Background: Tuberculous meningitis (TBM) is a major, global public health problem and its diagnosis remains problematic even after years of experience. Early diagnosis and treatment are of paramount importance as a delay in diagnosis can lead to irreversible central nervous system damage and mortality. Aims: To evaluate the role of cerebrospinal fluid (CSF) IS6110-based tuberculosis (TB) polymerase chain reaction analysis (PCR) as an efficient diagnostic tool for the diagnosis of TBM. Materials and Methods: This is a cross-sectional study carried out with patients clinically suspected to have TBM. A total of 102 cases were enrolled, out of which 15 patients were excluded from the study because of alternative diagnoses. An in-house IS6110-PCR method using a specific pair of primers designed to amplify the insertion sequence, IS6110 in the Mycobacterium tuberculosis genome was used to analyze CSF. Results: Our study showed PCR positivity in 51.7% cases. PCR gave a sensitivity of 100% and specificity of 68.9% against the cases of TBM confirmed by culture. Conclusion: IS6110 TB-PCR is a novel diagnostic tool that can diagnose a greater number of individuals with clinically suspected TBM with its rapidity, accuracy, and reliability, and its positivity even after 4 weeks of starting treatment makes it the diagnostic tool of choice.
Keywords: Cerebrospinal fluid (CSF), diagnosis, polymerase chain reaction (PCR), tuberculous meningitis (TBM)
Tuberculosis (TB) continues to haunt the human race as a medical malady with a tragic consequences in the social and economic realms. At the dawn of new millennium we are still mute witness to its myriad complications and vicious killing powers. Extrapulmonary TB refers to disease outside the lungs. In developed countries, 10-15% of tuberculosis cases have extrapulmonary involvement, but in patients from high-incidence countries the rate is much higher. People who are human immunodeficiency virus (HIV)-positive and infected with TB develop extrapulmonary diseases much more frequently, in up to 50% of cases.  Neurological TB accounts for 5-10% of the cases of extrapulmonary TB  and occurs more frequently in children. Among extrapulmonary types of TB, tuberculous meningitis (TBM) leads to multiple complications in the central nervous system (CNS) and continues to be a major health problem in underdeveloped and developing countries.  CNS TB is the severest form of TB infection, causing death or severe neurological defects in more than half of those affected, in spite of recent advancements in antituberculosis treatment (ATT). In addition, owing to an increasing number of immunocompromised hosts caused by the prevalence of acquired immunodeficiency syndrome (AIDS), the wider use of immunosuppressive agents, and other immunocompromised states, TBM remains a serious clinical and social problem. Owing to the wide spectrum of its neurological symptoms, CNS TB remains a formidable diagnostic challenge. ,,,
The diagnosis of TB is mainly based on the clinical presentation, histopathology, demonstration of acid-fast bacilli (AFB) in smears, and isolation of Mycobacterium tuberculosis from culture. These diagnostic criteria have limitations that include atypical clinical presentations of disease and the poor sensitivity and specificity of AFB microscopy, particularly with paucibacillary specimens. Culture for M. tuberculosis usually takes 4-6 weeks to grow on solid media, delaying the time to result. The diagnosis of extra-pulmonary TB is particularly difficult to establish, especially in developing countries because of the paucibacillary nature of extrapulmonary specimens and because the signs and symptoms of disease can be nonspecific. , The diagnosis of TBM is also difficult as the yield from AFB staining and culture can be very low, probably because of the low concentration of bacilli in the CSF.
Molecular methods such as polymerase chain reaction (PCR)-based diagnostic techniques have begun to be applied in the local setting, for better and easy assessment of tubercular load in the CNS infected by TB. PCR-based assays offer high sensitivity by the amplification of a small amount of DNA, and has been extensively evaluated for the detection of M. tuberculosis from clinical samples.  Many of the tests described in the literature are based on the amplification of IS6110, an insertion element that is believed to be restricted to members of the M. tuberculosis complex. ,, The presence of multiple copies of this element in the majority of M. tuberculosis strains undoubtedly enhances the sensitivity of PCR.
This study aimed to evaluate cerebrospinal fluid IS6110 PCR as a diagnostic tool for the diagnosis of TBM with comparison to other diagnostic tests. Additionally, we tried to find out the correlation of molecular methods such as PCR to the cytological findings of CSF and radiological findings of TBM to ascertain the actual competence of this test and/or superiority over other tests.
The study was conducted in a tertiary care hospital of North India. The patients enrolled for the study were >12 years of age with symptoms and signs of meningitis including one or more of the following for >5 days: Fever, persistent headache, irritability, neck stiffness, vomiting, convulsions, focal neurological deficit, and altered sensorium or lethargy). After investigations, if the patient was found to have alternative diagnosis they were excluded from the study (e.g., pyogenic meningitis, cryptococcal meningitis, viral meningoencephalitis, cerebral malaria, or bacterial brain abscess).
The diagnosis of TBM was made based on the clinical characteristics, laboratory findings including the CSF examination, and radiological characteristics. The diagnosis of TBM was established using a standardized clinical case definition.  A scoring system was used as mentioned in [Table 1]. Then TBM cases were classified into the following categories:
Probable TBM :
Diagnostic score of ≥12 when imaging is available and ≥10 when imaging is not available.
Possible TBM :
Diagnostic score of 6-11 when imaging is available and 6-9 when imaging is not available.
All cases were classified as definitive, probable, possible, or non-TB meningitis, depending on their total diagnostic score. The cases were clinically evaluated and neurologically staged during admission to the hospital based on the British Medical Research Council criteria.  Accordingly, patients with mild cases featuring nonspecific symptoms who were conscious and did not have neurologic deficits were considered stage I [Glasgow Coma Scale (GCS)-15]; cases with mild alterations in consciousness and minor neurologic deficits such as cranial nerve palsies were considered stage II (GCS 11-14); while cases with major neurologic deficits such as paresis/plegia, cases with convulsions, and cases in a precoma/coma state were considered stage III (GCS ≤10).
Complete blood counts, serum Na + , serum K + , blood urea, serum creatinine, and liver function tests were done in all the patients. Computed tomography (CT) scan head and magnetic resonance imaging (MRI) brain were done. CSF analysis was done for total leukocyte count (TLC), differential leukocyte count (DLC), total protein, sugar, Gram staining, AFB staining, adenosine deaminase (ADA) levels, IS6110 TB-PCR, and Mycobacterium tuberculosis culture/sensitivity.
The pellet was washed in sterilized, distilled water by centrifugation 3000 rpm × 5 min. Following this, it was suspended in 200 μL of TE buffer and 200 μL of lysis buffer, boiled for 10 min at 100°C,; then 10 μL of proteinase K was added and kept at 56°C for 2-3 h. To inactivate proteinase K, after removal of the eppendorf (microcentrifuge tube-Sigma Aldrich India) from the water bath it was boiled for 10 min at 100°C and centrifuged at 6 × 10 3 rpm for 5 min. Then 400 μL of the supernatant was pipetted in the sterile eppendorf, after which the purification step followed. For purification, equal volumes of phenol-chloroform (24:1) were added and vortexed, followed by centrifugation at 6 × 10 3 rpm for 5 min. The aqueous phase was carefully transferred to another sterile eppendorf without disturbing the middle layer. An equal volume of chloroform was added, vortexed to remove traces of phenol, and centifuged at 10 × 10 3 rpm for 5 min. The aqueous phase was again transferred to a new eppendorf and 2.5 volume of chilled 100% ethanol and 10 μL sodium acetate (0.3 M final concentration) were added. The tubes were kept at −20°C overnight for precipitation of DNA. The next day after the samples were centrifuged at 10 × 10 3 rpm for 10 min, the supernatant was decanted and the DNA pellet was washed with 70% ethanol. The tubes were gently rotated between the palms for the decantation. The DNA pellet was dried for 1-2 h at room temperature to remove ethanol. DNA samples were suspended in 25 μL of triple-distilled water and stored at −20°C for PCR analysis.
Amplification by conventional PCR
A pair of M. tuberculosis oligonucleotide primers based on IS6110 (synthesized by Genei, Bengaluru, KA, India) was used. The amplified product size of IS6110 is 123 base pairs. The primers’ sequences are as follows. forward primer: 5′CCTGCGAGCGTAGGCGTCGG3′ and reverse primer: 5′CTCGTCCAGCGCCGCTTCGG3′. Amplification reaction was done in a final volume of 25 μL. 5 μL of extracted DNA was added to 20 μL of the PCR mixture (PCR mixture contained 25 mM MgCl 2 , 20 pmol of each primer, 10X Taq buffer, 10mM (each) deoxynucleotides (dNTPs) and 2.5 unit of Taq DNA polymerase). The reaction cycle was carried out for 35 times in a DNA thermal cycler. M. tuberculosis (H 37 Rv strain) DNA was used as the positive control. The amplification profile was carried out in 3 steps: Initial denaturation was done at 94°C for 1 min followed by 35 cycles each for 1 min. Then annealing was carried out at 63°C for 1 min and primer extension at 70°C for 1 min. Finally extension was done at 72°C for 10 min.
The amplified product was analyzed on 1.2% agar gel containing 0.5 μg/mL ethidium bromide (Sigma Aldrich India) along with a molecular weight marker (methyl-CpG-binding domain protein 13) 10bp DNA ladder, Genei, Bengaluru, KA, India) electrophoretically, and analyzed by a 264 nm wavelength ultraviolet (UV) transilluminator. Gel document system -Alphainnotech (Alphaimager HP Canada) was used for the documentary pictures. If a sharp band of 123bp of amplified DNA was visualized it was considered to be positive.
The cases were put on treatment with classical four-drug ATT [combination of isoniazid (INH), rifampicin (RIF), pyrazinamide (PRZ), and streptomycin] for 18 months along with supportive treatment for raised intracranial pressure, nutrition, intravenous fluids for comatose patients, and antiepileptic treatment for seizures, if present. Advanced-stage cases were also given intravenous dexamethasone therapy for 8 weeks.
The data were entered in Microsoft Excel Computer Program (Microsoft Corporation and Private Limited, Gurgaon, India) and checked for any inconsistency. The results are presented in mean ± standard deviation (SD) and percentages. The chi-square test was used to compare the categorical/dichotomous variables. The Mann-Whitney U nonparametric test is used to compare two continuous variables. The P value <0.05 was considered significant. All the analysis was carried out using SPSS version 15.0 (Statistical Package for the Social Sciences).
The study was not funded by any outside source.
A total of 102 cases of both sexes were enrolled for this study, of which 15 patients were excluded either because of alternative diagnosis (10 patients: 6 pyogenic meningitis, 2 cerebral malaria, 2 cryptococcal meningitis) or due to loss to follow-up (5 patients). In the present study, the patients were in the age group of 13-86 years. The mean age of patients was 33.20 ± 14.99 years. The maximum number of cases (35.6%) was in the 21-30-years age group, followed by 31-40 years (29.9%). The male:female ratio was 1.48:1. The mean duration of illness was 59.22 ± 30.24 days. All the patients (100%) had a history of fever and headache as the main complaint. A history of nausea and vomiting was present in 87.3%, weight loss in 14.94%, and night sweats in 35.63% of patients. Approximately 90% of patients presented in a state of altered sensorium and 16% had a history of convulsions. Twenty-nine (33.3%) patients had a focal neurological deficit and most of them (19 patients) had cranial nerve palsy, the sixth cranial nerve being the most commonly affected, in 14 patients, followed by the third cranial nerve in 5 patients. Nine patients had hemiparesis (right/left) and 1 had quadriparesis. Sixty-nine percent (69%) of the patients were in stage 3, followed by 28.7% and 2.3% in stage 2 and stage 1 respectively. CSF protein values were in the range 24-925 mg%, with a mean value of 207.76 ± 167.29. TLC was in the range 5-990/mL with the mean value of 186.20 ± 64.51. The mean lymphocyte percentage in CSF was 83.98±19.38, ranging 10-100%. ADA levels showed high variability, ranging 1-51U/L with the mean of 10.53 ± 7.56. The range of CSF-sugar level was 3.3-145 mg% with the mean value of 48.53 ± 29.65. The most common abnormality found on imaging studies was basal meningeal enhancement in 51.72% of cases, followed by hydrocephalus (47.12%), infarct (11.49%), and tuberculoma in 6.89%. About one-third of the patients (34.48%) had no abnormality on imaging.
CSF PCR was carried out on the day of admission and before the start of treatment. IS6110 PCR was found to be positive in 51.7% of the patients, and AFB smear in 21.8%. However, positive culture was found only in 18.4% of patients. All AFB smear- and culture-positive patients were positive for PCR and none of the patients had positive AFB smear and/or culture with negative PCR. The sensitivity and specificity of CSF PCR were found to be 100% and 68.9% respectively, against AFB culture as a gold standard [Table 2].
All the patients who presented in stage 1 were found to be positive for PCR, whereas about half of the patients in stage 2 (52%) and stage 3 (50%) were PCR-positive [Table 3]. There was no statistical significance was observed between stage of the disease at presentation and positivity of CSF PCR (P = 0.37).
The CSF protein level was significantly higher in PCR-positive patients (243.53 ± 173.18) compared to negatives (169.4 ± 153) (P < 0.05). Similarly statistically significant observations were found for ADA and TLC. However, the lymphocyte percentage was almost similar between PCR-positive and -negative patients without any statistical significance [Table 4] and [Figure 1].
PCR positivity was significantly associated with hydrocephalus (P = 0.008) and meningeal enhancement (P = 0.04), whereas a statistically significant correlation was not found between PCR positivity and the presence of infarct, tuberculoma, or calcification [Table 5].
On applying diagnostic scoring criteria, about half of the patients (51.7%) were diagnosed as definitive TBM, followed by possible and probable TBM in 36.8% and 11.5%, respectively.
TB remains a major global public health problem, especially in India. The clinical utility of detecting M. tuberculosis by PCR is the reduction in the time to detection it allows and its accuracy in detecting the pathogen in AFB smear-negative paucibacillary specimens. The detection of TBM is difficult to establish because of its pleomorphic clinical presentation and variable CSF cellular content and biochemical parameters, similar to that of partially treated pyogenic meningitis cases. Delayed diagnosis and treatment may be associated with many serious CNS complications.  Hence, rapid detection of M. tuberculosis is of vital importance for the proper diagnosis and management of TBM.
PCR is considered to be one of the most specific diagnostic methods among the many rapid methods studied. In this study, PCR was performed by amplification of the IS6110 insertion sequence, which belongs to the IS3 family and is found in almost all members of the M. tuberculosis complex. Most strains carry 10 to 15 copies, which are present in a wide variety of chromosomal sites.  Previous studies have documented increased positivity using the IS6110 target in extrapulmonary samples.  Although IS6110 PCR is not a novel diagnostic tool, it is believed that more studies are required to establish its utility in the diagnosis of TBM. This study was planned to evaluate the efficacy of the IS6110 PCR assay to detect M. tuberculosis DNA in CSF samples collected at our institution.
The hallmark of the PCR assay is that it was capable of detecting the presence of M. tuberculosis DNA in all the samples proven to be TBM using the gold standard of culture. This finding is unique, as the 100% sensitivity of the PCR-based assay could be used as a reliable and rapid method of diagnosing TBM. A similar degree of sensitivity was observed using PCR by other studies also using various clinical samples.  These findings are in contrast to other reports that demonstrated reduced sensitivity in similar settings. More recently, using a set of improved primers belonging to the repetitive sequence IS6110, a few studies were able to demonstrate 100% sensitivity in culture-positive TBM cases.  In our study, the IS6110 PCR assay could detect the presence of mycobacterial DNA in 51.7% of clinical TBM cases, whereas the AFB smear and culture could detect only up to 21.8% and 18.4% respectively. Even though only 51.7% of clinically suspected TBM patients were positive for IS6110 PCR, the value is much higher compared to CSF culture and smear, so that one can confirm TBM with certainty. The positive rate was much higher than in other studies, and this suggests the utility of the assay as an important tool to guide the clinician toward a diagnosis where conventional methods fail.
Rafi et al.  found that the IS6110 uniplex PCR assay had 98% sensitivity [negative prediction values (NPV) 99%] and specificity of 100% [positive prediction values (PPV) 100%] against the gold standard of culture and an overall sensitivity of 76.4% and specificity of 89.2% when clinical TBM was included.
In a Vietnamese study, Nguyen et al.  compared 104 patients treated for TBM on clinical grounds and the results of the initial CSF microscopy, culture, and PCR. He reported that the sensitivity of PCR was 32%, that of culture 17%, and that of microscopy 1%.
While many of those studies demonstrated improved sensitivity over traditional smear and culture methods, several other studies highlighted the low sensitivities associated with these PCR assays. The sensitivity of CSF microscopy and that of culture fall rapidly after the start of treatment, whereas mycobacterial DNA may remain detectable within the CSF until 1 month after the start of treatment. PCR can be performed on CSF for all forms of CNS TB. ,
However, like any diagnostic test for TB, a negative CSF PCR cannot exclude the possibility of tuberculosis, and clinical judgment remains paramount. An additional challenge of using PCR-based methods in the diagnosis of TBM is the requirement of appropriate laboratory infrastructure to perform these more sophisticated methods, which is often lacking in areas where TBM is highly endemic.
To judge the efficacy of PCR as an assessment tool for clinically correlated patients of TBM, we also assessed the relation of PCR to neuroimaging and cytological findings of CSF mostly associated with TBM. We found that CSF proteins, ADA, and TLCs all were well-correlated with PCR positivity. Similarly, neuroimaging with MRI/CT scan hydrocephalus and meningeal enhancement also showed statistically significant correlation with TB-PCR positivity.
Finally, with this study we can summarize, here, that TB-PCR is an important diagnostic tool for diagnosing tuberculous meningitis which presents with wide variety of clinical manifestation with high sensitivity and specificity. Currently PCR is the most rapid diagnostic test for TBM that provides an acceptable sensitivity in comparison to any other method. PCR therefore presents a positive and timely tool aiding the clinical diagnosis of TBM.
In our study, the IS6110 PCR assay increased the detection rate of M. tuberculosis and using it we were able to confirm clinically suspected TBM with better sensitivity and specificity than that of CSF AFB smear and culture. As early recognition and timely treatment are of critical importance in preventing the mortality and morbidity associated with the disease, CSF PCR can be used as the ideal diagnostic test compared to existing tests that can detect TB at an early stage. Results of IS6110 PCR can be obtained as early as within 1 day, and if there is doubt, the test can be repeated and confirmed. Its rapidity, accuracy, reliability, and its positivity even after 4 weeks of starting treatment make it the first test of choice for adaptation in the diagnosis of TBM patients.
Source of Support: None, Conflict of Interest: None
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]