Indian Journal of Medical Microbiology Medknow Publications on behalf of Indian Association of Medical Microbiology ISSN: 0255-0857 EISSN: 1998-3646 Vol. 28, Num. 4, 2010, pp. 366-369

Indian Journal of Medical Microbiology, Vol. 28, No. 4, October-December, 2010, pp. 366-369

Brief Communication

Evaluation of BioFM liquid medium for culture of cerebrospinal fluid in tuberculous meningitis to identify Mycobacterium tuberculosis

RS Kashyap¹, SS Ramteke¹, HM Gaherwar¹, PS Deshpande¹, HJ Purohit², GM Taori¹, H Daginawala¹

¹ Biochemistry Research Laboratory, Central India Institute of Medical Sciences, Nagpur - 440 010, India
² Environmental Genomics Unit, National Environmental Engineering Research Institute, Nehru Marg, Nagpur - 440 020, India
Correspondence Address:
H Daginawala, Biochemistry Research Laboratory, Central India Institute of Medical Sciences, Nagpur - 440 010, India
hfd_ciims@rediffmail.com

Date of Submission: 17-May-2010
Date of Acceptance: 10-Aug-2010

Code Number: mb10107

PMID: 20966571
DOI: 10.4103/0255-0857.71820

Abstract

Keywords: BioFM, cerebrospinal fluid, liquid culture, Mycobacterium tuberculosis, tuberculous meningitis

Introduction

Tuberculous meningitis (TBM) is the most common form of central nervous system chronic infections. ^[1] Despite the magnitude of this problem, the general diagnostic outlook is discouraging. Prompt diagnosis of TBM is needed for timely initiation of antituberculous therapy (ATT) thereby preventing secondary complications. ^[2] Acid fast staining (AFB) and Mycobacterium tuberculosis culture are always considered as the gold standards for the confirmation of TBM. The sensitivity of AFB, as reported by different laboratories, is only 5-15% and not up to a level of high standards. ^[3] The solid culture of M. tuberculosis from CSF is still the best confirmatory test for the diagnosis of TBM, but its sensitivity is limited. In addition, this procedure takes at least 5-6 weeks to demonstrate the growth of M. tuberculosis.^[4] To avoid these issues, the Centers for Disease Control and Prevention (CDC) have recommended the use of liquid medium for primary culture. Liquid culture medium has been designed to significantly accelerate the growth of mycobacteria and has been reported in previous studies. ^[5]

BioFM broth is possibly the most useful of the liquid media for TBM diagnosis. It is a nonselective Middlebrook 7H9 medium enriched with OADC (Oleic acid, Albumin, Dextrose, Catalase), and optimised for the growth of mycobacteria, allowing for the isolation of mycobacteria from all types of biological samples. ^[6] In the present study, we have evaluated the sensitivity and specificity of liquid culture medium (BioFM broth) for the manifestation of M. tuberculosis growth used for diagnosis of TBM.

Materials and Methods

A total of 200 patients, including 150 TBM patients, who were admitted to the Neurology Department and in whom diagnostic lumbar puncture was performed, were selected for liquid culture medium analysis. The age of the patients ranged between 15 and 65 years, and there were 118 males and 82 females in the study group. Cerebrospinal fluid (CSF) examination included total and differential cell counts, biochemistry analysis, and microscopic evaluation after Gram, India ink, and AFB staining.

Diagnosis of TBM was based on clinical features including subacute or chronic fever, and signs of meningeal irritation with or without other features of CNS abnormality. CSF findings in these patients included increased protein levels, decreased glucose (CSF/blood glucose ratio: <0.5), and pleocytosis with lymphocyte predominance. In addition, response to antituberculous drugs was also considered as a criterion for the diagnosis. Overall, these patients were considered as "suspected/probable" cases of TBM.

This group included patients with pyogenic, viral, or fungal meningitis. Pyogenic meningitis was suspected in patients who had acute high-grade fever with features of meningitis. These patients often had altered sensoria, as well as CSF findings of increased proteins, very low sugar (CSF/blood glucose ratio: <0.2), and pleocytosis with polymorph nuclear predominance. Response to appropriate antibiotics was also considered as one of the diagnostic criteria for pyogenic meningitis. Viral meningitis was suspected in patients who presented with acute onset of fever and symptoms of meningeal irritation. In such patients, the CSF showed a mild increase in protein, while levels of glucose were often normal (CSF/blood glucose ratio: >0.5), and pleocytosis was predominantly lymphocytic. Those with clinical suspicion of herpes encephalitis were treated with acyclovir, and other viral meningitis patients were treated symptomatically. Fungal meningitis shows CSF profiles similar to TBM. However, India ink staining can show the presence of potential aetiological agents such as Cryptococcus. Fungal meningitis was further confirmed by culturing on selective media. All other patients who had no clinical features of meningitis and had no evidence of CNS or extra-CNS tuberculosis were grouped into the noninfectious/control group. Patient's consent was obtained for all samples that were collected from all study groups for use in this study. The Ethical Committee of the Institute approved the study.

CSF samples (0.5 ml) were inoculated into 5 ml BioFM liquid media (Bio Rad, Marnes-la-Coquette, France), and subsequently incubated at 37΀C. All culture samples were monitored continuously for 6 weeks. The positivity of the culture was defined by the growth of mycobacteria, illustrated by a dark blue to violet colouration in liquid cultures.

All the cultures isolated from liquid medium were centrifuged at 10,000 rpm for 30 minutes. After discarding the supernatant, the resulting pellet was inoculated to nutrient broth (liquid medium) (Hi-Media Laboratories Pvt. Ltd., Mumbai, India). After inoculation, the cultures were incubated and monitored continuously for turbidity (growth).

A loop-full of culture was taken from BioFM liquid medium with a sterilised inoculating needle, and streaking was conducted on nutrient agar plates in a four-way direction near the flame of the burner (Hi-Media Laboratories Pvt. Ltd.). After streaking, the plates were incubated at 37΀C for 24-48 hours. Quantitative data have been analysed by Chi-square test between Bio-FM positive and negative groups in TBM and non-TBM patients. The growth in BioFM broth of non-TBM group at different time intervals was correlated by SPSS software. P < 0.05 was considered as statistically significant.

Results

In the present study, CSF samples from 200 patients were analysed for the culture of mycobacteria for final diagnosis of TBM. Of these, 150 patients were initially diagnosed with TBM. Out of 50 patients diagnosed as non-TBM, 30 were of non-TBM infectious meningitis (pyogenic meningitis = 22 and viral meningitis = 8) and 20 had other CNS disorders.

[Table - 1] shows M. tuberculosis growth in BioFM liquid media from the CSF of TBM and non-TBM patients. Out of 150 TBM patients, 120 showed the presence of mycobacterial growth in BioFM medium. In the non-TBM group, 32 of 50 cases were found to be positive while the remaining 18 cases were found to be negative for BioFM liquid media growth.

[Table - 2] shows the details of growth occurrences in 32 non-TBM cases which were positive for BioFM liquid culture growth. In 23 CSF samples, growth was observed within 24-48 hours, in 6 CSF samples growth was observed within 48-96 hours with an unusual appearance and in the remaining 3 CSF samples growth was noted within 4th to 6th week of incubation.

[Figure - 1]a shows the appearance of M. tuberculosis growth in BioFM liquid culture from CSF samples obtained from both TBM and non-TBM patients. The growth appeared after 3 weeks of incubation in the majority of cases in TBM patients. However, in majority of the non-TBM cases, growth was noted within 1 week of incubation. All positive cultures in liquid media within 1st-6th week were inoculated in nutrient broth (liquid medium) to check mycobacterial or other non-specific growth [Figure - 1]b. Our results show that cultures that appeared positive within a week in BioFM broth showed turbidity (growth) in the nutrient broth. However, cultures that were positive within 3rd-4th week in BioFM media indicated a lack of bacterial growth when inoculated in nutrient broth. The results were also confirmed by inoculating the same cultures onto nutrient agar (solid medium) for the presence of bacterial growth. Bacterial growth in nutrient agar (solid medium) was noted in all the cultures that were positive within 1 week; however, growth was not noted from cultures in which growth was noted within 4-6 weeks, in BioFM broth [Figure - 2].

Discussion

A positive culture of M. tuberculosis from solid or liquid media is regarded as the "gold standard" for the diagnosis of TBM. ^[7],[8] Solid media culture is very laborious and takes 5-6 weeks for completion. ^[9] However, the liquid culture system allows for the rapid growth and detection of M. tuberculosis. The average time for detection in liquid medium is 8-15 days as compared to the previously mentioned 4-5 weeks with traditional solid culture medium. ^[10] BioFM broth is widely used as the liquid media of choice in numerous diagnostic laboratories. Earlier studies have reported that the use of BioFM broth has been shown to reduce the time of mycobacterium isolation to about 10-20 days (mean detection time = 18.4), which is significantly less than the approximate time it takes using solid media (Lowenstein-Jensen). ^[11],[12]

In the present study, we tested the sensitivity and specificity of liquid culture medium (BioFM broth) for the detection of mycobacterial growth as an indicator of tuberculous infection of the central nervous system. Out of 200, a total of 150 CSF samples were of TBM patients and the remaining were from non-TBM patients. In all, 120/150 TBM cases were positive in BioFM broth with a sensitivity of 80% and an average detection time of 2-3 weeks.

The intriguing results from this study are that 32 (64%) of non-TBM subjects were also found to be culture positive in BioFM broth with an average detection time of 3-4 days. Since these unexpected positive results had apparent growth within only 4 days, compared to the 2-3 weeks period that was noted in positive TBM cases, it became a matter of further examination and confirmation of the liquid culture method. After investigation by Ziehl-Neelsen staining for AFB, we confirmed that the growth found in the case of non-TBM was non-AFB in origin (personal observation). We further verified our findings by sub-culturing the growth from BioFM broth on nutrient agar. Bacterial growth was found in non-TBM cases on solid media also but no growth was observed when culture of TBM was transferred to nutrient agar. BioFM broth is a nonselective Middlebrook 7H9 medium optimised for the growth of mycobacteria. BioFM V.C.A., which is a selective supplement, (vancomycin, colistin, amphotericin B) is added to broth to inhibit the growth of the flora associated with the main pathological samples (vancomycin for inhibition of gram positive bacteria, colistin for inhibiting the growth of gram negative bacteria and amphotericin to inhibit fungal growth). Therefore, the antibiotics collectively work to prevent the growth of other non-M. tuberculosis or environmental bacteria. But our results show that the broth also allows the growth of non-tuberculosis bacteria leading to a decrease in specificity of BioFM broth. Although it is very difficult to explain the presence of nonspecific growth in liquid culture medium, it can be hypothesised that may be the antibiotics are not present in optimum concentration to inhibit nonspecific bacteria or the broth is allowing the growth of some antibiotic resistant bacteria. Though vancomycin is known to inhibit all gram positive bacteria, few vancomycin resistant Enterococcus, Streptococcus, Staphylococcus and gram positive rods have also been reported.

According to our observation, it appears that liquid culture medium alone is not useful for the confirmation of mycobacterial infection. Another important observation of our study is that nearly 25 cases which have shown growth within 4-6 weeks of incubation did not respond to ATT. All of these patients were diagnosed as having TBM based on clinical presentations, CSF analysis and computed tomography (CT) scans. In the above cases, most tubes exhibited a peculiar pattern of growth characterised by a scant number of black-pigmented granules instead of the fine evenly dispersed blue colonies usually observed for mycobacteria. So, there may be the possibility of presence of other non-tuberculous mycobacteria having different growth pattern. ^[13] Therefore, it can be proposed that these patients who do not respond to ATT may belong to either multi drug resistant groups, or have mixed infections. Unfortunately, we do not have any clinical and experimental evidence to confirm this observation.

The findings presented in this current study indicate that even though liquid culture medium shortens the time for the detection of M. tuberculosis positive cultures, is less labour intensive and allows the processing of a large numbers of samples, there are clear limitations to the method. There is a likelihood for contamination by nonspecific bacteria, or also for the coexistence of M. tuberculosis and non-tuberculous meningitis (NTM) bacilli in any specific sample. For these stated reasons, detection of M. tuberculosis growth may become disputable or conversely may lead to the erroneous reporting of results. Therefore, our study suggests that we should not rely on the growth observed in the liquid culture (BioFM) medium to diagnose M. tuberculosis infection in patients. A communication between clinicians and the clinical laboratories is required and strongly recommended to ensure the dependability of the liquid culture technique. It is recommended that all individual tubes be checked showing mycobacterial growth pattern at the end of the incubation period to observe a significant amount of nonspecific growth.^[13]

Acknowledgement

We thank Prashant D Deoras for statistical assistance in the study.

References

Copyright 2010 - Indian Journal of Medical Microbiology

The following images related to this document are available:

Photo images