Antimicrobial-Resistant Nocardia Isolates, Taiwan, 1998–2009

TO THE EDITOR—Recently, Uhde et al [1] described the antimicrobial drug-resistant pattern of Nocardia isolates in the United States in a 10-year retrospective study. Among the 765 Nocardia isolates they studied, the most predominant species were Nocardia nova (28%), Nocardia brasiliensis (14%), and Nocardia farcinica (14%) [1]. Among those isolates, 61% were resistant to sulfamethoxazole and 42% were resistant to trimethoprim-sulfamethoxazole (TMP-SMZ). Brown-Elliot et al [2], however, proposed that there is geographical variation in the distribution of Nocardia species throughout the world, with some species being more prevalent in geographical locations with a specific climate. Therefore, we retrospectively assessed the species distribution in Taiwan and evaluated the drug-resistance pattern of individual Nocardia species. In addition, to avoid misidentification based on microscopic and phenotypic characteristics, we used 16S rRNA gene sequencing to accurately identify Nocardia isolates to the species level [3].

All clinical Nocarida isolates collected from patients treated in 4 large medical centers in Taiwan during the period 1998–2009 were studied. The 4 hospitals were National Taiwan University Hospital (Taipei; 2500 beds), National Cheng Kung University Hospital (Tainan; 1200 beds), Chi-Mei Medical Center (Liouying; 1500 beds), and Kaohsiung Medical University Chung-Ho Memorial Hospital (Kaohsiung; 1700 beds). The isolates were identified to the species level using conventional phenotypic identification methods and confirmatory 16S rRNA gene sequence analysis, as described elsewhere [3]. Antimicrobial susceptibilities of the isolates to TMP-SMZ, ceftriaxone, linezolid, amikacin, amoxicillin–clavulanic acid, imipenem, and ciprofloxacin were determined using the broth microdilution method, as recommended by the Clinical and Laboratory Standards Institute [4, 5].

During the study period, a total of 138 consecutive and nonduplicated isolates were investigated. The sources of the isolates included skin and soft–tissue specimens, respiratory tract specimens (sputum, bronchioalveolar lavage fluid, pleural effusion, and lung biopsy), brain biopsy specimens, blood samples, and lymph node specimens. Among the isolates obtained, N. brasiliensis (58 isolates [42%]) was the most common species, followed by Nocardia cyriacigeorgica (22 isolates [16%]), N. farcinica (11 isolates [8%]), Nocardia beijingensis (8 isolates), Nocardia otitidiscaviarum (7 isolates), N. nova (6 isolates), Nocardia puris (6 isolates), Nocardia asiatica (5 isolates), Nocardia flavorosea (5 isolates), Nocardia abscessus (3 isolates), and Nocardia asteroides complex, Nocardia rhamnosiphila, Nocardia carnea, Nocardia elegans, Nocardia jinanensis, Nocardia alba, and Nocardia Takedensis (1 isolate each).

The antimicrobial drug-resistant patterns of 7 antimicrobial agents against different Nocardia species are shown in Table 1. Among all Nocardia isolates, ≤10% were resistant to amikacin, TMP - SMZ, and linezolid; 17% were resistant to amoxicillin–clavulanic acid; and 10% were resistant to ceftriaxone. In contrast, ∼25% of the Nocardia isolates were resistant to imipenem—particularly N. brasiliensis (48%) and N. ottidiscaviarum (100%)—and most (86.2%) were resistant to ciprofloxacin. Among N. brasiliensis isolates, 98% were resistant to ciprofloxacin and 48% were resistant to imipenem. All N. cyriacigeorgica isolates were resistant to ceftriaxone.

Table 1.

Susceptibility Results of Nocardia Isolates by Indicated Species and Antimicrobial Agents

				No. of isolates resistant (% resistant)
Species (no. of isolates)	AMIK	AMOX-CLV	CTRI	CPFX	IMI	TMP-SMZ	LZD
Nocardiabrasiliensis (58)	0 (0)	1 (2)	0 (0)	57 (98)	28 (48)	1 (2)	1 (2)
Nocardia cyriacigeorgica (22)	0 (0)	0 (0)	0 (0)	22 (100)	0 (0)	0 (0)	1 (5)
Nocardia farcinica (11)	0 (0)	0 (0)	3 (27)	5 (45)	0 (0)	1 (9)	0 (0)
Nocardia beijingensis (8)	0 (0)	1 (13)	0 (0)	4 (50)	0 (0)	0 (0)	0 (0)
Nocardia ottidiscaviarum (7)	0 (0)	7 (100)	6 (87)	3 (43)	7 (100)	0 (0)	0 (0)
Nocardia nova (6)	1 (17)	3 (50)	0 (0)	5 (83)	0 (0)	0 (0)	0 (0)
Nocardia puris (6)	0 (0)	5 (83)	5 (83)	6 (100)	0 (0)	0 (0)	0 (0)
Nocardia asiatica (5)	0 (0)	4 (80)	0 (0)	5 (100)	0 (0)	0 (0)	0 (0)
Nocardia flavorosea (5)	0 (0)	0 (0)	0 (0)	5 (100)	0 (0)	0 (0)	0 (0)
Nocardia abscessus (3)	0 (0)	1 (33)	0 (0)	3 (100)	0 (0)	0 (0)	0 (0)
Nocardia alba (1)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	1 (100)	0 (0)
Nocardia asteroids complex (1)	0 (0)	1 (100)	0 (0)	1 (100)	0 (0)	0 (0)	0 (0)
Nocardia carnea (1)	0 (0)	1 (100)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Nocardia elegans (1)	0 (0)	0 (0)	0 (0)	1 (100)	0 (0)	0 (0)	0 (0)
Nocardia jinanesis (1)	0 (0)	0 (0)	0 (0)	1 (100)	0 (0)	0 (0)	0 (0)
Nocardia rhamnosiphila (1)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Nocardia takedensis (1)	0 (0)	0 (0)	0 (0)	1 (100)	0 (0)	0 (0)	0 (0)
Total (138)	1 (1)	24 (17)	14 (10)	119 (86.2)	35 (25)	3 (2)	2 (1)

NOTE. AMIK, amikacin; AMOX-CLV, amoxicillin-clavulanate; CPFX, ciprofloxacin; CTRI, ceftriaxone; IMI, imipenem; LZD, linezolid; TMP-SMZ, trimethoprim-sulfamethoxazole.

Our analyses revealed 2 important findings. First, they confirmed that there is geographical variation in the distribution of Nocardia species. In Taiwan, the most common pathogen was N. brasiliensis (42%) whereas N. nova complex is the most common isolates in the United States (211 [[28%] of 765 isolates) and Canada (109 [34%] of 325), and N. cyriacigeorgica (12 [32%] of 37) was the most common pathogen in Spain [1, 6, 7]. Second, although the majority of Nocardia isolates in the United States, Canada, and Spain were resistant to TMP-SMZ, ceftriaxone, and imipenem [1, 6, 7], we found that most of the Nocardia isolates in Taiwan were susceptible to TMP-SMZ. Accordingly, TMP-SMZ remains the empirical treatment of choice for nocardiosis in Taiwan.

In conclusion, our results demonstrate the importance of using molecular methods to identify Nocardia to the species level to understand the epidemiological distribution of different species and their antimicrobial resistance profiles.

Potential conflicts of interest. All authors: no conflicts.