July 1988, Volume 38, Issue 7



Giovanni De Virgilio  ( Italian Cooperation for Development, Italian Ministry of Foreign Affairs, Dirozione General Cooperazione, aio Sviluppo, Via Contarini 25,00194 Roma Italy. “TB Control Programme among Mgan Refugees” P.O. Box 813, University Post Office, Peshawar. )
Paolo Manteffini  ( Italian Cooperation for Development, Italian Ministry of Foreign Affairs, Dirozione General Cooperazione, aio Sviluppo, Via Contarini 25,00194 Roma Italy. “TB Control Programme among Mgan Refugees” P.O. Box 813, University Post Office, Peshawar. )
Paolo Piva  ( Italian Cooperation for Development, Italian Ministry of Foreign Affairs, Dirozione General Cooperazione, aio Sviluppo, Via Contarini 25,00194 Roma Italy. “TB Control Programme among Mgan Refugees” P.O. Box 813, University Post Office, Peshawar. )
Michele Sertoli  ( Italian Cooperation for Development, Italian Ministry of Foreign Affairs, Dirozione General Cooperazione, aio Sviluppo, Via Contarini 25,00194 Roma Italy. “TB Control Programme among Mgan Refugees” P.O. Box 813, University Post Office, Peshawar. )
Giovanni Fadda  ( ICD Consultant for the Referral Laboratory of the “TB Control Programme”, Institute of Medical Microbiology, University of Sassari, Viale San Pietro 43B, 07100 Sassari, Italy. )


In order to support a “Tuberculosis Con­trol Programme”, an adequate laboratory service must be guaranteed. This should include peripheral laboratories capable of preparing and examining smears for acid fast bacilli (AFB) and one “Re­ferral Laboratory” providing facilities for culture, identification and drug sensitivity tests for my. cobacteria.
The direct examination of sputum.smears represents the most important tool for detecting patients with tuberculosis, in order to control the spread of disease in the community.1 However, to improve case-finding, especially when no AFB are found in direct smears, when a failure or relapse occurs during treatment and/or drug-resistance is suspected, culture of the sputum and drug-susceptibility tests should be perfor­med2,3.
Although laboratory procedures for the study of Mycobacterium tuberculosis as well as non-tuberculous mycobacteria (NTM) have been already established and detailed in several pub­lications3-7 in organizing a laboratory service for a “Tuberculosis Control Programme” several rules must be emphasized.

The high cell-wall lipid content (up to 30%) of mycobacteria make them able to bind the dye carbolfuchsin and to resist de-staining with acid-alcohol.
All clinical specimens submitted for the diagnosis of a mycobacterial infection should be examined for AFB. Direct microscopy is much less sensitive than culture and a positive AFB smear provides only a presumptive diagnosis since the actual species of mycobacteria cannot be deter­mined by morphological observation. However, as NTM diseases are rare in high incidence areas of tuberculosis, the acid-fastness of mycobacterium, together with their characteristic size and shape, is useful for the early detectioh of tubercular infection and can help to follow the patient’s response to chemotherapy. Examination of stained smears of sputum is useful also when making appropriate dilutions of decontaminated— con­centrated samples for direct drug sensitivity testing3,8
Acid fast bacilli in specimens are usually rod-shaped and pleiomorphic (2.0-4.4 x 0.3 — 0.5 jz), which may appear bonded with dark-stained parts. In slides obtained from pure culture, M. tuberculosis may show strands of bacilli in “cord”’ :— used3,7
Two types of acid-fast stain are commonly Carbolfuchsin stains by Ziehi-Neelsen (hot stain) or Kinyoun (cold-stain),
2) Fluorochrome stain auramine 0 with or without rodhamine.
Smears stained with carbolfuchsin pro­cedures must be scanned with an oil-immersion objective for 100-300 fields (1x2 centimeters area). Smears stained with the fluorochrome­technique can be scanned with 25 x objective which will increase the field of view, allowing inspection of a larger area in a shorter time.
As the smear examination, using a Ziehi Neelsen procedure is an essential part of a “Tuber­culosis Control Programme”, microscopic ex­amination of sputa must be carried out meticu­lously and, in order to be meaningful, smear results must be quantified. In

Table 1 are the sug­gested interpretation for the reporting of smear results.
A clump of bacilli is considered a single colony forming unit, and must be taken in account as a single unit. In order to classify the slide as positive at least three AFB should be observed on the same slide; this shall avoid possible misinter­pretations between bacillary forms and cellular wastes.
Since M. tuberculosis is released from lungs irregularly, at least three• separate specimens must be examined:
- one spot specimen when the patients first attend the clinic
- an early morning specimen.
- a second spot specimen collected when the early morning specimen is submitted for examination.
Specimen collection
M. tuberculosis as well as non tuberculous mycobacteria can be found in a variety of clinical specimens including sputum, urine, pus and tissue biopsies from organs suspected of being the site of mycobacterial infection2,3.
Recovery and identification of M. tuber­culosis from clinical specimens provides the only definitive proof of tuberculous infection.
In areas of high incidence of tuberculosis nearly all the specimens to be examined come from the respiratory tract. Early morning expec­torated sputum, as well as sputum samples induced by nebulized saline (10%) inhalation, are more useful specimens than the 24 hours sputum pools. In fact the former provides:
- faster detection during smear observation.
- better growth when culture media are inoculated.
- reduced contamination rates when culture is performed.
If patient’s sputum cannot be obtained, gastric washing samples may be utilized3,6.
Since sputa, urine and pus from fistulized localization are usually soiled by commensal bacteria, specimens may be collected in clean but not necessarily sterile screw-capped containers. Specimen digestion and decontamination
The high cell-wall ffpid content of my­cobacterial cells make them more resistant to chemical agents such as alkali, acids and quater­nary ammonium compounds than other types of bacteria which can contaminate sputum, urine and other clinical specimens2,3,9.
Commensal bacteria, which usually re­plicate every 20 minutes can overgrow the slowly growing mycobacteria (average replication time: 20 hQurs). To eliminate undesirable bacteria and liquify mucus, before preparation of the cul­ture, contaminated specimens must be treated with a decontaminating agent capable of killing microorganisms other than mycobacteria.
The agents utilized for digestion and decontamination of clinical specimens are usually alkali and acids such as NaOH (4%), sulphuric acid (4%) and oxalic acid (5%). Because of the high toxicity of strong acid and alkaline solutions, many mycobacteria are also killed. Therefore, carefully controlled period of decontamination and proper neutralizing procedures are man­datory.
N-acetyl-L-cysteine or dithiotreitol (Spu­tolysin; Calbiochem, La Jolla, Calif.) employed simultaneously with 2% NaOH permits a milder alkali treatment, while liquifying tenacious sputum.
Sulphuric acid (4%) and oxalic acid (5%) are useful in the processing respectively of urine samples and of specimens containing Pseudomonas aeruginosa as a contaminant3.
Alternatively trisodium phosphate (13%) combined with benzalkonium chloride (Zephiran) or the chloride or bromide salt of cetyipridiniurn (1%) plus Naci (2%) appear to be milder de­contaminating agents10. With the latter methods careful timing is not required.
Cetyipridinium chloride (CPC), a quater­nary ammonium compound, may be added to the clinical specimen before shipment from out­patient clinics to the “Referral Laboratory”. Decontamination, due to the CPC and lique­fation of the sputum due to Naci, occur in transit. M. tuberculosis can survive 8 days transit without any significant loss of viability. Concentration and inoculation can begin upon arrival in the labora­tory, without further processing the clinical speci­mens.
Since CPC is bacteriostatic for mycobac­teria inoculated onto agar-base media, specimens processed with this chemical compound should be inoculated only onto egg-base10.
After treatment with decontaminating solution, mixtures are then centrifuged to concen­trate mycobacteria. Before this, to dilute toxic substances and decrease the specific gravity of the specimen, the digested-decontaminated sample must be diluted with sterile distilled water or sterile buffered normal saline.
However, for optimal recovery of myco­bacteria, because many of these remain in sus­pension following centrifugation and can be poured off with the discarded supernatant fluid a relatively high centrifugal force such as 3000 x g has been recommended5.
Sterile body fluids and tissue biopsies can. be homogenized and inoculation on to culture media can be done without any decontamination. Recovery and identification of M. tuberculosis
The growth requirements of M. tuber­culosis are such that it will not grow on primary isolation on simply chemically defined media. The only media which allow abundant cultures are those supplemented with blood, bovine serum, bovine albumin or homogenized hen’s egg. Of these, egg-enriched media containing glycerine and asparagine, or agar media supplemented with serum or bovine albumin seem to be the most suitable2,9.
The most commonly used media for isola­tion of M. tuberculosis and NTM are:
1) egg-base media (Lowenstein-Jensen, Petra­gnani, American Thoracic Society),
2) agar-base media (Middlebrook and Cohen, 7H-10 and 7H-11),
3) liquid media (7h-9 broth, Dubos broth),
4) selective media (the Guft modification of
Lowenstein—Jensen, Mycob actosel, selective 7H-11), It has been shown that an increased amount of CO2 (5—10%) in the incubation at­mosphere improves the growth of mycobacteria on egg-media, and is essential for the growth on 711-10-7H-11 agar.
According to Sommers and Good (1985) ideal culture media for the recovery of M. tuber­culosis and NTM from clinical specimens should include one egg-base medium such as Lowenstein­Jensen, a non selective agar medium and one selec­tive medium. However, since the cost of media supplemented with serum or bovine albumin are very high, an egg-enriched medium, such as the standard Lowenstein-Jensen, as prescribed by the International Union against Tuberculosis seem to be the most suitable in a “Tuberculosis Control Programme”9.
Lowenstein-Jensen medium is very sen­sitive, inexpensive and can be stored at 4° C for several months5.
Media inoculated with the largest possible quantity of the sample should be incubated at 35—37°C and examined after 3 and 7 days and thereafter weekly for 6—8 weeks. Cultures that show grossly visible characteristic colonies, are removed for acid fast staining and further iden­tification2,3,5,6.
M. tuberculosis and NTM, may be iden­tified by their acid fastness, growth rate, optimal growth, temperature, colonial photoreactivity, biochemical properties and growth inhibition tests7.
All mycobacteria have cataIasic activity which is thermolabile for M tuberculosis and strong and thermostable ‘for NTM. M. tuberculosis reduces nitrates to nitrites, is naturally resistant to 2 mg/L of thiophen-2-carboxylic acid hydra­zide (TCH), to 500mg/L of 4(p)-nitrobenzoic acid (PNB), and produce a large amount of niacine1.
Drug resistance in mycobacteria is in­dependent of exposure to antibiotics and arises by spontaneous mutations that occur at random in the wild bacteria population8,9,11,12. The proportion of these mutants can be relatively high also in culture of drug-susceptible bacilli isolated from previously untreated patients (Pri­mary Drug Resistance). Their frequencies have been estimated, for example at about from 10-5 to 10-6 for isoniazid (INH), 10-5 for streptomycin (SM) and 10-7 for rifampicin (RM)13. The resistant mutants do not survive well and remain in minority in the normal environment11. However, when a patient is treated with active concentrations of a single antibiotic, susceptible bacteria are inhibited and resistant microorganisms are selected (Acquired Drug Resistance).
Because strains resistant to a given anti­biotic are sensitive to other drugs, the frequency of resistant mutants to more than one drug de­creases and could be estimated in about 1 every 10 bacteria when, for example, INH and SM are simultaneously used. Advantage is taken of these properties in combined therapy of tubercu­losis9,12.
Therefore, the only difference between M. tuberculosis strains which are sensitive or resistant to a single drug is the presence of a higher number of resistant mutants in resistant isolates than in sensitive strains. Clinical drug trials have statistically demonstrated that if more than 1% of a patient’s tubercle bacilli are “in vitro” resis­tant to an antibiotic, the drug is not clinically useful. For such reasons the assessment of suscep­tibility to antibiotics with conventional diffusion techniques (total end point) are not suitable for mycobacteria: with these it is necessary to utilize a method allowing calculation of the proportion of tubercie bacilli, derived from the patient’s specimen, which are resistant “in vitro” to each drug to a lesser or greater extent than 1% (pro­‘portion method: 1% end point).
The proportion method is the only method that enables an exact assessment of the rate of M. tuberculosis sensitive or resistant to a given drug. This technique, which is utilized in highly specialized laboratories (especially. in Italy, France and United States) gives results which, not only are reproducible, but also correlate well with the clinical data. However, in order to calculate precisely the proportion of resistant mycobacteria, procedures must be followed correctly; several dilution of the inoculum have to be inoculated both in drug containing and in control media5. With a correctly performed proportion method it is possible to perform susceptibility test either from sputa rich in AFB (direct test) or from cultures of M. tuberculosis (indirect test).
With the direct test the inoculum size must be adjusted on the basis of the number of AFB seen in the smear9,14.
Recommended dilutions for preparing in­oculum from decontaminated concentrated sputum for the direct mycobacterial sensitivity test are included in table II.

The direct susceptibility test reflects the true situation of tubercle bacilli distribution in patient’s lesions and makes it possible to know the results of a susceptibility test in three to four weeks, whereas almost 8 weeks are required when using the indirect procedure.
The indirect sensitivity method, that uses organisms previously grown on culture medium, is performed when smears are negative but cultures positive and when growth on control medium in direct sensitivity test is inadequate. In order to maintain the true situation of tubercie bacilli distribution as in patient’s lesions, the inoculum for indirect susceptibility test should be prepared by selecting a proportion of all the colonies growing on the culture tube.
The inoculum is adjusted by turbidimetiy to yield countable colonies on one of the used dilutions.
Drugs under test are usually incorporated into the culture medium. The less expensive egg-based Lowenstein-Jensen medium is the most widely utilized. However, to avoid drug inacti­vation during the inspissation procedure, as well as the loss of potency of the antibiotic corn­plexed to egg proteins, several laboratories recom­mend the use of 7H-10 agar medium5. This easily prepared medium must be icubated in 5 to 10% CO2; drugs ate added after autoclaving and just before pouring the plates, thus fully preser­ving antibiotic activity.
Drugs have to be added to the culture medium at critical concentration which inhibits the growth of sensitive tubercle bacilli without affecting multiplication of all resistant mutants8,9.
Pyrazinamide, which is only active in acid environment (pH 5.5) needs to be incorporated in acid culture media. However, as the acid pH of the medium itself could affect the growth of M. tuberculosis one acid medium without pyrazina­mide should be included as a control9.
To assess the presence of the drug in the medium, organisms of known susceptibility, such as a drug susceptible strain of M. tuberculosis H 37 Rv, should be included.
Sensitivity test cultures .are incubated at 35-37° C and results reported at 3 weeks. As resistant colonies often grow more slowly than susceptible ones, test results can be reported in less than 3 weeks only if drug resistance is shown5.
Drug susceptibility tests are not generally performed for new cases of tuberculosis. If the incidence of primary resistance in a given population is less than 5%, a correct combined 3 drugs therapy can cover the risk of selective resistant mutant bacilli. For such a reason the American Thoracic Society suggests that suscepti­bility tests should usually be requested only in cases of tuberculosis which have been cured for a long period of time and which are likely to have acquired resistance.

Table III-VI shows such sugges­tions.
However, if epidemiological studies on the frequency and nature of Primary Drug Resistance have to be carried out, systematic antibiogramme must be performed in several new cases of tuber­culosis selected according to an appropriate statistical procedure.
There are some new exciting developments in the mycobacteriology laboratory. Procedures utilized to identify mycobacteria, such as thin-layer chromatography (TLC), mycolic and fatty acid analyses require sophisticated instruments and very large amounts of mycobacterial cells4,16. DNA probes are under evaluation, but up to now they lack the needed specificity and sensitivity. Monoclonal antibodies are being investigated to develop specific serologic tests and skin tests for hypersensitivity.
Fluorimetric assay of enzimatjc activity, TLC of mycobactins and serologic procedures measuring the immunologic relatedness of the mycobacterial catalase, are under evaluation7. Phage typing for epidemiological studies are so far utilized only by highly specialized labora­tories17,18.
Among the non conventional procedures one of the most widely used is the Bactec radio-metric System (Johnston Laboratories Inc., Cocheysville, M.O.). This method uses media containing14 C-labelled substrata which, when metabolized by bacteria, yield detectable levels of CO2. The radioactivity of the carbon dioxide is measured in the automatable ion chamber, and the relative value is printed out in Growth Index (G.I.) units, 100 of which are equivalent to 0.025 mCi of liberated14 CO2 19. .
The same principle was employed to detect growth in pure culture20,21, and for drug suscep­tibility testing16,20,22 of M. tuberculosis by using a Middlebrook 7H-12 liquid medium, which contains palmitic-1-14 C acid as a labeled sub­stratum.
This method was further developed to provide a rapid differentiation of tubercie bacffli from non-tuberculous mycobacteria by selective susceptibility to p-nitro-a-acetylamino-b-hydropro­piophenone23.
The radiometric procedure has been re­ported to reduce the time required to report culture results of M. tuberculosis from spu­tum20,21 and from extrapulmonary specimens22. Results are available in 5 to 10 days compared with an average23 days required by the conven­tional tests. Cultures can be also identified as M. tuberculosis in 5 to 7 days, and tubercie bacilli can be recovered and susceptibility test completed in an average time of 18 days, compared with the average 38 days of the conventional procedures4.
This new method and the many other techniques under evaluation can reduce the time required both for definitive diagnosis of tuber­culosis and susceptibility testing. Their high costs, however, make them unsuitable for a “Tuber­culosis Control Programme”.


In a “Tuberculosis Control Programme” the patient’s response to therapy should be always carefully monitored. If after 4 or 5 months of regular drug intake the patient’s smears or cultures are still positive the drug-regimens should pro. bably be changed. The role of the “Referral Laboratory” at this stage is essential. It is there­fore important to emphasize that cultures, as well as susceptibility tests for M. tuberculosis hive to be performed only by reliable experienced labora­tories. Those have to be proficient in myco­bacterial studies, with internal quality control check, and at least 10 susceptibility tests done each week11.
Only in such conditions the proficiency of the laboratory may be best maintained and major mistakes avoided. Wrong results, in fact, could be more dangerous for the patient than total absence of results.


1. Tecnichal guide for sputum examination for tuberculosis by direct Microscopy. Bull. Int. Union Tuberc., 1978; Supplement no. 2.
2. David, H.L. The bacteriology of mycobacter­ioses. Atlanta, Centers for Disease Control, PHS, HEW (HEW publication No 76-8316) 1975.
3. Sommers, H.M. and Mc Clatchy, J.K. Laboratory diagnosis of the micro-bacterioses cumulative techniques and procedures in clinical micro­biology n. 16 Washington, American Society for Microbiology, 1983.
4. Good, R.C. Problems and recent developments in laboratory tests for tuberculosis, in Ginesu and Fadda C. (Editors). La tubercolosi; un problema risolto o ancora attuale? Lotta contro Ia tubercolosi e Ie malattie sociali, 1984; 54 (sup); 48-53.
5. Kent, P.T. and Kubica, G.P. Public health myco­bacteriology. A guide for level III Laboratory. U.S. Department of health and human services Atlanta, Georgia, Public Health Service, Centers for Disease Control, 1985.
6. Sommers, H.M. and Good, R.C. Mycobacterium, in Lennette, E.M., Edited by Lennett, E.M. Balows, W. J. Hansler Jr. and H.J. Shadomy in manual of clinical microbiology. Washington, American Society for Microbiology, 1985, p. 216.
7. Wayne, L.G. and Kubica, G.P. Genus myco­bacterium, in Bergey’s manual of systematic bacteriology. Edited by Smeath, P.H. A., Mair, N.S. and Sharpe, E. Baltimore, Williams and Willdns, 1986, V.2,p. 1436.
8. David, H.L. Fundamentals of drug susceptibility testing in tuberculosis. Atlanta, Centers for Disease Control, PHS, HEW (HEW publication No 00-2165), 1971.
9. Grosset, J. and Truffot-Pernot, C. Le laboratoire; son rde dans Ie diagnostic et le traitement dela tubeculose. Bull. Int. Union Tuberc., 1982; 57: 234.
10. Smithwick, R.W., Stratigos, C.B. and David, M.L. Use of cetyipridiniurn chloride and sodium chloride for the decontamination of sputum specimens that are transported to the laboratory for the isolation of mycobacterium tuberculosis. J. Clin. Microbiol., 1975; 1: 411.
11. Canetti, G., Rist. N. and Grosset, J. Measurement of sensitivity of the tuberculous bacillus to antibacillary drugs by the method of proportions. Methodology, resistance criteria, results and interpretation. Rev. Tuberc. (Paris), 1963; 27: 217.
12. Grosset, J. Bacteriologic basis of short-course chemotherapy for tuberculosis. Qin. Chest Med., 1980; 1: 231.
13. David, H.L. Probability distribution of drug-resistant mutant in unselected populations of Mycobacterium tuberculosis. Appl. Microbiol., 1970;20: 810.
14. Cheesbrough, M. Mycobacteria, in medical laboratory manual for tropical countries. England, Thetford, 1984, p. 289.
15. Wayne, L.G. and Krasnow, I. Preparation of tuberculosis testing mediums by means of impre­gnated discs. Am. J.Clin. Pathol., 1966; 45:769.
16. Fadda, G., Foddai, G., Mureddu, F., Sanciu, S. Zanetti. Radiometric investigation of suscepti­bility of Mycobacterium tuberculosis. Myco­bacteria of clinical interest. Edited by M. Casal. Elsevier, 1986.
17. Jones, W.D. Jr., Good, R.C., Thompson, N.J. and Kelly, G.D. Bacteriophage types of Mycobac­terium tuberculosis in the United States. Am. Rev. Respir. Dis., 1982; 125: 640.
18. Jones, W.D. Jr., and Woodley, C.L. Phage-type patterns of Mycobacterium tuberculosis from Southeast Asian immigrants. Am. Rev. Respir. Dis., 1983; 127: 348.
19. DeLand, FJ4. and Wagner, H.NJr. Early detection of bacterial growth, with carbon—14— labelled glucose. Radiology, 1969; 92: 154.
20. Fadda, G., Cossellu, S., Roe, S. L., Rubattu, L., Satta, G.F. and Zanetti, S. Evaluation of a rapid radiometric method in the clinical bacterio­logy laboratory. Clin. Chem. Newsletter, 1984; 4: 72.
21. Middlebrook, G., Reggiardo, Z. and Tigert, W.D. Automable radiometric detection of growth of Mycobacterium tuberculosis in selective media. Am. Rev. Resp. Dis., 1977;115: 1066.
22. Fadda, G. and Roe, S.L. Recovery and suscepti­bility testing of Mycobacterium tuberculosis from extrapu]monary specimens by the BACTEC radiometric method. J. Clin. Microbiol., 1984; 19: 720.
23. Laszlo, A. and Siddiqi, S. Evaluation of a rapid radiometric differentiation test for the Myco­bacterium tuberculosis complex by selective inhibition with p-nitro-a-acetilamino-b-hydroxy­propiphenone. J.Clin. Microbiol., 1984; 19: 694.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: