- Email: [email protected]
Clinical features of immunocompromised and nonimmunocompromised patients with pulmonary tuberculosis
J Infect Chemother (2007) 13:405–410 DOI 10.1007/s10156-007-0558-z
© Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases 2007
ORIGINAL ARTICLE Yoshihiro Kobashi · Keiji Mouri · Shinichi Yagi Yasushi Obase · Naoyuki Miyashita · Niro Okimoto Toshiharu Matsushima · Mikio Oka
Clinical features of immunocompromised and nonimmunocompromised patients with pulmonary tuberculosis
Received: May 17, 2007 / Accepted: August 24, 2007
Abstract Although patients with risk factors for tuberculosis have increased with the increased use of immunosuppressive therapy, there have been few reports about differences in clinical findings between immunocompromised patients and nonimmunocompromised patients with pulmonary tuberculosis. Therefore, we investigated differences between the clinical features of immunocompromised patients with pulmonary tuberculosis and those of nonimmunocompromised patients in the past decade. We analyzed findings in 840 patients (312 immunocompromised and 528 nonimmunocompromised) with pulmonary tuberculosis who were found to be culture-positive for Mycobacterium tuberculosis between January 1997 and December 2006. The characteristic clinical findings of the immunocompromised patients, compared with findings in the nonimmunocompromised group, were as follows: (1) an increase in the number of patients with respiratory symptoms during the period of follow-up of underlying diseases; (2) an increase in the number of patients in a hyponutritional state and with a negative response for the tuberculin skin test; (3) an increase in the number of microbiologically smear-positive sputum specimens; (4) an increase in the number of patients with atypical radiological findings, such as a few cavities or calcification, bilateral and expansive consolidation, miliary shadows, and mediastinal and/or hilar lymphadenopathy ; (5) an increase in the number of patients with misdiagnosed pneumonia at admission; and (6) an increase in the mortality rate. We concluded that, among the immunocompromised patients with pulmonary tuberculosis, there Y. Kobashi (*) · K. Mouri · S. Yagi · Y. Obase · N. Miyashita · M. Oka Division of Respiratory Diseases, Department of Medicine, Kawasaki Medical School, 577 Matsushima, Kurashiki 701-0192, Japan Tel. +81-86-462-1111; Fax +81-86-464-1041 e-mail: [email protected] N. Okimoto Division of Respiratory Diseases, Department of Medicine, Kawasaki Medical School Kawasaki Hospital, Okayama, Japan T. Matsushima Division of Respiratory Diseases, Department of Medicine, Kurashiki Daiichi Hospital, Kurashiki, Japan
were many patients with atypical radiological findings and with smear-positive findings for acid-fast bacilli examination that was carried out to isolate M. tuberculosis (which had become infectious). We must perform the acid-fast bacilli examination for patients who have a fever and continuous cough, and antituberculous drugs should be administered as soon as possible when the results are positive. Key words Pulmonary tuberculosis · Immunocompromised patients · Nonimmunocompromised patients
Introduction Both the use of immunosuppressive therapies (cancer chemotherapy or new immunomodulatory agents) and the numbers of patients with diabetes mellitus, malignant diseases, chronic renal failure with hemodialysis, and HIV infection have recently increased in Japan. Consequently, due to suspected changes in the clinical features of pulmonary tuberculosis (TB), it has become difficult to distinguish pulmonary TB from other respiratory diseases.1–3 However, there have been few recent reports involving comparative studies of the clinical features of immunocompromised and nonimmunocompromised patients with pulmonary TB. Therefore, we decided to investigate the characteristic features of immunocompromised patients with pulmonary TB during the past decade.
Patients and methods Between January 1997 and December 2006, 840 patients with pulmonary TB were treated at Kawasaki Medical School Hospital (a 1070-bed medical college hospital) and at four associated community hospitals without special tuberculosis wards in Okayama prefecture in Japan. These 840 patients with pulmonary TB who satisfied the diagnostic criteria (see below) were divided into two groups (312 immunocompromised patients and 528 nonimmunocom-
406
promised patients), and the clinical features of their pulmonary TB were compared. The criteria for active pulmonary TB were as follows: (1) appearance of new active lesions in the lung field and (2) identification of Mycobacterium tuberculosis on a culture examination of acid-fast bacilli, using sputum or bronchoscopic specimens. We made use of the definition of lower lung field TB stated by Segerra et al.,4 as follows: (1) the lesions are in the lower lung lobe, except for S6a, the middle lobe, and the lingular segment; and (2) no active lesions are in the upper lung field. From medical records, each patient’s background was retrospectively reviewed with regard to age, sex, life history (smoking and alcohol abuse history), clinical symptoms, methods of detection of pulmonary TB, laboratory findings, microbiological findings, radiological findings, treatment, and prognosis. Regarding the detection methods, we divided the patients into three groups. The first group consisted of patients who consulted our hospital complaining of clinical symptoms such as fever, cough, chest pain, and dyspnea. The second group consisted of patients without clinical symptoms in whom pulmonary TB was detected among the periodic health examinations offered every year by local governments. The third group consisted of patients without clinical symptoms in whom pulmonary TB was detected by physicians at other departments during follow-up for other underlying diseases. As for the radiological findings, chest roentgenography or chest computed tomography (CT) was performed initially in all patients. Several respiratory specialists retrospectively evaluated the findings in all patients as: consolidation, miliary lesions, mediastinal and/or hilar lymphadenopathy, pleural effusion, cavities, and calcification. Regarding abnormal laboratory findings indicating side effects of antituberculous treatment, a peripheral blood cell count and chemical screening of items including aminotransferase, total bilirubin, and serum creatinine levels, were performed at each visit during therapy. Patients were advised to consult an otolaryngologist and an ophthalmologist initially, and as needed subsequently.
Statistical analysis Statistical analysis was performed on a Macintosh computer with a Stat View J 4.5 statistical program (Abacus Concepts, Cary, NC, USA). The differences between immunocompro-
mised patients and nonimmunocompromised patients were analyzed using the χ2 test. Concordance between the two patient groups was assessed using the kappa coefficient. Throughout this study, confidence intervals (CI) are given as 95%. A P value of less than 0.05 was considered significant.
Results In the immunocompromised patients group (group 1), the underlying diseases were: malignant diseases including leukemia in130 patients, diabetes mellitus in 86, immunosuppressive treatment (corticosteroids or other immunosuppressive drugs) in 68, chronic renal failure with hemodialysis in 42, and HIV infection in 3. Seventeen patients had a second underlying disease. However, in group 1, no patients with pulmonary TB had experienced the use of tumor necrosis factor-alpha (TNF-α) inhibitor drugs for rheumatoid arthritis or Crohn’s disease. The backgrounds of the patients with active pulmonary TB in the two patient groups are shown in Table 1. Regarding age, sex ratio, smoking history, alcohol abuse history, and previous antituberculous treatment, there were no significant differences between group 1 and the nonimmunocompromised patients (group 2). The methods used to detect pulmonary TB in the two groups are shown in Table 2. A comparison of the frequency of detection by the different methods in the two groups showed a significantly lower detection rate for the yearly periodic health examinations performed by local governments in group 1 compared with group 2, and a significantly higher detection rate during the follow-up of other diseases in group 1 compared with group 2. Although there was no significant difference between the groups in the detection rates for clinical symptoms overall, among individual symptoms, fever and dyspnea were significantly more frequently detected in group 1 than in group 2. The laboratory findings of the patients with active pulmonary TB in the two groups are shown in Table 3. Although there were no significant differences as to the inflammatory response, nutritional condition, in terms of total protein and albumin levels, was significantly poorer in group 1 than in group 2. Purified-protein derivative (PPD) findings, which reflect the cellular immune response, showed that there were significantly more PPD-negative erythema diameter of <10 mm maximum size patients in group 1 than in group
Table 1. Backgrounds of patients with pulmonary tuberculosis in the two groups Background
Group 1 (n = 312)
Group 2 (n = 528)
P value
Age, years (mean ± S.D.) (⭌65 years) Sex (male: female) Smoking history Alcohol abuse history Previous antituberculous treatment
60.4 ± 10.8 48 (15.3%) 198 : 114 218 (69.9%) 63 (20.2%) 28 (9.0%)
63.0 ± 10.6 64 (12.1%) 344 : 184 375 (71.0%) 96 (18.1%) 54 (10.2%)
NS NS NS NS NS NS
407 Table 2. Methods of detection of pulmonary tuberculosis in patients in the two groups Detection method
Group 1 (n = 312)
Group 2 (n = 528)
P value
Clinical symptoms (with repetition) Fever (37°C⬉) Cough Sputum (hemosputum) Chest pain Dyspnea Periodic health examination During follow-up of other disease
224 (71.7%) 190 (60.8%) 210 (67.3%) 184 (59.0%) 40 (12.8%) 58 (18.6%) 16 (5.1%) 72 (23.1%)
318 (60.2%) 216 (40.9%) 312 (59.1%) 242 (45.8%) 48 (9.1%) 28 (5.3%) 158 (29.9%) 53 (10.0%)
NS <0.05 NS NS NS <0.05 <0.05 <0.05
NS, Not significant
Table 3. Laboratory findings of patients with pulmonary tuberculosis in the two groups Laboratory findings
Group 1 (n = 312)
Group 2 (n = 528)
WBC (8500/µl⬉) CRP (1.0 mg/dl⬉) ESR (20 mm/h⬉) TP (6.5 g/dl >) Alb (3.5 g/dl >) PPD-negativea
60 (19.2%) 242 (77.6%) 230 (73.7%) 203 (65.1%) 216 (69.2%) 164/260 (63.1%)
152 (28.7%) 368 (69.6%) 402 (76.1%) 174 (33.0%) 184 (34.8%) 76/512 (14.8%)
P value NS NS NS <0.05 <0.05 <0.05
WBC, White blood cells; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; TP, total protein; PPD, purified protein derivative; NS, not significant a PPD-negative shows erythema diameter of <10 mm maximum size
Table 4. Clinical diagnosis at admission Clinical diagnosis
Group 1 (n = 312)
Group 2 (n = 528)
P value
Pulmonary tuberculosis Pneumonia Lung cancer Pulmonary mycosis
120 (38.5%) 162 (51.9%) 16 (5.1%) 14 (4.5%)
266 (50.4%) 214 (40.5%) 28 (5.3%) 20 (3.8%)
NS NS NS NS
NS, Not significant
Table 5. Microbiological findings of patients with pulmonary tuberculosis in the two groups Microbiological findings
Group 1 (n = 312)
Group 2 (n = 528)
P value
Smear-positive, culture-positive Sputum Bronchoscopic specimen Smear-negative, culture-positive Sputum Bronchoscopic specimen MDRMTB
164 (52.6%) 116 (37.3%) 48 (15.3%) 148 (47.4%) 112 (35.8%) 36 (11.6%) 2 (0.6%)
216 (40.9%) 110 (20.8%) 106 (20.1%) 312 (59.1%) 240 (45.5%) 72 (13.6%) 4 (0.8%)
NS <0.05 NS NS NS NS NS
NS, Not significant; MDRMTB, multidrug resistant Mycobacterium tuberculosis (resistant to isoniazid >0.2 µg/ml and rifampicin >40 µg/ml)
2. There were no significant differences in laboratory findings among individual underlying diseases. The clinical diagnoses at admission are shown in Table 4. Pulmonary TB was more frequently misdiagnosed as pneumonia at admission in group 1, and it was more frequently correctly diagnosed in group 2. However, the difference between the two groups was not significant.
The microbiological findings of the patients with active pulmonary TB in the two groups are shown in Table 5. The smear-positive rate in group 1 was higher than that in group 1. The smear-positive rate in sputum specimens, in particular, was significantly higher in group 1 than in group 2. As for drug sensitivity tests of isolated strains of M. tuberculosis for antituberculous drugs, multidrug-resistant
408 Table 6. Radiological findings of patients with pulmonary tuberculosis in the two groups Radiological findings Site of lesion Unilateral Bilateral Extension of lesiona 1 2 3 Characteristics Consolidation (lobar or segmental) Miliary lesion Predominantly upper lung field Predominantly lower lung field Mediastinal and/or hilar lymphadenopathy Pleural effusion Cavity Calcification
Group 1 (n = 312)
Group 2 (n = 528)
P value
114 (36.5%) 198 (63.5%)
290 (54.9%) 238 (45.1%)
<0.05 <0.05
94 (30.1%) 150 (48.1%) 68 (21.8%)
238 (45.1%) 236 (44.6%) 54 (10.2%)
NS NS <0.05
148 (47.4%) 56 (17.9%) 120 (38.4%) 70 (22.4%) 46 (14.7%) 74 (23.7%) 94 (30.1%) 46 (14.7%)
132 (25.0%) 22 (4.2%) 360 (68.1%) 54 (10.2%) 14 (2.7%) 84 (16.0%) 290 (54.9%) 222 (42.0%)
<0.05 <0.05 <0.05 <0.05 <0.05 NS <0.05 <0.05
NS, Not significant a Extension: 1, denotes within one-third of the unilateral lung field; 2, denotes within the unilateral lung field; 3, denotes beyond the unilateral lung field
Table 7. Treatment and prognosis in patients with pulmonary tuberculosis in the two groups
Treatment (first regimen) INH + RFP + PZA + SM (or EB) INH + RFP + SM (or EB) INH + RFP None Clinical effect (good response) Side effect Mortality rate
Group 1 (n = 312)
Group 2 (n = 528)
P value
174 (55.8%) 116 (37.2%) 12 (3.8%) 10 (3.2%) 234/270 (86.7%) 60/270 (22.2%) 42 (13.5%)
342 (64.8%) 152 (28.7%) 20 (3.7%) 14 (2.7%) 472/504 (93.7%) 110/504 (21.8%) 24 (4.5%)
NS NS NS NS NS NS <0.05
INH, isoniazid; RFP, rifampicin; PZA, pyrazinamide; SM, streptomycin; EB, ethambutol; NS, not significant
M. tuberculosis that showed resistance to isoniazid (INH; more than 0.2 µg/ml) and rifampicin (RFP; more than 40 µg/ ml), was recognized in two strains in group 1 and four strains in group 2. The resistance to antituberculous drugs was not significantly different between the two groups. The radiological findings of the patients with active pulmonary TB in the two groups are shown in Table 6. Regarding the site and extension of the lesion, bilateral lesions and expansive lesions beyond the unilateral lung field were recognized more frequently in group 1 than in group 2. Although there were significant differences in radiological findings, such as lobar or segmental consolidation, miliary lesions, and mediastinal and/or hilar lymphadenopathy, between group 1 and group 2, the frequencies of characteristic findings of pulmonary TB, such as cavities and calcification and the predominance of the upper lung field, were significantly lower in group 1 than in group 2. The frequency of lower lung field TB was significantly higher in group 1 than in group 2. The treatments and prognosis of the patients with active pulmonary TB in the two groups are shown in Table 7. Although the treatment regimens were almost the same in the two groups, the clinical effect was slightly better in
group 2 than in group 1. On the other hand, even though we excluded those patients who died of underlying diseases, the mortality rate was significantly higher in group 1 than in group 2.
Discussion The characteristics of pulmonary TB in Japan, as reported in 1990, are as follows: (1) an increase in the morbidity rate of elderly patients with pulmonary TB, (2) an increase in the number of patients who are smear-positive for M. tuberculosis, (3) the appearance of mass infections that include medical staff, (4) local differences in morbidity rates, and (5) local differences in the rates of resistance of isolated M. tuberculosis strains to antituberculous drugs. However, these characteristic findings of pulmonary TB have been reported mostly for nonimmunocompromised patients from special hospitals dealing with mycobacterial diseases. There have been only a few recent clinical reports regarding immunocompromised patients with underlying diseases in community hospitals without special wards for TB.
409
Therefore, we performed a retrospective comparative study of patients over the past 10 years, divided into an immunocompromised patient group and a nonimmunocompromised patient group. Throughout this study, we found the characteristic clinical findings in immunocompromised patients with PT, compared with those in the nonimmunocompromised patient group, to be as follows; (1) an increase in patients with respiratory symptoms such as fever or dyspnea during the follow-up period of underlying diseases, (2) an increase in patients in a hyponutritional state and with a negative response to the tuberculin skin test, (3) an increase in microbiologically smear-positive results in sputum specimens, (4) an increase in the percentage of patients with atypical findings such as a few cavities or calcification, and bilateral and expansive lesions on radiological findings, (5) an increase in patients with misdiagnosed pneumonia at admission, and (6) an increase in the mortality rate. As for the methods used for the detection of pulmonary TB in patients in the two groups, in regard to individual clinical symptoms, the immunocompromised patients had a higher prevalence of fever and dyspnea than the nonimmunocompromised patients. Kiyan et al.6 made similar observations. The symptoms and signs of pulmonary TB in immunocompromised patients are not specific. Fever is one of the most common symptoms of pulmonary TB in such patients, especially in HIV-infected patients.7,8 The radiological findings of pulmonary TB can vary depending on the underlying immune status.9 In immunocompromised patients, pulmonary TB often has an atypical presentation.10,11 The radiological findings of pulmonary TB in HIV-infected patients have been described well;12,13 HIV seropositive patients without AIDS demonstrate findings of reactivation TB, characterized by upper-lobe infiltrates and cavity formation. Kiyan et al.6 reported the following characteristic radiological findings in immunocompromised patients that were not seen in nonimmunocompromised patients: lobar or segmental consolidation, miliary lesions, and hilar and/or mediastinal lymphadenopathy; the radiological findings in our study resembled those reported by Kiyan et al. In addition, we found that other atypical radiological findings, such as bilateral, expansive infiltration shadows and predominantly lower-lung field shadows were significantly more predominant in the immunocompromised patients, and the radiological findings typical of pulmonary TB, such as cavities, calcification, and predominantly upperlung field shadows, were not recognized. Concerning the mechanism of onset of lower-lung field tuberculosis (LLFTB), Pratt-Johnson14 and Segerra et al.4 mentioned the perforation of tuberculous lymph nodes of the primary complex near the trachea or bronchus. In our study, mediastinal and/or hilar lymphadenopathy were recognized significantly more frequently in the immunocompromised patients than in the nonimmunocompromised patients. As another mechanism of onset, it is also suspected that LLFTB lesions form from exogenous reinfection when the actual lesion of the primary complex is present.15 Although it has been reported recently that TNF-α inhibitors were strongly associated with the activation of TB,16,17 we did not encoun-
ter any patients with pulmonary TB that developed during the use of TNF-α inhibitor drugs for underlying diseases in our immunocompromised patient group. Regarding the microbiological examinations, although we have recently been performing bronchoscopic examinations to obtain a definite diagnosis of pulmonary TB for patients who were not smear-positive for M. tuberculosis on sputum specimens, in the present study, the number of microbiologically smear-positive sputum specimens was significantly higher in the immunocompromised patient group. This may be related to the extent of the lesion caused by pulmonary TB. In the drug sensitivity testing of isolated strains of M. tuberculosis for antituberculous drugs, the strain of M. tuberculosis isolated from immunocompromised patients in our affiliated hospitals fortunately has not shown any resistance as compared with that in hospitals specializing in mycobacterial diseases.18 Therefore, we do not think that resistance for antituberculous drugs was a factor that influenced the clinical effect of the antituberculous drugs administered or the prognosis of the immunocompromised patients with pulmonary TB. We think that the extension of the lesion caused by M. tuberculosis or the delay in the correct diagnosis because of atypical radiological findings influenced the mortality rate. As for the discrepancy between the clinical effect of the antituberculous drugs and the mortality rate, although we excluded the patients who died of underlying diseases, we cannot deny the possibility that complications due to the underlying diseases may have been involved in the higher mortality rate in the immunocompromised patient group. In conclusion, we suspect that, in community hospitals without a special ward for TB, atypical findings of pulmonary TB in immunocompromised patients have been increasing. We found many patients with smear-positive findings for acid-fast bacilli examination, showing that M. tuberculosis had become infectious agents. In accordance with these findings, we must perform examinations of acidfast bacilli for patients who have a fever and continuous cough, and antituberculous drugs should be administered as soon as possible when the results are positive.
References 1. Guzman CP, Cruz AT, Velarde HV, Vargas MN. Progressive agerelated changes in pulmonary tuberculosis images and the effect of diabetes. Am J Respir Crit Care Med 2000;162:1728–40. 2. Ikezoe J, Takeuchi N, Johkoh M, Kohno N, Tomiyama N, Kozuka T, et al. CT appearance of pulmonary tuberculosis in diabetes and immunocompromised patients: comparison with patients who had no underlying disease. AJR Am J Roentogenol 1992;159:1175–9. 3. Nemery B, Yew WW, Albent R, Brun-Buisson C, Macnee W, Martinets FJ, et al. Tuberculosis, nontuberculous lung infection, pleural disorders, pulmonary function, respiratory muscles, occupational lung disease, pulmonary infections, and social issues in AJRCCM in 2004. Am J Respir Crit Care Med 2005;171:554–62. 4. Segerra F, Scherman DS, Aguero J. Lower lung field tuberculosis. Chest 1974;65:522–6. 5. Toyota T. Studies on the changes in clinical features of tuberculosis (in Japanese, abstract in English). Kekkaku 1990;65:619–31. 6. Kiyan E, Kilicasian Z, Gurgan M, Tunaci A, Yildiz A. Clinical and radiographic features of pulmonary tuberculosis in non-AIDS
410
7. 8. 9. 10.
11. 12.
immunocompromised patients. Int J Tuberc Lung Dis 2003;7: 764–70. Miralles P, Moreno S, Perez-Tascon M. Fever of unknown origin in patients with human immunodeficiency virus. Clin Infect Dis 1995; 20:872–5. Hall CH, Wilcox PA, Swanepoel CR, Kahn D, Van Zyl Smit R. Mycobacterial infections in renal transplant recipients. Chest 1994; 106:435–9. Goo JM, Im JG. CT of tuberculosis and nontuberculous mycobacterial infections. Radiol Clin North Am 2002;40:73–87. Pitchenik AE, Rubinson HA. The radiographic appearance of tuberculosis in patients with the acquired immune deficiency syndrome (AIDS) and pre-AIDS. Am Rev Respir Dis 1985;131: 393–6. Lee MP, Chan JW, Ng KK, Li PC. Clinical manifestation of tuberculosis in HIV-infected patients. Respirology 2000;5:423–6. Harris AD. Tuberculosis in human immunodeficiency virus infection in developing countries. Lancet 1990;162:8–12.
13. Leung AN, Brauner MW, Gamsu G, Mlika-Cabanne N, Ben Romdhane H, Carette MF, et al. Pulmonary tuberculosis: comparison of CT findings in HIV-seropositive and HIV-seronegative patients. Radiology 1996;198:687–91. 14. Pratt-Johnson JH. Observations on lower lobe tuberculosis. Br J Dis Chest 1959;53:385–9. 15. Palmer PES. Pulmonary tuberculosis–usual and unusual radiographic presentat ions. Semin Roentogenol 1979;14:204–43. 16. Keane J, Gershon S, Wise RP, Mirabile-Levens E, Karzrica J, Schwieterman WD, et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med 2001;345:1098–104. 17. Mutlu GM, Murlu EA, Bellmeyer A, Rubenstein I. Pulmonary adverse events of anti-tumor necrosis factor-α antibody therapy. Am J Med 2006;119:639–46. 18. Hirano K, Kazumi Y, Abe C, Mori T, Aoki M, Aoyagi T. Resistance to antituberculous drugs in Japan. Tuber Lung Disease 1996;77: 130–5.
© Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases 2007
ORIGINAL ARTICLE Yoshihiro Kobashi · Keiji Mouri · Shinichi Yagi Yasushi Obase · Naoyuki Miyashita · Niro Okimoto Toshiharu Matsushima · Mikio Oka
Clinical features of immunocompromised and nonimmunocompromised patients with pulmonary tuberculosis
Received: May 17, 2007 / Accepted: August 24, 2007
Abstract Although patients with risk factors for tuberculosis have increased with the increased use of immunosuppressive therapy, there have been few reports about differences in clinical findings between immunocompromised patients and nonimmunocompromised patients with pulmonary tuberculosis. Therefore, we investigated differences between the clinical features of immunocompromised patients with pulmonary tuberculosis and those of nonimmunocompromised patients in the past decade. We analyzed findings in 840 patients (312 immunocompromised and 528 nonimmunocompromised) with pulmonary tuberculosis who were found to be culture-positive for Mycobacterium tuberculosis between January 1997 and December 2006. The characteristic clinical findings of the immunocompromised patients, compared with findings in the nonimmunocompromised group, were as follows: (1) an increase in the number of patients with respiratory symptoms during the period of follow-up of underlying diseases; (2) an increase in the number of patients in a hyponutritional state and with a negative response for the tuberculin skin test; (3) an increase in the number of microbiologically smear-positive sputum specimens; (4) an increase in the number of patients with atypical radiological findings, such as a few cavities or calcification, bilateral and expansive consolidation, miliary shadows, and mediastinal and/or hilar lymphadenopathy ; (5) an increase in the number of patients with misdiagnosed pneumonia at admission; and (6) an increase in the mortality rate. We concluded that, among the immunocompromised patients with pulmonary tuberculosis, there Y. Kobashi (*) · K. Mouri · S. Yagi · Y. Obase · N. Miyashita · M. Oka Division of Respiratory Diseases, Department of Medicine, Kawasaki Medical School, 577 Matsushima, Kurashiki 701-0192, Japan Tel. +81-86-462-1111; Fax +81-86-464-1041 e-mail: [email protected] N. Okimoto Division of Respiratory Diseases, Department of Medicine, Kawasaki Medical School Kawasaki Hospital, Okayama, Japan T. Matsushima Division of Respiratory Diseases, Department of Medicine, Kurashiki Daiichi Hospital, Kurashiki, Japan
were many patients with atypical radiological findings and with smear-positive findings for acid-fast bacilli examination that was carried out to isolate M. tuberculosis (which had become infectious). We must perform the acid-fast bacilli examination for patients who have a fever and continuous cough, and antituberculous drugs should be administered as soon as possible when the results are positive. Key words Pulmonary tuberculosis · Immunocompromised patients · Nonimmunocompromised patients
Introduction Both the use of immunosuppressive therapies (cancer chemotherapy or new immunomodulatory agents) and the numbers of patients with diabetes mellitus, malignant diseases, chronic renal failure with hemodialysis, and HIV infection have recently increased in Japan. Consequently, due to suspected changes in the clinical features of pulmonary tuberculosis (TB), it has become difficult to distinguish pulmonary TB from other respiratory diseases.1–3 However, there have been few recent reports involving comparative studies of the clinical features of immunocompromised and nonimmunocompromised patients with pulmonary TB. Therefore, we decided to investigate the characteristic features of immunocompromised patients with pulmonary TB during the past decade.
Patients and methods Between January 1997 and December 2006, 840 patients with pulmonary TB were treated at Kawasaki Medical School Hospital (a 1070-bed medical college hospital) and at four associated community hospitals without special tuberculosis wards in Okayama prefecture in Japan. These 840 patients with pulmonary TB who satisfied the diagnostic criteria (see below) were divided into two groups (312 immunocompromised patients and 528 nonimmunocom-
406
promised patients), and the clinical features of their pulmonary TB were compared. The criteria for active pulmonary TB were as follows: (1) appearance of new active lesions in the lung field and (2) identification of Mycobacterium tuberculosis on a culture examination of acid-fast bacilli, using sputum or bronchoscopic specimens. We made use of the definition of lower lung field TB stated by Segerra et al.,4 as follows: (1) the lesions are in the lower lung lobe, except for S6a, the middle lobe, and the lingular segment; and (2) no active lesions are in the upper lung field. From medical records, each patient’s background was retrospectively reviewed with regard to age, sex, life history (smoking and alcohol abuse history), clinical symptoms, methods of detection of pulmonary TB, laboratory findings, microbiological findings, radiological findings, treatment, and prognosis. Regarding the detection methods, we divided the patients into three groups. The first group consisted of patients who consulted our hospital complaining of clinical symptoms such as fever, cough, chest pain, and dyspnea. The second group consisted of patients without clinical symptoms in whom pulmonary TB was detected among the periodic health examinations offered every year by local governments. The third group consisted of patients without clinical symptoms in whom pulmonary TB was detected by physicians at other departments during follow-up for other underlying diseases. As for the radiological findings, chest roentgenography or chest computed tomography (CT) was performed initially in all patients. Several respiratory specialists retrospectively evaluated the findings in all patients as: consolidation, miliary lesions, mediastinal and/or hilar lymphadenopathy, pleural effusion, cavities, and calcification. Regarding abnormal laboratory findings indicating side effects of antituberculous treatment, a peripheral blood cell count and chemical screening of items including aminotransferase, total bilirubin, and serum creatinine levels, were performed at each visit during therapy. Patients were advised to consult an otolaryngologist and an ophthalmologist initially, and as needed subsequently.
Statistical analysis Statistical analysis was performed on a Macintosh computer with a Stat View J 4.5 statistical program (Abacus Concepts, Cary, NC, USA). The differences between immunocompro-
mised patients and nonimmunocompromised patients were analyzed using the χ2 test. Concordance between the two patient groups was assessed using the kappa coefficient. Throughout this study, confidence intervals (CI) are given as 95%. A P value of less than 0.05 was considered significant.
Results In the immunocompromised patients group (group 1), the underlying diseases were: malignant diseases including leukemia in130 patients, diabetes mellitus in 86, immunosuppressive treatment (corticosteroids or other immunosuppressive drugs) in 68, chronic renal failure with hemodialysis in 42, and HIV infection in 3. Seventeen patients had a second underlying disease. However, in group 1, no patients with pulmonary TB had experienced the use of tumor necrosis factor-alpha (TNF-α) inhibitor drugs for rheumatoid arthritis or Crohn’s disease. The backgrounds of the patients with active pulmonary TB in the two patient groups are shown in Table 1. Regarding age, sex ratio, smoking history, alcohol abuse history, and previous antituberculous treatment, there were no significant differences between group 1 and the nonimmunocompromised patients (group 2). The methods used to detect pulmonary TB in the two groups are shown in Table 2. A comparison of the frequency of detection by the different methods in the two groups showed a significantly lower detection rate for the yearly periodic health examinations performed by local governments in group 1 compared with group 2, and a significantly higher detection rate during the follow-up of other diseases in group 1 compared with group 2. Although there was no significant difference between the groups in the detection rates for clinical symptoms overall, among individual symptoms, fever and dyspnea were significantly more frequently detected in group 1 than in group 2. The laboratory findings of the patients with active pulmonary TB in the two groups are shown in Table 3. Although there were no significant differences as to the inflammatory response, nutritional condition, in terms of total protein and albumin levels, was significantly poorer in group 1 than in group 2. Purified-protein derivative (PPD) findings, which reflect the cellular immune response, showed that there were significantly more PPD-negative erythema diameter of <10 mm maximum size patients in group 1 than in group
Table 1. Backgrounds of patients with pulmonary tuberculosis in the two groups Background
Group 1 (n = 312)
Group 2 (n = 528)
P value
Age, years (mean ± S.D.) (⭌65 years) Sex (male: female) Smoking history Alcohol abuse history Previous antituberculous treatment
60.4 ± 10.8 48 (15.3%) 198 : 114 218 (69.9%) 63 (20.2%) 28 (9.0%)
63.0 ± 10.6 64 (12.1%) 344 : 184 375 (71.0%) 96 (18.1%) 54 (10.2%)
NS NS NS NS NS NS
407 Table 2. Methods of detection of pulmonary tuberculosis in patients in the two groups Detection method
Group 1 (n = 312)
Group 2 (n = 528)
P value
Clinical symptoms (with repetition) Fever (37°C⬉) Cough Sputum (hemosputum) Chest pain Dyspnea Periodic health examination During follow-up of other disease
224 (71.7%) 190 (60.8%) 210 (67.3%) 184 (59.0%) 40 (12.8%) 58 (18.6%) 16 (5.1%) 72 (23.1%)
318 (60.2%) 216 (40.9%) 312 (59.1%) 242 (45.8%) 48 (9.1%) 28 (5.3%) 158 (29.9%) 53 (10.0%)
NS <0.05 NS NS NS <0.05 <0.05 <0.05
NS, Not significant
Table 3. Laboratory findings of patients with pulmonary tuberculosis in the two groups Laboratory findings
Group 1 (n = 312)
Group 2 (n = 528)
WBC (8500/µl⬉) CRP (1.0 mg/dl⬉) ESR (20 mm/h⬉) TP (6.5 g/dl >) Alb (3.5 g/dl >) PPD-negativea
60 (19.2%) 242 (77.6%) 230 (73.7%) 203 (65.1%) 216 (69.2%) 164/260 (63.1%)
152 (28.7%) 368 (69.6%) 402 (76.1%) 174 (33.0%) 184 (34.8%) 76/512 (14.8%)
P value NS NS NS <0.05 <0.05 <0.05
WBC, White blood cells; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; TP, total protein; PPD, purified protein derivative; NS, not significant a PPD-negative shows erythema diameter of <10 mm maximum size
Table 4. Clinical diagnosis at admission Clinical diagnosis
Group 1 (n = 312)
Group 2 (n = 528)
P value
Pulmonary tuberculosis Pneumonia Lung cancer Pulmonary mycosis
120 (38.5%) 162 (51.9%) 16 (5.1%) 14 (4.5%)
266 (50.4%) 214 (40.5%) 28 (5.3%) 20 (3.8%)
NS NS NS NS
NS, Not significant
Table 5. Microbiological findings of patients with pulmonary tuberculosis in the two groups Microbiological findings
Group 1 (n = 312)
Group 2 (n = 528)
P value
Smear-positive, culture-positive Sputum Bronchoscopic specimen Smear-negative, culture-positive Sputum Bronchoscopic specimen MDRMTB
164 (52.6%) 116 (37.3%) 48 (15.3%) 148 (47.4%) 112 (35.8%) 36 (11.6%) 2 (0.6%)
216 (40.9%) 110 (20.8%) 106 (20.1%) 312 (59.1%) 240 (45.5%) 72 (13.6%) 4 (0.8%)
NS <0.05 NS NS NS NS NS
NS, Not significant; MDRMTB, multidrug resistant Mycobacterium tuberculosis (resistant to isoniazid >0.2 µg/ml and rifampicin >40 µg/ml)
2. There were no significant differences in laboratory findings among individual underlying diseases. The clinical diagnoses at admission are shown in Table 4. Pulmonary TB was more frequently misdiagnosed as pneumonia at admission in group 1, and it was more frequently correctly diagnosed in group 2. However, the difference between the two groups was not significant.
The microbiological findings of the patients with active pulmonary TB in the two groups are shown in Table 5. The smear-positive rate in group 1 was higher than that in group 1. The smear-positive rate in sputum specimens, in particular, was significantly higher in group 1 than in group 2. As for drug sensitivity tests of isolated strains of M. tuberculosis for antituberculous drugs, multidrug-resistant
408 Table 6. Radiological findings of patients with pulmonary tuberculosis in the two groups Radiological findings Site of lesion Unilateral Bilateral Extension of lesiona 1 2 3 Characteristics Consolidation (lobar or segmental) Miliary lesion Predominantly upper lung field Predominantly lower lung field Mediastinal and/or hilar lymphadenopathy Pleural effusion Cavity Calcification
Group 1 (n = 312)
Group 2 (n = 528)
P value
114 (36.5%) 198 (63.5%)
290 (54.9%) 238 (45.1%)
<0.05 <0.05
94 (30.1%) 150 (48.1%) 68 (21.8%)
238 (45.1%) 236 (44.6%) 54 (10.2%)
NS NS <0.05
148 (47.4%) 56 (17.9%) 120 (38.4%) 70 (22.4%) 46 (14.7%) 74 (23.7%) 94 (30.1%) 46 (14.7%)
132 (25.0%) 22 (4.2%) 360 (68.1%) 54 (10.2%) 14 (2.7%) 84 (16.0%) 290 (54.9%) 222 (42.0%)
<0.05 <0.05 <0.05 <0.05 <0.05 NS <0.05 <0.05
NS, Not significant a Extension: 1, denotes within one-third of the unilateral lung field; 2, denotes within the unilateral lung field; 3, denotes beyond the unilateral lung field
Table 7. Treatment and prognosis in patients with pulmonary tuberculosis in the two groups
Treatment (first regimen) INH + RFP + PZA + SM (or EB) INH + RFP + SM (or EB) INH + RFP None Clinical effect (good response) Side effect Mortality rate
Group 1 (n = 312)
Group 2 (n = 528)
P value
174 (55.8%) 116 (37.2%) 12 (3.8%) 10 (3.2%) 234/270 (86.7%) 60/270 (22.2%) 42 (13.5%)
342 (64.8%) 152 (28.7%) 20 (3.7%) 14 (2.7%) 472/504 (93.7%) 110/504 (21.8%) 24 (4.5%)
NS NS NS NS NS NS <0.05
INH, isoniazid; RFP, rifampicin; PZA, pyrazinamide; SM, streptomycin; EB, ethambutol; NS, not significant
M. tuberculosis that showed resistance to isoniazid (INH; more than 0.2 µg/ml) and rifampicin (RFP; more than 40 µg/ ml), was recognized in two strains in group 1 and four strains in group 2. The resistance to antituberculous drugs was not significantly different between the two groups. The radiological findings of the patients with active pulmonary TB in the two groups are shown in Table 6. Regarding the site and extension of the lesion, bilateral lesions and expansive lesions beyond the unilateral lung field were recognized more frequently in group 1 than in group 2. Although there were significant differences in radiological findings, such as lobar or segmental consolidation, miliary lesions, and mediastinal and/or hilar lymphadenopathy, between group 1 and group 2, the frequencies of characteristic findings of pulmonary TB, such as cavities and calcification and the predominance of the upper lung field, were significantly lower in group 1 than in group 2. The frequency of lower lung field TB was significantly higher in group 1 than in group 2. The treatments and prognosis of the patients with active pulmonary TB in the two groups are shown in Table 7. Although the treatment regimens were almost the same in the two groups, the clinical effect was slightly better in
group 2 than in group 1. On the other hand, even though we excluded those patients who died of underlying diseases, the mortality rate was significantly higher in group 1 than in group 2.
Discussion The characteristics of pulmonary TB in Japan, as reported in 1990, are as follows: (1) an increase in the morbidity rate of elderly patients with pulmonary TB, (2) an increase in the number of patients who are smear-positive for M. tuberculosis, (3) the appearance of mass infections that include medical staff, (4) local differences in morbidity rates, and (5) local differences in the rates of resistance of isolated M. tuberculosis strains to antituberculous drugs. However, these characteristic findings of pulmonary TB have been reported mostly for nonimmunocompromised patients from special hospitals dealing with mycobacterial diseases. There have been only a few recent clinical reports regarding immunocompromised patients with underlying diseases in community hospitals without special wards for TB.
409
Therefore, we performed a retrospective comparative study of patients over the past 10 years, divided into an immunocompromised patient group and a nonimmunocompromised patient group. Throughout this study, we found the characteristic clinical findings in immunocompromised patients with PT, compared with those in the nonimmunocompromised patient group, to be as follows; (1) an increase in patients with respiratory symptoms such as fever or dyspnea during the follow-up period of underlying diseases, (2) an increase in patients in a hyponutritional state and with a negative response to the tuberculin skin test, (3) an increase in microbiologically smear-positive results in sputum specimens, (4) an increase in the percentage of patients with atypical findings such as a few cavities or calcification, and bilateral and expansive lesions on radiological findings, (5) an increase in patients with misdiagnosed pneumonia at admission, and (6) an increase in the mortality rate. As for the methods used for the detection of pulmonary TB in patients in the two groups, in regard to individual clinical symptoms, the immunocompromised patients had a higher prevalence of fever and dyspnea than the nonimmunocompromised patients. Kiyan et al.6 made similar observations. The symptoms and signs of pulmonary TB in immunocompromised patients are not specific. Fever is one of the most common symptoms of pulmonary TB in such patients, especially in HIV-infected patients.7,8 The radiological findings of pulmonary TB can vary depending on the underlying immune status.9 In immunocompromised patients, pulmonary TB often has an atypical presentation.10,11 The radiological findings of pulmonary TB in HIV-infected patients have been described well;12,13 HIV seropositive patients without AIDS demonstrate findings of reactivation TB, characterized by upper-lobe infiltrates and cavity formation. Kiyan et al.6 reported the following characteristic radiological findings in immunocompromised patients that were not seen in nonimmunocompromised patients: lobar or segmental consolidation, miliary lesions, and hilar and/or mediastinal lymphadenopathy; the radiological findings in our study resembled those reported by Kiyan et al. In addition, we found that other atypical radiological findings, such as bilateral, expansive infiltration shadows and predominantly lower-lung field shadows were significantly more predominant in the immunocompromised patients, and the radiological findings typical of pulmonary TB, such as cavities, calcification, and predominantly upperlung field shadows, were not recognized. Concerning the mechanism of onset of lower-lung field tuberculosis (LLFTB), Pratt-Johnson14 and Segerra et al.4 mentioned the perforation of tuberculous lymph nodes of the primary complex near the trachea or bronchus. In our study, mediastinal and/or hilar lymphadenopathy were recognized significantly more frequently in the immunocompromised patients than in the nonimmunocompromised patients. As another mechanism of onset, it is also suspected that LLFTB lesions form from exogenous reinfection when the actual lesion of the primary complex is present.15 Although it has been reported recently that TNF-α inhibitors were strongly associated with the activation of TB,16,17 we did not encoun-
ter any patients with pulmonary TB that developed during the use of TNF-α inhibitor drugs for underlying diseases in our immunocompromised patient group. Regarding the microbiological examinations, although we have recently been performing bronchoscopic examinations to obtain a definite diagnosis of pulmonary TB for patients who were not smear-positive for M. tuberculosis on sputum specimens, in the present study, the number of microbiologically smear-positive sputum specimens was significantly higher in the immunocompromised patient group. This may be related to the extent of the lesion caused by pulmonary TB. In the drug sensitivity testing of isolated strains of M. tuberculosis for antituberculous drugs, the strain of M. tuberculosis isolated from immunocompromised patients in our affiliated hospitals fortunately has not shown any resistance as compared with that in hospitals specializing in mycobacterial diseases.18 Therefore, we do not think that resistance for antituberculous drugs was a factor that influenced the clinical effect of the antituberculous drugs administered or the prognosis of the immunocompromised patients with pulmonary TB. We think that the extension of the lesion caused by M. tuberculosis or the delay in the correct diagnosis because of atypical radiological findings influenced the mortality rate. As for the discrepancy between the clinical effect of the antituberculous drugs and the mortality rate, although we excluded the patients who died of underlying diseases, we cannot deny the possibility that complications due to the underlying diseases may have been involved in the higher mortality rate in the immunocompromised patient group. In conclusion, we suspect that, in community hospitals without a special ward for TB, atypical findings of pulmonary TB in immunocompromised patients have been increasing. We found many patients with smear-positive findings for acid-fast bacilli examination, showing that M. tuberculosis had become infectious agents. In accordance with these findings, we must perform examinations of acidfast bacilli for patients who have a fever and continuous cough, and antituberculous drugs should be administered as soon as possible when the results are positive.
References 1. Guzman CP, Cruz AT, Velarde HV, Vargas MN. Progressive agerelated changes in pulmonary tuberculosis images and the effect of diabetes. Am J Respir Crit Care Med 2000;162:1728–40. 2. Ikezoe J, Takeuchi N, Johkoh M, Kohno N, Tomiyama N, Kozuka T, et al. CT appearance of pulmonary tuberculosis in diabetes and immunocompromised patients: comparison with patients who had no underlying disease. AJR Am J Roentogenol 1992;159:1175–9. 3. Nemery B, Yew WW, Albent R, Brun-Buisson C, Macnee W, Martinets FJ, et al. Tuberculosis, nontuberculous lung infection, pleural disorders, pulmonary function, respiratory muscles, occupational lung disease, pulmonary infections, and social issues in AJRCCM in 2004. Am J Respir Crit Care Med 2005;171:554–62. 4. Segerra F, Scherman DS, Aguero J. Lower lung field tuberculosis. Chest 1974;65:522–6. 5. Toyota T. Studies on the changes in clinical features of tuberculosis (in Japanese, abstract in English). Kekkaku 1990;65:619–31. 6. Kiyan E, Kilicasian Z, Gurgan M, Tunaci A, Yildiz A. Clinical and radiographic features of pulmonary tuberculosis in non-AIDS
410
7. 8. 9. 10.
11. 12.
immunocompromised patients. Int J Tuberc Lung Dis 2003;7: 764–70. Miralles P, Moreno S, Perez-Tascon M. Fever of unknown origin in patients with human immunodeficiency virus. Clin Infect Dis 1995; 20:872–5. Hall CH, Wilcox PA, Swanepoel CR, Kahn D, Van Zyl Smit R. Mycobacterial infections in renal transplant recipients. Chest 1994; 106:435–9. Goo JM, Im JG. CT of tuberculosis and nontuberculous mycobacterial infections. Radiol Clin North Am 2002;40:73–87. Pitchenik AE, Rubinson HA. The radiographic appearance of tuberculosis in patients with the acquired immune deficiency syndrome (AIDS) and pre-AIDS. Am Rev Respir Dis 1985;131: 393–6. Lee MP, Chan JW, Ng KK, Li PC. Clinical manifestation of tuberculosis in HIV-infected patients. Respirology 2000;5:423–6. Harris AD. Tuberculosis in human immunodeficiency virus infection in developing countries. Lancet 1990;162:8–12.
13. Leung AN, Brauner MW, Gamsu G, Mlika-Cabanne N, Ben Romdhane H, Carette MF, et al. Pulmonary tuberculosis: comparison of CT findings in HIV-seropositive and HIV-seronegative patients. Radiology 1996;198:687–91. 14. Pratt-Johnson JH. Observations on lower lobe tuberculosis. Br J Dis Chest 1959;53:385–9. 15. Palmer PES. Pulmonary tuberculosis–usual and unusual radiographic presentat ions. Semin Roentogenol 1979;14:204–43. 16. Keane J, Gershon S, Wise RP, Mirabile-Levens E, Karzrica J, Schwieterman WD, et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med 2001;345:1098–104. 17. Mutlu GM, Murlu EA, Bellmeyer A, Rubenstein I. Pulmonary adverse events of anti-tumor necrosis factor-α antibody therapy. Am J Med 2006;119:639–46. 18. Hirano K, Kazumi Y, Abe C, Mori T, Aoki M, Aoyagi T. Resistance to antituberculous drugs in Japan. Tuber Lung Disease 1996;77: 130–5.