Background
Tuberculosis (TB) remains one of the leading infectious diseases causing significant morbidity and mortality worldwide. One third of the world's population is latently infected with
Mycobacterium tuberculosis [
1], of which about 10% may develop active disease at anytime [
2]. Although the infection of
M. tuberculosis usually results in pulmonary TB, other organs and tissues can also be affected, resulting in extrapulmonary or disseminated TB [
3‐
5].
Extrapulmonary TB (EPTB) is a significant health problem, as is pulmonary TB (PTB), in both developing and developed countries [
4,
6]. For example, in India, while 15-20 percent of the immunocompetent adult TB cases were EPTB, the rate of EPTB was increased to more than 50% among the HIV co-infected patients [
7]. In the Netherlands, the frequency of EPTB was found to be 15% among the eastern and central Europeans, 58.9% among the Somali, 36.6% among people of other African origins, and 44% among the Asians [
8]. The reported proportions of EPTB among all TB cases in other developed countries ranged from 12% to 28.5% [
4,
9‐
12].
Turkey is a developing country with a population of more than 70 million and about 20,000 notified new TB cases annually. The proportion of EPTB among all TB cases in Turkey had increased from 19.6% in 1996 to 32.5% in 2007 [
13]. However, the reason for such an increase remains largely unknown. A directly observed therapy short-course strategy was started as a pilot study in 2003 and it was implemented throughout the country in 2008. BCG vaccination has been routinely applied to all children. TB control is mainly done by dispensaries around the country, sanatorium hospitals, and university hospitals. In TB dispensaries, the diagnosis of TB is usually based on a combination of medical history, physical examination, chest-X ray, and microscopic examination of clinical specimens for the presence of tubercle bacilli. People suspected to have TB will then be refered to the university or sanatorium hospitals, where mycobacterial culture can be performed for confirmation of the diagnosis. A limited number of medical centers are able to perform accurate and rapid culture and susceptibility testing. All TB patients are treated free-off charge.
Malatya is the third biggest city in the eastern Anatolia region of Turkey. The TB incidence rate (28.5/100.000) in Malatya has been higher than the average of the country and the EPTB rate which is higher than the average of the country has increased steadily in recent years, from 33.3% in 2005 to 42.2% in 2007 [
13]. The present study was conducted to gain insight into the demographic and microbial characteristics of EPTB cases in Malatya, Turkey, thereby to extend the knowledgebase of EPTB based on which better TB control strategies can be developed.
Discussion
To gain a better understanding of the epidemiology of EPTB in Turkey, we analyzed demographic, clinical and microbial characteristics of 397 TB patients diagnosed in Malatya, Turkey between January 1st, 2001 and December 31st, 2007. About one fourth of the study subjects had EPTB. Different types of EPTB occurred at a varied frequency, with genitourinary and meningeal TB being the most commonly seen types. Age distribution was significantly different (P < 0.01) between PTB and EPTB, people aged 46 and above appeared to have an increased risk for having EPTB. Furthermore, the distribution of different forms of EPTB differed significantly among age groups (P = 0.03), resulting in significantly different age compositions for different specific types of EPTB. M. tuberculosis isolates obtained from EPTB cases were significantly more likely to have unique DNA fingerprinting patterns. Different from previous studies of EPTB, in our study, being female was not found to be a risk factor for EPTB. Furthermore, although it was not statistically significant, meningeal and bone and/or joint TB were more commonly observed among the male patients than among the female patients. In contrast, lymphatic, genitourinary, and peritoneal TB cases were predominant among females.
The previous reported proportion of EPTB among all TB cases in different countries varied between 12% and 25.8% [
5,
9,
11,
20]. Previous studies conducted in different regions of Turkey showed that the rates of EPTB cases among all TB cases ranged from 3.2% to 53.8% [
21‐
25]. EPTB was found in 25.9% of our study population. Furthermore, the distribution of different forms of EPTB has varied among studies conducted in different populations, including different regions of Turkey. In our study population, genitourinary and meningeal TB was commonly seen EPTB types including 27.2% and 19.4% of the EPTB cases, respectively. In contrast, in several earlier studies conducted in other regions of Turkey, the most frequently seen forms of EPTB were pleural TB [
21,
22], lymphatic TB [
23], and central nerves system TB [
26], respectively. A previous study performed on 85 culture-proven EPTB cases in Arkansas showed that bone and/or joint tuberculosis was the most common type of extrapulmonary tuberculosis (27.1%) followed by cervical lymphatic tuberculosis (17.7%) [
4]. Another study conducted on 480 EPTB cases (of which 76% were culture positive) in San Francisco showed that lymphatic TB was the most frequent form (45.1%) followed by bone and joint TB (15.6%) and pleural TB (14.3%) [
11]. Our study included only culture- confirmed EPTB cases because Turgut Ozal Medical Center mainly accepted patients whose diagnosis require more complicated procedure for TB diagnosis. In contrast, in the Turkish studies mentioned above, the EPTB cases were mainly diagnosed based on clinical and radiological findings. Methodological differences, such as the difference in the inclusion criteria mentioned above, can be one of the possible reasons for the reported variations in the distribution of different forms of EPTB across different studies. However, it was worth to note that in all the EPTB studies conducted in Turkey so far, TB meningitis was found to be one of the most frequently observed type of EPTB.
In previous studies, being female, non-Hispanic black, and HIV infected were found to be the independent risk factors for EPTB [
4,
11,
27,
28]. In our study population, HIV screening is not routinely done, therefore we did not have the information regarding the HIV infection status of patients. However, according to the reports of the Turkish Ministry of Health, there were only three confirmed HIV/AIDS cases in Malatya at the end of 2007 [
29]. Thus, it is unlikely that the high frequency of EPTB in Malatya was driven by high HIV infection rate in the population. The increasing EPTB up to 42% in 2007 in Malatya [
13] may depend on an increasing awareness of EPTB and increasing laboratory facilities in recent years. Additionally as it was indicated previously [
4]; the high incidence high dynamics of extrapulmonary tuberculosis may be specific to our study population.
Different from several earlier studies of EPTB [
4,
11,
12], our study did not find that female sex was statistically associated with EPTB. The failure to detect an association between female sex and EPTB in our study could be partly due to the low statistical power associated with the small sample size of the study. In addition, we were unable to identify cases that might have had concurrent pulmonary and extrapulmonary involvement due to lack of information about additional disease sites. This limitation may have also contributed to the failure to identify some previously risk factors of EPTB in our study. Nevertheless, our observation of the differential distribution of a given type of EPTB between men and women suggest the possibility that gender differentials in EPTB exist in the population of Maltya, Turkey.
We found that age distribution was significantly different (P < 0.01) between PTB and EPTB. More than 52% of the PTB patients were in the 16-45 years-age group, while the age distribution among the EPTB cases was bimodal. This observation, together with our observation that the clustering rate was significantly higher among the pulmonary cases than among EPTB (38.1% vs. 19.4% respectively, P < 0.01), suggests that most of PTB cases in this population might be largely related to ongoing TB transmission, given that the associated age group (16-45 years) is known as a high risk group for TB transmission [
2]. Different from our study, a previous study conducted in Antananarivo (Madagascar) found no significant differences in the clustering rates between PTB and EPTB groups [
30] while another study conducted in San Francisco, the United States showed that young age was an independent risk factor for nonrespiratory TB and only pleural TB among the EPTB was associated with the highest clustering rate [
11].
As found in our previous studies using pulmonary isolates [
15,
31], in the present study, the ill-defined T clade, LAM, and Haarlem were the most commonly found clades among both PTB and EPTB cases. A previous population-based study conducted in Arkansas, U.S.A. found that Beijing family strain infection is statistically significantly associated with EPTB [
32]. While the number of the Beijing strains in the current study is too small to be assessed for statistical significance, the majority (3/4, 75%) of Beijing strains identified in our study were related to EPTB is intriguing.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
SG collected the data, participated in the design of the study, interpreted the data and drafted manuscript. ZH conceived and designed the study, interpreted the data, and provided critical revision of manuscript for important intellectual content. MA conducted all the statistical analysis, interpreted data, and assisted in the manuscript preparation. MK contributed to the collection of study isolates, patient information and interpretation of the results. ZKA contributed to the isolate collection and clinical classification of study patients. RD contributed to the design of the study, interpretation of the data, drafting and revision of the manuscript.
All authors have reviewed and approved the final version of the manuscript.