Usefulness of Tuberculin Skin Test and Three Interferon-Gamma Release Assays for the Differential Diagnosis of Pulmonary Tuberculosis

Transcription

1 ORIGINAL ARTICLE Usefulness of Tuberculin Skin Test and Three Interferon-Gamma Release Assays for the Differential Diagnosis of Pulmonary Tuberculosis Yoshihiro Kobashi, Masaaki Abe, Keiji Mouri, Yasushi Obase, Naoyuki Miyashita and Mikio Oka Abstract Background We compared the usefulness of tuberculin skin test (TST) and three interferon-gamma release assays (IGRAs) [QuantiFERON-TB Gold (QFT-G), QuantiFERON-TB Gold In-tube (QFT-G), T-SPOT.TB] as the supportive method for diagnosing pulmonary tuberculosis (TB). Methods The subjects were 66 patients who required clinical differentiation of pulmonary TB. The final clinical diagnosis of pulmonary TB in patients and non-pulmonary TB in patients was established by clinical specimens. Results In patients with pulmonary TB, the positive response rate was 9.% on TST, 8.8% on QFT- G, 86.% on QFT-G and 9.% on T-SPOT.TB. In patients with non-pulmonary TB disease, the positive response rate was.9% on TST, 6.8% on QFT-G, 6.8% on QFT-G,.6% on T-SPOT.TB. results on three IGRAs were recognized in one patient each on QFT-G and QFT-G among patients with pulmonary TB and in two patients each on QFT-G and QFT-G among patients with non-pulmonary TB. However, there were no indeterminate results on T-SPOT.TB in either patient group. Patients with falsenegative or indeterminate results on IGRAs had severe underlying diseases or were receiving immunosuppressive treatments. Conclusion There were no significant differences among the three IGRA tests in this study. However, because the three IGRA tests showed a significantly higher positive response rate for patients with pulmonary TB and a lower positive response rate for patients with non-pulmonary TB than TST, the three IGRA tests seemed to be more useful than TST for the differentiation of patients with pulmonary TB. Key words: interferon-gamma release assays (IGRAs), tuberculin skin test (TST), pulmonary TB (Intern Med : 99-, ) (DOI:.69/internalmedicine..7) Introduction Physicians sometimes encounter difficulty in establishing a correct diagnosis because patients cannot expectorate sputum for differentiation of pulmonary tuberculosis (TB). Although bronchoscopic and percutaneous procedures are invasive methods, they are useful to obtain a definitive diagnosis of pulmonary TB (, ). However, tuberculin skin test (TST) has frequently been used as a noninvasive supportive immunological diagnostic tool in the diagnosis of TB infection, even though TST has many well-known limitations (, ). Two blood tests, based on the detection of interferongamma produced by T-cells in response to antigens specific to Mycobacterium tuberculosis (MTB) and encoded by the RD region, have recently become available as noninvasive supportive diagnostic methods for TB infection. QuantiFERON-TB Gold (QFT-G) and QuantiFERON-TB Gold In-tube (QFT-G) (Cellestis Ltd., Chadstone, Australia) are both based on whole-blood ELISA and QFT-G has been approved as an in vitro diagnostic method since, while QFT-G has been approved since 9 in Japan. T- Division of Respiratory Diseases, Department of Medicine, Kawasaki Medical School, Japan Received for publication April, ; Accepted for publication January 7, Correspondence to Dr. Yoshihiro Kobashi, yoshihiro@med.kawasaki-m.ac.jp 99

2 Intern Med : 99-, DOI:.69/internalmedicine..7 SPOT.TB (Oxford Immunotec, Abingdon, UK) is based on in vitro overnight ELISPOT assay. Although it has been approved as an in vitro diagnostic method in Europe, it has not yet been approved in Japan. There are few published data on the usefulness for differentiation between pulmonary TB and other pulmonary diseases (nontuberculous mycobacterial (NTM) disease, lung cancer, pulmonary mycosis or lung abscess) using TST and three IGRAs for the same patients. Therefore, we performed a prospective comparative study of the usefulness of TST and three IGRAs as supportive noninvasive diagnostic methods for pulmonary TB. Study population Materials and Methods Ninety-two patients who required the differentiation of pulmonary TB containing miliary TB clinically were initially enrolled in this study. However, 6 of the enrolled patients were excluded for the following reasons: the final diagnosis could not be established because the causative microorganisms could not be identified from the clinical specimens in 8 patients, the results of the three IGRAs might have been influenced by previous antituberculous treatment in 8 patients. Finally, we performed a prospective comparative analysis in the remaining 66 patients. This study was approved by the Ethical Committee of Kawasaki Medical School. All patients were examined between 9 and. These patients were diagnosed at Kawasaki Medical School Hospital and four affiliated hospitals. All patients showed negative serological findings for HIV or an absence of obvious risk factors for this disease. The final diagnosis of 66 patients consisted of with pulmonary TB (including with pulmonary tuberculoma) and patients with non-pulmonary TB without a past history of TB [ with pulmonary nontuberculous mycobacterial disease (NTM), 9 with pneumonia, with lung cancer, with lung abscess, with pulmonary mycosis, with sarcoidosis]. We performed a prospective comparative study of the usefulness of TST and three IGRA tests (QFT-G, QFT- G and T-SPOT.TB). To obtain the correct diagnosis, we isolated MTB from clinical specimens such as expectorated sputum or bronchoscopy in patients with pulmonary TB and confirmed satisfaction of diagnostic criteria for NTM diseases proposed by American Thoracic Society (ATS) guidelines () in patients with pulmonary NTM diseases isolated causative microorganisms from clinical specimens such as expectorated sputum or bronchoscopy and recognized the typical clinical course (improvement of clinical findings following the administration of appropriate antibiotics) or confirmed the histological diagnosis from resected surgical specimens in 9 patients with pneumonia, with lung abscess, with pulmonary mycosis, and the histological diagnosis from the resected surgical specimens in patients with lung cancer, with sarcoidosis. The diagnostic methods used to obtain the clinical diagnosis were as follows; patients were diagnosed from bronchoscopic specimen (brushing or bronchoalveolar lavage fluid), by video-assisted thoracoscopic surgery (VATS), and by conventional sputum culture in the pulmonary TB group, while 6 patients were diagnosed from bronchoscopic specimens, by CT-guided lung biopsy, by VATS, and 9 by conventional sputum culture in the non-pulmonary TB group. We collected data regarding any history of previous TB disease and risk factors for TB disease such as a history of immunosuppressive condition or treatment with immunosuppressive drugs within three months before enrollment of this study. Information regarding any previous Mantoux TST results and BCG vaccination, as well as information about clinical symptoms, laboratory findings, radiological findings was collected from each patient at the time of enrollment in this study. All 66 patients had clinical findings resembling pulmonary TB that were difficult to completely distinguish from pulmonary TB even by a respiratory specialist. Sample collection and TST A heparinized blood sample was collected from each patient by vein puncture and aliquots were used for the three IGRA tests. Immediately following the simultaneous collection of three blood samples, Mantoux TST was administered. For TST,. ml of tuberculin purified protein derivative (PPD; Nippon BCG, Tokyo, Japan; equivalent to three tuberculin units of PPD-S) was injected intradermally into the volar aspects of the forearm, and the transverse induration diameter was measured 8 hours later. The results of this test were judged according to the ATS guidelines (6); namely, if the transverse induration diameter of TST was < mm, TST was considered negative and if mm, TST was considered positive. QFT-G and QFT-G test QFT-G and QFT-G test was performed according to the recommendations of the manufacturer (Cellestis Ltd., Carnegie, Australia). The judgment was performed according to the guideline proposed by the Centers for Disease Control and Prevention (CDC) for using QFT-G and QFT-G (7). T-SPOT.TB T-SPOT.TB was performed according to the recommendations of the manufacturer (Oxford Ltd., Oxfordshire, UK) and the test result was judged according to UK guidelines for using T-SPOT.TB to diagnose TB published by NCCCC (8). Statistical analysis Information from the questionnaires, TST results, and three IGRAs (QFT-G, QFT-G were statistically analyzed to determine the following: ) the positive response rate and negative response rate of QFT-G, QFT-G and T-SPOT.TB

3 Intern Med : 99-, DOI:.69/internalmedicine..7 Table. Background Characteristics of the Patients Enrolled in this Study Pulmonary TB (n=) Non-pulmonary TB (n=) p value Age, years (Mean±S.D.) Gender (Male:Female) Previous physical (Bedridden) Smoking his ory (+) Alcohol abuse (+) Underlying disease ( on) Respir ory disease Old pulmonary uberculosis Bronchial as ma Lung cancer COPD O hers Non- disease Ga disease Chronic renal failure Maligna disease Cardiovascular disease Diabe es melli s Collagen vascular disease Neurological disease O hers Immunosuppressive re men (+) Previous ous rea (+) 6.±. 7: (8%) 7 (77%) (%) 8 (8%) (%) 6 (77%) (%) (9%) 6.±. : 7(6%) (7%) (%) (77%) 9(%) (7%) 6 (%) (%) COPD : Chronic pulmonary disease, N.S. : compared with those of TST; ) the proportion of QFT-G, QFT-G and T-SPOT.TB showing an indeterminate result; ) concordance and discordance among TST, QFT-G, QFT-G and T-SPOT.TB in patients with pulmonary TB and nonpulmonary TB disease; and ) the sensitivity for TST, QFT- G, QFT-G and T-SPOT.TB in patients with pulmonary TB. The concordance rates between two test results (TST and QFT-G, TST and QFT-G, TST and T-SPOT.TB, QFT- G and QFT-G, QFT-G and T-SPOT.TB, QFT-G and T- SPOT.TB) were assessed using kappa coefficient. The TST, QFT-G, QFT-G and T-SPOT.TB results were compared using Chi-square test. Throughout this study, confidence intervals (CI) are given as 9%. A p value of less than. was considered to indicate a significant difference. Results The background characteristics of the pulmonary TB group ( patients) and non-pulmonary TB group ( patients) are shown in Table. Although males were significantly dominant among patients with pulmonary TB, there were no significant differences in other background characteristics between the two groups. The comparison of results among TST and the three IGRAs (QFT-G, QFT-G and T-SPOT.TB) in patients with pulmonary TB and non-pulmonary TB is shown in Table. The positive response rate on three IGRAs for the pulmonary TB group was significantly higher than that on TST. Among the three IGRAs, the positive response rate on T- SPOT.TB was highest (9.%) and there were no indeterminate results on T-SPOT.TB in either patients group. Among patients with pulmonary TB, seven patients (.8%) showed discordant results among TST and the three IGRAs (Fig. ). Two patients were QFT-G and QFT-G negative/t-spot.tb positive, suggesting that T-SPOT.TB is more sensitive than the two QFT tests for diagnosing TB infection. These two patients consisted of one patient receiving immunosuppressive treatment and one with advanced lung cancer. They both showed severe lymphocytopenia with CD lymphocytopenia and hypoalbuminemia due to underlying disease (Table ). Five patients showed TST negative/ three IGRAs positive results. The clinical findings of patients with pulmonary TB who showed negative test results on the three IGRAs and showed an indeterminate result on both QFT-G and QFT-G tests are shown in Table. There was one patient who showed a negative result on all three IGRAs. He had severe underlying disease and exhibited both lymphocytopenia and severe hypoalbuminemia. Although one patient showed a negative result on QFT-G only, the reason was unknown because there was no lymphocytopenia or hypoalbuminemia in that patient. One patient showed a positive response on T-SPOT. TB and indeterminate result on both QFT tests. He was an elderly patient with bed ridden status due to severe underlying disease and presented with both lymphocytopenia and hypoalbuminemia. The positive response rate of TST and three IGRA tests for patients with non-pulmonary TB is shown in Fig.. The positive response rate of three IGRAs was significantly lower than that on TST. There were three patients with pulmonary diseases other than TB who demonstrated positive results on all three IGRAs. The clinical findings of these three patients are shown in Table. The final diagnoses were pulmonary M. kansasii disease in two patients and pneumonia in one. One of three patients showed positive responses of TST. None of the patients demonstrated lympho-

4 Intern Med : 99-, DOI:.69/internalmedicine..7 Table. Comparison of the Results of TST and Three IGRAs Tests for Patients with Pulmonary TB and Patients with Non-pulmonary TB Disease Diagnos c method Judgment Pulmonary TB (n=) Non-pulmonary TB (n=) TST (9.%) 9 (.9%) p=. 8 (.9%) 6 (9.%) QFT-G p=.8 8 (8.8%) (.7%) (6.8%) 9 (88.6%) p=.9 (.%) (.6%) QFT-G 9 (86.%) (9.%) (6.8%) 9 (88.6%) (.%) (.6%) T-SPOT-TB (9.%) (.%) 6 (.6%) 8 (86.%) (%) (%) COPD : Chronic pulmonary disease, N.S. : Not TST(+) QFT-G(+) QFT-G(+) T-SPOT. TB(+) TST(+) : (9.%) QFT-G(+) : 8(8.8%) QFT-G(+) : 9(86.%) T-SPOT. TB(+) : (9.%) Figure. Positive response rate of TST and three IGRA tests for patients with pulmonary TB ( cases). cytopenia and hypoalbuminemia. Three patients were QFT negative/t-spot.tb positive. The clinical findings of these three patients are shown in Table. The final diagnoses were pulmonary cryptococcosis in one patient, pulmonary NTM disease due to M. intracellulare in one, and pulmonary abscess in one, respectively. None of these patients demonstrated lymphocytopenia or hypoalbuminemia, but one of the three patients demonstrated a positive response of TST. Although two patients on QFT-G test and two patients on QFT-G test demonstrated indeterminate results among 8 patients with non-pulmonary TB, these patients showed negative responses on T-SPOT.TB. They all showed severe lymphocytopenia and hypoalbuminemia due to underlying diseases (not shown in this manuscript). The sensitivity (with 9% confidence intervals) of QFT- G for the differential diagnosis between pulmonary TB and non-pulmonary TB was 8.8% (7. to 88.%), that of QFT-G was 86.% (7. to 88.%), that of TST was 9.% (. to 67.%), and that of T-SPOT.TB was 9.% (9. to 98.%). The negative response rate of QFT-G among patients with non-pulmonary TB was 88.6% (8. to 9.%), that of QFT-G was 88.6% (8. to 9.%), that of TST was 9.% (. to 66.%), and that of T-SPOT.TB was 86.% (8. to 9.%). Among 6 patients with determinate results on three IGRAs, although there was good agreement between the three IGRA results, there was poor agreement between TST and the three IGRA results. The concordance rates between TST and QFT-G, TST and QFT-G, TST and T-SPOT.TB, QFT-G and QFT-G, QFT- G and T-SPOT.TB, and QFT-G and T-SPOT.TB were 6.% (kappa.8), 6.6% (kappa.), 7.6% (kappa.), 98.% (kappa.8), 9.% (kappa.76), 9.9% (kappa.78), respectively. The concordance rate among all three IGRA results showed good agreement (9.%; kappa.766). Discussion Although there have recently been many comparative studies of TST and IGRA tests examining the usefulness as the supportive diagnostic methods for TB disease (9-), there are few reports comparing simultaneous TST and three IGRA tests (QFT-G, QFT-G and T-SPOT.TB) in the same patients to differentiate between pulmonary TB and other pulmonary diseases. Therefore, we prospectively performed three IGRA tests (QFT-G and QFT-G are commercially available in Japan, but T-SPOT.TB has not yet been approved by the Ministry of Health, Labour and Welfare, Japan) as useful noninvasive diagnostic methods for pulmo-

5 Intern Med : 99-, DOI:.69/internalmedicine..7 Table. Clinical Findings of Patients with Pulmonary TB who Showed Negative Results on Three IGRAs and Results on QFT-G and QFT-G Tests False ne result on three IGRAs ( case) Age, Sex Underlying disease 7.M Alcoholic li r cirrhosis WBC count (/ L) Lymphocyte count (/ L) CD Lymphocyte count (/ L) Albumin (g/dl) HIV AFB test result Radiological TST ndings* Smear Culture. (+) (+) (Sputum) Cavity Clinical diagnosis Pulmonary TB + TB pleur s False ne result on QFT-G and QFT-G ( cases) 8.M RPGN (Steroid on) (+) (+) (BALF) 8,M Lung cancer 8. Cavity Pulmonary TB (+) (+) Pulmonary TB (Sputum) Cavity(+) False ne result on QFT-G ( case) 6.M (+) (+) (BALF) Left Cavity Pulmonary TB result on QFT- G and QFT-G ( case) 9, F Old cerebral infarction (+) (+) Pulmonary TB Cavity AFB: acid-fast bacilli, RPGN: Rapidly progressive glomerulonephritis HIV: Human immunodeficiency virus, TST: Tuberculin skin test *: Data denote portion, extension and cavity (findings): ; denotes within onethird of unilateral lung field, ; within unilateral lung field TST(+) 6 QFT-G(+) QFT-G(+) T-SPOT. TB(+) TST(+), T-SPOT. TB(+) : TST(+) : 8 (.9%) QFT-G(+) : (6.8%) QFT-G(+) : (6.8%) T-SPOT. TB(+) : 6 (.6%) Figure. Positive response rate of TST and three IGRAs tests for patients with non-pulmonary TB disease ( cases). nary TB compared to the results of TST. As a result, the sensitivity was as follows: 9.% on TST, 8.8% on QFT- G, 86.% on QFT-G, and 9.% on T-SPOT.TB, while the negative response rate for patients with non-pulmonary TB was 9.% on TST, 88.6% on QFT-G, 88.6% on QFT- G, and 86.% on T-SPOT.TB. The highest sensitivity was shown by T-SPOT.TB and the highest negative response rate was shown by QFT-G and QFT-G. Although there were no significant differences in the sensitivity among the three IGRA tests, three IGRA tests were all significantly better than TST in both sensitivities for the diagnosis of active TB in this series. Diel et al. summarized the sensitivity and specificity of these two IGRA tests in previous reports (). The sensitivity of T-SPOT.TB (88%) was better than that of QFT-G (8%), while the specificity of QFT-G (99%) was better than that of T-SPOT.TB (86%) for detecting confirmed active TB. The main differences between the two commercial IGRAs (QFT and T-SPOT.TB) are the technique of IFN-γ detection (ELISA method versus ELISPOT method) and the specimens used (whole blood versus isolated mononuclear cells). Although T-SPOT.TB seems to involve a more complex technique than QFT, it is more sensitive for the detection of single IFN-γ-producing cells. results of QFT test were recognized in one patient with pulmonary TB and in three patients with nonpulmonary TB because they all showed positive control failure. However, there were no patients who showed an indeterminate response on T-SPOT.TB in either patient group. All patients who showed an indeterminate response on QFT test had severe lymphocytopenia including CD lymphocytopenia and hypoalbuminemia just as we described in our previous reports (, 6). However, the reason for falsenegative IGRA results in patients with active TB (Table ) is unknown. Although we speculated that the reason for the negative response on IGRA tests was mainly related to the limited dose of MTB in the sputum due to a small lesion rather abnormal laboratory findings (), three of four patients with pulmonary TB had severe lymphocytopenia and hypoalbuminemia due to underlying disease (Table ). Concerning the final diagnosis in this study, several respiratory specialists have used appropriate diagnostic methods to divide patients with pulmonary TB from those with nonpulmonary TB according to restricted diagnostic criteria in other series. The final diagnosis of patients with pulmonary TB was confirmed by the isolation of MTB from cul-

6 Intern Med : 99-, DOI:.69/internalmedicine..7 Table. Clinical Findings of Patients with Non-pulmonary TB Disease who Showed Positive IGRA Test Results Case 6 Age: Sex,M 7,M 7,M 8,M 7,M 68,M Underlying disease DM CVD HT an er ML CRF COPD WBC ount Cells (/ ) Lympho ytecdlympho yte ount ells ount ells (/ ) (/ ) Serum albumin (g/dl) HIV TST (Dura on, mm) Radiologi al ndings* Left Clini al diagnosis Pulmonary NTM disease (M. kansasii) Pulmonary NTM disease (M. kansasii) Pneumonia Unknown Pulmonary r Cavity Cavity Pulmonary abs ss (Unknown) Pulmonary NTM disease (M. intracellulare) QFT-G, QFT-G, T-SPOT.TB all positive T-SPOT.TB positive DM: Diabetes mellitus, CVD: Cerebrovascular disease, HT: Hypertension, TB: Tuberculosis, ML: Malignant lymphoma, CRF: Chronic renal failure, WBC: White blood cells, HIV: Human immunodefficiency virus, TST: Tuberculin skin test, NTM: Nontuberculous mycobacterium *: Data denote portion,extension and cavity (findings); ; within one-third of unilateral lung field, ; within unilateral lung field ture of sputum, brushing, bronchoalveolar lavage fluid, and transbronchial lung biopsy. The final diagnosis of patients with pulmonary NTM disease (Causative microorganisms consisted of Mycobacterium avium in patients, Mycobacterium intracellulare in 8, and Mycobacterium kansasii in ) was confirmed by satisfying the diagnostic criteria of ATS guidelines (). The final diagnosis of pneumonia, lung abscess, and pulmonary mycosis was confirmed according to the criteria used in our previous study (7). The final diagnosis of lung cancer, lung abscess, and pulmonary mycosis was also confirmed by histological diagnosis of surgically resected specimens. There were three patients who showed positive responses on all IGRAs despite having non-pulmonary TB (Table ). However, this finding can be explained in two patients who had pulmonary disease due to Mycobacterium kansasii which has the same ESAT-6 and CFP- MTB-specific antigens. However, the reason for false positive IGRA result in one patient with pneumonia (etiology was unknown) remains unknown. Regarding the agreement among TST and three IGRAs, while there was good agreement between three IGRAs (QFT-G, QFT-G and T-SPOT.TB) (kappa=.766), there was poor agreement between TST and three IGRAs. The reason is that in vitro IGRAs are superior to TST for obtaining a diagnosis of pulmonary TB. Because TST presents a positive response in BCG-vaccinated patients and patients with NTM disease, this discrepancy was caused by the high false positive rate on TST. In this study, (6.%) patients previously received BCG vaccination. The present study has several limitations. First, although this study was a medium-scale study of patients with various underlying diseases in a community hospital-based population suspected of active pulmonary TB using three IGRAs, it was geographically restricted to a small area in Japan. We hope that a large scale nationwide study will be performed in Japan. Secondly, our study consisted of a high proportion of immunosuppressed patients with a high risk of TB. Therefore, the result of this study cannot be generalized with regard to the general population. However, it is noteworthy that this study assessed such high-risk populations that are more likely to be candidates for testing with IGRAs. In conclusion, the present study demonstrated that three IGRAs are significantly more sensitive than TST in diagnosing pulmonary TB and should replace TST for diagnosing pulmonary TB in community hospitals. Among the three IGRA tests, T-SPOT.TB showed the highest sensitivity in diagnosing pulmonary TB. On the other hand, T-SPOT.TB showed several false positive responses among patients with non-pulmonary TB. Since T-SPOT.TB has not yet been approved by the Ministry of Health, Labour and Welfare, we cannot use T-SPOT.TB commercially in Japan. It will be necessary to confirm whether T-SPOT.TB is useful as a supportive diagnostic method of pulmonary TB considering the findings of this study. The authors state that they have no Conflict of Interest(COI). Acknowledgement The authors would like to thank T. Matsushima (Kurashiki Daiichi Hospital, Kurashiki, Japan; Asahigaoka Hospital, Okayama, Japan), N. Okimoto (Kawasaki Medical School Kawasaki Hospital, Okayama, Japan) and J. Nakamura, H. Yoneyama (Kasaoka Daiichi Hospital, Kasaoka, Japan) for their constructive comments.

7 Intern Med : 99-, DOI:.69/internalmedicine..7 References. Wong CF, Yew WW, Chan CY, Au LY, Cheung SW, Cheng AF. Rapid diagnosis of smear-negative pulmonary tuberculosis via fiberotic bronchoscopy: utility of polymerase chain reaction in bronchial aspirates as an adjunct to transbronchial biopsies. Respir Med 9: 8-89, Joos L, Patuto N, Chhajed PN, Tamm M. Diagnostic yield of flexible bronchoscopy in current clinical practice. Swiss Med Wkly 6: -9, 6.. Huebner RE, Schein MF, Bass JB Jr. The tuberculin skin test. Clin Infect Dis 7: , 99.. Tissot F, Zanetti G, Francioli P, Zellweger JP, Zysset F. Influence of bacilli Calmete-Guerin vaccination on size of tuberculin skin test reaction: to what size? Clin Infect Dis : -7,.. Griffith DE, Alsamit T, Brown-Elliot BA, et al. An official ATS/ IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 7: 67-6, American Thoracic Society and the Centers for Disease Control and Prevention. Diagnostic standards and classification of tuberculosis in adults and children. Am J Respir Crit Care Med 6: 76-9,. 7. Mazurek GH, Jereb J, Lobue P, et al. Guidelines for using QuantiFERON-TB Gold test for detecting Mycobacterium tuberculosis infection, United States. MMWR Recomm Rep : 9-,. 8. National Collaborating Center for Chronic Conditions. Tuberculosis: Clinical diagnosis and management of tuberculosis, and measures for its prevention and control. UM: Royal College of Physicians, London, 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 67: 8-, 6.. Lee JY, Choi HJ, Park IN, et al. Comparison of two commercial interferon-γ assays for diagnosing Mycobacterium tuberculosis infection. Eur Respir J 8: -, 6.. Arend SM, Thijsen SF, Leyten EM, et al. Comparison of two interferon-gamma assays and tuberculin skin test for tracing TB contacts. Am J Respir Crit Care Med 7: 68-67, 7.. Harada N, Higuchi K, Yoshiyama T, et al. Comparison of the sensitivity and specificity of two whole blood interferon-gamma assays for M.tuberculosis infection. J Infect 6: 8-, 8.. Higuchi K, Kawabe Y, Mitarai S, Harada N, Mori T. Comparison of performance in two diagnostic methods for tuberculosis infection. Med Microbiol Immunol 98: -7, 9.. Diel R, Loddenkemper R, Nienhaus A. Evidence-based comparison of commercial interferon-γ release assays for detecting active TB: a metaanalysis. Chest 7: 9-968, 7.. Kobashi Y, Mouri K, Obase Y, Fukuda M, Miyashita N, Oka M. Clinical evaluation of QuantiFERON TB-G test for immunocompromised patients. Eur Respir J : 9-9, Kobashi Y, Sugiu K, Mouri K, Obase Y, Miyashita N, Oka M. results of QuantiFERON TB-G test performed in routine clinical practice. Eur Respir J : 8-8, Kobashi Y, Yoshida K, Miyashita N, Niki Y, Matsushima T. Evaluating the use of a Streptococcus pneumoniae urinary antigen detection kit for the management of community-acquired pneumonia in Japan. Respiration 7: 87-9, 7. The Japanese Society of Internal Medicine