January - March 2005: 
Volume 18, Issue 1

Click on the image to download the Issue in PDF format.


Changes in the epidemiology of tuberculosis in Greece due to continuing immigration
Mass immigration from developing countries to industrialized countries entails the risk of spreading resistant strains of mycobacteria, which is associated with serious epidemiological, social and economic consequences. In the face of continuing large-scale immigration to Greece, we surveyed the number of new bacteriologically confirmed cases of tuberculosis; single resistance to isoniazid or rifampin; and multi-drug resistance, i.e. resistance to at least isoniazid and rifampin, of M. tuberculosis strains that were isolated from the first culture of sputum of immigrants in the period 1993-2002. To allow better insight, immigrants were separated to foreigners coming from South-Eastern Asia and Africa, and Greek immigrants to the Former Soviet Union who returned to their country. Findings were compared to relative data from native Greek patients in the same period. The number of newly diagnosed tuberculosis cases in native Greeks and returning Greek immigrants is constantly decreasing, probably due to higher living standards in Greece, whereas TB incidence among foreigners coming from South-Eastern Asia and Africa is rising, which may be attributed to the introduction of more effective surveillance systems in our country. Resistance of M. tuberculosis strains to isoniazid and rifampin is increasing in all three groups, most significantly in repatriated Greeks. Multidrug resistance also shows a significant increase in all three groups, particularly in repatriated Greeks. Pneumon 2005, 18(1):74-83.
Full text

In recent years, a considerable number of new tuberculosis cases are identified among immigrants moving to industrialized and economically robust countries; hence, there is a high risk of spreading resistant strains of mycobacteria in the countries of destination.1

Due to the ongoing inflow of economic immigrants to our country, many of whom come from countries known to have a high incidence of tuberculosis, we surveyed the number of new bacteriologically confirmed cases of tuberculosis; resistance of isolated M. tuberculosis strains to isoniazid or rifampin, the first-line antituberculosis drugs; and multi-drug resistance, i.e. resistance to at least both isoniazid and rifampin. M. tuberculosis strains were isolated from the first culture of sputum of immigrants in the period 1993-2002. Findings were compared to relative data from native Greek patients in the same period.


The material of our study constituted 4,108 M. tuberculosis strains, recovered from the first culture of sputum of the same number of patients. To allow for a more meticulous investigation of M. tuberculosis resistance patterns, the population of the immigrants was divided to foreigners coming mainly from Southeastern Asian countries (e.g. Pakistan, Bangladesh, Iraq) and, to a lesser extent, Africa, and to those who had Greek descent and represented the comeback of emigrants to East Europe, mainly the Former Soviet Union. Of the total 4,108 patients included in the study, 3,368 (82%) were native Greeks, 551 (13.6%) were foreigners and 189 (4.6%) were Greek emigrants who returned to Greece. Microbiology studies were performed in the National Reference Center for Mycobacteria (NRCM). NRCM is located in "Sotiria" Hospital and receives samples from patients hospitalized in either public or private hospitals or outpatients attended by community doctors, which are tested using the appropriate microbiology methods for tuberculosis. In case of a positive test result, the clinical doctor has to complete a special tuberculosis report form, which includes a set of clinical, laboratory and epidemiologic data of the patient. The Committee for the Investigation of Hospital Infections receives the form and has the responsibility to send it to the Center for Control of Special Infections (CCSI). NRCM draws all relevant information from these forms.


Microbiology testing included microscopic examination of the samples; performance of standard and newer culture techniques, such as the semi-automated radiometric method Bactec 460 (Becton-Dickinson) and the automated fluorescence method Bactec MGIT 960 (Becton- Dickinson); identification of strains based on their biochemical properties and the use of molecular biology techniques (Accuprobe-Gen Probe, InnoLipaa Mycobacteria- Innogenetics); and, finally, susceptibility testing. Susceptibility testing was carried out using the resistance ratio method in Löwenstein-Jensen medium, both solid and liquid, by the semi-automated method Bactec 460 and the automated fluorescence method Bactec MGIT 960. Susceptibility testing included all first-line antituberculosis drugs; nevertheless, we focused on isoniazid and rifampin, due to their significance relating to the treatment of tuberculosis. Multidrug resistant strains were defined based on resistance to at least both isoniazid and rifampin.2


Of the total 4,108 study subjects, native Greeks constitute the great majority of the patients, since they come up to 3,368 representing 82% of the total patient population. Foreigners and Greek immigrants returning home were definitely fewer, coming up to 551 (13.6%) and 189 (4.6%), respectively (Table 1). The total number of new cases diagnosed each year is on a slight decline. The number of native Greeks is significantly higher in all years of the study, since they come from a tenfold larger source population. Nevertheless, new cases of the infection show a clear progressive decrease (Figure 1). In 2002, native Greek patients comprised 56% of the native Greek patient population in 1993; in terms of absolute numbers, the number of patients decreased from 474 to 209 during this time. However, around the middle of the period from 1993 to 2002, i.e. the time of the peak inflow of immigrants in our country, the decline in the incidence of tuberculosis in native Greeks slowed down. In addition, the incidence of tuberculosis in Greek emigrants who returned to Greece fell by 75%; 47 new cases were diagnosed in 1993, but only 12 in 2002. In particular, in 1993, native Greeks, foreigners and repatriated Greeks constituted 85%, 6.5% and 8.4%, respectively, of the total patient population, whereas in 2003 native and repatriated Greeks constituted 68.5% and 3.9% of the cases, with the incidence of tuberculosis in foreigners increased to 27.5% (Table 1, Figure 1).

Figure 1. Graphical representation of new cases of bacteriologically confirmed TB in native Greeks, foreigners, and repatriated Greeks from Eastern countries in the period 1993 - 2002.

In relation to resistance, in 1993, 30/557 (5.3%) patients were susceptible to isoniazid and 6/557 (1%) were susceptible to rifampin, with 5/557 (0.9%) patients presenting multidrug resistance (i.e. resistance to isoniazid and rifampin). In 2002, these figures increased significantly, coming up to 35/305 (10.8%), 22/305 (7.2%) and 18/305 (5.9%), respectively (Table 2, Figure 2).

Figure 2. Graphical representation of comparative data on resistance of M. Tuberculosis strains to isoniazid, rifampin, and multidrug resistance in native Greeks, foreigners and repatriated
Greeks form Eastern European countries in the years 1993 and 2002

Resistance to isoniazid in native Greeks doubled from 4.6% to 10.5% in the period from 1993 to 2002, in foreigners it remained relatively constant (9.5%), but in repatriated Greeks it quadrupled from 10.6% to 41.7%. In relation to rifampin, resistance in native Greeks became 7 times higher (5.2%) in 2002, in foreigners it tripled (9.5%) and in repatriated Greeks rose to 12-fold the resistance rate in 1993. Lastly, in 2002, multidrug resistance in native Greeks was 8 times higher, reaching 4.8%, in foreigners it doubled (6%), and in repatriated Greeks it was 12 times that in 1993 (25%) (Table 3, Figures 3a, 3b, 3c).

Figure 3a. Graphical representation of M. Tuberculosis strains resistant to isoniazid (INH) recovered from native Greeks, foreigners and repatriated Greeks from Eastern Europe in 1993 - 2002.
Figure 3b. Graphical representation of M. Tuberculosis strains resistant to rifampin (RIF) recovered from native Greeks, foreigners and repatriated Greeks from Eastern Europe in the period 1993 - 2002.
Figure 3c. Graphical representation of M. Tuberculosis strains with multidrug resistant (INH+RIF) recovered from native Greeks, foreigners and repatriated Greeks from Eastern Europe in the period 1993 - 2002.


Tuberculosis, one of the oldest diseases, remains a serious concern for the medical community in view of its consequences for public health. In the last decades, statistics from developed countries constantly conveyed promising data for a progressive decline in the incidence of the disease. However, in the mid-80s', the global optimistic scenery was reversed, since a gradual increase in both the incidence of new cases and M. tuberculosis resistance to first-line antituberculosis drugs occurred. In 1990, 7.5 million cases of tuberculosis were reported worldwide and 2.5 million people died of tuberculosis; according to projections, by the year 2000 deaths would have exceeded 30 million in total. Therefore, in 1991, the World Health Organization (WHO) considered appropriate to raise the alarm concerning the disease.1-5

It is generally accepted that the reversal of declining trends in the burden of tuberculosis is the result of two main epidemiological influences: the increasing prevalence of HIV infection and the mass immigration from countries where tuberculosis is prevalent. The old postulate that migration has a strong influence on the epidemiology of infections sounds very apropos in our time.1,5-7 In all industrialized countries, tuberculosis (TB) and its association with migration are currently being increasingly debated. In fact, in developed countries there is a considerable discrepancy in the incidence of the disease between immigrants and native population.1,7,8 In the US, 40-50% of new TB cases occur in foreign-born persons; the percentage of foreign-born persons among TB cases is 80% in Australia and exceeds 50% in Europe. In particular, in 1993 in the US 70% of the 25,313 TB cases were foreigners.5,6 In addition, quite interestingly, 4.4% of the total M. tuberculosis strains isolated in 1995 in Spain came from immigrants, whereas this percentage reached 24.2% in 2001.1

The increased rates of M. tuberculosis resistance to isoniazid or rifampin, as well as multidrug resistant TB account for the most part of the seriousness of the problem. It has been shown that resistance to isoniazid in the countries of origin is about 30%.1,3,6,8 Globally, the countries with the higher TB prevalence and the higher incidence of M. tuberculosis resistance are countries in South-eastern Asia, the region of South Sahara and South Africa, and, in Europe in particular, countries in Eastern Europe and the Former Soviet Union (FSU). Both the prevalence and the mortality of the disease are high in these areas.3,6,8 In the FSU, specifically, where many of the patients in our study come from, an outbreak of tuberculosis occurred in 1991; in the period 1991-1994, the number of new TB cases increased by 53.7% and TB associated mortality doubled from 4.4 to 9.8 every 100,000 people. In addition, primary resistance to one drug reached 24% and primary mulridrug resistance 5.1%; secondary resistance was found in 33% of the cases.8 The issue of resistant TB is an additional burden in the countries of destination, because, apart from the adverse consequences in the epidemiology of the disease due to the spread of resistant strains in the community, it is associated with higher cost of care. The treatment of TB patients who have resistant TB is costly, difficult and long.5 In the neighboring country of Israel, a significant number of TB patients (45%) come from the FSU. In immigrants from the FSU, resistance to isoniazid is 30%, while in the general population of the country it is 10%; resistance to one drug and multidrug resistance in the same immigrant population are 50% and 17%, respectively, whereas the relevant figures in the general population are 29% and 8.5%, respectively.5,9

It has been established that 90% of TB cases among immigrants represent endogenous reactivation of latent infection.2 This has been shown using molecular biology techniques, mainly restriction fragment length polymorphisms (RFLP), which provide highly reliable evidence relating to the epidemiology of the infection. In certain areas, however, such as New York, a significant percentage of patients, probably as high as 30%, are newly and exogenously reinfected. Several of these patients are HIV positive.2,6,10 One in ten patients in the US have turned out to be exceedingly virulent and responsible for transmitting the infection to many susceptible persons exposed to them.2,5,6,10

According to the above discussion, the possibility of person-to-person transmission of the disease is considerable. Hence, there is room for an optimistic outlook, since stopping transmission is easier than treating the disease. Besides, it is common knowledge that early diagnosis is crucial for prevention. Therefore, great responsibility lies in state agencies; health authorities should be alert and surveillance programs should be implemented. 7 In fact, in the US, the Center for Diseases Control (CDC) has recommended strict surveillance strategies for the detection of undiagnosed active tuberculosis, and past infections who have not been appropriately treated, as well as the identification of persons with positive tuberculin skin testing. In addition, all immigrants are required to receive appropriate treatment for active asymptomatic tuberculosis, which may have eluded diagnosis. 1,5,6 Notably, the current literature highlights the need for developing empiric treatment regimens on the basis of the epidemiological characteristics of the disease in the country of origin.6,9 In the event resistance to isoniazid exceeds 4%, empiric regimens should indispensably include four first-line antituberculosis drugs, i.e. isoniazid, rifampin, pyrazinamide and ethamboutol.8 The imposition of such strict measures is expected to have beneficial effects on public health and the economy of developed countries. It is estimated that the implementation of such a strategy in the US for a year would contribute to the prevention of the activation of 9,000-10,000 cases of latent tuberculosis, thus saving 60-90 million dollars.6

Compliant completion of treatment constitutes an additional challenge. To this end, in the US, as well as in other well-governed countries, directly observed therapy systems (DOTS) are being introduced and gradually diffused, with excellent results. DOTS entail direct observation of compliance to treatment, so as to ensure that the prescribed drugs are properly taken; continuous and reliable drug supply; and an efficient recording system. In New York, DOTS were applied in 100 patients in 1991; however, this figure rose to 1200 in 1993. In the same period and in the same city, the number of new tuberculosis cases, excepting old persons and immigrants, fell to 15%.2,5 Besides, the policy recommended by WHO for the containment of the disease relies on the very same principles, i.e. development of systems for the early detection of new cases, compliant completion of prescribed treatment, and, last but not least, comprehensive processing of epidemiologic profiling data. In this context, Korea sets an excellent example as regards the implementation of a well-organized prevention program that yielded spectacular results.3

Nevertheless, the containment of tuberculosis is not only a matter of early diagnosis and effective treatment. Several additional factors are also critical, i.e. conditions of social life, workplace conditions, dense population, poor hygiene, alcoholism, and HIV infection.6 According to data from Israel, in the period 1993-1997, multidrug resistance among immigrants from FSU was 17%, whereas in the US in the same period multidrug resistance among immigrants from the same country of origin was 3.5%, probably due to the better living conditions in the US.5,9 In Russia, apart from poverty and alcoholism, the observed high rates of resistance to isoniazid and multidrug resistance are primarily attributed to dense habitation, since for the most part of the population a room is used by four persons.8

Nevertheless, there are still some reservations regarding the extent to which immigration alters the epidemiology of tuberculosis in the countries of destination, due to the lack of indisputable evidence. In fact, in Germany, contemporary epidemiological studies using RFLP analysis showed that, apart from erroneous therapeutic manipulations, increased resistance rates are attributed to the spread of resistant strains carried mainly by immigrants from the FSU. Conversely, in Spain and Denmark, epidemiological studies indicated inconsiderable changes in the epidemiological characteristics of tuberculosis in the general population due to immigration.1,10,11

The results of the present study, both absolute numbers and increases in percentages, are particularly important and should be evaluated as such by physicians, health care authorities and the state, in general. Although the exact number of immigrants in the study period is not known, not only due to incomplete records, but also due the great number of illegal immigrants, presented data are considered sufficient to mobilize all competent agencies. As shown in presented tables and figures, the number of new tuberculosis cases decreased significantly from 1993 to 2002 among native and repatriated Greeks, probably due to improvements in the social and economic status of Greek people. Conversely, in this 10- year-period there is a significant increase in tuberculosis cases among foreigners, that is attributed to the better screening and recording system health authorities have applied to economic immigrants the last few years (Table 1, Figure 1). The slow down in the rate of decline of the incidence of tuberculosis in native Greeks around the middle of this period is probably a result of both a peak in the immigration flow that occurred that time and the then lack of an effective screening system for immigrants. The institution of systems for granting residence permit and providing care to immigrants that followed has led to a continuing reduction in new TB cases in both native and repatriated Greeks. This is, in part, also explained by the incorporation of repatriated Greeks in the Greek society, including their easy access to the national health care system through social insurance coverage. Conversely, the number of new cases in foreigners is constant, presumably due to their cut off life, poor living conditions, and constant replenishment of this population, since new immigrants tend to take the place of those who return home.

Rates of resistance to isoniazid or rifampin, and multidrug resistance in the general population in 2002 were 10.8%, 7.2% and 5.9%, respectively (Table 2, Figure 2), which are definitely high, compared to industrialized countries in Western Europe. In fact, according to data from Euro TB, a TB recording system running in 52 European countries that are members of WHO, our resistance rates are twice those in Spain, where in 2001 resistance to isoniazid was 6.3%, resistance to rifampin 3.9%, and multidrug resistance 3%. During the same year, respective rates in France were 4.4%, 1.3% and 1.1%, in Germany 7.6%, 2.9% and 2.7%, and in Austria 3.8%, 1.1% and 0.8%. Italy had resistance rates similar to ours in 2001, i.e. 11.5%, 5.7% and 4.2%, respectively.12

Dividing our study population into sub-groups according to country of origin and immigration status, repatriated Greeks were found to have strikingly higher resistance rates; in 2002, resistance to isoniazid, rifampin and multidrug resistance were 41.7%, 25% and 25%, respectively (Table 3, Figure 3). These figures are similar to those reported in the FSU, the country most repatriated Greeks came from. In particular, according to Euro TB data, resistance to isoniazid in 2001 in Georgia was 54.2%, in Estonia 37.4%, and in Ukraine 24%; resistance to rifampin during the same period in these countries was 25.7%, 27.6% and 21.1%, respectively; and multidrug resistance 24.8%, 27.2% and 17.8%, respectively.12 Resistance to isoniazid, to rifampin and multidrug resistance was also significantly increased in native Greeks;from to 1993 to 2002 there was a 2-, 7-, and 8-fold increase in resistance rates, respectively. Increases in resistance rates among foreigners were smaller (Table 3, Figure 3).

Increased resistance rates and multidrug resistant TB in our country are probably imported (primary resistance) from the countries of origin of immigrants, where resistant TB is prevalent. However, it certainly may also be iatrogenic, i.e. caused by inadequate treatment regimens (secondary resistance). Common causes of inadequate treatment are failure to administer a four-drug regimen in the first two months of treatment, early withdrawal of ethambutol before results regarding susceptibility to firstline antituberculosis drugs are obtained, and three-, instead of nine-month treatment in case of cavitary disease with bacteriologically positive sputum in the end of the first two months. It is likely that the institution of regulations requiring isolation of new cases, especially those infected with multidrug resistant strains, in appropriate special clinics until sputum smears were found negative on three occasions, would facilitate effective management of the disease. However, this is a sensitive and complex issue, which requires collaboration of health care professionals with jurists and sociologists if an integrated approach is to be approved by competent agencies. Furthermore, there is currently no policy dictating directly observed treatment (DOT), not even in cases with multidrug resistant TB. Guidelines regarding chemoprophylaxis for immigrants with a positive standard Mantoux test are also lacking, although most of them are young. It should be remarked, however, that costs for diagnosis, hospital care, treatment and follow-up of immigrants with TB are covered by public funds, which is a credit to our country.

The alarming findings of our study should prompt decision makers to institute integrated systems for epidemiologic data collection and processing, and for the implementation of preventive measures and the use of appropriate empiric treatment regimens. In addition, regulations requiring isolation of bacteriologically positive cases should be instituted. It is furthermore required that directly observed treatment in cases of multidrug resistant TB, as well as difficult to treat cases be introduced; compliant completion of treatment should be monitored. Special attention should be drawn to adequate chemoprophylaxis for immigrants, in view of their young age in most of the cases. Pulmonologists should be updated and alert, so that they can adapt to the new reality. Moreover, hospitals should allocate rooms with efficient ventilation systems for compulsory isolation of resistant TB cases.

It is clear that the risk of resistant M. tuberculosis strains spread in our country is real and high. Our study has not included an epidemiologic investigation using molecular techniques to indicate that increased resistance rates in our country have resulted by the spread of resistant strains carried by immigrants. This will be the focus of our next project. Until then, however, measures to effectively control the disease should be implemented and competent agencies should be enhanced with both human resources and adequate funds to succeed in their task.