T?berk?lozis ?nfeksiyonu Tan?s?nda Bron? Lavaj?n?n PCR ile
De?erlendirilmesinin Klinik Kullan?m?
HASANO?LU*, Zeki YILDIRIM*, Nurhan K?KSAL*, M?nire G?KIRMAK*, Selami G?NAL**,
We evaluated clinical utility of polymerase chain reaction (PCR) in bronchial washings of? patients with suspected tuberculosis infection. The study included direct microscopical examination, culture and PCR analysis of bronchial washing and sputum samples of 56 patients whose diferential diagnosis included tuberculosis. Fiberoptic bronchoscopy was performed to all patients. Bronchial washings were aspirated after giving 20 ml of 0.9% NaCl solution to the suspicious segment or lobe. The aspirated fluid was used for direct microscopical, PCR and culture examinations. The 123-bp fragment in the IS6110 insertion sequence of Mycobacterium tuberculosis complex was used as target in PCR procedure. According to the results of this study, there were 6 true positives, 6 false positives and no false negatives for PCR. Sensitivity and specificity of the bronchial washings PCR was found as 100% and 88%, respectively. These results suggested that bronchial washings PCR examination can be a useful method for exclusion of the diagnosis of tuberculosis.
Key Words: Polymerase chain reaction, Mycobacterium tuberculosis, bronchial washings
Clinical Utility of Bronchial Washings Polymerase Chain Reaction Evaluation in the Diagnosis of Mycobacterium tuberculosis Infection
Bu ?al??ma t?berk?loz ??phesi olan vakalarda, bron? lavaj? polimeraz zincir reaksiyonu (PCR) incelemesinin klinik kullan?mdaki de?erini ara?t?rmak i?in yap?ld?. ?al??ma ay?r?c? tan?lar?nda t?berk?lozun bulundu?u 56 hastan?n balgam ve bron? lavaj? ?rneklerinin direkt mikroskopik, k?lt?r ve PCR incelemelerinin sonu?lar?n? i?ermektedir. Hastalar?n t?m?ne bronkoskopi yap?ld?. Bron? lavaj? ??pheli lob veya segment bron?una 20 ml NaCl %0.9 sol?syonu verilip aspire edildi. Al?nan s?v? direkt mikroskopik inceleme, k?lt?r ve PCR incelemelerinde kullan?ld?. PCR i?leminde hedef olarak Mycobacterium tuberculosis complex IS6110 insersiyon dizili?indeki 123-bp fragman? kullan?ld?. ?al??mam?z?n sonu?lar?na g?re 6 hasta do?ru pozitif, 6 hasta yanl?? pozitif olarak kabul edildi, yanl?? negatiflik ise saptanmad?. Bron? lavaj? PCR incelemesinin duyarl?l??? %100, ?zg?ll??? ise %88 olarak bulundu. Bu bulgularla, bron? lavaj? PCR incelemesinin en az?ndan t?berk?loz tan?s?n? ekarte etmede kullan?labilece?i d???n?ld?.
Anahtar Kelimeler: Polimeraz zincir reaksiyonu, Mycobacterium tuberculosis, bron? lavaj?
Tuberculosis is currently an important health problem throughout the world mostly associated with human immunodefficiency virus infection in developed countries. It is also a common health problem in Turkey.
Although a presumptive diagnosis of tuberculosis can be made by the patients? history, clinical and radiological findings, the isolation of M. tuberculosis is required for the definitive diagnosis. At present microscopical examination is the rapid diagnostic method available, but it has a low sensitivity. Culture true new radiometric systems and biphase culture may require more than 2 weeks to confirm the diagnosis. Research in serology is not widely used because of high costs and low specificity and sensitivity (1-3). Therefore development of a more rapid method with high sensitivity and specificity for the diagnosis of tuberculosis is highly desirable.
The detection of M. tuberculosis by enzymatic amplification has been found useful in the diagnosis of the pulmonary, pleural and meningeal tuberculosis. Many investigators have reported different rates for sensitivity and specificity of polymerase chain reaction (PCR) for the detection of M. tuberculosis in respiratory specimens (4-10). Several PCR procedures have been described to detect M. tuberculosis genome, the target sequence being the main difference due to the influence on specificity and sensitivity (11,12). The insertion sequence IS6110, specific for M. tuberculosis and repeated several times in the chromosome is the most frequently used target for diagnosis and epidemiologic purposes (3,13).
There are several PCR studies using sputum as a sample, however there isn?t any with bronchial washings or bronchoalveolar lavage fluids. Therefore, we have studied the utility of PCR in bronchial washings of patients whose differential diagnosis included tuberculosis with no typical radiological and clinical signs of disease. The results of PCR were compared with the classic microbiological methods and evaluated considering the follow up of patients clinically and radiologically.
PATIENTS and METHODS
A total of 56 patients whose differential diagnosis included tuberculosis participated in the study. None of the patients had typical radiological or clinical findings of tuberculosis, rather the disease had to be ruled out. Mean age of the patients was 52.54?16.03 years ranging between 14 and 75. Thirty-eight of the patients were male and 18 were female. A pulmonologist has taken the histories and completed the physical examinations of all the patients. Tuberculosis history and smoking or alcohol habits were asked. Chest X-rays were performed in all patients and 34 cases were evaluated by computed tomography (CT). Sputum smear and cultures for acid fast bacilli (AFB) were performed in all patients for at least 3 times. The samples of sputum were also analysed with PCR. Fiberoptic bronchoscopy was performed in 56 patients and bronchial washings were taken from the suspicious segment or lobe, and from the major bronchi in patients with normal radiologic findings. Direct microscopical examination, culture and PCR analysis of the fluid were done.
Clinical samples and microbiological studies
A total of 224 specimens including sputum (n: 56 x 3) and bronchial washings (n: 56) were examined. Sputum and bronchial washing samples were digested and decontaminated by treating with an equal volume of N-acetyl-L-cystein (NALC)-2% NaOH. The specimens treated with NALC-NaOH were centrifuged, the sediment was washed with 0.067 M phosphate buffer (pH 6.8) and resuspended in 2 ml of buffer. This suspension was used to prepare smear, inoculate on Lowenstein-Jensen and for DNA preparation.
Bacterial lysis and DNA extraction
We used the method described previously with some modifications (8,12). Approximately 1 ml of sediment was transferred into a microcentrifuge tube and centrifuged at 13,000 x g for 5 min. at room temperature. The supernatant was carefully removed by leaving approximately 100 ml supernatant over the pellet. After vortex, the tube was heated at 80˚C for 20 min. to kill mycobacteria. To this tube, 0.5 ml of TE buffer (10 mM Tris-HCl pH 8.3, 1 mM EDTA) containing 20 mg/ml lysozyme was added and tube was incubated in a water bath at 37˚C for 2 hours. Subsequently, 100 ml of an equal mixture of sodium dodecyl sulphate (2%) and 1N NaOH were added and mixed. The tube was then placed on boiling water bath for 5 min. After the sample was cooled to room temperature, it was neutralized with 50 ml of the 1M HCl-1M Tris-HCl. DNA was extracted by treatment twice with phenol and once with phenol/chloroform, and precipitated with absolute ethanol by centrifugation, and washed with 70% ethanol. After the pellet was dried at room temperature, DNA was suspended in 20 ml of water for cell culture grade (Sigma), and 5 ml of the DNA suspension was used for amplification.
The PCR was performed using 5 ml of the DNA samples (extracted from the specimens) in 50 ml of reaction mixture containing 2 U of Thermus aquaticus DNA polymerase (Panotaq, Panaroma, USA), 1 mM MgCl2, 20 pmol (each) specific primer (T4: 5?- CCT GCG AGC GTA GGC GTC GG - 3?, T5: 5?- CTC GTC CAG CGC CGC TTC GG-3?) for amplification of 123-bp fragment in the IS6110 insertion sequence of M. tuberculosis complex chromosome (), 0.2 mM each deoxynucleoside triphosphate (Stratagene CA, USA), 1% (wt/vol.) Triton X-100, 0.1 mg/ml bovine serum albumin, 10 mM Tris-HCl, and 50 mM KCl. After denaturation of the template DNA at 94˚C for 5 min, the PCR amplification was performed for 35 cycles using a Deltacycler II system (Ericomp Inc. USA). A cycle consisted of denaturation at 94˚C for 2 min, primer annealing at 55˚C for 2 min, and primer extension at 72˚C for 3 min. Amplification conditions described by Eisenach et al. were used with some modification.
DNA obtained from M. tuberculosis H37Rv was used as a positive control and water for cell culture grade (Sigma) as a negative control.
The amplified product was diluted within bromphenol blue indicator by 3/4 ratio and 6 ml of this sample were electrophoressed in 2% agarose gel containing ethidium bromide. The 123-bp product was visualized under UV light transilluminator. QX174-DNA digested with Hae III, was used as molecular weight marker.
Twelve of 56 patients had PCR positive bronchial washings for M. tuberculosis. Only 6 of them were considered to have tuberculosis with clinical, radiological and laboratory findings. In 5 of these 6 patients, one or more samples submitted for culture were positive for M. tuberculosis. Two of these 6 patients? sputa were positive for mycobacteria by acid-fast staining (Table 1). The cases were also followed up for at least 6 months and all of them had resolution of symptoms and three of them had clearing of chest infiltrates while receiving anti-tuberculous therapy. A patient (number 1) who do not have positive cultures for sputum or bronchial washings but have positive PCR of bronchial washing specimen were diagnosed as tuberculosis, since he had radiological and clinical improvement while taking antituberculous therapy. Clinical, radiological, bronchoscopic findings and final diagnosis of those patients are shown in Table 2.
The other 6 patients whose bronchial washings PCR were positive for M. tuberculosis, but all specimens submitted for culture were negative, were considered as false positive (Table 1). The final diagnosis of those patients are seen in (Table 3 ). Two of the patients were diagnosed as lung cancer. One of them was operated and diagnosed as adenoid cystic malformation. One patient?s diagnosis was interstitial lung disease. One of the patients in this group was decided to have a sequela of pulmonary infection in his chest radiograph. The last one was decided to have idiopathic hemoptysis. These patients were also followed up for 6 months and no evidence of active tuberculosis has been observed.
The final diagnosis of 44 patients whose bronchial washing and sputum PCR examinations were negative for M. tuberculosis are shown in Table 4. None of the cultures from these patients were positive for M. tuberculosis. Four patients had been previously treated for culture proven tuberculosis.
According to the results of this study, there were 6 true and 6 false positives for bronchial washings PCR. Results of PCR performed on bronchial washings aspirated from the 44 patients of whom the diagnosis of the tuberculosis was subsequently excluded, were all negative. Therefore, sensitivity of the bronchial lavage fluid PCR evaluation for M. tuberculosis was found to be 100% and its specificity was 88%.
The polymerase chain reaction has been extensively studied as a rapid assay for detecting tuberculosis in clinical samples like sputum, urine, pleural fluid and cerebrospinal fluid. Obviously, PCR is a powerful tool with a potential to amplify specific segments of DNA from every organism and, studies for M. tuberculosis has been well established (3,14-16). The most studied target for diagnosis has been the repetitive DNA insertion sequence IS6110 which has been shown to be highly specific for M. tuberculosis and also highly sensitive and specific for diagnosis of pulmonary tuberculosis (14, 15).
In many countries, sputum microscopy remains the first step of the diagnosis of pulmonary tuberculosis. This test is not so sensitive, since a positive result requires a minimum bacterial concentration of approximately 104/ml. It is not specific since dead bacilli, atypical mycobacteria and other acid fast organisms may be seen (17,18). Sixty percent of notified cases of postprimary pulmonary tuberculosis in Edinburg, in 1985 were bacteriologically positive. Mean of the cases of pulmonary tuberculosis bacteriologically confirmed in Europe, in 1983 was 52% (19). We had positive sputum PCR results in 4 patients with active tuberculosis although 2 of them had AFB negative sputum smear. So we have smear positivity for M. tuberculosis at a rate of 33.3% (2/6) in patients with active tuberculosis. Bronchial washings PCR for M. tuberculosis were positive in all 6 patients with active tuberculosis while smear of bronchial washings were negative for AFB in 3 of these patients.
Cultures even with radiometric methods, still require weeks and also culture examinations for M. tuberculosis doesn?t have high sensitivity (20). The sensitivity of culture of both specimens, sputum and/or bronchial washings was found as 83.3% in our study.
Bronchial washing examinations are more sensitive than sputum examinations since the sputum specimen may be little more than saliva and even if sputum, it may come from the wrong lung unit. Bronchoscopy requires special expertise and equipment but specimen origin is certain and the yield is superior to that of spontaneous sputum microscopy (18).
In our study, PCR evaluation of M. tuberculosis in bronchial washings confirmed high sensitivity (100%) and specificity (88%). Although all known precautions were taken, there were 6 false positive PCR results in bronchial washings. The false positivity was reported by the other investigators also. PCR contamination is a formidable problem and evidence of PCR carry-over contamination can not be tolerated in a diagnostic laboratory. So decontamination methods must be applied not to cause false positive PCR results (21). Since PCR method has high sensitivity, in the cases of treated tuberculosis, primary infection, noncavitary tuberculosis, sequela of tuberculosis infection, the assay may be positive while the cultures can be negative. It will be still uncertain if these results should be classified as true positives in terms of the presence of tuberculosis DNA or false positives in terms of active tuberculosis (22-24). Our 6 cases were decided as false positives in this study, since all the cultures were negative and final diagnosis of the patients were other pulmonary diseases. These patients were followed up for 6 months and no evidence for active tuberculosis has been observed. The true positivity of the 6 cases had been decided by sputum and bronchial lavage microscopical and culture examinations, besides the clinical and radiological evidences of tuberculosis. Also improvement of the cases by antituberculous treatment confirmed the diagnosis.
Table 5 shows sensitivity and specificity of PCR for the detection of M. tuberculosis using culture positivity for the calculation of sensitivity, specificity, positive and negative predictive values, in different studies (21). When the culture positiveness is not the only criteria for true positiveness, these results will obviously change.
Accurate and definitive diagnosis of tuberculosis by PCR would reduce the numbers of patients who are treated on the basis of clinical suspicion and exposed to these potentially toxic drugs. Especially the results of bronchial washings PCR detection from the suspicious segment of the lung should be more sensitive than the sputum examinations. According to the results of our study, when the PCR detection for M. tuberculosis in bronchial washings is negative, diagnosis will not be tuberculosis. However, sputum and bronchial washings PCR examinations, like all other examinations must be considered in the total clinical context. Tuberculosis can be diagnosed with PCR technique within 4-8 hours of receipt of the sample in the suspicious cases with radiological and clinical evidences. We hope refinements of the PCR technique and reduction in the costs will allow more extensive application of PCR in the countries that tuberculosis is still a public health problem and this will at least lead to rapid exclusion of the diagnosis of tuberculosis.
ADDRESS FOR CORRESPONDENCE:
Dr. H. Canan HASANO?LU
?n?n? ?niversitesi Turgut ?zal T?p Merkezi
G???s Hastal?klar? Anabilim Dal?