Vol. 10, No. 2
JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 1979, p. 168-174 0095-1137/79/08-0168/07$02.00/0
Biotypes of Haemophilus Encountered in Clinical Laboratories THOMAS R. OBERHOFER* AND ANDREW E. BACK Microbiology Section, Department of Pathology, Madigan Army Medical Center, Tacoma, Washington 98431
Received for publication 14 May 1979
The biochemical characteristics of 464 strains of Haemophilus influenzae and 83 strains of Haemophilus parainfluenzae isolated over an 18-month period are described. Of 22 characteristics obtained, only 6 were necessary to biochemically identify and biotype the isolates. The key substrates or tests were urease, ornithine, indole, o-nitrophenyl-,8-D-galactopyranoside, sucrose, and xylose. Five biotypes of H. influenzae and four of H. parainfluenzae were commonly recognized. Some strains were encountered which could not be accommodated in the recognized taxa but which constituted separate biotypes of the two species. H. influenzae biotype I was recovered principally from blood, cerebrospinal fluid, and upper respiratory secretions, and biotypes II and III were recovered from eye and sputum cultures. Biotype I was recovered primarily from children less than 1 year of age, whereas biotypes II and III were from persons 1 to 5 years old and from those over 20 years of age. Multiple isolates recovered from the same patient were almost always of the same biotype. Strains of H. parainfluenzae were isolated primarily from sputum, with others being isolated from body sources such as dental abscesses, gastric aspirates, and peritoneal fluid. An inverse relationship was noticed between hemolysis and mannose fermentation among H. parainfluenzae biotype III strains, whereas the relationship was absent among the other biotypes.
Traditionally, the identification of Haemophilus is accomplished by testing the requirements for specific growth factors on factor-deficient agar medium or by using the test for satellitism. It is then common to report isolates recovered from clinical sources as Haemophilus influenzae or Haemophilus parainfluenzae, without consideration being given to the relationship of source of isolation to strain characteristics. Until the emergence of beta-lactamase-producing H. influenzae, little recognition had been given to the biochemical identification and differentiation of the haemophilus. Kilian (12, 13) can be credited with the sparking of interest in, and the recognition and rediscovery of, the haemophilus as biochemically active organisms rather than as bacteria of biochemical insignificance. The excellent report on the taxonomic study of the genus (13) clarified the relationships of the species and showed the practical usefulness of biochemical differentiation within the genus. Since then, biotyping of H. influenzae has been used to describe the isolates recovered as causative agents of diffuse bronchopneumonia (20) and bacteremia (1, 5). This study was undertaken to define the biochemical behavior of Haemophilus isolates re-
covered from routine clinical specimens. Each was subjected to a large number of tests to identify it to the species level, to delineate distinct biotypes of each species, and to ascertain the distribution of biotypes in relation to the source of isolate and age of patient. Insofar as possible, all isolates were nonselectively included in the study in order to accurately reflect the normal incidence of the various species or
biotypes. MATERIALS AND METHODS Test organisms. The organisms included in this study were isolated over an 18-month period in the Microbiology Laboratory, Madigan Army Medical Center. Isolates were recovered using chocolate agar (5% sheep blood in Mueller-Hinton base) supplemented with 1% IsoVitaleX (BBL Microbiology Systems) and 5% sheep blood agar containing a staphylococcus steak for satellite growth. Incubation was at 35°C in 5 to 7% C02. Growth factor requirements. Isolates were lightly suspended in 2 to 3 ml of sterile saline to produce a turbidity equivalent to a 0.5 MacFarland standard, to reduce any carry-over effect of nutrients from the medium (25). A swab moistened with the suspension was streaked in three directions on a Trypticase soy agar (BBL) plate. Each plate was dried for approximately 15 min, after which X-, V-, and XV-
168
VOL. 10, 1979
BIOTYPES OF HAEMOPHILUS
169
factor disks (Difco) were placed on the plates and nitrate reduction. Subsequently, it was found that firmly pressed to the agar. Disks were spaced so that suitable results were obtained after 18 to 24 h without factors were not mutually contributory. Plates were loss of sensitivity. incubated overnight under CO2 and examined for Classification. The classification of the Haemogrowth around the individual disks. Satellite growth philus isolates closely followed the scheme of Kilian was determined using the beta-toxigenic (ATCC (13). Divergence was necessary when information was 25923) strain of Staphylococcus aureus as the V-factor lacking. The serotype of each isolate was taken into feeder. Tests for C02 requirements were not performed account during the classification attempts. routinely. RESULTS Serotyping. Encapsulated isolates of H. influenzae were serotyped by slide agglutination, using specific (i) H. influcharacteristics. Biochemical antisera to types a through f (Burroughs-Wellcome). enzae. The biochemical features of 464 fresh by determined was activity Hemolytic Hemolysis. streaking the isolates on sheep blood agar and then clinical isolates of H. influenzae are presented making a straight-line cross-streak with the S. aureus in Table 1. All strains were dependent upon both feeder strain. After overnight incubation, growth was X and V factors for growth, and all strains except examined for beta-hemolysis within the wide zone of eight could be accommodated in the currently beta-lysin produced by the feeder strain. Those iso- recognized taxa of five biotypes. The intraspecies lates showing hemolysis were further checked for a assignment of biotypes (bio) was based primarily characteristic CAMP reaction on sheep blood agar by on three biochemical features: ornithine decarmaking straight-line streaks of such isolates at right boxylation, indole production, and urea hydrolangles to (within 1 to 2 mm) the staphylococcus streak. ysis. All strains were characteristically negative Oxidase test. A drop of a 1% solution of tetramethyl-p-phenylenediamine dihydrochloride was for fi-D-galactosidase, and, of the isolates tested, placed on filter paper. The test organism was spot- none was hemolytic. Isolates of bio I (99 strains) were uniformly rubbed to a portion of the reagent-impregnated paper, positive in the three key tests as well as in the with results recorded after 10 s. Biochemical tests. Characterization of the Hae- test for xylose fermentation. The 217 strains of mophilus isolates was accomplished by use of the bio II formed the largest group of isolates. These Minitek system (BBL) according to the methods de- are seen as ornithine negative but indole and scribed previously (2). Briefly, this system utilizes urease positive. Xylose fermentation was variaplastic plates with wells accommodating reagent-im- ble, with only half of the isolates fermenting this pregnated disks containing phenol red indicator. Minbio itek Neisseria inoculum broth was found to be a carbohydrate. The 128 strains belonging to and indole and ornithine were negative III Haemoinoculum vehicle for the reliable supportive philus isolates. Originally, tests were read after 48 h of urease positive. Most strains fennented xylose. incubation, followed by addition of reagents to the Although only four isolates of bio IV were enappropriate wells to determine indole production and countered, they were uniformly positive in the TABLE 1. Biochemical characteristics of 464 isolates of H. influenzae and 83 isolates of H. parainfluenzaea % Positive test reactions
Test or substrate I
(99)b Ornithine Indole Urease ONPGc Nitrate reduction Oxidase Hemolysis Dextrose Sucrose Maltose Mannose
Xylose Galactose Fructose
Glycerold
100 100 100 0 100 82 0 97 0
H. influenzae (biotypes) V IV II III (4) (8) (128) (217) 100 100 0 0 100 0 0 100 0 100 100 100 0 0 0 0 100 100 100 99 83 100 82 88 0 0 0 0 100 0
4
5
1 98 100 76 89
6 50 92 82 37
100 0 10 5 91 95 76 56
100 0 0 0 100 100 100 100
100 0 0 0 100
100 63 100
H. parainfluenzae (biotypes) IV UnclassiIII II
VI (8)
(35)
(24)
(14)
(7)
fied (3)
100 0 0 0 100 100 0
100 0 0 97 100 14 4
100 0 100 50 96 31 22
0 0 100 7 100 40 77
0 0 0 14 100 0 0
100 100 100 100 100 0 0
100 0 0 0 100 100 100 80
100 100 100 73 0 91 100 50
100 100 100 67 4 59 100 33
100 100 100 36 0 79 100 14
100 100 100 86 0 100 100 0
100 100 100 33 33 100 100 50
I
a Al strains were tested and found negative for: lactose, mannitol, sorbitol, rhamnose, trehalose, melibiose, inositol, arabinose, H2S. b Numbers in parentheses indicate number of isolates tested. ONPG, o-Nitrophenyl-,B-D-galactopyranoside. d Not all isolates tested. c
170
J. CLIN. MICROBIOL.
OBERHOFER AND BACK
ornithine and urease tests but negative in the indole test. Strains conforming to bio V (ornithine and indole positive) were seen in eight instances, as were those which were unclassified (ornithine positive, indole negative). Members of these latter two groups were the only H. influenzae isolates which failed to produce urease. Isolates of the unclassified group were homogeneous in their activity, and they clearly constituted a separate sixth biotype (bio VI). (ii) H. parainfluenzae. Table 1 also shows the biochemical characteristics of 83 isolates of H. parainfluenzae. All strains were dependent on the V factor for growth. These strains could be divided into five biotypes based on their indole reaction, urease reaction, and ability to decarboxylate ornithine. Bio I (35 strains) was characterized by the positive test for ornithine and the negative test for urease. Bio II, the second largest group of isolates, was categorized because of the positive tests for both ornithine and urease. Bio III, the smallest of the three main biotypes, was ornithine negative but urease positive. We designated those strains of H. parainfluenzae which failed to act on any of the three key substrates as bio IV, based on the not infrequent occurrence of this biotype. The fifth biotype was positive in all three key tests and was the only V-factor-dependent Haemophilus strain showing a positive indole reaction. Hemolytic activity on sheep blood agar was seen with 4% of bio I, 22% of bio II, and 77% of bio III strains. The hemolysis appeared as visible zones of incomplete clearing surrounding the colonies growing within the zone of partial lysis
produced by the staphylococcus feeder strain. The fonn of hemolysis was distinct from the CAMP phenomenon because of the absence of synergistic activity (complete clearing) between the two lysins. A typical CAMP reaction was not produced by any of the strains nor did growth occur outside of the zone of partial lysis. Biochemical features distinguishing H. parainfluenzae from H. influenzae were abilities of many H. parainfluenzae isolates to produce fD-galactosidase and to consistently ferment sucrose, maltose, fructose and, to a great degree, With the exception of the unclassified strains, these isolates did not produce indole or ferment xylose. None produced acid from lactose or mannitol, which aids in the differentiation of these organisms from H. paraphrophilus, H. aphrophilus, and H. pleuropneumoniae. Source of isolates. (i) H. influenzae Table 2 shows the source of isolation of H. influenzae according to biotype and serotype. Half of the isolates (49%) were from eye cultures, with 63 (14%) recovered from sputum and 70 (15%) from nasopharyngeal cultures. Isolates from blood, ear, and spinal fluid (72 strains) accounted for 16% of the total isolates tested. The distribution of bio I isolates according to source was broad (Table 2). However, 34 of the 36 blood isolates and 14 of 15 cerebrospinal fluid isolates were of bio I, and all but 23 of the bio I isolates were serotype b. There were equal numbers of nasopharyngeal and sputum isolates of the same serotype (b). Of interest is that all three serotype f isolates were of bio I, with two of the isolates recovered from sputum and one mannose.
TABLE 2. Source of isolation of 464 strains of H. influenzae according to biotype
Serotype/source
Serotype NTa e
380 1 79 1
f
3
a
b
Source Eye Nasopharynx Sputum Blood Ear
No. of strains of biotype:
Total no. tested
227
70 63 36 21 15 7 10 3 12
ivvV
20b 0 76 0 3
10 15 14 34 4 14 1 3 2 2
214
128
3
7
8
0 3 0
0 0 0
0 0 1
1 0 0
0 0 0
0
0
0
0
0
130
86
1
0
0
38 22 1 6 1 4 4 1 10
11 22 1 4 0 1 3 0 0
0 2 0 1 0 0 0 0 0
3 3 0 1 0 1 0 0 0
3 0 0 5 0 0 0 0 0
Cerebrospinal fluid Tracheal aspirates Otherfluids Throat Other a NT, nontypable. b Sputum, six strains; eye, two; nasopharyngeal, seven; tracheal aspirate, one; ear, one.
BIOTYPES OF HAEMOPHILUS
VOL. 10, 1979
Of the bio II isolates, 130 cultures, with 28% of the remaining isolates arising from sputum and nasopharyngeal cultures. Also, 86 of the 128 bio III isolates (67%) were from eye cultures, with sputum and nasopharyngeal cultures accounting for 26% of the strains. All except three isolates of bio II and all isolates of bio III could not be serotyped. Isolates from ear cultures were evenly distributed among the three most common biotypes, with a significant number listed in the unclassified (bio VI) group. All isolates of bio V were from respiratory secretions. (ii) H. parainfluenzae Table 3 presents the sources of isolates of H. parainfluenzae. It is seen that the preponderance of these (43%) was from sputum, with the remainder recovered from various sources. The isolates from all sources were fairly evenly distributed among the four numbered biotypes and unclassified strains. It is interesting to note that all four of the cervical/urethral isolates were bio II. When compared with H. influenzae, the H. parainfluenzae cultures showed a greater proportion arising from dental abscesses and from abdominal areas such as peritoneal fluid, gastric aspirates, and bile. Age distribution according to biotype. The distribution of H. influenzae isolates, according to age of host and source of isolation, is shown in Table 4. The majority of strains refrom an eye culture. (61%) were from eye
171
covered from blood (59%) and cerebrospinal fluid (69%) were from patients less than 1 year of age, and all isolates except one were from those aged 4 or less. All were bio I, serotype b. Strains of bio I isolated from nasopharyngeal (86%), eye (100%), and ear (75%) cultures were from children aged 4 or less, whereas the 10 isolates from sputum were from patients 10 years of age or older. Bio II and III isolates, comprising the majority of all H. influenzae strains, were recovered principally from pediatric ear and eye cultures. In contrast, isolates from sputum were TABLE 3. Source of isolation of 81 strains of H. parainfluenzae according to biotype No. of strains of biotype: UnclasI II HI IV
Total Source
no.
siflied
tested
16 3 1 2 2 0 0 1 2 0 1 2
35 9 6 3 4 4 3 2 2 2 2 2 2 5
Sputum Dental abscess Eye Peritoneal fluid
Throat Cervix/urethra Wound Cord Gastric washings Ear Cerebrospinal fluid Transtracheal Sinus Other
1 2
8 0 3 0 1 0 0 0 0 1 1 0 0 0
9 3 1 0 0 4 2 1 0 1 0 0 1 2
1 1 1 1 1 0 1 0 0 0 0 0 0 1
1 2 0 0 0 0 0 0 0 0 0 0 0 0
TABLE 4. Age distribution of biotypes according to source of isolationa Organism/source
H. influenzaeb Blood CSF NP
Sputum
Eye
No.
JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 1979, p. 168-174 0095-1137/79/08-0168/07$02.00/0
Biotypes of Haemophilus Encountered in Clinical Laboratories THOMAS R. OBERHOFER* AND ANDREW E. BACK Microbiology Section, Department of Pathology, Madigan Army Medical Center, Tacoma, Washington 98431
Received for publication 14 May 1979
The biochemical characteristics of 464 strains of Haemophilus influenzae and 83 strains of Haemophilus parainfluenzae isolated over an 18-month period are described. Of 22 characteristics obtained, only 6 were necessary to biochemically identify and biotype the isolates. The key substrates or tests were urease, ornithine, indole, o-nitrophenyl-,8-D-galactopyranoside, sucrose, and xylose. Five biotypes of H. influenzae and four of H. parainfluenzae were commonly recognized. Some strains were encountered which could not be accommodated in the recognized taxa but which constituted separate biotypes of the two species. H. influenzae biotype I was recovered principally from blood, cerebrospinal fluid, and upper respiratory secretions, and biotypes II and III were recovered from eye and sputum cultures. Biotype I was recovered primarily from children less than 1 year of age, whereas biotypes II and III were from persons 1 to 5 years old and from those over 20 years of age. Multiple isolates recovered from the same patient were almost always of the same biotype. Strains of H. parainfluenzae were isolated primarily from sputum, with others being isolated from body sources such as dental abscesses, gastric aspirates, and peritoneal fluid. An inverse relationship was noticed between hemolysis and mannose fermentation among H. parainfluenzae biotype III strains, whereas the relationship was absent among the other biotypes.
Traditionally, the identification of Haemophilus is accomplished by testing the requirements for specific growth factors on factor-deficient agar medium or by using the test for satellitism. It is then common to report isolates recovered from clinical sources as Haemophilus influenzae or Haemophilus parainfluenzae, without consideration being given to the relationship of source of isolation to strain characteristics. Until the emergence of beta-lactamase-producing H. influenzae, little recognition had been given to the biochemical identification and differentiation of the haemophilus. Kilian (12, 13) can be credited with the sparking of interest in, and the recognition and rediscovery of, the haemophilus as biochemically active organisms rather than as bacteria of biochemical insignificance. The excellent report on the taxonomic study of the genus (13) clarified the relationships of the species and showed the practical usefulness of biochemical differentiation within the genus. Since then, biotyping of H. influenzae has been used to describe the isolates recovered as causative agents of diffuse bronchopneumonia (20) and bacteremia (1, 5). This study was undertaken to define the biochemical behavior of Haemophilus isolates re-
covered from routine clinical specimens. Each was subjected to a large number of tests to identify it to the species level, to delineate distinct biotypes of each species, and to ascertain the distribution of biotypes in relation to the source of isolate and age of patient. Insofar as possible, all isolates were nonselectively included in the study in order to accurately reflect the normal incidence of the various species or
biotypes. MATERIALS AND METHODS Test organisms. The organisms included in this study were isolated over an 18-month period in the Microbiology Laboratory, Madigan Army Medical Center. Isolates were recovered using chocolate agar (5% sheep blood in Mueller-Hinton base) supplemented with 1% IsoVitaleX (BBL Microbiology Systems) and 5% sheep blood agar containing a staphylococcus steak for satellite growth. Incubation was at 35°C in 5 to 7% C02. Growth factor requirements. Isolates were lightly suspended in 2 to 3 ml of sterile saline to produce a turbidity equivalent to a 0.5 MacFarland standard, to reduce any carry-over effect of nutrients from the medium (25). A swab moistened with the suspension was streaked in three directions on a Trypticase soy agar (BBL) plate. Each plate was dried for approximately 15 min, after which X-, V-, and XV-
168
VOL. 10, 1979
BIOTYPES OF HAEMOPHILUS
169
factor disks (Difco) were placed on the plates and nitrate reduction. Subsequently, it was found that firmly pressed to the agar. Disks were spaced so that suitable results were obtained after 18 to 24 h without factors were not mutually contributory. Plates were loss of sensitivity. incubated overnight under CO2 and examined for Classification. The classification of the Haemogrowth around the individual disks. Satellite growth philus isolates closely followed the scheme of Kilian was determined using the beta-toxigenic (ATCC (13). Divergence was necessary when information was 25923) strain of Staphylococcus aureus as the V-factor lacking. The serotype of each isolate was taken into feeder. Tests for C02 requirements were not performed account during the classification attempts. routinely. RESULTS Serotyping. Encapsulated isolates of H. influenzae were serotyped by slide agglutination, using specific (i) H. influcharacteristics. Biochemical antisera to types a through f (Burroughs-Wellcome). enzae. The biochemical features of 464 fresh by determined was activity Hemolytic Hemolysis. streaking the isolates on sheep blood agar and then clinical isolates of H. influenzae are presented making a straight-line cross-streak with the S. aureus in Table 1. All strains were dependent upon both feeder strain. After overnight incubation, growth was X and V factors for growth, and all strains except examined for beta-hemolysis within the wide zone of eight could be accommodated in the currently beta-lysin produced by the feeder strain. Those iso- recognized taxa of five biotypes. The intraspecies lates showing hemolysis were further checked for a assignment of biotypes (bio) was based primarily characteristic CAMP reaction on sheep blood agar by on three biochemical features: ornithine decarmaking straight-line streaks of such isolates at right boxylation, indole production, and urea hydrolangles to (within 1 to 2 mm) the staphylococcus streak. ysis. All strains were characteristically negative Oxidase test. A drop of a 1% solution of tetramethyl-p-phenylenediamine dihydrochloride was for fi-D-galactosidase, and, of the isolates tested, placed on filter paper. The test organism was spot- none was hemolytic. Isolates of bio I (99 strains) were uniformly rubbed to a portion of the reagent-impregnated paper, positive in the three key tests as well as in the with results recorded after 10 s. Biochemical tests. Characterization of the Hae- test for xylose fermentation. The 217 strains of mophilus isolates was accomplished by use of the bio II formed the largest group of isolates. These Minitek system (BBL) according to the methods de- are seen as ornithine negative but indole and scribed previously (2). Briefly, this system utilizes urease positive. Xylose fermentation was variaplastic plates with wells accommodating reagent-im- ble, with only half of the isolates fermenting this pregnated disks containing phenol red indicator. Minbio itek Neisseria inoculum broth was found to be a carbohydrate. The 128 strains belonging to and indole and ornithine were negative III Haemoinoculum vehicle for the reliable supportive philus isolates. Originally, tests were read after 48 h of urease positive. Most strains fennented xylose. incubation, followed by addition of reagents to the Although only four isolates of bio IV were enappropriate wells to determine indole production and countered, they were uniformly positive in the TABLE 1. Biochemical characteristics of 464 isolates of H. influenzae and 83 isolates of H. parainfluenzaea % Positive test reactions
Test or substrate I
(99)b Ornithine Indole Urease ONPGc Nitrate reduction Oxidase Hemolysis Dextrose Sucrose Maltose Mannose
Xylose Galactose Fructose
Glycerold
100 100 100 0 100 82 0 97 0
H. influenzae (biotypes) V IV II III (4) (8) (128) (217) 100 100 0 0 100 0 0 100 0 100 100 100 0 0 0 0 100 100 100 99 83 100 82 88 0 0 0 0 100 0
4
5
1 98 100 76 89
6 50 92 82 37
100 0 10 5 91 95 76 56
100 0 0 0 100 100 100 100
100 0 0 0 100
100 63 100
H. parainfluenzae (biotypes) IV UnclassiIII II
VI (8)
(35)
(24)
(14)
(7)
fied (3)
100 0 0 0 100 100 0
100 0 0 97 100 14 4
100 0 100 50 96 31 22
0 0 100 7 100 40 77
0 0 0 14 100 0 0
100 100 100 100 100 0 0
100 0 0 0 100 100 100 80
100 100 100 73 0 91 100 50
100 100 100 67 4 59 100 33
100 100 100 36 0 79 100 14
100 100 100 86 0 100 100 0
100 100 100 33 33 100 100 50
I
a Al strains were tested and found negative for: lactose, mannitol, sorbitol, rhamnose, trehalose, melibiose, inositol, arabinose, H2S. b Numbers in parentheses indicate number of isolates tested. ONPG, o-Nitrophenyl-,B-D-galactopyranoside. d Not all isolates tested. c
170
J. CLIN. MICROBIOL.
OBERHOFER AND BACK
ornithine and urease tests but negative in the indole test. Strains conforming to bio V (ornithine and indole positive) were seen in eight instances, as were those which were unclassified (ornithine positive, indole negative). Members of these latter two groups were the only H. influenzae isolates which failed to produce urease. Isolates of the unclassified group were homogeneous in their activity, and they clearly constituted a separate sixth biotype (bio VI). (ii) H. parainfluenzae. Table 1 also shows the biochemical characteristics of 83 isolates of H. parainfluenzae. All strains were dependent on the V factor for growth. These strains could be divided into five biotypes based on their indole reaction, urease reaction, and ability to decarboxylate ornithine. Bio I (35 strains) was characterized by the positive test for ornithine and the negative test for urease. Bio II, the second largest group of isolates, was categorized because of the positive tests for both ornithine and urease. Bio III, the smallest of the three main biotypes, was ornithine negative but urease positive. We designated those strains of H. parainfluenzae which failed to act on any of the three key substrates as bio IV, based on the not infrequent occurrence of this biotype. The fifth biotype was positive in all three key tests and was the only V-factor-dependent Haemophilus strain showing a positive indole reaction. Hemolytic activity on sheep blood agar was seen with 4% of bio I, 22% of bio II, and 77% of bio III strains. The hemolysis appeared as visible zones of incomplete clearing surrounding the colonies growing within the zone of partial lysis
produced by the staphylococcus feeder strain. The fonn of hemolysis was distinct from the CAMP phenomenon because of the absence of synergistic activity (complete clearing) between the two lysins. A typical CAMP reaction was not produced by any of the strains nor did growth occur outside of the zone of partial lysis. Biochemical features distinguishing H. parainfluenzae from H. influenzae were abilities of many H. parainfluenzae isolates to produce fD-galactosidase and to consistently ferment sucrose, maltose, fructose and, to a great degree, With the exception of the unclassified strains, these isolates did not produce indole or ferment xylose. None produced acid from lactose or mannitol, which aids in the differentiation of these organisms from H. paraphrophilus, H. aphrophilus, and H. pleuropneumoniae. Source of isolates. (i) H. influenzae Table 2 shows the source of isolation of H. influenzae according to biotype and serotype. Half of the isolates (49%) were from eye cultures, with 63 (14%) recovered from sputum and 70 (15%) from nasopharyngeal cultures. Isolates from blood, ear, and spinal fluid (72 strains) accounted for 16% of the total isolates tested. The distribution of bio I isolates according to source was broad (Table 2). However, 34 of the 36 blood isolates and 14 of 15 cerebrospinal fluid isolates were of bio I, and all but 23 of the bio I isolates were serotype b. There were equal numbers of nasopharyngeal and sputum isolates of the same serotype (b). Of interest is that all three serotype f isolates were of bio I, with two of the isolates recovered from sputum and one mannose.
TABLE 2. Source of isolation of 464 strains of H. influenzae according to biotype
Serotype/source
Serotype NTa e
380 1 79 1
f
3
a
b
Source Eye Nasopharynx Sputum Blood Ear
No. of strains of biotype:
Total no. tested
227
70 63 36 21 15 7 10 3 12
ivvV
20b 0 76 0 3
10 15 14 34 4 14 1 3 2 2
214
128
3
7
8
0 3 0
0 0 0
0 0 1
1 0 0
0 0 0
0
0
0
0
0
130
86
1
0
0
38 22 1 6 1 4 4 1 10
11 22 1 4 0 1 3 0 0
0 2 0 1 0 0 0 0 0
3 3 0 1 0 1 0 0 0
3 0 0 5 0 0 0 0 0
Cerebrospinal fluid Tracheal aspirates Otherfluids Throat Other a NT, nontypable. b Sputum, six strains; eye, two; nasopharyngeal, seven; tracheal aspirate, one; ear, one.
BIOTYPES OF HAEMOPHILUS
VOL. 10, 1979
Of the bio II isolates, 130 cultures, with 28% of the remaining isolates arising from sputum and nasopharyngeal cultures. Also, 86 of the 128 bio III isolates (67%) were from eye cultures, with sputum and nasopharyngeal cultures accounting for 26% of the strains. All except three isolates of bio II and all isolates of bio III could not be serotyped. Isolates from ear cultures were evenly distributed among the three most common biotypes, with a significant number listed in the unclassified (bio VI) group. All isolates of bio V were from respiratory secretions. (ii) H. parainfluenzae Table 3 presents the sources of isolates of H. parainfluenzae. It is seen that the preponderance of these (43%) was from sputum, with the remainder recovered from various sources. The isolates from all sources were fairly evenly distributed among the four numbered biotypes and unclassified strains. It is interesting to note that all four of the cervical/urethral isolates were bio II. When compared with H. influenzae, the H. parainfluenzae cultures showed a greater proportion arising from dental abscesses and from abdominal areas such as peritoneal fluid, gastric aspirates, and bile. Age distribution according to biotype. The distribution of H. influenzae isolates, according to age of host and source of isolation, is shown in Table 4. The majority of strains refrom an eye culture. (61%) were from eye
171
covered from blood (59%) and cerebrospinal fluid (69%) were from patients less than 1 year of age, and all isolates except one were from those aged 4 or less. All were bio I, serotype b. Strains of bio I isolated from nasopharyngeal (86%), eye (100%), and ear (75%) cultures were from children aged 4 or less, whereas the 10 isolates from sputum were from patients 10 years of age or older. Bio II and III isolates, comprising the majority of all H. influenzae strains, were recovered principally from pediatric ear and eye cultures. In contrast, isolates from sputum were TABLE 3. Source of isolation of 81 strains of H. parainfluenzae according to biotype No. of strains of biotype: UnclasI II HI IV
Total Source
no.
siflied
tested
16 3 1 2 2 0 0 1 2 0 1 2
35 9 6 3 4 4 3 2 2 2 2 2 2 5
Sputum Dental abscess Eye Peritoneal fluid
Throat Cervix/urethra Wound Cord Gastric washings Ear Cerebrospinal fluid Transtracheal Sinus Other
1 2
8 0 3 0 1 0 0 0 0 1 1 0 0 0
9 3 1 0 0 4 2 1 0 1 0 0 1 2
1 1 1 1 1 0 1 0 0 0 0 0 0 1
1 2 0 0 0 0 0 0 0 0 0 0 0 0
TABLE 4. Age distribution of biotypes according to source of isolationa Organism/source
H. influenzaeb Blood CSF NP
Sputum
Eye
No.
