We have no real picture of EPTB because we don’t know what EPTB actually is. One of the reasons for incorrect estimation of epidemiology on EPTB is a misadjustment in terminology. Some authors consider EPTB as TB of any organ, excluding only broncho-pulmonary lesions, so pleural TB and broncho-pulmonal lymph nodes were related to the one of the form of EPTB. Others think that division of lung and its cover pleura on two separate organs is incorrect, and ascribe both organs to PTB and instead of EPTB use a term Extrathoracal Tuberculosis (ETTB), or extrarespiratory tuberculosis. WHO has considered an extrapulmonary case of TB as a patient with TB of organs other than only the lungs (e.g. pleura, lymph nodes, abdomen, genitourinary tract, skin, joints and bones, meninges). The question is – what about bronchial TB? As pleural TB is EPTB (formally it is, of course) – bronchial TB is an extrapulmonary form too? I think this position is absurd, and pulmonary TB (or more correctly – TB of breathing system, respiratory TB) should include a disease of the lungs, pleura and bronchi – and all other organs we have to attribute as extrapulmonary ones. This opinion has been totally accepted by the Russian Federation as well as in some other countries with extensive experience dealing with TB. The absence of a unique suitable definition, absence of a unanimous understanding of what really EPTB is, leads to confusing estimations of the proportion of forms of TB.

There was very unstable prevalence of EPTB in different times and regions. In 1984 EPTB remained a major health problem in Australia, where 24.3 % of all new TB notifications were of extrapulmonary origin (Dwyer et al. 1987). In the past century in Oklahoma a greater proportion of newly diagnosed cases of EPTB occurred in nonwhites (Snider 1975). In those days EPTB was frequent in Africa and was of great severity due to delayed diagnoses and multifocal forms (Aubry et al. 1979). In the 1980s in the city of Boston, EPTB represented 4.5 % of all new cases of active TB and tended to occur in older patients (Alvarez and McCabe 1984). In Spain EPTB increased from 30.6 % of cases in 1991–1996 to 37.6 % in 2003–2008 (García-Rodríguez et al. 2011). EPTB had an increasing rate in Turkey in 2001–2007. The reason remains largely unknown (Gunal et al. 2011). In 2009 in the USA, 73.6 % were PTB and 18.7 % were EPTB (Peto et al. 2009). In developed countries, from 2 – 10 % patients with PTB have also UGTB; in developing countries such proportion increases up to 15–20 % (Figueiredo and Lucon 2008). About 20 % of patients cured of PTB, had EPTB later, mostly–UGTB (Lenk and Schroeder 2001). We have to agree that incidence is changing depending on the country and time – but not so dramatically. I’m sure the main reason is a global mistake in the definition of the EPTB and different approaches to the diagnosis and confirmation of the disease.

In 1984 in Australia, the commonest sites of disease were the lymph nodes, urogenital tract, pleura and bone (Dwyer et al. 1987). In Oklahoma, the majority of EPTB forms were meningitis and lymphadenitis (Snider 1975). In the 1980s in the city of Boston, sites of involvement included lymph nodes, genitourinary tract, bone and articular sites, the meninges, peritoneum, adrenal glands, pericardium, and miscellaneous sites, in this order (Alvarez and McCabe 1984). In Spain in 2003–2008 TB lymphadenitis increased up to 27 % (García-Rodríguez et al. 2011). In Nepal common sites for EPTB were lymph nodes (42.6 %) and peritoneum and/or intestines (14.8 %) (Sreeramareddy et al. 2008). The most common types of EPTB in Turkey in 2001–2007 were UGTB (27.2 %) and meningeal TB (19.4 %) (Gunal et al. 2011). But other authors from the same region reported a little different data. Among 141 EPTB patients in Istanbul for 7 years, meningeal TB accounted for 23 %, TB lymphadenitis 21 % (Sevgi et al. 2013).

In 2009 in the USA, EPTB included lymphatic (40.4 %), pleural (19.8 %), bone and/or joint (11.3 %), genitourinary (6.5 %), meningeal (5.4 %), peritoneal (4.9 %), and unclassified EPTB (11.8 %) cases (Peto et al. 2009). In France in 2012 the most frequent clinical presentations of EPTB were lymphadenitis, pleuritis and osteoarticular TB (Mazza-Stalder et al. 2012). In some countries the rate of growth of bone&joint TB reached the leading position among EPTB (Kulchavenya et al. 2013a). Location of TB on the spine remains the most common form of skeletal TB, representing 62.2 % of all osteo-articulary locations (Didilescu and Tănăsescu 2012; Wiler et al. 2010).

In Bashkortostan, female genital TB (FGTB) prevailed in 1998–2006 years, but from 2007 the bone & joint TB was a leader among EPTB (29.5 %), then–FGTB (27.4 %), then – lymphonodal TB (17.9 %). The share of UGTB was 9.5 % only (Tuktamysheva et al. 2011). We support the idea to separate urological TB and gynecological TB (female genital TB) because it allows us to better estimate the epidemiology.

Throughout the world, high TB burden countries account for about 80 % of the world’s TB cases. Vietnam is one of such countries with the biggest prevalence of TB – 100 cases among 100,000 inhabitants (Do Chau Giang 2004). Among EPTB patients UGTB only was diagnosed in 77.2 %, combination of PTB and UGTB was found in 19.2 %, and UGTB and other forms of EPTB – in 3.5 % of patients (Nguyen Phuc Cam et al. 2009).

Navarro-Vilasaró et al. (2008) reported UGTB was the third most frequent EPTB infection, following pleural and nodal involvement in Spain in 2008. In the opinion of Abbara and Davidson (2011), UGTB is the second most common form of EPTB, with more than 90 % of cases occurring in developing countries. At the same time Goth and Joshi (2004) listed this form of EPTB as “others” due to low prevalence.

Within the last decade the spectrum of EPTB in Siberia has changed significantly (Kulchavenya et al. 2013b). TB of the central nervous system almost doubled from 4.9 to 8.7 %, mostly due to co-morbidity with HIV. Bone and joints TB increased by about half from 20.3 to 34.5 %, and among this group TB spondylitis with neurological disorders predominated. The proportion of UGTB decreased from 42.9 to 31.7 %. On the contrary, there was a decrease of peripheral lymph nodes TB from 16.7 % in 1999 to 11.2 % in 2011. At the end of the last century ocular TB accounted for 7.4 % and in 2008 (in 2009 listed in “others”) for 4.4 % of the patients with EPTB. Accordingly, in 1999 other forms of TB accounted for 7.8 % and in 2009 for 15.8 % (in 2011–13.9 %). The increase is partly due to inclusion of patients with ocular TB in this group, and partly due to better diagnosis of TB of the skin, abdominal organs, breast etc. (Kulchavenya et al. 2013a, b).

Fritjofsson and Kollberg (1973) analyzed the incidence of UGTB in Sweden, and found that the incidence showed geographical variation within the country. The same situation was in Hungary, where the number of new cases in one province was as high as 10 per 100,000 inhabitants per year, while in another part of the country only 2–3 cases had been reported. Unexplainable different prevalence of UGTB in the neighbor regions was found in Siberia too (Kulchavenya and Krasnov 2012)–multiple predominance of the number of EPTB cases in one regions relatively another in 20–50 km – and there is no idea for the reason of this phenomena.

The first note of urogenital TB was made by Porter in 1894 [2]; in 1937 Wildbolz [3] suggested the term genitourinary TB c The term urogenital TB is more correct, because kidney TB, which is usually primary, is diagnosed more often than genital TB. Actually the term “urogenital tuberculosis” is incorrect too as it collects many forms with its own clinical features and requiring its own therapy and management. UGTB joins kidney TB, urinary tract TB, male and female genital TB – and every form has its own features. The combined term “UGTB” does not allow estimation of a real epidemic picture – is it prevalence of male genital TB or kidney TB? Also a high incidence rate of kidney TB 1–2 stages is better than a low incidence rate, except when complicated forms are revealed.

Some authors have found that UGTB affects more men than women (Figueiredo and Lucon 2008; Benchekroun et al. 1998; el Khader et al. 1997; Tanthanuch et al. 2010; Nurkić 2006); others – exactly the contrary (Mazza-Stalder et al. 2012; Snider 1975; García-Rodríguez et al. 2011; Singh et al. 2011). It seems that renal TB, alike any other kidney disease, should be found more often in female patients, because menses, gravidity, and inflammation of female genitals may hinder the urine passage. Urinary stasis carries a possibility for fixation of M.tuberculosis to urothelium, and, so, for developing renal TB.