Aims and Objectives: To study the diagnostic practices for diagnosis of extra-pulmonary tuberculosis (EPTB) in a high-prevalence tuberculosis unit (TU). Material and Methods: It was a cross-sectional study using a pre-designed and pre-tested structured questionnaire. The information was collected from new EPTB cases registered in Rampur TU between 1 July 2007 and 31 March 2008. Diagnostic practices of the medical practitioners for establishing the diagnosis of different types of EPTB were studied. Results: For the diagnosis of pleural TB the main tools used were X-ray chest (37 cases; 100.0%) and aspiration of pleural fluid (pleural tap) for color of pleural fluid (35 cases; 94.6%). For establishing the diagnosis of lymph node TB, eight (66.6%) cases were subjected to fine needle aspiration cytology (FNAC) examination of involved lymph nodes. Excision biopsy of lymph nodes was undertaken in the remaining four (33.3%) cases. The diagnosis of abdominal TB was primarily established on the basis of X-ray (six cases; 85.7%) and ultrasonography (USG) of abdomen for the presence of ascites in five (71.4%) cases. The patients of tubercular meningitis were diagnosed predominantly on clinical grounds whereas in bone and genitourinary TB, all appropriate investigations were performed for confirmation of EPTB of these organs. Conclusions: Except for lymph node TB and the patients in whom the diagnosis had been established in tertiary care institutions of the state, patients were being diagnosed on clinical grounds.
Keywords: Diagnostic protocol, extra-pulmonary tuberculosis, overestimation
The state of Himachal Pradesh in India has been achieving the twin objectives of the Revised National Tuberculosis Control Program (RNTCP) in patients with pulmonary tuberculosis (PTB) since last many years. , The same may not be true for extra-pulmonary tuberculosis (EPTB) as no data on diagnostic practices in EPTB is available from the peripheral institutions of the state. There is little information on a state-wide basis in Himachal Pradesh about how the diagnosis of EPTB is made. With the exception of lymph node tuberculosis, which is confirmed through aspiration of affected lymph nodes, most patients with EPTB involving other sites are managed without bacteriological or histological confirmation in peripheral health institutions.
The accurate diagnosis of EPTB is often complex and difficult requiring invasive and expensive serological/ radiological investigations. Diagnostic modalities such as adenosine deaminase (ADA) levels and polymerase chain reaction (PCR) for confirmation of EPTB are useful in diagnosis.  These tests are not available in the peripheral health institutions and the diagnosis is established on clinical grounds alone in the majority of cases in these health facilities. The lack of these diagnostic facilities can lead to over-diagnosis of EPTB in a high proportion of cases.
The World Health Organization (WHO), in 1991, first published guidelines for national tuberculosis control programs that included criteria for the diagnosis of smear-positive and smear-negative pulmonary and extra-pulmonary tuberculosis. These guidelines were subsequently revised in 1997 and 2003 to facilitate prompt and accurate diagnosis.  Prompt diagnosis of EPTB requires fully functioning laboratories.
The study was conducted in Rampur TU of District Shimla. Rampur is located at 31° 45′ north latitude and 77° 63′ east longitude and is 130 kilometers from Shimla. Rampur has an average elevation of 1350 meters. TU Rampur was started in April 2000 and caters to a population of 1.64 lacs , and this TU has its boundaries with District Kinnaur and District Kullu. A large area under this TU remains covered with snow during winters. Some of the areas are still not connected with roads and have a difficult terrain.
All new EPTB cases registered in Rampur TU between 1 July 2007 and 31 March 2008 (a total of nine months) were included in the study. The selection of the TU was based on the proportion of EPTB cases reported from various TUs of the Shimla district. Rampur TU had reported highest proportion (47.2%) of new EPTB cases during the first (41%) and second (53.3 %) quarters of the year 2007 (average 47.2 %). , All new cases of EPTB put on Directly Observed Treatment Short Course (DOTS) between 1 July 2007 and 31 March 2008 and patients not switched to non-DOTS treatment at the time of the study were included in the study. Patients who were seriously ill and hospitalized at the time of data collection or patients who were transferred in or trnsferred out during the study period were excluded from the study. Patients who could not be traced even after paying two visits to the respective DOTS centre and at their residence were also excluded from the study.
During the period of study, Rampur TU was visited and a list of the EPTB cases diagnosed and registered during the already defined period was prepared. The data was scrutinized for errors and after cleaning the data a fresh list of cases to be recruited was prepared. In the selected TU, as per the TB register record, 70 new EPTB cases were registered during the study period (1 July 2007 to 31 March 2008). Of the initial list of 70 new EPTB cases, one case was over-reported, two sputum-negative pulmonary tuberculosis (PTB) cases were wrongly recorded as EPTB cases, two patients died during the treatment, one case was later proved to be suffering from ovarian malignancy and three patients could not be contacted during the course of study. After excluding these nine cases, the revised list for study comprised 61 patients who were enrolled for the purpose of the present study.
A list of DOTS centers from where those patients were receiving anti-tubercular treatment (ATT) was also prepared. The DOTS centers were visited on Monday, Wednesday or Friday (DOTS Days in Himachal Pradesh, India) to locate the EPTB cases. The cases were interviewed and their treatment cards were checked after obtaining prior informed consent. A predesigned and pretested structured questionnaire was used to collect information on diagnostic protocol. The evidence on diagnostic protocol was corroborated by conducting an interview with the medical practitioner making the diagnosis for EPTB in case of the concerned patient. Information on treatment being received by the patient was also collected on the same questionnaire. The information obtained on interview was also crosschecked with the relevant records like Tuberculosis (TB) Register, Laboratory Register and Treatment Cards. If any patient did not visit the center on the scheduled DOTS days, he/she was traced to his/her home and interviewed there.
Sixty-one new EPTB cases were enrolled in the study [Table 1]. For the diagnosis of pleural TB (n = 37), the main tools used were X-ray chest (37 cases; 100.0%) and aspiration of pleural fluid (pleural tap) for color of pleural fluid (35 cases; 94.6%). Microscopy and biochemistry of pleural fluid was undertaken only in four cases (10.8%). In 13 (35.1%) patients of pleural TB, the diagnosis was supported by clinical findings and in 11(29.7%) cases antibiotic trial was given prior to start of ATT [Table 2]. For establishing the diagnosis of lymph node TB (n = 12), eight (66.6%) cases were subjected to fine needle aspiration cytology (FNAC) examination of involved lymph nodes. Excision biopsy of lymph nodes was undertaken in the remaining four (33.3%) cases [Table 3]. The diagnosis of abdominal TB (n = 7) was primarily established on the basis of X-ray (six cases; 85.7%) and ultrasonography (USG) of abdomen for the presence of ascites in nine (71.4%) cases. Positive Mantoux test was also undertaken as a criterion for establishing the diagnosis in two (28.6%) children. Clinical findings suggestive of abdominal TB were taken into consideration in all the cases [Table 4]. The patients of tubercular meningitis were diagnosed predominantly on clinical grounds whereas in bone and genitourinary TB, all appropriate investigations were performed for confirmation of EPTB of these organs [Table 5].
The study sample comprised 61 new EPTB patients registered for DOTS in Rampur TU in District Shimla during the third and fourth quarters of 2007 and the first quarter of 2008 (1 July 2007 to 31 March 2008). Of the 37 cases of pleural TB registered in TU Rampur, all the patients had undergone radiological examination of the chest (X-ray chest). The color of the pleural fluid consistent with tubercular pleural effusion was taken as the most appropriate diagnostic criterion in the diagnosis of pleural TB in most of these cases. The pleural fluid was subjected to microscopic and biochemical examination in a very small number of patients. Clinical signs consistent with pleural TB were also considered for establishing the diagnosis in 13 (35.1%) cases. Investigations like USG chest for pleural effusion, computed tomography (CT) scan and magnetic resonance imaging (MRI) were performed only in the cases in which the diagnosis was established at a tertiary hospital in the state. In one pediatric patient diagnosis of pleural TB was established on the basis of positive Mantoux test. Hematological examination for Erythrocyte sedimentation rate (ESR), Total leucocyte count (TLC) and Differenetial leucocyte Ccount (DLC); clinical signs suggestive of pleural TB and antibiotic trial before starting the DOTS were also considered as supportive diagnostic criteria.
In the review of EPTB cases in RNTCP in India,  97% of the pleural effusion patients had at least one X-ray performed, in 24.5% cases pleural aspirate was examined for Acid fast bacillus (AFB) smear, total lymphocyte count and biochemistry; 1.6% patients were diagnosed on clinical grounds alone and pleural biopsy was done in two of the 624 cases. The diagnosis of lymph node TB was established on the basis of either FNAC of lymph nodes or histopathological examination of the excised lymph node tissue. In a few cases, an antibiotic trial was also given before definitive diagnosis of lymph node TB was established. In the same study, of all 1630 cases of lymph node TB 696 (43%) were diagnosed on clinical grounds alone, 341 (21%) after antibiotic trial, 418 (25.6%) on FNAC examination and 252 (15%) were diagnosed after histopathological examination of lymph node biopsy. In another study from rural Ethiopia, among 147 clinically suspected cases, 107 (72.8%) were confirmed by FNAC and AFB smear examination. 
In our study, patients having abdominal TB were diagnosed purely on the basis of the presence of ascites on X-ray abdomen or USG abdomen. Hematological findings suggestive of TB were taken as supportive diagnostic criteria in most of the cases. Clinical signs consistent with abdominal TB were also considered in establishing the diagnosis in all these cases. In pediatric patients, positive Mantoux test was taken as the diagnostic criterion.
In a retrospective study from Nepal, the diagnostic modalities in 32 patients with abdominal TB were studied. Two cases were diagnosed on clinical evidence alone, 9 cases on the basis of clinical and biochemical evidence, 1 case by clinical and histopathological evidence, 17 cases by clinical and imageological evidence, 5 cases on the basis of clinical and laparotomy evidence, 1 case by clinical and microbiological evidence and 5 cases by clinical aberrant. 
In the present study, two patients were diagnosed with meningeal TB in TU Rampur. One pediatric patient aged four years was diagnosed on clinical signs consistent with tubercular meningitis. A trial of antibiotics was given prior to starting anti-tubercular drugs to which the child did not respond. Whereas in the second patient aged 19 years, the diagnosis was established on microscopic and biochemical examination of cerebrospinal fluid (CSF). Clinical signs consistent with tubercular meningitis were also taken into consideration for establishing the diagnosis.
In a study in Pereira, Colombia, 14 (73.6%) cases were diagnosed on the basis of clinical criteria and 9% on the basis of Ziehl Neelsen (ZN) staining positive for M. tuberculosis.  Another study from Hong Kong reveals that two (12.5%) cases of meningeal TB were diagnosed by histopathology, eight (50.0%) by culture and three (18.8%) by radiology or clinical signs. 
In the present study one case of bone TB and two cases of genital TB were registered in the TU. The diagnosis of bone TB was established at a tertiary hospital in the state. The relevant investigations were done for establishing the diagnosis which included X-ray, CT scan, MRI, histopathological and microscopic examination and hematological examination. In this patient diagnosed with bone TB, prior antibiotic trial was given before starting anti-tubercular drugs. In two patients the diagnosis of genito-urinary TB was established at a tertiary level hospital in the state. In both the cases hematological examination, histopathological examination of endometrial biopsy and laparoscopic examination of pelvic organs was done to establish the diagnosis.
Except for lymph node TB and the patients in whom the diagnosis had been established in tertiary care institutions of the state, the majority of the patients with EPTB were being diagnosed on clinical grounds and with basic investigations such as the color of pleural fluid or blood tests for TLC, DLC and ESR. Most of the EPTB cases were not being confirmed at the selected TU as the facilities for the required tests were not available in this centre.
It is necessary to strengthen laboratory infrastructure and the services in the peripheral health institutions since most patients seek healthcare advice in these institutions. Also the influence of host and bacterial genotype on the development of disseminated disease with M. tuberculosis has been shown. , This necessitates further studies in this vital area of genetic typing of the host and bacterial genomes. Since, in the present study, the diagnosis of EPTB had already been established and the required investigations had not been done in most of these cases, further prospective studies with confirmation of diagnosis as per the RNTCP guidelines need to be carried out to know the prevalence and the true burden of EPTB disease in the state.
Limitations of the study
The data was collected primarily on the basis of recall method. So there is a probability of recall bias by the patients. The association of HIV by carrying out HIV serology that has been presumed to be associated with rise in EPTB cases in developed countries was not studied.
Source of Support: None, Conflict of Interest: None
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]