Journal of Clinical Pharmacy and Therapeutics (1987) 12,249-254.

RESEARCH AND REPORTS

A COMPARISON OF TWO INCUBATION TEMPERATURES FOR THE ISOLATION OF GRAM-NEGATIVE CONTAMINANTS

FROM RAW MATERIALS AND NON-STERILE PHARMACEUTICALS

Anne Ferguson, Asmita Pate1 and Rosamund M. Baird North East Thames Regional Health Authority, Pharmaceutical Microbiology Laboratory,

St Bartholomew’s Hospital, West Smithjield, London EC1, U.K.

SUMMARY

Selective and non-selective broth enrichment techniques may be used in the isolation of microbial contaminants from pharmaceutical products. A non-selective method may give better recovery rates for damaged organ- isms. A trial was carried out to determine whether the recovery of Gram-negative contaminants could be improved by using an incubation temperature of 30°C for 48 h, rather than the more widely used 37°C for 24 h. Contaminants were isolated from 3.2% of samples incubated at the lower temperature compared with 0.8% at the higher temperature. The recovery rate from raw materials improved noticeably (9.0% compared with 0.9%).

INTRODUCTION

The North East Thames Regional Health Authority Pharmaceutical Microbiology Laboratory provides a microbiological quality assurance service for hospitals within the region. A wide variety of non-sterile products and raw materials are examined routinely and are required to conform to accepted regional microbiological standards (1 , 2). Of those contaminants considered to be undesirable, the Pseudomonads and related genera are the organisms most commonly isolated in practice (3). The need to test pharma- ceutical products for the presence of these, and other organisms, is well established (4,5). This report presents the results of a trial carried out in the laboratory to test whether the existing method used to isolate microbial contaminants could be made more sensitive.

For the past 10 years an all-purpose non-selective enrichment technique was used for the examination of all non-sterile products and raw materials. The technique involved aerobic incubation of the sample in broth at 37°C for 24 h followed by subculture onto both selective and non-selective solid media. This particular temperature was originally chosen in order to detect organisms such as Staphylococcus aureus, Salmonella sp.,

Correspondence: R. M. Baird.

249

250 A . Ferguson, A . Patel and R . M . Baird

Escherichia coli, and Pseudomonas aeruginosa, all considered to be unacceptable contaminants in pharmaceutical products. In practice, however, with the exception of Ps. aeruginosa, these organisms are rarely found; Gram-negative opportunist pathogens, such as other Pseudomonads, Achromobacter sp. and Acinetobacter sp., are more likely to be isolated. Such free-living organisms are widely found in the environment and have an optimal growth range of 30-35°C.

Thus it was decided to carry out a 12-month trial whereby samples were incubated aerobically for 6 months in an enrichment broth at a temperature of 37°C for 24 h. These results were then compared with tests carried out in the following 6 months using 30°C as the incubation temperature for a period of 48 h.

M E T H O D One millilitre or 1 g of each sample was added to 20ml of nutrient broth (Oxoid) containing 3% Tween 80. Ointments and powder products were homogenized in the broth using a Seward Stomacher 400. Products containing halogen compounds were added to 20 ml nutrient broth (Oxoid) containing 1 yo sodium thiosulphate. Phenolic products were diluted in 100ml nutrient broth (Oxoid). A total viable count was obtained for each sample by plating 0.1 ml of the inoculated broth onto horse blood agar (Tissue Culture Services). For the first 6 months of the trial, all broths and agar plates were incubated aerobically at 37°C for 24 h. For the next 6 months, the incubation was carried out at 30°C for 48 h.

Broths which showed signs of turbidity after incubation were subcultured onto horse blood agar and nutrient agar (Oxoid) containing 0.03% cetrimide and incubated as before. The resultant isolates were identified according to their Gram-stain reaction and colonial morphology. Coagulase tests were carried out on all Gram- positive cocci. Gram-negative rods were identified using either the APl 20E, or 20NE Microtubule systems according to their oxidase reaction and colonial morphology.

RESULTS

Existing regional specifications for non-sterile products and raw materials require a total viable count of less than 104cfu/ml or g and the absence of the unacceptable organisms, previously mentioned, as well as other Gram-negative contaminants. In the first 6 months of the trial 893 samples were examined (674 non-sterile products and 219 raw materials), 9 (1.0%) failed to meet the required specification, and of those 7 (0.8yo) contained Gram-negative rods. In the following 6 months, 868 samples were examined (623 non-sterile products and 245 raw materials), 31 (3.6%) failed the specification; 28 (3.2%) contained Gram-negative rods.

The figures for non-sterile products and raw materials were examined separately for each 6-month period as shown in Tables 1 and 2. Although there was no significant increase in the overall proportion of failures recorded in the non-sterile category (0.7% and l .O~o), there was a 10-fold increase in the proportion of raw material failures seen following incubation at 30°C for 48 h (0.9y0 and 9.0%). However, not all types of products were equally affected, for example, syrups and detergents from both non-sterile

Isolation of Gram-negative contaminants 25 1

and raw material categories showed a marked increase. Gram-negative isolates from both non-sterile products and raw materials are shown in Tables 3 and 4. All except one of the raw material detergent solution failures were Teepol products. The figures obtained from incubation at 30°C may, therefore, have been artificially high for this product but no comparison could be made with the previous 6 months as no Teepol samples were received for testing.

Table 1. Comparison of contamination rates in non-sterile products incubated by two methods

Incubation at 37°C for 24 h Incubation at 30°C for 48 h

Number Number Number Number Product tested failed tested failed

Syrups 47 0 25 2 Detergent solutions 43 1(2.3%) 38 2 (5.3%) Oral liquids 46 1 4 (0.9%) 428 1(0.2%) Ointments 114 0 123 0 Natural products 7 0 8 1 Powders 2 0 1 0

Totals 674 5 (0.7%) 623 6 (1.0%)

Table 2. Comparison of contamination rates in raw materials incubated by two methods

Incubation at 37°C for 24 h Incubation at 30°C for 48 h

Number Number Number Number Raw material tested failed tested failed

Syrups 86 0 63 2 (3.2%)

Oral liquid bases 13 0 27 0 Ointment bases 35 1(2.8%) 48 2 (4.2%) Natural products 57 0 46 3 (6.5%) Powders 12 1(8.3%) 38 3 (7.9%)

Totals 219 2(0.9%) 245 22 (9.0%)

Detergent solutions 16 0 23 12 (52.2%)

DISCUSSION The results indicate that by incubating the broth cultures at 30°C for 48 h rather than at 37°C for 24h the recovery of Pseudomonads and other Gram-negative opportunist pathogens from pharmaceutical products was improved. While a sig- nificant overall improvement in the detection of such organisms was observed in

252 A . Ferguson, A . Patel and R . M . Baird Table 3. Gram-negative isolates in 10 non-sterile products

Type Product Isolate

Syrups Trifluoperazine Ps. vesicularis

Detergent solutions Teepol shampoo Ps.Juorescens (2)

Oral liquids Peppermint water Ps. cepacia

Dextrose 0.75% Ps. cepacia

Moraxella urethralis

Ps. spp.

Ps. sp. (2) Phenobarbitone sodium Ps. maltophilia

Natural products Normacol special Citrobacter freundii

Table 4. Gram-negative isolates in 22 raw materials

Type Product Isolate

Syrup Syrup BP (preserved)

Detergent solutions Teepol (unpreserved)

Teepol L.

Ointments Soft soap Wool alcohols Emulsifying Emulsifying wax

Ps. stutzeri Ps. cepacia Ps. putida Ps. stutzeri (4) Moraxella sp. Ps.fluorescens ( 3 ) Ps. stutzeri Ps. sp. Ps. aeruginosa Ps. pickettii Ps. fluorescens

Natural products Tragacanth Enterobacter agglomerans Powders Calcium resonium Ps. stutzeri

Resonium A Ent. agglomerans Sodium phosphate Achromobacter xylosoxidans Calcium carbonate Ps. sp.

the case of raw materials, there was no apparent advantage in the case of non- sterile products. These results were not unexpected in the light of the following considerations.

In order to produce a finished product, the raw material components may be subjected to processes which adversely affect their microbial flora, killing many and reducing the viability of others. The latter, even when put in the more favourable environment of a broth enrichment culture may show an extended lag phase, and growth may be at a slower rate than before processing (7). Many finished products additionally contain a preservative which injures the cells sublethally. For this reason, inactivating agents, such as Tween 80, are routinely added to enrichment broth cultures, thereby facilitating

Isolation of Gram-negative contaminants 253

the recovery and growth of injured organisms. As a general rule, however, the import- ance of resuscitation procedures, such as those commonly used in the food industry, has been overlooked in the isolation techniques used in pharmaceutical microbiology (8,9). Preliminary incubation at a lower temperature, as used here, is one such method routinely employed.

In the light of these results, it is interesting to study the tests for microbial contami- nation specified in the British Pharmacopoeia, European Pharmacopoeia, and United States Pharmacopoeia (6, 10, 11). The EP test for the presence of Pseudomonas sp. recommends the use of cetrimide broth incubated at 30-32°C for 72°C which is then subcultured onto cetrimide agar and re-incubated at the same temperature for a further 48 h. Here, lower temperature and longer incubation period are specified in order to allow for the slower growth of organisms in a selective medium. In practice, however, cetrimide broth has been observed to be too inhibitory to recover injured cells. In a busy laboratory there is not always time or, indeed, the necessary resources to set up several tests for the presence of different groups of organisms. Furthermore, in using a selective broth enrichment technique, inevitably other Gram-negative organisms go undetected. Thus a non-selective enrichment method can be used instead, followed by subculture onto selective solid media. The test methods defined in the USP for the detection of Ps. aeruginosa and Staph. aureus use this principle. The product sample is inoculated into tryptone soya broth and incubated at a temperature between 30-35°C for 24-48 h.

The results of this trial suggest that if incubation is carried out at the higher end of the temperature range, then certain species of Pseudomonas and other genera such as Achromobacter or Acinetobacter, if present, will remain undetected. Until recently, such organisms were not considered to be potentially hazardous to patients and therefore, their presence in medicaments was not actively sought. There is, however, growing evidence that a number of organisms, other than the species listed in the Pharmacopoeias, are of clinical significance to certain groups of patients, particularly to immunosuppressed and debilitated patients in hospital (4, 12, 13).

A further comparison of the two methods of incubation on the recovery rates of the micro-organisms, run in parallel rather than sequentially, would be worthwhile to confirm the results of this study.

ACKNOWLEDGMENT

The technical help of Mrs K. Pate1 is gratefully acknowledged.

REFERENCES

1. Baird, R.M. (1985) Microbial contamination of non-sterile pharmaceutical products made in the North

2. Baird, R.M. (1985) Microbial contamination of pharmaceutical products made in a hospital pharmacy-

3. Baird, R.M., Brown, W.R.L. & Shooter, R.A. (1976) Pseudomonas aeruginosa in hospital pharmacies.

4. Parker, M.T. (1972) The clinical significance of the presence of micro-organisms in pharmaceutical and

5. Report by a Public Health Laboratory Service Working Party (1971) Microbial contamination of

6. British Pharmacopoeia (1986) Vol 11, Appendix XVI. H.M.S.O., London.

East Thames Regional Health Authority. Journal of Clinical and Hospital Pharmacy, 10,95-100.

a nine year survey. PharmaceuticalJournal, 234,5456.

British MedicalJournal, 1,511-512.

cosmetic preparations. Journal of the Society of Cosmetic Chemists, 23,415426.

medicines administered to hospital patients. PharmaceuticalJournal, 207,96-99.

254 A. Ferguson, A. Patel and R . M . Baird 7. Gilbert, P. (1984) The revival of micro-organisms sublethally injured by chemical inhibitors. In:

The Revival of Injured Microbes (eds. M.H.E. Andrew & A.D. Russell) S A B Symposium Series No. 12, pp. 175-197. Academic Press, London.

8. Ray, B. (1986) Impact of bacterial injury and repair in food microbiology: its past, present and future. Journal of Food Protection, 49,65 1-655.

9. Mossel, D.A.A. & Corry, J.E.L. (1977) Detection and enumeration of sublethally injured patho- genic and index bacteria in foods and water processed for safety. Alimenta-Sonderausgabe Microbiol, 19-34.

10. European Pharmacopoeia. (1980) 2nd edn, Part I. Maisonneuve, France. 11. United States Pharmacopoeia (1985) 21st edn, United States Pharmaceutical Convention Inc. Rockville,

12. Bassett, D.C.J. (1971) Causes and prevention of sepsis due to Gram-negative bacteria; common source

13. Ramphal, R. & Kluge, R.M. (1979) Acinetobacter calcoaceticus variety anitratus: an increasing

Maryland.

outbreaks. Proceedings of the Royal Society of Medicine, 64,98&986.

nosocomial problem. American Journal of Medical Science, 217,57-66.