebactam, and as such, the clinical breakpoints for the Enterobacterales do not apply to Morganellaceae (12). Isolates that are susceptible to imipenem can be presumed to be susceptible to IMR.

The ratio of the area under the concentration curve (AUC) to MIC is the PK/PD index that best correlates with antibacterial activity of relebactam, which was demonstrated by both mouse thigh infection models and in vitro hollow fiber models. Two-log killing of P. aeruginosa was achieved for relebactam combined with imipenem (at 2× the humanized dose) at free AUC (fAUC)/MIC ratios of 7 in a murine thigh model and 7.5 in the hollow fiber models (14). Monte Carlo simulations demonstrated >90% probability of target attainment for imipenem (target, 30% of time above MIC) and relebactam (target, fAUC/MIC ratio of 7.5), with a susceptible breakpoint of ≤2 μg/ml for P. aeruginosa, across creatinine clearances of <15 to 250 ml/min (15).

IMR was approved by the United States Food and Drug Administration (FDA) in July 2019 for the treatment of adults with complicated intra-abdominal infections (cIAI) and complicated urinary infections (cUTI), including pyelonephritis, caused by susceptible aerobic and anaerobic Gram-negative bacilli. Two phase 2 clinical trials determined coadministration of relebactam (at 125 mg or 250 mg) with 500 mg imipenem was noninferior to treatment with 500 mg imipenem alone in adults ≥18 years of age with cUTI (16) or cIAI (17). These trials demonstrated no differences in microbiological response at discontinuation of i.v. therapy for patients with cUTI and no differences in clinical response at discontinuation of i.v. therapy for patients with cIAI. As the 250-mg relebactam dose was well tolerated, it was selected for the phase 3 trials RESTORE IMI-1 and -2. RESTORE IMI-1 was a randomized, double-blind comparator-controlled study that evaluated the safety and efficacy of IMR versus colistin plus imipenem in patients ≥18 years of age with imipenem not-susceptible (i.e., intermediate or resistant) isolates causing infections, including hospital-acquired bacterial pneumonia (HABP), ventilator-associated bacterial pneumonia (VABP), cUTI, or cIAI (18). Forty-seven patients were enrolled, including 31 in the IMR arm and 16 in the comparator arm. Favorable overall response was defined as survival through day 28 postrandomization for patients with HABP/VABP, clinical response at day 28 postrandomization for cIAI, and composite clinical and microbiological response at days 5 to 9 post-end of therapy for subjects with cUTI. This endpoint was achieved for 71.4% of patients in the IMR arm and 70.0% in the colistin plus imipenem arm. All-cause mortality was 2/21 (9.5%) in the IMR arm and 3/10 (30%) in the colistin plus imipenem arm. RESTORE-IMI 2 enrolled patients with HABP or VABP and randomized to i.v. imipenem, cilastatin, and relebactam (500, 500, and 250 mg, respectively) or i.v. piperacillin-tazobactam (5 and 500 mg, respectively), which were dosed every 6 h for 7 to 14 days (19). In this study, the primary endpoint was day 28 all-cause mortality in the modified intention-to-treat population (i.e., patients receiving study therapy, excluding those with only Gram-positive cocci at baseline). Among the 537 randomized patients, the most common pathogens were K. pneumoniae (25.6%) and P. aeruginosa (18.9%). The rates of day 28 all-cause mortality were 15.9% in the IMR arm and 21.3% in the piperacillin-tazobactam arm (achieving the noninferiority endpoint; P < 0.001). Favorable clinical responses at early follow-up (days 7 to 14) were 61.0% and 55.8% for IMR and piperacillin-tazobactam, respectively. Based on these combined data, and in consideration of the desire to harmonize with existing FDA-approved breakpoints, susceptible breakpoints of ≤1/4 μg/ml, ≤2/4 μg/ml, and ≤4/4 μg/ml were selected for the Enterobacterales (excluding Morganellaceae), P. aeruginosa, and anaerobes, respectively. Intermediate and resistant breakpoints are shown in Table 1. Of note, these differ from the European Committee on Antimicrobial Susceptibility Testing (EUCAST) susceptible and resistant breakpoints, which are ≤2/4 μg/ml and >2/4 μg/ml, respectively, for the Enterobacterales, P. aeruginosa, and anaerobes in the v.11 EUCAST breakpoint tables. EUCAST additionally published MIC and disk diffusion breakpoints for IMR and Acinetobacter spp., and the FDA publishes MIC breakpoints for Acinetobacter calcoaceticus-A. baumannii complex, which are the same as those for P. aeruginosa. CLSI does not recognize IMR breakpoints for Acinetobacter spp.

Testing of IMR should be considered for carbapenem-resistant Enterobacterales, excluding those known to harbor a metallo-β-lactamase (e.g., NDM) or the Morganellaceae. In addition, testing may be considered for P. aeruginosa isolates that are resistant to imipenem. In all cases, if an isolate is known to be susceptible to imipenem, it is reliably susceptible to IMR. FDA-cleared testing options for IMR include Etest (bioMérieux, Durham, NC) and MTS (Liofilchem, Italy) gradient diffusion strips, DD (Hardy, Santa Ana, CA), and Sensititre broth microdilution panels (Thermo Fisher, Lenexa, KS). Aerobic isolates for verification of these methods are available from the CDC & FDA Antibiotic Resistance Isolate Bank.

Ceftolozane-tazobactam.

Ceftolozane-tazobactam breakpoints for Haemophilus influenzae were added to M100 31st edition, following new data from Merck on the outcomes of the ASPECT-NP phase 3 clinical trial of hospital-onset bacterial pneumonia (20). These breakpoints supplement existing ceftolozane-tazobactam breakpoints for Enterobacterales, P. aeruginosa, and viridans group Streptococcus spp., which were published in 2016 and following cUTI and cIAI studies. While not a cUTI or cIAI pathogen, ∼7% of enrolled patients in ASPECT-NP had H. influenzae isolated from respiratory cultures. More than 99% of both β-lactamase-positive and -negative H. influenzae isolates had MICs to ceftolozane-tazobactam of ≤0.5 μg/ml, which was set as the susceptible breakpoint (21). No resistant or intermediate breakpoint was established due to a lack of data; isolates with MICs of >0.5 μg/ml are rare. There are limited testing options for H. influenzae with ceftolozane-tazobactam, and CLSI does not presently recommend routine testing.

Lefamulin.

Lefamulin (Xenleta [pronounced zen-LET-a]; Nabriva Therapeutics) is the first systemic antibacterial agent of the pleuromutilin class approved for human use. The pleuromutilins are naturally occurring antibacterials originally isolated from the edible mushroom Pleurotus mutilis. Lefamulin was approved by the FDA in 2019 for treatment of adults with community-acquired bacterial pneumonia (CABP) caused by Streptococcus pneumoniae, Staphylococcus aureus, H. influenzae, Legionella pneumophila, Mycoplasma pneumoniae, and Chlamydophila pneumoniae. Lefamulin targets a wide range of bacterial pathogens, including Gram-positive, fastidious Gram-negative, and atypical bacterial respiratory pathogens. It exhibits both bactericidal and bacteriostatic activities, depending upon the organism, and achieves high intracellular concentrations in macrophages. It is available in i.v. (150 mg every 12 h for 5 to 7 days) or oral (600 mg every 12 h for 5 days) formulations.

Lefamulin has a novel mechanism of action. It inhibits bacterial protein synthesis by binding to the peptidyl transferase center (PTC) of the 50S bacterial ribosome at the A- and P-sites. The interaction of lefamulin at the PTC prevents tRNA from binding, thus preventing the peptide transfer necessary for bacterial growth. There is a low propensity for development of bacterial resistance in vitro or for cross-resistance with other antimicrobial classes due to this unique mechanism of action. Resistance may occur, however, as a result of protection of the ribosomal targets by ABC- to -F proteins, such as those encoded by the genes vga(A, B, or E) in staphylococci, lsa(E), in streptococci, and sal(A) in staphylococci other than S. aureus (22). Other acquired resistance mechanisms include mutations leading to modification of the target in or near the PTC. Lefamulin is not active against Enterobacterales, P. aeruginosa, A. baumannii, Enterococcus faecalis, and most anaerobes, including Bacteroides spp.

The phase 3 clinical trials Lefamulin Evaluation Against Pneumonia (LEAP)-1 and -2 for CABP treatment of adults demonstrated that lefamulin met noninferiority endpoints against comparator antibiotics (23, 24). LEAP-1 compared lefamulin i.v.-to-oral transition against moxifloxacin with or without linezolid, while LEAP-2 compared oral lefamulin against moxifloxacin. High response rates were seen in both arms of the pooled LEAP-1 and -2 data sets, with 89% (577/646) and 90.5% (582/643) of patients achieving early clinical response of the primary endpoint for lefamulin and moxifloxacin, respectively, in the intention-to-treat analysis data sets. Neutropenic mouse models and pharmacokinetic studies in humans further supported the proposed breakpoints. However, studies have shown that only 13% of lefamulin is excreted as unchanged antimicrobial in urine; therefore, CLSI determined that lefamulin should not be reported from urinary tract isolates as treatment may be suboptimal (25).

The FDA established MIC and DD breakpoints for lefamulin in 2019 for S. aureus (methicillin-susceptible isolates), S. pneumoniae, and H. influenzae (Table 1). Methicillin-resistant S. aureus (MRSA) was not given FDA breakpoints due to the lack of sufficient numbers of MRSA isolates in clinical trials. The CLSI AST Subcommittee subsequently evaluated lefamulin and set MIC breakpoints identical to those of the FDA for S. pneumoniae and H. influenzae, which are published in the 2021 M100 31st edition. CLSI MIC breakpoints were set for MRSA in addition to methicillin-susceptible S. aureus (MSSA) based on supportive in vivo PK/PD data using primarily MRSA isolates and the high success rates of MRSA isolates in the phase 2 clinical study for acute bacterial skin and skin structure infections (ABSSSI) (26). Additionally, the ECVs for 99.0% of isolates tested (ECV_99.0s) for MSSA and MRSA were identical at 0.25 μg/ml. DD breakpoints for S. pneumoniae, H. influenzae, and S. aureus set by CLSI in 2020 matched those of the FDA at the time, with the exception of MRSA isolates, for which FDA breakpoints did not exist.

Only susceptible breakpoints for lefamulin have been set, since nearly all isolates examined demonstrated MICs below the breakpoints. A resistant category was not justified due to lack of sufficient isolates above the susceptible breakpoint. A result for lefamulin other than susceptible should be classified as “nonsusceptible” and should be investigated as a possible testing error or new resistance mechanism.

Lefamulin may be considered for treatment of patients with CABP when patients are intolerant to β-lactams, macrolides, or fluoroquinolones (27) or when significant safety concerns discourage use of these more traditional agents. Lefamulin may also be considered if typical local susceptibility patterns show high resistance to agents commonly used for empirical therapy. Given the overall relatively poor yield of sputum culture for recovery of organisms causing CABP, treatment of CABP is often empirical (28). Lefamulin’s broad spectrum of activity allows for coverage of the most common bacterial causes of CABP. Global surveillance studies of pathogens commonly causing CABP demonstrate that >99.0% of isolates tested susceptible to lefamulin when applying FDA breakpoints (29).

Antimicrobial susceptibility testing methods for lefamulin include CLSI standard DD and reference BMD. FDA-cleared disks (20 μg) are currently available from Hardy Diagnostics. It should be noted that the EUCAST disk content for lefamulin differs from that of CLSI, thus contributing to some differences in DD breakpoints. FDA-cleared MIC methods include Sensititre BMD panels and MTS gradient diffusion strips. Currently, the CDC & FDA AR Isolate Bank does not offer a lefamulin verification panel.

METHODS AND TESTING CONSIDERATIONS

Noteworthy new or revised methods in the M100 31st edition include (i) a new standardized method for direct DD susceptibility testing from positive blood culture broth, (ii) revisions to the oxacillin breakpoints for Staphylococcus spp. other than Staphylococcus aureus or Staphylococcus lugdunensis, (iii) mention of two emerging members of the S. aureus complex, Staphylococcus argenteus and Staphylococcus schweitzeri, and (iv) use of results from testing linezolid for predicting susceptibility to tedizolid for several Gram-positive organisms.

Additional methods-related revisions in the M100 31st edition include updated guidance on antimicrobials that should not be reported on isolates recovered from cerebrospinal fluid, updates to guidance for using molecular assays for resistance detection, and clarification of the use of “^” in the intermediate interpretative category definition.

Direct disk method.

DD testing performed directly from positive blood culture broth is currently being assessed for select Gram-negative bacteria that grow aerobically. As an ongoing project, DD breakpoints are being evaluated for groups of organisms for overnight reads at 16 to 18 h of incubation and for earlier reads at 8 to 10 h of incubation. A large prospective multicenter study funded by the Antimicrobial Resistance Leadership Group (ARLG) was undertaken by CLSI to assess performance of these methods in a clinical setting using blood bottles that flagged positive for growth (unpublished data). In the M100 31st edition, the following antimicrobials were approved for overnight direct reads for Enterobacterales when applying the current DD breakpoints: ampicillin, aztreonam, ceftazidime, ceftriaxone, tobramycin, and trimethoprim-sulfamethoxazole. The AST Subcommittee continues to evaluate disk zone cutoffs for additional antimicrobials, additional groups of organisms, and the additional time point of 8 to 10 h of incubation. Suggestions for verification of this testing approach will be forthcoming.

A step-by-step description of methods for this testing approach is provided in Table 3E in the M100 31st edition. Laboratories should note that the DD tests must be set up within 8 h of the blood bottle flagging positive for growth in automated detection systems. Only bottles demonstrating Gram-negative bacilli on Gram stain should be tested by this method; if a mix of organisms is seen on Gram stain, this testing should not be performed. A purity plate must be included. An identification method for the organism in question should be used in parallel so that the organism’s identity can be confirmed at least to the genus level (for Enterobacterales breakpoints) or species (for forthcoming P. aeruginosa breakpoints) before DD breakpoints are applied and interpretive categories released. If the organism is intrinsically resistant to the antimicrobial in question (e.g., K. pneumoniae is intrinsically resistant to ampicillin), the isolate should be reported as resistant regardless of the zone size. Further details and guidance will be provided in future editions of M100 and M02 documents.

Testing considerations for Staphylococcus spp.

Since 2012, recommendations for use of phenotypic methods to detect oxacillin (methicillin) resistance among Staphylococcus spp. other than S. aureus/S. lugdunensis have been revised several times (30). Prior breakpoints were based on a limited data set of ∼10 isolates per species (31) and had good correlation with the presence of mecA for Staphylococcus epidermidis but overcalled resistance for non-S. epidermidis species to various degrees based on the species in question. Recent refinement has been undertaken by CLSI due to the availability of methods that can readily identify coagulase-negative staphylococci to the species level (e.g., matrix-assisted laser desorption ionization–time of flight mass spectrometry [MALDI-TOF MS]). In 2021, a revised oxacillin breakpoint was published for coagulase-negative Staphylococcus (Table 1), based on new studies that evaluated large collections of the most commonly isolated species of coagulase-negative Staphylococcus, other than S. epidermidis (i.e., S. capitis, S. haemolyticus, S. hominis, and S. warneri) (30). In these evaluations, testing for mecA and/or PBP2a was used as a reference to evaluate cefoxitin MIC, cefoxitin DD, oxacillin MIC, and oxacillin DD methods. When adjusting the oxacillin-resistant breakpoint to ≥1 μg/ml and the susceptible breakpoint to ≤0.5 μg/ml, ME rates were reduced from 7.1% to 0.3%, while VME rates were only modestly altered from 2.1% to 2.8%. When data for these four species were combined with data generated for S. epidermidis (32), S. pseudintermedius (33), and S. schleiferi (34), the new oxacillin MIC breakpoints resulted in a change of ME from 4.4% to 0.9% ME and a change in VME rate from 1.2% to 1.6%. Consequently, the oxacillin MIC breakpoints in the M100 31st edition were revised to ≤0.5 μg/ml as susceptible and ≥1 μg/ml as resistant for S. epidermidis, S. pseudintermedius, S. schleiferi, and other Staphylococcus spp. Analysis of the cefoxitin DD correlates to the new oxacillin MIC breakpoints demonstrated that there was no need to revise the cefoxitin DD breakpoints; this method is acceptable for testing Staphylococcus spp., with the notable exceptions of S. pseudintermedius and S. schleiferi (33, 34).

Throughout these evaluations, it has been demonstrated repeatedly that there is no single phenotypic method that is 100% accurate for detecting mecA-mediated methicillin resistance in all staphylococcal species. This has led to the inclusion of a statement in the M100 31st edition that tests for mecA and PBP2a are the most definitive tests for detection of methicillin resistance for all Staphylococcus spp. and that isolates that test positive for mecA or PBP2a or that test resistant by any of the recommended phenotypic methods should be reported as methicillin resistant. Tests for mecA and PBP2a may be the most definitive methods, but they may not be readily available or practical to use on every isolate or in every laboratory. Furthermore, only one PBP2a test is FDA cleared for Staphylococcus species other than S. aureus. Laboratories should attempt to identify staphylococci isolated from clinically significant infections to the species level, especially if a test for mecA or PBP2a cannot be performed, to guide appropriate phenotypic testing.

The M100 31st edition includes mention of two emerging members of the S. aureus complex, S. argenteus and S. schweitzeri, which are difficult to identify with routine methods in use in clinical laboratories. There is increasing evidence that S. argenteus is a causative agent of serious invasive diseases, consistent with the pathogenicity of S. aureus. In contrast, S. schweitzeri has not yet been implicated in human infections (35). The Gram stain appearances, colony morphologies, and biochemical reactions of these two newly recognized members of the S. aureus complex are identical to those of S. aureus. Some S. argenteus isolates appear beta-hemolytic, although this characteristic is not consistent. S. argenteus received its name due to the grayish or silver appearance of its colonies; however, the color is subtle and best appreciated on chocolate agar after 48 h of incubation. S. schweitzeri has a yellow pigment. Neither species is included in either of the two commercial FDA-cleared MALDI-TOF MS databases. However, if an isolate is identified as either of these two species using a research-use-only MALDI-TOF MS database, whole-genome sequencing, or other advanced technology, it should be reported as “S. aureus complex (S. argenteus)” or “S. aureus complex (S. schweitzeri)” to emphasize that these species are closely related to S. aureus. When identified, AST methods and breakpoints as described for S. aureus should be applied. Data from testing 28 clinical S. argenteus isolates, which included 3 mecA-positive isolates, showed 100% agreement between mecA results and cefoxitin DD and oxacillin MIC results, when applying S. aureus breakpoints (35).

Alternative testing agent for the prediction of tedizolid susceptibility.

Tedizolid and linezolid are members of the oxazolidinone antimicrobial class. Both antimicrobials are approved for treatment of ABSSSI caused by S. aureus, E. faecalis, Streptococcus pyogenes, Streptococcus agalactiae, and Streptococcus anginosus group. Linezolid breakpoints were approved by the AST Subcommittee in the early 2000s, and the ability to test linezolid is widely available. Tedizolid MIC breakpoints were approved by AST Subcommittee in 2016; however, fewer options for susceptibility testing of tedizolid are available. EUCAST recommends that isolates susceptible to linezolid can be reported as susceptible to tedizolid (36). In 2020, data were presented to the AST Subcommittee to support inclusion of a similar comment in the M100 31st edition. Testing of 21,969 isolates of S. aureus (14 tedizolid-resistant isolates), 2,986 of E. faecalis (including 6 tedizolid-resistant isolates), and 2,035 of S. pyogenes, 1,556 of S. agalactiae, and 408 of the S. anginosus group (0 resistant isolates) demonstrated that all isolates that tested susceptible to linezolid by MIC were susceptible to tedizolid by MIC. In contrast, 8 isolates that were resistant to linezolid (7 of S. aureus and 1 of E. faecalis) were susceptible to tedizolid. As such, comments were added to the Staphylococcus, Enterococcus, Streptococcus species beta-hemolytic group and Streptococcus species viridians group tables to indicate that isolates testing susceptible to linezolid by MIC are considered susceptible to tedizolid, whereas some isolates that test resistant to linezolid may be susceptible to tedizolid.

Updated guidance for the use of assays for the detection of resistance mechanisms.

As additional data are obtained, there will likely be additional reporting recommendations in future editions of M100 concerning situations during which discordance exists between results of phenotypic versus molecular tests for carbapenem resistance. In the M100 31st edition, determination of the presence of a carbapenemase now includes not only the presence of the applicable gene, but also demonstration of carbapenemase enzyme production as measured by phenotypic carbapenemase methods. If results from either of these tests for carbapenemase resistance do not correlate with phenotypic results from DD or MIC tests, action must be taken to resolve and/or explain the discrepancy. For example, a KPC gene may be detected in a positive blood culture broth, but the isolate recovered from the broth may test susceptible to carbapenems. If repeat testing shows the same discordant results, it could be that the KPC gene is present but not expressed. M100 lists suggested actions to take, depending on the nature of the discordant results. In this case, both results should be reported following a conversation with the clinician as the clinical course of action is not definitively known.

Cerebrospinal fluid—AST report “Warning.”

Tables 1A, 1B, and 1C in M100 have included a “Warning” comment that advises laboratories to refrain from reporting antimicrobial agents that do not readily cross the blood-brain barrier for any bacterial isolate from cerebrospinal fluid (CSF). The concern as with all AST results is that if the agent is listed on the patient report, a clinician unaware of all nuances of prescribing may consider using the agent, which could result in dire consequences for the patient. In the M100 31st edition, the wording (but not the message) of the CSF Warning comment has been revised, and doripenem, ertapenem, imipenem, and lefamulin have been added to the list of agents that should not be reported on CSF isolates. Meropenem is not included and is acceptable to report on CSF isolates, when indicated, as this agent has been shown to be effective in treating meningitis (37).

Clarification of the use of “I^.”

In 2020, the “^” symbol (i.e., the caret symbol) was introduced into the M100 30th edition to provide more clarification to the intermediate category definition. CLSI defines the intermediate (I) category by a breakpoint that includes isolates with MICs or corresponding zone diameters within the intermediate range that approach usually attainable blood and tissue levels and/or for which response rates may be lower than for susceptible isolates. The “I” category also includes a buffer zone for inherent variability in test methods which will prevent small uncontrolled technical factors from causing major discrepancies in interpretations. The confusion lies in that the “I” category is not used the same for all drug-bug combinations. To provide more clarity as to which “I” definition applies, “^” was added to “I” breakpoints in M100 Tables 2 to indicate which agents have the potential to concentrate at certain anatomical sites. The anatomical site where this predominately occurs is urine. This was further clarified in the current M100 31st edition, wherein the document links the potential to concentrate in the urine with the abbreviation “I^.” Further guidance as to how to report I^ antimicrobials can be found in the CLSI Spring 2021 News Update, along with case examples on how to implement I^ (https://clsi.org/media/3mrkwh4j/ast-news-update-volume-6-issue-1-april-2021_5-17-21.pdf) At present, I^ is listed for the β-lactam agents, aminoglycosides, and fluoroquinolones for Enterobacterales, Enterococcus spp., and P. aeruginosa, which represent the most frequently identified urinary pathogens and treatment options. Separate urine clinical breakpoints are available for cefazolin and Enterobacterales, but further expansion of urine-specific breakpoints for those agents that concentrate in the urine is not under consideration by CLSI.

Quality control.

Recent review of quality control (QC) data revealed that the previous acceptable DD range for amikacin (30-μg disk) and P. aeruginosa ATCC 27853 needed adjustment: the revised range listed in the M100 31st edition is 20 to 26 mm. Similarly, the DD QC ranges for eravacycline (20-μg disk) with Escherichia coli ATCC 25922 and S. aureus ATCC 25923 were adjusted to 17 to 24 mm and 19 to 26 mm, respectively. DD QC ranges for a 5-μg ceftobiprole disk were added; those for the 30-μg disk were removed from M100 and added to the M100 QC archives. Several new MIC QC ranges were added for agents that are not yet FDA approved, including aztreonam-nacubactam and cefepime-nacubactam. It is not uncommon for pharmaceutical manufacturers to request QC ranges while studies are being conducted prior to approval.

ADDITIONAL RESOURCES

A major focus of the AST Subcommittee is educational efforts, which are coordinated by the Outreach Working Group (ORWG). The ORWG facilitates an annual webinar, which has been presented since 2004 and is available online and on demand. The ORWG also provides biannual News Updates which describe some of the new content in M100 and its practical application in a laboratory setting. The CLSI website (www.clsi.org) houses links to the AST News Updates, webinars, dates and information for registering for upcoming meetings, and more.

Major changes to the CLSI M100 31st edition and subsequent actions for each laboratory to consider based on these changes are listed in Table 2. Some additional CLSI AST Subcommittee topics under discussion in 2021 are listed in Table 3.

TABLE 2.

Summary of major changes to the CLSI M100 31st edition and implementation considerations^a

M100 31st edition 2021 change	Organism/organism group	Implementation considerations
Azithromycin DD and MIC breakpoints	Shigella spp.	Projected test vol; may be higher priority in regions of endemicity or in public health laboratoriesb
Azithromycin DD breakpoints	Neisseria gonorrhoeae	Projected test vol; may be higher priority in public health laboratoriesb
Ceftolozane-tazobactam MIC breakpoints	Haemophilus influenzae	Projected test vol; most isolates are highly susceptible; susceptible-only breakpoint is provided in M100b
Imipenem-relebactam DD and MIC breakpoints	Enterobacterales, except Morganellaceae; Pseudomonas aeruginosa; anaerobes	Projected test vol; need for routine testing for isolates encountered and patient population servedb
Lefamulin DD and MIC breakpoints AST	Staphylococcus aureus, Haemophilus influenzae, Streptococcus pneumoniae	Projected test vol; most isolates of targeted species highly susceptible to lefamulinb
Disk diffusion using positive blood culture broth	Enterobacterales and select antimicrobial agents	Rapid AST results for GNR from bloodstream infections can potentially impact patient outcome; reasonable option to consider, particularly if no other rapid AST/AR detection method used for blood cultures positive with GNR
Oxacillin MIC breakpoints	Staphylococcus spp. other than S. aureus and S. lugdunensis	New breakpoints for oxacillin and other Staphylococcus spp. will improve performance but are less reliable than mecA or PBP2a; ascertain (discuss with manufacturer) performance of oxacillin for Staphylococcus spp. compared to mecA/PBP2a with AST system in use
Clarification of species included in Staphylococcus aureus complex	Staphylococcus spp.	S. argenteus should be interpreted using the S. aureus oxacillin breakpoints and reported as a member of the S. aureus complex
Recommendation that linezolid susceptibility as determined by MIC testing predicts tedizolid susceptibility	S. aureus, Enterococcus faecalis, Streptococcus pyogenes, Streptococcus agalactiae, Streptococcus anginosis group	If tedizolid used in a facility and linezolid is tested, could streamline reporting
Clarification of instructions and expanded list of drugs to suppress on CSF isolates (with doripenem, ertapenem, imipenem, lefamulin added to list of “do not report”)	All	Will help avoid inappropriate prescribing of antimicrobial agents that do not effectively cross the blood-brain barrier
Clarification of definition of intermediate ^ (I^) interpretive category—now only pertains to urine isolates	Enterobacterales, Pseudomonas aeruginosa, Enterococcus spp.	A report of I^ might be might be difficult to implement, depending on information technology resources available; can be conveyed in more general terms, such as an information comment
Amikacin DD QC ranges	Pseudomonas aeruginosa ATCC 27853	New data support updated QC ranges

^aPrioritization for implementation should be discussed at the institutional level with physicians, pharmacy, antimicrobial stewardship teams, and hospital leadership. Special considerations may be necessary for public health laboratories. Refer to the text for supplement details and references. Abbreviations: ASP, Antimicrobial Stewardship Program; AST, antimicrobial susceptibility test; AR, antimicrobial resistance; DD, disk diffusion; GNR, Gram-negative rods; vol, volume.

^bBased on the projected test volume, consider submission to outside laboratory, such as a reference or public health laboratory, when results are needed.

TABLE 3.

Some CLSI AST SC topics under discussion in 2021

Topic	Specific category
Breakpoint updates	Aminoglycosides
	Amoxicillin-clavulanate
	Cefiderocol
	Piperacillin-tazobactam (Enterobacterales)
	Oxacillin (Staphylococcus saprophyticus)
Reporting	Reformatting and expanding suggestions for Tables 1A to 1C in M100 (antimicrobial agents to test and report)
	Clarification of cefazolin/cefuroxime reporting (E. coli, K. pneumoniae, P. mirabilis)
	Phenotypic vs genotypic results for carbapenemases
Test methods	Disk diffusion using positive blood culture broth—additional organisms, antimicrobial agents, and incubation times
	Inducible cefazolin resistance in methicillin (oxacillin)-susceptible S. aureus
	Use of Haemophilus test medium (HTM) for H. influenzae
	Harmonization of CLSI disk content and QC protocols with EUCAST
	Testing Stenotrophomonas maltophilia and Burkholderia cepacia
Upcoming updates of current CLSI AST standards and guidelines	M39: Analysis and Presentation of Cumulative Antimicrobial Susceptibility Test Data (fall 2021)
	M02: Performance Standards for Antimicrobial Disk Susceptibility Tests
	M07: Methods for Dilution Susceptibility Tests for Bacteria That Grow Aerobically
	M45: Methods for Antimicrobial Dilution and Disk Susceptibility Testing of Infrequently Isolated or Fastidious Bacteria
	M52: Verification of Commercial Identification and Antimicrobial Susceptibility Testing Systems
Educational	M100: Interactive Educational Program (fall 2021)

CONCLUSION

As new antimicrobial agents become available, new mechanisms of resistance are recognized, and new technology is introduced, there is a need to ensure that those involved with any aspect of antimicrobial testing have the tools they require to address the new developments. The CLSI AST Subcommittee has been providing these tools, many of which are found in M100, consistently for over 5 decades. The ultimate goal is to help guide optimal patient management. It is not uncommon for a recommendation to be revised multiple times as new information becomes available.

Improvements to the CLSI AST and reporting recommendations are only possible because volunteers bring forth ideas and generously provide a rationale for and/or data to support a recommendation. CLSI encourages anyone who has a suggestion or data that might be beneficial to this process to contact CLSI or one of the current AST Subcommittee volunteers. For example, clinical laboratory scientists working at the bench frequently provide collective QC data to CLSI when they recognize a potential problem with an existing QC range. When sufficient data are obtained, the QC Working Group reviews the data and the range is adjusted, when appropriate. There are many other ways in which volunteer participation can impact change in laboratory practice and patient care.

To learn more about how you can become a volunteer and the roles of various levels of volunteers for the CLSI AST Subcommittee, you can access a detailed “New Attendee Orientation” on the CLSI website referenced above. We encourage all of you reading this to please consider what you can offer to the AST Subcommittee—whether it be QC data from your laboratory, an improved method for standardizing inoculum preparation, a suggestion for how to better word a report comment, etc. No contribution is too small, and some of the smallest contributions have led to recommendations that have undoubtedly helped to combat antimicrobial resistance and improve patient care.

Funding Statement

No funding