The Demonstration of Mycobacterium Tuberculosis in Anorectal Abscess

An abundant literature is available reporting various diagnostic laboratory procedures for the demonstration of Mycobacterium tuberculosis in fistula in ano. Recently Buie, Smith, and Jackman reviewed the subject calling attention to the almost universal failure to demonstrate the organism by direct staining methods. They recommended guinea pig inoculation. Quite recently Joynt confirmed the general conclusions of previous writers and employed culture methods in place of the guinea pig. In relation to the antecedent condition, anorectal abscess, demonstration of the organism by direct microscopical methods has not been reported with any important degree of success. In institutions where patients are continuously under observation we may anticipate the spontaneous rupture of these abscesses by aspirating the contents. During the past 9 years we have successfully aspirated 16 of a total of 20 abscesses under observation. 

Clinical Material

This report is concerned with the investigation of 22 abscesses from 19 patients. All patients furnishing material for examination had positive sputum except 1 whose pulmonary history was unknown, as the specimen was sent in by an outside physician. One patient had 3 distinct abscesses form over a period of 1 year. Another had 2 abscesses during a 4 months’ period. 

Sixteen abscesses were successfully aspirated. Rupture during preparation for aspiration occurred 4 times and spontaneous rupture occurred from 1 week to 4 months before investigation in 2 instances. 

Procedure

The syringe and needle No. 16 were sterilized at 220 degrees C. dry heat for 1 hour. The investigation of the material was made with the dilution-flotation procedure,³ and guinea-pig inoculation. All animals were tested before they were used with 0.1 cubic centimeter purified protein derivative second strength. The same dose was given as a diagnostic test on the twentieth day after inoculation.

The pus diluted 10 times with 0.5 per cent sodium hydroxide was shaken in a strong shaker for 5 minutes and set at 37 degrees for 1 hour, or in case contaminating organisms were present, for 2 hours. Thorough homogenization was accomplished by 10 minutes shaking. The centrifuge was employed at 1000 revolutions per minute for 2 minutes, to throw down undigested tissue fragments. The upper third of the solution was pipetted off, placed at 55 degrees C. for 30 minutes to complete the digestion process and used for dilution-flotation. The final dilution with distilled water was made so that the amount of the original material contained in the fraction would be diluted 100 times. (On account of the high protein content of this type of material it is necessary to dilute to a much greater degree than is necessary in the case of sputum.) From 0.5 to 1 cubic centimeter of xylol was added, and the mixture shaken for 10 minutes. Films were made from the xylol layer by means of a capillary pipette. The exact number of organisms found in a 10 minute search was recorded if present to the extent of one in two fields or less. 

For guinea-pig inoculation the remaining two thirds of the digested specimen was employed. It was neutralized with 3 per cent hydrochloric acid in the presence of brom-thymol-blue, and injected subcutaneously in divided dosage in 2 or more regions. (If all the inoculum is given in one region, pyogenic ulceration may occur early and expel the organisms before infection takes place. Ulceration of all inoculation sites in case of divided dosage is a rare event.) We had but 1 early pyogenic abscess in the series. The animal, however, was positive. In 9 animals the inoculation was made intraperitoneally. Not a single animal was lost by death due to other causes. 

Results of Investigation

Of 20 abscesses and 2 recently formed sinuses investigated 4 showed from 2 to 10 organisms per field, so that confirmation by inoculation was not needed. The results obtained by dilution-flotation and guinea-pig inoculation of material obtained from the remaining 18 abscesses are tabulated in Table I. The table presents 3 groups of results. The first consists of 13 successfully aspirated abscesses averaging 2.66 cubic centimeters. The second consists of 4 abscesses which ruptured into the anal canal under pressure from the speculum resulting in much loss of material and averaging 0.49 cubic centimeter of material recovered. The third group represents a single specimen in which a very small amount of serous discharge was obtained from a fistulous tract which had formed 4 months previously.

 

Table I.—Findings by Dilution-Flotation and Guinea-Pig Inoculation in 18 Anorectal Abscesses and Recently Formed Fistulas in 14 Known Tuberculosis Patients 

Specimens	No. of examinations	Average amount c.cm.	D.F. + G.P. +	D.F. –   G.P. –	D.F. – G.P. +	D.F. + G.P. –
Aspirated	13	2.66	12	1	0	0
Ruptured at operation	4	0.49	2	1	1	0
Ruptured 4 mos. before investigation	1	0.01	0	0	1	0
Totals	18		14	2	2	0

D.F., dilution-flotation 

G. P., guinea-pig. 

 

In the first group of successful aspirations dilution-flotation and guinea-pig inoculation were in 100 per cent agreement, 12 positive and 1 negative result. In the second group of 4 abscesses rupturing under pressure, agreement resulted in 3 instances, 2 positive and 1 negative; and in the fourth instance in which only 0.05 cubic centimeter of material was obtained, dilution-flotation was negative but the guinea-pig was positive. In the third group dilution-flotation was negative and the guinea-pig was positive.

The total of the 3 groups shows 18 specimens in which dilution-flotation and inoculation were in agreement 16 times. Fourteen specimens were positive and 2 negative. It should be noted that the 2 failures by dilution-flotation occurred with materials small in amount, following spontaneous rupture.

Table II.—Range of the Number of Organisms Found in 18 Specimens Positive by Dilution-Flotation 

Times Found
1 in 2 fields to 10 per field	8
Organisms found in 10 minute search
15 to 79	1
1 to 14	91

 

Elsewhere^3,4 the relative sensitivity of homogenized smears counterstained with strong methylene blue, with picric acid, and the same material treated by dilution-flotation counterstained with picric acid is given as 1, 5.5, and 80, respectively. Applying these ratios to Table II, 8 of the 18 specimens, or 44.4 per cent should be positive by Ziehl-Neelsen with methylene blue counterstain. If picric acid is used, another specimen should be positive, making 9 positives in a total of 18 or 50 per cent. Nine specimens should be found positive only by dilution-flotation. This is the group in which the organisms are so rare that they may be found by Ziehl-Neelsen only by prolonged search of an hour or more. Even with the 15-fold enrichment attained by dilution-flotation only 4 organisms were found in 1 specimen in 1I hour’s search. In the remaining specimens the organisms were found during the first 5 minutes’ search. Ziehl-Neelsen counterstained with picric acid was carried out in 7 of the 18 specimens, 4 of which, or 57.1 per cent, were positive. This percentage agrees well with the probable percentage deduced from the known sensitivity ratios.

The actual number of organisms found in the 9 specimens per cubic centimeter of material is small. Corper established about 100, 000 organisms per cubic centimeter as the lower limit of sensitivity of the Ziehl-Neelsen method and since dilution-flotation with picric acid counterstain is 80 times more sensitive, a single bacillus found in 10 minutes’ search would indicate 100,000 / 80 = 1250 organisms per cubic centimeter. The number of organisms, 1 to 14, found in a 10 minute search would indicate the presence of 1,250 to 17,500 organisms per cubic centimeter of original material investigated.

Evaluation of Study

In view of the unprecedented success with which we have demonstrated Mycobacterium tuberculosis in this series of abscesses, we wish to point out the causes of failure of earlier workers. First, the selection of the optimum time for aspiration has not been duly considered. It is well established that the number of organisms in tuberculous lesions increase greatly at the time of caseation and liquefaction. Following removal of the abscess contents by spontaneous rupture, the organisms decrease rapidly so that their demonstration is difficult. Second, an adequate enrichment and a light counterstain is absolutely necessary for maximum results.

Undoubtedly many, perhaps most, of these abscesses rupture spontaneously before the surgeon is consulted, thus the best conditions for direct and immediate investigation have disappeared.

It is evident that a method that can give a definite diagnosis within 2 hours after the specimen is received by the laboratory offers a distinct advance over former practice. In 1 of these cases a negative result by dilution-flotation led us to refrain from surgical procedure. The abscess healed spontaneously and 2 months later the guinea pig confirmed the negative findings.

Summary and Conclusion

A comparison of the results obtained by dilution-flotation and guinea pig inoculation in the demonstration of Mycobacterium tuberculosis in 18 anorectal abscesses is presented.

The two methods were in 100 per cent agreement in 13 abscesses aspirated by needle. Twelve were positive and 1 was negative by both. Under less favorable conditions in which the abscesses ruptured spontaneously with much loss of material, they were in agreement in result in 3 of 5 abscesses investigated. Two were positive and 1 was negative by both. In the 2 other abscesses dilution-flotation was negative and inoculation was positive. 

Dilution-flotation with picric acid counterstain offers a means for the immediate diagnosis of anorectal tuberculosis in the abscess stage, or its probable exclusion.

 

References Cited:

1. Buie, L. A.. Smith, N. D., and Jackman, R. J. The rule of tuberculosis in anal fistula. Surg., Gynec. & Obst., 1939, 68: 191.
2. Joynt, G. H. C. Ano-rectal tuberculosis. Am. Rev. Tuberc., 1940, 41: 760.
3. Pottenger, J. E. The demonstration of rare tubercle bacilli in sputum. Am. Rev. Tuberc., 1931, 24: 583.
4. Idem. Technique of sputum examination. Am. Rev. Tuberc., 1939, 40: 581.