Aerobic and Anaerobic Bacteriology of Wounds and Cutaneous Abscesses Itzhak Brook, MD, MSc, Edith H. Frazier, MSc \s=b\ The aerobic and anaerobic microbiologic characteristics of 584 wounds and 676 skin or soft-tissue abscesses were studied and correlated with the infection site. In wounds, aerobic or facultative bacteria only were present in 223 specimens (38%), anaerobes only in 177 specimens (30%), and mixed flora in 184 specimens (32%). In total there were 1470 isolates, 558 aerobic and 912 anaerobic, an average of 2.5 isolates per wound (1.6 anaerobic and 0.9 aerobic isolates). In abscesses, aerobic or facultative bacteria were recovered in 177 specimens (26%), anaerobes only in 243 specimens (36%), and mixed flora in 256 specimens (38%). In total there were 1702 isolates, 602 aerobic and 1100 anaerobic, an average of 2.5 isolates per abscess. The highest rates of anaerobes in wounds were in the inguinal, buttocks, and trunk areas and in abscesses in the perirectal, external genitalia, neck, and inguinal areas. The predominant aerobic organisms were Staphylococcus aureus (363 isolates), group A streptococci (98 isolates), and Escherichia coli (97 isolates). The predominant anaerobic organisms were Bacteroides species (986 isolates), Peptostreptococcus species (559 isolates), Clostridium species (153 isolates), and Fusobacterium species (109 isolates). The predominance of certain isolates in certain anatomical sites was correlated with their distribution in the normal flora adjacent to the infected site. These data highlight the polymicrobial nature of wounds and cutaneous abscesses.
(Arch Surg. 1990;125:1445-1451)
Skinpatients. Knowledge
and soft-tissue infections are common in hospitalized of the common bacterial causes of these infections enables the selection of empiric antimicrobial therapy before the results of bacterial cultures are available. The importance of aerobic and facultative bacteria, such as Staphylococcus aureus, in skin and soft-tissue infections is well established. However, only a few studies demonstrated the role of anaerobic bacteria in these infections.1"4 Two of these reports evaluated the aerobic and anaerobic microbio¬ logie characteristics of cutaneous abscesses in adults,13 and one in children.2 Only one report evaluated these organisms' role in wounds.4 However, in three of these studies,1,3'4 the species of most ofthe anaerobic organisms were not reported, and the role of aerobic and facultative strains other than S aureus and Streptococcus species was not evaluated. This retrospective report described the experience over 14 years of a military hospital in the diagnosis of the aerobic and anaerobic bacterial etiology of wounds and skin and softtissue abscesses. MATERIALS AND METHODS Between June 1973 and June 1987, 584 specimens of wounds and 676 specimens of skin and soft-tissue abscesses that were submitted
Accepted for publication July 1,1990. From the Departments of Pediatrics and Infectious Diseases, Naval Medical Center, Bethesda, Md. The opinions and assertions contained herein are the private ones of the writers and are not to be construed as official or reflecting the views ofthe Navy Department, the naval services at large, or the Defense Nuclear Agency. Reprint requests to Armed Forces Radiobiology Research Institute, Bethesda, MD 20814-5145 (Dr Brook).
for the isolation of aerobic and anaerobic bacteria showed bacterial growth. Excluded from analysis were 29 specimens of wounds, be¬ cause these wounds were treated with local antimicrobial therapy prior to sample collection. Specimens were processed by the clinical microbiology laboratories at the Naval Hospital in Bethesda, Md. Data regarding the administration of antimicrobial therapy prior to sample collection was available in 392 (67%) of 584 wounds and 431 (64%) of 676 abscesses. Systemic antimicrobial agents were used for 85 wounds (22%) and 92 abscesses (21%). These agents included ßlactamase-resistant penicillin (54 patients), erythromycin ethylsuccinate (31 patients), ampicillin sodium (28 patients), a cephalosporin (28 patients), penicillin G (18 patients), an aminoglycoside (15 pa¬ tients), and clindamycin phosphate (four patients). Of the 1260 patients from whom these specimens were obtained, 919 were males. Patients ranged in age between 2 weeks and 76 years
(mean age, 39 years, 4 months).
Specimens were obtained from wounds or abscesses by direct needle aspiration of purulent contents into a syringe that was immedi¬ ately sealed and transported to the laboratory within 30 minutes or by a swab that was dipped in the pus and introduced into anaerobic transport media (Port-A-Cul, BBL Microbiological Systems, Cockeysville, Md) and generally inoculated within 2 hours after collection. For specimen collection through surgical drainage, the involved site was first scrubbed with providone-iodine. After local anesthesia with 1% lidocaine injection or ethyl chloride spray, pus was collected for culture by percutaneous aspiration of the abscess cavity or by swabbing aspiration from the open cavity through a surgical incision. Sheep blood (5%), chocolate, and MacConkey agar plates were inoculated for the isolation of aerobic organisms. The plates were incubated at 37°C aerobically (MacConkey agar) and under 5% carbon dioxide (blood and chocolate agar) and examined at 24 and 48 hours. For the isolation of anaerobes, specimens were plated onto prereduced vitamin ,-enriched Bruceila blood agar, anaerobic blood agar plates containing kanamycin and vancomycin, and anaerobic blood plates containing colistin and nalidixic acid, and then inoculated into enriched thioglycolate broth. The plated media were incubated in GasPak jars (BBL Microbiology Systems) and examined at 48, 96, and 120 hours. The thioglycolate broth was incubated for 14 days. Anaerobes were identified by techniques previously described.5,6 Aerobic bacteria were identified using conventional methods.7 The data were organized according to anatomic location: head, neck, trunk, extremities, inguinal, perirectal, buttocks, and external genitalia.
RESULTS Of the 584 wounds, 36 (10 postsurgical wounds) were in the head, 16 (five postsurgical wounds) in the neck, 222 (105 postsurgical wounds) in the trunk, 15 (one postsurgical wound) in the arm, 60 in the hand, 181 (29 postsurgical wounds) in the leg, 18 (eight postsurgical wounds) inguinal, 22 (five postsurgical wounds) in the buttocks, seven perirectal, and 13 (three postsurgical wounds) in the external genitalia. In wounds, aerobic or facultative bacteria were present in 223 specimens (38%), anaerobes in only 177 specimens (30%), and mixed aerobic-anaerobic flora in 184 specimens (32%). In total there were 1470 isolates, 558 aerobic or facultative and 912 anaerobic bacteria, an average of 2.5 isolates per wound
(1.6 anaerobes and 0.9 aerobes or facultative bacteria) (Tables land 2).
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table 1.—Aerobic and Facultative
Organisms Isolated
From 584 Wounds
No. of Isolates
Organism No. of specimens Streptococcus a-Hemolytic • -Hemolytic Group A Group Group D Staphylococcus aureus S epidermidis Proteus species Pseudomonas aeruginosa Other Pseudomonas species
Head
Neck
Trunk
Arm
Hand
Leg
Inguinal
Buttocks
30
16
222
15
60
181
18
22
Perirectal
External Genitalia
Total
13
584 62
34
27
10
52
16
15 16
21
26
46
155
11
17
38
13
16
37
28
18
56
46
20
Escherichia coli
10
pneumoniae Enterobacter species Klebsiella
12
Other enterobacteriaceae*
40
21
Haemophilus influenzae H parainfluenzae Eikenella corrodens
Neisseria Candida
species species
Total
31
"Other enterobacteriaceae include Klebsiella
species, and Aeromonas species.
17
171
15
85
species other than pneumoniae,
Deep intra-abdominal or truncal infection was also present in 46 patients. Organisms similar to those isolated from wounds or cutaneous abscesses were also recovered from deep infections in 31 (67%) of these patients. In abscesses, aerobic or facultative organisms were recov¬ ered from 177 specimens (26%), anaerobic bacteria from only 243 specimens (36%), and mixed aerobic-anaerobic flora from 256 specimens (38%). In total there were 1702 isolates, 602 aerobic or facultative and 1100 anaerobic bacteria, an average of 2.5 isolates per abscess (1.6 anaerobes and 0.9 aerobes or facultative bacteria; range, one to seven isolates per speci¬ men) (Tables 3 and 4). The number of isolates per infected site varied between one and seven. The average number of isolates per infected site is reported in Table 5. The rate of recovery of anaerobic bacteria was almost always higher than the rate of recovery of aerobic or facultative organisms. The highest rate of recovery of anaerobes in wounds was in the inguinal, buttocks, and trunk areas, and in abscesses in the perirectal, external genitalia, neck, and inguinal areas. One hundred thirty-four wounds (23%) and 160 abscesses (24%) yielded only one organism. Staphlyococcus aureus was the most common organism, found in 218 cases (74%). Twenty-nine wounds and 41 abscess¬ es yielded a pure culture of a single anaerobe, including 44 isolates of Bactemides species (13 Bacteroides fragilis, six Bacteroides melaninogenicus, six Bacteroides intermedius, five Bacteroides thetaiotaomicron, four Bacteroides distasonis, four Bacteroides vulgatus, and two each of Bacteroides disiens, Bacteroides oralis, and Bacteroides asaccharolyticus) 12 isolates of Clostridium species, seven isolates of Pusobacterium species, five isolates of Peptostreptococcus species, and two isolates of Propionibacterium acnes. In the 184 wounds and 256 abscesses where polymicrobial aerobic-anaerobic infection was observed, several bacterial
184
15
Citrobacter
species,
558
26 Providencia
species, Morganella species, Acinetobacter
combinations were noted: the fragilis group was recovered with Escherischia coli 36 times, with S aureus 18 times, and with group D streptococcus 15 times. The melaninogenicus group was isolated with group A streptococcus in 30 instances and with S aureus in 11 instances. Anaerobic cocci were recovered with E coli in 34 cases, with group A streptococcus in 15 cases, and with S aureus in nine cases. Tables 1 to 4 record the major types of isolated bacteria. The predominant aerobic organisms were S aureus (208 iso¬ lates in abscesses and 155 in wounds), group A streptococci (46 isolates in abscesses and 52 in wounds), and E coli (41 isolates in abscesses and 56 in wounds). The predominant anaerobic organisms were Bacteroides species (586 isolates in abscesses and 400 in wounds), gram-positive cocci (296 iso¬ lates in abscesses and 280 in wounds), Clostridium species (46 isolates in abscesses and 107 in wounds) and Pusobacterium species (84 isolates in abscesses and 25 in wounds). The distribution of aerobic and facultative organisms showed the following trends: S aureus was recovered from abscesses in all body sites but predominated (highest number of isolates per sample) in infections of the leg (0.6 isolates per specimen), neck (0.5 isolates per specimen), and hand (0.5 isolates per specimen). Staphylococcus aureus was recovered least often from abscesses and wounds of the perirectal and external genitalia areas. Group D streptococci, E coli, and Neisseria gonorrhoeae were mostly isolated from abscesses in the external genitalia and perirectal areas, and Haemophilus influenzae was mostly isolated from abscesses of the head and neck. Proteus and Pseudomonas species were mostly removed from abscesses in the trunk, head, and perirectal areas. In wounds, the highest recovery rates for all gramnegative organisms were in the trunk and buttocks areas. The following trends were noted regarding distribution of anaerobic organisms: Anaerobic gram-positive cocci were iso-
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table 2.—Anaerobic
Organisms Isolated
From 584 Wounds
No. of Isolates
Organism No. of specimens Peptostreptococcus magnus
Head
Neck
30
16
Trunk 222
Arm
Hand
Leg
Inguinal
Buttocks
15
60
181
18
22
Perirectal
External Genitalia
Total
13
584
26
24
73
16
20
44
6
25
55
3
18
micros
asaccharolyticus morbilorum
prevotii saccharolyticus
14
anaerobius
13
Other Peptostreptococcus
species
26
17
53
Streptococcus intermedius S constellatus Veillonella párvula
10
V alcalescens
Other Veillonella
species
Gaffkya species Bifidobacterium
species 12
Eubacterium lentum
Other Eubacterium
species
Proprionibacterium acnes Other Proprionibacterium Lactobacillus
19
11
52
species
species
Clostridium septicum C difficile C perfringens Other Clostridium
22
species
19 17
25
48
58
Fusobacterium nucleatum
15
F varium
Other Fusobacterium Bacteroides
species
fragilis*
84
36
141
B distasonis*
13
B ovatus*
10
vulgatus* thetaiotaomicron*
19
34
S uniformis*
melaninogenicus
26
S intermedius
10
26
asaccharolyticus ureolyticus
13
29
19
B oralis bivius
12
24
disiens oris-buccae Other Bacteroides
species
Total
"These
26 39
species all belong to the
24
383
61
16 22
89
259
30
37
10
19
912
fragilis group.
lated from wounds and abscesses in all sites, and their rate of recovery was between 0.3 to 0.6 isolates per specimen. How¬ ever, the highest rates of isolates per abscess were in the neck (0.7 isolates per specimen) and perirectal (0.6 isolates per specimen) areas. In wounds, the highest rates of recovery were in the neck (0.6 isolates per specimen), inguinal (0.6 isolates per specimen), and external genitalia (0.6 isolates per specimen) areas. Clostridium species were mostly recovered in abscesses of the trunk, perirectal, buttocks, and external
areas and in wounds of the trunk, leg, and hand Fusobacterium species were mostly isolated from ab¬ scesses of the head, neck, and perirectal areas, and the fragilis group was mostly isolated from abscesses of the perirectal (0.6 isolates per specimen), trunk (0.6 isolates per specimen), and external genitalia (0.5 isolates per specimen) areas. In wounds,the fragilis group was mainly isolated in the trunk (0.5 isolates per specimen), inguinal (0.5 isolates per specimen), and leg (0.3 isolates per specimen) areas.
genitalia areas.
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table 3.—Aerobic and Facultative
Organisms Recovered
From 676 Cutaneous Abscesses
No. of Isolates
Organism No. of specimens Streptococcus a-Hemolytic
Head
Neck
Breast
Trunk
Arm
Hand
Leg
Inguinal
158
43
40
123
13
21
12
32
Buttocks 35
Perirectal 136
External Genitalia 63
676 56
-y-Hemolytic
20
Group A Group Group C Group D Staphylococcus aureus S epidermidis Neisseria gonorrhoeae Proteus species Pseudomonas aeruginosa
12
46
25 49
22
11
42
12
10
33
208
10
33
12
21
30
Other Pseudomonas species Escherichia coli Klebsiella
Total
17
16
pneumoniae
22 41 12
Enterobacter species
Other enterobacteriaceae*
15
39 11
10
Haemophilus influenzae H parainfluenzae Eikenella corrodens
species species
Lactobacillus
Candida Total
123
41
28
"Other enterobacteriaceae include Klebsiella species other than
species, and Aeromonas species.
130
14
28
13
22
26
122
55
602
pneumoniae, Citrobacter species, Providencia species, Morganella species, Acinetobacter
melaninogenicus group was mostly isolated in ab¬ of the arm (0.6 isolates per specimen), head (0.4 isolates per specimen), perirectal (0.4 isolates per specimen), and neck (0.3 isolates per specimen) areas. In wounds, the melaninogenicus group predominated in the hand (0.2 isolates per specimen) and head (0.2 isolates per specimen) areas. Bacteroides bivius was mostly recovered in abscesses of the perirectal area and in wounds of the external genitalia, leg, and trunk areas. Bacteremia caused by organisms identical to those found in wounds occurred in 78 instances (13% of all wounds), and bacteremia caused by organisms identical to those recovered in abscesses occurred in 76 cases (11% of all abscesses) (Table 6). The highest numbers of isolates were the fragilis group (60 isolates, including 45 fragilis, seven thetaiotamicron, six distastonis, and two vulgatus), Clostridum species (21 isolates), S aureus (21 isolates), and enterobacteriaceae The
scesses
(15 isolates).
Analysis showed no significant reduction in the isolation rate of organisms in patients who received antimicrobial ther¬ apy prior to sample collection. Evaluation of the microbiologi¬ cal data according to the patients' ages showed no significant correlation between the age and the bacterial isolates, except for the higher recovery rate of H influenzae and group Streptococcus in children younger than 2 years and the infre¬ quent recovery of the fragilis group in this age group. COMMENT This study demonstrates the importance of anaerobic bac¬ teria in cutaneous abscesses and wounds. Although S aureus was the predominant isolate, especially in infections in the
extremities and the trunk, anaerobes were frequently isolat¬ ed, and the number of anaerobic bacteria exceeded the num¬ ber of aerobic and facultative bacteria in cutaneous abscesses and wounds in the perirectal, external genitalia, buttocks, inguinal, neck, and trunk areas. The recovery rate of anaer¬ obes in these sites as reported in this study is similar to the isolation rates of anaerobes in other studies that have investi¬ gated the microbiologie characteristics of subcutaneous ab¬ scesses in adults' and children.2 Anaerobes were isolated in these studies in numbers similar to or greater than aerobes in abscesses proximal to the oral, rectal, and vulvovaginal ar¬ eas. These findings are not surprising, since anaerobes pre¬ dominate in the normal mucous membrane flora in these sites, reaching concentrations of 1 10" in the rectum and 1 109 in the oral cavity, outnumbering aerobes 1000 to 1 in the rectum and between 10 and 100 to 1 in the oral cavity.8,9 These normal flora isolates are sometimes associated with local cutaneous abscesses and wounds in these areas. In contrast to previous reports, the present study demon¬ strates a higher frequency of isolation of the fragilis group in particular and anaerobes in general in trunk and leg wounds and abscesses. This may be due to the inclusion in our report of only specimens submitted for the recovery of both aerobic and anaerobic bacteria. Furthermore, about half of the trunk wounds and cutaneous abscesses followed abdominal surgical procedures that probably involved seeding with organisms of
gastrointestinal origin.
The location of the cutaneous abscesses and wounds is of
paramount importance in determining the organisms in¬
volved in the infection. Aspirates from infection in and around the oral, rectal, and vulvovaginal regions tend to yield mixed
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table 4.—Anaerobic
Organisms Recovered by 676 Cutaneous Abscesses No. of Isolates
Organism specimens Peptostreptococcus magnus No. of
Head 158 15
Neck 43
Breast
Trunk
Arm
40
123
13
Hand 21
Leg Inguinal 12
32
Buttocks
Perirectal
External Genitalia
Total
35
136
63
676
13
10
60
13
40
12
micros
asaccharolyticus morbilorum
23 14
prevotii saccharolyticus anaerobius
species Microaerophilic streptococci Streptococcus intermedius Veillonella párvula Other Peptostreptococcus
23
13
42
25
91
10 17
V alcalescens
Other Veillonella Bifidobacterium
species species 21
Eubacterium lentum
Other Eubacterium
species 23
Proprionibacterium acnes Other Proprionibacterium species Lactobacillus species Clostridium perfringens C butyhcum Other Clostridium species
12
32
13
Fusobacterium varium 36
F nucleatum F mortiferum F necrophorum
Other Fusobacterium Bacteroides
species
12
37
13
fragilis*
54
54
20
149
distasonis* 21
ovatus* B
vulgatus*
10
thetaiotaomicron*
27
S uniformis*
melaninogenicus
44 1
19
asaccharolyticus ureolyticus
73
17
34 21
ß intermedius
20
1
56 71
27
10
fi oralis
19
S bivius
20
S disiens oris-buccae
Other Bacteroides Total
"These
species
72
56
169
22
15
11
58
49
301
121
1100
species all belong to the S fragilis group.
aerobic and anaerobic flora similar to those that are part of the normal microbial flora at the adjacent mucous membrane. Conversely, specimens obtained from areas remote from those sites primarily contained constituents of the microflora indigenous to the skin. Mixed aerobic and anaerobic infections are also more prev¬ alent in the breast area, fingers, and nail beds.2 This finding may be due to the direct introduction of mouth flora (which are
71
19
15
226
predominantly anaerobic) by sucking or biting.
The data presented illustrate the relative frequencies of isolation of Bacteroides, Peptostreptococcus, Clostridium, and Fusobacterium species in various infections. About 91% of the anaerobic bacteria recovered from the infections exam¬ ined were members of these genera. The polymicrobial na¬ ture of the infections also was highlighted in the data; the average number of isolates per specimen was 2.5. Although the pathogenic role of all bacterial isolates has not been estab¬ lished, the synergistic characteristics in polymicrobial infec-
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table
5.—Bacteriological Characterization of
Wounds
specimens Type of bacterial growth, No. of specimens Aerobic only Anaerobic only No. of
Head
Neck
Trunk
Arm
Hand
Leg
Inguinal
Buttocks
30
16
222
15
60
181
18
22
76
10
19 13
98
55 91
28
12
Aerobic and anaerobic No. of bacterial isolates per specimen Aerobic
Perirectal
External Genital
Total
13
584
223 177
59 24
184
1.0 1.3
1.1
0.8
1.0
1.4
1.2
1.7
1.5
1.5
1.7
1.7
0.9 1.4
0.6
1.5
1.0 1.4
0.8
Anaerobic
1.5
0.9 1.6
Total
2.3
2.6
2.5
2.5
2.9
2.4
2.5
2.9
2.3
2.1
2.5
"Values do not add to total because of
rounding error.
tions have been well established.11"13 Several hypotheses have been proposed to explain such microbial synergy. It may be due to mutual protection from phagocytosis and intracellular killing,14 production of essential growth factors,16 and/or low¬ ering of oxidation reduction potentials in host tissues.16 The virulence of Bacteroides, Clostridium, Peptostreptococcus, and Fusobacterium species is well documented in animal studies13 and for clinical infections.1718 In past studies Bacteroides species were the most frequent¬ ly recovered anaerobes"; in this report they accounted for 49% of all anaerobic isolates. The relative distribution of the different Bacteroides species has important implications for the management of infections because of the differing antimi¬ crobial susceptibilities of various Bacteroides species. Al¬ though most members of the fragilis group produce ß-lactamase and resist penicillin, their susceptibility to ceph¬ alosporins varies19 but is predictable.2" The members of the fragilis group are the most prevalent Bacteroides species isolated, and fragilis is the most prevalent organism in the group, accouting for 69% of isolates in abscesses and in wounds and 75% of isolates in blood. However, the other members of the fragilis group are more resistant than fragilis to the newer cephalosporins.19 The growing resistance of Bacteroides species previously susceptible to penicillins has been noticed in the last decade. " These are members of the melaninogenicus group, oralis, disiens, bivius, oris, and buccae. The main mechanism of resistance is through the production of the enzyme ß-lactamase. When choices are made between antimi¬ crobial agents for the therapy of infections involving Bacter¬ oides species, complete identification and testing for antimi¬ crobial susceptibility and ß-lactamase production are
important.
The recovery rate of Bacteroides species in infected sites is similar to their distribution in normal flora.8,91718 While mem¬ bers of the fragilis group were more often isolated from sites proximal to the gastrointestinal tract, strains of pigmented Bacteroides species were more prevalent in infections proximal to the oral cavity, and disiens and bivius were more often isolated from vulvovaginal area. Knowledge of this distribution allows a logical choice of antimicrobial agents adequate for the therapy of infections in these sites. As in previous studies,1"4 anaerobic cocci were the second most frequently recovered anaerobes (29% of anaerobic iso¬ lates). The predominance of Peptostreptococcus magnus was previously established.21 Anaerobic cocci have been shown to possess virulence in animals infected with them alone,22 but more so in synergy with other aerobic and anaerobic bacte-
Table
6.—Organisms
Isolated From Blood of 676 Patients' Abscesses and 584 Patients' Wounds
Organism
No. of Isolates in Blood/ Total No. of Isolates in Abscesses (%)
No. of Isolates in Blood/ Total No. of Isolates in Wounds (%)
Aerobic
Staphylococcus aureus Group A Streptococcus Pseudomonas species Escherichia coli Other enteroacteriaceae Subtotal
(5) (7) 7/43 (16) 2/41 (5) 5/58 (9) 28*/396 (7)
4/58
(6) (4) (13) (7) (7)
26t/367
(7)
(2) (5) 3/84 (4) 1/20 (5) 38/216 (18) 4/183 (2) 7/148 (5) 60 /991 (6) 8811/1387 (6)
(3) (17) 1/25 (4) 1/24 (4) 22/204 (11) 1/87 (1) 5/194 (3) 57§/906 (6) 83#/1273 (6)
11/208
10/155
3/46
2/52 6/46 4/56
Anaerobic
Peptostreptococcus species Clostridium species Fusobacterium species Bacteroides bivius
fragilis group melaninogenicus group Other Bacteroides species Subtotal Total
5/296 2/44
8/263
19/109
"There were 26 patients. tThere were 25 patients. JThere were 53 patients. §There were 56 patients. HThere were 76 patients. #There were 78 patients.
ria.13 While clostridia (8% of all anaerobes) were mostly found in abscesses and wounds around the rectal or vulvovaginal area, fusobacteria (5% of all anaerobes) were isolated from wounds and abscesses in and around the oral area. This
corresponds to the presence as part of the normal host flora in
these sites.8,9
Staphylococcus aureus, the most prevalent aerobe, was usually found alone. This organism has a well-recognized propensity for abscess formation, both in local and in visceral infections. In contrast to anaerobes, its potential for abscess formation is not as dependent on synergistic bacterial mixtures.18
The recovery rate of aerobic and facultative organisms was also correlated with the site of infection. Oral flora, such as H influenzae, predominated in head and neck infections, genital
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
584 Wounds and 676 Cutaneous Abscesses Abscesses Head
Neck
Breast
Trunk
Arm
Hand
J-eg
Inguinal
Buttocks
Perirectal
External Genital
Total
158
43
40
123
13
21
12
32
35
136
63
676
45
20
51
14
18
41
15
15
55
23
243
15
29
10
12
72
36
256
0.9
62
0.8
53
177
13
1.0 1.7
0.7
M
1.1
1.3
1.1
0.7
0.7
0.9
0.9
1.4
1.4
1.3
1.7
0.7
0.9
1.8
1.4
2.2
1.9
1.6
2.2
2.6
2.1
2.4
2.8
2.0
2.0
2.5
2.1
3.1
2.8
2.5
pathogens such as gonorrhoeae were mostly recovered in external genitalia and perirectal infections, while enterobac¬ teriaceae predominated in rectal, external genital, leg, and trunk infections. Skin flora, such as S aureus and Streptococ¬ cus, showed uniform distribution throughout the different body sites, although they predominated in the extremities. This study illustrates the importance of obtaining speci¬ mens adequate for the recovery of aerobic as well as anaerobic
bacteria from wounds and cutaneous abscesses. The observed isolation trends in different body sites and the initial reading of Gram stain results could guide the clinician in selecting an empiric antimicrobial therapy. However, the final choice of antimicrobial agents should be based on isolation of specific organisms, aerobes as well as anaerobes. Antimicrobial agents may also play an important role in surgical prophylaxis, preventing postsurgical wounds and abscesses, which occurred in about 25% of our patients. A first-generation cephalosporin, such as cefazolin, is generally effective as surgical prophylaxis in sites distant from the oral or rectal area. However, in surgical procedures that involve mucus surfaces (oral, gastrointestinal, or vulvovaginal), cov¬ fragilis erage against both enterobacteriaceae and the group is of great importance. Single-agent prophylaxis with cefoxitin sodium is generally the treatment of choice for such
prophylaxis.
Surgical drainage is the therapy of choice of cutaneous abscesses and wounds. This is important because the environ¬ ment of an abscess is detrimental for many antimicrobial agents. The abscess capsule, the low pH, and the presence of binding proteins or inactivating enzymes (such as ß-lacta¬ mase) may impair the activity of many antimicrobial agents (especially aminoglycosides). However, the administration of systemic antimicrobial agents may be indicated in selected cases, especially to prevent or treat bacteremia. Antimicrobi¬ al therapy for mixed infections due to aerobic and anaerobic bacteria requires the administration of antimicrobial agents effective against both aerobic and anaerobic bacterial compo¬ nents of the infection.1'18 Antimicrobial agents that provide coverage for S aureus as well as anaerobic bacteria include cefoxitin sodium, clindamycin phosphate, imipenem and cilastatin sodium, and the combinations of ß-lactamase inhibitor and penicillin and metronidazole hydrochloride and ß-lactamase-resistant penicillin. Cefoxitin sodium and imipenem and cilastatin sodium provide also coverage for enterobacter¬ iaceae. However, aminoglycosides or other agents effective against these organisms should be added to the other agents when treating infections that include these bacteria. The authors acknowledge the efforts ofthe staff ofthe Clinical Microbiology Laboratories and the Clinical Boards at the Naval Medical Center and the secretarial assistance of Minerva G. Barreiro.
References 1. Meislin HW, Lerner SA, Graves MH, et al. Cutaneous abscesses: anaerobic and aerobic bacteriology and outpatient management. Ann Intern Med.
1977;97:145-150. 2. Brook I, Finegold SM. Aerobic and anaerobic microbiology of cutaneous abscesses in children. Pediatrics. 1981;67:891-895. 3. Ghoneim ATM, McGoldrick J, Blick PWH, Flowers MW, Marsden AK, Wilson OH. Aerobic and anaerobic bacteriology of subcutaneous abscesses. Br J Surg. 1981;68:498-500. 4. Kontiainen S, Rinne E. Bacteria isolated from skin and soft tissue lesions. Eur J Clin Microbiol Infect Dis. 1987;6:420-422. 5. Holdeman LV, Cato EP, Moore WEC, eds. Anaerobe Laboratory Manual. 4th ed. Blacksburg, Va: Virginia Polytechnic Institute and State University; 1977. 6. Sutter VL, Citron DM, Edelstein MAC, Finegold SM. Wadsworth Bacteriology Manual. 4th ed. Belmont, Calif: Star Publishing Co; 1985. 7. Lennette EH, Balows A, Hausler WJ, Shadomy CH. Manual of Clinical Microbiology. 4th ed. Washington, DC: American Society for Microbiology; 1985. 8. Gibbons RJ. Aspects of the pathogenicity and ecology of the indigenous oral flora of man. In: Ballow A, ed. Anaerobic Bacteria: Role in Disease. Springfield, Ill: Charles C Thomas Publisher; 1974. 9. Gorbach SL. Intestinal microflora. Gastroenterology 1977;60:1100-1129. 10. Brook I. Bacteriology of paronychia in children. Am J Surg 1981;141:703\x=req-\ 705. 11. Brook I. Enhancement of growth of aerobic and facultative bacteria in mixed infections with Bacteroides species. Infect Immun. 1985;50:929-931. 12. Altemeir WA. The pathogenicity of the bacteria of appendicitis. Sur-
gery. 1942;11:374-378. 13. Brook I, Hunter
V, Walker RI. Synergistic effects of anaerobic cocci, Bacteroides, clostridia, fusobacteria, and aerobic bacteria on mouse mortality and induction of subcutaneous abscess. J Infect Dis. 1984;149:924-928. 14. Tofte RW, Peterson PK, Schemling D. Opsonization of four Bacteroides
species: role of the classical complement pathway and immunoglobulin. Infect
Immun. 1980;27:784-792. 15. Lev M, Drudell DC, Milford AF: Succinate as a growth factor for Bacteroides melaninogenicus. J Bacteriol. 1971;108:175-178. 16. Mergenhagen SE, Thonard JC, Scherp HW. Studies on synergistic infection, I: experimental infection with anaerobic streptococci. J Infect Dis.
1958;103:33-44.
17. Brook I. Pediatric Anaerobic Infection: Diagnosis and Management. St
Louis, Mo: CV Mosby Co; 1989.
18. Finegold SM. Anaerobic Bacteria in Human Disease. Orlando, Fla: Academic Press Inc; 1977. 19. Aldridge KE, Sanders CV, Janney A, Faso S, Marier RL. Comparison of activities of penicillin G and new beta-lactam antibiotics against clinical isolates of Bacteroides species. Antimicrob Agents Chemother. 1984;26:410-413. 20. Tally FP, Cuchural GJ, Jacobus NV, et al. Susceptibility of the Bacteroides fragilis group in the United States in 1981. Antimicrob Agents Chemother. 1983;23:536-540. 21. Brougault AM, Rosenblatt JE, Fitzgerald RH. Peptococcus magnus: a significant pathogen. Ann Intern Med. 1980;93:244-248. 22. Brook I, Walker RI. Pathogenicity of anaerobic gram-positive cocci. Infect Immun. 1984;45:320-324.
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
(Arch Surg. 1990;125:1445-1451)
Skinpatients. Knowledge
and soft-tissue infections are common in hospitalized of the common bacterial causes of these infections enables the selection of empiric antimicrobial therapy before the results of bacterial cultures are available. The importance of aerobic and facultative bacteria, such as Staphylococcus aureus, in skin and soft-tissue infections is well established. However, only a few studies demonstrated the role of anaerobic bacteria in these infections.1"4 Two of these reports evaluated the aerobic and anaerobic microbio¬ logie characteristics of cutaneous abscesses in adults,13 and one in children.2 Only one report evaluated these organisms' role in wounds.4 However, in three of these studies,1,3'4 the species of most ofthe anaerobic organisms were not reported, and the role of aerobic and facultative strains other than S aureus and Streptococcus species was not evaluated. This retrospective report described the experience over 14 years of a military hospital in the diagnosis of the aerobic and anaerobic bacterial etiology of wounds and skin and softtissue abscesses. MATERIALS AND METHODS Between June 1973 and June 1987, 584 specimens of wounds and 676 specimens of skin and soft-tissue abscesses that were submitted
Accepted for publication July 1,1990. From the Departments of Pediatrics and Infectious Diseases, Naval Medical Center, Bethesda, Md. The opinions and assertions contained herein are the private ones of the writers and are not to be construed as official or reflecting the views ofthe Navy Department, the naval services at large, or the Defense Nuclear Agency. Reprint requests to Armed Forces Radiobiology Research Institute, Bethesda, MD 20814-5145 (Dr Brook).
for the isolation of aerobic and anaerobic bacteria showed bacterial growth. Excluded from analysis were 29 specimens of wounds, be¬ cause these wounds were treated with local antimicrobial therapy prior to sample collection. Specimens were processed by the clinical microbiology laboratories at the Naval Hospital in Bethesda, Md. Data regarding the administration of antimicrobial therapy prior to sample collection was available in 392 (67%) of 584 wounds and 431 (64%) of 676 abscesses. Systemic antimicrobial agents were used for 85 wounds (22%) and 92 abscesses (21%). These agents included ßlactamase-resistant penicillin (54 patients), erythromycin ethylsuccinate (31 patients), ampicillin sodium (28 patients), a cephalosporin (28 patients), penicillin G (18 patients), an aminoglycoside (15 pa¬ tients), and clindamycin phosphate (four patients). Of the 1260 patients from whom these specimens were obtained, 919 were males. Patients ranged in age between 2 weeks and 76 years
(mean age, 39 years, 4 months).
Specimens were obtained from wounds or abscesses by direct needle aspiration of purulent contents into a syringe that was immedi¬ ately sealed and transported to the laboratory within 30 minutes or by a swab that was dipped in the pus and introduced into anaerobic transport media (Port-A-Cul, BBL Microbiological Systems, Cockeysville, Md) and generally inoculated within 2 hours after collection. For specimen collection through surgical drainage, the involved site was first scrubbed with providone-iodine. After local anesthesia with 1% lidocaine injection or ethyl chloride spray, pus was collected for culture by percutaneous aspiration of the abscess cavity or by swabbing aspiration from the open cavity through a surgical incision. Sheep blood (5%), chocolate, and MacConkey agar plates were inoculated for the isolation of aerobic organisms. The plates were incubated at 37°C aerobically (MacConkey agar) and under 5% carbon dioxide (blood and chocolate agar) and examined at 24 and 48 hours. For the isolation of anaerobes, specimens were plated onto prereduced vitamin ,-enriched Bruceila blood agar, anaerobic blood agar plates containing kanamycin and vancomycin, and anaerobic blood plates containing colistin and nalidixic acid, and then inoculated into enriched thioglycolate broth. The plated media were incubated in GasPak jars (BBL Microbiology Systems) and examined at 48, 96, and 120 hours. The thioglycolate broth was incubated for 14 days. Anaerobes were identified by techniques previously described.5,6 Aerobic bacteria were identified using conventional methods.7 The data were organized according to anatomic location: head, neck, trunk, extremities, inguinal, perirectal, buttocks, and external genitalia.
RESULTS Of the 584 wounds, 36 (10 postsurgical wounds) were in the head, 16 (five postsurgical wounds) in the neck, 222 (105 postsurgical wounds) in the trunk, 15 (one postsurgical wound) in the arm, 60 in the hand, 181 (29 postsurgical wounds) in the leg, 18 (eight postsurgical wounds) inguinal, 22 (five postsurgical wounds) in the buttocks, seven perirectal, and 13 (three postsurgical wounds) in the external genitalia. In wounds, aerobic or facultative bacteria were present in 223 specimens (38%), anaerobes in only 177 specimens (30%), and mixed aerobic-anaerobic flora in 184 specimens (32%). In total there were 1470 isolates, 558 aerobic or facultative and 912 anaerobic bacteria, an average of 2.5 isolates per wound
(1.6 anaerobes and 0.9 aerobes or facultative bacteria) (Tables land 2).
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table 1.—Aerobic and Facultative
Organisms Isolated
From 584 Wounds
No. of Isolates
Organism No. of specimens Streptococcus a-Hemolytic • -Hemolytic Group A Group Group D Staphylococcus aureus S epidermidis Proteus species Pseudomonas aeruginosa Other Pseudomonas species
Head
Neck
Trunk
Arm
Hand
Leg
Inguinal
Buttocks
30
16
222
15
60
181
18
22
Perirectal
External Genitalia
Total
13
584 62
34
27
10
52
16
15 16
21
26
46
155
11
17
38
13
16
37
28
18
56
46
20
Escherichia coli
10
pneumoniae Enterobacter species Klebsiella
12
Other enterobacteriaceae*
40
21
Haemophilus influenzae H parainfluenzae Eikenella corrodens
Neisseria Candida
species species
Total
31
"Other enterobacteriaceae include Klebsiella
species, and Aeromonas species.
17
171
15
85
species other than pneumoniae,
Deep intra-abdominal or truncal infection was also present in 46 patients. Organisms similar to those isolated from wounds or cutaneous abscesses were also recovered from deep infections in 31 (67%) of these patients. In abscesses, aerobic or facultative organisms were recov¬ ered from 177 specimens (26%), anaerobic bacteria from only 243 specimens (36%), and mixed aerobic-anaerobic flora from 256 specimens (38%). In total there were 1702 isolates, 602 aerobic or facultative and 1100 anaerobic bacteria, an average of 2.5 isolates per abscess (1.6 anaerobes and 0.9 aerobes or facultative bacteria; range, one to seven isolates per speci¬ men) (Tables 3 and 4). The number of isolates per infected site varied between one and seven. The average number of isolates per infected site is reported in Table 5. The rate of recovery of anaerobic bacteria was almost always higher than the rate of recovery of aerobic or facultative organisms. The highest rate of recovery of anaerobes in wounds was in the inguinal, buttocks, and trunk areas, and in abscesses in the perirectal, external genitalia, neck, and inguinal areas. One hundred thirty-four wounds (23%) and 160 abscesses (24%) yielded only one organism. Staphlyococcus aureus was the most common organism, found in 218 cases (74%). Twenty-nine wounds and 41 abscess¬ es yielded a pure culture of a single anaerobe, including 44 isolates of Bactemides species (13 Bacteroides fragilis, six Bacteroides melaninogenicus, six Bacteroides intermedius, five Bacteroides thetaiotaomicron, four Bacteroides distasonis, four Bacteroides vulgatus, and two each of Bacteroides disiens, Bacteroides oralis, and Bacteroides asaccharolyticus) 12 isolates of Clostridium species, seven isolates of Pusobacterium species, five isolates of Peptostreptococcus species, and two isolates of Propionibacterium acnes. In the 184 wounds and 256 abscesses where polymicrobial aerobic-anaerobic infection was observed, several bacterial
184
15
Citrobacter
species,
558
26 Providencia
species, Morganella species, Acinetobacter
combinations were noted: the fragilis group was recovered with Escherischia coli 36 times, with S aureus 18 times, and with group D streptococcus 15 times. The melaninogenicus group was isolated with group A streptococcus in 30 instances and with S aureus in 11 instances. Anaerobic cocci were recovered with E coli in 34 cases, with group A streptococcus in 15 cases, and with S aureus in nine cases. Tables 1 to 4 record the major types of isolated bacteria. The predominant aerobic organisms were S aureus (208 iso¬ lates in abscesses and 155 in wounds), group A streptococci (46 isolates in abscesses and 52 in wounds), and E coli (41 isolates in abscesses and 56 in wounds). The predominant anaerobic organisms were Bacteroides species (586 isolates in abscesses and 400 in wounds), gram-positive cocci (296 iso¬ lates in abscesses and 280 in wounds), Clostridium species (46 isolates in abscesses and 107 in wounds) and Pusobacterium species (84 isolates in abscesses and 25 in wounds). The distribution of aerobic and facultative organisms showed the following trends: S aureus was recovered from abscesses in all body sites but predominated (highest number of isolates per sample) in infections of the leg (0.6 isolates per specimen), neck (0.5 isolates per specimen), and hand (0.5 isolates per specimen). Staphylococcus aureus was recovered least often from abscesses and wounds of the perirectal and external genitalia areas. Group D streptococci, E coli, and Neisseria gonorrhoeae were mostly isolated from abscesses in the external genitalia and perirectal areas, and Haemophilus influenzae was mostly isolated from abscesses of the head and neck. Proteus and Pseudomonas species were mostly removed from abscesses in the trunk, head, and perirectal areas. In wounds, the highest recovery rates for all gramnegative organisms were in the trunk and buttocks areas. The following trends were noted regarding distribution of anaerobic organisms: Anaerobic gram-positive cocci were iso-
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table 2.—Anaerobic
Organisms Isolated
From 584 Wounds
No. of Isolates
Organism No. of specimens Peptostreptococcus magnus
Head
Neck
30
16
Trunk 222
Arm
Hand
Leg
Inguinal
Buttocks
15
60
181
18
22
Perirectal
External Genitalia
Total
13
584
26
24
73
16
20
44
6
25
55
3
18
micros
asaccharolyticus morbilorum
prevotii saccharolyticus
14
anaerobius
13
Other Peptostreptococcus
species
26
17
53
Streptococcus intermedius S constellatus Veillonella párvula
10
V alcalescens
Other Veillonella
species
Gaffkya species Bifidobacterium
species 12
Eubacterium lentum
Other Eubacterium
species
Proprionibacterium acnes Other Proprionibacterium Lactobacillus
19
11
52
species
species
Clostridium septicum C difficile C perfringens Other Clostridium
22
species
19 17
25
48
58
Fusobacterium nucleatum
15
F varium
Other Fusobacterium Bacteroides
species
fragilis*
84
36
141
B distasonis*
13
B ovatus*
10
vulgatus* thetaiotaomicron*
19
34
S uniformis*
melaninogenicus
26
S intermedius
10
26
asaccharolyticus ureolyticus
13
29
19
B oralis bivius
12
24
disiens oris-buccae Other Bacteroides
species
Total
"These
26 39
species all belong to the
24
383
61
16 22
89
259
30
37
10
19
912
fragilis group.
lated from wounds and abscesses in all sites, and their rate of recovery was between 0.3 to 0.6 isolates per specimen. How¬ ever, the highest rates of isolates per abscess were in the neck (0.7 isolates per specimen) and perirectal (0.6 isolates per specimen) areas. In wounds, the highest rates of recovery were in the neck (0.6 isolates per specimen), inguinal (0.6 isolates per specimen), and external genitalia (0.6 isolates per specimen) areas. Clostridium species were mostly recovered in abscesses of the trunk, perirectal, buttocks, and external
areas and in wounds of the trunk, leg, and hand Fusobacterium species were mostly isolated from ab¬ scesses of the head, neck, and perirectal areas, and the fragilis group was mostly isolated from abscesses of the perirectal (0.6 isolates per specimen), trunk (0.6 isolates per specimen), and external genitalia (0.5 isolates per specimen) areas. In wounds,the fragilis group was mainly isolated in the trunk (0.5 isolates per specimen), inguinal (0.5 isolates per specimen), and leg (0.3 isolates per specimen) areas.
genitalia areas.
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table 3.—Aerobic and Facultative
Organisms Recovered
From 676 Cutaneous Abscesses
No. of Isolates
Organism No. of specimens Streptococcus a-Hemolytic
Head
Neck
Breast
Trunk
Arm
Hand
Leg
Inguinal
158
43
40
123
13
21
12
32
Buttocks 35
Perirectal 136
External Genitalia 63
676 56
-y-Hemolytic
20
Group A Group Group C Group D Staphylococcus aureus S epidermidis Neisseria gonorrhoeae Proteus species Pseudomonas aeruginosa
12
46
25 49
22
11
42
12
10
33
208
10
33
12
21
30
Other Pseudomonas species Escherichia coli Klebsiella
Total
17
16
pneumoniae
22 41 12
Enterobacter species
Other enterobacteriaceae*
15
39 11
10
Haemophilus influenzae H parainfluenzae Eikenella corrodens
species species
Lactobacillus
Candida Total
123
41
28
"Other enterobacteriaceae include Klebsiella species other than
species, and Aeromonas species.
130
14
28
13
22
26
122
55
602
pneumoniae, Citrobacter species, Providencia species, Morganella species, Acinetobacter
melaninogenicus group was mostly isolated in ab¬ of the arm (0.6 isolates per specimen), head (0.4 isolates per specimen), perirectal (0.4 isolates per specimen), and neck (0.3 isolates per specimen) areas. In wounds, the melaninogenicus group predominated in the hand (0.2 isolates per specimen) and head (0.2 isolates per specimen) areas. Bacteroides bivius was mostly recovered in abscesses of the perirectal area and in wounds of the external genitalia, leg, and trunk areas. Bacteremia caused by organisms identical to those found in wounds occurred in 78 instances (13% of all wounds), and bacteremia caused by organisms identical to those recovered in abscesses occurred in 76 cases (11% of all abscesses) (Table 6). The highest numbers of isolates were the fragilis group (60 isolates, including 45 fragilis, seven thetaiotamicron, six distastonis, and two vulgatus), Clostridum species (21 isolates), S aureus (21 isolates), and enterobacteriaceae The
scesses
(15 isolates).
Analysis showed no significant reduction in the isolation rate of organisms in patients who received antimicrobial ther¬ apy prior to sample collection. Evaluation of the microbiologi¬ cal data according to the patients' ages showed no significant correlation between the age and the bacterial isolates, except for the higher recovery rate of H influenzae and group Streptococcus in children younger than 2 years and the infre¬ quent recovery of the fragilis group in this age group. COMMENT This study demonstrates the importance of anaerobic bac¬ teria in cutaneous abscesses and wounds. Although S aureus was the predominant isolate, especially in infections in the
extremities and the trunk, anaerobes were frequently isolat¬ ed, and the number of anaerobic bacteria exceeded the num¬ ber of aerobic and facultative bacteria in cutaneous abscesses and wounds in the perirectal, external genitalia, buttocks, inguinal, neck, and trunk areas. The recovery rate of anaer¬ obes in these sites as reported in this study is similar to the isolation rates of anaerobes in other studies that have investi¬ gated the microbiologie characteristics of subcutaneous ab¬ scesses in adults' and children.2 Anaerobes were isolated in these studies in numbers similar to or greater than aerobes in abscesses proximal to the oral, rectal, and vulvovaginal ar¬ eas. These findings are not surprising, since anaerobes pre¬ dominate in the normal mucous membrane flora in these sites, reaching concentrations of 1 10" in the rectum and 1 109 in the oral cavity, outnumbering aerobes 1000 to 1 in the rectum and between 10 and 100 to 1 in the oral cavity.8,9 These normal flora isolates are sometimes associated with local cutaneous abscesses and wounds in these areas. In contrast to previous reports, the present study demon¬ strates a higher frequency of isolation of the fragilis group in particular and anaerobes in general in trunk and leg wounds and abscesses. This may be due to the inclusion in our report of only specimens submitted for the recovery of both aerobic and anaerobic bacteria. Furthermore, about half of the trunk wounds and cutaneous abscesses followed abdominal surgical procedures that probably involved seeding with organisms of
gastrointestinal origin.
The location of the cutaneous abscesses and wounds is of
paramount importance in determining the organisms in¬
volved in the infection. Aspirates from infection in and around the oral, rectal, and vulvovaginal regions tend to yield mixed
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table 4.—Anaerobic
Organisms Recovered by 676 Cutaneous Abscesses No. of Isolates
Organism specimens Peptostreptococcus magnus No. of
Head 158 15
Neck 43
Breast
Trunk
Arm
40
123
13
Hand 21
Leg Inguinal 12
32
Buttocks
Perirectal
External Genitalia
Total
35
136
63
676
13
10
60
13
40
12
micros
asaccharolyticus morbilorum
23 14
prevotii saccharolyticus anaerobius
species Microaerophilic streptococci Streptococcus intermedius Veillonella párvula Other Peptostreptococcus
23
13
42
25
91
10 17
V alcalescens
Other Veillonella Bifidobacterium
species species 21
Eubacterium lentum
Other Eubacterium
species 23
Proprionibacterium acnes Other Proprionibacterium species Lactobacillus species Clostridium perfringens C butyhcum Other Clostridium species
12
32
13
Fusobacterium varium 36
F nucleatum F mortiferum F necrophorum
Other Fusobacterium Bacteroides
species
12
37
13
fragilis*
54
54
20
149
distasonis* 21
ovatus* B
vulgatus*
10
thetaiotaomicron*
27
S uniformis*
melaninogenicus
44 1
19
asaccharolyticus ureolyticus
73
17
34 21
ß intermedius
20
1
56 71
27
10
fi oralis
19
S bivius
20
S disiens oris-buccae
Other Bacteroides Total
"These
species
72
56
169
22
15
11
58
49
301
121
1100
species all belong to the S fragilis group.
aerobic and anaerobic flora similar to those that are part of the normal microbial flora at the adjacent mucous membrane. Conversely, specimens obtained from areas remote from those sites primarily contained constituents of the microflora indigenous to the skin. Mixed aerobic and anaerobic infections are also more prev¬ alent in the breast area, fingers, and nail beds.2 This finding may be due to the direct introduction of mouth flora (which are
71
19
15
226
predominantly anaerobic) by sucking or biting.
The data presented illustrate the relative frequencies of isolation of Bacteroides, Peptostreptococcus, Clostridium, and Fusobacterium species in various infections. About 91% of the anaerobic bacteria recovered from the infections exam¬ ined were members of these genera. The polymicrobial na¬ ture of the infections also was highlighted in the data; the average number of isolates per specimen was 2.5. Although the pathogenic role of all bacterial isolates has not been estab¬ lished, the synergistic characteristics in polymicrobial infec-
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
Table
5.—Bacteriological Characterization of
Wounds
specimens Type of bacterial growth, No. of specimens Aerobic only Anaerobic only No. of
Head
Neck
Trunk
Arm
Hand
Leg
Inguinal
Buttocks
30
16
222
15
60
181
18
22
76
10
19 13
98
55 91
28
12
Aerobic and anaerobic No. of bacterial isolates per specimen Aerobic
Perirectal
External Genital
Total
13
584
223 177
59 24
184
1.0 1.3
1.1
0.8
1.0
1.4
1.2
1.7
1.5
1.5
1.7
1.7
0.9 1.4
0.6
1.5
1.0 1.4
0.8
Anaerobic
1.5
0.9 1.6
Total
2.3
2.6
2.5
2.5
2.9
2.4
2.5
2.9
2.3
2.1
2.5
"Values do not add to total because of
rounding error.
tions have been well established.11"13 Several hypotheses have been proposed to explain such microbial synergy. It may be due to mutual protection from phagocytosis and intracellular killing,14 production of essential growth factors,16 and/or low¬ ering of oxidation reduction potentials in host tissues.16 The virulence of Bacteroides, Clostridium, Peptostreptococcus, and Fusobacterium species is well documented in animal studies13 and for clinical infections.1718 In past studies Bacteroides species were the most frequent¬ ly recovered anaerobes"; in this report they accounted for 49% of all anaerobic isolates. The relative distribution of the different Bacteroides species has important implications for the management of infections because of the differing antimi¬ crobial susceptibilities of various Bacteroides species. Al¬ though most members of the fragilis group produce ß-lactamase and resist penicillin, their susceptibility to ceph¬ alosporins varies19 but is predictable.2" The members of the fragilis group are the most prevalent Bacteroides species isolated, and fragilis is the most prevalent organism in the group, accouting for 69% of isolates in abscesses and in wounds and 75% of isolates in blood. However, the other members of the fragilis group are more resistant than fragilis to the newer cephalosporins.19 The growing resistance of Bacteroides species previously susceptible to penicillins has been noticed in the last decade. " These are members of the melaninogenicus group, oralis, disiens, bivius, oris, and buccae. The main mechanism of resistance is through the production of the enzyme ß-lactamase. When choices are made between antimi¬ crobial agents for the therapy of infections involving Bacter¬ oides species, complete identification and testing for antimi¬ crobial susceptibility and ß-lactamase production are
important.
The recovery rate of Bacteroides species in infected sites is similar to their distribution in normal flora.8,91718 While mem¬ bers of the fragilis group were more often isolated from sites proximal to the gastrointestinal tract, strains of pigmented Bacteroides species were more prevalent in infections proximal to the oral cavity, and disiens and bivius were more often isolated from vulvovaginal area. Knowledge of this distribution allows a logical choice of antimicrobial agents adequate for the therapy of infections in these sites. As in previous studies,1"4 anaerobic cocci were the second most frequently recovered anaerobes (29% of anaerobic iso¬ lates). The predominance of Peptostreptococcus magnus was previously established.21 Anaerobic cocci have been shown to possess virulence in animals infected with them alone,22 but more so in synergy with other aerobic and anaerobic bacte-
Table
6.—Organisms
Isolated From Blood of 676 Patients' Abscesses and 584 Patients' Wounds
Organism
No. of Isolates in Blood/ Total No. of Isolates in Abscesses (%)
No. of Isolates in Blood/ Total No. of Isolates in Wounds (%)
Aerobic
Staphylococcus aureus Group A Streptococcus Pseudomonas species Escherichia coli Other enteroacteriaceae Subtotal
(5) (7) 7/43 (16) 2/41 (5) 5/58 (9) 28*/396 (7)
4/58
(6) (4) (13) (7) (7)
26t/367
(7)
(2) (5) 3/84 (4) 1/20 (5) 38/216 (18) 4/183 (2) 7/148 (5) 60 /991 (6) 8811/1387 (6)
(3) (17) 1/25 (4) 1/24 (4) 22/204 (11) 1/87 (1) 5/194 (3) 57§/906 (6) 83#/1273 (6)
11/208
10/155
3/46
2/52 6/46 4/56
Anaerobic
Peptostreptococcus species Clostridium species Fusobacterium species Bacteroides bivius
fragilis group melaninogenicus group Other Bacteroides species Subtotal Total
5/296 2/44
8/263
19/109
"There were 26 patients. tThere were 25 patients. JThere were 53 patients. §There were 56 patients. HThere were 76 patients. #There were 78 patients.
ria.13 While clostridia (8% of all anaerobes) were mostly found in abscesses and wounds around the rectal or vulvovaginal area, fusobacteria (5% of all anaerobes) were isolated from wounds and abscesses in and around the oral area. This
corresponds to the presence as part of the normal host flora in
these sites.8,9
Staphylococcus aureus, the most prevalent aerobe, was usually found alone. This organism has a well-recognized propensity for abscess formation, both in local and in visceral infections. In contrast to anaerobes, its potential for abscess formation is not as dependent on synergistic bacterial mixtures.18
The recovery rate of aerobic and facultative organisms was also correlated with the site of infection. Oral flora, such as H influenzae, predominated in head and neck infections, genital
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
584 Wounds and 676 Cutaneous Abscesses Abscesses Head
Neck
Breast
Trunk
Arm
Hand
J-eg
Inguinal
Buttocks
Perirectal
External Genital
Total
158
43
40
123
13
21
12
32
35
136
63
676
45
20
51
14
18
41
15
15
55
23
243
15
29
10
12
72
36
256
0.9
62
0.8
53
177
13
1.0 1.7
0.7
M
1.1
1.3
1.1
0.7
0.7
0.9
0.9
1.4
1.4
1.3
1.7
0.7
0.9
1.8
1.4
2.2
1.9
1.6
2.2
2.6
2.1
2.4
2.8
2.0
2.0
2.5
2.1
3.1
2.8
2.5
pathogens such as gonorrhoeae were mostly recovered in external genitalia and perirectal infections, while enterobac¬ teriaceae predominated in rectal, external genital, leg, and trunk infections. Skin flora, such as S aureus and Streptococ¬ cus, showed uniform distribution throughout the different body sites, although they predominated in the extremities. This study illustrates the importance of obtaining speci¬ mens adequate for the recovery of aerobic as well as anaerobic
bacteria from wounds and cutaneous abscesses. The observed isolation trends in different body sites and the initial reading of Gram stain results could guide the clinician in selecting an empiric antimicrobial therapy. However, the final choice of antimicrobial agents should be based on isolation of specific organisms, aerobes as well as anaerobes. Antimicrobial agents may also play an important role in surgical prophylaxis, preventing postsurgical wounds and abscesses, which occurred in about 25% of our patients. A first-generation cephalosporin, such as cefazolin, is generally effective as surgical prophylaxis in sites distant from the oral or rectal area. However, in surgical procedures that involve mucus surfaces (oral, gastrointestinal, or vulvovaginal), cov¬ fragilis erage against both enterobacteriaceae and the group is of great importance. Single-agent prophylaxis with cefoxitin sodium is generally the treatment of choice for such
prophylaxis.
Surgical drainage is the therapy of choice of cutaneous abscesses and wounds. This is important because the environ¬ ment of an abscess is detrimental for many antimicrobial agents. The abscess capsule, the low pH, and the presence of binding proteins or inactivating enzymes (such as ß-lacta¬ mase) may impair the activity of many antimicrobial agents (especially aminoglycosides). However, the administration of systemic antimicrobial agents may be indicated in selected cases, especially to prevent or treat bacteremia. Antimicrobi¬ al therapy for mixed infections due to aerobic and anaerobic bacteria requires the administration of antimicrobial agents effective against both aerobic and anaerobic bacterial compo¬ nents of the infection.1'18 Antimicrobial agents that provide coverage for S aureus as well as anaerobic bacteria include cefoxitin sodium, clindamycin phosphate, imipenem and cilastatin sodium, and the combinations of ß-lactamase inhibitor and penicillin and metronidazole hydrochloride and ß-lactamase-resistant penicillin. Cefoxitin sodium and imipenem and cilastatin sodium provide also coverage for enterobacter¬ iaceae. However, aminoglycosides or other agents effective against these organisms should be added to the other agents when treating infections that include these bacteria. The authors acknowledge the efforts ofthe staff ofthe Clinical Microbiology Laboratories and the Clinical Boards at the Naval Medical Center and the secretarial assistance of Minerva G. Barreiro.
References 1. Meislin HW, Lerner SA, Graves MH, et al. Cutaneous abscesses: anaerobic and aerobic bacteriology and outpatient management. Ann Intern Med.
1977;97:145-150. 2. Brook I, Finegold SM. Aerobic and anaerobic microbiology of cutaneous abscesses in children. Pediatrics. 1981;67:891-895. 3. Ghoneim ATM, McGoldrick J, Blick PWH, Flowers MW, Marsden AK, Wilson OH. Aerobic and anaerobic bacteriology of subcutaneous abscesses. Br J Surg. 1981;68:498-500. 4. Kontiainen S, Rinne E. Bacteria isolated from skin and soft tissue lesions. Eur J Clin Microbiol Infect Dis. 1987;6:420-422. 5. Holdeman LV, Cato EP, Moore WEC, eds. Anaerobe Laboratory Manual. 4th ed. Blacksburg, Va: Virginia Polytechnic Institute and State University; 1977. 6. Sutter VL, Citron DM, Edelstein MAC, Finegold SM. Wadsworth Bacteriology Manual. 4th ed. Belmont, Calif: Star Publishing Co; 1985. 7. Lennette EH, Balows A, Hausler WJ, Shadomy CH. Manual of Clinical Microbiology. 4th ed. Washington, DC: American Society for Microbiology; 1985. 8. Gibbons RJ. Aspects of the pathogenicity and ecology of the indigenous oral flora of man. In: Ballow A, ed. Anaerobic Bacteria: Role in Disease. Springfield, Ill: Charles C Thomas Publisher; 1974. 9. Gorbach SL. Intestinal microflora. Gastroenterology 1977;60:1100-1129. 10. Brook I. Bacteriology of paronychia in children. Am J Surg 1981;141:703\x=req-\ 705. 11. Brook I. Enhancement of growth of aerobic and facultative bacteria in mixed infections with Bacteroides species. Infect Immun. 1985;50:929-931. 12. Altemeir WA. The pathogenicity of the bacteria of appendicitis. Sur-
gery. 1942;11:374-378. 13. Brook I, Hunter
V, Walker RI. Synergistic effects of anaerobic cocci, Bacteroides, clostridia, fusobacteria, and aerobic bacteria on mouse mortality and induction of subcutaneous abscess. J Infect Dis. 1984;149:924-928. 14. Tofte RW, Peterson PK, Schemling D. Opsonization of four Bacteroides
species: role of the classical complement pathway and immunoglobulin. Infect
Immun. 1980;27:784-792. 15. Lev M, Drudell DC, Milford AF: Succinate as a growth factor for Bacteroides melaninogenicus. J Bacteriol. 1971;108:175-178. 16. Mergenhagen SE, Thonard JC, Scherp HW. Studies on synergistic infection, I: experimental infection with anaerobic streptococci. J Infect Dis.
1958;103:33-44.
17. Brook I. Pediatric Anaerobic Infection: Diagnosis and Management. St
Louis, Mo: CV Mosby Co; 1989.
18. Finegold SM. Anaerobic Bacteria in Human Disease. Orlando, Fla: Academic Press Inc; 1977. 19. Aldridge KE, Sanders CV, Janney A, Faso S, Marier RL. Comparison of activities of penicillin G and new beta-lactam antibiotics against clinical isolates of Bacteroides species. Antimicrob Agents Chemother. 1984;26:410-413. 20. Tally FP, Cuchural GJ, Jacobus NV, et al. Susceptibility of the Bacteroides fragilis group in the United States in 1981. Antimicrob Agents Chemother. 1983;23:536-540. 21. Brougault AM, Rosenblatt JE, Fitzgerald RH. Peptococcus magnus: a significant pathogen. Ann Intern Med. 1980;93:244-248. 22. Brook I, Walker RI. Pathogenicity of anaerobic gram-positive cocci. Infect Immun. 1984;45:320-324.
Downloaded From: http://archsurg.jamanetwork.com/ by a University of California - San Diego User on 06/12/2015
