Tanzania Journal of Health Research Health User's Trust Fund (HRUTF) ISSN: 1821-6404 Vol. 11, Num. 2, 2009, pp. 65-69

Tanzania Journal of Health Research, Vol. 11, No. 2, April-June, 2009, pp. 65-69

Original Article

The usefulness of microscopic observation for drug susceptibility of Mycobacterium tuberculosis complex in routine clinical microbiology laboratory

Mshana SE, Imirzalioglu C, Domann E, Chakraborty T

Weil Bugando University College of Health Sciences, P.O. Box 1464, Mwanza, Tanzania and Justus-Liebig University, Institute of Medical Microbiology, Giessen

Correspondence Address:Weil Bugando University College of Health Sciences, P.O. Box 1464, Mwanza, Tanzania and Justus-Liebig University, Institute of Medical Microbiology, Giessen

mshana72@yahoo.com

Date of Submission: 26-Sep-2008

Date of Decision: 27-Mar-2009

Date of Acceptance: 27-Mar-2009

Code Number: th09012

Abstract

Keywords: tuberculosis, drug, susceptibility, Mycobacterium tuberculosis

Introduction

Tuberculosis (TB) is one of the most common infectious diseases in the world with 8 million new cases and 2-3 million deaths annually (WHO, 2006). The treatment of tuberculosis is based on multiple drugs administered for the period of 6-12 months depending on the location of the disease. A cure rate of more than 95% is possible but the effectiveness is compromised when drugs are taken inappropriately due to poor patient compliance (Espinal et al., 2000). Morbidity and mortality are higher in drug resistant cases particularly multi-drug resistance-TB (MDR-TB) (WHO, 2004). Primary drug resistance in the world ranges from 0-16%, acquired resistance rate ranges from 0-48% (Dooley et al. , 1992; Cohn et al ., 1997; WHO, 2006). In most parts of the developing countries the situation of MDR-TB is not known and there is no routine drug susceptibility test available in the existing clinical microbiology laboratories (WHO, 2008).

Advanced laboratories can detect TB cases within 7 to 14 days using liquid culturing systems such as BACTEC or Mycobacterium growth index tube (MGIT). The Use of molecular techniques can shorten this period to hours in a certain amount of cases (Luz et al ., 2000). Most of reference TB laboratories in developing countries use solid media (Lowenstein Jensen media) for isolation and drug susceptibility assay. With this approach detection takes about 3-8 weeks and drug susceptibility testing takes additional 3 - weeks (Madison et al. , 2006). The delay in detection and determination of drug susceptibility results makes control of tuberculosis difficult. A patient with undetected MDR-TB will continue to attend clinic until after 3 months of deteriorating health, the illness threshold is surpassed and drug susceptibility test (DST) is requested. All this time the patient will have exposed others to MDR tuberculosis (David et al ., 2007).

Microscopic observation drug susceptibility (MODS) was first described in 2000. The method is fast, cheap and comparable to gold standard methods (Luz et al ., 2000). The MODS assay for detection of tuberculosis and multidrug-resistant tuberculosis directly from sputum relies on the following facts: Mycobacterium tuberculosis complex species grow faster in liquid media than in solid media, cord formation can be visualized microscopically and incorporation of drugs permits rapid and direct drug susceptibility testing together with detection of bacterial growth (Moore et al ., 2004, 2006; Girum et al., 2007). This method was evaluated in large field studies in 2006 and different countries are currently evaluating the method for applicability in the specific countries (Moore et al 2006, WHO, 2008, Brady et al 2008). This study was conducted to evaluate the usefulness of MODS assay as a routine procedure in Clinical Microbiology Laboratory aiming at promoting this method for developing countries.

Methods and Materials

Bacterial isolation

The study was conducted at Clinical Microbiology Laboratory at Institute of Medical Microbiology Giessen, Germany. Thirty preserved Mycobacterium tuberculosis isolates; recovered 2006 and four smear positive sputum specimens from patients attending Infectious Clinic Unit at University Hospital Giessen were tested using MODS as described previously (Luz et al. , 2000; Moore et al. , 2004). The results were compared with results from Borstel Germany TB Reference Laboratory. Middlebrook 7H9 broth (Becton Dickinson, Sparks, Md.) supplemented with 10% oleic acid, albumin, dextrose ( OADC) and 0.2% glycerol (Sigma) was prepared as per manufacturer instruction: 4.7g of the powder was mixed (7H9 Base powder) in 900 ml of distilled water containing 2ml glycerol, then was autoclaved at 121 ^o C for 10 minutes. Aseptically 100ml of OADC was added and 7.2 ml of 7H9 broth was dispensed into 15 ml sterile tube followed by 160µl of PANTA antibiotic mixture (BD Franklin Lakes, NJ USA 07417) was added to make final concentration of PANTA 20ìl/ml.

Susceptibility assay

Indirect and direct susceptibility assays were performed as recommended in previous studies (Luz et al ., 2000; Girum et al ., 2007). Recommended critical concentrations and one higher concentration were tested for each drug. For indirect assay isolates from Mycobacterium growth index tube (MGIT) were taken and vortexed and left to settle for 20 min., the supernatant was diluted using 7H9 broth to 0.5 McFarland equivalents. The 800µ of diluted sample was added into 7.2 ml of 7H9 broth. For direct susceptibility assay, the pellets of decontaminated sputum were reconstituted with 1ml of 7H9 broth and 800µl of the mixture was added in 7.2 ml 7H9 broth. Using prepared working sheet, 540µl of the mixture (7.2ml of 7H9 and 800µl of sample) was distributed into corresponding 24 wells plates (Sigma). Two wells without antibiotics and the other two wells of different concentration of antibiotics. Stock solutions of antibiotics were made as follows: Isoniazid (INH) 1mg/ml with distilled water, rifampicin (RIF) 1.0mg/ml with methanol, ethambutol (ETH) 2.5mg/ml with distilled water and 2.0mg/ml streptomycin (STR) with distilled water. Solutions were filter sterilized using 0.22µm filters and 1ml aliquots were stored at -20 ^o C. Working solutions were made by diluting stock solutions with sterile distilled water. Concentrations of working solutions were as follows: isoniazid 1 and 4µg /ml, rifampicin 10.0 and 5.0µg/ml, ethambutol 25 and 50µg/ml and streptomycin 20 and 60µg/ml (David et al ., 2007).

In each well 60µl of working solution of antibiotics was added. 60µl of 7H9 broth was applied to the drug free wells. The final concentrations of the examined drugs were 0.1 and 0.4µg/ml INH, 0.5 and 1µg/ml RIF, 2.5 and 5µg/ml ETH and 2 and 6µg/ml STR. Plates were covered and sealed completely with parafilm (Chicago, IL 60631) and incubated at 37 ^o C for up to 40 days. Culture plates were examined under inverted light microscope (Leitz GmbH) at magnification of 32x daily from day 4 to 7 and alternate day from day 9 to 21 and twice weekly from day 26 to 40.

Positive culture was identified by cord formation characteristic of M. tuberculosis [Figure - 1]. Drug free wells were examined first. In case of growth in these wells and absence of growth in drug contained wells, the isolate was considered susceptible. Bacterial and fungal contamination was recognized by rapid overgrowth and clouding in the wells. The first day of visible growth was noted. Plates were examined in subsequent days to confirm susceptibility results. This was done two days after growth in the drug free wells was detected. Two examiners read the plates independently. In discordant cases a third person was used to resolve the difference. The results were compared to results from reference laboratory.

The use of SYBR Green stain (Molecular Probes, USA) to quantify mycobacterial growth was also investigated. Portions (200ìl) of cultures from 10 strains growing at day 0, 4 and 7 in culture broth with and without antibiotics were heat-inactivated before staining with 10ìl of SYBR Green and examined for fluorescence emission (Molecular Devices, USA).

Results

The median turn around time (TAT) was 6 days (range= 4-15 days). The susceptibility results were 100% concordant with reference laboratory for rifampicin, 97% concordant for ethambutol and 94% concordant for isoniazid and streptomycin [Table - 1].

Sybr green showed clear increase in fluorescence intensity in drug resistant wells compared to drug sensitive wells [Figure - 2] and [Figure - 3].

Two isolates were confirmed to be Mycobacterium tuberculosis complex using an inverse hybridization genotyping assay (Hain Lifescience GmbH ) and they had TAT of 6 and 7 days. One which had cord like morphology and overgrowth within 48h was identified as Mycobacterium chelonae. The fourth direct sample was positive after 8 days and exhibited star-like growth. This was identified as Mycobacterium celatum and it was correctly labelled resistant to isoniazid, rifampicin, ethambutol and streptomycin by MODS.

Discussion

The median turn around time (TAT) observed in this study was 6 days. This is slightly lower the TAT (7-9 days) observed in other (Luz et al. , 2000; Moore et al ., 2004). The TAT was comparable to MGIT, and BACTEC which have median turn around time of 14 days from receiving of specimen to susceptibility results. Using Lowenstein Jensen (LJ) slopes the median TAT for indirect assay from receiving specimens to susceptibility results was 3.5 weeks. Two weeks can be gained if LJ proportion method is used to asses drug susceptibility (Luz et al ., 2000).

In four direct samples assessed, TAT was comparable to previous studies (Luz et al ., 2000; Moore et al ., 2004, 2006). Two members of the Mycobacterium tuberculosis complex namely, Mycobacterium chelonae and Mycobacterium celatum were identified in this study. M. celatum was found resistant to isoniazid, rifampicin, ethambutol and streptomycin by MODS (Moore et al ., 2004). The cording growth characteristic has been found to be useful in identifying M. tuberculosis complex. M. chelonae and M. smegmatis also display cord like growth but can be easily differentiated from M. tuberculosis complex by markedly faster growth (Luz et al ., 2000; Moore et al ., 2004). M. chelonae grows fast with visible colonies within 48h (David et al ., 2007).

Generally in this study susceptibility results were 100% concordant to results from TB reference laboratory for rifampicin, 97% concordant for ethambutol and 94% concordant for isoniazid and streptomycin. In all discordant cases MODS showed resistance, rather than susceptibility. In two cases the bacteria were resistant to low concentration of streptomycin and isoniazid (Moore et al ., 2004, 2006). The discordance in these cases can be explained by the fact that in MODS, as a qualitative method, any growth in drug containing wells is labelled as resistance. Due to lack of colonies, proportion can not be calculated in MODS. Decrease in antibiotic potency due to storage, inocolum variation or true resistance are other possible explanations. In the two cases of low level resistance, decrease in drug potency was suspected and retesting with new stock solution of isoniazid and streptomycin was performed. These samples were found to be sensitive with new stock solution of isoniazid and streptomycin. It was noted, that the stock solution had been stored for 10 weeks at -20 ^o C at the time of susceptibility assay. One isolate which was resistant to ethambutol on indirect assay, was also found to be resistant on direct assay to low (25ìg/ml) and high (50ìg/ml) ethambutol concentration on repeat.

The method can be used in clinical microbiology routine for both indirect susceptibility assay as per this study and for direct susceptibility assay as recommended previously (Moore et al ., 2004). Using it for indirect susceptibility assay in places where indirect assay using LJ proportion method is performed as in most developing countries, it will shorten the time for susceptibility results by more than 2 weeks.

To address microscopic bias the use of SYBR Green to quantify mycobacterial growth can be used. However, further research is needed in this area. In this study clear increase in fluorescence for cultures of drug-resistant strains when compared to culture of sensitive strains was observed. Thus this modification allows quantification of growth and extends the utility of MODS as a routine laboratory tool. This method is fast and gives comparable results with standards methods. It can also be readily used as a routine method in regional and tertiary clinical microbiology laboratories in developing countries where drug-resistant tuberculosis is a major problem. Establishment of MODS in developing countries for routine M. tuberculosis susceptibility testing will help to guide TB therapy and monitor for MDR and extensively drug resistance.^[13]

References

Copyright 2009 - Tanzania Journal of Health Research

The following images related to this document are available:

Photo images