|Year : 2005 | Volume
| Issue : 4 | Page : 112-115
Problems in laboratory diagnosis of tuberculosis
JM Joshi, P Sundaram, RT Kamble
Department of Respiratory Medicine, B.Y.L. Nair Ch Hospital, Mumbai-8, India
J M Joshi
Department of Respiratory Medicine, B.Y.L. Nair Ch Hospital & TN Medical College, Mumbai-400 008
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Setting : Department of Respiratory Medicine, B.Y.L. Nair Hospital, Mumbai, India.
Objective : To study pre-treatment sputum smear, culture and drug susceptibility testing for mycobacterium tuberculosis in fresh cases of pulmonary tuberculosis, the extent of laboratory related problems and correlation of the laboratory results with clinical outcome.
Design : This study is a prospective analysis of 57 cases of pulmonary tuberculosis that denied previous treatment with anti tuberculosis drugs. Cases with associated human immunodeficiency virus (HIV) infection and diabetes mellitus (DM) were excluded. Pre-treatment smear, culture and drug susceptibility were performed by standard culture techniques. Patients were treated with short course chemotherapy (SCC) on the basis of World Health Organisation (WHO) category I. Laboratory results were correlated with initial clinical data and treatment outcomes.
Results : Of the 57 cases selected, there were 34 males and 23females, age range 18-65 years, mean age 27.86 years. Clinical data was lacking in 16 patients who defaulted on treatment and hence were excluded from the analysis. Of the 41 cases with complete data, 37 patients were declared cured (91.25%) while 4 patients failed on therapy (9.75%), 17/41 (41.46%) had laboratory results consistent with clinical data and treatment results whereas 24/41 (58.53%) had poor correlation between laboratory results, clinical data and treatment outcomes. The major laboratory related problems were: 1) Smear positive / culture negative (S+/C-) in 16/41 (39%) cases at the start of treatment; 2) HR pattern of resistance in 4/41 (9.75%) and R resistance 3/41 (7.31%) on initial culture susceptibility tests but response to SCC suggesting incorrect susceptibility results.
Conclusions : Discrepant reports between clinical findings, laboratory reports and treatment outcomes were found in 58.53% cases. Treatment should not be decided only on the basis of the initial culture susceptibility testing to avoid unnecessary and toxic second line therapy. Improvement in standardisation and new reliable diagnostic techniques are required for management of TB.
|How to cite this article:|
Joshi J M, Sundaram P, Kamble R T. Problems in laboratory diagnosis of tuberculosis. Lung India 2005;22:112-5
| Introduction|| |
Tuberculosis (TB) has been declared a global emergency by World Health Organisation (WHO).  The diagnosis of TB is based on the demonstration of acid fast bacilli (AFB) by smear microscopy and culture of mycobacterium tuberculosis (M.TB). The global increase in the incidence of drug resistant tuberculosis has necessitated performance of drug susceptibility testing more frequently.  However, variations in laboratory methods in evaluatin of M.TB and drug susceptibility can lead to diagnostic errors in a certain percentage of patients. , In addition, some laboratory results are difficult to correlate clinically. , These are, occurrence of smear positive culture negative (S+/C-) states in cases of pulmonary tuberculosis (PTB) prior to onset of therapy , and, development of smear positive culture negative (S+C-) tests during therapy. , Inconsistent laboratory results, reports which correlate poorly with clinical data and reports which are difficult to interpret, often due to lack of awareness of the reasons for their occurrence, lead physicians to rely excessively on radiology for management of TB. This results in over diagnosis and improper treatment of TB and wastage of scarce resources available for TB control particularly in the developing countries. Laboratory related mistakes or errors of interpretations of laboratory results therefore, are likely to be of serious consequence.
We performed a prospective analysis of sputum smear, culture and drug susceptibility test results of fresh cases of pulmonary tuberculosis, without associated diabetes mellitus (DM) and human immunodeficiency virus (HIV) infection, to assess the extent of laboratory related problems encountered in the diagnosis of TB.
| Material and Methods|| |
57 cases of pulmonary tuberculosis that denied previous treatment with anti-tuberculous therapy (ATT) in whom the diagnosis was made by sputum smear examination and chest radiography were included in the study. Initial evaluation included haemoglobin, total and differential blood counts, blood sugar; blood urea nitrogen, liver function tests and enzyme linked immunoassay (ELISA) for human immunodeficency virus (HIV). Cases with associated diabetes mellitus and HIV infection were excluded from the study. Pre-treatment culture and drug susceptibility test was also performed by standard technique from a reputed laboratory of a postgraduate teaching institute. Sputum samples after concentrating by N-acetyl-L-cysteine NaOH (NALC-NaOH) method were inoculated into Lowenstein-Jenson (LJ) medium. Readings were taken on a weekly basis till eight weeks. Absolute concentration method was used for susceptibility; sample was considered susceptible if concentrations were less than following i.e. Streptomycin <2 ug/ml, Isoniazid <0.1ug/ml, Rifampicin <0.5 ug/ml, Ethambutol <2 ug/ml and Pyrazinamide <100 ug/ml. All cases were started on short course chemotherapy (SCC) category I WHO regimen  , 2 months of Isoniazid (INH), Rifampicin (RMP), Ethambutol (EMB), and Pyrazinamide (PZA), and 4 months of INH and RMP but were not given directly observed (DO) therapy. 16 of the patients defaulted while on treatment and were excluded from the study analysis. Failure of sputum smear to convert at three months was considered as failed treatment and repeat culture and drug susceptibility test was ordered. While awaiting results of culture, SCC intensive phase with 4 drugs was continued. If sputum smear remained positive, patients were initiated on second line therapy. Demonstration of negative smears along with clinical and radiological improvement was considered as successful treatment i.e. cure. All the laboratory results were correlated with the initial clinical data and outcome of therapy at the end of the study period.
| Results|| |
57 cases of PTB, 34 males and 23 females, mean age 27.86 years (age range 18-65) without associated DM and HIV co-infection diagnosed by sputum microscopy and chest radiography, and who gave no history of previous treatment with ATT were included in the study. 16 patients defaulted on treatment and were excluded from the analysis due to lack of complete clinical data. Of the 41 cases included, 37 patients were declared cured (91.25%) while 4 patients failed on therapy (9.75%) [Table 1]. 17/41 (41.46%) had laboratory results consistent with clinical data and treatment results whereas 24/41 (58.53%) had poor correlation between laboratory results, clinical data and treatment outcome [Table 2].
Among those cases who had inconsistent clinical and laboratory correlation, the largest number of cases had positive sputum smears but negative cultures. Thus 15/37 of the cured cases included in the analysis had smear positive culture negative (S+/C-) status. All these patients had denied history of previous treatment with ATT (which was the inclusion criteria for the study). HR pattern of resistance was seen in 4/37 and isolated R resistance (without H resistance) resistance was seen in 3/37 patients. HR resistance also called multi drug resistance (MDR TB) and Rifampicin mono-resistance, though rare, result in failure of SCC. Hence the successful outcome of SCC in these 7 cases suggests incorrect laboratory results. One case, 1/4 who was initially S+C- failed to respond to SCC, repeat C/S at 3 months showed resistance to RMP, PZA, and Streptomycin (SM). The patient was treated with second line drugs with good response. One case represented another aspect of poor laboratory and clinical correlation. He had shown sensitivity to all first line drugs (non-HR pattern of resistance). This patient failed on SCC subsequently requiring second line therapy. Unexplained high levels of resistance was seen to other anti TB drugs Z (n=18), KM (n=9), Ethionamide (ETA) (n=19) CYS (n=5) [Table 3].
| Discussion|| |
58.33% cases in this study showed culture and susceptibility results, which could not be correlated with the clinical data and treatment outcome. Although routine initial culture sensitivity testing is not recommended in the diagnosis of TB, such tests are now performed due to the rising incidence of MDRTB  . The danger of relying solely on laboratory results in diagnosing and treating TB diseases are obvious. Thus, treatment should be based on accurate history correlating it with laboratory results. Standardised laboratory techniques with quality control is most important.  However, even if these pre-requisites are ensured diagnosis or therapeutic decisions should be made only after careful clinical correlation  , failing which serious errors are likely to be made in the management of TB.
The common laboratory related problems encountered in this study were the phenomenon of S+C- status 16/41 (39%). Chen et al  have reported 4-6% cases as having S+C- status prior to therapy. S+C- phenomenon has also been reported in 25-37% cases with active pulmonary tuberculosis while on treatment, , particularly with RMP containing regimens. S+C- in patients receiving chemotherapy are due to organisms which may have lost the ability to grow on culture media and are practically dead. In patients not on chemotherapy the causes of negative cultures when sputum is positive, are exposure of specimens to sunlight or heat, storage of specimens too long before inoculation, inadequate culture media and deficient incubation.  Although all our cases denied previous treatment, it is possible that the treatment history was incorrect. Undisclosed previous treatment may have possibly resulted in S+C- status in some of the cases. Smears of cases who are S+C- while on therapy usually convert to negative while continuing the same therapy. However, if smears are persistently positive after 14 weeks of cultures turning negative, change of treatment of smear results alone should be considered. One of our cases of S+C- status failed on therapy and required treatment with second line drugs. Repeat culture at 3 months in this case was positive and drug susceptibility testing showed resistance to R, Z, and SM.
4/41 (9.75%) cases showed HR pattern of resistance initially but responded to SCC. Additional 3/41 (7.31%) cases showed RMP resistance. The consequence of HR and R resistance is failure of SCC. , Hence it was surprising that the initial cultures in these cases had shown drug resistant organisms. Misdiagnosis of MDR-TB due to laboratory related errors has been reported recently.  The possible explanation has been cross contamination with M. avium complex, suspected mislabelling and discrepant susceptibility tests due to poorly standardised techniques in different in different laboratories. Another important issue has been the reliability of the techniques currently used to measure drug resistance. Although H and R resistance can be reliably measured, resistance to Z, E, S is more difficult due to limitation of technique.  The therapeutic index for a given drug, which is the difference between in vitro MIC and the drug levels obtained in blood is low for certain second line drugs such as ethionamide, cycloserine, kanamycin and PAS leading to misinterpretation of results due to failure to distinguish between sensitive and resistant strains.  The unexpected high resistance to first and second line drugs seen in our study confirms these facts. For these reasons routine testing for second line drugs is not recommended. ,
The other problems encountered in our laboratory results were pertaining to incorrect drug susceptibility results. These were, sensitivity to all first line drugs or non-HR resistance but failure of SCC chemotherapy. One case in this study (2.43%) showed sensitivity to all first line drugs, seemed to be compliant with the therapy, however failed to respond to SCC. WHO has suggested that most cases of failure on SCC are likely to be cases of "failure to take treatment rather than treatment failure".  Such cases should be put on WHO category II regimen under direct observation (DO). MDR TB should be considered only after this DO treatment fails.
In conclusion, the findings of our study confirm the discrepancy between clinical findings, laboratory results and treatment outcome in cases of PTB. It is therefore extremely important to make a clinical correlation of the laboratory results and order repeat tests if there is poor correlation between the two. Improvement in laboratory techniques and standardisation along with newer, more reliable diagnostic techniques will perhaps offer a solution to these problems. Till such time, unnecessary antituberculous therapy, particularly with more expensive and toxic second line drugs can be avoided by being aware of the limitations of laboratory results in diagnosis of TB and avoiding therapeutic decisions based solely on them.
| References|| |
|1.||Tuberculosis: A global emergency. World Health Forum 1993; 14:438. |
|2.||Tenover FC, Crawford JT, Heubner RE, Ceiter LF, Harsburgh CR, Good RC. The resurgence of tuberculosis: is your laboratory ready? J Clin Microbiol 1993; 31:767-770. |
|3.||Nitta AT, Davidson PT, deKong ML, Kilman RJ. Misdiagnosis of multidrug resistant tuberculosis possibly due to laboratory related errors. JAMA 1996; 276: 1980-1983. |
|4.||Dunlop NE, Harris RH, Benjamin WH Jr, Herden JW, Hafner D. Laboratory contamination of Mycobacterium tubercle cultures. Am J Respir Crit Care Med 1995; 152:1702-1704. |
|5.||Kim TK, Blademon RS, Heatwole KM, Kim T, Rochester DF. Acid fast bacilli in sputum smears of patients with pulmonary tuberculosis. Am Rev Respir Dis 1984; 129: 264-268. |
|6.||Dott AK, Stead WW. Short course chemotherapy - The Arkansas experience. Chest 1981; 80: 724-727. |
|7.||Chan W et al. Bulletin of the World Health Organisation 1971; 45: 551. |
|8.||Toman K. Tuberculosis case finding and cure: Questions and Answers. WHO, Geneva, 1974, 39. |
|9.||Weir JA, Storey PD, Walrath MS, Weiser O, Temple CW. The incidence of smear positive culturally negative sputum in patients with pulmonary tuberculosis. Transactions of the 14th conference on chemotherapy against tuberculosis : VA - Armed Forces Washington 25 DC: VA Dept. of Medicine and Surgery, Central Office 1955: 255-8. |
|10.||Warring Fc, Jr, Sutramonglcote U. Non-culturable acidfast forms in the sputum of patients with tuberculosis and chronic pulmonary disease. Am Rev Respir Dis 1976; 102: 714-724. |
|11.||Vareldiz BP, Grusse J de Kantor I, Crofton J, Laszk A, Felten M, Reviglicnc MC, Kochi A. Drug resistance tuberculosis: laboratory issues. Tubercle and Lung Diseases 1994; 75; 1-7. |
|12.||Guidelines for management of drug resistant tuberculosis. WHO / TB / 96.210. |
|13.||Mitchison D A, Nunn A J. Influence of initial drug resistance on the response to short course chemotherapy of pulmonary tuberculosis. Am Rev Respir Dis 1986, 137; 423-430. |
|14.||Sandman L, Schluger N W, Davidow A L, Bank S. Risk factors for Rifampin-monoresistant tuberculosis. Am J Respir Crit Care Med 1999, 159; 468-472. |
|15.||Coretti G, Fox W, Khamarko et al. Advances in techniques of testing mycobacterial drug sensitivity and the we of sensitivity tests in tuberculosis control programmes. Bull World Health Organ 1969; 41: 21-43. |
|16.||Canetti G, Rist N, Grosset J. Primary drug resistance in tuberculosis. Am Rev of Tuberculosis and Respir Dis 1964; 90: 792-799. |
[Table 1], [Table 2], [Table 3]