J Clin Pathol: first published as 10.1136/jcp.38.9.1073 on 1 September 1985. Downloaded from

J Clin Pathol 1985;38:1073-1084

Technical method

Use of microwaves for acid and alcohol fast staining

S HAFIZ, RC SPENCER, MARGARET LEE, organism depends on several factors but remains an HILARY GOOCH, BI DUERDEN Department important component in the identification of a ofMedical Microbiology, Sheffield University Medi- restricted group of organisms. cal School, Sheffield The technical problems associated with the classic Ziehl-Neelsen method include the dangers of heat- ing by flaming torch in laboratories in which volatile Certain bacteria that are characterised by a high organic solvents are used, the deposits that accumu- content of lipid in their cell walls cannot be stained late on the underside of the slide as a result of heat- by simple stains, and either heat or prolonged con- ing, and the crystalline deposits of stain that collect tact is required to drive the stain into the cells; once on the film as a result of evaporation and drying. stained they resist decolourisation by mineral acids This method also requires 20-30 minutes of or acid and alcohol. These organisms are designated laboratory time. Numerous minor modifications acid fast or acid and alcohol fast. They include have been made to the Ziehl-Neelsen method by Mycobacterium tuberculosis and related mycobac- increasing the concentration of carbolic magenta or teria, Mycobacterium leprae, and certain of the by cold staining techniques,'0-'2 but the basic prob- actinomycetes. Koch first stained the tubercle bacil- lems remain. lus by immersion in an alkaline solution of Some of the problems may be overcome by the copyright. methylene blue for 24 hours.' Erlich used warm sol- use of microwave energy, a type of high frequency utions of magenta (fuchsine) or methyl violet for (2450 MHz) radiowave. It is absorbed by molecules 15-30 minutes; the bacilli then resisted decolourisa- of water, causing them to vibrate at a very high fre- tion with strong mineral acids.2 He believed that the quency and produce heat, which radiates out; con- bacilli were resistant to ordinary stains because they sequently, there is an increase in the turgor pressure were surrounded by a capsule permeable only to within the cells and the bacterial cell becomes more alkalis. Ziehl, however, showed that they could be permeable, allowing easy passage of soluble pro- stained with an acidic dye or 2% alcoholic methyl ducts as shown by the increased permeability of bac- violet solution in carbolic acid water.3 Later, Ziehl terial enzymes.'3 Prolonged exposure results in dis- http://jcp.bmj.com/ used carbolic magenta. Neelsen increased the ruption and destruction of the cells. strength of phenol in the stain,4 and the Ziehl- We describe a new rapid approach to the Ziehl- Neelsen method became the standard one for gen- Neelsen stain using a microwave oven for the stage eral use. of heating in carbolic magenta. The procedure also The property of acid fastness seems to be due to eliminates the need to treat histopathological sec- the presence in the bacilli of unsaponifiable wax,5 tion with xylol and alcohol and sheds some light on first isolated by Aronson in 1898.6 It was referred to react with the way in which acid fast organisms on October 2, 2021 by guest. Protected as one of the higher alcohols by Bulloch and Mac- stains. leod.' Tamura isolated an alcohol from mycobac- teria that had the property of acid fastness and named it "mycol."8 It was at first regarded as an Material and methods unsaturated fat but is now generally believed to be lipoidal or waxy in nature. Not everyone agreed on MICROWAVE OVEN the simple chemical explanation of acid fastness. A domestic oven (Sharp model 9410E) with a max- Sordelli and Arena, for example, stated that it imum output of 640 watts, four power settings, and depended on a semipermeable membrane around a digital timer was used. the organism that allowed magenta to diffuse inwards but did not allow acid magenta to diffuse SPECIMENS outwards.9 The degree of acid fastness of an Clinical smears of sputum from patients with pul- monary tuberculosis and from non-tuberculous con- Accepted for publication 24 April 1985 trols were prepared in the normal way and dried in 1073 J Clin Pathol: first published as 10.1136/jcp.38.9.1073 on 1 September 1985. Downloaded from

1074 Hafiz, Spencer, Lee, Gooch, Duerden air. Histological sections of lung from patients with organisms, the limited number of clinical slides were tuberculosis and patients with infections with stained by the following methods: one set of sputum Nocardia asteroides prepared in paraffin wax were smears was stained by the standard Ziehl-Neelson provided by the department of histopathology. method; a duplicate set was flooded with carbolic magenta, placed in the microwave oven for 30 sec- ORGANISMS onds at full power, decolourised with acid alcohol, Pure cultures of M tuberculosis, non-acid fast bac- and counter stained with malachite green. One set of teria including Bacillus spp, clostridia, corynebac- histological sections from patients with tuberculosis teria, staphylococci, streptococci, Haemophilus spp, was stained by the standard Ziehl-Neelson method. enterobacteriaceae, and neisseriae, N asteroides A second set was flooded with carbolic magenta NCTC 11293, andActinomyces israeli NCTC 10236 without having been treated with xylol and graded were cultured on recommended media and grown in alcohol, placed in the microwave oven for 30 sec- optimum conditions. Smears were made in the onds at full power, decolourised with acid alcohol, normal manner and dried in air. Pure cultures of M and counterstained with malachite green. The third tuberculosis were mixed with other organisms and set was treated as the second, except that the smears made. primary stain used was saffron. Sections of lung from a patient with nocardia STAINS FOR MYCOBACTERIA infection were flooded with carbolic magenta, Carbolic magenta, magenta, and saffron were used placed in the microwave oven for 30 seconds at full for initial staining. Carbolic magenta was prepared power, decolourised with acid alcohol, and counter according to the method described by Cowan and stained with malachite green. Steel.4 A solution of saffron (1%) was made in distilled water. Magenta was made by dissolving 1 g magenta in 10 ml absolute alcohol, which was then Results made up to 100 ml with distilled water. Table 1 shows the results with pure cultures, sputum COUNTER STAIN smears, and histological sections. There were no dif-copyright. Malachite green was prepared according to the ferences between the results obtained with the stan- method described by Cowan and Steel.'4 dard Ziehl-Neelsen staining technique and the method based on microwave treatment in terms of STAINING METHODS proof of acid and alcohol fastness. The slides were Standard Ziehl-Neelsen method Smears were cleaner, however, after treatment with microwaves, flooded with carbolic magenta and heated until and there were no crystalline deposits of stain in the steaming with a torch for 10 minutes, washed with smears. Table 1 shows only the results obtained water, decolourised with acid alcohol (hydrochloric when the decolourising agent was acid alcohol. The acid 3% in 95% alcohol), and counter stained with mycobacteria were consistently acid and alcohol http://jcp.bmj.com/ malachite green. fast, unlike the other bacteria examined. Microwave method Smears were flooded with Table 2 records the effects of different decolouris- carbolic magenta, placed in a sandwich box with the ing solutions. Only Mycobacterium spp were resis- lid on, and exposed to microwave irradiation for 30 tant to decolourising by acid alcohol or 20% sul- seconds at full power (640 watts). They were then phuric acid. N asteroides was resistant to 1% sul- decolourised with acid alcohol and counterstained phuric acid and about 50% of cells in the smears with malachite green. were resistant to 5% sulphuric acid. A israeli and on October 2, 2021 by guest. Protected A series of 15 sets of slides was made from the other non-acid fast species were decolourised by 1 % cultured organisms. One set was stained by the stan- sulphuric acid. Identical staining reactions were dard Ziehl-Neelsen method, two sets were stained achieved by substituting magenta or 1 % saffron for with magenta or saffron, heated by the conventional carbolic magenta and exposing the smears to mic- torch method, decolourised by acid alcohol, and rowave energy, although the acid and alcohol fast counter stained by malachite green, four sets were species were not stained by these particular stains flooded with one of three stains (carbolic magenta, when conventional heat was used. magenta, or saffron) and exposed to microwave The staining of sections of tissue (without dewax- irradiation in the oven for 30 seconds at full power; ing) by the microwave method with carbolic one set treated with each stain was then decolour- magenta or magenta was comparable with the ised by 1%, 5%, and 20% sulphuric acid and by acid results obtained by the conventional Ziehl-Neelsen alcohol. method (after dewaxing), but 1% saffron was not On the basis of the results with pure cultures of suitable for histological sections as the dye was not J Clin Pathol: first published as 10.1136/jcp.38.9.1073 on 1 September 1985. Downloaded from

Technical method 1075 Table 1 Comparison ofacid and alcohol fast staining by standard Ziehl-Neelson and microwave methods Type ofspecimen No ofspecimens No positive by: examined Ziehl-Neelsen method Microwave method Tuberculous sputum 30 30 30 Non-tuberculous sputum 30 0 0 Nocardia sputum 1 0 0 Cultures of Mycobacterium tuberculosis 30 30 30 Cultures of Mycobacterium bovis 1 1 1 Cultures of Mycobacterium kansasi 1 1 1 Cultures of Nocardia asteroides 1 0 0 Cultures of Actinomyces israeli 5 0 0 Cultures of non-acid fast bacteria 10 0 0 Histological sections of tuberculosis 4 4 4 Histological sections of nocardiasis 2 2 2

retained when the sections were treated with acid is not simply a result of heating but also reflects alcohol. increased permeability of the cell wall. In the second stage the stain probably forms a complex with mycol Discussion or other components of acid fast organisms that is insoluble in acid alcohol and therefore indelible; The microwave method of acid fast staining was as even the exposure of smears of acid fast bacteria to reproducible and reliable as the standard Ziehl- microwave radiation during decolourisation failed to Neelsen staining method. The advantages for the decolourise the organisms, showing that it is not routine diagnostic laboratory over the standard merely a matter of permeability. method were that the slides were cleaner; there were no crystalline deposits on the smears; proces- References copyright. sing of smears was much quicker as the time was reduced by 15 minutes for each batch; no flames Koch R. Berl Klin Wochsenschr 1882;19:221. were required, and thus there was no risk of the 2 Erlich P. Aus em Verein fur innere medizin zu Berlin. Dtsch Med Wochenschr 1882;8:269. slides cracking as a result of direct heat or of fires Ziehl F. Dtsch Med Wochenschr 1882;8:451. starting due to naked flames in the laboratory; and 4John A. Einzweiffeloser Fall von congenitaler Tuberkulose. the smears were safe because mycobacteria are kil- 5Anderson RJ. Chemistry of lipoids of tubercle bacilli. Physiol led by microwave irradiation in 30-60 seconds in the Rev 1932; 12: 166-89. 6 Aronson H. Zur Biologie der Tuberkelbacillen. Ber Klin absence of phenol (unpublished data). The charac- Wochenschr 1898;35:484-6. teristics of decolourisation remained the same, and a Bulloch W, Macleod JJR. The chemical constitution of the tuber- http://jcp.bmj.com/ particular advantage for staining histological sec- cle bacillus. J Hyg 1904; 4: 1-10. tions was that the tedious stages of dewaxing were Tamura S. Zur chemie der Bakterien. Ztschr f Physiol Chem Strassb 1913;87:85-114. eliminated. Sordelli A, Arena A. Interpretacion de la acido resistencia del This study also suggests that the acid fastness is Mycobacterium tuberculosis. Rev Soc Argent de Biol probably a process in two stages. In the first stage, as 1934; 10: 168-71. a result of heating or exposure to microwaves, the 10 Kinyoun JJ. A note on Uhlenhuth's method for sputum examina- stain penetrates the cell wall of acid fast organisms; tion for tubercle bacilli. Am J Public Health 1915;5:867-70. Aubert E. Cold stain for acid fast bacteria. Can J Public Health on October 2, 2021 by guest. Protected penetration of stain during exposure to microwaves 1950;41:31-2.

Table 2 Acid fast staining ofsmears from bacterial cultures by microwave method using different decolourising agents Organisms Acid fastness after decolourisation with: Sulphuric acid Acid alcohol 1% 5% 20% Mycobacterium tuberculosis + + + + Mycobacterium bovis + + + + Mycobacterium kansasi + + + + Actinomyces israeli Nocardia asteroides + +(50%) - - Non-acid fast bacteria J Clin Pathol: first published as 10.1136/jcp.38.9.1073 on 1 September 1985. Downloaded from

1076 Hafiz, Spencer, Lee, Gooch, Duerden 2wHok TI. A simple and rapid cold staining method for acid fast teria. 2nd ed. Cambridge: Cambridge University Press, bacteria. Am Der Resp Dis 1962;85:753-4. 1974:161-3. 3 Spencer RC, Hafiz S, Cook C. Effect of microwave energy on the metabolism of Enterobacteriaceae. J Med Microbiol Requests for reprints to: Dr S Hafiz, Department of Medi- 1985; 19:269-72. cal Miicrobiology, University of Sheffield Medical School, 4 Cowan 5, Steel KJ. Manual for the identification of medical bac- Beech Hill Road, Sheffield So10 2RX, England.

During the course of chemotherapy she hidden leukaemic deposits or foci of infec- Letters developed a fever, but no causative organ- tion. ism was identified. Metronidazole (a total M WIENBRENt Measurements for haemoglobin of 18'5 g) was given intravenously over the RM PERINPANAYAGAMt next 12 days in combination with a penicil- L CAMBAt During the meeting of the International lin and gentamicin. The induction treat- CA LEEt Committee for Standardisation in ment was repeated on 17 December for a Department of Bacteriologyt, Haematology (ICSH) secretariat it was further week. Ten days later she became and the Department ofHaematologyt, noted that an increasing number of papers feverish again and Staphylococcus epider- Queen Mary's Hospital, published in both monographs and journals midis was isolated from blood cultures. Roehampton Lane, are now using the World Health Organisa- Treatment with fucidin and netilimicin was London SWIS SPN tion International Committee for Standard- started and there was an initial improve- References isation in Haematology/International So- ment, although the temperature never set- ciety of Haematology recommendation tled. Metronidazole (a total of 10 g) was 'Frytak S, Moertel CG, Childs DS, Albers JW. to express haemoglobin as gram/litre (g/l). added with the onset of diarrhoea because Neurologic toxicity associated with high dose Furthermore, haemoglobinometers, which pseudomembranous colitis was suspected. metronidazole therapy. Ann Intern Med use this scale, are being introduced. A few days later the patient became 1978;88:361-2. The ICSH secretariat would like to point acutely unwell, had rigors, and the paren- 2 Kusumi RK, Plouffe JF, Wyatt RH, Fass RJ. associated a teral antibiotics were changed to ery- Central nervous system toxicity out the need for rationalisation of the with metronidazole therapy. Ann Intern Med expression of haemoglobin results that thromycin, rifampicin, and ceftazidime. 1980;93:59-60. would conform with the principles of SI There was dramatic improvement and after 3Halloran TJ. Convulsions associated with high units, and which should be encouraged as five days all antibiotics were stopped. cumulative doses of metronidazole. Drugcopyright. quickly as possible. The patient was clinically well, without Intell Clin Pharm 1982; 16:409. RL VERWILGHEN fever, and ready for discharge when she 4Scharer K. Selective injury to Purkinje cells in University Hospital, was found on the floor having a grand mal the dog after oral administration of high- Herestaat, 49, fit. Over the next 24 hours she had several dose metronidazole derivative. (Original in B 3000 Leuven, more fits despite treatment with chlor- German.) Verh Dtsch Ges Pathol 1972;56: Belgium. methiazole and phenytoin. A computed 407-10. (Will authors please note. Editor) tomographic scan showed no focal lesion. Enzyme markers in acute non-lymphoid An electroencephalogram showed leukaemia encephalopathy. The urea and electrolyte concentrations sampled 30 minutes before We read with interest the recent review inhttp://jcp.bmj.com/ Convulsions and encephalopathy in a the first fit were normal. The platelet count the journal by Dr Drexler et al.' It is stated patient with leukaemia after treatment with had not fallen below 60 x 109/l. that "the main group AML did not show metronidazole The patient made a recovery to normal one distinct enzyme marker phenotype". within 48 hours, but during this period she While this may be true for the purine sal- Metronidazole is an antimicrobial agent suffered some retention of urine. A repeat vage enzymes, we would like to draw atten- commonly used to treat anaerobic bacterial electroencephalogram four weeks later tion to the use of certain lysosomal infections. We consider that the side effect showed moderate abnormality. At this enzymes in the diagnosis and classification of convulsions resulting from treatment time the patient was in complete remission of acute myeloid leukaemias in both adults2 on October 2, 2021 by guest. Protected with metronidazole' may often be unrec- awaiting consolidation of treatment. and children.3 We have now studied 57 ognised, particularly as a consequence of The patient received a total of 29 g patients with acute non-lymphoid cumulative exposure23 in a patient receiv- metronidazole and developed epilepsy, leukaemia and have measured the activities ing many other drugs and suffering many encephalopathy, and, probably, sensory and isoenzyme profiles for the enzymes complications of treatment. We describe a neuropathy manifested by retention of f3 hexosaminidase and a mannosidase. A patient with acute myeloid leukaemia, who urine. Acute encephalopathy and epilepsy summary of our results is shown in the developed epileptiform seizures, secondary to metronidazole probably table. Patients with AML, AMMOL, and encephalopathy, and sensory neuropathy resulted from a high cumulative dose.3 It is AMOL showed significant increases in after treatment with metronidazole. known that the drug crosses the blood these enzyme activities when compared A 43 year old woman presented with brain barrier; in dogs it has been shown with peripheral blood granulocytes or acute myeloid leukaemia (FAB that high doses of metronidazole induce leukaemic cells of lymphoid origin. classification M4) in December 1984. She Purkinje cell lesions.4 It is especially impor- Moreover, in patients with AML in particu- was treated with Cytarabine 100 mg/M2 tant to be aware of this side effect in the lar there was a reduction in f hex- intravenously for seven days and leukaemic patient to avoid unnecessary osaminidase B isoenzyme peak when com- daunorubicin 50 mg/M2 for three days. invasive investigations in the search for pared with the A component.