INTRODUCTION

Tuberculosis (TB) is a granulomatous infectious disease that can involve all tissues and organs of the body. Although it is a preventable and curable disease, it still exists as one of the major health problems in the world (1). In Turkey, the incidence of extrapulmonary TB (EPTB) has been reported as 30-45% (2-4). EPTB frequently involves lymph nodes and pleura, and less frequently includes bones, joints, genitourinary system, skin and soft tissues. In recent years, there has been an increase in the number of EPTB cases (1,3,5-7). One should consider EPTB in the differential diagnosis of almost every infection, particularly in countries where TB is endemic. The ratio of EPTB to pulmonary TB (PTB) changes according to geographical, social, ethnic, and economic parameters (8-10). EPTB develops as a result of lympho-hematogenous dissemination of the primary infection and subsequent latency of the disseminated TB bacilli, which then may acquire reactivation in case of reduced body resistance or increased susceptibility. The disease may occur in any stage of life and may involve any organ (4-11). The latency period in different organs ranges from 6 to 600 months (4). It is not yet clear why TB bacilli show reactivation in the lungs in some cases and in other organs in other cases (4,8). Female gender, history of contact with TB, smoking and end-stage renal disease have been implicated as factors affecting reactivation in organs (4,12). Studies on rare EPTB cases are limited (13-19). In this context, different clinical courses of TB render diagnostic difficulties in terms of different involved organs and rarity of EPTB. Difficulty in diagnosis results in delayed treatment, therapeutic problems, and increased costs as well as increased morbidity and mortality. The purpose of this study was to determine the demographic and clinical features of rare EPTB cases, to show that possible complications of EPTB can be prevented by early diagnosis and therapy, and to underline that EPTB should be considered among other disorders in differential diagnosis.

METHODS

Study Design, Setting and Population

We designed a retrospective study of EPTB cases diagnosed and/or followed up in a tertiary training hospital for chest diseases and thoracic surgery. The study was approved by the Local Ethics Committee of the Institution and was conducted in accordance with the ethical principles stated in the Declaration of Helsinki. Since our study was retrospective, patient consent forms were not obtained. The data was authorized by the hospital’s academic committee, provided that the patient’s identity remained confidential. The study included EPTB cases followed up in our clinic. The criteria for exclusion were as follows: 1) Immunosuppression, 2) Presence of malignancy, 3) Pleural TB, 4) Lymph node TB, 5) Age under 18.

Data Source

The clinical and laboratory data of EPTB cases hospitalized in our clinic were obtained from patient files retrospectively.

Additional Covariate

The EPTB cases were evaluated in terms of age, gender, contact with TB, co-morbidities, concurrent PTB, history of contact with TB, tuberculin skin test (TST), radiological and bacteriological findings, diagnostic and therapeutic features and extrapulmonary organ involvement.

Definitions

There was at least one of the following diagnostic criteria in cases with definitive diagnosis of EPBT (20):

- Determination of the presence of acid-fast bacteria in the direct microscopic examination and/or culture of biopsy material obtained from extrapulmonary organs,

- Presence of necrotizing granulomatous inflammation in the biopsy material,

- Consistent with TB, TST positivity and response to anti-TB treatment,

- Clinical picture: Cough, expectoration, night sweats, weight loss, malaise, fever.

TST was accepted as positive when the induration diameter was ≥10 mm in patients without Bacillus Calmette-Guérin (BCG) scar, and ≥15 mm in patients with BCG scar (21).

Statistical Analysis

A descriptive analysis was performed to evaluate patient demographics and data.

RESULTS

The study included 50 cases with EPTB. Of the cases, 69% were female and the mean age was 34±15.8 years. Four cases (0.8%) had a history of TB and 16 (31%) had contact with TB. The demographic features of the cases are summarized in Table 1. All cases were human immunodeficiency virus negative and one case was hepatitis B surface antigen positive. Concurrent PTB was present in eight patients (16%). Three of eight cases had acid-fast bacilli-positive sputum and five of eight cases had positive culture for TB. Pulmonary radiology was normal in 23 patients. The mean duration of treatment was 9 months, including isoniazid, rifampicin, ethambutol, pyrazinamide (HREZ) regimen in the first 2 months and isoniasid, rifampicin (HR) regimen in the remaining 7 months. The most frequent symptoms were abdominal pain and weight loss followed by cough, fever, night sweats, expectoration, chest pain, and anorexia as expected in TB. The distribution of symptoms on admission is shown in Table 2. The diagnosis of EPTB was mostly histological. In 92% of our cases, EPTB was diagnosed histopathologically and in eight cases with concurrent PTB, smear/culture was also positive in addition to histopathological positivity. No culture was made from biopsy specimens. In our EPTB cases, gastrointestinal system was the most frequently involved system. The distribution of cases according to organ involvement is shown in Table 3.

DISCUSSION

EPTB causes diagnostic and therapeutic difficulties because it has different clinical courses in different organs. The consequences of this situation are delayed treatment and cost problems, as well as increased morbidity and mortality. In our study, most cases were female, there was a history of contact with TB in 28% of the cases, classical TB symptoms like weight loss, fever, night sweats were frequently present although the symptoms changed according to the involved organ, the most frequently involved system was the gastrointestinal system, and the most frequently used diagnostic method was histopathological examination. The incidence of EPTB has been reported to be 4.5-53% (2,3,5,9,22). This variable incidence of TB is related to geographical, social, ethnic, and economic parameters. The disease is more frequent in women than in men (4,6,9,10). Similarly, we had more female patients in our study. Although the role of gender in EPTB is not fully clarified, cellular immunity, hormonal changes and socio-economic and cultural status are thought to be related to this difference (12,23,24). EPTB is most common in the 30-58 age range (3,5,6,10,12). The mean age of our patients was consistent with this age range. Lin et al. (12) reported the prevalence rate of EPTB as 5.9% at ages ≤24 and as 54.9% at ages ≥60. This difference in prevalence may be related to changes in the immune system with aging. In their study on the association of smoking with TB, Kolappan et al. (25) reported that the rate of TB increased parallel to the number of cigarettes smoked and length of smoking period. On the other hand, EPTB is more frequent in non-smokers (4,26). Lin et al. (12) reported a negative correlation between smoking habit and EPTB. In our patient group, 72% were smokers, indicating a finding that might be related to high rate of smoking in Turkey.

An important clue in the diagnosis of EPTB is the history of patient’s previous contact with TB. Musellim et al. (4) reported an EPTB rate of 76.6% within 5 years after contact with TB. According to another study, there was no significant difference between PTB and EPTB in terms of contact history with TB (8.9% vs. 8.4%, respectively) (9). Two patients with breast TB in our study were sisters with a family history of PTB (27). Cases of PTB under treatment and included in the screening program for TB in the family should also be carefully examined for the possible presence of EPTB.

Co-morbidities such as long-term corticosteroid use, Chronic Obstructive Pulmonary disease, alcoholism, diabetes mellitus, chronic renal failure, malignancies, and immunosuppression increase the development of TB (28). Lin et al. (12) found that the presence of co-morbidity is insignificant for the development of EPTB and PTB. On the other hand, Gonzalez et al. (29) reported hepatic cirrhosis as a risk factor for the development of EPTB. In our patient group, there was only one case with diabetes mellitus. The symptoms and findings in EPTB vary according to the involved organ and concurrent PTB. In our patient group, the most common symptoms were cough, expectoration, weight loss, night sweats, and anorexia accompanied by organ-specific symptoms in most cases, consistent with data from relevant studies (3,6,9,12). The diversity of systemic or pulmonary symptoms leads patients to admit doctors in various medical branches, resulting in diagnostic difficulties and delayed treatment unless TB is suspected. Demiralay (9) reported that the time between the onset of symptoms and establishment of EPTB diagnosis was 154±39.2 days. The diagnosis time in EPTB is shortest in pleural TB and longest in skeletal system TB (9,29). This finding may be related to a higher incidence of pleural TB, which is suspected and therefore diagnosed earlier. Mycobacterium TB culture is the gold standard in the diagnosis of TB (1). The diagnosis of EPTB is more difficult than the diagnosis of PTB. Histopathological examination and culture of biopsy material are important for the diagnosis of EPTB (5,8). In our patient group, no culture was made from biopsy specimens, and this was related to the fact that EPTB was not considered in differential diagnosis in the preoperative period. The clinical suspicion of EPTB is an important step in diagnosis, leading to establishment of diagnosis by using a proper diagnostic method. Delay in diagnosis may result in serious morbidity and mortality. One of our patients with late testicular TB had additional vertebral and psoas involvement at the time of diagnosis (30). In our patient group, gastrointestinal system was the most frequently involved system. In a comparative study of PTB and EPTB on a total of 474 cases (48.5% EPTB and 51.5% PTB), Sreeramareddy et al. (11) reported that patients with EPTB had 42.6% lymph node involvement, 14.8% peritoneal and/or intestinal involvement, and 12.4% bone and/or joint involvement. Some rare EPTB cases had 7.2% miliary, 7.2% meningeal/brain, 4.8% skin, 2.9% genital and 2.4% laryngeal involvement (11). The rate of EPTB is variable (3,4,9-11). This variation may be related to social and environmental factors as well as to the center where the study is conducted. Our study was conducted at a chest diseases center where PTB cases are frequently referred. This may be the cause of our limited number of EPTB cases. The rate of co-existence of PTB and EPTB has been reported to range from 16 to 34% (3,6,11,12). This rate was 16% in our study. It has been reported that treatment longer than 9 months has no additional advantage, and that long-term treatment reduces patient compliance and increases costs (31,32). Our study has some limitations: The data were obtained retrospectively from patient files. Some patients lacked data on lifestyle. Thus, in some cases, it was not possible to study the effects of social, ethnic, economic, and environmental factors and therefore the reasons for delayed treatment. In addition, since this study was conducted on a small group of cases and at a tertiary hospital, it does not reflect the situation in the whole population. Prospective future studies are required to overcome these limitations.

CONCLUSION

In conclusion, as seen in our patient group, TB is a systemic infection that can involve any organ in the body. Delayed diagnosis and treatment of TB result in serious morbidity and mortality. We think that EPTB is overlooked in cases with no PTB. In cases without specific symptoms and findings, suspicion of EPTB is the most important step toward definitive diagnosis. In our patient group, patients exhibited organ-specific symptoms, frequently along with symptoms expected in TB. Particularly in female patients, the presence of these symptoms with a history of contact with TB and TST-positivity should lead the clinician to suspect TB even in the absence of PTB. The suspicion of TB will lead to early diagnosis with diagnostic algorithm and treatment that decreases costs, mortality and morbidity.

Ethics

Ethics Committee Approval: Ethics committee approved. (İstanbul Süreyyapaşa Chest Diseases and Thoracic Surgery Training and Research Hospital, 13.11.2015/7).

Informed Consent: Retrospective study.

Peer-review: External and internal peer-reviewed.

Authorship Contributions

Surgical and Medical Practices: S.G., M.Y., E.A., Concept: S.G., Ö.S., Design: S.G., B.B.A.D., Data Collection or Processing: O.A., E.A.Ö., Analysis or Interpretation: S.G., M.Y., Literature Search: S.G., E.U.B., Writing: S.G., M.Y.

Conflict of Interest: No conflict of interest was declared by the authors.

Financial Disclosure: The authors declared that this study received no financial support.