Diagnosis of Disease Caused by Mycobacterium avium Complex* Michio Tsukamura M.D., Ph.D. Isolation of Mycobacterium amum complex from sputum specimens in association with the appearance of a new cavitary (or in6Itrative) lesion was studied in 299 patients from whom the organism was isolated one or more times. Of the patients studied, 114 showed only single isolation. Of these 114, only two patients (2 percent) had association with appearance of a cavitary lesion. Of 29 patients who showed two isolations, 26 (90 percent) had the association. Of 40 patients who showed three isolations, 39 (98 percent) had the association. All 116 patients who showed four or more isolations had the association with appearance of a cavitary lesion. Accordingly, of a total of 185 patients who showed two or more isolations, 181 (98 percent) had the association. Of these 181, 176 (97 percent) showed two or more isolations in the sputum examinations made in the
initial three days. Therefore, the sputum examination in the 6rst three days after onset of disease is most important for the diagnosis ofdisease caused by Mycobacterium avium complex. Since the probability that casual isolation of the organism occurs twice is extremely low, we can make the diagnosis of pulmonary infection caused by this organism by evidence of two or more isolations of the organism in the 6rst few days after the onset of disease, which is associated with appearance ofa new cavitary (or in6ltrative) lesion. Moreover, theoretical consideration made in this study has led us to conclude that patients who have had a single isolation ofthe organism together with a new cavitary lesion should be regarded as having an infection. (Cheat 1991; 99:667-69)
criteria currently in use for mycobacteD iagnostic rial infection demonstrate repeated isolation of
this condition was never observed in patients with tuberculosis. Since it is obviously beneficial if we can diagnose the infection by minimum examination, we have investigated this condition further.
the same species of mycobacteria from sputum specimens. l-3 For example, the criteria of Yamamoto et all demonstrate, as minimum condition, four isolations, each consisting of more than 100 colonies. However, the necessity of repeated isolation does not seem to be based on theoretical consideration, but merely on an assumption that the more frequent isolation is surer. On the other hand, we estimated actually the frequency of isolation of mycobacteria in tuberculous patients in whom the frequency of casual isolation should be high as 0.008 per examination. Compared with this frequency, two or more isolations in three to ten examinations are statistically significantly higher and are regarded as an extremely unusual event. Coexistence ofthis unusual event with another unusual event, appearance of a cavitary lesion, was considered to show the two events correlated. Thus, we proposed new criteria. 4.5 The diagnostic criteria in the presence of a new cavitary (infiltrative) lesion is to demonstrate two or more isolations of the same species of mycobacteria in the initial three examinations made daily or monthly in the first three months. The criteria in the presence of a cavitary lesion with sclerotic process are demonstrated by three or more isolations in monthly sputum examinations in a period of six months, with at least one of these isolates having more than 100 colonies; *From the Department of Microbiology, Fujita Health University School of Medicine, Toyoake, Aichi, and The National Chubu Hospital, Obu, Aichi, Japan Manuscript received July 9; revision accepted September 24. Reprint requests: Dr: Tsukamura, NatioRal Chubu Hospital, Obu, Aichi, japan 4747
MATERIALS AND METHODS
Patients from whom Mycobacterium avium- Mycobacterium intraceUulare complex (Mycobacterium avium complex or MAl complex) was isolated from their sputum specimens in a period from 1971 to 1988 (18 years) were the subjects of the study. All patients were hospitalized in the National Chubu Hospital for a period of six months to more than ten years. If the period of observation in the outpatient clinic of this hospital was included, the period of observation of all patients was more than 12 months. The patients who showed single isolation were restricted to those observed from 1981 to 1988. The sputum examination was carried out daily in the first three days after hospitalization and thereafter monthly. One morning sputum specimen was added with equal volume of 4 percent or 2 percent NaOH solution and shaken in room temperature to completely dissolve. Two loopfuls of the dissolved sputum were inoculated onto Ogawa egg medium (before 1978) or Tween 80 egg medium6 (after 1979) by spiral loop that can deliver a O.02-ml sample by each inoculation. This method of cultivation was a little superior in the positive rate to the standard method, in which LOwenstein-Jensen medium was inoculated by pipette after neutralization. 7 Identification of mycobacteria isolated was made according to the methods described previously. It RESULTS
One hundred fourteen patients showed a single isolation of Mycobacterium avium complex during a period of observation for 12 months or more. The amount of colonies in these 114 patients is shown in Table 1. It was divided into two categories, one less than 50 colonies and another more than 100 colonies. All the patients who showed a single isolation had CHEST I 99 I 3 I MARCH, 1991
687
Table I-lNtribution of Colony Number of Single I.alata of Mycobacterium avium Complnfrom Sputum SpecimefIIJ No. of Colonies
No. of Patients with Single Isolate
1-5 6-10 11-20
Frequency of Isolation
45 17 15 9
21-50
Table 3-BelationBhip betweera Frequency oIl.alation of Mycobacterium avium Complex arad Appearance of a Cooita'll or Infiltrative Leaion in lbtienta
1 2
o
51-100 >100 Total
3
28 114
4 ~5
bronchiectasis, emphysema, pneumoconiosis, fibrosis, pyothorax, or cancer in chest roentgenographic examinations. In these 114 patients with a single isolation, association with appearance of a new cavitary lesion occurred in only two patients (2 percent). These two showed isolation of more than 100 colonies (Table 2). Twenty-nine patients showed two isolations. In these patients, association with a new cavitary or infiltrative lesion was observed in 26 (90 percent). Another 40 patients showed three isolations of the organism. In these, association was observed in 39 (98 percent). One hundred sixteen patients showed four or more isolations. All of these patients had association with a new cavitary lesion (Table 3). In a total of 185 patients who showed two or more isolations, 181 (98 percent) showed association with appearance of a new cavitary or infiltrative lesion. In these 181 patients who were considered to have infection caused by Mycobacterium avium complex, 176 (97 percent) showed two or more isolations in the initial three examinations (Table 4). Five patients did not show their isolates in the initial three examinations. They had shown caseous or infiltrative lesions on their chest roentgenograms in the year of observation. However, they later showed cavitary lesions in their follow-up examinations. cavi~
DISCUSSION
In this study, two or more isolations were, in most cases (181/185, 98 percent), associated with appearance of a new cavitary or infiltrative lesion. In these 181 patients with two or more isolations associated with the appearance of a lesion, 176 (97 percent) had such isolation in the initial three examinations. There-
No. of Patients 114 29 40 11 105 299
No. (%) Patients with Appearance of Cavitary or Infiltrative Lesion 2 26 39 11 105
183
(2) (90) (98) (100) (100)
fore, evidence of two or more isolations in the initial period of hospitalization (the onset of disease) is most important for diagnosing the disease. In other words, almost all patients with two or more isolations in the first several days of hospitalization or with the onset of symptoms have infection by mycobacteria. Previousl~ a single isolation was not regarded as a sign of infection. This is probably based on consideration of contamination from environmental mycobacteria or on suspecting colonization of the respiratory tract. However, casual isolation should not be accompanied by the appearance of a new lesion, and it should be without clinical symptoms, even if the casual isolation occurred occasionally in patients with tuberculosis or other respiratory diseases. 9 We showed that the casual isolation occurs at a rate of 0.008 per examination in patients with pulmonary tuberculosis, who have much high chance of contamination or colonization by environmental mycobacteria. 4 ,5 Casual isolation occurs rarely. On the other hand, appearance ofa new cavitary lesion also is a very rarely occurring, unusual event. If these two unusual events occur in the same patient at the same time, the probability that these two events appear independently in the same person should be extremely lo~ It is therefore more probable that these two events correlate with each other than that these two are independent. The correlation of these two events means that the lesion is caused by the mycobacteria isolated. Accordingl~ if any single isolation occurs together with the appearance of a new lesion, we may consider that infection exists. To realize this consideration, it is essential to avoid contamination ofenvironmental mycobacteria in the course of cultivation. For this purpose, it is
Table 2-PatientB with Single I.olation of Mycobacterium avium Compln Aaociated with Appearance of a New Caoitary Leaion* CaselAge, yr/Sex
l/73IM
Chest Roentgenographic Finding
Mode of Excretion of Mycobacteria Smear Culture Smear Culture
v
+
Fresh cavity with pericavitary infiltration
II
IV +
*Roman numeral letters in the smear show the Gaftky numbers. The symbol medium.
Middle-sized, thin-walled cavity
+
in the culture shows more than 100 colonies on isolation
DIagnoIIe of DIee888 C8u8ed by II avlum Complex (Ilichio Tsuk."""ra)
Table 4-Mode a/Isolation a/Mycobacterium avium Complex in lbtienta Who Showed Two or More Isolates in Initial Three Sputum Emminationa ABsociated with Appearance 0/ a Cavitary or Infiltrative uaion Frequency of Isolation That Occurred in Observation Period, No. ofTImes
No. of Patients
No. (%) of Patients in Whom Two or More Isolations Occurred in Initial Three Sputum Examinations
26 39 11
2
3 4 ~5
Total
lOS
181
23
37 11 105 176
(88)
(95) (100) (100) (97)
essential to sterilize the vessels that are used for taking the sputum and to cultivate the sample immediately after taking the sputum specimen. REFERENCES 1 Yamamoto M, Ogura Y, Sudo K, Hibino S. Diagnostic criteria for disease caused by atypical mycobacteria. Am Rev Respir Dis 1967; 98:773-78
2 American Thoracic Society. Diagnostic standards and classification of tuberculosis and other mycobacterial diseases. 14th ed. Am Rev Respir Dis 1981; 123:343-58 3 Abn CH, McLarty J~ Abn SS, Abn SI, Hurst GA. Diagnostic criteria for pulmonary disease caused by Mycobacterium kansasii and Mycobacterium intraceUulare. Am Rev Respir Dis 1982; 125:388-91 4 Tsukamura M. A trial of standarization of diagnosing lung disease due to mycobacteria other than tubercle bacilli. Keldcaku 1978; 53:367-76 5 Tsukamura M. Diagnosis of nontuberculous mycobacteriosis. In Casal M, ed. Mycobacteria of clinical interest. Amsterdam, Netherlands: Elsvier Science Publishers; 1986; 251-63 6 Tsukamura M, Toyama H, Fukaya Y. lween egg medium' for isolating mycobacteria from sputum specimens. Microbiol Immuno11979; 23:833-38 7 Tsukamura M, Toyama H, Fukaya Y. Spiral loop-inoculation for isolating mycobacteria from sputum specimens. Iryo 1979; 33:50913 8 Tsulcamura M. Numerical classification of 280 strains of slowly growing mycobacteria. Microbiol Immunoll983; 27:315-24 9 Tsukamura M. Background factors for casual isolation of Mycobacterium intraceUulare from sputum of patients with tuberculosis. Am Rev Respir Dis 1973; 108:679-83
Plan to Attend ACCP·s
57th Annual Scientific Assembly San Francisco November 4-8, 1991
CHEST I 99 I 3 I MARCH. 1991
881
initial three days. Therefore, the sputum examination in the 6rst three days after onset of disease is most important for the diagnosis ofdisease caused by Mycobacterium avium complex. Since the probability that casual isolation of the organism occurs twice is extremely low, we can make the diagnosis of pulmonary infection caused by this organism by evidence of two or more isolations of the organism in the 6rst few days after the onset of disease, which is associated with appearance ofa new cavitary (or in6ltrative) lesion. Moreover, theoretical consideration made in this study has led us to conclude that patients who have had a single isolation ofthe organism together with a new cavitary lesion should be regarded as having an infection. (Cheat 1991; 99:667-69)
criteria currently in use for mycobacteD iagnostic rial infection demonstrate repeated isolation of
this condition was never observed in patients with tuberculosis. Since it is obviously beneficial if we can diagnose the infection by minimum examination, we have investigated this condition further.
the same species of mycobacteria from sputum specimens. l-3 For example, the criteria of Yamamoto et all demonstrate, as minimum condition, four isolations, each consisting of more than 100 colonies. However, the necessity of repeated isolation does not seem to be based on theoretical consideration, but merely on an assumption that the more frequent isolation is surer. On the other hand, we estimated actually the frequency of isolation of mycobacteria in tuberculous patients in whom the frequency of casual isolation should be high as 0.008 per examination. Compared with this frequency, two or more isolations in three to ten examinations are statistically significantly higher and are regarded as an extremely unusual event. Coexistence ofthis unusual event with another unusual event, appearance of a cavitary lesion, was considered to show the two events correlated. Thus, we proposed new criteria. 4.5 The diagnostic criteria in the presence of a new cavitary (infiltrative) lesion is to demonstrate two or more isolations of the same species of mycobacteria in the initial three examinations made daily or monthly in the first three months. The criteria in the presence of a cavitary lesion with sclerotic process are demonstrated by three or more isolations in monthly sputum examinations in a period of six months, with at least one of these isolates having more than 100 colonies; *From the Department of Microbiology, Fujita Health University School of Medicine, Toyoake, Aichi, and The National Chubu Hospital, Obu, Aichi, Japan Manuscript received July 9; revision accepted September 24. Reprint requests: Dr: Tsukamura, NatioRal Chubu Hospital, Obu, Aichi, japan 4747
MATERIALS AND METHODS
Patients from whom Mycobacterium avium- Mycobacterium intraceUulare complex (Mycobacterium avium complex or MAl complex) was isolated from their sputum specimens in a period from 1971 to 1988 (18 years) were the subjects of the study. All patients were hospitalized in the National Chubu Hospital for a period of six months to more than ten years. If the period of observation in the outpatient clinic of this hospital was included, the period of observation of all patients was more than 12 months. The patients who showed single isolation were restricted to those observed from 1981 to 1988. The sputum examination was carried out daily in the first three days after hospitalization and thereafter monthly. One morning sputum specimen was added with equal volume of 4 percent or 2 percent NaOH solution and shaken in room temperature to completely dissolve. Two loopfuls of the dissolved sputum were inoculated onto Ogawa egg medium (before 1978) or Tween 80 egg medium6 (after 1979) by spiral loop that can deliver a O.02-ml sample by each inoculation. This method of cultivation was a little superior in the positive rate to the standard method, in which LOwenstein-Jensen medium was inoculated by pipette after neutralization. 7 Identification of mycobacteria isolated was made according to the methods described previously. It RESULTS
One hundred fourteen patients showed a single isolation of Mycobacterium avium complex during a period of observation for 12 months or more. The amount of colonies in these 114 patients is shown in Table 1. It was divided into two categories, one less than 50 colonies and another more than 100 colonies. All the patients who showed a single isolation had CHEST I 99 I 3 I MARCH, 1991
687
Table I-lNtribution of Colony Number of Single I.alata of Mycobacterium avium Complnfrom Sputum SpecimefIIJ No. of Colonies
No. of Patients with Single Isolate
1-5 6-10 11-20
Frequency of Isolation
45 17 15 9
21-50
Table 3-BelationBhip betweera Frequency oIl.alation of Mycobacterium avium Complex arad Appearance of a Cooita'll or Infiltrative Leaion in lbtienta
1 2
o
51-100 >100 Total
3
28 114
4 ~5
bronchiectasis, emphysema, pneumoconiosis, fibrosis, pyothorax, or cancer in chest roentgenographic examinations. In these 114 patients with a single isolation, association with appearance of a new cavitary lesion occurred in only two patients (2 percent). These two showed isolation of more than 100 colonies (Table 2). Twenty-nine patients showed two isolations. In these patients, association with a new cavitary or infiltrative lesion was observed in 26 (90 percent). Another 40 patients showed three isolations of the organism. In these, association was observed in 39 (98 percent). One hundred sixteen patients showed four or more isolations. All of these patients had association with a new cavitary lesion (Table 3). In a total of 185 patients who showed two or more isolations, 181 (98 percent) showed association with appearance of a new cavitary or infiltrative lesion. In these 181 patients who were considered to have infection caused by Mycobacterium avium complex, 176 (97 percent) showed two or more isolations in the initial three examinations (Table 4). Five patients did not show their isolates in the initial three examinations. They had shown caseous or infiltrative lesions on their chest roentgenograms in the year of observation. However, they later showed cavitary lesions in their follow-up examinations. cavi~
DISCUSSION
In this study, two or more isolations were, in most cases (181/185, 98 percent), associated with appearance of a new cavitary or infiltrative lesion. In these 181 patients with two or more isolations associated with the appearance of a lesion, 176 (97 percent) had such isolation in the initial three examinations. There-
No. of Patients 114 29 40 11 105 299
No. (%) Patients with Appearance of Cavitary or Infiltrative Lesion 2 26 39 11 105
183
(2) (90) (98) (100) (100)
fore, evidence of two or more isolations in the initial period of hospitalization (the onset of disease) is most important for diagnosing the disease. In other words, almost all patients with two or more isolations in the first several days of hospitalization or with the onset of symptoms have infection by mycobacteria. Previousl~ a single isolation was not regarded as a sign of infection. This is probably based on consideration of contamination from environmental mycobacteria or on suspecting colonization of the respiratory tract. However, casual isolation should not be accompanied by the appearance of a new lesion, and it should be without clinical symptoms, even if the casual isolation occurred occasionally in patients with tuberculosis or other respiratory diseases. 9 We showed that the casual isolation occurs at a rate of 0.008 per examination in patients with pulmonary tuberculosis, who have much high chance of contamination or colonization by environmental mycobacteria. 4 ,5 Casual isolation occurs rarely. On the other hand, appearance ofa new cavitary lesion also is a very rarely occurring, unusual event. If these two unusual events occur in the same patient at the same time, the probability that these two events appear independently in the same person should be extremely lo~ It is therefore more probable that these two events correlate with each other than that these two are independent. The correlation of these two events means that the lesion is caused by the mycobacteria isolated. Accordingl~ if any single isolation occurs together with the appearance of a new lesion, we may consider that infection exists. To realize this consideration, it is essential to avoid contamination ofenvironmental mycobacteria in the course of cultivation. For this purpose, it is
Table 2-PatientB with Single I.olation of Mycobacterium avium Compln Aaociated with Appearance of a New Caoitary Leaion* CaselAge, yr/Sex
l/73IM
Chest Roentgenographic Finding
Mode of Excretion of Mycobacteria Smear Culture Smear Culture
v
+
Fresh cavity with pericavitary infiltration
II
IV +
*Roman numeral letters in the smear show the Gaftky numbers. The symbol medium.
Middle-sized, thin-walled cavity
+
in the culture shows more than 100 colonies on isolation
DIagnoIIe of DIee888 C8u8ed by II avlum Complex (Ilichio Tsuk."""ra)
Table 4-Mode a/Isolation a/Mycobacterium avium Complex in lbtienta Who Showed Two or More Isolates in Initial Three Sputum Emminationa ABsociated with Appearance 0/ a Cavitary or Infiltrative uaion Frequency of Isolation That Occurred in Observation Period, No. ofTImes
No. of Patients
No. (%) of Patients in Whom Two or More Isolations Occurred in Initial Three Sputum Examinations
26 39 11
2
3 4 ~5
Total
lOS
181
23
37 11 105 176
(88)
(95) (100) (100) (97)
essential to sterilize the vessels that are used for taking the sputum and to cultivate the sample immediately after taking the sputum specimen. REFERENCES 1 Yamamoto M, Ogura Y, Sudo K, Hibino S. Diagnostic criteria for disease caused by atypical mycobacteria. Am Rev Respir Dis 1967; 98:773-78
2 American Thoracic Society. Diagnostic standards and classification of tuberculosis and other mycobacterial diseases. 14th ed. Am Rev Respir Dis 1981; 123:343-58 3 Abn CH, McLarty J~ Abn SS, Abn SI, Hurst GA. Diagnostic criteria for pulmonary disease caused by Mycobacterium kansasii and Mycobacterium intraceUulare. Am Rev Respir Dis 1982; 125:388-91 4 Tsukamura M. A trial of standarization of diagnosing lung disease due to mycobacteria other than tubercle bacilli. Keldcaku 1978; 53:367-76 5 Tsukamura M. Diagnosis of nontuberculous mycobacteriosis. In Casal M, ed. Mycobacteria of clinical interest. Amsterdam, Netherlands: Elsvier Science Publishers; 1986; 251-63 6 Tsukamura M, Toyama H, Fukaya Y. lween egg medium' for isolating mycobacteria from sputum specimens. Microbiol Immuno11979; 23:833-38 7 Tsukamura M, Toyama H, Fukaya Y. Spiral loop-inoculation for isolating mycobacteria from sputum specimens. Iryo 1979; 33:50913 8 Tsulcamura M. Numerical classification of 280 strains of slowly growing mycobacteria. Microbiol Immunoll983; 27:315-24 9 Tsukamura M. Background factors for casual isolation of Mycobacterium intraceUulare from sputum of patients with tuberculosis. Am Rev Respir Dis 1973; 108:679-83
Plan to Attend ACCP·s
57th Annual Scientific Assembly San Francisco November 4-8, 1991
CHEST I 99 I 3 I MARCH. 1991
881
