Frequency of tuberculin-reactive T-lymphocytes in pleural fluid and blood from patients with tuberculous pleurisy

A limiting dilution assay was used to determine the frequency of T-lymphocytes reactive to purified protein derivative of tuberculin (PPD). Pleural fluid from patients with tuberculous pleurisy showed higher frequencies of PPD-reactive T-lymphocytes than peripheral blood from the same patients or tuberculin-positive healthy control subjects. The mean frequencies were 1/2,204 T cells in pleural fluid from tuberculous pleurisy, 1/14,970 T cells in peripheral blood from the same patients, and 1/13,130 T cells in peripheral blood from healthy controls. The concentration of tuberculin-reactive lymphocytes in tuberculous pleural fluid could represent selective accumulation or in situ expansion of this population of cell.     

High level of interferon gamma in tuberculous pleural effusion

It has been observed that T-lymphocytes of patients with tuberculosis produce interferon gamma (IFN gamma) in vitro. Based on this idea, we studied IFN gamma in pleural fluid and serum. We studied 80 patients with pleural effusion; 30 patients with tuberculous pleurisy had high IFN gamma concentrations in pleural fluid. Patients with malignant pleural effusions, nonspecific pleural effusion, parapneumonic effusions and pleural transudates had low levels. The IFN gamma levels were higher in those with massive tuberculous effusion and apparent pulmonary lesion on x-ray film. We found that the T4/T8 lymphocyte ratio was higher in pleural fluid than in peripheral blood. Numbers of T3 and T4 lymphocytes were higher in tuberculous pleural effusions compared with those in other patients. There is no correlation between IFN gamma levels and lymphocyte subsets in pleural effusion. Perhaps pleural T-lymphocytes produce IFN gamma after stimulation by mycobacterial antigens and this lymphokine activates macrophages, increasing their bactericidal activity against Mycobacterium tuberculosis.     

Cooperation between accessory cells and T lymphocytes in patients with tuberculous pleurisy

We studied interleukin 1 (IL-1) activity of pleural fluid macrophages and peripheral blood monocytes obtained from ten patients with tuberculous pleurisy and ten patients with malignant pleurisy, using purified protein derivative (PPD) as a stimulating agent. Tuberculous pleural fluid macrophages and peripheral blood monocytes tended to produce higher IL-1 activity than malignant pleural fluid macrophages and blood monocytes and showed significantly more IL-1 activity than healthy control monocytes. However, no significant difference in IL-1 activity was observed between tuberculous pleural macrophages and blood monocytes. With the cooperation of these accessory cells, pleural fluid T lymphocytes in patients with tuberculous pleurisy showed a significant level of interleukin 2 (IL-2) activity in the presence of PPD. Tuberculous pleural fluid macrophages promoted greater IL-2 production than blood monocytes from either tuberculous pleural fluid or blood T lymphocytes despite relative equivalence in measured IL-1 production. Combination of tuberculous pleural fluid macrophages and pleural fluid T lymphocytes was the most effective for increasing IL-2 activity when compared with other combinations. These results suggest that tuberculous pleural fluid macrophages and T lymphocytes may contribute to the immunopathogenesis of tuberculosis at a local site of disease.     

Clinical immunology of tuberculosis

The standard tuberculin skin test has been known as the prototype of delayed type hypersensitivity testing which is mediated by T cells and macrophages and plays an important role in the pathogenesis of tuberculosis. Tuberculosis is indeed a chronic infectious disease, but variation in the host immune responses to tubercle bacilli results in the various clinical manifestations of the disease ranging from an immunologically hyperreactive state observed in pleural fluid lymphocytes in tuberculous pleurisy to an almost totally unresponsive state observed in those severely ill with refractory tuberculosis. In tuberculous pleurisy, T cells in pleural fluid respond remarkably in vitro to PPD tuberculin whereas T cells in peripheral blood responded poorly to PPD stimulation. Compartmentalization of PPD-reactive T cells in the pleural fluid and immunosuppression by T cells and/or macrophages in the peripheral blood were responsible for this immunological difference observed between the lymphocytes in pleural fluid and those in peripheral blood of tuberculous pleurisy. In advanced, drug-resistant tuberculosis as well as in nontuberculous mycobacterial infection, the proliferative responses of T cells in vitro to PPD stimulation were impaired. This depressed T cell response was due to depressed interleukin-2 (IL-2) production and not due to depressed IL-2 responsiveness. Therefore, the addition of exogenous IL-2, returned the depressed PPD-induced lymphocyte proliferation in vitro in these patients to the level of the response observed in lymphocytes from patients with newly-diagnosed tuberculosis. Our results suggest that recombinant IL-2 offers a novel approach to the therapy of advanced, drug-resistant tuberculosis and nontuberculous mycobacterial infection. Preliminary clinical trials of immunotherapy with recombinant IL-2 reveals the effectiveness of this therapy and encourages us to extend the trial to a larger scale. Tubercle bacilli have various biological activities. Research on tuberculosis and tubercle bacilli have contributed much to the progress of biochemistry, pathology and immunology. Mycobacterium is a fascinating organism, which now presents another big appeal to those studying immunology: Study of immunological interaction between gamma delta T cells and the highly conserved protein in mycobacteria, HSP, heat shock protein will contribute to the elucidation of the mechanism of immunological surveillance and the mechanism of autoimmune diseases. In addition, it will also contribute to the development of a new mycobacterial vaccine which will give direct, protective immunity against tuberculosis.     

The significance of subpleural mononuclear cell infiltration in the differential diagnosis of pleurisy showing nonspecific histological findings

To determine whether or not patients who showed abundant lymphocytes in pleural effusion without any specific findings by pleural biopsy could be histopathologically differentiated between those with tuberculous and nontuberculous origin, we reexamined histology of pleural biopsies of all patients whose pleural effusion showed predominantly lymphocytes and did not contain malignant cells from January, 1984 to March, 1989. A total of 45 patients with a nonspecific histology of pleural biopsy were categorized based on their final diagnosis into three groups; tuberculous (n = 20), carcinomatous (n = 10) and nontuberculous, benign pleurisy (n = 15). Pleural biopsy of patients with nontuberculous, benign pleurisy frequently showed band-like infiltration of mononuclear cells in the subpleural adipose tissue (11 out of 15 patients), while the finding was significantly less frequent in those with tuberculous pleurisy (1 out of 20, p less than 0.01) and pleurisy associated with carcinoma (3 out of 10, p less than 0.05). Based on these findings, the presence of band-like infiltration of mononuclear cells in the subpleural adipose tissue in pleural biopsy of patients with abundant lymphocytes in pleural effusion strongly suggests that the pleurisy is nontuberculous origin.     

Subpleural mononuclear cell infiltration. Significance in the differential diagnosis of pleuritis showing nonspecific histologic findings

To determine if patients who had lymphocyte-rich pleural effusion and a pleural biopsy without any specific findings could be histopathologically differentiated between those with tuberculous and nontuberculous pleuritis, we histologically re-evaluated the pleural biopsies of all patients whose pleural effusion was predominant with lymphocytes and contained no malignant cells. A total of 40 patients with a nonspecific histologic findings of pleural biopsy specimen were categorized based on their ultimate diagnosis as having tuberculous (n = 15), carcinomatous (n = 10) or nontuberculous, benign pleuritis (n = 15). The pleural biopsy specimen of patients with nontuberculous, benign pleuritis frequently showed a band-like infiltration of mononuclear cells in the subpleural adipose tissue with minimal pleural inflammatory infiltrate (10 out of 15 patients), while the same finding was infrequent in those with tuberculous pleuritis (0 out of 15, p = 0.0001) and pleuritis associated with carcinoma (three out of 10, p = 0.082). Based on these results, the presence of band-like infiltration of mononuclear cells in the subpleural adipose tissue with minimal pleural inflammatory infiltrate in pleural biopsy specimens of patients with lymphocyte-rich pleural effusion suggests that the pleuritis is nontuberculous in its nature.     

Diagnostic significance of interferon-gamma in tuberculous pleural effusions

STUDY OBJECTIVES: Tuberculosis (TB), the single most frequent infectious cause of death worldwide, also is a major cause of pleural effusion, which in TB usually has lymphocytic and exudative characteristics. Differential diagnosis between TB and nontuberculous pleural effusion can be sometimes difficult, representing a critically important clinical problem. METHODS: We studied 46 patients presenting with pleural effusion to the National Sanyo Hospital between April 2000 and January 2001 (34 men and 12 women; mean age, 64 years). Ten patients (22%) had tuberculous pleurisy, 19 patients (41%) had malignant pleuritis, and 17 patients (37%) had pleural effusion due to an etiology other than tuberculosis or cancer. Pleural fluid concentrations of four suggested markers were measured using commercially available kits. RESULTS: The pleural fluid levels (mean +/- SE) of adenosine deaminase (83.3 +/- 18.2 U/L vs 25.8 +/- 20.4 U/L, p < 0.0001), interferon-gamma (137 +/- 230 IU/mL vs 0.41 +/- 0.05 IU/mL, p < 0.0001), immunosuppressive acidic protein (741 +/- 213 micro g/mL vs 445 +/- 180 micro g/mL, p < 0.001) and soluble interleukin 2 receptor (7,618 +/- 3,662 U/mL vs 2,222 +/- 1,027 U/mL, p < 0.0001) were significantly higher for tuberculous pleuritis than for other causes of effusion. Receiver operating characteristic analysis demonstrated that pleural fluid content INF-gamma was the best indicator of tuberculous pleurisy among four relevant biological markers. CONCLUSIONS: INF-gamma in pleural fluid is the most sensitive and specific among four biological markers for tuberculous pleuritis. Thus, our results suggest that determination of INF-gamma at the onset of pleural effusion is informative for the diagnosis of tuberculous pleuritis. Further studies including larger numbers of patients are needed to verify this result.     

Cytokine content in pleural effusion. Comparison between tuberculous and carcinomatous pleurisy

Tuberculous pleurisy is a good model for resolution of local cellular immunity. It would be expected that tuberculous pleural fluid contains a variety of immunologically important cytokines because of the accumulation of immunocompetent cells in the pleural cavity. We studied interleukin 1 (IL-1), interleukin 2 (IL-2), and interferon gamma (IFN-gamma) levels in pleural fluid of 20 patients with tuberculous pleurisy and compared them with those in pleural fluid of 20 patients with malignant pleurisy. We also evaluated adenosine deaminase (ADA) levels in both effusions. Tuberculous pleural fluid had higher levels of IL-1, IL-2, IFN-gamma, and ADA than malignant pleural fluid. Although the difference of IL-1 level between tuberculous and malignant pleural fluid was modest, that of IL-2, IFN-gamma, and ADA was dominant. These findings suggest that activated T lymphocytes in tuberculous pleural fluid concern the production of lymphokines at the morbid site and they effectively exert local cellular immunity through the action of such lymphokines.     

Elevated levels of soluble interleukin-2 receptors in tuberculous pleural effusions

The levels of soluble IL-2R were measured in pleural fluid from patients with tuberculosis pleurisy. There were significantly elevated soluble IL-2R values in tuberculous pleural fluid as compared with pleural fluid of nontuberculous etiology including malignant, bacterial and transudative pleural effusions. In patients with tuberculous pleurisy, the level of soluble IL-2R in pleural fluid was markedly greater than that in serum. Furthermore, a significant positive correlation was observed between soluble IL-2R levels and adenosine deaminase levels in tuberculous pleural fluid. These findings suggest that elevated levels of pleural fluid soluble IL-2R in tuberculous pleurisy could reflect the local proliferation of activated T-cells and may be clinically useful in the diagnostic procedures for patients with pleural tuberculosis.     

T and B lymphocytes in pleural effusions.

To determine the diagnostic significance of the determination of T and B lymphocytes in pleural fluid, we studied these cells in peripheral blood and in pleural fluid by means of surface markers. Our study comprised 30 patients suffering from pulmonary tuberculosis, pulmonary malignancy, connective tissue disease, nonspecific pleurisy or congestive cardiac failure. In pulmonary tuberculosis, both the percentage and absolute numbers of T lymphocytes in pleural fluid were significantly higher than in peripheral blood. In patients with pulmonary tuberculosis, pulmonary malignancy or nonspecific pleuritis, the percentages and absolute numbers of B lymphocytes were significantly lower in pleural fluid than in peripheral blood. Considered together with other clinical and laboratory indices, these determinations may aid in the differential diagnosis of pleurisy of various etiology.     

Elevated soluble CD26 levels in patients with tuberculous pleurisy.

SETTING: Several reports have shown that tuberculous infection elicits a Th1-like immune response with increased levels of IFN-gamma. Recently, expression of CD26 on CD4+ lymphocytes has been shown to correlate with the production of Th1-like cytokines. We therefore hypothesized that CD26 expression might increase in tuberculous pleural effusion, and might thus be a possible marker for detecting tuberculous pleurisy. OBJECTIVE AND DESIGN: To test this hypothesis, we measured soluble CD26 levels in the serum and pleural fluid of patients with tuberculous pleurisy (TB; n = 13), carcinomatous pleurisy (CA, n = 17), empyema (EM, n = 6), and congestive heart failure (HF, n = 10). RESULTS: The pleural CD26 levels, but not the serum CD26 levels, in patients with tuberculous pleurisy were significantly higher than those in other groups, and were correlated with levels of adenosine deaminase and interferon-gamma in the tuberculous pleural effusion. Furthermore, when the cut-off value for p-CD26 was set at 544.5 ng/ml, the positive rate for the TB group was significantly higher than that for the CA, EM and HF groups (P < 0.05). CONCLUSION: These results suggest that elevation of soluble CD26 in pleural fluid is implicated in Th1-like immune response, and may be a useful marker for tuberculous pleurisy.     

Comparison of six biological markers for the diagnosis of tuberculous pleuritis

STUDY OBJECTIVE: We sought a marker to differentiate tuberculous pleural effusions from nontuberculous pleural effusions, which otherwise can be difficult. PATIENTS: We studied 55 patients with pleural effusions, 20 (36%) with tuberculous pleuritis and 35 (64%) with a nontuberculous etiology. MEASUREMENTS AND RESULTS: Pleural fluid levels of adenosine deaminase, interferon (INF)-gamma, interleukin (IL)-12p40, IL-18, immunosuppressive acidic protein, and soluble IL-2 receptors were measured and were subjected to receiver operating characteristic analysis. INF-gamma had the greatest sensitivity and specificity for tuberculous pleuritis among the six biological markers studied. CONCLUSION: The determination of INF-gamma levels in pleural fluid is the most informative in the diagnosis of tuberculous effusion.     

Tumour necrosis factor-alpha in comparison to adenosine deaminase in tuberculous pleuritis

BACKGROUND: Adenosine deaminase (ADA) is already used for the differential diagnosis of tuberculosis pleurisy. Tumour necrosis factor-alpha (TNF) is another marker which has been investigated for this purpose. OBJECTIVE: We evaluated the diagnostic value of pleural fluid and serum TNF concentrations in tuberculous pleuritis and compared them to ADA. METHODS: Sixty-two patients (24 tuberculous pleuritis, 38 non-tuberculous pleuritis) with exudative pleurisy were included. Serum and pleural fluid TNF concentrations were determined in all patients and ADA activity in 54 patients. Pleural fluid TNF concentrations and pleural fluid/serum TNF were compared to pleural fluid ADA activity and pleural fluid/serum ADA. RESULTS: When the tuberculous and non-tuberculous groups were compared, pleural fluid TNF concentrations (65.4 +/- 136.9 pg/ml vs. 54.5 +/- 144.2 pg/ml, respectively; p < 0.001), pleural fluid ADA activity (74.2 +/- 33.3 U/l vs. 23 +/- 16.3 U/l; p < 0.0001), pleural fluid/serum TNF (2.55 +/- 5.23 vs. 0.26 +/- 0.2; p < 0.001) and pleural fluid/serum ADA (4.58 +/- 8.14 vs. 1.15 +/- 0.7; p < 0.0001) were significantly higher in the tuberculous group. When cut-off points were assessed, 8 pg/ml and 40 U/l were found for pleural fluid TNF concentrations and pleural fluid ADA activity, respectively. Sensitivity, specificity, area under the curve were 87.5%, 76.3%, 0.772 for pleural fluid TNF concentrations and 90.9%, 89.5%, 0.952 for pleural fluid ADA activity, respectively; the difference between these areas under the curves was significant (p < 0.05). CONCLUSIONS: Pleural fluid TNF levels and pleural fluid/serum TNF were higher in tuberculous effusions than in other exudates, but their diagnostic value appears to be poorer than that of ADA.     

Increased pleural fluid adenosine deaminase in brucellosis is difficult to differentiate from tuberculosis

Pleural involvement in brucellosis is very rare. Current knowledge on brucella pleuritis is limited to a few case studies, and pleural adenosine deaminase (ADA) in brucellosis has not been studied previously. We report the pleural fluid characteristics, including ADA, of two cases with brucella pleurisy. Analysis of the pleural fluids revealed exudative effusions with increased ADA level, decreased glucose concentration, and lymphocyte predominance. The similarity with tuberculous pleurisy was remarkable. We suggest that brucellosis should be considered in the differential diagnosis of tuberculosis, especially in regions endemic for both diseases.     

Diagnosis and treatment of tuberculous pleurisy--with special reference to the significance of measurement of pleural fluid cytokines

Tuberculous pleurisy as well as malignant pleuritis is a representative disease presenting pleural effusion. The diagnosis of tuberculous pleurisy is made from examination of pleural effusion, but the sensitivity of smear or culture of Mycobacterium tuberculosis from pleural fluid is generally low. Although the pleural fluid concentration of adenosine deaminase (ADA) is useful in terms of sensitivity or specificity, the value could be high in empyema or rheumatoid pleuritis. Thoracoscopic biopsy of pleura is more sensitive rather than conventional percutaneous needle biopsy, but is more invasive. Tuberculous pleural effusion is caused by delayed allergy which macrophage and T-helper 1 cells mainly relate and the stimuli of bacterial body consecutively induces T-helper 1 cytokines. Pleural fluid interferon-gamma (INF-gamma) is important not only in pathogenesis but also in diagnosis. We demonstrated that INF-gamma is a more sensitive and specific indicator for tuberculous pleurisy than ADA using receiver operating characteristics (ROC) analysis. Cytometric bead array (CBA) is a tool to simultaneously measure abundance of various cytokines and is expected to be a very useful method to provide informations for understanding a feedback mechanism of cytokine network. It is needed to clear the immunity in pleural fluid and to establish the less invasive and more useful method to diagnose tuberculous pleurisy.     

Diagnostic accuracy of T-cell interferon-γ release assays in tuberculous pleurisy: a meta-analysis

Background and objective: The diagnosis of tuberculous pleurisy by analysis of pleural fluid using standard diagnostic tools is difficult. Recently, T‐cell interferon‐γ release assays (IGRA) have been introduced for the diagnosis of tuberculous pleurisy. The aim of the present meta‐analysis was to establish the overall diagnostic accuracy of IGRA on both pleural fluid and peripheral blood, for diagnosing tuberculous pleurisy. Methods: A systematic review was performed of English language publications. Sensitivity, specificity and other measures of the accuracy of IGRA for the diagnosis tuberculous pleurisy using both pleural fluid and blood were pooled using a random‐effects model or a fixed‐effects model. Receiver operating characteristic curves were used to summarize overall test performance. Results: Seven out of eight studies met the inclusion criteria. The summary estimates of sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, positive predictive value, negative predictive value and diagnostic odds ratio were, for pleural fluid: 0.75, 0.82, 3.49, 0.24, 0.85, 0.70 and 19.04, respectively; and for blood: 0.80, 0.72, 2.86, 0.28, 0.78, 0.74 and 11.06, respectively. Conclusions: As almost 20% of non‐tuberculosis patients would be erroneously treated for tuberculosis and 25% of patients with tuberculous pleurisy would be missed, pleural fluid IGRA are not useful for the clinical diagnosis of tuberculous pleurisy.     

Lymphocyte subpopulations analysis in pleural fluid and peripheral blood in patients with lymphocytic pleural effusions.

Lymphocyte subpopulations analysis by an 11-monoclonal antibody (MoAb) panel was carried out in pleural fluid and in peripheral blood in 30 patients affected by newly diagnosed, untreated pleural effusion of different etiology determinated with bacteriological, cytological or histological criteria. Lymphocytes were the predominant cell type, in pleural fluid, in neoplastic pleural effusions as well as in congestive heart failure pleural effusions and, especially, in tuberculous pleural effusions. Lymphocyte analysis in pleural fluid and in peripheral blood suggests the involvement of different mechanisms for the lymphocyte accumulation in the pleural space according to different etiologies. Tuberculous pleural effusions showed an evident CD4+ and TEC T5.9+ lymphocyte accumulation from peripheral blood. In these patients, cutaneous skin test response to purified protein derivative was strongly related to this situation. In neoplastic pleural effusions there was a lower percentage of CD4+ lymphocytes, reflecting circulating lymphocyte pool; however, in neoplastic pleural effusions, various lymphocyte patterns may be sometimes observed depending on different histologies. Passive lymphocyte accumulation seems to be the most important mechanism in congestive-heart-failure pleural effusions.     

Differential diagnosis of tuberculous pleurisy by measurement of cytokine concentrations in pleural effusion

Study objective: Measurement of cytokine concentration in serum and pleural effusion may be useful in the differential diagnosis of tuberculous pleurisy.Patients and methods: We compared the biochemical properties and concentrations of cytokines in serum and pleural effusion samples of 18 patients with tuberculous pleurisy, 7 patients with parapneumonic pleurisy, and 25 patients with malignant pleurisy.Results: A high value of adenosine deaminase (ADA) was observed in pleural effusion of patients with tuberculosis. The serum concentrations of interleukin (IL)-1-beta, IL-2, interferon (IFN)-gamma and tumor necrosis factor (TNF)-alpha were similar among the three groups. However, the concentration of IFN-gamma in pleural effusion was high in tuberculous patients, and that of TNF-alpha was high in tuberculous and parapneumonic pleural fluid, but both cytokines were low in malignant pleural fluid. The sensitivity, specificity and accuracy of IFN-gamma in the diagnosis of tuberculous pleurisy were 94%, 100% and 98%, respectively. Similarly, those of TNF-alpha for the diagnosis of infectious pleurisy including tuberculous and parapneumonic pleurisy were 88%, 80% and 84%, respectively.Conclusions: Our results indicate that simultaneous measurement of IFN-gamma and TNF-alpha in pleural effusion is a useful diagnostic tool for differentiating tuberculous pleurisy from parapneumonic and malignant pleurisy.     

结核性和恶性胸膜炎患者VEGF的水平及意义

目的 通过检测结核性胸膜炎及恶性胸膜炎患者血清和胸水中血管内皮生长因子（VEGF）含量,分析VEGF在两组患者血清、胸水中的差异,探讨VEGF在二者中的意义和诊断价值.方法 对确诊结核性胸膜炎和恶性胸膜炎各30例的患者在同一日留取胸水标本10 ml及静脉血5 ml,采用双抗体夹心酶联免疫吸附试验检测患者胸水及血清中VEGF水平,分析其差异及相关性.结果 结核组血清和胸水VEGF检测值分别为（45.33±18.33） ng/L、（62.73±24.65） ng/L;恶性组血清和胸水VEGF检测值分别为（66.00±29.83） ng/L、（95.54±42.11） ng/L;恶性组血清及胸水中VEGF含量均高于结核组（t值分别为3.9、5.2,P值均＜0.05）.VEGF在两组的血清和胸水中均呈正相关性（r值分别为0.53、0.38,P值均＜0.05）.结论 在结核性胸膜炎及恶性胸膜炎患者血清、胸水中VEGF水平有差异,恶性高于结核性;两组患者胸水中VEGF含量均高于血清,胸水VEGF含量随着血清VEGF含量增高而增高.检测胸腔积液患者血清和胸水中VEGF含量对结核性胸膜炎和恶性胸膜炎的诊断和鉴别诊断有一定的价值.     

An analysis of in vitro T cell responsiveness in nontuberculous mycobacterial infection

In vitro cell-mediated immunity was examined in patients infected with nontuberculous mycobacteria, Mycobacterium avium-intracellulare complex in Japan. Peripheral blood lymphocytes of patients, as compared with those of tuberculous patients or tuberculin-positive healthy donors, showed depressed in vitro blastogenic responses to purified protein derivative of tuberculin (PPD), not only to PPDs of Mycobacterium tuberculosis but also to PPD-B and PPD-Y of M intracellulare and M kansasii, respectively. Nonspecific lymphocyte blastogenic responses to concanavalin A, phytohemagglutinin and pokeweed mitogen were normal. Analysis of defective in vitro PPD-induced lymphocyte blastogenic responses in these patients revealed that PPD-induced interleukin 2 (IL-2) production was impaired whereas PPD-induced IL-2 responsiveness was normally developed after PPD stimulation. Therefore, addition of exogenous recombinant human IL 2 substantially recovered the in vitro depressed PPD-induced blastogenic responses in these patients with nontuberculous mycobacterial infection.