|Year : 2021 | Volume
| Issue : 2 | Page : 149-155
Extrapulmonary Tuberculosis and Its Association with HIV in Patients Hospitalized in a Tertiary Care Center: A Cross-Sectional Study
Nitin Yashas Murthy, Soumya Umesh, Gherard D Ravindran
Department of General Medicine, St Johns Medical College Hospital, Bangalore, India
|Date of Submission||02-Jul-2020|
|Date of Decision||14-Sep-2020|
|Date of Acceptance||11-Oct-2020|
|Date of Web Publication||27-Aug-2021|
|Date of Print Publicaton||25-May-2021|
Dr. Soumya Umesh
Associate professor, Department of General Medicine, St Johns Medical College Hospital, Sarjapur Road, Bengaluru 560034
Source of Support: None, Conflict of Interest: None
Context: Extrapulmonary tuberculosis (EPTB) presentation can be unusual, and the diagnosis may often be challenging when associated with HIV. Aims: This study was conducted to find out the pattern of newly diagnosed EPTB in hospitalized patients, the diagnostic modalities used to arrive at the diagnosis, frequency of HIV positivity among them, and its association with CD4 counts. Settings and Design: All patients newly diagnosed to have EPTB admitted in a tertiary care medical college hospital in various departments over 1 year were included. Methods and Materials: Clinical presentation, comorbidities, and investigations done to arrive at diagnosis of EPTB were recorded. They were classified based on the affected site. HIV was tested for all patients, and CD4 counts in positive patients. Results: A total of 230 patients were newly diagnosed to have EPTB. The median age was 35 years (IQR 29–45), and the proportion of males was 127 (55%). Imaging was used alone in 47 (20%) patients and in combination with cytology or biopsy in 133 (57.8%) patients to establish the diagnosis. Pleural effusion, 50 (21.7%) patients, peripheral lymph node TB (LNTB), 44 (19.1%) patients, and CNS TB, 37 (16.1%) patients, were the most common forms. HIV was associated in 61 (26.5%) patients and diabetes in 27 (11.5%) patients. Both diseases influenced the presentation of EPTB. There was an association between HIV positivity and type of EPTB with peripheral LNTB being the most common in HIV patients. High CD4 counts were associated with pleural effusion and low counts with disseminated forms. Statistical Analysis Used: chi-square, ANOVA, and Student t test. Conclusion: The pattern of EPTB changes with HIV and vigilance is required to detect severe and disseminated forms of EPTB with lower CD4 counts.
Keywords: TB, extrapulmonary TB, HIV, CD4 count
|How to cite this article:|
Murthy NY, Umesh S, Ravindran GD. Extrapulmonary Tuberculosis and Its Association with HIV in Patients Hospitalized in a Tertiary Care Center: A Cross-Sectional Study. MAMC J Med Sci 2021;7:149-55
|How to cite this URL:|
Murthy NY, Umesh S, Ravindran GD. Extrapulmonary Tuberculosis and Its Association with HIV in Patients Hospitalized in a Tertiary Care Center: A Cross-Sectional Study. MAMC J Med Sci [serial online] 2021 [cited 2021 Nov 30];7:149-55. Available from: https://www.mamcjms.in/text.asp?2021/7/2/149/316913
Key Messages: HIV, CD4 count, and diabetes mellitus influence the pattern of EPTB
| Introduction|| |
Tuberculosis (TB) is a public health disorder of paramount importance. Although it is a treatable disease, 1.6 million people died of TB in 2018. India accounts for a quarter of the global burden. More than 0.3 million cases of new extrapulmonary TB (EPTB) were registered for treatment under RNTCP (Revised National Tuberculosis control programme) in 2018.
These include peripheral lymph node TB (LNTB), pleural TB, central nervous system (CNS) TB, abdominal TB (ATB), skeletal TB, genitourinary TB (GUTB), pericardial TB, miliary TB, cutaneous TB, and ocular TB. LNTB or pleural TB have been reported to be the most common forms of EPTB in South Asia.,,
HIV infected persons are at a markedly higher risk of developing TB, and it may occur as primary, reactivation, or reinfection from an exogenous source. The two diseases share a sinister synergy, and therefore, the incidence of EPTB is on the rise as the incidence of HIV is increasing. EPTB constitutes 15% to 20% of all cases of TB, but in HIV-infected individuals, the frequency is higher., Studies have found EPTB to be more common than PTB in HIV-infected patients.
EPTB presentations can be unusual and are also related to the degree of immunosuppression. There is an absence of definitive clinical or diagnostic tools and many overlapping differentials. Therefore, the diagnosis may pose a tremendous challenge. Most studies done in India are retrospective and are from DOTS or ART centers that lack facilities for newer or definitive tests., This limitation may have resulted in inaccurate reporting. There is an emergence of microbiological diagnostic tests such as TB-Polymerase chain reaction (TB-PCR) and cartridge based nucleic acid amplification test (CBNAAT). Also, imaging with ultasonography (USG), CT scan, MRI, endoscopy, and laparoscopy is more accurate and is being used increasingly. Therefore, the role of tertiary care centers and medical colleges with these facilities becomes crucial in making a definitive diagnosis of EPTB.
| Objectives|| |
This study was conducted to find out the patterns and frequency of newly diagnosed EPTB in hospitalized patients, the diagnostic modalities used to arrive at the diagnosis, rate of HIV positivity among them, and its association with CD4 counts.
| Methods|| |
We conducted a cross-sectional, descriptive epidemiological study after obtaining institutional ethical clearance. All hospitalized patients who were newly diagnosed to have EPTB in various departments from September 2016 over 1 year were notified to us. A chest X-ray was done in all patients. The patients who were diagnosed to have EPTB and found to have concurrent PTB by sputum positivity or parenchymal disease were excluded from the study as per WHO guidelines as they were classified as Pulmonary TB. ,,
EPTB cases were diagnosed based on
- Positive acid-fast bacilli smear or culture from the extrapulmonary site;
- Histopathology showing tubercular granuloma (with or without caseation);
- Features compatible with TB on radiological imaging or endoscopy
- Pleural/ascitic fluid analysis showing evidence of lymphocytic exudative effusion or CSF showing lymphocytic pleocytosis with hypoglycorrhachia.
- Patients with a negative diagnostic workup, but with a strong clinical suspicion of EPTB who were empirically started on full course of ATT drugs as per RTNCP guidelines.,
Informed consent was obtained and then the patients were recruited. They were interviewed, and their symptoms and comorbidities were recorded. The investigations used to confirm the diagnosis of EPTB were recorded. These included biopsy and histopathology reports, fluid or CSF analysis, acid fast bacilli (AFB) reports, radiological studies, and endoscopies. HIV test was done for all patients after obtaining consent. In seropositive patients, the most recent CD4 count was recorded. Anti-retroviral therapy (ART) status and current regimen were noted. The percentage of EPTB among all newly detected cases of TB among inpatients was also calculated by obtaining data on frequency of newly detected cases of PTB.
EPTB was classified as per the ICD 10 based on the affected site. Involvement of more than two noncontiguous organs was classified as disseminated TB.,,
The data were entered in a structured proforma and then transcribed into MS-Excel sheet (MS Excel 2016). SPSS (SPSS 23) was used for statistical analysis. A test of normality was used for determination of the distribution of data. Continuous normally distributed data were represented as mean and SD, whereas nonparametric was represented with median and interquartile range (IQR). Categorical data were expressed as numbers and percentages. Chi-square/Fischer’s test was used to check the association between categorical variables. Independent T test was used to compare the means of two independent groups, whereas one-way ANOVA was used for three or more independent variables. Mann–Whitney U test and Kruskal–Wallis test were used for nonparametric data. P value of <0.05 was considered to be significant.
| Results|| |
Frequency and pattern of EPTB
The total number of newly detected cases of TB in-hospital admissions was 430, of which PTB were 200 (46.52%) patients, and EPTB were 230 (53.48%) patients. The median age of patients with EPTB was 35 years (IQR 29–45). The proportion of males was 127 (55%). Most of the EPTB patients, 80 (34.8%), were in the 30 to 39 years age group followed by 63 (27.4%) patients in the 20 to 29 years age group. There was no statistically significant association between the type of EPTB and age group or the sex. A past history of TB was found in 16 (6.9%) of patients. The most common type was pleural effusion (21%) followed by peripheral lymph node TB, CNS TB, abdominal TB, and bone TB. [Table 1].
|Table 1 Frequency of the types of EPTB and HIV seropositivity in each group|
Click here to view
Diagnostics in EPTB
Imaging was the most important investigation and was used alone in 47 (20%) patients and in combination with cytology or biopsy in 133 (57.8%) patients to establish the diagnosis. No case was diagnosed on clinical grounds alone [Table 2].
|Table 2 Frequency of different diagnostic methods used to diagnose EPTB (n = 230)|
Click here to view
HIV was the most frequent comorbidity and was seen in 61 (26.5%) patients of EPTB. This was followed by diabetes mellitus, 27 (11.5%) patients. There was a statistically significant association between the presence of diabetes and the type of EPTB with pleural effusion being the most common, eight (29.6%) patients, followed by abdominal and bone TB, four (14.8%) patients each, followed by GU TB, three (11.1%) patients (P = 0.019).
HIV and EPTB
Of the 61 patients who were found to be HIV positive, 24 (39.3%) patients were between 30 and 39 years of age (P = 0.05). Newly diagnosed cases were 26 (42.6%), whereas the others were already receiving ART. The mean CD4 count was 204.46 ± 102.16 cells/mm3. The number of patients on first-line ART was 31 (50.8%), and four patients [6.6 %] were on second-line ART. There was no association between ART regimen or duration of HIV with the type of EPTB (P = 0.523). There was a significant association between HIV and the type of EPTB with peripheral lymphadenitis being the most common EPTB among those with HIV (P = 0.00). The mean CD4 count was lowest in disseminated TB and highest in pleural effusion (P = 0.03) [Table 3].
|Table 3 Pattern of EPTB in HIV positive patients, CD4 counts, and percentage of newly detected HIV in each group.|
Click here to view
In these 50 patients, the most common method of diagnosis was a combination of radiology and cytology, 44 (88%) patients, whereas only five (10%) patients required additional histopathology. The median adenosine deaminase (ADA) value was 41 U/L (IQR 32.75–54.5). There was no difference in ADA levels with HIV. None of the samples was positive for AFB. Those who required additional pleural biopsy had a significantly lower mean ADA of 22.6.
Peripheral lymph node TB
Forty-four patients had peripheral lymphadenitis. Cervical was found in 22 (50%) patients, axillary in nine (20.5%), supraclavicular in three (6.8%), inguinal in two (4.5%), and lymph nodes at two sites were found in eight (18.2%) patients. The diagnosis was confirmed using FNAC in 27 (61.4%) patients and biopsy in 17 (38.6%) patients [Figure 1]. The technique used had a significant association with the location of the lymph node with 18 (81.8%) of cervical lymph node TB being diagnosed by FNAC and eight (88.9%) of axillary lymph node being diagnosed by biopsy. AFB smear was positive in 19 patients (43.18%) with 17 (89.5%) of them being HIV positive. There was a statistically significant association between HIV positive status and AFB smear positivity (P = 0.00). However, there was no association between HIV and the site of lymphadenopathy.
|Figure 1 Histopathology of peripheral lymph node biopsy showing granulomatous inflammation with epitheloid cells, foamy macrophages, and giant cells in a background of necrosis|
Click here to view
Of the 33 patients, the most frequent type of ATB was intestinal and ascitic in equal numbers, 11 (33.3%) patients each. Abdominal CT [Figure 2] was done in 16 (48.4%) patients, whereas in 13 (39.4%) of the cases, only USG was needed. A biopsy was done under radiological or endoscopic guidance in 14 of the cases (42.4%) and with laparoscopy in one (3%) patient. Out of all the biopsies, AFB was detected in four (28.6%) cases. There was an association between the status of HIV and the type of abdominal TB with abdominal lymphadenopathy, five (62.5%) patients, being the most common in HIV positive patients. There was no relationship between ADA and HIV.
|Figure 2 (A) Axial contrast-enhanced CT of abdomen showing omental and peritoneal thickening. (B) Coronal section of contrast enhanced CT of abdomen showing gross ascites|
Click here to view
Central nervous system tuberculosis
There were 37 cases of CNS TB. TB Meningitis (TBM) alone was the most common presentation in 17 (45.9%) of the cases followed by tuberculomas, 11 (29.7%) patients [Figure 3]. A combination of pathologies, including TBM, tuberculomas, and tuberculous abscesses was seen in nine (24.3%) patients. The HIV frequency was 9 [24.30%], and there was no association between HIV seropositivity and the type of CNS TB. TBM cases with HIV had a significantly lower mean CD4 count of 107 cells/mm3 as compared to tuberculomas, in which the mean CD4 count was 260 cells/mm3 (P = 0.05). MRI was the imaging modality done in all 37 (100%) patients. The imaging findings in those who were HIV negative (N = 28) was tuberculomas, 10 (35.7%) patients, followed by a normal MRI in six (21.4%) of the patients, and three (10.7%) had basal exudates. In those with HIV and CNS TB (N = 9), the predominant MRI finding was a normal MRI in four (44.1%) patients, followed by tuberculomas in two (22.2%) patients. The median CSF ADA level was found to be 6.25 U/L (IQR 4.00–8.00). There was no difference in the levels of ADA based on HIV seropositivity.
|Figure 3 Axial T1W postcontrast MRI sequence of brain shows multiple conglomerate bifrontal tuberculomas with significant white matter edema|
Click here to view
Mediastinal lymph node TB
There were eight cases of mediastinal tuberculosis lymphadenitis without any parenchymal involvement. All the cases were diagnosed using CT thorax. Bronchoscopy and trans bronchial needle aspiration of the lymph nodes were done in four (50%) patients. AFB was demonstrated in two samples.
There were 20 cases of skeletal TB. TB spine was seen in 19 patients, and there was one case of TB hip. Psoas abscess was present in four patients. All cases were diagnosed using CT. Of the nine who underwent FNAC/biopsy under CT guidance, five were AFB positive.
A total of seven patients had GUTB. Urine AFB was positive in five (71.4%) patients. USG was done in three (42.8%) patients, CT and MRI in one (14.3%) each, while two (28.6%) patients required cystoscopy.
One patient had a retropharyngeal abscess, and another had nasal TB. Both were diagnosed based on biopsy and AFB was negative in both.
There were two patients with pericardial effusion. Both were diagnosed using echocardiography and paracentesis. Both were HIV positive, and both were already on ART. Both patients had neutrophil predominance in the aspirated fluid. Elevated ADA levels of 63.5 U/L and 55 U/L was seen respectively.
There were 27 patients with Disseminated TB involving two or more noncontiguous organs or miliary TB. Seven patients (25.9%) had choroid tubercles in the fundus.
| Discussion|| |
Our study found that more patients with EPTB than PTB were getting admitted for evaluation and treatment. The most frequent types of EPTB were pleural effusion followed closely by peripheral LNTB. The most important investigation to confirm diagnosis was imaging in combination with cytology or biopsy. HIV was the most common comorbidity, and in these patients, peripheral LNTB and CNS TB were the more common forms. CD4 counts had a relationship with the manifestation of EPTB wherein lower values were associated with disseminated forms, and higher values were associated with pleural effusion.
WHO estimates EPTB as 15% of all TB. Higher proportion found in our study may be attributed to rising numbers of EPTB and HIV. Ours is a tertiary care referral hospital-based study, and EPTB patients may have been more serious or needed more extensive evaluation to confirm diagnosis while most PTB patients may be diagnosed and treated on an outpatient basis itself.
The median age of those who had EPTB in the current study was 35 years. This is similar to other studies done in the Indian Subcontinent., This suggests the possibility that after primary infection in the lungs, the probability of reactivation at an extrapulmonary site may be higher at a younger age. There was no association between the type of EPTB and the age group in our study. However, a study done in Turkey demonstrated that the distribution of different kinds of EPTB differed significantly among age groups where younger patients had meningeal and GUTB, whereas older patients had LNTB, ATB, and skin TB.
Imaging was crucial both to confirm the diagnosis and to obtain samples of tissues or fluids with precision. It is noteworthy that MRI was done in all CNS TB cases. Neuroimaging with MRI is of proven superiority in better delineation of the pathology and therefore, accuracy in the diagnosis of tuberculosis. Chest X-ray and abdominal USG were done in most patients for diagnosis, but an additional CT scan in many patients in which the diagnosis was still in question often clinched the diagnosis. Studies in Saudi Arabia as well confirm the superiority of CT abdomen over USG in ATB. Increasing use of advanced imaging modalities may have increased the cost of care, but it has ensured that no case was diagnosed only based on strong clinical suspicion.
In the presence of DM, there was a different pattern of EPTB manifestation observed with pleural effusion being the most common followed by ATB and skeletal TB as compared to the findings in the entire study population in which pleural effusion was the most common followed by LNTB. This finding suggests that presence of DM may have a relationship to the EPTB manifestation which was corroborated in a study done in Kannur in Kerala.
The interaction between HIV and TB in persons co-infected with them is bidirectional and synergistic. The course of HIV infection is accelerated subsequent to the development of TB, and the inverse relationship between HIV viremia and CD4+ count gets accelerated. It is noteworthy that EPTB was the reason for the diagnosis of HIV in many patients, but many more developed EPTB along the course of illness despite being on ART. This emphasises the interaction between both diseases. There was no association between the type of EPTB and duration of HIV or the ART regimen. The mean CD4 count was 204 and higher than other studies like Spalgais et al. This may be attributed to the fact that many of our patients were already on ART. CD4 was lowest in disseminated EPTB and highest in pleural effusion, and this reiterates the role of immunosuppression in the manifestation of EPTB. Our findings can be explained by the fact that TB pleural effusions are usually caused by a delayed-type hypersensitivity reaction to mycobacterial antigens that would require an intact immune system. Studies were done by Jaryal et al., Sharma et al. in India,, and Kingkaew et al. in Thailand have all concurred that disseminated forms of EPTB were common in patients with low CD4 counts. However, Spalgais found no difference in CD4 counts between single or multiple site manifestations. This may be because the overall mean CD4 count in their study was low. CNS TB, which is a serious form of EPTB, is also associated with low CD4 count.,EPTB is a paucibacillary form, and AFB positivity is less common than in PTB. However, LNTB showed significant AFB positivity in HIV-positive patients as seen in other studies by Jaryal et al. and Rajashekaran et al. Therefore, there is a high likelihood of detecting AFB in TB lymph nodes of HIV-positive patients with FNAC or biopsy.
There was no difference in the CSF, pleural, or ascitic fluid analysis with HIV. ADA levels were similar, and AFB was not detected in any of the samples irrespective of HIV status. However, in ATB abdominal lymph nodes were more common with HIV as compared to ascites and intestinal TB, which were most common in all patients. These findings corroborate with a study carried out in Varanasi where abdominal lymphadenopathy constituted the majority of cases of those with HIV and abdominal TB.
In patients with CNS TB, TBM was the most common form. The presence of HIV made no difference to the CNS manifestation. This is similar to a study by Nelson et al., in which HIV did not typically alter the presentation.
The limitation of our study is that we have included only hospitalized patients. We have been unable to do CBNAAT, as it was not available in our institution at that time.
| Conclusion|| |
Our study has shown that EPTB is an important cause of admissions in a tertiary care hospital with tubercular pleural effusion being the most common form. EPTB is being diagnosed using combinations of imaging, cytology, and biopsy. Presence of HIV or DM influenced the manifestation of EPTB. Therefore, screening and treatment of both these comorbidities are imperative. Further studies are required to delineate the role of glycemic control in EPTB. LNTB was the most common EPTB in HIV patients. CD4 counts determined the severity of EPTB; hence, vigilance is required in patients with lower CD4 counts to detect more severe and disseminated forms of EPTB. Further studies are necessary to assess the outcomes of patients with various manifestations of EPTB.
Research Quality and Ethics Statement
The authors of this manuscript declare that this scientific work complies with reporting quality, formatting and reproducibility guidelines set forth by the EQUATOR Network. The authors also attest that this clinical investigation was determined to require the Institutional Review Board/Ethics Committee review, and the corresponding protocol/approval number is (IEC no.384/2018) number. We also certify that we have not plagiarized the contents in this submission and have done a Plagiarism Check.
The authors gratefully acknowledge Dr. Farah Naaz Fathima.
Financial support and sponsorship
Conflicts of interest
The authors reported no conflicts of interest.
| References|| |
World Health Organization. Global tuberculosis report 2018. Geneva: World Health Organization; 2018.
Central TB Division. India TB report 2019: RNTCP annual status report. New Delhi: Ministry of Health and Family Welfare; 2019.
Sreeramareddy CT, Panduru KV, Verma SC, Joshi HS, Bates MN. Comparison of pulmonary and extrapulmonary tuberculosis in Nepal − a hospital-based retrospective study. BMC Infect Dis 2008;8:8.
Velingker A, Lawande D, DCosta L. Clinico-epidemiological profile of extra pulmonary tuberculosis in Western India. Int J Contemp Med Res 2018;5:B1-B4.
Sharma SK, Mohan A. Extrapulmonary tuberculosis. Indian J Med Res 2004;120:316-53.
Namme LH, Marie-Solange D, Hugo Bertrand MN, Elvis T, Achu JH, Christopher K. Extrapulmonary tuberculosis and HIV coinfection in patients treated for tuberculosis at the Douala General Hospital in Cameroon. Ann Trop Med Public Heal 2013;6:100-4.
Spalgais S, Jaiswal A, Puri M, Sarin K, Agarwal U. Clinical profile and diagnosis of extrapulmonary TB in HIV infected patients: Routine abdominal ultrasonography increases detection of abdominal tuberculosis. Indian J Tuberc 2013;60:147-53.
Prakasha Sr, Suresh G, Shetty S, D′sa I, Kumar Sg. Mapping the pattern and trends of extrapulmonary tuberculosis. J Glob Infect Dis 2013;5:54.
Sharma SK, Mohan A, Chauhan LS, Narain JP, Kumar P, Behera D et al.
Contribution of medical colleges to tuberculosis control in India under the Revised National Tuberculosis Control Programme [RNTCP]: lessons learnt & challenges ahead. Indian J Med Res 2013;137:283-94.
] [Full text]
Wares F, Balasubramanian R, Mohan A, Sharma SK. Extrapulmonary tuberculosis: management and control. Tuberculosis control in India. New Delhi: Ministry of Health and Family Welfare; 2005. 95-114.
Central TB Division DGHS. Guidance for TB notification in India 2012 [Internet]. Delhi, India: Central TB division DGHS; 2012. Available from: http://tbcindia.nic.in/
. [Accessed September 16, 2016].
Sharma A, Raina R, Jaryal A, Sarkar M. Manifestations of tuberculosis in HIV/AIDS patients and its relationship with CD4 count. Lung India 2011;28:263. [Full text]
World Health Organization. International statistical classification of diseases and related health problems, 10th revision, Fifth Edition. Geneva, Switzerland: World Health Organisation; 2016.
García-Rodríguez JF, Álvarez-Díaz H, Lorenzo-García MV, Mariño-Callejo A, Fernández-Rial Á, Sesma-Sánchez P. Extrapulmonary tuberculosis: epidemiology and risk factors. Enferm Infecc Microbiol Clin 2011;29:502-9.
Sharma SK, Mohan A, Sharma A. Miliary tuberculosis: a new look at an old foe. J Clin Tuberc Other Mycobact Dis 2016;3:13-27.
Wang JY, Hsueh PR, Wang SK, Jan IS, Lee LN, Liaw YS et al.
Disseminated tuberculosis: a 10-year experience in a medical center. Medicine 2007;86:39-46.
Gunal S, Yang Z, Agarwal M, Koroglu M, Arc ZK, Durmaz R. Demographic and microbial characteristics of extrapulmonary tuberculosis cases diagnosed in Malatya, Turkey, 2001-2007. BMC Public Health 2011;11:154.
Trivedi R, Saksena S, Gupta RK. Magnetic resonance imaging in central nervous system tuberculosis. Indian J Radiol Imaging 2009;19:256-65.
] [Full text]
Ibrahim M, Osoba AO. Abdominal tuberculosis. On-going challenge to gastroenterologists. Saudi Med J 2005;26:274-80.
Mavila R, Kottarath M, Nair S, Thaha MM. Site predilection of extrapulmonary tuberculosis: study from a tertiary care centre. Int J Res Med Sci 2015;3:3386-90.
Toossi Z. Virological and immunological impact of tuberculosis on human immunodeficiency virus type 1 disease. J Infect Dis 2003;188:1146-55.
Sharma SK, Soneja M, Prasad KT, Ranjan S. Clinical profile & predictors of poor outcome of adult HIV-tuberculosis patients in a tertiary care centre in north India. Indian J Med Res 2014;139:154-60.
] [Full text]
Kingkaew N, Sangtong B, Amnuaiphon W, Jongpaibulpatana J, Mankatittham W, Akksilp S et al.
HIV-associated extrapulmonary tuberculosis in Thailand: epidemiology and risk factors for death. Int J Infect Dis 2009;13:722-9.
Rajashekharan S, Gunasekaran M, Jaykumar DDJ, Jeyaganesh D, Bhanumathi V. Tuberculous cervical lymphadenitis in HIV positive and negative patients. Indian J Tuberc 2001;48:201-4.
Agarwal D, Narayan S, Chakravarty J, Sundar S. Ultrasonography for diagnosis of abdominal tuberculosis in HIV infected people. Indian J Med Res 2010;132:77-80.
] [Full text]
Nelson CA, Zunt JR. Tuberculosis of the central nervous system in immunocompromised patients: HIV infection and solid organ transplant recipients. Clin Infect Dis 2011;53:915-26.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]