Improved recovery of nontuberculous mycobacteria in culture with adjunctive use of a selective agar

LETTER

NTM Elite (bioMèrieux, France) agar is a selective medium for the isolation of nontuberculous mycobacteria (NTM) from respiratory specimens. Decontamination methods are used to reduce bacterial growth in conventional culture for acid-fast bacilli (AFB), but these techniques can also kill 20%–90% of mycobacteria (1, 2). Conversely, there is no decontamination step prior to inoculating NTM Elite, and the total incubation time is only 4 weeks instead of 6 weeks (3). Here, mycobacterial recovery on NTM Elite was compared to recovery by conventional culture methods.

Prospective respiratory specimens (N = 376) from cystic fibrosis (CF, n = 156) and non-CF (n = 220) patients were plated onto NTM Elite using a swab and incubated in non-CO₂ conditions at 30°C ± 2°C. Plates were read at 4, 7, 14, 21, and 28 days as per manufacturer’s instructions (3). NTM showed a typical appearance on NTM Elite (e.g., rough/smooth and white/yellow), while breakthrough bacterial contaminants appeared blue/red.

All samples were also digested, decontaminated, and inoculated onto Middlebrook-7H11/Mitchison-7H11S agar, Lowenstein Jensen media (LJ, Remel, USA), and Mycobacterial Growth Indicator Tube (MGIT) broth (BD Diagnostics, USA). Solid media were incubated in CO₂ at 37°C ± 2°C and monitored for growth at 1, 3, and 6 weeks. MGIT tubes were incubated in the MGIT960 (BD Diagnostics) for 6 weeks. If any growth was detected, AFB was confirmed by performing Ziehl-Neelsen smears. NTM identification was performed by the GenoType NTM-DR ver1.0 line-probe (HAIN Lifescience, Germany) or laboratory-developed Sanger sequencing of 16S or rpoB as previously described (4).

When both methods were combined, 83 samples were positive and 89 organisms were detected (23.7% positivity, 89/376), with a positivity rate of 34.1% (75/220) from non-CF and 9.0% (14/156) from CF patients (Table 1). A total of 3 partially acid-fast bacteria (PAF), 26 rapidly growing (RGM), and 60 slowly growing (SGM) mycobacteria were detected (Table 2).

TABLE 1.

Comparison of growth on NTM Elite agar, conventional acid-fast bacilli (AFB) culture, and when both methods were combined

		NTM Elite	Conventional AFB culture	Conventional AFB culture + NTM Elite
Sensitivity of organism detection	CF	69.2% (9/14)	92.3% (12/14)	100% (14/14)
	Non-CF	56.0% (42/75)	92.0% (69/75)	100% (75/75)
	All	57.3% (51/89)	91.0% (81/89)	100% (89/89)
Positivity rate (percentage of organisms detected/samples)	CF	5.8% (9/156)	7.7% (12/156)	9.0% (14/156)
	Non-CF	19.1% (42/220)	31.4% (69/220)	34.1% (75/220)
	All	13.6% (51/376)	21.5% (81/376)	23.7% (89/376)
Cultures with breakthrough bacterial contaminationa	All	3.7% (14/376)	31.3% (118/376)	32.7% (123/376)
Average growth rate of SGMb	All	16.6 days	11.2 days	11.2 days
Average growth rate of RGMb	All	5.6 days	6.0 days	4.9 days

^a Note that this represents any contamination to show the rate of breakthrough bacteria; it does not represent a true contamination rate as defined by the CLSI M48 (5).

^b Out of 43 cultures that showed growth by both methods.

TABLE 2.

Organisms recovered from conventional culture and NTM Elite

Organism	Type of organisma	No. grown from conventional culture	No. grown from NTM Elite
Mycobacterium intracellulare	SGM	23	14
Mycobacterium avium	SGM	21	7
Mycobacterium abscessus subsp. abscessus	RGM	12	12
Mycobacterium abscessus subsp. massiliense	RGM	5	4
Mycobacterium intracellulare subsp. chimaera	SGM	5	3
Mycobacterium simiae	SGM	4	3
Mycobacterium fortuitum	RGM	2	1
Mycobacterium stomatepiae	SGM	1	1
Mycobacterium gordonae	SGM	2
Nocardia nova	PAF	2
Gordonia terrae	PAF	1
Mycobacterium alvei	RGM	1
Mycobacterium timonense	SGM	1
Nocardia cyriacigeorgica	PAF	1
Mycobacterium chelonae	RGM		2
Mycobacterium llatzerense	RGM		2
Mycobacterium marseillense	SGM		1
Mycobacterium paraterrae	SGM		1
Total	All	81	51

^a SGM: slowly growing mycobacteria; RGM: rapidly growing mycobacteria; PAF: partially acid fast organism.

The sensitivity of conventional culture was 91.0% (81/89) but varied by media type: 77.5% (69/89) for MGIT broth, 76.4% (68/89) for 7H11/7H11 biplates, and 29.2% (26/89) for LJ. The sensitivity of NTM Elite alone was 57.3% (51/89) and was greatest for RGM (80.8%, 21/26) rather than SGM (50%, 30/60) or PAF (0%, 0/3).

Here, NTM Elite had higher sensitivity than LJ and, as previously seen in a recent European study, detected unusual SGM like Mycobacterium llatzerense that did not grow by conventional culture (6). However, previous publications showing higher positivity from NTM Elite (6–8) suggest that on-site evaluations should be performed to account for possible differences in patient populations, organisms, and laboratory practices (1, 9). For instance, different geographic regions have different proportions of organisms and susceptibility patterns (4, 10). NTM Elite also had low rates of breakthrough contamination (3.7% versus 31.3% for other solid media, though mycobacterial growth could still be evaluated in all cases due to the redundancy of media types). It is not designed to capture M. tuberculosis, and therefore growth may be suitable for workup under biosafety level-2 conditions. These factors suggest that NTM Elite may be a useful adjunctive or replacement media in laboratories with higher rates of RGM, or to capture unusual mycobacteria. However, the cost, changes to workflow, and additional labor should be considered.

ETHICS APPROVAL

The National Jewish Health Human Research Protection Program reviewed and determined this study (HS-4138) to be of Exempt status. BRANY IRB (EXT23-062-528) reviewed and determined it to meet the waiver criteria per 45 CFR 164.512 (i)2(ii), authorizing it to use or access protected health information.