Address for correspondence: Leslie L. Barton, showed that the incidence in patients Thoracic Society and Infectious Dis-
University of Arizona, College of Medicine, treated for NTM pulmonary disease eases Society of America (3).
2335 East Seneca St, Tucson, AZ 85719; USA: (the group used as a surrogate of NTM Annual proportions of NTM iso-
email: firstname.lastname@example.org disease) has been decreasing over lates causing colonizations ranged
time, which is reassuring. from 29.2% in 2001 to 19.8% in 2007.
During the study period, annual in-
Eduardo Hernández-Garduño cidences of NTM colonization and
and R. Kevin Elwood disease increased from 6.6/100,000
Author afﬁliation: British Columbia Centre inpatients and 2.7/100,000 outpatients
for Disease Control, Vancouver, British Co- in 2000 to 34.5/100,000 inpatients and
Increasing 10.2/100,000 outpatients in 2008. My-
cobacterium avium complex (MAC)
Incidence of DOI: 10.3201/eid1606.100228
was the most prevalent species, colo-
Nontuberculous References nizing 1,282 (34.8%) of 3,681 patients.
Mycobacteria, 1. Lai CC, Tan CK, Chou CH, Hsu HL,
Annual proportions of MAC isolates
causing colonization ranged from
Taiwan, 2000–2008 Liao CH, Huang YT, et al. Increasing in- 20.0% in 2000 to 12.6% in 2006. An-
cidence of nontuberculous mycobacteria,
Taiwan, 2000–2008. Emerg Infect Dis. nual incidence of MAC colonization
To the Editor: Lai et al. (1) re-
2010;16:294–6. increased from 1.9/100,000 inpatients
ported an increase in the number of 2. Grifﬁth DE, Aksamit T, Brow-Elliott BA, in 2000 to 12.3/100,000 inpatients
nontuberculous mycobacteria (NTM) Catanzaro A, Daley C, Gordin F, et al. An in 2008; incidence of MAC disease
isolates and patients with pulmonary ofﬁcial ATS/IDSA statement: diagnosis,
treatment, and prevention of nontuber- also increased from 0.5/100,000 in-
NTM diseases after implementation of
culous mycobacterial diseases. Am J Re- patients in 2000 to 2.1/100,000 in-
the BACTEC system (Becton Dickin- spir Crit Care Med. 2007;175:367–416 patients in 2008. M. abscessus, the
son, Sparks, MD, USA) late in 2001. 10.1164/rccm.200604-571. DOI: 10.1164/ second most common species in our
These authors also reported that the in- rccm.200604-571ST
3. Hernández-Garduño E, Rodrigues M, study (2), caused colonization and
crease was mainly in persons infected
Elwood RK. The incidence of pulmonary disease in 669 and 155 patients, re-
with Mycobacterium avium complex non-tuberculous mycobacteria in British spectively. Annual incidence of M.
(MAC) and M. abscessus. They stated Columbia, Canada. Int J Tuberc Lung Dis. abscessus colonization and infection
that diseases caused by NTM were 2009;13:1086–93.
also increased from 1.49/100,000 in-
deﬁned according to current diagno-
Address for correspondence: Eduardo
patients and 0.3/100,000 outpatients
sis criteria published in 2007 (2). This
Hernández-Garduño, Department of Tuber-
in 2000 to 7.0/100,000 inpatients and
ﬁnding suggests that Lai et al. were
culosis Control, British Columbia Centre
1.9/100,000 outpatients in 2008.
able to review the clinical and radio-
for Disease Control, 1063-655 W 12th Ave,
Our study and a previous study
logic information for all patients.
Vancouver, British Columbia V5Z 4R4, Canada;
in British Columbia (4) suggest that
We wonder whether they were
improvement in diagnostic methods
also able to identify and exclude peo-
would detect increased incidence of
ple with NTM colonization, i.e., per-
NTM, especially of MAC; most iso-
sons with positive cultures for NTM
lates identiﬁed in these studies were
who did not meet the American Tho-
associated with colonization. We also
racic Society disease criteria. It would
demonstrated a gradual increase in the
have been interesting to know the
incidence of all NTM, MAC, and M.
trend in colonized persons. In a previ- In Response: We thank Hernán- abscessus over time in Taiwan, which
ous study from British Columbia (3), dez-Garduño and Elwood for drawing may be attributable to increasing vigi-
we found an increase in the number of attention to the issue of nontubercu- lance and awareness of these bacteria
NTM isolates mostly in persons with lous mycobacteria (NTM) coloniza- as human pathogens and the increased
MAC colonization. This ﬁnding coin- tion (1), which was not described in population of immunocompromised
cided with implementation of a new our study (2). Among 4,786 patients patients. Thus, clinicians should con-
laboratory technique in 2000, which with NTM isolates treated at our hos- sider diagnosing NTM diseases with
suggested that the new technology is pital during 2000–2008, colonization sensitive and advanced laboratory
more sensitive in detecting MAC. In was found in 76.9% (3,681), and only methods because of the increasing
contrast with the ﬁndings of Lai et 23.1% (1,105) had NTM diseases ac- population of patients at risk.
al., our study from British Columbia cording to the criteria of the American
Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 16, No. 6, June 2010 1047
Chih-Cheng Lai, Che-Kim Tan, Food Reservoir for ﬁeld gel electrophoresis , indicating
that identical genotypes (between iso-
Escherichia coli lates from retail meat and human in-
Chun-Hsing Liao, Causing Urinary fections) were observed only once.
Yu-Tsung Huang, Pan-Chyr Yang, Tract Infections Although we agree that the ﬁnd-
Kwen-Tay Luh, ing of a partial overlap between mul-
and Po-Ren Hsueh To the Editor: We read with in- tilocus sequence types of isolates
Author afﬁliations: Cardinal-Tien Hospital, terest the article by Vincent et al. that from retail meat and from human UTI
Taipei, Taiwan (C.-C. Lai); Chi-Mei Medical compared Escherichia coli isolates isolates is noteworthy (especially re-
Center, Tainan, Taiwan (C.-K. Tan); Nation- from 3 sources (human urinary tract covery of an ST131 isolate of avian
al Taiwan University College of Medicine, infections [UTIs], retail meat, and res- origin), the emphasis posed for the
Taipei (C.-H. Chou, H.-L. Hsu, Y.-T. Huang, taurant/ready-to-eat foods) by multiple role of food transmission in the dis-
P.-C. Yang, K.-T. Luh, P.-R. Hsueh); and Far molecular typing methods (1). This semination of the E. coli strains that
Eastern Memorial Hospital, Taipei (C.-H. study has to be considered in the con- cause community-acquired UTIs, in
Liao) text of the larger debate about the pos- our opinion, does not seem strongly
sible animal origin of E. coli isolates supported by the experimental data.
that cause extraintestinal infections in Nevertheless, the topic is relevant, and
humans (2–5), and the same authors we would highlight the importance of
(Vincent et al.) have declared, in the further research on this issue.
1. Hernández-Garduño E, Elwood RK. In- introduction, that their efforts were
creasing incidence of nontuberculous directed toward investigating the hy- Maria Giufrè, Caterina Graziani,
mycobacteria, Taiwan, 2000–2008. Emerg pothesis that retail chicken is the main Marisa Accogli,
Infect Dis. 2010;16:1047. and Marina Cerquetti
2. Lai C-C, Tan C-K, Chou C-H, Hsu H-L, reservoir for extraintestinal E. coli.
Liao C-H, Huang Y-T, et al. Increasing in- We strongly appreciate the Author afﬁliations: Istituto Superiore di San-
cidence of nontuberculous mycobacteria, amount of the experimental data and ità, Rome, Italy
Taiwan, 2000–2008. Emerg Infect Dis. some interesting ﬁndings, but we are
2010;16:294–6. DOI: 10.3201/eid1606.100158
3. Grifﬁth DE, Aksamit T, Brow-Elliott BA, not totally convinced of the authors’
Catanzaro A, Daley C, Gordin F, et al. An conclusions, particularly the assump- References
ofﬁcial ATS/IDSA statement: diagnosis, tion that the study strongly supports
treatment, and prevention of nontubercu- the preliminary hypothesis. First, the 1. Vincent C, Boerlin P, Daignault D, Do-
lous mycobacterial diseases. Am J Respir zois CM, Dutil L, Galanakis C, et al. Food
Crit Care Med. 2007;175:367–416. DOI: observation that only a low proportion
reservoir for Escherichia coli causing uri-
10.1164/rccm.200604-571ST (73/844, 8.6%) of the E. coli isolates nary tract infections. Emerg Infect Dis.
4. Hernández-Garduño E, Rodrigues M, El- analyzed belonged to clonal groups 2010;16:88–95.
wood RK. The incidence of pulmonary (deﬁned as >2 E. coli isolates that had 2. Collignon P, Angulo FJ. Fluoroquino-
nontuberculous mycobacteria in British lone-resistant Escherichia coli: food for
Columbia, Canada. Int J Tuberc Lung Dis. indistinguishable multilocus variable
thought. J Infect Dis. 2006;194:8–10.
2009;13:1086–93. number tandem repeats and enter- 3. Graziani C, Luzzi I, Corrò M, Tomei F,
obacterial repetitive intergenic con- Parisi G, Giufrè M, et al. Phylogenetic
Address for correspondence: Po-Ren Hsueh, sensus 2 patterns), including members background and virulence genotype of
Departments of Laboratory Medicine and from >1 source, suggests an overall ciproﬂoxacin-susceptible and resistant Es-
cherichia coli strains of human and avian
Internal Medicine, National Taiwan University high degree of genetic heterogeneity origin. J Infect Dis. 2009;199:1209–17.
Hospital, No. 7, Chung-Shan South Rd, Taipei among isolates from different sources. DOI: 10.1086/597423
100, Taiwan; email: email@example.com Second, looking at the single isolates 4. Johnson JR, Kuskowski MA, Menard
within clonal groups reported in Table M, Gajewski A, Xercavins M, Garau J.
Similarity between human and chicken
2, twelve (2.9%) of the 417 isolates Escherichia coli isolates in relation to cip-
from retail meat shared multilocus roﬂoxacin resistance status. J Infect Dis.
variable number tandem repeats, en- 2006;194:71–8.
terobacterial repetitive intergenic 5. Johnson JR, Sannes MR, Croy C, John-
ston B, Clabots C, Kuskowski MA, et al.
consensus 2, and multilocus sequence Antimicrobial drug–resistant Escherichia
types with some human UTI iso- coli from humans and poultry products,
lates; however, only 1 isolate (strain Minnesota and Wisconsin, 2002–2004.
EC01DT06–1737–01) was also found Emerg Infect Dis. 2007;13:838–46.
to be indistinguishable from a human
isolate (strain MSHS 161) by pulsed-
1048 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 16, No. 6, June 2010