Rhodococcus infection: a 10-year retrospective analysis of clinical experience and antimicrobial susceptibility profile

ABSTRACT

Rhodococcus equi is an opportunistic pathogen known to cause pulmonary and extrapulmonary disease among immunocompromised patients. Treatment is frequently challenging due to intrinsic resistance to multiple antibiotic classes. While non-equi Rhodococcus spp. are prevalent, their clinical significance is poorly defined. There is also limited data on antibiotic susceptibility testing (AST) of Rhodococcus infection in humans. We conducted a single-center, retrospective cohort study evaluating clinical characteristics, microbiologic profile, and AST of Rhodococcus infections between June 2012 and 2022 at our tertiary academic medical center. Identification of Rhodococcus spp. was performed by Sanger 16S rRNA gene sequencing and/or matrix-assisted laser desorption ionization-time of flight mass spectrometry, and AST was performed by agar dilution. Three hundred twenty-two isolates of Rhodococcus spp. were identified from blood (50%), pulmonary (26%), and bone/joint (12%) sources. R. equi/hoagii, R. corynebacterioides, and R. erythropolis were the most frequently isolated species, with 19% of isolates identified only to genus level. One hundred ninety-nine isolates evaluated for AST demonstrated high-level resistance to amoxicillin/clavulanate, cephalosporins, and aminoglycosides. More than 95% susceptibility to imipenem, vancomycin, linezolid, rifampin, and clarithromycin was observed. Non-equi species showed a significantly more favorable AST profile relative to R. equi. Clinically significant Rhodococcus infection was rare with 10 cases diagnosed (majority due to R. equi) and managed. The majority of patients received 2- or 3-drug combination therapy for 2–6 months, with favorable clinical response. Significant differences in AST were observed between R. equi and non-equi species. Despite high antimicrobial resistance to several antibiotic classes, imipenem and vancomycin remain appropriate empiric treatment options for R. equi. Future research evaluating mechanisms underlying antimicrobial resistance is warranted.

INTRODUCTION

Rhodococcus species are facultative-intracellular, Gram-positive rods that closely resemble Nocardia and Mycobacterium species (1). The highest burden of these organisms is found in the soil, water, and manure of grazing animals (2). Rhodococcus equi/hoagii (herein referred to as R. equi) underwent a recent taxonomic modification to R. hoagii to simplify the identification of species (due to a prior heterotypic synonym of Corynebacterium hoagii) (3, 4). R. equi is a zoonotic human pathogen that can be acquired via inhalation or local inoculation. Significant pulmonary and extrapulmonary diseases due to R. equi have been described predominantly among immunocompromised patients (85% of cases), particularly those with HIV infection (5) and solid organ or hematopoietic stem cell transplantation (SOT or HSCT) (6, 7), but rarely also in immunocompetent hosts (8). Pulmonary infection is frequently associated with cavitary upper-lobar pneumonia (PNA), which in combination with weak acid-fast staining can lead to misdiagnosis as pulmonary tuberculosis (9). Extrapulmonary manifestations, occurring with or without pulmonary disease, include subcutaneous abscess, brain abscess, meningitis, endophthalmitis or corneal ulcers, bone and joint infections, and isolated bacteremia in the setting of catheter-related infections (5, 6, 9, 10).

There are also several non-equi species of Rhodococcus including R. fascians, R. erythropolis, R. globerulus, R. corynebacterioides, and R. rhodochrous. While the clinical significance of these species is largely unknown, rare cases of diseases including pneumonia with bacteremia (11, 12), peritoneal dialysis catheter infection (13), keratitis and chronic endophthalmitis (14), ventriculoperitoneal shunt infection (12), and meningitis (15) secondary to non-equi species have been described.

Unfortunately, management remains a challenge as there are no established guidelines regarding choice and duration of therapy for Rhodococcus infection. Commonly employed antimicrobial therapy involves two to three in vitro active antimicrobial agents (typically a combination of macrolides, fluoroquinolones, imipenem, vancomycin, rifampin, linezolid, or aminoglycosides) administered for 2–8 weeks based on clinical improvement; longer courses of 2–6 months are generally recommended among immunocompromised hosts or those with severe or disseminated disease (8, 16). Much of the current knowledge regarding microbiologic profile and antibiotic susceptibility testing (AST) of R. equi is derived from clinical case series (5, 10) or veterinary literature (17). Additionally, there are limited AST data for the non-equi Rhodococcus spp. with treatment guided by case-by-case evaluation of in vitro susceptibility. Further complicating the clinical picture is the recent increase in multidrug-resistant R. equi isolates reported in the United States due to mass use of macrolide and rifampin prophylaxis on horse breeding farms (18, 19).

Consequently, an updated evaluation of microbiologic profile and AST of R. equi and non-equi spp. is needed to optimize clinical management of human infections. We therefore reviewed our institutional clinical experience with Rhodococcus infections and in parallel evaluated the antimicrobial susceptibility profile of Rhodococcus spp. at our reference laboratory over the past decade.

MATERIALS AND METHODS

Study design

We conducted a single-center, retrospective study to evaluate the microbiologic and clinical experience with Rhodococcus spp. at a tertiary medical center and reference laboratory between June 2012 and 2022. The study was reviewed by the Mayo Clinic Institutional Review Board and deemed exempt (IRB # 22-006143). Those younger than 18 years of age or those who did not provide research authorization were excluded from clinical evaluation.

Microbiologic isolates

Bacterial identification (ID) and AST were performed on isolates submitted to the Mayo Clinic Laboratories (MCL) from both peripheral clients and patients managed at Mayo Clinic Rochester (MCR). Sources of isolates included blood, pulmonary (sputum, bronchoalveolar lavage [BAL], pleural fluid, and lung tissue), bone and joint (synovial fluid, bone biopsy, and spinal or epidural abscess), skin and soft tissue, cerebrospinal fluid, intra-abdominal fluid, urine, and lymph node tissue.

Microbial identification and susceptibility testing

Specimens received in the clinical microbiology laboratory were cultured in BACTEC 960 Mycobacteria Growth Indicator Tube (MGIT) broth (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) and on Middlebrook 7H11/7H11S agar biplates. Cultures were incubated at 37°C for 6 weeks. Positive MGIT broth was subcultured in a Middlebrook 7H11 agar plate, and isolated colonies from that subculture growth were used for Sanger dideoxynucleotide sequencing of a 500 bp region of the 16S rRNA gene as previously described (20). Databases used for sequencing analysis included the MicroSeq ID 16S rDNA 500 library, a Mayo Clinic custom database, and NCBI GenBank BLAST analysis. When using BLAST analysis, the sequence was required to be published in the peer-reviewed literature in order to be used for identification of the isolate. A 100% match to the database entry was required for all identifications. Since 2015, matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry has also been used for identification of aerobic actinomycetes as previously described (21). The required threshold for identification by MALDI-TOF mass spectrometer was a score of 2.0 or higher (21). AST was performed using the current Clinical & Laboratory Standards Institute (CLSI) guidelines during the study period (22). The Sensititre RAPMYCO plate (Thermo Fisher Scientific, Waltham, MA, USA; formerly Trek Diagnostics) was used for determining the minimal inhibitory concentration of each isolate. CLSI interpretive criteria for susceptible or resistant isolates used the most current CLSI interpretive table for Nocardia and other aerobic actinomycetes such as Rhodococcus species (23).

Clinical data

Clinical experience with Rhodococcus infection was evaluated among adult patients treated at MCR. Electronic medical records were reviewed for the following data: clinical comorbidities, Charlson comorbidity index (CCI) (24), Pitt bacteremia score (25, 26), immunocompromising conditions, imaging and microbiologic evaluation, antimicrobial therapy (inclusive of all directed antibiotics), duration of treatment, and clinical outcome. Diagnosis of Rhodococcus infection was determined by evaluating presentation, microbiologic ID, and radiographic findings by an infectious diseases physician (preferred) or primary service. Need for surgical intervention and type of surgery were assessed. Outcomes within 90 days of therapy completion were defined as cure (if clinical and radiographic improvement with therapy and lack of regression following discontinuation) or relapse (initial improvement but recurrent symptoms and/or radiographic worsening following discontinuation of therapy). All-cause mortality at 90 days following treatment completion was evaluated.

Data analysis

Clinical data were collected and managed using REDCap electronic data capture tools hosted at Mayo Clinic (27, 28). Clinical characteristics and patient demographics were summarized using descriptive statistics through R version 4.1.3 (29).

RESULTS

Isolate identification

Over the 10-year study period, 322 Rhodococcus isolates were identified, with 199 isolates evaluated for AST. The majority of the isolates (90%) were referred by MCL clients at peripheral institutions for ID and/or AST, with 32 isolates submitted from patients treated at MCR. The number of Rhodococcus isolates obtained for identification each year is reported over the past decade, with increasing frequency noted (Fig. 1A). For the isolates cultured in our laboratory, the average time to culture positivity was 10 days (range 2–29 days) using MGIT broth cultures and 19 days (range 3–40 days) using Middlebrook agar culture. The predominant sources of Rhodococcus isolates included blood (49.7%), pulmonary (25.5%), and bone and joint (11.8%) (Fig. 1B). Species level identification was feasible in 81% of isolates (262/322), with 57% (150/262) identified as R. equi/hoagii. The most common non-equi species included R. corynebacterioides (17%), R. erythropolis (14%), and R. rhodochrous (2%). Sixty of 322 isolates (19%) were only identifiable to genus level (Fig. 1C).

Fig 1.

Microbiologic profile of Rhodococcus infection including (A) annual trends in Rhodococcus isolate identification, (B) sources of isolates, and (C) species-level identification.

Antibiotic susceptibility testing

AST was performed on 199 isolates (Fig. 2), with significant heterogeneity observed in AST including a high level of resistance noted to penicillin/β-lactamase combination agents and cephalosporins, including extended-generation agents such as cefepime. However, imipenem susceptibility remained high (99.5%), with most isolates having minimum inhibitory concentration (MIC) ≤2 µg/mL. Among intravenous agents, uniform susceptibility was noted to vancomycin, but high rates of resistance were noted to both amikacin (33.8%) and tobramycin (16.7%). Among the oral agents assessed, Rhodococcus spp. were uniformly susceptible to linezolid and rifampin, with 95% of isolates susceptible to clarithromycin. Interestingly, discordance was observed in susceptibility among fluoroquinolones with high rate of susceptibility to moxifloxacin (94.4%), but not ciprofloxacin (62.1%). Trimethoprim/sulfamethoxazole (TMP-SMX) and minocycline retained reasonable activity, in the range of 80%–90% (Fig. 2).

Fig 2.

Antimicrobial susceptibility profile of Rhodococcus species with MIC distribution of oral and intravenous antibiotics evaluated. Abbreviations: MIC, minimum inhibitory concentration; TMP-SMX, trimethoprim/sulfamethoxazole; S, susceptible; I, intermediate; R, resistant. CLSI interpretive criteria for susceptible or resistant isolates used the most current CLSI interpretive table for Nocardia and other aerobic actinomycetes such as Rhodococcus species (21).

Due to a paucity of data on AST of non-equi species, we further delineated the susceptibility profile among 105 isolates of R. equi and 94 isolates of non-equi species (Fig. 3). Notably, the high resistance to amoxicillin/clavulanate and cephalosporins appeared to be unique to R. equi, with higher rates of susceptibility (≥80%) observed among non-equi species. The non-equi species also demonstrated increased susceptibility to amikacin, tobramycin, ciprofloxacin, and moxifloxacin. Relative to R. equi, a minor reduction in susceptibility to minocycline (85.6% vs 83%) and clarithromycin (96.2% vs 93.6%) was noted among non-equi species.

Fig 3.

Rhodococcus equi and non-equi species antimicrobial susceptibility profile. Abbreviations: S, susceptible; I, intermediate; R, resistant.

We also evaluated susceptibility profiles on a species level (Table S1). Appreciating limitations in low number of isolates (<10 isolates) for several non-equi species, patterns unique to species level were observed. For instance, R. erythropolis was noted to have increased resistance to aminoglycosides, minocycline, and TMP-SMX, even relative to R. equi, whereas most of the remaining non-equi species were uniformly susceptible to most tested antibiotics. Rhodococcus spp. only identified to genus level demonstrated a heterogeneous susceptibility profile. While it is possible that some of these isolates represent uncharacterized species or R. equi not identified due to sequence variability, our lab does not pursue further identification as part of routine isolate ID.

Clinical experience

Of 32 isolates identified at MCR, 10 patients were diagnosed with clinically significant Rhodococcus infection and received therapy (Table 1). The reason for exclusion of the 22 cases is described in detail in Fig. S1, with the primary exclusion from treatment due to suspected contamination or colonization with non-equi isolates. The median age at diagnosis was 62 years (interquartile range [IQR] 44–67), with eight patients noted to have an immunocompromising condition (Table 1; Table S2). R. equi was the causative pathogen in seven patients, with a majority of patients presenting with a primary pulmonary process (i.e., pneumonia with consolidation, nodular disease, or empyema), with two patients developing concurrent bacteremia, and one case of isolated central line-associated bloodstream infection (CLABSI). Two patients developed empyema requiring thoracotomy with decortication for source control. Two patients with R. globerulus and R. erythropolis infection were noted to have isolated bacteremia suspected to be secondary to CLABSI. Interestingly, one case of isolated CNS infection was noted in an otherwise immunocompetent host with Rhodococcus spp. identified only to the genus level.

TABLE 1.

Narrative review of diagnosis and treatment of Rhodococcus infection at our tertiary medical center^a,b,c

Patient	Age	Sex	CCI	Immune compromising condition	Tx to infection (days)	Microbiology: site cultured and result	Pathogen	Dx	Treatment and duration	Secondary regimen and duration	90-day outcome
Patient 1	68	M	7	AML s/P allogeneic SCT acute GVHD	121	Blood culture +	R. globerulus	CLABSI	IV vancomycin oral moxifloxacin duration: 60 days	Moxifloxacin duration: 80 days	Cure
Patient 2	45	F	3	Lung SCC; not on active chemotherapy	N/A	Pleural FNA +	R. equi/hoagii	Empyema	Sgx: thoracotomy* IV vancomycin oral levofloxacin duration: 14 days	N/A	Cure
Patient 3	36	M	4	Liver transplant	1,457	BAL +	R. equi/hoagii	PNA with nodular disease	IV vancomycin oral levofloxacin duration: 30 days	Azithromycin + bactrim duration: 350 days	Cure
Patient 4	63	M	9	Relapsed CLL on chemotherapy	N/A	Bronchial brush + pleural fluid –	R. equi/hoagii	PNA with empyema	Sgx: thoracotomy* IV vancomycin oral ciprofloxacin oral clarithromycin duration: 37 days	Ciprofloxacin + clarithromycin duration: 121 days	Cure
Patient 5	60	M	6	Kidney transplant	3,909	Blood culture +	R. equi/hoagii	PNA bacteremia	IV vancomycin oral minocycline oral azithromycin duration: 67 days	Moxifloxacin + azithromycin duration: 30 days	Cure
Patient 6	72	M	8	Kidney-pancreas transplant	1,760	Bronchial biopsy +	R. equi/hoagii	Pulmonary mass	IV vancomycin IV meropenem oral azithromycin duration: 60 days	Azithromycin duration: 1 year	Cure
Patient 7	20	F	3	None	N/A	Blood culture +	R. equi/hoagii	CLABSI	Oral moxifloxacin oral rifampin duration: 14 days	N/A	Cure
Patient 8	44	F	3	Systemic mastocytosis with MGUS	N/A	Blood culture +	R. equi/hoagii	Isolated bacteremia	Oral moxifloxacin oral rifampin duration: 21 days	N/A	Cure
Patient 9	64	M	4	None	N/A	CSF + sputum -	Rhodococcus spp.	PNA Meningitis	IV vancomycin IV meropenem (5 d) duration: 21 days	N/A	Cure
Patient 10	76	F	8	AML on chemotherapy	N/A	Blood culture +	R. erythropolis	CLABSI	IV vancomycin duration: 28 days	N/A	Cure

^a M, male; F, female; CCI, Charlson comorbidity index; Tx, treatment; Dx, diagnosis; AML, acute myelogenous leukemia; s/P, status post; SCT, stem cell transplantation; GVHD, graft versus host disease; CLABSI, central line-associated bloodstream infection; SCC, squamous cell carcinoma; N/A, not applicable; FNA, fine needle aspiration; PNA, pneumonia; CLL, chronic lymphocytic leukemia; BAL, bronchoalveolar lavage; Sgx, surgery; MGUS, monoclonal gammopathy of unclear significance; CSF, cerebrospinal fluid.

^b Definitions: Outcomes within 90 days of therapy completion were defined as cure (if clinical and radiographic improvement with therapy and lack of regression following discontinuation) or relapse (initial improvement, but recurrent symptoms and/or radiographic worsening following discontinuation of therapy).

^c * Subset of patients needing surgical management and corresponding surgical intervention.

Most patients received a two- or three-drug combination regimen with intravenous vancomycin plus oral fluoroquinolone or macrolide, with a median duration of primary regimen of 29 days (IQR 20–54 days). Patients with underlying immunocompromise, due to HSCT or SOT, frequently received a secondary regimen with oral agents (most commonly macrolide or fluoroquinolone; Table 1) for a median of 101 days (Table S2). All patients had a positive response to therapy and were noted to be cured at 90 days following treatment completion, with no clinical or microbiologic relapse (Table 1). No mortality was observed at 90 days of treatment completion.

DISCUSSION

We conducted a single-center, retrospective review of patients with Rhodococcus infection over the past decade to define the clinical experience, microbiology, and antimicrobial susceptibility profile of Rhodococcus isolates. Within this cohort, a majority of Rhodococcus isolates were from blood or pulmonary sources, consistent with the clinical preponderance for pulmonary disease with hematogenous dissemination (5, 6, 30, 31). Interestingly, we noted an increasing frequency of species-level identification of isolates submitted to our reference laboratory. However, 19% of isolates were only identifiable to genus level. This is consistent with prior reports of 80% species-level identification of Rhodococcus using MALDI-TOF (32).

Among isolates identified to species level, R. equi accounted for 57% of isolates, with remaining non-equi species most frequently identified including R. corynebacterioides, R. erythropolis, and R. rhodochrous. Pulmonary and extrapulmonary diseases due to R. equi are well described among immunocompromised hosts, particularly those with HIV infection (5) and SOT (6, 7), with mortality rates ranging from 25% to 55% (8). Clinical infection due to non-equi species is poorly defined and limited to case reports. R. corynebacterioides, the most commonly isolated non-equi species within our cohort, has been demonstrated to cause sepsis secondary to isolated bacteremia in patients with hematologic (33) and solid organ malignancy (34). R. erythropolis, R. rhodochrous, and R. fascians have been implicated in CLABSI, peritoneal dialysis catheter infection (13), keratitis and chronic endophthalmitis (14), and ventriculoperitoneal shunt infection (12). Given this predilection for localized disease, it is plausible that direct inoculation or environmental contamination may play an important role in acquisition of non-equi infection.

To the best of our knowledge, this is the largest study to date characterizing the antimicrobial susceptibility profile of human Rhodococcus isolates. R. equi demonstrated a heterogeneous antibiotic susceptibility profile. We observed high-level resistance to amoxicillin/clavulanate and cephalosporins (including extended-generation agents like cefepime), but imipenem remained highly susceptible (99%). This is consistent with previous studies (17), with hypothesized mechanisms supporting both intrinsic and acquired resistance via antimicrobial degradation due to β-lactamase production (30, 35). Albeit rare, imipenem resistance has been associated with altered penicillin binding proteins (PBPs) (36). Susceptibility to other carbapenems has not been systematically evaluated. However, imipenem-susceptible Rhodococcus isolates have previously demonstrated equivalent MIC₅₀ and MIC₉₀ to meropenem (36), with clinical evidence of success with use of meropenem in R. equi infection (10).

In addition to imipenem, intravenous vancomycin has been proposed as a potential backbone for the induction regimen (8, 16). In support of this, we observed uniform in vitro susceptibility to vancomycin in all tested isolates. Interestingly, however, aminoglycosides including amikacin and tobramycin had high rates of resistance that may preclude use as part of empiric therapy. Prior in vitro studies identified a similar pattern of higher MIC₉₀ to amikacin (4 µg/mL), but not gentamicin (≤1) (17, 37). The mechanism underlying variable resistance to aminoglycosides is unknown and warrants further evaluation.

Antibiotics with intracellular activity have been recommended for treatment of R. equi infection because intrahistiocytic survival is a major virulence determinant in pathogenesis (38). Consistent with this, we observed >95% susceptibility of R. equi to clarithromycin, rifampin, and linezolid. Most laboratories, including ours, assess macrolide class resistance using clarithromycin because no standardized CLSI breakpoints have been established for azithromycin (39). Susceptibility to azithromycin for most aerobic actinomycetes can be derived using clarithromycin, albeit this correlation is not well established for Rhodococcus spp. It is notable that azithromycin appears to have 4- to 8-fold higher MIC₅₀ for Rhodococcus in in vitro studies, suspected to be related to a technical issue with azithromycin solubility during AST rather than increased resistance (17, 38). Nonetheless, the overall susceptibility to macrolides is reassuring given the growing concern for multidrug-resistant R. equi in the setting of widespread rifampin/macrolide prophylaxis among horse breeding farms (18, 19). Finally, we confirmed the previously noted discordance in fluoroquinolones with higher rate of resistance to ciprofloxacin relative to moxifloxacin (40). While the mechanism for discordance is unclear, in vitro ciprofloxacin resistance in Rhodococcus veterinary isolates is driven by point mutations in DNA gyrase subunit A (37). Ultimately, the in vitro AST data for oral and intravenous antibiotics for R. equi (Fig. 2) support current guidelines recommending empiric imipenem and vancomycin as a backbone of therapy, particularly in severe infection and/or in immunocompromised hosts (16).

Unique to this study, we also further evaluated the AST pattern among non-equi Rhodococcus species (Fig. 3). There is a significant paucity of data for these species with antimicrobial therapy guided by in vitro studies. For the non-equi species, we observed a favorable antimicrobial susceptibility profile with higher rates of susceptibility to amoxicillin/clavulanate, cephalosporins, aminoglycosides, and fluoroquinolones (including ciprofloxacin). Interestingly, while overall susceptibility is improved, species-level variability was observed. For instance, R. erythropolis demonstrated rates of resistance comparable to R. equi for aminoglycosides, TMP/SMX, and minocycline. Unfortunately, a mechanistic understanding of species-level variability is lacking among Rhodococcus spp. and further evaluation is necessary.

Finally, we conducted a narrative review of clinical experience with 10 patients with clinically significant Rhodococcus infection managed at our institution. Notably, 22 of 32 Rhodococcus isolates identified at MCR were considered non-pathogenic, primarily due to suspected colonization or contamination with non-equi species (Fig. S1). Among those treated, R. equi was the predominant pathogen with the majority of patients (8 of 10) immunocompromised due to HSCT or SOT. Invasive pulmonary disease with bacteremia was frequently observed with need for surgical thoracotomy in two patients with empyema. Interestingly, isolated bacteremia was observed in the setting of CLABSI with both R. equi and non-equi species (R. globerulus and R. erythropolis), supporting environmental contamination of foreign devices as a potential mechanism of infection (13, 33). Approach to treatment was quite heterogeneous with patients receiving a two- or three-drug combination therapy as a primary “induction” phase for 29 days and maintenance phase with 1- or 2-drug regimen for a median of 101 days. In contrast to prior reports of 25% mortality in immunocompromised hosts, we fortunately did not observe any deaths and all patients achieved clinical cure at 90 days post-treatment completion. Nonetheless, clear clinical guidelines regarding diagnosis and management, which incorporate AST data, are warranted in future studies.

Limitations

Despite being the largest cohort to date, we are limited in our capacity to draw conclusions between in vitro AST and optimal management approaches due to the rarity of Rhodococcus infection and the retrospective nature of the study. We evaluated both R. equi and non-equi antimicrobial susceptibility profiles but are unable to propose a species-level antibiogram for the latter due to limited number of isolates. In keeping with this, we recommend cautious interpretation of species-level AST data, particularly due to high variability in species with <10 isolates. Within this study, we did not evaluate underlying mechanisms of heterogeneous resistance observed and this is an important question that requires future consideration as it likely has implications for other aerobic actinomycetes.

Conclusion

Within this study, we present a comprehensive overview of the antimicrobial susceptibility profile of R. equi and non-equi species, highlighting significant heterogeneity in AST that warrants careful consideration when tailoring treatment regimens for Rhodococcus infection. Acknowledging the rarity of infection, our clinical experience supports two- or three-drug regimens for 2–6 months in immunocompromised hosts as an appropriate strategy to achieve clinical cure. Future research is warranted to elucidate the mechanisms behind antibiotic resistance and to establish clinical guidelines for Rhodococcus infection management.

AFTER EPUB

[This article was published on 13 February 2024 with an incorrect publication date. The date was corrected in the current version, posted on 23 February 2024.]

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jcm.01537-23.

Supplementary Figure S1. jcm.01537-23-s0001.docx.

Rationale for inclusion and exclusion of clinical isolates of patients treated at our tertiary medical center.

Supplementary Table S1. jcm.01537-23-s0002.docx.

Rhodococcus equi and non-equi species specific antimicrobial susceptibility profile.

Supplementary Table S2. jcm.01537-23-s0003.docx.

Descriptive analysis of comorbidities, clinical presentation, and outcomes of Rhodococcus infection.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.