Mycobacterium tuberculosis of the Beijing Genotype in Iran and the World Health Organization Eastern Mediterranean Region: A Meta-Analysis

Introduction

In contrast to the assumption that all strains of Mycobacterium tuberculosis are equal in terms of transmissibility, virulence, and capacity to develop drug resistance, ¹ there is concern that the Beijing genotype may have a tendency for emerging drug resistance and may be spreading worldwide, probably as a result of increased virulence.^2–6

Having typical patterns of Insertion Sequence 6110 restriction Fragment Length Polymorphism (IS6110 RFLP), 24 loci Mycobacterial Interspersed Repetitive Units-Variable Number of Tandem Repeats (MIRU-VNTR), ⁷ and a unique Spacer Oligonucleotide Typing (Spoligotyping) patterns are distinctive features of the Beijing family isolates. ⁸

The Beijing tuberculosis (TB) family has its origin in China where it is the dominant type of M. tuberculosis but it has also been shown to have a global distribution. ⁸ It comprises ∼50% of the TB species in East Asia and 13% of global isolates. ⁹ These isolates have emerged as endemic and predominant genotypes in East and South Asia, South Africa, and the countries of the former Soviet Union. ⁴ Based on the classification by Glynn et al., there are four different ecological patterns for Beijing genotype TB: (1) endemic and not associated with drug resistance, (2) epidemic and associated with drug resistance, (3) epidemic but susceptible to drug, and (4) at a very low level or absent. ⁵

In a recent review on the global distribution of the M. tuberculosis Beijing family, prevalence data were reported from around the globe and the worldwide presence of it was clearly shown. ¹⁰ But even if the Beijing strain family is seen almost in every country and setting, studies of the frequency and characteristics differ quite significantly. In this article, we give information on Beijing TB in Iran and review the literature for the rest of the World Health Organization Eastern Mediterranean region (WHO EMR). It is noticeable; Iran is bordered with the Beijing TB high-prevalence countries such as Azerbaijan and Armenia. ¹¹

Only limited information is available from the Eastern Mediterranean region office (EMRO) countries, including 22 countries within the Middle East, West Asia, North East Asia, and North Africa. Nearly half of these countries are among the economically developing countries.^12,13 A review of the prevalence of M. tuberculosis Beijing genotype with special consideration for its drug resistance pattern may probably allow the development of better strategies for TB control in EMRO countries.

Methods

Results

Considering the average rates of Beijing isolates in patients with pulmonary and nonpulmonary TB between 2000 and 2015, the highest level in the Middle East was 10% and 9.0% in Iran and Pakistan, respectively.^6,14,15

Distribution of Beijing strains was also very different in the various regions of Iran.^6,14,16–19 For example, in the northwestern and western provinces of the country, prevalence has been estimated at 9–10%,^6,16 in the Khorasan province, located in the east of Iran, 7.1%, ¹⁷ and in Tehran, the capital of Iran, 5.0%. ¹⁸

Among the countries in West Asia, Saudi Arabia estimated the distribution of this subspecies at nearly 5%.^20,21 United Arab Emirates and Oman reported a similar prevalence of this strain of ∼4%.^22,23 It seems that countries in North Africa do not have a significant prevalence of this strain family. Somalia, ¹¹ Djibouti, ²⁴ Libya, ¹¹ and Tunisia ¹¹ reported the absence of this strain and in countries such as Morocco (0.8%) and Egypt (0.06%) this strain was estimated at less than 1%.^25,26

Meta-prevalence of Beijing TB strains was calculated to be 4% (CI 95%: 3–5) (Fig. 1). Significant heterogeneity across estimates of prevalence was observed (p ≤ 0.000, I ²  = 82.08%). The prevalence of Beijing strains was different based on detection techniques (meta-prevalence by spoligotyping = 4% vs. other techniques = 6%, p = 0.003) and years of study (meta-prevalence by or before the year 2000 = 2% vs. after year 2000 = 4%, p = 0.004).

OPEN IN VIEWER

FIG. 1.

Meta-prevalence of Beijing TB strains in EMRO, analyzed by Meta prop_one. EMRO, Eastern Mediterranean region office; TB, tuberculosis.

Meta-regression and subgroup analysis on prevalence of Beijing strains by techniques, TB types, countries, and year of study summarized in Table 1.

One of the issues to be discussed is the spreading trend of these strains in Pakistan where an increasing trend from 3% in 2003 to 8.9% in 2013 was observed.^15,27

In Pakistan and Iran, there was a significant relationship between Beijing strains and resistance to streptomycin (OR = 2.6, CI 95%: 1.4–6.5) ¹⁴ and of multidrug-resistant TB (OR = 3.0, CI 95%: 1.18–8.87).^6,28 In contrast to this, in a number of reports from Morocco, Oman, and Saudi Arabia, no relationship with drug resistance was found (p > 0.05).^21,22,25

Study of the age distribution of infected patients with the Beijing strain can be important in terms of the ongoing transmission. A review of studies reporting the age variable was performed, and it was seen that Beijing TB outbreaks in Pakistan were higher in younger age groups (p = 0.03, chi-square test). ¹⁵ However, in another study in Pakistan, no relationship was found between age and this strain family. ⁹

Discussion

Due to the lack of reported data, some of the countries could not be included in the investigation. The meta-prevalence of TB caused by Beijing family strains in the EMRO region was estimated to be 4.0% according to this meta-analysis. The full picture, however, is still uncertain since this estimation was based on studies in a total of 13 countries. It should also be noted that in the estimation, studies that had sample sizes of less than 30 and those in which information was not generally published were excluded. Moreover, in cases where the target group in a specific study was patients with drug resistance or specific groups such as inmates, they were excluded from the estimation of overall prevalence.

Reviewing the Beijing family proportion in TB, a considerable geographical trend was obtained among EMRO countries. North African countries were at an extremely low level in terms of infection with this strain.^{11,24–26,28} Moreover, the West Asia region including Palestine and Syria was at a very low level^29,30 and almost consistent with the estimation of North Africa. In the Southwest Asia, the average of this strain was estimated to be more^22–24,31 and in the Middle East, including Pakistan and Iran^{6,9,15–17,27,31–33} this strain was more widespread. Iran is neighboring with Beijing high-prevalence countries in the north, including Azerbaijan 73.9%, ¹¹ Armenia 46%, ¹¹ and Turkmenistan 45.1%. ³⁴ Nearby Russia to the north has a Beijing high prevalence with an average prevalence of 51.4%,^35–37 However, to the west, Iran shares a border with Turkey, which is among the countries with a very low reported prevalence of Beijing TB with an average of only 0.5%.^34,38,39

Among the reviewed studies, eight investigated the relationship between these strains and drug resistance.^{6,9,23,25,27,28,30,34} In five cases, statistically significant relationships were reported^{9,23,28,30,34} that were related to Pakistan and Iran studies. In other parts of the world, for example, Russia, Estonia, Azerbaijan, and India, a strong statistical relationship between the Beijing lineage and drug resistance was reported^{4,30,34–36,40,41}; whereas in China, the origin of Beijing strains is endemic, but it is generally not related with any drug resistance. ⁸ In a study by Ghebremichae et al., in Sweden (2010), the prevalence of multidrug resistance was significantly higher in Beijing strains than in drug-resistant non-Beijing strains, but the majority of the patients (75%) were from Asia. ⁴² It is not entirely clear whether the Beijing family specifically increases the likelihood of drug resistance in EMRO; however, this result has been acquired in a considerable number of studies.

Investigation of the time trend in each country was generally not possible due to the limited reports available. Only in Pakistan and Iran were several studies performed at different points in time, allowing for possible trends to be analyzed. Since the estimations related to Iran were obtained from different regions of the country and these estimations had significant differences at the same time, evaluating the time trend of infection with this strain was considered incorrect. However, in Pakistan, the trend of TB cases caused by this strain had increased in Punjab province (2003 to 2005, 3%; 2013, 8.9%).^9,27

Meta-regression of prevalence by years of study between countries found significant heterogeneity between prevalence of Beijing strains before year 2000 and after 2000.

It is probable that other factors also have synergistic effect on the heterogeneity of prevalence based on years of study, but the lack of sufficient studies did not allow multiple regression analysis to identify other related variables, for example, migration, race, or geographical factors as well as time-trend analysis.

Unfortunately, among the available studies, only six studies investigated the relationship between age and TB infection with Beijing family^6,21–22,27 and there was no possible meta-regression analysis. In the study conducted in Pakistan (2003–2005), younger patients were more likely to develop active TB with the Beijing strain. ⁹ Further, in another study from Saudi Arabia, odds of infection with this strain were higher in patients who were 21 to 40 years old. ²²

Conclusion

The meta-prevalence of the M. tuberculosis Beijing family strain was generally heterogenic in the WHO EMRO region, and considerable differences were seen between North African countries (with no or very few cases) and Middle East areas. Additional studies are needed to clarify the trends more generally, as well as for having a better understanding of the relationship between the Beijing family and drug-resistant TB in the region.

Compliance with Ethical Standards

There are no relationships or interests that can have direct or potential influence or impart bias to the work. This work has been approved by the regional ethics committee (code of 5/4/5375 in date of September 8, 2012) in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards in Tabriz University of medical science. The right of all patients was regarded. All patients had right to decide to the (identifiable) personal data gathered, to what they had said during a study or an interview. The researchers did not know the name of the participants, and their identification was blinded.