Part III: Culture
homogenisation and decontamination

The usual microbiological techniques of plating clinical material on selective or differential culture media and subculturing to obtain pure cultures cannot be applied to tuberculosis bacteriology. M. tuberculosis requires special media not used for other organisms and grows slowly, taking three to six weeks or longer to give visible colonies. Cultures are usually made in bottles rather than in petri dishes because tubercle bacilli are present in relatively small numbers in most specimens; this necessitates large inocula which are spread out over the surface of the media. Because of the long incubation time required, the bottles are tightly stoppered to prevent drying of the cultures (which would occur in petri dishes).

The majority of clinical specimens submitted to the tuberculosis culture laboratory are contaminated to varying degrees by more rapidly growing normal flora organisms. These would rapidly overgrow the entire surface of the medium and digest it before the tubercle bacilli start to grow. Most specimens must, therefore, be subjected to a harsh digestion and decontamination procedure that liquefies the organic debris and eliminates the unwanted normal flora.

All currently available digesting/decontaminating agents are to some extent toxic to tubercle bacilli; therefore, to ensure the survival of the maximum number of bacilli in the specimen, the digestion/decontamination procedure must be precisely followed. In order for enough tubercle bacilli to survive to give a confirmatory diagnosis, it is inevitable that a proportion of cultures will be contaminated by other organisms. As a general rule, a contamination rate of 2%-3% is acceptable in laboratories that receive fresh specimens; if specimens (especially sputum) take several days to reach the laboratory then losses due to contamination may be as high as 5%-10%. It is also important to note that a laboratory which experiences no contamination is probably using a method that kills too many of the tubercle bacilli.

When culturing tubercle bacilli, three important aspects should be borne in mind:

• Specimens must be homogenised to free the bacilli from the mucus, cells or tissue in  which they may be embedded. The milder this homogenisation the better the results
• Neither homogenisation nor decontamination should unnecessarily diminish the viability of tubercle bacilli
• The success of homogenisation and decontamination depends on:
  • the greater resistance of tubercle bacilli to strongly alkaline or acidic digesting solutions
  • the length of exposure time to these agents
  • the temperature build-up in the specimen during centrifugation
  • the efficiency of the centrifuge used to sediment the tubercle bacilli

Many different methods of homogenisation and decontamination of sputum specimens for culturing have been described but there is no universally recognised best technique. The choice of a suitable method is to a large extent determined by the technical capability and the availability of staff in a laboratory, as well as the quality and type of equipment available. Each method has its limitations and advantages and it is recommended that regional/central laboratories standardise on one method only. Methods which consistently yield the highest percentage of positive cultures are those which require:

• well trained staff
• relatively expensive equipment (eg. centrifuges) and related supplies
• continued maintenance of equipment and of good staff performance

Any method which require the use of a centrifuge present some problems which must be considered:

• The centrifuge must be fast enough to attain a relative centrifugal force (RCF) of 3 000 x g. If the RCF is not high enough, many tubercle bacilli remain in suspension following  centrifugation and are poured off with the discarded supernatant fluid. Recent studies  have shown that 3 000 x g for 15 minutes would sediment 95% of mycobacteria in a digested sputum specimen. The specific gravity of tubercle bacilli ranges from 1.07 to 0.79, making centrifugal concentration of specimens ineffective if the RCF is not 3 000 x g
• Precautions must be taken to minimise the potential for staff infection in the event of  tube breakage during centrifugation. These include:
  • using a floor model centrifuge with lid and a fixed angle rotor. The mass of the fixed  angle rotor permits centrifugation of tubes with small weight differences without causing  vibration and possible tube breakage
  • always ensuring that tubes in the centrifuge are balanced. The weight of centrifuge tubes can be balanced by adding sterile saline to specimens or by inserting tubes with sterile water or 70% ethanol among the tubes containing sputum (using 70% ethanol in stead  of water in the tubes used as balances will reduce the risk in the event of breakage)
  • using aerosol-free safety cups if available
  • enclosing the centrifuge in a specially ventilated cabinet if possible

Digestion and decontamination procedures
Sputum specimens
Sputum specimens should not be pooled because of the risk of cross-contamination. Since the exposure time to digestants/decontaminants has to be strictly controlled it is best to work in sets equivalent to one centrifuge load (eg. eight specimens at a time).

Always digest/decontaminate the whole specimen, ie. do not attempt to select portions of the specimen as is done for direct microscopy. If the sputum will pour, it should be gently decanted from the specimen container into the centrifuge tube. If the specimen is too viscuous to pour, an equal volume of digestant/decontaminant could be added to the sputum in the specimen container and the mixture poured carefully into the appropriate centrifuge tube.

Since sputum specimens are the most common clinical specimens submitted for tuberculosis culture, homogenisation and decontamination procedures have been largely targeted towards their processing. Specimens other than sputum demand even more care during processing because of the low numbers of tubercle bacilli present in positive specimens.

SODIUM HYDROXIDE (MODIFIED PETROFF) METHOD
This method is used widely in developing countries because of its relative simplicity and the fact that the reagents are easy to obtain.

NaOH is toxic, both for contaminants and for tubercle bacilli; therefore, strict adherence to the indicated timing is required

Reagents
4% sodium hydroxide (NaOH) solution
Sodium hydroxide pellets (analytical grade) 4g
Distilled water 100ml
Dissolve NaOH in distilled water and sterilise by autoclaving at 121EC for 15 minutes.

Sterile saline
Sodium chloride pellets (analytical grade) 0.85g
Distilled water 100ml
Dissolve NaCI in distilled water and sterilise by autoclaving at 121EC for 15 minutes.
An alternative method for preparing 4% NaOH is as follows:

Add the contents of a 250g bottle NaOH pellets to 500ml distilled water and fill water to the 625ml mark. Be careful since heat is released in this reaction. When needed, prepare a fresh 4% NaOH solution by adding 40% NaOH to sterile distilled water in the proportion 1:10.

Procedure
Refer to Diagram 1.

Advantages

• The NaOH method is simple and inexpensive and provides fairly effective control of contaminants
• The time needed to process a single specimen is approximately one hour; 20 specimens would take  approximately two hours, with centrifuge capacity being the limiting factor
• Sterilised NaOH solution will keep for several weeks

Limitations

• The specimen exposure times must be strictly followed to prevent over-kill of tubercle bacilli
• The NaOH procedure is very robust and may kill up to 60% of tubercle bacilli in clinical specimens.  This initial kill is independent of additional contributory factors such as heat build-up in the centrifuge and centrifugal efficiency

A variety of more expensive and labour intensive homogenisation and decontamination methods¹ are available to countries with the required financial and human resources. Some of these will be discussed briefly.

¹Kent PT, Kubica GP. Public health mycobacteriology: Guide for the Level III Laboratory. US Department of Health and Human Services, Centres for Disease Control, USA, 1985.

Excessive contamination is sometimes encountered in clinical material from certain patients, from certain areas or at certain times, and may present a difficult problem. For these problem specimens alternative decontamination methods such as 5% oxalic acid or 4% sulphuric acid may be used.¹

Other specimens
Gastric lavage
These specimens should be processed within four hours of collection since their acidity is damaging to tubercle bacilli. Usually, gastric lavage does not need to be decontaminated, provided it has been collected aseptically in a sterile container. Centrifuge the total volume at 3 000 x g for 30 minutes. If contamination is suspected the sediment should be mixed with 2ml of 4% sulphuric acid and allowed to stand for 15 minutes, after which 15ml sterile saline is added. Centrifuge this mixture at 3 000 x g for 15 minutes and neutralise the sediment with 4% NaOH containing a phenol red indicator. Inoculate the sediment immediately onto culture medium.

Urine
Centrifuge the total volume at 3 000 x g for 15 minutes. Discard the supernatant fluid and add 2ml of 4% sulphuric acid to the sediment. Let stand for 15 minutes, add 15ml sterile saline and centrifuge at 3 000 x g for 15 minutes. Neutralise the sediment with 4% NaOH containing a phenol red indicator. Inoculate the sediment immediately onto culture medium.

¹Kent PT, Kubica GP. Public health mycobacteriology: Guide for the Level III Laboratory. US Department of Health and Human Services, Centres for Disease Control, USA, 1985.

Laryngeal swabs
Cover the swab (in its original tube) with 5% oxalic acid and allow to act for 15 minutes. Remove the swab to another tube containing sterile saline. Lift after a few minutes, allow to drain and use to inoculate culture media.

For optimal results the oxalic acid (which might contain tubercle bacilli washed off from the swab) should be transferred to a centrifuge tube and centrifuged at 3 000 x g for 15 minutes. Wash the sediment once with sterile saline, centrifuge at 3 000 x g for 15 minutes and inoculate immediately onto culture medium.

Tissue
Lymph nodes, biopsies and other surgically resected tissue should be cut into small pieces with a sterile scalpel or scissors. Homogenise the specimen in a sterile porcelain mortar or tissue grinder, using 0.5-1ml sterile saline and a small quantity of sterilised sand (if necessary in mortar). This suspension can be directly inoculated onto culture media if the sterility measurements described before have been met; if not, decontaminate using 4% sulphuric acid as described for urine.

Mortars, pestles and tissue grinders must be cleaned and sterilised thoroughly to prevent false positive results or contamination due to organisms left over from previous specimens

Pus
This may be treated in the same way as aspirated fluids. If the material is very thick, it should be treated in the same way as sputum.

Cerebrospinal fluid
Cerebrospinal fluid should be concentrated by membrane filtration, by high speed centrifugation or by precipitation methods. Precipitation can be achieved by adding 0.1ml sterile rabbit serum or an equivalent albumin solution for every 10ml of cerebrospinal fluid. Mix until uniformly cloudy, centrifuge at 3 000 x g for 15 minutes and culture the sediment if contamination is not suspected.

If no sediment can be obtained by centrifugation, a sterile 20% solution of sulpho-salicylic acid may be added drop by drop until turbidity sets in. This precipitate is then more easily spun down to form a sediment.

If contamination of cerebrospinal fluid is likely, the sediment is mixed with 2ml of 4% sulphuric acid and allowed to stand for 15 minutes. Add 15 ml of sterile saline and centrifuge at 3 000 x g for 15 minutes. Inoculate sediment onto culture media.

Clots
In the case of specimens that form large clots, eg. pleural and ascitic fluids, it is recommended that clot formation be avoided by the addition of sodium citrate at the time of specimen collection. Add two drops of 20% sodium citrate for every 10ml fluid collected.

When clots are present, they can be digested with Petroff?s NaOH method after homogenisation as described for tissue.

Other body fluids (including pleural fluid)
Mucopurulent fluid : Treat as for sputum when volume is 10ml or less.

Clear fluid : If collected aseptically centrifuge at 3 000 x g for 15 minutes and inoculate sediment directly onto culture media. If volume is more than 10ml treat as for gastric lavage.

Tubercle bacilli may adhere to glass or plastic surfaces. To optimise recovery, containers could be rinsed with sterile saline. Centrifuge the saline at 3 000 x g for 15 minutes and inoculate 2-3 drops onto culture media.

Specimens differ greatly in their degree of contamination and decontaminants should be selected to suit the nature of the specimens. The need for decontamination is also determined by the freshness of the specimen and by the efficiency of refrigeration before processing.

The following specimens usually do not need decontamination when aseptically collected into sterile containers:

• Spinal, sinovial or other internal body fluids
• Bone marrow
• Pus from cold abscesses
• Surgically resected specimens  (excluding autopsy material)
• Material obtained from pleural, liver  and lymph nodes as well as biopsies (if not fistulised)

Whenever doubt exists about the contamination of specimens, an untreated portion may be inoculated onto nonselective bacteriological media, eg. nutrient agar, and incubated for 24 hours to check for the presence of fast-growing nonmycobacterial organisms. The remaining portion of the specimen is kept untreated and refrigerated until the absence of contaminants is confirmed. Should this not be the case the remaining specimen can then be appropriately decontaminated.