Background: Extra-pulmonary tuberculosis is difficult to diagnose by conventional methods, because they are less sensitive and more time consuming. Loop-mediated Isothermal Amplification (LAMP) is a novel gene amplification method that has been developed to diagnose Mycobacterium tuberculosis complex in pulmonary and paucibacillary extrapulmonary specimens even in resource poor settings. Aims and Objectives: To evaluate the sensitivity and specificity of LAMP to detect Mycobacterium tuberculosis complex in paucibacillary extra-pulmonary specimens and to compare the results with other conventional methods in extrapulmonary specimens. Materials and Methods: 45 specimens from suspected extrapulmonary tuberculosis patients were collected and tested by LAMP method after DNA extraction. Simultaneously, these specimens were tested by smear microscopy, solid and liquid culture methods. Culture positivity, either in solid or liquid culture, was considered as a confirmed case of extrapulmonary tuberculosis. Results: Sensitivity of LAMP was 95.6% whereas for liquid culture, solid culture and smear microscopy were 69.6%, 65.2% and 17.4% respectively. Specificity for LAMP was 95.4% and for other 3 conventional methods were 100%. Positive predictive values, negative predictive values and likelihood ratios were also evaluated. Turn-around time (TAT) for LAMP was 8 hours only whereas for liquid culture was 2-4 weeks, and for solid culture it was 4-8 weeks. Conclusion: LAMP was a simple, rapid and cost-effective procedure with good sensitivity and specificity. It was found to be better than conventional methods to diagnose extrapulmonary tuberculosis.
Keywords: Extrapulmonary tuberculosis, loop-mediated isothermal amplification (LAMP), mycobacterium tuberculosis, complex
Mycobacterium tuberculosis complex (MTC) causes tuberculosis (TB), an infectious disease and a major global health problem with an estimated 9.6 million new TB cases and 1.5 million TB deaths worldwide in 2014. A simple and affordable diagnostic method to detect both pulmonary and extrapulmonary tuberculosis with good sensitivity and specificity is still lacking. Most laboratories in the world rely on conventional methods like smear microscopy, solid culture and liquid culture for diagnosis of TB even today. But they are less sensitive or lack in quicker turn-around-time (TAT). Smear microscopy is least expensive, technically non-demanding, but it has low sensitivity particularly in extrapulmonary specimens. Solid and liquid cultures have better sensitivity and specificity, but require prolonged time. Moreover, sensitivity for extrapulmonary specimens arepoor.,, Therefore, a rapid and reliable test to diagnose TB, both pulmonary and extra-pulmonary is the need of the hour, particularly after the emergence of HIV epidemic and its co-infection with tuberculosis. Modern and rapid diagnosis of TB by molecular methods like PCR has given promising results in last few years.,,, But these tests are too complex to perform in peripheral resource constrained laboratories around the world.
A novel method has been developed in the year 2000, termed loop-mediated isothermal amplification (LAMP) that amplifies DNA with high specificity, efficiency and rapidity under isothermalconditions., This method employs a DNA polymerase and a set of four specially designed primers that recognize a total of six distinct sequences on the target DNA. It does not require a thermocycler and denaturated DNA template but DNA is amplified at isothermal condition by strand displacement auto-cycling amplification process. Large amount of DNA is generated within an hour and can even be visualised in naked eye after addition of SYBR Green fluorescent dye if Gel doc spectrophotometer is notavailable.,, Hence, rapid diagnosis of pulmonary and extrapulmonary TB can be made possible in resource poor settings.
In our study, we evaluated the sensitivity and specificity of LAMP to detect Mycobacterium tuberculosis complex in paucibacillary, extra-pulmonary specimens and compared the result with other time honoured conventional methods.
After obtaining proper institutional ethical clearance from competent authority and consent from each patient, the study was conducted in a tertiary care hospital of Kolkata, India from December 2013 to November 2014. 45 extrapulmonary specimens from skin punched biopsy (n=25), pus (n=6), lymph node (n=6), endometrial tissue (n=3), pleural fluid (n=2), menstrual fluid (n=1), synovial fluid (n=1) and gastric aspirate (n=1) were randomly collected from clinically suspicious patients for this study.
Tissues and unsterile specimens were first homogenised and decontaminated following standard procedure of NALC-NaOH method. From all these 45 specimens, smears and tissue sections were prepared first, stained with Ziehl-Neelsen (ZN) stain following standard procedure and examined under microscope for acid-fast bacillus. They were also inoculated into Lowenstein-Jensen (LJ) media (Himedia, India) and Middlebrook 7H9 broth (Himedia, India) with added growth supplement and antibiotic mixture (Himedia, India), and were incubated at 37ºC for 6-8 weeks. On LJ media, colonies of MTB appeared as yellowish and buff coloured, dry, rough and raised with wrinkled surface. [Figure 1] On Middlebrook 7H9 broth, within 2-4 weeks growth of MTB complex appeared as clumps, masses or in disbursed manner with increased turbidity. They were further confirmed as MTB after doing ZN staining and biochemical tests for identification following standard procedures. Mycobacterium tuberculosis H37Rv strain procured from Jalma Institute Agra, 1029, Jalma was taken as positive control strain for this study.
LAMP methodology was done using Nu LAMP TB kit, RAS Lifesciences Pvt. Ltd, India, Lot no: LTB0313, following the manufacturer’s instructions. DNA was first extracted from the clinical samples after sample preparation. Fortissue specimen, 10-20 mg small pieces were taken and 300 μl lysis buffer and 20 μl proteinases (all provided with Nu LAMP TB kit) were added and tissues were disrupted thoroughly for several minutes and transferred to 1.5 ml eppendorf and processed for DNA extraction. For body fluid specimen, 1 ml of body fluid were taken in eppendorf , spinned at 6000 rpm for 1 minute, the supernatant were discarded and 300 μl lysis buffer and 20 μl proteinases were added to the pellet and processed for DNA extraction as per literature.
DNA extraction was done using reagents provided by Nu LAMP TB kit and following their literature. The specimens were incubated for 1 hour at 60-65ºC in heating block. After incubation, the homogenate was cooled in ice for 5 minutes. 175 μl protein precipitation solution was added to the homogenate and tube was inverted 10-20 times for uniform mixing. The solution was further cooled in ice for 5 minutes. The tubes were spinned for 10 minutes at 12,000 rpm in microfuge at 4ºC. The clear supernatant was aspirated and removed into a fresh 1.5 ml eppendorf and equal volume of isopropanolol was added followed by inverting the tube for 20-30 times. Again the tubes were spinned for 10 minutes, at 12000 rpm at 4ºC. The supernatant was aspirated and removed carefully without disturbing the white pellet. The tube was re-spinned for 5-10 seconds to collect all residual fluid at bottom of tube. Using a fine bore pipette tip, the residual fluid was carefully aspirated and removed. The pellets were re-suspended in 100 μl tris-edetate buffer, vortexed for 5 seconds, spinned briefly to collect the liquid at bottom of the tube. It was heated for 5 minutes at 55-60ºC with intermittent vortexing to completely dissolve the nucleic acid and the DNA preparation was ready to use for LAMP assay or stored at -20ºC for future use.
For performing LAMP assay, kit components were taken out and placed on -20ºC mini cooler. Except Best Polymerase (Bst) mix, the components were thawed and spinned down when necessary. Working solution of SYBR Green using Tris-Borate EDTA was prepared. Separate reaction tubesor PCR tubes were used for each clinical sample, positive control RAS LTB PC and negative control molecular biology grade water (MBGW) both provided with the kit. Reaction master mix was prepared and the specific components were dispensed at specific volumes per reaction. For one reaction, 17 μl of TB Bst Mix, 4 μl of MBGW and 4 μl of DNA preparation or RAS LTB PC for positive control or MBGW for negative control were added to make 25 μl of reaction mixture in PCR tubes. These PCR tubes were placed in the thermocycler, incubated at uniform temperature of 65ºC for 60 minutes.
After completion of the run, 1 μl of working SYBR Green solution (a fluorescent ds DNA intercalating dye) was added into each tube. Results were analyzed by visually observing the colour change in black background and in the Gel-Doc Spectrophotometer by UV ray (302 nm). Those with green fluorescence were positive for MTB complex and those remained orange or did not give green fluorescence were considered negative. [Figure 2].
Culture positivity, either in solid or liquid culture, was considered as a confirmed case of extrapulmonary tuberculosis. Out of 45 clinical specimens, 23 were culture positive by either solid or liquid culture or both, of which 8 extrapulmonary samples gave growth positive in both solid and liquid culture, 8 gave growth positive only in liquid culture whereas 7 gave growth positive only in solid culture. So, total 16 extrapulmonary samples gave growth positive in liquid culture and 15 gave growth positive in solid culture. In smear microscopy only 4 were AFB positive. On the contrary, in LAMP assay, 23 were LAMP positive of which one was false positive. [Table 1].
The Sensitivity and Specificity of LAMP to detect Mycobacterium tuberculosis complex in extrapulmonary samples were 95.6% (95% Confidence Interval = 89.8 to 99.9%) and 95.4% (95% Confidence Interval = 89.6 to 99.9%) respectively. The Positive and Negative Predictive Value are 95.6% and 95.4% respectively. Likelihood Ratio +ve (LR+) was 20.7 and Likelihood Ratio –ve (LR–) was 0.04. [Table 1].
The Sensitivity and Specificity of Liquid culture to detect Mycobacterium tuberculosis complex in extrapulmonary samples were 69.6% (95% Confidence Interval = 50 to 89.2%) and 100% (95% Confidence Interval = 89.6 to 99.9%) respectively. The Positive and Negative Predictive Value are 100% and 75.9% respectively. Likelihood Ratio –ve (LR–) was 0.30. [Table 1].
The Sensitivity and Specificity of Solid culture to detect Mycobacterium tuberculosis complex in extrapulmonary samples were 65.2% (95% Confidence Interval = 45.6 to 84.8%) and 100% (95% Confidence Interval = 89.6 to 99.9%) respectively. The Positive and Negative Predictive Value are 100% and 73.3% respectively. Likelihood Ratio –ve (LR–) was 0.35. [Table 1].
The Sensitivity and Specificity of Smear Microscopy to detect Mycobacterium tuberculosis complex in extrapulmonary samples were 17.4% (95% Confidence Interval = 1 to 54%) and 100% (95% Confidence Interval = 89.6 to 99.9%) respectively. The Positive and Negative Predictive Value are 100% and 53.6% respectively. Likelihood Ratio –ve (LR–) was 0.44. [Table 1].
Turn-around time (TAT) for LAMP was 8 hours only whereas for liquid culture was 2-4 weeks, and for solid culture it was 4-8 weeks.
Diagnosis of extrapulmonary TB without microbiological confirmation may result in over-diagnosis. And diagnosis of tuberculosis from paucibacillary extrapulmonary specimen is even more difficult. It still remains a challenge, particularly with the existing time-honoured conventional methods, as they lack sensitivity and are time-consuming requiring 4-8 weeks time to achieve maximumsensitivity., Nucleic acid amplification tests (NAAT) allows highly sensitive, specific and rapid detection of MTB complex.,, A study done by Wiener et al. have reported good sensitivity and specificity of NAAT from extrapulmonary specimens, but that did not add weightage in clinical diagnosis and management by the clinicians as they considered clinical diagnosis as gold standard for extrapulmonary TB. A number of NAAT have been employed to detect MTB complex. Among the NAAT, PCR has been most widely used but it requires expensive laboratory set up with sophisticated technical skills. LAMP has been developed with an added advantage of simple technique, possessing isothermal reaction characteristics as well as promising result. But, very few studies of LAMP on extrapulmonary TB has been done till date. In this study, a set of six specific primers targeting 6 different regions of rpo B genes (β Subunit of RNA dependant DNA polymerase gene) of Mycobacterium tuberculosis complex bacteria (M. tuberculosis, M. africanum, M. bovis, M. microti, M. canetti) were selected and MTB-LAMP system was developed by Nu LAMP TB kit, RAS Lifesciences Pvt. Ltd.
Even though culture is considered as gold standard for MTB diagnosis, the sensitivity for liquid culture and solid culture for extrapulmonary samples in our study were 69.6% and 65.2% respectively. Moreover they took 2-8 weeks time. This delay may even affect patient outcome. Our study result showed that the sensitivity of LAMP was 95.6% and specificity was 95.4% for diagnosis of paucibacillary extrapulmonary specimens in comparison with culture and TAT was only 8 hours. Sensitivity and specificity for smear microscopy in our study was only 17.4% and 100% though in other studies like Kiran Chawla et al. sensitivity and specificity of smear microscopy for extrapulmonary samples was 52% and 86%. Smear microscopy becomes less sensitive when bacterial load is less than 10,000 bacilli/ml sample, which is always so in extrapulmonary samples. But smear microscopy has been commonest mode for rapid diagnosis of MTB in resource poor developing countries.
Out of 45 extrapulmonary samples, largest numbers were skin punch biopsy samples 25 and LAMP could diagnose 12 out of 13 culture positive samples proved by liquid or solid cultures, with 92.3% sensitivity in comparison with culture. Most commonly the diagnosis of cutaneous tuberculosis was done either clinically or by histopathological examination.Therefore, LAMP can be an excellent tool for rapid detection of extrapulmonary tuberculosis. Out of 2 culture positive tubercular lymphadenitis cases, LAMP could correctly diagnose 2 cases, but it gave 1 false positive result, which was subsequently proved to be a case of Hodgkin’s lymphoma by immunohistochemistry. Contamination could be a causative factor though utmost care was taken to prevent it. LAMP also showed excellent result with other samples like pus where all 4 were LAMP positive out of 4 culture positive cases. Gastric aspirate, endometrial tissue, pleural fluid and synovial fluid one each culture positive case also became LAMP positive, hence giving 100% sensitivity in comparison with culture. So, overall sensitivity was 95.6% which is quite good.
In our study, we compared different statistical parameters of LAMP and other conventional methods to diagnose extrapulmonary tuberculosis and found that LAMP could be more useful tool in comparison to other conventional methods in this regard. It is a simple, rapid and cost-effective procedure with good sensitivity and specificity to diagnose paucibacillary extrapulmonary tuberculosis.
If compared with PCR, LAMP is much simpler technique and it being isothermal requires only water-bath for attaining 65ºC; whereas PCR requires different temperature settings at different steps requiring complicated thermocycler and more technical expertise and expensive laboratory setup. Also, in LAMP, results can be visualised by using SYBR green dye, which is seen as green fluorescence for positive results, whereas for PCR, UV- transillumination or gel doc is mandatory for result interpretation. In a study by Leila Kohan et al, they have showed that LAMP is more sensitive (100%) than PCR (91.6%) for diagnosing MTB. This LAMP assay with Nu-LAMPTB kit has increased specificity and sensitivity and is economical, rapid with minimal technical expertise and requiring minimal laboratory setup.
We thank Dr. Kalidas Rit, Associate professor, Dept of Microbiology, Institute of Post-Graduate Medical Education and Research and Dr Rupali Dey, Associate professor, Dept of Microbiology, Burdwan Medical College, for their valuable contribution in this study. Dr Manju Banerjee: Director of Institute of Post-Graduate Medical Education and Research and Dr. Pradip K Mitra: Ex- Director of Institute of Post-Graduate Medical Education and Research, for their kind support and permission for carrying out this work.
Financial support and sponsorship
Conflicts of Interest
There are no conflicts of interest.
Source of Support: None, Conflict of Interest: None
[Figure 1], [Figure 2]