Prevalence and occurrence rate of Mycobacterium tuberculosis Haarlem family multi-drug resistant in the worldwide population: A systematic review and meta-analysis.

BACKGROUND: Transmission of Mycobacterium tuberculosis (M. tuberculosis) can occur in different ways. Furthermore, drug resistant in M. tuberculosis family is a major problem that creates obstacles in treatment and control of tuberculosis (TB) in the world. One of the most prevalent families of M. tuberculosis is Haarlem, and it is associated with drug resistant. Our objectives of this study were to determine the prevalence and occurrence rate of M. tuberculosis Haarlem family multi-drug resistant (MDR) in the worldwide using meta-analysis based on a systematic review that performed on published articles.
MATERIALS AND METHODS: Data sources of this study were 78 original articles (2002-2012) that were published in the literatures in several databases including PubMed, Science Direct, Google Scholar, Biological abstracts, ISI web of knowledge and IranMedex. The articles were systematically reviewed for prevalence and rate of MDR. Data were analyzed using meta-analysis and random effects models with the software package Meta R, Version 2.13 (P < 0.10).
RESULTS: Final analysis included 28601 persons in 78 articles. The highest and lowest occurrence rate of Haarlem family in M. tuberculosis was in Hungary in 2006 (66.20%) with negative MDR-TB and in China in 2010 (0.8%), respectively. From 2002 to 2012, the lowest rate of prevalence was in 2010, and the highest prevalence rate was in 2012. Also 1.076% were positive for MDR and 9.22% were negative (confidence interval: 95%).0020.
CONCLUSION: Many articles and studies are performed in this field globally, and we only chose some of them. Further studies are needed to be done in this field. Our study showed that M. tuberculosis Haarlem family is prevalent in European countries. According to the presence of MDR that was seen in our results, effective control programs are needed to control the spread of drug-resistant strains, especially Haarlem family.

INTRODUCTION

Mycobacterium tuberculosis (M. tuberculosis) is a primarily human pathogen that causes tuberculosis (TB) disease.[1] This disease damages the lungs, central nervous system, lymphatic system and circulatory system.[2] Rate of infection with M. tuberculosis is associated with host's inherited susceptibility, environmental risk factors and genetic variations.[3] There are approximately 8 million new TB cases each year and over 2 million deaths because of this disease.[4] Different studies showed that the one-third of the worldwide population is infected with M. tuberculosis and new cases occur at a rate of 1/s.[5] Transmission of TB can occur through different ways such as: Human-to-human and human-to-animal, poor hygienic conditions, blood and fecal contamination (transmission) and inhalation of contaminated droplets that are released from the lungs of an infected individual (coughing and inhalation). Prevention and control of TB infection can be achieved through the prompt detection and treatment of infectious patients, air related precautions, and treatment of people who are suspected or confirmed to have TB disease and vaccination with the TB vaccine (Bacillus Calmette — Guerin [BCG]). There are several kinds of tests that are used to detect TB bacteria: The tuberculin skin test and TB blood tests (Centers for Disease Control or test). The typing method is based on DNA presence of polymorphism at one particular chromosomal locus, the “Direct Repeat” (DR) region, which is in M. tuberculosis complex bacteria.[6] This locus was first described in Mycobacterium bovis BCG. The DR region in M. bovis BCG consists of directly repeated sequences of 36 base pairs, which are interspersed by nonrepetitive DNA spacers, each 35-41 base pairs in length.[7] The main genotype families of M. tuberculosis include Beijing, Haarlem, Africa, East-African-Indian, Latin American and Mediterranean strains. The Haarlem genotype is associated with widespread epidemics and drug resistance.[48] Several meta-analyses are conducted on different aspects of M. tuberculosis around the world: Trunz et al. showed by a meta-analysis that BCG vaccination was effective against childhood TB, meningitis and TB in the military and the number of cases that were prevented was highest in South East Asia (46%), sub-Saharan Africa (27%) and in the western Pacific region (15%).[9] Ling et al. (2008) in Canada by a meta-analysis investigated on the multi-drug resistant (MDR) TB based on genotype MTBDR assays and they showed that the sensitivity and specificity were 98.1% and 98.7% for rifampicin resistance, respectively.[10] Yang et al. (2011) in China using a systematic review and Meta-Analysis showed that the prevalence of drug-resistant TB in new cases was 27.9% and in previously treated cases was 60.3%.[11] Nowadays, this MDR quality is a major problem in TB treatment and control. Also, drug resistance in TB patients will increase in cases of improper use of antibiotics. Drug resistance is raising in areas with weak TB control programs. Also, transmission of TB to other individuals by drug-resistant patients is easily possible. Determining the M. tuberculosis prevalence and occurrence rate of MDR-TB can help to control and plan therapies. So it is important to know specific families that are over represented among drug-resistant cases, and in particular if these resistant strains are successfully transmitted within the community. Haarlem family is associated with drug resistance and rapid clonal expansion; this genotype has epidemic potential.[12] The aim of this study was to determine the prevalence and distribution rate of the Haarlem family MDR in worldwide using a meta-analysis based on a systematic review that performed on published articles.

MATERIALS AND METHODS

Search strategy

The different literature databases and original articles that were published in the time span of 2002-2012 in English language for determination of M. tuberculosis Haarlem family prevalence and occurrence rate of MDR in this family in worldwide population were obtained from valid and credible web sites: PubMed (http://www.ncbi.nlm.nih.gov/pubmed), Science Direct (http://www.sciencedirect.com), Google Scholar (http://www.scholar.google.co.uk), Biological Abstracts (http://www.science.thomsonreuters.com) and ISI web of knowledge (http://www.pcs.webofknowledge.com) in Thomsonreuters, the years of coverage of these web sites are 1955, 1977, 2006, 2009 and 1958, respectively. Also, some article was found in IranMedex (http://www.iranmedex.com). We contacted several study authors in the cellular and molecular research center and microbiology department of Kurdistan University of medical sciences, located in Sanandaj, Iran to identify some additional studies. The last search was performed in the winter of 2012. For the means of searching in different websites, several key words were entered. For example TB Haarlem, molecular epidemiology, MDR and spoligotyping were entered in PubMed (Advanced search part). In some cases, search process got limited because in order to reach to some of the papers’ full text, we should have had a specific username and password and hence we only used their abstracts which were available or free.

Data extraction and assessment of studies quality

Process for selecting the studies

All the information in different articles and papers used in this study were surveyed by authors. Two reviewers did the searching independently. When there was any disagreement between them for selecting specific articles, they discoursed about selecting it. Also, eligibility for each paper was determined in Excel data sheets (CEB603, Chino-Excel Technology). The following data were extracted from original publications: Number of cases, websites, Author, Study place (country), Year of the research, Sample size, and prevalence of Haarlem and MDR association.

Inclusion criteria

Research articles with full text,

Articles with abstract in English.

Excluded studies

Review articles,

Congress abstracts,

Studies that reported in languages other than English and

Studies that were not available for us in abstract or full text,

Studies that their sampling location was uncertain,

Studies that locations of sampling was performed at the same time,

Studies that their data were not clear.

Data synthesis and meta-analysis

Variables of our study were prevalence of TB infection and Haarlem family MDR in different countries (2002-2012). According to different studies performed in various countries, it was concluded that high levels of prevalence of TB infection and Haarlem family MDR can exist in different countries. Hence, the prevalence of Haarlem family of TB in 78 studies was the main outcome. The variance of the prevalence was computed using the binomial distribution (confidence interval [CI]: 95%). Meta-analysis with the random effect model was applied to combine the prevalence among studies. There was sensitivity (how the uncertainty in the output of a mathematical model or system can be apportioned to different sources of uncertainty in its inputs) and heterogeneity among studies. I2 and Q (P < 0.10) statistical tests were used to check out this heterogeneity (I2 static is percentage of observed total variation across studies that are due to heterogeneity rather than chance. It is calculated as I2 = 100% × (Q − df)/Q, where Q is Cochran's heterogeneity statistic and df degrees of freedom. Negative values of I2 are put equal to zero so that I2 lies between 0% and 100%. A value of 0% indicates no observed heterogeneity, and larger values show increasing heterogeneity. Q is the weighted of squares on a standardized scale. It is reported with a P value with low P values indicating the presence of heterogeneity. This test however is known to have low power to detect heterogeneity, and it is suggested to use a value of 0.01 as a cut of for significance. Conversely, Q has too much power as a test of heterogeneity if the number of studies is large. Subgroup analyses were assessed using Chi-square tests, and it was done for continents). Stratified analyses were subsequently performed with respect to geographical locations and MDR-TB among the Haarlem family isolates.[13] Meta-analysis was carried out using the software package Meta R (Version 2.13.2, Copyright 2011, The R Foundation for Statistical Computing).

Publication bias

There was some evidence of publication bias in the subsample of studies. Some of these studies did not find a positive association between antibiotic consumption and antibiotic resistance. Also, excluded studies that were introduced above are publication Bias.

RESULTS

Description of included and excluded studies

According to our investigation in PubMed, Science Direct, Google Scholar, Biological Abstracts, ISI web of knowledge and IranMedex we had 78 inclusions (research articles with full text and abstracts in English) and in end 26 exclusions (Review articles, congress abstracts, studies that reported in languages other than English, studies that were not available for us in abstract or full text, studies that their sampling location was uncertain, studies that locations of sampling was performed at the same time, studies that their data were not clear) for this meta-analysis [Figure 1].

Figure 1

Flow diagram for study selection

Samples information

In our study, 78 articles were reviewed from 33 different countries. The names of these countries are mentioned in Table 1.

Table 1

Data that were extracted from published documents about country, year, sample size, prevalence and M. tuberculosis Haarlem family MDR

The total population for prevalence of Haarlem family of M. tuberculosis that was included in this meta-analysis was 28,601. The major risk factors for M. tuberculosis infection were geographical areas and MDR-TB among Haarlem family. The typing method described in these studies was based on DNA polymorphism that exists at one particular chromosomal locus, the DR region, which is uniquely present in M. tuberculosis complex bacteria.

Prevalence of Mycobacterium tuberculosis Haarlem family in the worldwide population

In this study, 78 different studies with a total number of 28,601 persons from 2002 to 2012 were subjected. For this meta-Analysis, prevalence of M. tuberculosis Haarlem Family in worldwide population was analyzed according to the year of occurrence and MDR-TB. The highest rate of prevalence was in Hungary in 2006 (66.20%) with negative MDR-TB and the lowest rate was in China in 2010 (0.8%), also MDR-TB was observed in these studies (CI: 95%) [Table 1].

Year

The data for 11 groups were analyzed for the period of 2002-2012. The lowest rate of prevalence was in 2010, and stood at 6.42% (95% CI: 5.91-6.95%). But the highest prevalence was attained in 2012, and stood at 20.62% (95% CI: 17.48-24.05%). The heterogeneity, I2 = 97.7, heterogeneity Chi-squared = 383. 86 (d.f = 9) P < 0.00, estimate of between-study variance tau-squared = 0.01 and P = 0.00. Also results of publication bias presented in Figures 2-4 and Table 2.

Figure 2

Pooled sensitivity (confidence interval: 95%) and heterogeneity for Mycobacterium tuberculosis Haarlem family in worldwide population based on year (Method: Inverse variance method [Freeman Turkey double arcsine transformation])

Figure 4

Results of publication bias in for Mycobacterium tuberculosis Haarlem family in worldwide population based on year (confidence interval: 95%)

Table 2

Pooled sensitivity (CI: 95%) and heterogeneity for M. tuberculosis Haarlem family in worldwide population based on year

Forest plot of fixed (a) and random (b) effect estimates for pooled sensitivity and heterogeneity estimates for prevalence of Haarlem family of Mycobacterium tuberculosis in worldwide population based on year (confidence interval: 95%)

Multi-drug resistant tuberculosis

Multi-drug resistant tuberculosis prevalence analysis in our 78 previously mentioned studies, showed that 1.076% (95% CI: 10.30-11.23%) were positive MDR. But, prevalence for negative MDR was 9.22% (95% CI: 8.3-10.2%). Heterogeneity, I-squared = 79.4, heterogeneity Chi-squared = 9.69 (d.f = 2) P < 0.001, Estimate of between-study variance Tau-squared = 0.00 and P = 0.00. Also there is no publication bias [Figures 5 and 6, Table 3].

Figure 5

Pooled sensitivity (confidence interval: 95%) and heterogeneity estimates for prevalence of Haarlem family of Mycobacterium tuberculosis in worldwide population based on multi-drug resistant tuberculosis (Method: Inverse variance method [Freeman Turkey double arcsine transformation])

Figure 6

Forest plot of fixed (a) and random (b) effect estimates for pooled sensitivity (confidence interval [CI]) and heterogeneity estimates for prevalence of Haarlem family of Mycobacterium tuberculosis in worldwide population based on multi-drug resistant tuberculosis (CI: 95%)

Table 3

Pooled sensitivity (CI: 95%) and heterogeneity estimates for prevalence of Haarlem family of M. tuberculosis in worldwide population based on MDR-TB

DISCUSSION

Mycobacterium tuberculosis causes TB disease and possibly death.[3] The aim of our study was to determine the prevalence and rate of occurrence of MDR Haarlem family of M. tuberculosis in the worldwide using meta-analysis and a systematic review based on published articles in the period of 2002-2012. Molecular epidemiology fulfills to establish a dynamic transmission of disease. Results in our study showed that the lowest and highest rates of prevalence were in 2010 and 2013, respectively. 1.076% of samples were positive for MDR and 9.22% were negative for MDR. TB cases have decreased since 2005 and new cases since 2002. For example China has a % 80 decline in its TB mortality rate. 80% of the population in Asian and African countries has positive tuberculin test, but 5-10% of the USA population has positive tuberculin test. In 2007, Swaziland with 1200 cases per 100,000 people had the highest incidence rate of TB. India with 2.0 million new cases had the largest total incidence.[58384] In our study the lowest rate of Haarlem family prevalence was 6.42% in 2010 and the highest prevalence rate was 20.62% in 2012. TB morbidity has decreased worldwide but an increase in the incidence and prevalence of MDR-TB can be observed.[85] Also in our results the highest and lowest rates of occurrence of Haarlem family of M. tuberculosis in worldwide population were in Hungary in 2006 (66.20%) and in China in 2010 (0.8%), respectively. Risk factors that are important for considering when we talk about M. tuberculosis occurrence rate are poor living conditions, unhealthy work environments, crowded places, poverty, and lack of access to proper medical care, and factors that impair the host's defense against TB infection such as HIV infection, malnutrition, smoking, diabetes, alcohol abuse, and indoor air pollution. These factors can be effective on prevalence of M. tuberculosis family in different regions according to our study as compared to other studies.[8687] These data suggest that different factors such as spread of strains of MDR-TB (primary drug resistance) or the development of drug resistance in the drug-susceptible M. tuberculosis strains due to treatment default or failure (acquired resistance) in patients are important factors for MDR-TB existence. Durmaz et al. in Turkey showed that genetic diversity and major spoligotype families of drug-resistant M. tuberculosis from different regions of Turkey in clinical isolates was for T super-family (29%), Latin-American and Mediterranean (33.5%), Haarlem (14%), and the S lineage (3%).[88] Thus, genetic variation in TB plays an important role in the prevalence of the disease.[3] Based on these results genotyping of M. tuberculosis is an important factor for TB control. Several molecular techniques exist to identity M. tuberculosis such as variable-number tandem repeats of different families of genetic elements that are called mycobacterial interspersed repetitive units and polymerase chain reaction-based spoligotyping.[8990] In another study Marais et al. in South Africa by spoligotyping that they performed for drug susceptibility testing on 391 isolates; 12.5% were resistant to isoniazid, and 20% were resistant to both isoniazid and rifampin. In Marais study, Haarlem genotype family with 53.1% was significantly associated with drug resistance, so association between antibiotic resistance and prevalence of Haarlem family can be confirmed by spoligotyping.[8] The most common method that is used in clinical molecular epidemiology of TB is spoligotyping that is used extensively.[9192] This technique showed various profiles of Haarlem family of M. tuberculosis that were obtained from different populations.[93] Bazira et al. in Uganda by spoligotyping on 125 isolates revealed that 79 spoligotype patterns, with an overall diversity of 63.2%. Resistance mutations to either rifampicin or isoniazid were detected in 6.4% of the isolates. Multidrug resistance was seen in 1.6% of the cases.[94] Mutation in bacterium and diversion of the genotypes are largely in M. tuberculosis families and their resistance to drugs may be different among various families.[95] Research articles of our study were gathered for meta-analysis from different websites, including newest researches in PubMed, Science Direct, Google Scholar, Biological Abstracts and ISI web of knowledge in Thomsonreuters in regard to prevalence of TB in world health population. The difference in percentages in databases is logical based on published medical journals. Also we used the Endnote software for our search and deleted overlap studies. In this review, 78 studies were selected for analysis; therefore power of statistical analysis is great and association with drug resistance document in total published reports. Since Haarlem strains are largely transmitted in cities of different countries, causing health problems and outbreak of MDR-TB in populations. This family isolated from patients in different locations in Europe, Asia and USA, so its detection and prevention is necessary.[396] We hope results of this study can be effective on control and prevention of Haarlem Family of M. tuberculosis in world populations.

CONCLUSION

This study's results indicate the dynamics of transmission of Haarlem family and its association in drug resistance and they can improve our vision about TB in worldwide populations. Tracking and designing programs for controlling TB infection and MDR is necessary for public health because M. tuberculosis families can be transmitted between different populations in the world. TB infection has been decreasing in recent years, but our results showed that the prevalence rate in 2010 was more than other years that we surveyed about. Therefore, control of its transmission and drug-resistant patients are required. This meta-analysis provides comprehensive and reliable data on the prevalence and occurrence rate of M. tuberculosis Haarlem family MDR among different countries and may be helpful in prevention of its occurrence.

AUTHOR'S CONTRIBUTION

RR contributed in original idea and protocol, conception of the work, conducting the study, revising the draft, approval of the final version of the manuscript, and agreed for all aspects of the work. DR was statistical consulting. PS contributed in wrote and editing of this manuscript. SR contributed in wrote and editing of this manuscript conception of the work, conducting the study, revising the draft, approval of the final version of the manuscript, and agreed for all aspects of the work.