Abstract
A fluoroquinolone (FQ) resistance rate of 5.9% is reported in 205 Mycobacterium tuberculosis isolates from patients presenting to field clinics in Karachi, Pakistan (2006 to 2009). FQ resistance among multidrug-resistant (MDR) strains was 11.1% (5/45), and it was 4.9% (5/103) in M. tuberculosis strains susceptible to all first-line agents. Spoligotyping of resistant strains did not show dominance of one strain type. Our data reflect considerable FQ-resistant M. tuberculosis isolates and the need to consider inclusion of FQ within first-line sensitivity testing in such settings.
Pakistan has the highest multidrug-resistant (MDR) tuberculosis (TB) burden in the Eastern Mediterranean region of the World Health Organization (WHO) (12). Fluoroquinolones (FQ) are an integral component of second-line therapy for multidrug-resistant strains (11). FQ resistance among Mycobacterium tuberculosis strains is reported in regions with high quinolone usage (1). Such regions also include Pakistan, with over-the-counter antimicrobial prescriptions and extensive FQ usage (2). National TB surveillance data to assess prevalence of FQ resistance in the country are not available. The current study was conducted to determine FQ resistance in M. tuberculosis at a community level.
Methods.
This study was conducted in Karachi, the largest city in Pakistan. Our study sites included 10 field clinics run by the Marie Adelaide Leprosy Centre (MALC), a nongovernmental organization working in partnership with the government of Pakistan. These clinics, within Karachi, are located in low-socioeconomic, highly populated administrative units. Patients are either self-referred or referred by their nearby general practitioners. The prevalence of FQ resistance was estimated through a study of consenting adult patients with a clinical suspicion of pulmonary TB at initial enrollment.
Early-morning sputum specimens were collected and transported to the clinical laboratory of the Aga Khan University Hospital (AKUH) for smear examination, culture, and drug susceptibility testing (DST). AKUH laboratory is accredited by the Joint Commission of International Accreditation (JCIA) and by World Health Organization (WHO) Supranational Laboratory external quality assurance for first- and second-line drug susceptibility testing.
Both LJ and MGIT (Becton Dickinson) were used for the isolation of M. tuberculosis for all specimens. M. tuberculosis was identified by the Bactec NAP TB differentiation test (Becton Dickinson), growth in p-nitrophenyl butyrate (PNB)-containing medium, nitrate reduction, and niacin accumulation (10). Susceptibility testing was performed using the agar proportion method on enriched Middlebrook 7H10 medium (BBL) at the following concentrations: rifampin (R), 1 μg/ml; isoniazid (H), 0.2 μg/ml and 1 μg/ml; and ethambutol (E), 5 μg/ml. Pyrazinamide (Z) sensitivity was carried out using the Bactec 7H12 medium (pH 6.0) at 100 μg/ml (Bactec PZA test medium, Becton Dickinson) (9). FQ susceptibilities were determined with ciprofloxacin (2 μg/ml) from 2006 to 2008 and with ofloxacin (2 μg/ml) from 2009 onward (9). M. tuberculosis H37Rv was used as a control with each batch of susceptibility testing. MDR was defined as resistance to both isoniazid (0.2 μg/ml) and rifampin.
Spoligotyping was performed on all FQ-resistant strains using a commercially available kit provided by Isogen Life Science B.V., Maarssen, Netherlands. DNA was extracted by the cetyltrimethylammonium bromide (CTAB) method (6). Spoligotyping based on the 43 spacers of the direct repeat (DR) region of the M. tuberculosis complex was carried out using primers DRa (5′ GGTTTTGGGTCTGACGAC 3′) and DRb (5′ CCGGAGAGGGGACGGAAAC 3′) as originally described (8). Negative and positive controls, including template-free PCR-amplified reaction mix and M. tuberculosis H37Rv DNAs, were used with each spoligotype blot. The spoligotyping results were entered in the BioNumerics software version 3.5 Applied Maths Program, Biosystematica, United Kingdom. A dendrogram was generated using the unweighted pair group method with arithmetic averages (UPGMA) calculation. The spoligotypes were compared with the most prevalent M. tuberculosis subfamilies, as identified by the world spoligotyping database, SpolDB4.0, of Pasteur Institute of Guadeloupe (http://www.pasteur-guadeloupe.fr/tb/bd_myco.html), including >40,000 isolates split into 1,030 shared types and >3,530 orphan profiles (5).
Data management and statistical analysis.
Data were double entered into an Epi database and checked for errors and were then transferred to statistical software SPSS version 16.0. Duplicate specimens from the same patients were excluded. Frequencies with percentages were computed for each category.
Results and discussion.
Fluoroquinolone susceptibility was checked in 205 patients. Of these, 53 (25.9%) had a history of previous treatment with first-line antituberculous drugs (isoniazid, rifampin, ethambutol, and pyrazinamide). Fluoroquinolones as antituberculous agents were not included in any of the patients' regimes. All patients gave a history of receiving multiple courses of antibiotics during the course of their current illness; however, due to an absence of centralized pharmacy records, this could not be further evaluated.
Of the 205 strains included, 5.9% (n = 12) were resistant to fluoroquinolones. The rate of resistance was highest among MDR strains (11.1% [n = 5]), followed by strains with resistance to ≥2 first-line drugs (6.5% [n = 2]). FQ resistance (4.9% [n = 5]) was also detected among strains fully sensitive to first-line drugs (Table 1). FQ resistance among previously treated and untreated groups did not differ significantly (9.4% versus 4.6%; P value of 0.206). FQ resistance was significantly higher in patients with durations of illness of more that 3 months (18.8% versus 3.5%; P value of 0.004). Spoligotyping of 10/12 FQ-resistant strains did not show dominance of any one particular strain type within these isolates (Table 2).
TABLE 1.
Fluoroquinolone resistance in community-based M. tuberculosis strains from Karachi (June 2006 to September 2009)
| Resistance pattern | Individuals previously treated for tuberculosis |
Untreated individuals |
Total |
|||
|---|---|---|---|---|---|---|
| No. of strains tested | No. (%) of FQ-resistant isolates | No. of strains tested | No. (%) of FQ-resistant isolates | No. of strains tested | No. (%) of FQ-resistant isolates | |
| Sensitive to all first-line agents | 13 | 0 | 90 | 5 (5.6) | 103 | 5 (4.9) |
| Resistant to any one first-line agent | 3 | 0 | 23 | 0 | 26 | 0 |
| MDR | 24 | 4 (16) | 21 | 1 (4.8) | 45 | 5 (11.1) |
| Resistant to ≥2 first-line drugs (other than MDR) | 13 | 1 (7.7) | 18 | 1 (5.6) | 31 | 2 (6.5) |
| Total | 53 | 5 (9.4) | 152 | 7 (4.6) | 205 | 12 (5.9) |
TABLE 2.
Spoligotype distribution of fluoroquinolone-resistant M. tuberculosis strains
| Genogroup | Shared type | No. (%) of isolates with shared type |
|---|---|---|
| CAS family | CAS1_DELHI ST1343 | 1 (10) |
| CAS1_DELHI ST26 | 2 (20) | |
| EAI family | EAI3_IND ST11 | 1 (10) |
| H family | H4 ST127 | 2 (20) |
| T family | T3 ST37 | 1 (10) |
| U family | U (CAS_ANCESTOR)27 | 1 (10) |
| Orphan clusters | Unique cluster 11 | 1 (10) |
| Orphan types | Unique | 1 (10) |
| Total no. of isolates | 10 |
Our data indicating FQ resistance among both MDR strains and strains sensitive to first-line agents reflect high FQ usage, a hypothesis supported by increasing FQ resistance in other bacterial pathogens from this area (7, 13). While FQs were not used in TB treatment, data of FQ usage in our patients prior to enrollment for TB treatment were not available; however, FQ is reported to be one of the most commonly used antibiotics in Pakistan for many indications (2). Moreover, significantly higher rates of resistance in patients with illness durations of more than 3 months also point to the possibility of multiple FQ prescriptions in these patients for indications other than tuberculosis.
The lack of association between FQ resistance and strain type argues against a role of particular genotypes in the dissemination of resistant strains in the community and argues for the concept of usage-driven resistance. This is in contrast to an earlier study reporting high rates of FQ resistance among specific strain types (4). Our data therefore point to an urgent need to control FQ usage in this community.
In general terms, a resistance rate of more than 5% renders an agent unsuitable for use as an empirical drug in the management of tuberculosis (3). To date, there are no guidelines for FQ testing other than as a second-line agent in treatment failure cases. We therefore recommend that in areas with high rates of FQ resistance, such as ours, FQ testing should be considered part of first-line DST and included in rapid diagnostics for the detection of resistant strains.
Acknowledgments
This study was supported through a grant from the Benenden Healthcare Society, United Kingdom, and in part by the joint Pak-U.S. Academic & Research Program HEC/MoST/USAID.
The authors report no transparency declarations.
Informed consent was taken from all study subjects after approval from the Ethics Review Committee of the Aga Khan University, Pakistan.
Footnotes
Published ahead of print on 6 December 2010.
REFERENCES
- 1.Agrawal, D., Z. F. Udwadia, C. Rodriguez, and A. Mehta. 2009. Increasing incidence of fluoroquinolone-resistant Mycobacterium tuberculosis in Mumbai, India. Int. J. Tuberc. Lung Dis. 13:79-83. [PubMed] [Google Scholar]
- 2.Butt, Z. A., A. H. Gilani, D. Nanan, A. L. Sheikh, and F. White. 2005. Quality of pharmacies in Pakistan: a cross-sectional survey. Int. J. Qual. Health Care 17:307-313. [DOI] [PubMed] [Google Scholar]
- 3.Centers for Disease Control and Prevention (CDC). 1993. Initial therapy for tuberculosis in the era of multidrug resistance—recommendations of the Advisory Council for the Elimination of Tuberculosis. MMWR Recomm. Rep. 42:1-8. [PubMed] [Google Scholar]
- 4.Duong, D. A., et al. 2009. Beijing genotype of Mycobacterium tuberculosis is associated with high-level fluoroquinolone resistance in Vietnam. Antimicrob. Agents Chemother. 53:4835-4839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Flores, L., et al. 2007. Large sequence polymorphisms classify Mycobacterium tuberculosis strains with ancestral spoligotyping patterns. J. Clin. Microbiol. 45:3393-3395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Honore-Bouakline, S., J. P. Vincensini, V. Giacuzzo, P. H. Lagrange, and J. L. Herrmann. 2003. Rapid diagnosis of extrapulmonary tuberculosis by PCR: impact of sample preparation and DNA extraction. J. Clin. Microbiol. 41:2323-2329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jabeen, K., A. Zafar, E. Khan, S. Irfan, and R. Hasan. 2010. Increase in isolation of ESBL-producing multidrug resistant nontyphoidal salmonellae from Pakistan. BMC Infect. Dis. 10:101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kamerbeek, J., et al. 1997. Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J. Clin. Microbiol. 35:907-914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.National Committee for Clinical Laboratory Standards. 2003. Susceptibility testing of Mycobacteria, Nocardia, and other aerobic actinomycetes, 2nd edition. Approved standard M24-A. NCCLS, Wayne, PA. [PubMed]
- 10.Winn, W. C., et al. 2005. Color atlas and text book of diagnostic microbiology. Lippincott Williams & Wilkins, Philadelphia, PA.
- 11.World Health Organization. 2009. Treatment of tuberculosis: guidelines. World Health Organization, Geneva, Switzerland.
- 12.World Health Organization. 2010. Multidrug and extensively drug-resistant TB (M/XDR-TB). 2010 global report on surveillance and response. World Health Organization, Geneva, Switzerland.
- 13.Zafar, A., and K. Jabeen. 2007. Antimicrobial resistance in Neisseria gonorrhoeae and limited treatment options. J. Pak. Med. Assoc. 57:333-334. [PubMed] [Google Scholar]