Diagnostic Usefulness of a T-cell-based Assay for Extrapulmonary Tuberculosis in Immunocompromised Patients

CLINICAL RESEARCH STUDY Diagnostic Usefulness of a T-cell-based Assay for Extrapulmonary Tuberculosis in Immunocompromised Patients Sung-Han Kim, MD, a,b,c Kyoung-Ho Song, MD, a Su-Jin Choi, MS, b Hong-Bin Kim, MD, a Nam-Joong Kim, MD, a Myoung-don Oh, MD, a,b Kang-Won Choe, MD a,b a Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea; b Clinical Research Institute, Seoul National University Hospital, Seoul, Republic of Korea. c Current affiliation: Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea. ABSTRACT BACKGROUND: The low reactivity of the tuberculin skin test limits its clinical use in immunocompromised patients with extrapulmonary tuberculosis. A recently developed T-cell-based assay for diagnosing tuberculosis infection gave promising results. However, there were few data on the usefulness of this assay for diagnosing extrapulmonary tuberculosis in immunocompromised patients. METHODS: All adult patients with suspected extrapulmonary tuberculosis were prospectively enrolled at 2 university-affiliated hospitals over an 18-month period. In addition to the conventional tests for diagnosing extrapulmonary tuberculosis, enzyme-linked immunospot (ELISPOT) assay for the interferon- -producing T-cell response to early secretory antigenic target-6 and culture filtrate protein-10 was performed. The final diagnoses in patients with suspected extrapulmonary tuberculosis were classified by clinical category. RESULTS: There were 179 patients with suspected extrapulmonary tuberculosis enrolled: 59 (33%) were classified as immunocompromised. Of the 179 patients, 75 (42%) were classified as extrapulmonary tuberculosis, including 56 confirmed tuberculosis plus 19 probable tuberculosis, and 97 (54%) were classified as not tuberculosis. The remaining 7 (4%) had possible tuberculosis and were excluded from the final analysis. The tuberculin skin test (induration size 10 mm) was less sensitive in immunocompromised patients (38%; 95% confidence interval [CI], 19%-59%) than in immunocompetent patients (69%; 95% CI, 54%-81%, P.01). In contrast, the ELISPOT assay retained a high sensitivity: (88%; 95% CI, 68%-97%) in immunocompromised patients compared with 96% (95% CI, 87%-100%) in immunocompetent patients (P.32). CONCLUSION: The immunosuppressive condition does not affect the diagnostic sensitivity of the ELISPOT assay for extrapulmonary tuberculosis. 2009 Elsevier Inc. All rights reserved. The American Journal of Medicine (2009) 122, 189-195 KEYWORDS: ELISPOT; Immunocompromised; Tuberculosis Funding: No author received financial support. Conflict of Interest: There are no potential conflicts of interest for any authors. Authorship: All authors had access to the data and played a role in writing this manuscript. Requests for reprints should be addressed to Myoung-don Oh, MD, Department of Internal Medicine, Seoul National University College of Medicine, 28 Youngun-dong, Chongro-gu, Seoul 110-744, Republic of Korea. E-mail address: mdohmd@snu.ac.kr Extrapulmonary tuberculosis has become more common in developed countries, probably because of increased numbers of immunocompromised patients (ie, human immunodeficiency virus [HIV]-infected patients 1 or patients receiving tumor-necrosis factor- antagonist 2 ) and foreign-born persons. 3 However, diagnosing extrapulmonary tuberculosis remains a challenge, as clinical samples obtained from sites that are not easily accessible are sometimes paucibacillary, thus decreasing the sensitivity of diagnostic tests. 4 Furthermore, the increased use of immunosuppressive treatments highlights the importance of early diagnosis and prompt treatment of tuberculosis in patients with extrapulmonary tuberculosis, because delays in the diagnosis may contribute to morbidity and mortality. 2,5 In this problematic clinical situation, the tuberculin skin test is 0002-9343/$ -see front matter 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.amjmed.2008.07.028

190 The American Journal of Medicine, Vol 122, No 2, February 2009 one of the most widely used tests to detect Myobacterium tuberculosis infection and guide the initiation of empirical therapy. 6 However, low reactivity of tuberculin skin test limits its clinical use in immunocompromised patients with extrapulmonary tuberculosis. 6 The enzyme-linked immunospot assay (ELISPOT) detecting interferon- -secreting T-cells is based on 2 antigens, early secretory antigenic target-6 (ESAT-6) and culture filtrate protein-10 (CFP-10), which are encoded by genes in the region of difference-1 (RD1) present in M. tuberculosis but absent from Myobacterium bovis bacilli Calmette-Guérin vaccine and most nontuberculous mycobacteria. 7,8 This T-cell-based assay for diagnosing tuberculosis infection gave promising results. 8 Recently, we reported preliminary data indicating that it is a useful adjunct test for diagnosing extrapulmonary tuberculosis, with sensitivity reaching more than 90%. 9 However, little information is available on the usefulness of this assay for diagnosing extrapulmonary tuberculosis in immunocompromised patients. The purpose of this study was to examine the usefulness of the tuberculin skin test and the ELISPOT assay for diagnosing extrapulmonary tuberculosis in immunocompromised patients versus immunocompetent patients. CLINICAL SIGNIFICANCE Enzyme-linked immunospot (ELISPOT) assay had the same high sensitivity in immunocompromised patients as in immunocompetent patients, although tuberculin skin test had lower sensitivity in immunocompromised patients than in immunocompetent patients. ELISPOT assay may be used as an adjunct test for excluding active tuberculosis in patients with suspected extrapulmonary tuberculosis regardless of immunosuppressive condition. METHODS Study Population All adult patients with suspected extrapulmonary tuberculosis were prospectively enrolled at Seoul National University Hospital, Seoul, and Seoul National University Bundang Hospital, Kyeonggi, South Korea, between April 2006 and September 2007. Microbiologic and pathologic specimens for diagnosing extrapulmonary tuberculosis were processed by standard techniques and procedures. In brief, mycobacteria were cultured on solid culture medium, and the M. tuberculosis complex was identified with a commercial DNA probe (AccuProbe Mycobacterium complex culture identification kit; Gen- Probe; San Diego, Calif). We used the commercially available Cobas Amplicor MTB PCR test (Roche Diagnostics, Branchburg, NJ). Decisions about antituberculosis therapy were the responsibility of the primary care physicians (Figure 1). The results of the tuberculin skin test and ELISPOT assays were not concealed from the attending physicians. This investigation was approved by the Institutional Review Board of our hospitals. Some of the cases at Seoul National University Hospital were enrolled in a previous study. 9 Figure 1 Schematic flow chart of the relationship of the ELISPOT assay to empirical antituberculosis therapy and the final classification of extrapulmonary tuberculosis. TB tuberculosis.

Kim et al Immunodiagnosis for Extrapulmonary Tuberculosis 191 Clinical Category of Extrapulmonary Tuberculosis and Definition of Terms All cases were independently classified by the study investigators, without knowledge of the results of the tuberculin skin test and ELISPOT assays, on the basis of clinical, histopathological, radiological, and microbiological information collected over at least 3 months of follow-up. The clinical categories of patients with suspected extrapulmonary tuberculosis were given in the previous report. 9 To classify the diagnostic performances of the ELISPOT assay and tuberculin skin test, we used confirmed and probable tuberculosis as the reference standards for extrapulmonary tuberculosis, and not tuberculosis as that for not extrapulmonary tuberculosis. Immunocompromised patients were defined as those with underlying diseases such as HIV infection, malignancy, liver cirrhosis, and chronic renal failure, or those receiving immunosuppressive treatment. 9 ELISPOT Assay and Tuberculin Skin Test A peripheral venous blood sample was collected from each patient for the ELISPOT assay for the interferon- -producing T-cell response (ie, T-SPOT.TB, Oxford Immunotec, Abingdon, UK). Peripheral blood mononuclear cells (PBMCs) were separated from peripheral venous blood, and 2.5 10 5 PBMCs were plated per well in wells precoated with antihuman interferon- antibody. The PBMCs were cultured at 37 C for 18 hours, and spots were counted with an automated microscope (CRL ImmunoSpot S4 Core Analyzer, Cellular Technology Ltd., Cleveland, OH). We used the criteria for positive, negative, and indeterminate outcomes recommended by the manufacturer. The tuberculin skin test was done by the Mantoux technique, injecting a 2-TU dose of purified protein derivative RT23 (Statens Serum Institut, Copenhagen, Denmark) intradermally into the forearm. 10,11 The positive criterion for tuberculin skin test was 10 mm or greater size of induration 48 hours after injection. 11 Statistical Analyses Statistical analyses were performed with SPSS for Windows (version 12.0; SPSS Inc., Chicago, Ill). Categorical variables were compared by Pearson chi-squared test or Fisher s exact test where appropriate. The Mann-Whitney U test was used to compare continuous variables. Diagnostic performance was expressed in terms of sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio. Concordance between the tuberculin skin test and ELISPOT assay was assessed using the coefficient. 12 All tests of significance were 2-tailed; P values of.05 were considered significant. RESULTS Patient Characteristics There were 186 subjects with suspected diagnoses of extrapulmonary tuberculosis prospectively enrolled. All subjects received the tuberculin skin test and ELISPOT assay. Three (2%) gave indeterminate ELISPOT results and 4 (2%) missed a follow-up visit and were excluded from the final analysis (Figure 1). Of the remaining 179 patients, 59 (33%) were classified as immunocompromised. The final clinical categories of the 179 patients with suspected extrapulmonary tuberculosis were as follows: 56 (31%) were classified as confirmed tuberculosis, 19 (11%) as probable tuberculosis, and 97 (54%) as not tuberculosis. The remaining 7 (4%) had possible tuberculosis. Confirmed and probable tuberculosis were used as the reference standard of extrapulmonary tuberculosis, and not tuberculosis as that of not extrapulmonary tuberculosis. The baseline clinical characteristics of the patients with extrapulmonary tuberculosis and not extrapulmonary tuberculosis are presented in Table 1. Diagnostic Performances of the Tuberculin Skin Test versus the ELISPOT Assay The diagnostic performances of the positive tuberculin skin test and ELISPOT assay in the 179 patients with suspected extrapulmonary tuberculosis by diagnostic category are shown in Table 2. When patients with possible tuberculosis were excluded from the analysis, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio of the tuberculin skin test were 59% (95% confidence interval [CI], 47%-70%), 76% (95% CI, 67%-84%), 66% (95% CI, 61%- 77%), 70% (95% CI, 53%-79%), 2.47 (95% CI, 1.65-3.71), and 0.54 (95% CI, 0.40-0.72), respectively (Table 3). By comparison, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio of the ELISPOT assay were 93% (95% CI, 85%-98%), 66% (95% CI, 56%-75%), 68% (95% CI, 58%- 77%), 93% (95% CI, 84%-98%), 2.74 (95% CI, 2.07-3.64), and 0.10 (95% CI, 0.04-0.24), respectively (Table 3). When the concordance between the tuberculin skin test and ELIS- POT assay was assessed using the coefficient, moderate agreement was found ( 0.44, 95% CI, 0.33-0.56). Diagnostic Performances of the Tuberculin Skin Test versus the ELISPOT Assay in Immunocompromised Patients versus Immunocompetent Patients The effect of immunosuppression on the positive tuberculin skin test and ELISPOT assays in the 179 patients with suspected E-TB by diagnostic category is shown in Table 2. When patients with possible tuberculosis were excluded from the analysis, the tuberculin skin test (induration size 10 mm) showed lower sensitivity in immunocompromised patients (38%; 95% CI, 19%-59%) than in immunocompetent patients (69%; 95% CI, 54%-81%, P.01; Table 3). In contrast, the ELISPOT assay was as sensitive in immunocompromised patients (88%; 95% CI, 68%-97%) as in immunocompetent patients (96%; 95% CI, 87%-100%, P.32, Table 3).

192 The American Journal of Medicine, Vol 122, No 2, February 2009 Table 1 Clinical Characteristics of 172 Patients with Suspected Extrapulmonary Tuberculosis* Characteristics Final Diagnosis Extrapulmonary Tuberculosis (n 75) Not Tuberculosis (n 97) Mean age (mean SD), years 48.3 17.8 46.2 17.8 Male 36 (48) 50 (52) Suspected infection site Lymph node 17 (23) 21 (22) Skeletal 17 (23) 19 (20) Central nervous system 8 (10) 29 (30) Abdominal 13 (17) 14 (14) Pleural 4 (5) 5 (5) Miliary nodules in lung 5 (7) 4 (4) Genitourinary 7 (9) 1 (1) Pericardial 2 (3) 2 (2) Other sites 2 (3) 2 (2) Extent of disease Extrapulmonary disease only 45 (60) 71 (73) Extrapulmonary and pulmonary disease 9 (12) 10 (10) Disseminated disease 21 (28) 16 (17) Underlying condition or illness Human immunodeficiency virus infection 4 (5) 6 (6) Transplantation 3 (4) 3 (3) Hematologic malignancy 5 (7) 9 (9) Solid tumor 4 (5) 5 (5) Renal insufficiency 5 (7) 1 (1) Rheumatologic disease 1 (1) 8 (8) Diabetes 5 (7) 5 (5) No underlying illness 48 (64) 60 (62) Immunosuppressive condition 24 (32) 35 (36) Prior latent tuberculosis treatment 0 (0) 0 (0) Prior active tuberculosis treatment 5 (7) 5 (5) Results of diagnostic tests for tuberculosis Granulomatous inflammation seen in biopsy 56 (75) 11 (11) specimen Positive AFB stain in (biopsy) specimen 14 (19) 0 (0) Positive M. tuberculosis PCR in (biopsy) specimen 45 (60) 0 (0) Positive M. tuberculosis culture in (biopsy) specimen 44 (59) 0 (0) Data are presented as numbers (%) unless otherwise indicated. *Extrapulmonary tuberculosis includes confirmed (n 56) and probable (n 19) tuberculosis. Seven patients with possible tuberculosis were excluded. Hematologic abnormalities due to bone marrow involvement. Disseminated tuberculosis includes isolation of M. tuberculosis, positive M. tuberculosis PCR, or histologic demonstration of caseating granulomatous inflammation from the bone marrow, 2 noncontiguous organs, or miliary lung lesions. Immunosuppressive condition is defined as patients with underlying diseases such as human immunodeficiency virus infection, malignancy, liver cirrhosis, and chronic renal failure, or receiving immunosuppressive treatment. Biopsy specimens were available in 58 patients. Biopsy specimens were available in 58 patients. Associations between the Size of Indurations in the Tuberculin Skin Test, or Responses to ESAT-6 and CFP-10, and Immunosuppressive Status We evaluated the associations between the size of induration in the tuberculin skin test, or the responses to ESAT-6 and CFP-10, and the immunosuppressive status of patients with extrapulmonary tuberculosis (Figure 2). Although size of induration in the tuberculin skin test was associated with immunosuppression (P.004), there was no significant association between immunosuppressive condition and responsiveness to ESAT-6 (P 0.81) or CFP-10 (P.68). DISCUSSION We have assessed the clinical usefulness of the T-cell-based ELISPOT assay in patients with suspected extrapulmonary

Kim et al Immunodiagnosis for Extrapulmonary Tuberculosis 193 Table 2 Effect of Immunosuppression on the Positive Tuberculin Skin Test and ELISPOT Assay in All 179 Patients with Suspected Extrapulmonary Tuberculosis by Diagnostic Category Tuberculin Skin Test Positive ELISPOT Assay Positive All patients (n 179) Confirmed and probable TB* 44/75 (59%, 47%-70%) 70/75 (93%, 85%-98%) Confirmed TB 33/56 (59%, 45%-72%) 53/56 (95%, 85%-99%) Probable TB 11/19 (58%, 34%-80%) 17/19 (90%, 67%-99%) Possible TB 3/7 (43%, 10%-82%) 5/7 (56%, 21%-86%) Not TB 23/97 (24%, 17%-33%) 33/97 (34%, 25%-44%) Non-immunocompromised host (n 120) Confirmed and probable TB* 35/51 (69%, 54%-81%) 49/51 (96%, 87%-100%) Confirmed TB 25/35 (71%, 54%-85%) 33/35 (94%, 81%-99%) Probable TB 10/16 (63%, 35%-85%) 16/16 (100%, 79%-100%) Possible TB 3/7 (43%, 10%-82%) 5/7 (56%, 21%-86%) Not TB 19/62 (31%, 20%-44%) 22/62 (36%, 24%-49%) Immunocompromised host (n 59) Confirmed and probable TB* 9/24 (38%, 19%-59%) 21/24 (88%, 68%-97%) Confirmed TB 8/21 (38%, 18%-62%) 20/21 (95%, 76%-100%) Probable TB 1/3 (33%, 1%-91%) 1/3 (33%, 1%-91%) Possible TB 0/0 (0%, 0%-91%) 0/0 (100%, 3%-100%) Not TB 4/35 (11%, 3%-27%) 11/35 (31%, 17%-49%) TB tuberculosis. Data are presented as numbers positive for test/numbers tested (%, 95% confidence interval). *P.05 for difference in sensitivity between tuberculin skin test and ELISPOT assay. P.05 for difference in specificity between tuberculin skin test and ELISPOT assay. P.05 for difference in specificity between tuberculin skin test and ELISPOT assay. tuberculosis. Our data show that the ELISPOT assay is a more sensitive test for extrapulmonary tuberculosis than the tuberculin skin test. Furthermore, the ELISPOT assay was less affected by immunosuppression than the tuberculin skin test. The sensitivity of the tuberculin skin test in patients with cervical tuberculous lymphadenitis is reported to exceed 90%. 13,14 Therefore, granuloma with and without necrosis in patients with a positive tuberculin skin test and a compatible clinical syndrome provides an immediate presumptive diagnosis, and clinicians usually start empirical therapy while awaiting culture results. 14 However, low tuberculin reactivity limits its clinical use in patients with immunosuppression or severe tuberculosis other than cervical tuberculous lymphadenitis. 6 Therefore, it was important to know whether the new T-cell- Table 3 Accuracy of Diagnosis of Active Tuberculosis in 172 Patients with Suspected Extrapulmonary Tuberculosis* Sensitivity % Specificity % PPV % NPV % Negative Likelihood Ratio Positive Likelihood Ratio All patients (n 172) Tuberculin skin test 59 (47-70) 76 (67-84) 66 (61-77) 70 (53-79) 2.47 (1.6503.71) 0.54 (0.40-0.72) ELISPOT assay 93 (85-98) 66 (56-75) 68 (58-77) 93 (84-98) 2.74 (2.07-3.64) 0.10 (0.04-0.24) Nonimmunocompromised patients (n 113) Tuberculin skin test 69 (54-81) 69 (56-80) 65 (51-77) 73 (59-84) 2.24 (1.47-3.40) 0.45 (0.29-0.69) ELISPOT assay 96 (87-100) 64 (51-76) 69 (56-79) 95 (83-99) 2.71 (1.93-3.80) 0.06 (0.02-0.24) Immunocompromised patients (n 56) Tuberculin skin test 38 (10-59) 89 (73-97) 69 (39-91) 67 (52-80) 3.28 (1.14-9.44) 0.70 (0.51-0.98) ELISPOT assay 88 (68-97) 69 (51-83) 66 (47-81) 89 (71-98) 2.78 (1.67-4.65) 0.18 (0.06-0.54) CI confidence interval; PPV positive predictive value; NPV negative predictive value. *Seven patients with possible tuberculosis were excluded. P.05 between tuberculin skin test and ELISPOT assay.

194 The American Journal of Medicine, Vol 122, No 2, February 2009 Figure 2 Response to early secretory antigenic target-6 and culture filtrate protein-10 and induration size of tuberculin skin test according to immunosuppressive status in patients with extrapulmonary tuberculosis. Bars indicate medians. The Mann-Whitney U test was used to compare the differences between groups. ESAT-6 early secretory antigenic target-6; CFP- 10 culture filtrate protein-10; TST tuberculin skin test. based assay could be used in this subgroup. We showed that the ELISPOT assay had the same high sensitivity in immunocompromised patients with extrapulmonary tuberculosis as in immunocompetent patients with extrapulmonary tuberculosis. Our data thus suggest that the ELISPOT assay is a useful tool for excluding tuberculosis in patients with suspected extrapulmonary tuberculosis, regardless of immunosuppressive condition. Previous studies have addressed the clinical usefulness of diagnoses based on M. tuberculosis-specific antigens, such as ELISPOT-based diagnosis (ie, T-SPOT.TB) and enzyme-linked immunosorbent assay-based diagnosis (ie, QuantiFERON-TB Gold), for latent tuberculosis infection and active pulmonary tuberculosis. 8 However, data on the clinical usefulness of these new tests for diagnosing extrapulmonary tuberculosis are limited. 9,15-18 In the present large cohort, we found that the ELISPOT assay had 93% sensitivity for diagnosing active tuberculosis in patients with suspected extrapulmonary tuberculosis, which is consistent with the previous report. 9 We also showed that the ELISPOT assay displayed high sensitivity, even in immunocompromised patients with extrapulmonary tuberculosis. Two recent studies comparing these 2 commercial tests concurred in showing T-SPOT.TB to be more sensitive and less influenced by immunosuppression than QuantiFERON-TB Gold. 17,19 It also has been reported that the sensitivity of T-SPOT.TB is not affected in HIV-infected patients with tuberculosis infection. 20 However, our study is the first to compare the usefulness of the tuberculin skin test and the ELISPOT assay for diagnosing extrapulmonary tuberculosis in immunocompromised patients versus immunocompetent patients. In the present study, 33 (34%) of the 97 subjects without active tuberculosis were positive in the ELISPOT assay. Thus, the clinical use of this assay is limited in regions where tuberculosis is of intermediate to high prevalence. In general, sensitivity and specificity were not affected by disease prevalence. However, because the ELISPOT assay can detect latent tuberculosis infection as well as active tuberculosis, the specificity of this test for active tuberculosis depends on the prevalence of latent tuberculosis infection. Therefore, when a patient in a region where the prevalence of tuberculosis is low has a moderate probability of having extrapulmonary tuberculosis (an usual clinical presentation of extrapulmonary tuberculosis) and a positive ELISPOT result, the post-test probability should be high enough to diagnose extrapulmonary tuberculosis, because specificity should be higher in areas of low prevalence of tuberculosis than in those of high prevalence. However, the sensitivity of this test for active tuberculosis is not affected by the prevalence of tuberculosis. If a patient had a low pretest probability of extrapulmonary tuberculosis (ie, an unusual clinical manifestation or residence in a region of low prevalence of tuberculosis) and a negative ELISPOT

Kim et al Immunodiagnosis for Extrapulmonary Tuberculosis 195 result, the post-test probability would be so low that one could exclude a diagnosis of extrapulmonary tuberculosis. It is not known whether RD1 peptide-specific T-cell responses are influenced by differences of disease status or host condition. We showed that immunosuppressive status had little impact on RD1 peptide responsiveness. Recently, Goletti et al 21 demonstrated that patients with severe pulmonary tuberculosis gave a lower response to selected RD1 peptides than those with less severe disease. There are reports that the magnitude of the response to RD1 peptides is correlated with bacterial antigenic load. 22-24 However, it seems to us that an antigenic load and host immune response to antigen should be interdependent, so that both factors should contribute to in vitro responsiveness to RD1 peptides. Further studies are needed on this issue. Some readers may be concerned that 11% of the patients were classified as probable extrapulmonary tuberculosis based mainly on histopathologic findings and clinical responses to antituberculosis therapy, without microbiologic confirmation. Another possible source of concern is the fact that the attending physicians were not blind to the results of the tuberculin skin test and ELISPOT, and this could have affected decisions about empirical antituberculosis treatment (Figure 1). We agree that reference standards often involve some degree of error or user-dependence. However, real world considerations compelled us to use practical definitions. Therefore, we applied strict predefined criteria for culture-negative extrapulmonary tuberculosis or not extrapulmonary tuberculosis cases, and we classified them without knowledge of the results of the tuberculin skin test and ELIS- POT assay to avoid verification bias. Furthermore, 12 of the 19 patients with probable extrapulmonary tuberculosis were classified based on histopathologic findings. A misclassification bias is less likely in view of the intermediate prevalence of tuberculosis in South Korea 11 and the rarity of reported cases of endemic fungi or brucellosis in Korea that could produce similar histopathologic features. Most previous studies of extrapulmonary tuberculosis have used the same clinical criteria for diagnosing extrapulmonary tuberculosis because the low bacillary count from relatively inaccessible sites of extrapulmonary tuberculosis makes bacteriological confirmation difficult. 10,14,16,19 In this study, patients with possible tuberculosis were excluded from the final analysis. However, when these are included, the sensitivity and specificity of the positive ELISPOT assay for diagnosing extrapulmonary tuberculosis are unaffected (sensitivity 91% and specificity 66%). In conclusion, the outcome of this study suggests that the ELISPOT assay is a useful adjunct test for excluding active tuberculosis in patients with suspected extrapulmonary tuberculosis regardless of immunosuppressive condition. References 1. Mehta JB, Dutt A, Harvill L, Mathews KM. Epidemiology of extrapulmonary tuberculosis. A comparative analysis with pre-aids era. Chest. 1991;99:1134-1138. 2. Keane J, Gershon S, Wise RP, et al. Tuberculosis associated with infliximab, a tumor necrosis factor -neutralizing agent. N Engl J Med. 2001; 345:1098-1104. 3. Cantwell MF, Snider DE, Jr, Cauthen GM, et al. Epidemiology of tuberculosis in the United States, 1985 through 1992. JAMA. 1994; 272:535-539. 4. Brodie D, Schluger NW. The diagnosis of tuberculosis. Clin Chest Med. 2005;26:247-271. 5. Kobashi Y, Mouri K, Obase Y, et al. Clinical evaluation of QuantiFERON TB-2G test for immunocompromised patients. Eur Respir J. 2007;30:945-950. 6. Huebner RE, Schein MF, Bass JB. The tuberculin skin test. Clin Infect Dis. 1993;17:968-975. 7. Thomas MM, Hinks TSC, Raghuraman S, et al. Rapid diagnosis of Mycobacterium tuberculosis meningitis by enumeration of cerebrospinal fluid antigen-specific T-cells. Int J Tuberc Lung Dis. 2008;12:651-657. 8. Pai M, Riley LW, Colford JM. Interferon-gamma assays in the immunodiagnosis of tuberculosis: a systematic review. Lancet Infect Dis. 2004;4:761-766. 9. Kim SH, Choi SJ, Kim HB, et al. Diagnostic usefulness of a T-cell-based assay for extrapulmonary tuberculosis. Arch Intern Med. 2007;167:2255-2259. 10. Liebeschuetz S, Bamber S, Ewer K, et al. Diagnosis of tuberculosis in South African children with a T-cell-based-assay: a prospective cohort study. Lancet. 2004;364:2196-2203. 11. Kang YA, Lee HW, Yoon HI, et al. Discrepancy between the tuberculin skin test and the whole-blood interferon gamma assay for the diagnosis of latent tuberculosis infection in an intermediate tuberculosis-burden country. JAMA. 2005;293:2756-2761. 12. Altman DG. Some common problems in medical research. In: Altman DG. Practical Statistics for Medical Research, 1st edition. London, UK: Chapman & Hall; 1991:397-439. 13. Cinar F, Cinar S, Yilmaz B, Gursel O. Purified protein derivative: the vital part of the cervical tuberculous adenitis diagnosis. Otolaryngol Head Neck Surg. 2003;129:245-247. 14. Polesky A, Grove W, Bhatia G. Peripheral tuberculous lymphadenitis. Epidemiology, diagnosis, treatment, and outcome. Medicine (Baltimore). 2005;84:350-362. 15. Lalvani A, Pathan AA, McShane H, et al. Rapid detection of Mycobacterium tuberculosis infection by enumeration of antigen-specific T cells. Am J Respir Crit Care Med. 2001;163:824-828. 16. Meier T, Eulenbruch HP, Wrighton-Smith P, et al. Sensitivity of a new commercial enzyme-linked immunospot assay (T SPOT-TB) for diagnosis of tuberculosis in clinical practice. Eur J Clin Microbiol Infect Dis. 2005;24:529-536. 17. Ferrara G, Losi M, D Amico R, et al. Use in routine clinical practice of two commercial blood tests for diagnosis of infection with Mycobacterium tuberculosis: a prospective study. Lancet. 2006;367:1328-1334. 18. Dewan PK, Grinsdale J, Kawamura LM. Low sensitivity of a wholeblood interferon-gamma release assay for detection of active tuberculosis. Clin Infect Dis. 2007;44:69-73. 19. Lee JY, Choi HJ, Park IN, et al. Comparison of two commercial interferon-gamma assays for diagnosing Mycobacterium tuberculosis infection. Eur Respir J. 2006;28:24-30. 20. Dheda K, Lalvani A, Miller RF, et al. Performance of a T-cell-based diagnostic test for tuberculosis infection in HIV-infected individuals is independent of CD4 cell count. AIDS. 2005;19:2038-2041. 21. Goletti D, Butera O, Bizzoni F, et al. Region of difference 1 antigenspecific CD4 memory T cells correlate with a favorable outcome of tuberculosis. J Infect Dis. 2006;194:984-992. 22. Nicol MP, Pienaar D, Wood K, et al. Enzyme-liked immunospot assay responses to early secretory antigenic target 6, culture filtrate protein 10, and purified protein derivative among children with tuberculosis: implications for diagnosis and monitoring of therapy. Clin Infect Dis. 2005; 40:1301-1308. 23. Wilkinson KA, Kon OM, Newton SM, et al. Effect of treatment of latent tuberculosis infection on the T cell response to Mycobacterium tuberculosis antigens. Clin Infect Dis. 2006;193:354-359. 24. Chee CB, KhinMar KW, Gan SH, et al. Latent tuberculosis infection treatment and T-cell responses to Mycobacterium tuberculosis-specific antigens. Am J Respir Crit Care Med. 2007;175:282-287.