Published online before print August 16, 2002, 10.1148/radiol.2251011357
(Radiology 2002;225:205-209.)
© RSNA, 2002
Utility of Polymerase Chain Reaction for Detecting Mycobacterium tuberculosis in Specimens from Percutaneous Transthoracic Needle Aspiration1
Eun-Young Kang, MD,
Jung-Ah Choi, MD,
Bo Kyoung Seo, MD,
Yu-Whan Oh, MD,
Chang Kyu Lee, MD and
Jae Jeong Shim, MD
1 From the Departments of Diagnostic Radiology (E.Y.K., J.A.C., B.K.S., Y.W.O.), Clinical Pathology (C.K.L.), and Internal Medicine (J.J.S.), Korea University College of Medicine, Korea University Guro Hospital, 80 Guro-dong, Guro-ku, Seoul 152-050, Korea. Received August 10, 2001; revision requested September 28; revision received December 19; accepted February 25, 2002. Address correspondence to E.Y.K. (e-mail: keyrad@korea.ac.kr).
 |
ABSTRACT
|
|---|
PURPOSE: To determine the clinical utility of polymerase chain reaction (PCR) for detecting Mycobacterium tuberculosis in specimens from percutaneous transthoracic needle aspiration (PTNA).
MATERIALS AND METHODS: PCR for M tuberculosis detection in specimens from PTNA was performed prospectively in addition to cytologic and histologic analyses in 45 patients. On computed tomographic (CT) scans, tuberculosis (TB) versus malignant neoplasm or other infection was diagnosed in 28 patients; possible malignancy was diagnosed in 11, but TB was considered clinically because of young patient age or presence of tuberculous lesions in other areas of the lungs. In six of these patients, TB was diagnosed, but bacteriologic results were negative. PTNA was performed with a 21-gauge needle by one chest radiologist by using CT (n = 25), ultrasonographic (n = 5), or fluoroscopic guidance (n = 15). Final diagnoses were malignant neoplasm (n = 19), hamartoma (n = 1), TB (n = 17), and pneumonia (including actinomycosis and aspergillosis) (n = 8). Sensitivity, specificity, and positive predictive values of PCR in PTNA specimens for diagnosis of TB were calculated.
RESULTS: In 17 patients with TB, PCR results were positive in 11 and negative in six. PCR results were negative in all cases of malignant neoplasm, hamartoma, and pneumonia. Cytologic and histologic analysis of PTNA specimens resulted in a specific diagnosis of TB in two (12%) of 17 patients. By adding the PCR results, diagnosis of TB was established in 12 (71%) of 17 patients. Sensitivity, specificity, and positive and negative predictive values of PCR for diagnosis of TB in PTNA specimens were 65% (11 of 17), 100% (28 of 28), 100% (11 of 11), and 82% (28 of 34), respectively.
CONCLUSION: PCR for detection of M tuberculosis in PTNA specimens is a useful adjuvant to cytologic and histologic analysis for diagnosis of TB.
© RSNA, 2002
Index terms: Lung, biopsy, 60.1261 • Lung, infection, 60.23 • Tuberculosis, pulmonary, 60.23
|
INTRODUCTION
|
|---|
Percutaneous transthoracic needle aspiration (PTNA) is used to aid in the diagnosis of a variety of benign and malignant chest lesions. The accuracy of PTNA for diagnosis of malignancy is excellent (1–4). However, the accuracy for a specific diagnosis of a benign lesion is not as high. The reported results of PTNA for diagnosis of benign lesions are variable (1,2,4).
The differential diagnosis of chest lesions can sometimes be particularly difficult in areas where tuberculosis (TB) is endemic, because TB shows variable patterns of lung diseases that mimic benign and malignant disease. A specific diagnosis of TB is difficult even when PTNA is performed.
Polymerase chain reaction (PCR) to detect Mycobacterium tuberculosis has been widely studied (5–10). In the diagnosis of pulmonary TB, the procedure has been most extensively investigated in specimens of sputum and materials obtained during bronchoscopy (11–13). Specimens obtained at direct puncture of the lung lesion probably constitute valuable material for PCR, the results of which can supplement cytologic and histologic findings. Previously, we reported (14) that nested PCR for detection of M tuberculosis in solitary pulmonary nodules showed good sensitivity (87.5%) and specificity (96%) with the use of lung aspirates from solitary pulmonary nodules. However, PCR for detecting M tuberculosis with materials from PTNA is not widely used by radiologists and physicians.
The aim of this study was to determine the clinical utility of PCR for TB detection in specimens obtained at PTNA.
|
MATERIALS AND METHODS
|
|---|
Patients
The study included 45 prospectively selected patients who had undergone PTNA during the previous 3 years. PCR for detection of M tuberculosis was performed with lung aspirates, in addition to cytologic and histologic examinations. In most of the patients, the diagnosis at computed tomography (CT) was TB versus malignant neoplasm or TB versus other infection (n = 28). Although CT findings in 11 patients were suggestive of a malignant lesion, the possibility of TB was considered clinically because of patients’ young age (n = 3) or because of tuberculous lesions present in other areas of the lungs on a CT scan (n = 8). In six patients, the CT diagnosis was TB, but bacteriologic results from sputum and materials obtained at bronchoscopy were negative. In these cases, therefore, we prospectively performed PTNA for cytologic and histologic analysis and for PCR to detect TB. Our institutional review board approved the study, and informed consent was obtained.
Final diagnoses were primary malignant neoplasm (n = 18), metastasis (n = 1), hamartoma (n = 1), TB (n = 17), and pneumonia (n = 8) (including actinomycosis [n = 2] and aspergillosis [n = 1]). Malignant and benign neoplasms were all diagnosed histologically. Other than with PCR, TB was diagnosed by using one or more methods: acid-fast bacilli (AFB) stain and mycobacterial culture of sputum (n = 4); positive AFB stain or mycobacterial culture of pleural fluid and bronchoscopic aspirates (n = 2); histologic diagnosis with material from neck lymph node or pleural biopsy, bronchoscopy, or lung surgery (n = 6); or cytologic and histologic analysis of material from PTNA (n = 2). In seven patients, the diagnosis of TB was established clinically on the basis of therapeutic response. Infections other than TB were diagnosed on the basis of PTNA results (n = 3) and/or clinical improvement after medication (n = 8).
PTNA Method
PTNA was performed by using CT (n = 25), ultrasonographic (US) (n = 5), or fluoroscopic guidance (n = 15) by one chest radiologist (E.Y.K.). The radiologist who performed PTNA selected the guidance method on a case-by-case basis. In general, CT guidance was preferred, but fluoroscopic guidance was used in patients with a small lesion and in less cooperative patients or when CT was not available. US guidance was usually used in patients with a lesion attached to the pleura or chest wall. All PTNA procedures were performed with a 21-gauge Chiba needle (M. I. Tech; Seoul, Korea). Each patient underwent PTNA once. The needle was inserted into the lesion, and its position was confirmed with image guidance. When the needle reached the lesion, the stylet was removed and the needle was connected to a 10-mL plastic syringe. The needle was then advanced to penetrate the lesion and was moved forward and backward several times, with continuous suction maintained by hand. The aspirate was expelled into the plastic syringe.
The chest radiologist who performed PTNA decided on a case-by-case basis whether PCR for detection of TB was necessary, according to the reasons given previously. In all cases, lung aspirates from each PTNA were handled three different ways for histologic, cytologic, and PCR analyses. When a tiny tissue core was aspirated into the syringe, it was individually placed in formalin for histologic evaluation. The rest of the aspirate was submitted for cytologic analysis. Residual material in the lumen of the needle or syringe was submitted for PCR analysis.
PCR for Detecting M tuberculosis
PCR was performed with the lung aspirates (by a technician under the supervision of the pathologist [C.K.L.]) by using an M tuberculosis kit (Amplicor; Roche Diagnostic Systems, Somerville, NJ), which amplified part of the 16S rRNA gene. The complete M tuberculosis test was performed according to the manufacturer’s instructions. Briefly, DNA of lung aspirates was extracted with the sample preparation kit. A portion (50 µL) of pretreated patient samples and a positive and a negative control were processed with a PCR thermocycler (GeneAmp 9600; Roche Diagnostics, Pleasanton, Calif) for 37 cycles (duration, 1.5 hours). The cycle parameters were 20 seconds at 98°C (cycles 1 and 2), 20 seconds at 94°C (cycles 3–37), 20 seconds at 62°C, and 45 seconds at 72°C. After amplification, hybridization was performed in probe-coated microwells. The absorbance at 450 nm, or A450, was measured, and samples with A450 values of 0.35 or greater were considered to be positive for the presence of M tuberculosis.
Statistical Methods
The sensitivity, specificity, and positive and negative predictive values of TB-PCR in specimens from PTNA for the diagnosis of TB were calculated.
|
RESULTS
|
|---|
PTNA Cytologic and Histologic Diagnosis
Among 19 malignant lesions, cytologic and histologic analysis with PTNA material showed that 18 were positive for malignant cells and one was negative. In 14 of these 18 lesions, a diagnosis of the specific cell type was established, including distinction between small cell and non–small cell carcinoma. In four of 18 malignant lesions, the specific cell type could not be determined. The lesion that was negative for malignant cells in the PTNA sample was confirmed to be malignant with histologic analysis of bronchoscopic biopsy material (Table 1).
Of 26 nonmalignant lesions, 25 were negative for malignant cells and the sample from one was inadequate for diagnosis. One benign neoplasm was diagnosed as a hamartoma. In 17 patients with TB, 16 did not have malignant cells and the sample from one was inadequate for diagnosis. In two, a specific diagnosis of TB was established. In eight patients with an infection other than TB, the aspirates in none showed malignant cells and the aspirates in three showed actinomycosis (n = 2) and aspergillosis (n = 1) (Table 1).
For detection of malignancy, PTNA had a sensitivity of 95% (18 of 19) and a specificity of 96% (25 of 26). With regard to the 26 nonmalignant lesions, a specific diagnosis of benign lesion was rendered in six (23%) cases, and a nonspecific diagnosis of benignity was rendered in 19 (73%); the sample in one case was inadequate for diagnosis. A specific diagnosis of TB was rendered in two (12%) of 17 cases (Table 1).
PCR for Detecting TB in PTNA Specimens
In 17 patients with TB, PCR results were positive in 11 and negative in six (Table 1). In 11 patients with positive PCR results, the diagnosis in seven was established by means of bacteriologic and/or cytologic and histologic analysis; the diagnosis in four was established with therapeutic response. In six patients with negative PCR results, the diagnosis in three was established by means of bacteriologic and/or cytologic and histologic analysis, and the diagnosis in the other three was established with therapeutic response (Table 2). A specific diagnosis of TB was established on the basis of histologic results in two patients, one of whom had positive results at both histologic analysis and PCR for detection of TB. By adding the results of PCR for TB, a specific diagnosis of TB was rendered in 12 (71%) of 17 cases (Figs 1, 2).
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Transverse chest CT scan obtained at the level of the aortic arch in a 43-year-old man with pulmonary TB. A well-defined 2-cm-diameter nodule with a cavity (arrow) is visible in the left upper lobe. Bacteriologic results in sputum were negative. Fluoroscopy-guided PTNA was performed, and the results were negative both for malignant cells and for a histologic diagnosis of TB. Results of PCR for TB in PTNA lung aspirates were positive. This lesion showed improvement after medication for TB.
|
|
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Transverse contrast material-enhanced chest CT scan in a 36-year-old man with pulmonary TB shows triangular consolidation with an internal low-attenuating area (arrow) in the left upper lobe. Results of M tuberculosis culture and AFB stain with sputum were negative. US-guided PTNA was performed, and the results were negative both for malignant cells and for a histologic diagnosis of TB. Results of PCR for TB in PTNA lung aspirates were positive. Additionally, bacteriologic results with pleural fluid were positive. This lesion showed improvement after medication for TB.
|
|
Results of PCR for detection of TB were negative in all 28 patients without TB. In 19 patients with a malignant lesion, PCR for TB detection with PTNA specimens was negative in all 19 patients (Fig 3). PCR results were negative in one patient with hamartoma and in eight patients with pneumonia (Table 1).
View larger version (88K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Transverse chest CT scans in a 37-year-old man with adenocarcinoma. (a) A well-defined 3.5-cm-diameter mass (arrow) is present in the right upper lobe apical segment. (b) A nodule with a cavity (arrow) is visible in the left upper lobe. Numerous variable-sized small nodules are scattered in both lungs, which was suggestive of metastasis on follow-up chest radiographs and CT scans (not shown). In this patient, sputum AFB stain was positive initially (1 month before the CT scans were obtained), so medication was started, and the cavitary nodule in left upper lobe improved during 1 month. The CT diagnosis for the right upper lobe lesion was a malignant mass. Fluoroscopy-guided PTNA was performed on the right upper lobe lesion, and adenocarcinoma was diagnosed. Results of PCR for TB detection in PTNA lung aspirates were negative.
|
|
View larger version (94K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Transverse chest CT scans in a 37-year-old man with adenocarcinoma. (a) A well-defined 3.5-cm-diameter mass (arrow) is present in the right upper lobe apical segment. (b) A nodule with a cavity (arrow) is visible in the left upper lobe. Numerous variable-sized small nodules are scattered in both lungs, which was suggestive of metastasis on follow-up chest radiographs and CT scans (not shown). In this patient, sputum AFB stain was positive initially (1 month before the CT scans were obtained), so medication was started, and the cavitary nodule in left upper lobe improved during 1 month. The CT diagnosis for the right upper lobe lesion was a malignant mass. Fluoroscopy-guided PTNA was performed on the right upper lobe lesion, and adenocarcinoma was diagnosed. Results of PCR for TB detection in PTNA lung aspirates were negative.
|
|
For the diagnosis of TB, the sensitivity and specificity of PCR with PTNA lung specimens were 65% (11 of 17) and 100% (28 of 28), respectively. The positive and negative predictive values were 100% (11 of 11) and 82% (28 of 34), respectively (Table 3).
|
DISCUSSION
|
|---|
The radiograph is the major diagnostic tool for the recognition of TB. However, approximately 10% of patients who are subsequently proved to have TB had a pulmonary infiltrate that was not thought to be characteristic of TB on radiographs (15). TB shows variable patterns that mimic benign and malignant lung disease. One of the most common causes for radiographic misdiagnosis is the presence of a large nodule or masslike opacity that is considered to be a neoplasm (15). Sometimes, it seems more difficult to predict TB than it is to predict malignancy in solitary pulmonary nodules on CT scans; the reported (14) prediction rate for TB was 62.5%, and that for malignancy was 88.2%. Furthermore, reactivation TB is an occasional complication in patients with malignancy, and it is not uncommon to encounter TB and lung cancer at the same time and in the same lobe (16). Therefore, radiologists occasionally render a diagnosis of pulmonary TB with difficulty, even though CT is performed for diagnosis, because TB shows unusual radiologic findings that simulate benign and malignant lung disease and because TB coexists with lung carcinoma.
A definitive diagnosis of pulmonary TB requires culture of M tuberculosis or a smear showing AFB in appropriate materials. In areas where TB is endemic, sputum is the easiest and most valuable source of the organism. When sputum is nondiagnostic, bronchoscopy is valuable for the bacteriologic investigation of pulmonary TB. In particular, PTNA can be useful in patients in whom a diagnosis is difficult with results of routine bacteriologic and radiologic examinations.
PTNA has become a popular and reliable method for the diagnosis of pulmonary lesions. In the hands of an experienced radiologist, the yield of PTNA in patients with malignant lung lesions has been shown to be greater than 90% (1–4,17). In contrast, the yield of PTNA as regards a specific diagnosis of benign disease in patients with a lung mass is variable and relatively low and has been reported to range between 11.7% and 68% (1,4,17). The results of the present study were similar to those in the previous reports. For detection of malignancy, PTNA showed 95% sensitivity and 96% specificity. However, a specific diagnosis of benign lesion was established in 23% of cases, and a specific diagnosis of TB was established in 12%. This poor yield for a specific diagnosis of benign lesion, including TB, with PTNA was due to the following: We used a fine needle, so materials for histologic examination may have been inadequate in some cases; an on-site cytopathologist was unavailable; and a definitive diagnosis of TB was rendered only when necrotizing caseous granulomatous inflammation was seen histologically, to reduce the false-positive rate for TB. Generally, factors that detract from the utility of PTNA were false-negative results for cancer and infrequent specific diagnosis of benign lesions (2). Because nonspecific cancer-negative biopsy results lead to clinical uncertainty, many patients with benign conditions and nonspecific negative biopsy findings undergo diagnostic thoracotomy to rule out malignancy. So, various maneuvers have been attempted to improve biopsy results, including repeated biopsy and biopsy with a large-bore needle to improve the results of PTNA (2,17). Also, the presence of an expert cytopathologist and proper handling of specimens improve the yield of PTNA.
PCR has been used to identify mycobacterial DNA sequences in uncultured clinical specimens, and it offers great promise for rapid and reliable identification and characterization of mycobacteria. There are many reports (5–13) of the demonstration of M tuberculosis by means of PCR. In cases of pulmonary TB, PCR results are most often investigated by using sputum and materials obtained at bronchoscopy (11–13). PCR for TB detection in bronchial aspirates obtained at bronchoscopy in patients with smear-negative pulmonary TB has shown 97.2% sensitivity, 73.2% specificity, 82.7% positive predictive value, and 95.2% negative predictive value (12). PCR for TB detection can also be used with lung tissue and pleural fluid (6,9,10). Material obtained at direct puncture of a lung lesion is probably very valuable for PCR for TB detection, providing additional information for cytologic and histologic diagnosis, and PCR for detection of TB in lung aspirates obtained at PTNA could be a reasonable alternative method to improve the results of PTNA.
In a previous study (14) at our institution, we performed nested PCR for detecting M tuberculosis in solitary pulmonary nodules by using lung aspirates obtained at PTNA and found good sensitivity (87.5%) and specificity (96.0%). Since publication of that previous report and using its results, we performed PCR for detection of TB selectively in specimens obtained at PTNA. The chest radiologist who performed PTNA decided whether PCR was necessary in each case.
In the present study, we did not confine the cases to solitary pulmonary nodules but included both particular lesions for which radiologic or clinical diagnosis was difficult and presumed TB lesions that were not confirmed bacteriologically. These lesions manifested as nodules (n = 10), masses (n = 26), or consolidations (n = 9) on CT scans. We also used primary PCR methods instead of nested PCR to improve specificity in this study. The sensitivity and specificity of PCR for TB detection in lung aspirates from PTNA were 65% (11 of 17) and 100% (28 of 28), respectively. The positive and negative predictive values were 100% (11 of 11) and 82% (28 of 34), respectively. These results showed less sensitivity (65% vs 87.5%) but more specificity (100% vs 96%) than those in the previous report (14). The lower sensitivity in the present study was due to the PCR method we used (primary rather than nested PCR). In primary PCR one primer pair and the one round of PCR are used, while in nested PCR two pairs of amplification primers and two rounds of PCR are used; therefore, in general, nested PCR will have higher sensitivity than primary PCR.
The most important limitation to routine use of PCR for diagnosis of TB is the relatively low sensitivity (65%), but sputum bacteriologic investigations showed positive results in only four (24%) of 17 patients in whom the diagnosis was particularly difficult. PCR showed higher sensitivity than did sputum bacteriologic analyses. There were six (35%) false-negative results for PCR in this study. These false-negative results might be due to inhibitors of DNA polymerase in specimens. It is known that hemoglobin, especially porphyrin, is the most potent inhibitory substance among the inhibitors. Our PTNA specimens usually contained blood, so inhibitory substances may have remained in the reaction mixture, although specimens were washed during the pretreatment process. AFB-positive specimens showed a lower rate of false-negative results than did AFB-negative specimens. In our cases, AFB-negative specimens yielded positive PCR results, even though they contained blood. This suggests that if there is a sufficient amount of target DNA in the specimen, the inhibitory effect of hemoglobin can be overcome. Furthermore, the most important points of PCR are that it is highly specific and that PCR for TB detection yielded no false-positive results and so had a 100% positive predictive value in PTNA specimens in our study.
In conclusion, PCR provides a rapid means for detecting M tuberculosis in PTNA specimens from patients who were previously found to have negative results with other bacteriologic investigations. PCR in PTNA specimens is highly reliable for confirmation and interpretation of pulmonary TB and differentiation from neoplasm and other infections. PCR for detection of TB would be a good adjuvant to histologic and cytologic analyses for the definitive and differential diagnoses of TB in PTNA specimens, because of the increased accuracy of results obtained with PTNA.
|
FOOTNOTES
|
|---|
Abbreviations: AFB = acid-fast bacilli,
PCR = polymerase chain reaction,
PTNA = percutaneous transthoracic needle aspiration,
TB = tuberculosis
Author contributions: Guarantor of integrity of entire study, E.Y.K.; study concepts, E.Y.K., J.J.S.; study design, E.Y.K., J.J.S., C.K.L.; literature research, E.Y.K., B.K.S.; clinical studies, J.J.S., C.K.L.; data acquisition, J.A.C., B.K.S.; data analysis/interpretation, E.Y.K., Y.W.O.; statistical analysis, Y.W.O., E.Y.K.; manuscript definition of intellectual content, E.Y.K., Y.W.O., B.K.S.; manuscript editing, E.Y.K.; manuscript preparation, revision/review, and final version approval, E.Y.K., J.A.C.
|
REFERENCES
|
|---|
- Khouri NF, Stitik FP, Erozan YS, et al. Transthoracic needle aspiration biopsy of benign and malignant lung lesions. AJR Am J Roentgenol 1985; 144:281-288.[Abstract/Free Full Text]
- Greene R, Szyfelbein WM, Isler RJ, Stark P, Jantsch H. Supplementary tissue-core histology from fine-needle transthoracic aspiration biopsy. AJR Am J Roentgenol 1985; 144:787-792.[Abstract/Free Full Text]
- Wescott JL. Percutaneous transthoracic needle biopsy. Radiology 1988; 169:593-601.[Free Full Text]
- Perlmutt LM, Johnston WW, Dunnick NR. Percutaneous transthoracic needle aspiration: a review. AJR Am J Roentgenol 1989; 152:451-455.[Free Full Text]
- Pao CC, Yen TS, You JB, Maa JS, Fiss EH, Chang CH. Detection and identification of Mycobacterium tuberculosis by DNA amplification. J Clin Microbiol 1990; 28:1877-1880.[Abstract/Free Full Text]
- De Wit D, Steyn L, Shoemaker S, Sogin M. Direct detection of mycobacterium tuberculosis in clinical specimens by DNA amplification. J Clin Microbiol 1990; 28:2437-2441.[Abstract/Free Full Text]
- Hermans PWM, Schuitema ARJ, Soolingen DV, et al. Specific detection of mycobacterium tuberculosis complex stains by polymerase chain reaction. J Clin Microbiol 1990; 28:1204-1213.[Abstract/Free Full Text]
- Kim HJ, Hyun IK, Lee MK, et al. Diagnosis of tuberculous cervical lymphadenitis using polymerase chain reaction. Tuberc Respir Dis 1995; 42:35-41.
- Kuwano K, Minamide W, Kusunoki S, et al. Evaluation of nested polymerase chain reaction for detecting mycobacterial DNA in pleural fluid. Kansenshogaku Zasshi 1995; 69:175-180[English abstract].[Medline]
- Aslansadeh J, Viuda M, Fille M, Smith WB, Namdari H. Comparison of culture and acid-fast bacilli stain to PCR for detection of Mycobacterium tuberculosis in clinical samples. Mol Cell Probes 1998; 12:207-211.[CrossRef][Medline]
- Tan YK, Lee AS, Khoo KL, Ong SY, Wong SY, Ong YY. Rapid mycobacterial tuberculosis detection in bronchoalveolar lavage samples by polymerase chain reaction in patients with upper lobe infiltrates and bronchiectasis. Ann Acad Med Singapore 1999; 28:205-208.[Medline]
- Wong CF, Yew WW, Chan CY, Au LY, Cheung SW, Cheng AF. Rapid diagnosis of smear-negative pulmonary tuberculosis via fiberoptic bronchoscopy: utility of polymerase chain reaction in bronchial aspirates as an adjunct to transbronchial biopsies. Respir Med 1998; 92:815-819.[CrossRef][Medline]
- Yamada M, Furuse K, Kawahara M, et al. Detection of mycobacterium tuberculosis in transbronchial biopsy specimens by polymerase chain reaction. Nihon Kyobu Shikkan Gakkai Zasshi 1994; 32:747-751[English abstract].[Medline]
- Shim JJ, Cheong HJ, Kang EY, In KH, Yoo SH, Kang KH. Nested polymerase chain reaction for detection of mycobacterium tuberculosis in solitary pulmonary nodules. Chest 1998; 113:20-24.[Abstract/Free Full Text]
- Woodring JH, Vandiviere HM, Fried AM, Dillom ML, Williams TD, Melvin IG. Update: the radiographic features of pulmonary tuberculosis. AJR Am J Roentgenol 1986; 146:497-506.[Abstract/Free Full Text]
- Miller WT. Tuberculosis in 1992. In: Greene R, Muhm JR, eds. Syllabus: a categorical course in diagnostic radiology—chest radiology. Oak Brook, Ill: Radiological Society of North America, 1992; 157-166.
- Klein JS, Zarka MA. Transthoracic needle biopsy. Radiol Clin North Am 2000; 38:235-266, vii.[CrossRef][Medline]
TOP
ABSTRACT
INTRODUCTION
