Azlocillin

Background: Tuberculous pleural effusion occurs in 30% of patients with tuberculosis (TB). Rapid diagnosis of a tuberculous pleural effusion would greatly facilitate the management of many patients. Polymerase chain reaction (PCR) has been used to detect Mycobacterium tuberculosis in pleural fluid with highly variable sensitivity.

Objective: To improve our laboratory diagnosis of tuberculous pleural effusion.

Methods: We applied PCR to detect DNA specific for M tuberculosis in 33 of the studied pleural biopsy specimens using an IS986-based primer that was specific for mycobacterium complex, and compared it to the results of pleural fluid and biopsy cultures performed on either Lowenstein-Jensen (LJ) medium or BACTEC 12B liquid medium (Becton Dickinson Microbiology Systems; Coekeysville, MD), Ziehl-Neelsen (ZN) staining, and histopathology in 45 patients with pleural effusion.

Results: Of the 45 patients with pleural effusion who were studied, 26 patients received diagnoses of tuberculous pleural effusion that had been confirmed by either culture and or hlstopathology, 10 patients received diagnoses of executive effusion due to causes other than TB, and 9 patients received diagnoses of transudative effusion. Histopathology of the pleural biopsy specimen had a sensitivity of 53.8%. The sensitivity of the ZN staining of pleural fluid and biopsy specimens was 0.0% and 3.8%, respectively. The sensitivity of the culture on both BACTEC 12B liquid medium and LJ medium was higher in pleural biopsy specimens (92.3%) than in pleural fluid specimens (15.4%; p > 0.001). The improvements of the BACTEC culture system improved and shortened the detection time of M tuberculosis in pleural biopsy specimens. PCR of pleural biopsy specimens had 90% sensitivity and 100% specificity. The positive predictive value and the negative predictive value for pleural biopsy specimen cultures were 100% and 90.5% vs 100% and 86.7% for pleura] biopsy specimen PCRs.

Conclusion: The overall accuracy of PCR of pleural biopsy was similar to the results of pleural biopsy culture, however, PCR of the pleural biopsy was much faster in reaching diagnosis. PCR of pleural biopsy is a useful method when used in combination with the BACTEC culture system and histopathologic examination of pleural biopsy to reach a rapid diagnosis of tuberculous pleural effusion.

Key words: BACTEC system culture; pleural biopsy; pleural effusion; polymerase chain reaction; tuberculosis pleural effusion

Abbreviations: AFB = acid-fast bacilli; bp = base pair; LJ = Lowenstein-Jensen; NAP = P-nitro-acetylamino [alpha] hydroxy-proionophenone: NPV = negative predictive value: PCR = polymerase chain reaction; PPV = positive predictive value; TB = tuberculosis; ZN = Ziehl-Neelsen

**********

Tuberculosis (TB) is a major health problem especially in developing countries. The most effective approach for controlling the disease is accurate diagnosis and treatment of active TB cases. (1) tuberculous pleural effusion occurs in up to 30% of patients with TB (2) and constitutes the major portion of the extrapulmonary TB morbidity. (3) Diagnosis of tuberculous pleural effusion depends oil the demonstration of the presence of tubercle bacilli in the sputum, pleural fluid, or pleural biopsy specimen, or the demonstration of granuloma in the pleura by histopathologic examination. (4) In many cases, the findings of the histopathologie examination of the pleural biopsy specimen may be negative or nonspecific. (4) Microbiological methods for diagvosing tuberculous pleural effusion include Ziehl-Neelsen (ZN) acid-fast bacilli (AFB) stain and laboratory culture of the causative organism Mycobacterium tuberculosis. The detection of mycobacterium in pleural fluid by microscopy has low sensitivity and cannot be used to distinguish between the various members of the mycobacterium genus. (5) On the other hand, a laboratory culture of M tuberculosis is sensitive, but it takes up to 8 weeks to yield a positive result. (6, 7) The radiometric BACTEC 460 TB system (Becton Dickinson Diagnostic Instrument System; Sparks, MD) has been an important improvement in culture methods. The use of this method has reduced the average detection time of both smear-positive and smear-negative samples by nearly 50%. (5, 8) The BACTEC 460 TB system also has been adopted to differentiate M tuberculosis from other mycobacteria as well as to perform antimicrobial susceptibility testing, particularly for M tuberculosis. (9) Diagimstic assays based on nucleic acid amplification methods such as polymerase chain reaction (PCR) dramatically decrease the time required to identify an organism in clinical specimens, (10) This method has been tested extensively for the detection of M tuberculosis in sputum specimens, (10-13) la cerebrospinal fluid, (14) and pericardial fluid and biopsy specimens. (15) PCR also has been used to detect M tuberculosis in pleural fluid samples, with highly variable sensitivity (11 to 81%) in different studies (16-18) Comparatively little work has focused on the utility of PCR and the BACTEC 460 TB system in detecting M tuberculosis in pleural biopsy specimens.

In this study, we examined whether PCR of pleural biopsy specimens would be superior to that of pleural fluid in detecting M tuberculosis. We applied PCR to detect DNA (IS986) specific for M tuberculosis complex in pleural biopsy specimens mid compared the results to those of pleural fluid and biopsy cultures using the BACTEC system, ZN examination, and histopathology.

MATERIALS AND METHODS

The present study comprised 45 patients with pleural effusion, who had been admitted to the Thoracic Medicine Department, Mansoura University, Egypt, between February and December 2000.

Study Design

The study was performed prospectively in a blinded manner in which the clinical diagnosis was not available to the laboratory personnel. Informed consent was received from each patient involved in the study. This study was approved by Human Subject Research Committee. Each patient underwent the following procedures:

1. Mantoux tuberetdin skin test, in which an induration of > 10 mm was considered to be positive;

2. Radiologic examination including the following:

a. Chest radiographs in the posteroanterior and lateral views;

b. Ultrasound of the pleural space to identify the appropriate site for thoracentesis; and

c. CT scan to determine the presence of any parenchymal lesion underlying the effusion;

3. Laboratory investigation including CBC, blond sugar test, liver function, and kidney function;

4. Sputum examination, including Gram and ZN staining, and culture; and

5. Diagnostic thoracentesis and pleural biopsy using die Abram needle.

Methods

Processing of Pleural Fluid and Biopsy Samples': One hundred milliliters of the pleural fluid was aspirated and divided into two parts, one fur cytology examination and the other for microbiology studies. Pleural fluid samples sent to the Microbiology Department were subjected to decontamination and then were concentrated by eentrifugation for 20 rain at 10,000g at 4[degrees]C, Then the sediments were resuspended in sterile normal saline solution, and used for ZN smear and culture. The pleural biopsy was performed using Abram needles, and 4 to 5 punches of parietal pleura were divided equally into two parts. One part of the specimen was fixed in formation for histopathologic examination, while the other part was put into sterile normal saline solution, then homogenized and used for ZN smear, culture, and PCR studies.

Laboratory Methods

ZN Smear of Pleural Fluid and Biopsy: Smears of pleural fluid sediments or pleural biopsy specimens were stained by ZN stain according to the method of Jenkins. (19) The slide then was examined under an oil immersion lens for the presence of AFB.

Loewenstein-Jensen Culture of Pleural Fluid and Biopsy Specimen: Ready to use bottles of Loewenstein-Jensen (LJ) medium (Hispan Lab; Medicopharmatrade Co; Madrid, Spain) were inoculated with 0.1 mL concentrated pleural fluid or biopsy specimen and were kept in a C[O.sub.2] incubator for S weeks. Bottles were inspected twice a week for visible colonies, and suspicious growth was subjected to ZN staining. Negative culture was discarded alter 8 weeks. Resulting growth was left in light for 2 h and was examined for yellow pigments to identify photochromogen species.

Pleural Fluid and Biopsy Culture Using Badiometric BACTEC 460 System: The BACTEC 460 system and BACTEC 12B vials (Becton Dickinson Microbiology Systems; Cockeysville, MD) were used for radiometric culturing of M tuberculosis. BACTEC 12B vials were prepared by the addition of dm antimicrobial agent PANTA which contains polymyxin, amphotericin B, nalidixic acid, trimethoprim, and azlocillin. One half milliliter concentrated pleural fluid or biopsy specimen was inoculated into each BACTEC 12B Vial, was kept at 37[degrees]C, and then was tested three times a week, with vials showing a growth index of [greater than or equal to] 10 being considered positive and tested daily.

BACTEC Differentiation Test: When the inoculated BACTEC 12 B vials showed a growth index of 50 to 100, a Gram stain and subculture were performed to make sure that the culture was not contaminated, then the culture was homogenized with a tuberculin syringe, and 1 mL was transferred to the vial containing P-nitro acetylamino [alpha]-hydroxy-proionophenone (NAP). The NAP vial and the control vial (ie, the original culture) were tested daily with the BACTEC instrument for the next 2 to 6 days. A decreased or unchanging GI in the NAP vials indicated the presence of TB complex, while an increase in GI indicated the presence of a mycobacterium other than TB.

PCR of Pleural Biopsy Specimen: PCR was used for the detection of DNA (IS986) specific for the mycobacterium complex on 33 of the studied pleural biopsy samples. The size of the amplification product was 123 base pairs (bp). A PCR reaction was performed using an M tuberculosis kit (code H.02; Experteam: Venice, Italy) according to the manufacturer's instructions.

DNA Extraction: Pleural biopsy specimens (2 mm in diameter, and 2 to 3 punches per patient) were homogenized in normal saline solution using a sterile wide-gauge needle for each patient sample to avoid cross contamination. Pleural biopsy homogenates were centrifuged at 5,000g for 10 ruin, and the pellet was resuspended in 162 [micro]L lysis buffer. Eighteen microliters each lysozyme and mutzmolysin were added to the pellet, which then was incubated at 37[degrees]C for 1 h. Then, after 20 [micro]L proteinase K and 200 [micro]L buffer alkaline lysate were added to the pellet and incubated at 56[degrees]C for 30 min, and for a further 30 min at 95[degrees]C. Later, DNA was extracted from the pellet with mini spin columns (QIAamp; Qiagen; Valencia, CA). To monitor for cross-contamination, one water-containing negative control tube was used per five sample tubes.

PCR: A typical PCR reaction mixture contained 2.5 [micro]L 10 x buffer, 2.5 [micro]L deoxynucleoside triphosphates, and 1 [micro]L each primers BMT002 (CCTGCGAGCGTAGGCGTCGG) and BMT003 (CTCGTCCAGCGCCGCTTCGG), to which 0.3 [micro]L Taq DNA polymerase, 7.7 [micro]L distilled water, 10 [micro]L DNA solution extracted with QIAamp culumns from each pleural biopsy sample, and 25 [micro]L mineral oil were added. Each set of the PCR reaction contained a positive control containing DNA extracted from TB bacilli that had been isolated in our lahoratory and two negative controls, one containing the same amount of DNA extracted from the lymphocytes of healthy individuals and the other containing the same amount of distilled water. The thermal cycle was programmed for 5 rain at 940G for 30 cycles, denaturation at 94[degrees]C for 2 min, annealing at 68[degrees]C for 2 min, and extension at 72[degrees]C for 2 rain. The amplified products were detected by gel electrophoresis using 2% agarose gel with ethidium bromide, and a 123-bp amplified band was visualized on an ultraviolet transilluminator.

Statistical Analysis

The results of clinical evaluation and diagnostic tests were analyzed using computer software (SPSS, version 10; SPSS; Chicago, IL). The results of ZN examination, culture, histopathology, and PCR were compared to the final diagnosis of TB with individual patients used as the unit of analysis. The data were presented as percentages and were compared using the [chi square] test, with a p value of < 0.05 considered to be significant. Culture for AFB and histopathologic examination for caseating granuloma were used as the "gold standard" tests.

RESULTS

Patient Selection

The study included 45 patients (27 men and 18 women) with a mean ([+ or -] SD) age of 46 [+ or -] 2.30 years. None of our patients were tested for HIV, as HIV infection is uncommon in our setting. The patients were categorized into the following three groups.

Group 1: Tuberculous Pleural Effusion: This group included 26 patients (17 men and 9 women) with a mean age of 43.15 [+ or -] 2.8 years. Diagnosis of pleural TB was confirmed by culture and/or histopathology of pleural biopsy specimens, supported by laboratory, radiographic, and clinical data. Eight of 26 patients (30.8%) in this group presented with left-sided pleural effusion, 16 of 26 patients (61.5%) presented with right-sided pleural effusion, and 2 of 26 patients (7.7%) presented with bilateral pleural effusion. There was no underlying parenchymal infiltration, cavitation, or lymphadenopathy.

Group 2: Exudative Pleural Effusion: This group included 10 patients (six men and four women) with a mean age of 47.3 [+ or -] 5.96 years. Within this group, malignant pleural effusion was diagnosed in seven patients (small cell carcinoma, two patients; lymphoma, one patient; mucoepidermoid, one patient; adenocarcinoma, two patients; 0rod undifferentiated carcinoma, one patient), systemic lupus erythematous (8LE) was diagnosed in two patients, and parapneumonic effusion was diagnosed in one patient. Five of 10 patients (50%) presented with left-sided pleural effusion, 2 of 10 patients (20%) presented with right-sided pleural effusion, and 3 of 10 patients (30%) presented with bilateral effusion. Five of i0 patients (50%) had underlying parenchymal infiltrates or lymphadenopathy but no cavitation.

Group 3: Transudative Pleural Effusion: This group included nine patients (four men and five women) with a mean age of 52.78 [+ or -] 5.37 years. Bilharzial liver cirrhosis was diagnosed in five patients, and congestive heart failure was diagnosed in four patients. Eight of nine patients (88.9%) presented with right-sided pleural effusion, and one of nine patients (11.1%) presented with left-sided pleural effusion. There was no underlying parenchymal infiltration, lymphadenopathy, or cavitation.

Tuberculin Skin Test

Tuberculin skin test results were positive in 19 of 26 patients (7:3.1%) in the tuberculous pleura] effusion group and were negative in all other patients having exudative and transudative pleural effusion. ZN Staining of Sputum, Pleural Fluid, and Biopsy Specimen

ZN staining results of sputum and pleural fluid were negative in all three groups. ZN staining results of pleural biopsy specimens were positive in only I of 26 patients (3.8%) in the tuberculous pleura] effusion group, and were negative in both the exudative and transudative pleura] effusion groups (Tables 1 and 2).

Culture of Sputum, Pleural Fluid, and Biopsy Specimens on Either LJ Medium or BACTEC System

Sputum culture on LJ medium showed no mycobacterial growth. A culture of pleural fluid on both LJ medium and BACTEC 12B medium showed the growth of TB bacilli in 4 of 26 patients (15.4%) in the tuberculous pleural effusion group. A pleural biopsy specimen culture using BACTEC culture system showed a higher detection rate of M tuberculosis (24 of 26 patients; 92.3%) than did a culture on LJ medium (22 of 26 patients; 84.6%) in the tuberculous pleural effusion group. Pleural fluid and biopsy specimen cultures showed no mycobacterium growth in nontuberculous pleural effusions. A pleural biopsy culture was negative in two cases of tuberculous pleural effusion, in which the diagnosis was confirmed by histopathology showing caseating granuloma (Tables 1 and 2). The mean detection time of M tuberculosis was 13 days for BACTEC 12B medium vs 28 days for LJ medium.

PCR Detection of TB Bacilli in Pleural Biopsy Specimen

PCR was performed in only 33 of the studied pleural biopsy specimens (tuberculous pleural effusion, 20 patients; exudative effusion, 7 patients; and transudative effusion, 6 patients). The PCR result was positive in 18 of 20 patients in the tuberculous pleural effusion group, and in none of the patients in the transudate and exudate pleural effusion groups (Fig 1). PCR had a false-negative result in two patients with tuberculous effusion in whom the diagnosis of TB was confirmed by bistopathology showing caseating granuloma and by the response of these patients to antituberculous drugs. Based on the results of pleural biopsy specimen culture, the PCR of the pleural biopsy specimen had 100% sensitivity, but, based on the results of combined culture and histopathology, it had 90% sensitivity, 100% specificity, 93.3% accuracy, 100% positive predictive value (PPV), and 86.6% negative predictive value (NPV) [Table 2].

[FIGURE 1 OMITTED]

Histopathology of Pleural Biopsy Specimen

Histopathologic examinations of the pleural biopsy specimens showed a caseating granuloma with central necrosis in 14 of 26 patients (53.8%) with tuberculous pleural effusions. In the exudative pleural effusion group, histopathology was diagnostic of malignant effusion in seven patients and showed nonspecific pleurisy in three patients (systemic lupus erythematosus was diagnosed in two patients, and parapneumonic effusion was diagnosed in one patient). The diagnosis of systemic lupus erythematosus was based on a positive result of testing for anti-double-strand DNA antibodies and LE cells, in addition to other clinical criteria. The diagnosis of parapneumonic effusion was confirmed by Gram staining and culture of sputum showing Gram-positive cocci (Streptococcal pneumonia) and also by the response of the effusion to antibiotic therapy. In the transudative effusion, histopathology revealed normal pleura (Tables 1, 2).

DISCUSSION

Rapid diagnosis and treatment of tuberculous pleura[ effusion is crucial to reduce morbidity and mortality from untreated TB. Because the differential diagnosis of exudative pleura] effusion is broad, the rapid diagnosis of tuberculous pleural effusion would greatly facilitate the management of these patients. At present, the most reliable method for the diagnosis of tuberculous pleural effusion is the identification of M tuberealosis in the pleural specimens. The diagnosis of tuberculous pleural ellusion depends on the demonstration of tubercle bacilli in the sputum, pleural fluid, or pleural biopsy specimens, or the demonstration of granuloma ill pleural biopsy specimens, (4) with high sensitivity, of the latter up to 80% and low sensitivity" of the former. (21) Histopathology alone cannot reliably establish the diagnosis because it cannot distinguish disease caused by M tuberculosis from that caused by mycobacteria other than TB (22) or other granuloma.4 In this study, our goal was to improve our diagnosis of tuberculous pleural effusion and to determine the most sensitive, specific, and rapid diagnostic methods. We used PCI/to detect DNA (IS986) specific for M tuberculosis complex on pleural biopsy specimens and compared it to the results of pleural biopsy cultures using the BACTEC 460 system and LJ medium.

Our study showed that the results of sputum ZN staining and culture for AFB on LJ medium were negative in all cases. A sputum study, that was negative for AFB in our study was explained by the absence of any underlying pulmonary parenchymal disease, as determined by radiologic studies, and the sputum induction method that has been reported to improve the yield of sputum culture for mycobactetium2a was not used. Our result was ill agreement with that of the previous study (24) showing low sensitivity of the sputum examination in pleural effusion patients with normal chest radiograph findings.

ZN staining had a low detection rate of M tuberculosis in pleural fluid (0%) and pleural biopsy specimens (3.8%). A culture of the pleural fluid on both LJ medium and BACTEC 12B medium showed a lower sensitivity (15.4%) in comparison with the 92.3% seen in cultures of pleural biopsy specimens. The lower detection rate in pleural fluid samples studied with ZN staining and culture when compared to that of pleural biopsy ZN staining and culture can be explained by the presence of viable bacilli in the pleural tissue causing infection, in which the pleural fluid is the manifestation of a hytersensitivity reaction to the mycobacterium. (25) Pleural biopsy specimens cultured on BACTEC 12B medium had high sensitivity and short detection time when compared to pleural biopsy specimen cultures on LJ medium. The detection time of mycobacterimn was shortened from a mean of 25 days using LJ medium to a mean of 13 days when BACTEC 12B medium was used.

Histopathologic examination of the pleural biopsy specimen had a sensitivity of 53.8% and was a rapid method for reaching the diagnosis of tuberculous pleural effusion, however, its sensitivity was low in comparison to that of pleural biopsy culture (92.3%). These results are in agreement with the results reported by Katiyar et al (26) that showed that the culture of pleural biopsy was more sensitive in diagnosing tuberculous pleural effusion when compared to the pathologic examination of pleural biopsy specimens. In contrast, other studies have found that the histopathology of pleural biopsy specimens was more sensitive than the pleural biopsy specimen culture in the diagnosis of tuberculous pleural effusion. (25) The difference in detection rates in the histopathologic examinations in these studies could be attributed to the fact of repeated sampling needed for histopathologic examinations to detect caseating lesions, while for cultures one sample seems to be sufficiently In addition to the need for multiple sampling, histopathology alone cannot distinguish between a disease caused by M tuberculosis from other causes of granuloma. (4, 28)

In our study, a PCR of a pleural biopsy specimen had a 90% sensitivity, a 100% specificity, a 100% PPV, a 86.6% NPV, and 93.9% accuracy, which are values that approach those of pleural biopsy specimen culture using both BACTEC B12 medium and LJ medium. The results of pleural biopsy specimen PCB and culture were negative in only two patients with tuberculous pleural effusion, in which the diagnosis was established by histopathology showing caseating granuloma and by the positive response to antituberculous drugs. False-negative results of pleural biopsy specimen PCR and culture can be explained by the presence of a low number of TB bacilli in these lesions where immunologic reaction with extensive caseation was predominant. The sensitivity of nucleic amplification techniques depends on the number of mycobacteria, their homogenous distribution in the sample, the presence of the amplification inhibitor in the sample (29) and type of the primer. (30) In our study, we select the primer pair amplifying 123 bp, which is specific for members of the M tuberculosis complex. (31) Also, this primer pair has been reported (30) to be the most sensitive primer of three different primer pairs in detecting mycobacterium DNA. Applying PCR to a fresh pleural biopsy specimen in this study showed a higher sensitivity when compared to PCR detecting mycobacterium in a paraffin-embedded pleural biopsy specimen by avoiding DNA inhibitors in paraffin-embedded tissues. (29) In our study, the sensitivity of a PCR of a pleural biopsy specimen to detect M tuberculosis was higher when compared to that of pleural fluid PCR in earlier studies, which have demonstrated a sensitivity ranging from 11 to 81%. (6-18, 32, 33) Only one study, by Takagi et al (34) addressed the utility of a PCR of a pleural biopsy specimen to detect M tuberculosis with similar sensitivity and specificity to our study. However, the study by Takagi et al (34) used histopathology alone as a "gold standard" because they had no BACTEC culture studies of pleural biopsy specimens. The negative finding of the PCR of pleural biopsy specimens in two cases implies the importance of combining PCR of pleural biopsy specimens with microbiological and histopathologic examination.

In this study, we showed that a PCR of the pleural biopsy specimen has similar sensitivity and specificity to that of a culture of the pleural biopsy specimen, however, the potential advantage of PCR is that it may give faster results. Although PCR provides a rapid result, it has some limitations. It cannot replace conventional culture techniques and must be used in conjunction with culture to ensure optimal isolation rates for test sensitivity and specificity. Also, multiple negative, positive, and inhibition controls must be performed with each assay to ensure quality and reliability. (35) In conclusion, a PCR of pleural biopsy specimens is a useful tool when used in combination with the BACTEC culture system and histopathologic examination of pleural biopsy specimens to reach a rapid diagnosis of tuberculous pleural effusion.

ACKNOWLEDGMENT: The authors thank Dr. Hussein D. Foda for his critical review of the manuscript.

REFERENCES

(1) World Health Organization. TB, a global emergency: WHO report on the TB epidemic; WHO tuberculosis program information. Washington, DC: World Health Organization, 1994

(2) Ogaw K, Koga H, Hirakata Y, et al. Differential diagnosis of tuberculous pleurisy by measurement of cytokine concentration in pleurM fluid. Tuber Lung Dis 1997:78:29-34

(3) Iseman MD. A clinician's guide to tuberculosis. Philadelphia, PA: Lippincott Williams & Wilkins, 2000

(4) Light RW. Pleural disease. 4th ed. Philadelphia, PA: Lippincott William & Wilkins, 2001

(5) Chopin K. Clinical microscopy. 6th ed. Washington, DC: American Society for Microbiology, 1995

(6) Daniel TM. The rapid diagnosis of tuberculosis: a selective review. J Lab Clin Med 1990; 116:277-282

(7) Abe C, Hosojima S, Fukasawa Y, et al. Comparison of MB-Cheek, BACTEC, and egg-based media for recovery of mycobacteria. J Clin Microbiol 1992; 30:878-881

(8) Roberts GD, Goodman NL, Heifets L, et al. Evaluation of the BACTEC radiometric method for recovery of mycobacteria and drug susceptibility testing of Mycobacterium tubercolosis from acid-fast smear-positive specimens. J Clin Microbiol 1983; 18:689-696

(9) Kirihara JM, Hillier SL, Coyle MB. Improved detection times for Mycobacterium avium complex and Mycobacterium tubercolosis with the BACTEC radiometric system. J Clin Microbiol 1985; 22:841-845

(10) Ieven M, Goossens H. Relevance of nucleic acid amplification techniques for diagnosis of respiratory tract infections in the clinical laboratory. Clin Microbiol Rev 1997; 10:242-256

(11) Brisson-Noel A, Aznar C, Chureau C, et al. Diagnosis of tuberculosis by DNA amplification in clinical practice evaluation. Lancet 1991; 338:364-366

(12) Choi YJ, Hu Y, Mahmood A. Clinical significance of a polymerase chain reaction assay for the detection of Mycobacterium tubercolosis. Am J Clin Pathol 1996; 105:200-204

(13) Pietrzak J, Frei R, Senn HP, et al. Comparison of polymerase chain reaction with standard methods in the diagnosis of Mycobacterium tuberculosis infection. Eur J Clin Microbiol Infect Dis 1994; 13:1079-1083

(14) Scarpellini P, Racca S, Cinque P, et al. Nested polymerase chain reaction for diagnosis and monitoring treatment response in AIDS patients with tuberculous meningitis. AIDS 1995; 9:895-900

(15) Ceglelski JP, Devlin BH, Morris AJ, et al. Comparison of PCR, culture, and histopathology for diagnosis of tuberculous pericarditis. J Clin Microbiol 1997; 35:3254-3257

(16) Kolk AH, Schuitema AR, Kuijper S, et al. Detection of Mycobacterium tuberculosis in clinical samples by using polymerase chitin reaction and a nonradioactive detection system. J Clin Microbiol 1992; 30:2567-2575

(17) Querol JM, Minguez J, Garcia-Sanchez E, et al. Rapid diagnosis of pleural tuberculosis by polymerase chain reaction. Am J Respir Crit Care Med 1995; 152:1977-1981

(18) De Lassence A, Lecossier D, Pierre C, et al. Detection of mycobacterial DNA in pleural fluid from patients with tuberculous pleurisy by means of the polymerase chain reaction: comparison of two protncols. Thorax 1992; 47:265-269

(19) Jenkins PA. The laboratory diagnosis of mycobacterial disease. Commun Dis Rep CDR Rep 1992; 2:R101-R103

(20) Neimark H, Ali Baig M, Carleton S. Direct identification and typing of Mycobacterium tuberculosis by PCR. J Clin Microbiol 1996; 34:2454-2459

(21) Valdes L, Alvarez D, Sail Jose E, et al. Tuberculous pleurisy: a study of 254 patients. Arch Intern Med 1998; 158:2017-2021

(22) Seibert AF, Haynes J Jr, Middleton R, et al. Tuberculous pleural effusion: twenty-year experience. Chest 1991; 99:883-886

(23) Conde MB, Loivos AC, Rezende VM, et al. Yield of sputum induction in the diagnosis of pleural tuberculosis. Am J Respir Crit Care Med 2003; 167:723-725

(24) Epstein DM, Kline LB, Albelda SM, et al. Tuberculous pleural effusions. Chest 1987; 91:106-109

(25) Valdes L, Alvarez D, San Jose E, et al. Tuberculous pleurisy: a study of 254 patients. Arch Intern Med 1998; 158:2017-2021

(26) Katiyar SK, Singh RP, Singh KP, et al. Cultivation of Mycobacterium tuberculosis from pleural tissue and its histopathology in suspected cases of tuberculous pleural effusion. Indian J Pathol Microbiol 1997; 40:51-54

(27) Ferret J. Pleural tuberculosis. Eur Respir J 1997; 19:942-947

(28) Salazor LM, Quiroz RH, Bales JL, et al. Diagnostic methods of primary tuberculous pleural effusion in a region with high prevalence of tuberculosis. Rev Invest Clin 1997; 49:452-456

(29) Ruiz-Manzano J, Manterola JM, Gamboa F, et al. Detection of Mycobacterium tuberculosis in paraffin-embedded pleural biopsy specimens by commercial ribosomal RNA trod DNA amplification kits. Chest 2000; 118:648-655

(30) Ozkara HA, Kocagoz T, Ozcelik U, et al. Comparison of three different primer pairs for the detection of Mycobacterium tuberculosis by polymerase chain reaction in paraffin-embedded tissues. Int J Tuberc Lung Dis 1998; 2:451-455

(31) Eisenach KD, Cave MD, Bates JII, et al. Polymerase chain reaction amplification of a repetitive DNA sequence specific for Mycobacterium tuberculosis. J Infect Dis 1990; 161:977-981

(32) Nagesh BS, Sehgal S, Jindal SK, et al. Evaluation of polymerase chain reaction for detection of Mycobacterium tuberculosis in pleural fluid. Chest 2001; 119:1737-1741

(33) Villegas MV, Labrada LA, Saravia NG. Evaluation of polymerase chain reaction, adenosine deaminase, and interferon-[gamma] in pleural fluid for the differential diagnosis of pleural tuberculosis. Chest 2000; 118:1355-1364

(34) Takagi N, Hasegawa Y, Ichiyama S, et al. Polymerase chain reaction of pleural biopsy specimens fi)r rapid diagnosis of tuberculous pleuritis. Int J Tuberc Lung Dis 1998; 2:338-341

(35) Caws M, Drobniewski FA. Molecular techniques in the diagnosis of Mycobacterium tuberculosis mad the detection of drug resistance. Ann N Y Acad Sci 2001; 953:138-145

* From the Departments of Thoracic Medicine (Drs. Hasaneen, Shalaby, and EI-Morsi) and Clinical Pathology (Dr. Zaki), Mansoura University Hospital, Mansoura, Egypt.

This study was supported by a grant from Egyptian Ministry of Higher Education and Mansura Faculty of Medicine, Mansura University.

Manuscript received February 6, 2003; revision accepted June 20, 2003.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians e-mail: permissions@chestnet.org).

Correspondence to: Nadia A. Hasaneen, MD, PhD, VA Medical Center, 79 Middleville Rd, Building 62, Northport, NY 11768; e-mail: Nadia.Hasaneen@med.va.gov

COPYRIGHT 2003 American College of Chest Physicians
COPYRIGHT 2004 Gale Group