Species Distribution And Antibiotic Susceptibility Of Nocardia Isolates From Yantai, China.

PURPOSE: This study aimed to investigate the species distributions and drug sensitivities among 19 strains of Nocardia isolated from Yantai, China, from 2017 to 2019. PATIENTS AND METHODS: Definitive species identification was performed by sequencing a fragment of the 16S rRNA gene (1480 bp) and by matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS). The susceptibilities of the isolates to 15 commonly-used antibiotics were tested using the microbroth dilution method.
RESULTS: Among the 19 Nocardia isolates, five species were confirmed. Seventeen of the 19 Nocardia spp. strains were identified consistently by the two methods, while two isolates of N. cyriacigeorgica were misidentified as N. otitidiscaviarum by MALDI-TOF MS. N. farcinica was the most common species (8/19), followed by N. cyriacigeorgica (6/19), N. otitidiscaviarum (2/19), N. brasiliensis (2/19), and N. nova (1/19). All isolates were susceptible to trimethoprim-sulfamethoxazole and amikacin, followed by linezolid and tigecycline (94.7% susceptibility rates). The sensitivity and minimum inhibitory concentration patterns for ciprofloxacin, moxifloxacin, clarithromycin, and tobramycin were significantly correlated with the species.
CONCLUSION: These results regarding the distribution and antibiotic resistance features of Nocardia species further our understanding of the diversity of Nocardia species circulating in Yantai, China, and thus support the use of more accurate empirical treatments.

Introduction

Nocardia comprises a genus of naturally occurring bacteria that can be isolated from water, soil, dust, decaying vegetation, and animal waste, and which can be responsible for opportunistic infections in immunocompromised and immunocompetent patients.1 Nocardia infection can lead to pneumonia, brain abscesses, and skin and soft tissue infections.2 Nocardiosis has been increasingly reported in recent years worldwide,3–5 though the distribution and antibiotic susceptibility of Nocardia varies geographically and over time.6,7 Nocardia is traditionally identified by biochemical methods, but the success rate is low. Data on the distribution of Nocardia species and the associated drug susceptibility patterns in Yantai, China, are thus currently lacking.8 To the best of our knowledge, the present study provides the first comprehensive evaluation of the clinical features, species identification, distribution, and antibiotic susceptibility profiles of Nocardia in a coastal city in China.

Materials And Methods

Strain Collection

Nineteen non-repetitive strains of Nocardia were collected from clinical specimens from seven hospitals in Yantai, Shandong Province, China, from January 2017 to March 2019. Of these, 13 were from Yantai Yuhuangding Hospital, which is the Regional Medical Center in Yantai, and the other six strains were isolated from six different central hospitals in Yantai city. Nineteen isolates from clinical specimens including sputum, bronchoalveolar lavage fluid, puncture fluid (knee-joint, hepatophyma liver abscess, pleural effusion), and skin and soft tissue pus were inoculated onto Columbia blood agar plates and cultured. All the strains were forwarded to Yantai Yuhuangding Hospital. Approval by the institutional review board and informed consent was obtained for experimentation with human subjects. Definitive identification was performed by microbial matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry (Microflex MALDI Biotyper; Bruker, Madison, WI, USA) and sequencing of a fragment of 16S rRNA gene.

Mass Spectrometry

For the standard Bruker MALDI-TOF MS preparation procedure, samples were collected from the plate and re-suspended in 300 µl sterile water. Absolute ethanol (900 µl) was added, the tubes were vortexed briefly and then centrifuged 13,000 rpm for 2 mins. The supernatant was removed, and the pellets were dried at room temperature. Then 70% formic acid (20 µl) and acetonitrile (20 µl) was added to each pellet. Samples were vortexed and centrifuged 13,000 rpm for 2 mins, 1 µl of the supernatant was placed on a MALDI target plate and allowed to air dry. After drying, spots were overlaid with 1 µl matrix solution and again allowed to dry. Mass spectrometry was performed on a MALDI Biotyper 3.1 Microflex LT system using the manufacturer’s settings. Species-level identification was accepted if the score was ≥ 2.00; genus-level identification was accepted if the score was 1.70–2.00.

DNA Extraction, PCR, And Sequencing

DNA was extracted from Nocardia strains using a TIANGEN Genomic DNA kit (TIANGEN Biotech, Beijing, China) according to the manufacturer’s instructions. In brief, isolated colonies were mixed with 400μl NF-H2O in a microcentrifuge tube. Then 3 μl RNase A was added to the tube and mixed by pulse vortexing for 5 mins. The column was centrifuged at 15,000 g for 2 mins. The supernatant was added with 400 μl buffer GP2 and the whole mixture was transferred to a mini spin column. The column was centrifuged at 12,000 rpm for 30 seconds, and the filtrate was discarded. After sequential washes with 600 μl buffer GD, 700 μl buffer PW, DNA was eluted with 20 μl NF-H2O and stored at −20°C.

The isolated Nocardia strains were identified at the species level by sequencing of the 16S rRNA gene (1480 bp) as described previously.9 PCR amplification and sequencing of the 16S rRNA gene were performed with the universal primers 27F (5ʹ-AGAGTTTGATCCTGGCTCAG-3ʹ) and 1492R (5ʹ-GGTTACCTTGTTACGACTT-3ʹ). The polymerase chain reaction (PCR) products were purified and sequenced by a genetic testing laboratory (Zhejiang Tianke High-tech Development Co., Ltd., Hangzhou, Zhejiang, China). PCR reactions were conducted using 96-well plates with a total reaction volume of 25 μl per well containing 12.5 μl phanta Max Master Mix (Vazyme, Nanjing, China), 0.5 μl each primer, 2.5 μl DNA, and 9 μl RNase-free water. The thermal cycling conditions of PCR were as follows: 95 °C for 2 mins, followed by 30 cycles of 95 °C for 30 s, 55°C for 30s and 72 °C for 30 s. Identification of Nocardia isolates at the species level species was based on a similarity value of ≥99.0% as described previously.9,10 Nocardia species determined by the 16S rRNA gene at NCBI GenBank using BLAST software () and leBIBI database ().

Antibiotic Susceptibility Testing

Antibiotic susceptibility testing was performed using the microbroth dilution method (Sensititre Susceptibility plates; TREK Diagnostic Systems Ltd., Cleveland, OH, USA). The antibiotics tested were amikacin, trimethoprim-sulfamethoxazole (SXT), moxifloxacin, ciprofloxacin, cefepime, imipenem, ceftriaxone, amoxicillin-clavulanate, linezolid, doxycycline, cefoxitin, minocyline, tigecycline, tobramycin, and clarithromycin. The minimum inhibitory concentration (MIC) breakpoints were interpreted following the CLSI recommendations M24-A for aerobic actinomycetes.11 Briefly, colonies of identified isolates growing on blood agar were collected and transferred into sterile 0.85% sodium chloride solution and vortexed repeatedly. The supernatant was then adjusted to a turbidity of 0.5 McFarland, 50 µl bacterial suspension was spread into Sensititre® cation-adjusted Mueller–Hinton broth with TES (TREK Diagnostic Systems) and the MICs were determined according to the instructions for the Sensititre Susceptibility plates after 72 h of incubation at 35°C. Staphylococcus aureus ATCC 29213 and Escherichia coli ATCC 35218 were used as controls.

Results

Distribution Of Nocardia Species

Five species were confirmed among the 19 Nocardia isolates. The clinical features are shown in Table 1. N. farcinica was the most common species (42.1%, 8/19), followed by N. cyriacigeorgica (31.6%, 6/19), N. otitidiscaviarum (10.5%, 2/19), N. brasiliensis (10.5%, 2/19), and N. nova (5.3%, 1/19), respectively. Two samples of N. cyriacigeorgica were misidentified as N. otitidiscaviarum by the MALDI Biotyper. In addition to gene sequencing analysis, phylogenetic relationships based on the 16S rRNA gene were also used to identify the 19 clinical Nocardia isolates (Figure 1).

Table 1

Clinical Features Of 19 Nocardia Isolates

Strain No.	Species By MALDI Biotyper	Species By16SrRNA	Accession Number	Age	Gender	Specimen	Clinical Diagnosis Underlying Diseases	Concurrent Infection	Initial Empirical Medication	Adjust Medication After Identifying Pathogen	Outcome
A3615	N. farcinica	N. farcinica	NR_117248.1	71	Female	Sputum	Pneumonia/bronchiectasis/endometrial cancer	No	BIA+LZD	SXT+IPM	recovered
A2671	N. farcinica	N. farcinica	NR_117248.1	58	Male	Sputum	Pneumonia/diabetes mellitus/nephrotic syndrome	No	AMX+AZM	SXT	recovered
A0495	N. farcinica	N. farcinica	NR_117248.1	82	Male	Sputum	Pneumonia/diabetes mellitus	No	MXF+IPM	SXT	recovered
A2788	N. farcinica	N. farcinica	NR_117248.1	29	Male	Drainage,	SSTIS/no	No	Cephalosporins(NA)+LEV	SXT+AMC	recovered
A2754	N. farcinica	N. farcinica	NR_117248.1	37	Male	Drainage,	Joint infection/no	No	Cephalosporins(NA)	SXT+LZD	recovered
A0457	N. farcinica	N. farcinica	NR_117248.1	36	Male	Drainage	SSTIS/no	No	Cephalosporins(NA)	SXT	Recovered
A2850	N. farcinica	N. farcinica	NR_117248.1	54	Female	Puncture fluid	Hepatapostema/Nephrotic syndrome/Diabetes mellitus	No	TZP+VCZ	SXT+IPM+LZD	Recovered
A2849	N. farcinica	N. farcinica	NR_117248.1	77	Male	Synovial Fluid	Joint infection/no	No	NA	SXT	Recovered
A3092	N. brasiliensis	N. brasiliensis	NR_117247.1	78	Male	Pus	SSTIS/diabetes mellitus	No	TZP	SXT	Recovered
A2748	N. brasiliensis	N. brasiliensis	NR_117247.1	54	Male	Pus	SSTIS/diabetes mellitus	No	TZP	SXT	Recovered
A2229	N. otitidiscaviarium	N. otitidiscaviarium	NR_117344.1	55	Female	BALF	Pneumonia/COPD	No	TZP+LEV	SXT+DOX	Recovered
A2350	N. otitidiscaviarium	N. otitidiscaviarium	NR_117344.1	58	Male	Sputum	Pneumonia/diabetes mellitus	No	TZP	SXT	Recovered
A3594	N. nova	N. nova	NR_117343.1	72	Female	Sputum	pneumonia/diabetes mellitus/siccasyndrome	Aspergillus	TZP	SXT+IPM	Death
A1175	N. otitidiscaviarium	N. cyriacigeorgica	NR_117334.1	41	Male	Sputum	pneumonia		TZP+LEV	SXT	Recovered
A2525	N. cyriacigeorgica	N. cyriacigeorgica	NR_117334.1	68	Male	Whole Blood	pulmonary abscess/septicemianephrotic syndrome		TZP	SXT+LZD	Recovered
A2427	N. otitidiscaviarium	N. cyriacigeorgica	NR_117334.1	27	Male	Pleural Effusion	pneumonia/myasthenia gravis/thymic carcinoma	Aspergillus	AZM+ VCZ +LZD	IPM+SXT+LZD+VRC	Death
A3603	N. otitidiscaviarium	N. cyriacigeorgica	NR_117334.1	65	Male	Sputum	Pneumonia/diabetes mellitus/COPD	Aspergillus	TZP+LEV+SCF	MXF+MEM+VRC	Recovered
A0494	N. cyriacigeorgica	N. cyriacigeorgica	NR_117334.1	60	Male	BALF	Pneumonia/bronchiectasis		MXF	SXT+BIA+AZM	Recovered
A2483	N. cyriacigeorgica	N. cyriacigeorgica	NR_117334.1	66	Male	Whole Blood	Pneumonia/AML/diabetes mellitus		IPM+TEC	MXF+TEC	Death

Abbreviations: NA, not available; AMK, amikacin; AMC, amoxicillin-clavulanic acid; FEP, cefepime; CRO, ceftriaxone; CIP, ciprofloxacin CLR, clarithromycin; DOX, doxycycline; IPM, imipenem; LZD,linezolid; MIN, minocycline; MXF, moxifloxacin; TOB, tobramycin; SXT, trimethoprim-sulfamethoxazole; BIA, biapenem; AMX, amoxicillin; AZM, azithromycin; LEV, levofloxacin; TZP, piperacillin-tazobactam; VCZ, voriconazole.

Figure 1

16S rRNA sequence-based phylogenetic tree of clinical isolates of Nocardia with those of closely related species which computed by the NJ analyses.

According to the criteria for judging invasive and noninvasive cases, six (24.5%) Nocardia strains isolated from sterile sites were considered to be non-invasive and the other 13 (75.5%) isolated from sputum, bronchoalveolar lavage fluid, and skin and soft tissue were considered to be non-invasive. Invasive cases included N. farcinica (3/6) isolated from joint infections (2/3) and hepatapostema (1/3), and N. cyriacigeorgica (3/6) isolated from whole blood (2/3) and pleural effusion (1/3).

Patient Characteristics

The individual characteristics of the 19 patients are shown in Table 1. The isolates were distributed evenly throughout the study period and no outbreaks were detected. Nocardiosis was diagnosed in 15 (78.9%) men and four (21.1%) women (age range 27–82 years, mean 57.2±17.3). No strains were isolated from children, even though one hospital included a 300-bed pediatric ward.

Underlying Diseases

Fifteen patients (78.9%) had underlying diseases (independent or simultaneous), including diabetes mellitus (n=9), pulmonary lesions (n=4), nephrotic syndrome (n=3), autoimmune disease (n=2), malignant tumor (n=2), and hematologic disease (n=1). The other five patients were immunocompetent, but one had pneumonia and four had arthritis and local infection.

Antibiotic Susceptibility Profiles

The susceptibilities of the Nocardia isolates to the 15 antibiotics are summarized in Table 2. Nocardia was most sensitive to SXT, linezolid, amikacin, and tigecycline, with sensitivities of 100%, 100%, 94.7%, and 94.7%, respectively. However, the sensitivities to imipenem and moxifloxacin were only 31.5% and 47.3%, and to ceftriaxone, amoxicillin-clavulanic acid, cefepime, oxycycline, minocycline, and clarithromycin were 21%, 10.5%, 10.5%, 5.26%, 5.26%, and 5.26%, respectively.

Table 2

Antimicrobial Susceptibility, MIC50 And MIC90 Of 19 Nocardia Isolates

Species	SXT	LZD	CIP	MXF	AMC	CRO	FEP	FOX	IPM	TOB	AMK	DOX	MIN	TGC	CLA
Susceptible %	100	94.7	36.8	47.3	10.5	21	10.5	5.26	31.5	52.6	100	5.26	5.26	94.7	5.26
MIC 50	0.25	2	≥8	1	16	≥128	≥64	128	8	2	≤1	4	4	0.5	≥32
MIC 90	0.25	2	≥8	4	32	≥128	≥64	≥128	≥128	16	≤1	4	4	1	≥32
N. farcinica	≤0.25	2	2	2	16	≥128	≥64	64	8	16	≤1	4	2	2	≥32
N. farcinica	≤0.25	2	0.25	≤0.25	16	≥128	≥64	≥128	8	16	≤1	4	4	1	≥32
N. farcinica	≤0.25	2	1	≤0.25	16	≥128	≥64	≥128	32	16	≤1	4	4	1	≥32
N. farcinica	≤0.25	≤1	0.5	≤0.25	16	≥128	≥64	128	16	16	≤1	4	4	0.5	≥32
N. farcinica	≤0.25	≤1	0.5	≤0.25	16	≥128	≥64	≥128	16	≥32	≤1	4	2	0.5	≥32
N. farcinica	≤0.25	2	0.5	≤0.25	8	≥128	≥64	64	8	16	≤1	4	4	0.5	≥32
N. farcinica	≤0.25	2	0.25	≤0.25	16	≥128	≥64	128	16	16	≤1	4	2	0.5	≥32
N. farcinica	≤0.25	2	1	≤0.25	8	16	32	32	4	8	≤1	4	2	1	≥32
N. brasiliensis	0.5	4	≥8	2	16	≥128	≥64	≥128	≥128	≤1	≤1	4	2	0.5	≥32
N. brasiliensis	≤0.25	2	≥8	1	16	≥128	≥64	≥128	≥128	≤1	≤1	4	2	0.5	≥32
N. cyriacigeorgica	≤0.25	16	≥8	8	32	8	16	128	8	≤1	2	4	4	1	≥32
N. cyriacigeorgica	≤0.25	≤1	≥8	4	16	4	4	128	≤2	≤1	≤1	2	4	1	≥32
N. cyriacigeorgica	≤0.25	2	≥8	2	32	16	≥64	128	8	≤1	≤1	8	4	1	≥32
N. cyriacigeorgica	≤0.25	2	≥8	4	32	4	16	128	4	≤1	≤1	4	4	1	≥32
N. cyriacigeorgica	≤0.25	≤1	≥8	2	32	≥128	≥64	128	≤2	≤1	≤1	2	2	1	≥32
N. cyriacigeorgica	≤0.25	2	≥8	4	32	16	≥64	128	4	≤1	≤1	4	4	1	≥32
N. otitidiscaviarium	≤0.25	2	≥8	1	32	≥128	≥64	≥128	64	4	≤1	1	4	0.25	≥32
N. otitidiscaviarium	≤0.25	≤1	≥8	4	≥128	≥128	≥64	≥128	≥128	2	≤1	2	≤1	0.5	≥32
N. nova	≤0.25	≤1	≥8	4	≥128	8	8	16	≤2	≥32	≤1	16	4	0.5	≤0.06

Abbreviations: AMK, amikacin; AMC, amoxicillin-clavulanic acid; FEP, cefepime; CRO, ceftriaxone; CIP, ciprofloxacin; CLR, clarithromycin; DOX, doxycycline; IPM, imipenem; LZD,linezolid; MIN, minocycline; MXF, moxifloxacin; TOB, tobramycin; SXT, trimethoprim-sulfamethoxazole.

Correlations Between Antimicrobial Susceptibility Profiles And Nocardia Species Designations

The correlations between drug susceptibility patterns and Nocardia species are demonstrated in Table 3. Susceptibility patterns per se were not indicative of a particular species, but N. farcinica, N. cyriacigeorgica, and N. nova all had certain characteristics not seen in N. brasiliensis and N. otitidiscaviarium.

Table 3

Antimicrobial Susceptibility Patterns Of Different Nocardia Species

Nocardia Species	No Of Isolates	Drug Patterns Types	Antimicrobial Susceptibility Pattern
non-Susceptible(Intermediate And Resistant)(%)		Susceptible(%)
N. farcinica	8	V	CRO	100	SXT	100
			FEP	100	AMK	100
			TOB	100	LZD	100
			CLA	100	TGC	87.5
					MXF	87.5
		CIP	87.5
N. cyriacigeorgica	6	VI	CIP	100	SXT	100
			CLA	100	AMK	100
			AMC	100	TGC	100
					TOB	100
		LZD	83
N. brasiliensis	2	VIII	CIP	100	SXT	100
			CLA	100	AMK	100
			CRO	100	TGC	100
			FEP	100	LZD	100
IPMc	100	TOB	100
N. otitidiscaviarium	2	VII	CIP	100	SXT	100
			CLA	100	AMK	100
			CRO	100	TGC	100
			FEP	100	LZD	100
IPMc	100	TOB	100
N. nova	1	III	CIP	Ra	CLA	Sa
			AMC		CRO
TOB DOX	FEP

Abbreviations: a, only one stain; AMK, amikacin; AMC, amoxicillin-clavulanic acid; FEP, cefepime; CRO, ceftriaxone; CIP, ciprofloxacin; CLR, clarithromycin; DOX, doxycycline; IPM, imipenem; LZD,linezolid; MIN, minocycline; MXF, moxifloxacin; TOB, tobramycin; SXT, trimethoprim-sulfamethoxazole.

Discussion

Only a few species of Nocardia, including N. brasiliensis, N. farcinica, and N. pseudobrasiliensis, can currently be reliably identified using traditional biochemical methods.12 Although molecular biology techniques (16S rRNA gene PCR and sequencing) allow more species within the Nocardia genus to be described,8 molecular identification techniques are difficult to carry out routinely in China because of high cost and a lack of standardization. However, MALDI-TOF MS technology has recently made the identification of bacteria faster and more convenient. In the current study, we used both methods to identify the 19 isolated strains of Nocardia, and showed that the results of the two methods were consistent for N. farcinica, N. brasiliensis, N. otitidiscaviarium, and N. nova, but that two strains of N. cyriacigeorgica were misidentified as N. otitidiscaviarium. These results suggested that, although MALDI-TOF MS was accurate for the genus-level identification of Nocardia, it could not reliably distinguish among Nocardia species.

More than 50 species of Nocardia have been identified as human pathogens to date,12 including species distributed in various geographic regions.13 The five species found in the current study were the most prevalent in different parts of the world.12 However, it was interesting to note that N. otitidiscaviarum and N. brasiliensis isolates, which are commonly found in tropical countries,14 were also isolated from samples from our temperate region. This may be associated with increases in world population mobility, but homology analysis is needed to explain this issue. Moreover, N. farcinica was the predominant species in our area, which differed from the patterns in the United States,7 Taiwan,15 Spain,16 and Iran,17 where N. nova, N. brasiliensis, and N. asteroides were the most prevalent species, respectively. N. brasiliensis is associated with insect bites all over the world so this could be another fact.

Fourteen of the 19 patients in the current study had at least one significant underlying condition. The types of underlying diseases were similar to those reported by Hashemi-Shahraki et al.17 although the proportions were slightly different. The most frequent conditions were diabetes mellitus (9 patients, 64.2%), followed by chronic lung disease (chronic obstructive lung disease or bronchiectasis; four patients, 28.5%), nephrotic syndrome (three patients, 21.4%), cancer (two solid tumors patients and one hematologic malignancy, 10.8%), and autoimmune diseases (two patients, 14.3%), while four (80%) of five patients with skin and soft tissue and joint infections had no underlying conditions.

Sulfamethoxazole and Trimethoprim (SXT), linezolid, and imipenem are the usual drugs for the empirical treatment of nocardiosis.18 However, only five of the 19 patients (26%) received initial empirical medication with one or both combinations, indicating the need to identify the infectious pathogens. After the pathogen was identified, 17 of the 19 patients received SXT alone or combined with other drugs, while the other two patients received moxifloxacin combined with other drugs. Each patient changed from empirical to targeted treatment, and 16 of the 19 patients achieved good therapeutic effects and were discharged after improvement. Three patients died, including two who were also infected with Aspergillus and one with a bloodstream Nocardia infection, suggesting that coexisting Aspergillus and bloodstream infections may be risk factors for a poor prognosis in patients with Nocardia infections.

The CLSI guidelines recommend broth microdilution for determining the antimicrobial susceptibility of Nocardia spp.11 and McTaggart et al.2 demonstrated the suitability of the Sensititre Rapmyco microdilution panel for the determination of antimicrobial susceptibility patterns in clinical isolates of Nocardia spp. As in other studies,2,5 five species showed good sensitivity to amikacin, linezolid, and SXT with MIC90 values of 1, 2, and 0.25 µg/mL, respectively, and susceptibility and sensitivity rates of 100% for all isolates. However, other studies5,8 reported higher resistance to SXT among N. farcinica, but low rates of SXT treatment failure. This apparent discrepancy may indicate a defect in the method of drug sensitivity testing, or differences between the results of drug sensitivity tests in vitro and drug action in vivo. Imipenem is one of the recommended empirical drugs,18 but the present study found that only 31.5% of isolates were sensitive to imipenem, with drug-resistant or moderately drug-resistant strains distributed among the other four species, suggesting that imipenem is no longer suitable for clinical treatment of Nocardia infection. The sensitivities to the three tetracycline drugs were similar to a previous report,19 with sensitivities to minocycline and doxycycline of 5.26% but MIC values indicating moderate drug resistance, while the sensitivity to tigecycline was 94.7% and the MIC90 was 1 μg/mL, suggesting that tigecycline might be a useful empirical clinical treatment. The sensitivity rates to cephalosporins and amoxicillin-clavulanic were 21% or less, with slight differences among the strains. Only N. cyriacigeorgica had a sensitivity rate of 50% to ceftriaxone.

We found a correlation between the drug susceptibility patterns and species in relation to ciprofloxacin, moxifloxacin, clarithromycin, and tobramycin, in line with previous reports of links between Nocardia taxonomy and specific patterns of antimicrobial susceptibility.2 In the present study, no isolates of N. farcinica were ciprofloxacin-resistant, while the other four species were all ciprofloxacin-resistant. Two N. otitidiscaviarum isolates were resistant to ciprofloxacin as reported by McTaggart et al.2 but in contrast to Brown-Elliott et al.12 who indicated that they were susceptible to ciprofloxacin. The MIC values of ciprofloxacin and moxifloxacin in N. farcinica were similar, but significantly different from other strains. The N. nova was sensitive to clarithromycin with a MIC value ≤0.06, in sharp contrast to the other four species. There were obvious differences in tobramycin sensitivity between N. farcinica and N. nova and the other three species. Interestingly, the two N. cyriacigeorgica strains wrongly identified by mass spectrometry could be distinguished by their MIC values, given that the MIC values of N. cyriacigeorgica isolates to tobramycin were all ≤1, while N. otitidiscaviarum was sensitive, but had a high MIC value.

Although the number of strains in our experiment was limited, the antibiotic susceptibility patterns per se were not indicative of a particular species. However, the results suggested that Nocardia isolates identified on the basis of other phenotypic characteristics but with conflicting susceptibility test results should be further examined and confirmed.

Conclusion

In conclusion, the present retrospective study reports 19 nonrepetitivce clinical Nocardia strains isolated and characterized from Yantai, China, in terms of their species, distribution, and antibiotic profiles. These results will improve our understanding of the clinical features of nocardiosis and of the Nocardia species distributed in our area, and will thus aid decision-making in the context of empirical treatment.