Extra pulmonary tuberculosis: A retrospective study in Matara District, Sri Lanka

Background: Despite improvements in prevention, diagnosis and treatment, tuberculosis continues to be a leading cause of death globally. Since extra pulmonary tuberculosis does not contribute significantly to the transmission of the disease, it is not given priority in public health agenda. But it contributes significantly to tuberculosis related morbidity and mortality. Objectives: To identify the various presentations of extra pulmonary tuberculosis in Matara district, Sri Lanka and to evaluate the response to anti tuberculosis treatment. Methods: A descriptive study was conducted among the patients registered in District Chest Clinic, Matara. Data analysis were done by using SPSS version 25. Chi-square test was used to assess the association be tween variables. Results: Among 147 patients 50.3% were female (mean age = 44years). 93.2% were Sinhalese. 11.6% of the study population were from Matara municipal council MOH area. 32.7% of the sample was diagnosed to have pleural effusion. 9.5% were current smokers. 9.5% were diagnosed with diabetes.98% of patients had completed treatment. Conclusions: There was high prevalence of pleural tuberculosis. The diagnosis of lymphatic tuberculosis was lower. Prevalence of extra pulmonary tuberculosis was high in MOH areas with high population density. Significantly high treatment completion rate than the national target was noted at district chest clinic Matara.


Introduction
Tuberculosis is an infectious disease, which is air born and caused by the bacillus Mycobacterium tuberculosis. (Occasionally by Mycobacterium bovis and Mycobacterium africanum). 1 Tuberculosis can involve any part of the human body other than nails, teeth and hair, with lung being the most common. Extra pulmonary tuberculosis refers as any bacteriologically confirmed or clinically diagnosed case of tuberculosis involving organs other than lung parenchyma or tracheobronchial tree. 2 Diagnosis of extra pulmonary tuberculosis should be based on smear/culture positive specimen or histological or strong clinical evidence consistent with active extra pulmonary tuberculosis ,followed by a decision by a clinician to treat with a full cause of anti-tuberculosis chemotherapy.1A patient with pulmonary and co-existing extra pulmonary tuberculosis should be classified as pulmonary tuberculosis. 2 Presentation of extra pulmonary tuberculosis may be atypical and symptoms may depend on the affected anatomical site.Usual anatomical sites of extra pulmonary tuberculosis arebones and joints, central nervous system,genitourinary system, tuberculous lymphadenopathy, abdominal tuberculosis,ocular tuberculosis, ttuberculosis pleural effusion, miliary tuberculosis, pericardial tuberculosis, tuberculosis involving any other organs. 1 Presentation of extra pulmonary tuberculosis may be atypical or relatively insidious and tuberculosis may not be considered initially in differential diagnosis. This is an important phenomenon as delay in diagnosis may be crippling or even life threatening. Therefore, from a public health perspective, there is a need to address this group of patients as do they contribute to the total burden of the disease and they do have a significant impact on resources. The study included all cases of extra pulmonary tuberculosis of all age groups and both sexes treated at District Chest Clinic Matara.
Patients who were diagnosed as extra pulmonary tuberculosis and commenced treatment at other district chest clinics and transferred to complete treatment at District Chest Clinic Matara and ppatients who died of tuberculosis or its complications while on treatment irrespective of the period of treatment given were included in this study.
Patients who were not residents of the Matara district, but any how diagnosed at this clinic, were transferred out to the relevant district chest clinic of their own resident area to complete anti tuberculosis chemotherapy (transferred out), patients who were initially diagnosed as extra pulmonary tuberculosis but became sputum positive for Acid fast Bacilli culture/Gene expert (a molecular diagnosis technique which identifies the DNA molecule of the Mycobacterium tuberculosis)or for both during the period of treatment (Any case of pulmonary tuberculosis) and patients who died not due to tuberculosis or its complications while on treatment irrespective of the period of treatment,were excluded from the study.
The ethical clearance was obtained from the ethical review committee, University of Ruhuna. The study sample was 147 (70 patients from 2017 and 77 patients from 2018). Data collection was done using SPSS data sheet. Data was collected from patient's record according to the tuberculosis registration number which was given by District Chest Clinic, Matara.
Missing data in the records were traced over the phone as patient's records mentioned their active telephone numbers. There was no need for visiting patients or getting the help of public health inspectors.
Before gathering missing data from patients over the phone, informed consent was obtained verbally. Data was analysed using SPSS statistics 25. Chi-square test was used to assess the association between variables.

Results
The study group contained total number of 147 patients who were registered as extra pulmonary tuberculosis in District Chest Clinic, Matara. Table  01 showed the distribution of sociodemographic data among study population. The mean age was 44 years and standard deviation was 18.006. 50.3% were female and 93.2% were Sinhalese.

Table 02-Distribution of anatomical site involved in extra pulmonary tuberculosis
When considered the MOH area wise distribution,11.6% cases were in Matara Municipal Counsil area followed by Weligama (10.9%). Lowest number of cases were found in Kirinda MOH division. (Table 03)

Table 03-Distribution among MOH areas
Prevalence of co-morbidities among the study population was distributed as in table 04.   Correlations between co-morbidities and treatment outcome was cross tabulated and there was no significant association was found. Correlations between risk factors and treatment outcome was notfound in the study population.

Discussion
In Matara district, in the year 2017 there were 72 (32%)extra pulmonary tuberculosis cases registered out of 225 (total number of registered TB patients) and, In 2018, 81 cases registered out of total 216 (37.5%).3 147 extra pulmonary patients were included and analysed in this study which is 33.3% out of total tuberculosis cases and it is compatible with the national figures (25%-35%). 1 The mean age was 44 and standard deviation was 18.006. 95 % of cases are in between the age 8 to 80 (mean +_ 2SD). This is compatible with national figures which showed highest burden of the disease is in the most economically productive age group of our society (15-54 years). 2 There was no gender predominance offemale (50.3%) as compared to male (49.7%). When considered the ethnic groups, 93.2% of cases were Sinhalese, 4.8% cases were Sri Lankan Moor while 2% were Tamil which is compatible with Matara district census-20124( Sinhalese 94.16%, Sri Lankan moor 2.94%, Tamil 2.86% ). 8 patients (5.4%) had got con-tact history of tuberculosis in the past and 2 patients (1.4%) had got past history of tuberculosis. When considered the distribution of anatomical sites, most common was pleural effusion (32.7%), followed by lymphatic system (23.8%). This distribution is not compatible with national figures where the commonest form is lymphatic system which is 35% followed by pleural effusions which is 20%.1 This revealed increased diagnosis of pleural tuberculosis by using advanced procedures such as thoracoscopy in the District Respiratory Unit.
The occurrence of tuberculosis in any other organ has shown significant variability in Sri Lanka as well as globally.1This study revealed third common site as ocular tuberculosis (10.2%) followed by central nervous system (8.8%), bones and joints (6.8%) and abdomen (6.8%), genitourinary tract ( There was only one HIV positive patient in the study group (0.7%) which is higher than the national figure which is< 0.1%. There were 9.5% patientswith diabetics in the study group which is slightly higher than the national figure which is 8.5%. There were no rheumatoid arthritis patients in this population. There were 2.7% chronic kidney disease patients in the study group which is higher than national prevalence which is 1.3%.There were 1.4% patients with malignancy which is slightly higher than the national figure which was 1.1% . There were 2% patients with chronic respiratory diseases which was less than the national figure which was 6%.There were no correlations foundbetween co-morbidities andtreatment outcome.
79.6% were non-smokers while 10.9% had smoked previously and 9.5% have been smoking currently. 74.8% of them had not consumed alcohol while 10.9% had consumed previously and 14.3% have been consuming alcohol currently. There were no patients with other substance abuseamong both genders.There were no correlations found between risk factors and treatment outcome.
144 (98%) patients has completed anti tuberculosis chemotherapy. This isa very good treatment completion rate as >90% treatment success rate is given as the national target. 2 When compared with the studies conducted by south Asian countries this study showed a high treatment success rate. 6,7 One patient had failed treatment (0.7%). Two patients had died due to tuberculosis (1.4%) which is also less than national figure which lies between 3-5 %. 2

Conclusions
Economically productive age group (15-54) was affected more and there was high prevalence of pleural tuberculosis. The diagnosis of lymphatic tuberculosis is comparatively low. Prevalence of extra pulmonary tuberculosis was high in MOH areas with high population density which is similar to pulmonary tuberculosis. Prevalence of co-morbidities was higher than the normal population. Significantly high treatment completion rate than the national target indicatesan efficient patient follow-up at District Chest Clinic, Matara