Research article

Genotyping and drug resistance patterns of M. tuberculosis strains in Pakistan

Mahnaz Tanveer¹ , Zahra Hasan¹ , Amna R Siddiqui² , Asho Ali¹ , Akbar Kanji¹ , Solomon Ghebremicheal³ and Rumina Hasan¹

¹  Department of Pathology and Microbiology, The Aga Khan University, Stadium Road Karachi, Pakistan

²  Department of Community Health Sciences, The Aga Khan University, Stadium Road Karachi, Pakistan

³  Department of Bacteriology, Swedish Institute for Infectious Diseases Control, Stockholm, Sweden

author email corresponding author email

BMC Infectious Diseases 2008, 8:171doi:10.1186/1471-2334-8-171

Published:	24 December 2008

Abstract

Background

The incidence of tuberculosis in Pakistan is 181/100,000 population. However, information about transmission and geographical prevalence of Mycobacterium tuberculosis strains and their evolutionary genetics as well as drug resistance remains limited. Our objective was to determine the clonal composition, evolutionary genetics and drug resistance of M. tuberculosis isolates from different regions of the country.

Methods

M. tuberculosis strains isolated (2003–2005) from specimens submitted to the laboratory through collection units nationwide were included. Drug susceptibility was performed and strains were spoligotyped.

Results

Of 926 M. tuberculosis strains studied, 721(78%) were grouped into 59 "shared types", while 205 (22%) were identified as "Orphan" spoligotypes. Amongst the predominant genotypes 61% were Central Asian strains (CAS ; including CAS1, CAS sub-families and Orphan Pak clusters), 4% East African-Indian (EAI), 3% Beijing, 2% poorly defined TB strains (T), 2% Haarlem and LAM (0.2). Also TbD1 analysis (M. tuberculosis specific deletion 1) confirmed that CAS1 was of "modern" origin while EAI isolates belonged to "ancestral" strain types.

Prevalence of CAS1 clade was significantly higher in Punjab (P < 0.01, Pearsons Chi-square test) as compared with Sindh, North West Frontier Province and Balochistan provinces. Forty six percent of isolates were sensitive to five first line antibiotics tested, 45% were Rifampicin resistant, 50% isoniazid resistant. MDR was significantly associated with Beijing strains (P = 0.01, Pearsons Chi-square test) and EAI (P = 0.001, Pearsons Chi-square test), but not with CAS family.

Conclusion

Our results show variation of prevalent M. tuberculosis strain with greater association of CAS1 with the Punjab province. The fact that the prevalent CAS genotype was not associated with drug resistance is encouraging. It further suggests a more effective treatment and control programme should be successful in reducing the tuberculosis burden in Pakistan.