MATERIALS AND METHODS

In this study, the MgCl₂, dNTP, Taq polymerase, 10X PCR buffer were purchased from Fermentas, Canada; the primers were from TIB MOLBIOL, Germany; the tris-HCl, EDTA, lysozyme, proteinase K and SDS were all from Sigma-Aldrich, USA.

Processing of pleural fluid samples: From 2002 to 2005, 78 pleural fluid samples from patients with tuberculosis exudative effusion were received in the Microbiology Laboratory of Tehran University of Medical Sciences (TUMS). Samples were subjected to decontamination with N-acetyl-L-cisteine-NaOH (NALC) (Bahador et al., 2004). The pellets were resuspended in sterile normal saline solution and used for ZN smear, culture and PCR studies.

Loewenstein-jensen culture of pleural fluid samples: Concentrated pleural fluid specimens (0.1 mL) were inoculated into LJ medium and were kept in an incubator for 8 weeks. Bottles were inspected twice a week for visible colonies. Suspicious growth was subjected to ZN staining. Negative cultures were discarded after 8 weeks. Cultures with growth were left under light for 2 h and examined for yellow pigment to identify photochromogenic species.

DNA extraction: DNA extraction was performed by the SDS/lysozyme method (Bahador et al., 2004).

IS1081-PCR: We designed a novel primer set from IS1081 gene by Primer3 program for detection of Mycobacterium tuberculosis complex. The primers specific for IS-1081 (5`-CGAATCAGTTGTTGCCCAAT-3`, 5`-GTTCTTCGGTGCTGGTCAGT-3`) enclosed a 344-bp fragment located at positions 85 to 428 of the IS-1081 gene (GenBank accession number MBBIS1081). Samples were amplified by PCR in 25 μL reaction mixtures containing: 1 μL dNTP (10 mM), 1 μL of each primer (20 pM), 0.25 μL (500 U) Taq polymerase, 1.5 μL MgCl₂ (25 mM), 2.5 μL of 10X PCR buffer and 2 μL of DNA template.

The amplification parameters included an initial denaturation at 95°C for 5 min followed by 40 cycles each of denaturation at 95°C for 30 sec, annealing at 56°C for 30 sec and extension at 72°C for 45 sec. The extension step in the 40th cycle was held for 5 min before the samples were detected.

rpoB-PCR: The rpoB assay was performed as described previously (Kim et al., 2001).

The amplified products were detected by gel electrophoresis using 1.5% agarose gel with ethidium bromide.

RESULTS

We tried a new set of PCR primers based on the sequence of the IS1081 insertion element of the M. tuberculosis complex and one primer based on the published nucleotide sequence of rpoB in mycobacteria. Figure 1 shows an example of PCR amplification from a patient sample positive for the M. tuberculosis.

As shown in Table 1, 78 pleural tuberculosis cases were confirmed by histopathology of pleural biopsy specimens, supported by laboratory (sputum culture), biochemical markers like ADA, gamma interferon and tumor necrosis factor, radiographic and clinical data. 17 cases were confirmed by culture and 3 were confirmed by microscopy and 14 were AFB smear-negative and culture-positive.

Fig. 1:	1.2% Ethidium bromide-stained gel of PCR products. Lane MW: 100 bp molecular size marker; Lane 2: PCR product of IS1081 gene of the H37Rv; Lane 3: rpoB-PCR product of the patient sample; Lane 4: PCR product of IS1081 gene of the patient sample 5: negative control

Table 1:	Comparison of culture, ZN, rpoB-PCR and IS1081-PCR

The PCR results in Table 1 show that, of 61 patients positive for the IS1081 sequence by the PCR amplification, 3 patients were found positive for acid-fast bacilli using the ZN staining protocol and only 16 were found positive by culture on LJ medium. PCR amplification readily detected the rpoB sequence in 43 out of 78 patient samples. Of 35 patients who were negative using rpoB, 19 patients were found positive for the IS1081 sequence. One of pleural fluid samples had a positive result by rpoB-PCR that was negative for 1081-PCR and was considered to be non-tuberculosis mycobacteria (NTM).

The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the rpoB-PCR, 1081-PCR and ZN tests performed on the pleural fluid samples from patients compared with culture as gold standard are shown in Table 1. The sensitivity of the mycobacterial DNA detection PCR was 94.1% with the rpoB -PCR and 1081-PCR. The specificity of the rpoB-PCR was 55.7 and 26.2% for other PCR method. The PPV was much lower, with values ranging from 26.2% with the 1081-PCR to 100% with the ZN stain. The NPV of the test ranged from 81.3% for ZN stain to 97.1% for rpoB -PCR. When IS1081-PCR test results were compared with the confirmed radiographic, histopathology, biochemical markers and clinical diagnosis of tuberculous pleuritis, the sensitivity, for the IS1081-PCR and rpoB-PCR were 78.2 and 55.1%, respectively.

DISCUSSION

At present, the most reliable method for the diagnosis of tuberculous pleuritis is the identification of MTB in pleural specimens (Ahmad et al., 1998). Rapid diagnosis and treatment of tuberculous pleuritis is crucial in order to reduce morbidity and mortality caused by untreated tuberculosis. Because the differential diagnosis of exudative pleural effusion is broad, the rapid diagnosis of tuberculous pleural effusion would greatly facilitate the management of these patients. The mean detection time of MTB was 28 days for LJ specimen culture, 40 min for ZN stain and 7 h for PCR.

Theoretically, PCR for the MTB genome should be a highly effective and rapid method for diagnosing tuberculous pleuritis. Using different genomic DNA sequences and extraction methods, MTB-PCR studies of pleural fluid from patients with tuberculosis pleuritis have produced highly variable outcomes, with diagnostic sensitivity ranging from 20 to 81% .This probably reflects the importance of experience, technique and sensitivity in the utilization of this test (Neimark et al., 1996; DeWit et al., 1992; De Lassence et al., 1992; Querol et al., 1995; Kuwano et al., 1995).

Most researchers have reported using insertion element IS6110 primers for direct detection of tuberculosis. Villegas et al. (2000) have evaluated PCR directed toward the IS6110 sequence of MTB, reporting that the IS6110-PCR had a false-negative result in two patients with tuberculous effusion. In the studies of Parandaman et al. (2000) and Villegas et al. (2000) there were 22% false-negative results by PCR with IS-6110 primers. The high number of false negatives using PCR with IS6110 may in part be due to the absence of IS6110 copies in MTB. PCR is positive in 100% of the culture-positive TB pleural fluids (Parandaman et al., 2000), but PCR is positive in only 30-60% of culture-negative pleural fluids. The study of Reechaipichitkul et al. (2000) reveals that the sensitivity and specificity of pleural fluid cultures on LJ medium were 17 and 100%, respectively, whereas the sensitivity and specificity of the pleural fluid PCR-assay of the 16 S-23 S rRNA gene spacer sequences was 50 and 61%, respectively.

We compared the performance of the IS1081-PCR and rpoB-PCR assays for the detection of tuberculosis pleuritis. Our study showed that, using the IS1081-PCR assay, 61 of 78 samples (78.2%) were positive. The rpoB-assay had 43 (55.1%) positives of 78 samples (Table 1). The detection rate of the IS1081-PCR and rpoB-PCR assays differed significantly (78.2% versus 55.1%, respectively, p<0.05). The superior detection rate of the IS1081-PCR could be due to multiple copies (five to seven) of IS1081 within the chromosome of all strains belonging to the M. tuberculosis complex. The ZN staining results of pleural fluid specimens were positive in only 3 patients (3.8%) with tuberculous pleural effusion (Table 1). Of all the samples, only 17 (21.7%) led to a confirmed diagnosis of tuberculous pleuritis based on culture results (Table 1). In pleural fluid specimens, the sensitivity of IS1081-PCR test for diagnosing TB in our study was 78.2%, which was similar to that reported by Querol et al. (1995) (81%) but higher than that reported by Shi et al. (1994) (52.9%).

19 IS1081 PCR-positive samples tested negative in the rpoB-PCR assays (Table 1) indicated that the rpoB-PCR assays have lower detection ability and this could be because there is only one of the rpoB genes in the MTB genome compared to the five to seven copies of IS1081. One IS1081-PCR negative patients in our study that was positive by rpoB-PCR from pleural fluid may be due to an NTM infection. NTM organisms do not have IS1081 gene in their chromosome.

Using PCR, the presence of multiple copies of IS1081 in all strains belonging to MTB complex should be easily detectable in tuberculous patients. The utility of rpoB-PCR for the differential identification of mycobacterial species based on PCR restriction-fragment length polymorphism (PCR-RFLP) is its main advantage over other current procedures available for the direct detection of mycobacteria.

The main limitation of our study is that currently there is no diagnostic test to serve as an adequate “gold standard” to evaluate PCR. Since the culture method has low sensitivity, clinical assessment may be subjective.

In conclusion, the results of the present study show that the PCR assay can successfully be used to detect M. tuberculosis DNA in pleural fluid samples for a more rapid, specific and reliable tuberculous pleuritis diagnosis than the LJ medium culture and ZN staining methods. In this study we demonstrated that the IS1081-PCR assay is more sensitive, specific and reliable than the rpoB-PCR assay.