Battle Casualties in Korea: Studies of the Surgical Research Team, Volume IV
Bacterial Flora of War Wounds of Patients with Renal Failure*
Lieutenant Colonel R. B. Lindberg, MSC, USA
First Lieutenant Fred Parrott, MSC, USAR
A study of the flora of war wounds in patients who developed anuria during the first 4 days following injury was conducted during 1953 at the 11th Evacuation Hospital. There has been extensive speculation on the possibility that toxins from Clostridia in wounds contribute significantly to the renal shut-down syndrome. If this were the case, it was felt that such patients would show an anaerobic flora at variance with that observed in wounded men not suffering renal complications. Data on this latter category of patients were concurrently collected.
Materials and Methods
Wounds were observed for appearance, location and nature of injury. Cultures were made from biopsy specimens and from swabs, and blood cultures were obtained. Biopsy specimens were planted in chopped meat medium at the time of operation, and shipped to the 406th Medical General Laboratory for anaerobic culture. Swabs were cultured locally in thioglycollate broth and on aerobic blood agar plates. Blood cultures were made in evacuated rubber-stoppered bottles with trypticase soy broth under reduced oxygen tension. Antibiotic sensitivities of anaerobes were determined by a tube dilution technic using thioglycollate broth. Aerobes were tested by use of antibiotic-impregnated absorbent disks containing graded doses of antibiotic.
Observations
A total of 31 patients were studied. Of these, 25 provided only biopsy and swab cultures. Blood cultures were obtained on 8 cases. Seven patients were cultured at autopsy.
The Clostridia recovered from wound specimens in 25 anuric cases are shown in Table 1. Eighty-five strains of sixteen species, and one unidentified Clostridum, were recovered. The proportion of patients
*Submitted for publication in Annals of Surgery.
158
Table 1. Clostridia Recovered From Wounds and Tissue Specimens in25 Anuric Patients
Species | Total No. of Strains Recovered | Per Cent of Total | No. of Patients Positive | Per Cent of Patients |
|
Cl. perfringens* |
18 |
21.0 |
7 |
28 |
|
Cl. sporogenes |
25 |
29.3 |
9 |
36 |
|
Cl. novyi* |
10 |
11.7 |
4 |
16 |
|
Cl. bifermentans |
5 |
5.9 |
3 |
12 |
|
Cl. multifermentans |
5 |
5.9 |
3 |
12 |
|
Cl. parabotulinum |
1 |
1.1 |
1 |
4 |
|
Cl. lentoputrescens |
3 |
3.5 |
2 |
8 |
|
Cl. tertium |
2 |
2.3 |
1 |
4 |
|
Cl. tetani* |
3 |
3.5 |
3 |
12 |
|
Cl. aerofoetidum |
2 |
2.3 |
1 |
4 |
|
Cl. tetanomorphum |
1 |
1.1 |
1 |
4 |
|
Cl. carnis* |
2 |
2.3 |
2 |
8 |
|
Cl. butyricum |
3 |
3.5 |
2 |
8 |
|
Cl. paraputrificum |
2 |
2.3 |
2 |
8 |
|
Cl. histolyticum* |
1 |
1.1 |
1 |
4 |
|
Cl. sphenoides |
1 |
1.1 |
2 |
8 |
|
Cl. unclassified |
1 |
1.1 |
1 |
4 |
|
|
85 |
|
|
|
*Species pathogenic for laboratory animals.
positive for each species is shown. This detailed listing permitted a comparison with findings on patients not in renal failure. Predominant species included Cl. sporogenes (29 per cent), Cl. perfringens(21 per cent) and Cl. novyi (11.7 per cent) of the total strains. It will be noted that 34, or 40 per cent, of the total strains recovered were of species usually termed pathogenic. Thirty-six per cent of patients were positive for Cl. sporogenes, 28 per cent for Cl. perfringens, 16 per cent for Cl. novyi and 12 per cent for Cl. multifermentans. Seventeen of the twenty-five had wounds which harbored Clostridia. Of 81 tissue specimens cultured, only 40 were positive for Clostridia. These proportionate values are summarized in Table 2.
It will be seen that 68 per cent of the patients, and 49.7 per cent of the specimens, were positive for Clostridia. In terms of those cases positive for Clostridia, an average of five strains of Clostridia were recovered per patient. From the 40 positive tissue specimens, an average of 2.12 strains per specimen were obtained.
When the proportion of specimens to strains of pathogenic Clostridia was considered, it appeared that 41.9 per cent of specimens yielded
159
Table 2. Clostridia Recovered From 81 Specimens Collected From 25Anuric Patients
|
| Number | Per Cent |
|
Patients studied | 25 | 100 |
|
Number of patients harboring Clostridia | 17 | 68.0 |
|
Total specimens | 81 | 100 |
|
Total specimens positive for Clostridia | 40 | 49.7 |
|
Total strains of Clostridia recovered | 85 |
|
|
Average number of strains per positive patient | 5 |
|
|
Average number of strains per positive specimen | 2.12 |
|
toxigenic Clostridia. Of all patients cultured, 44 per cent harbored toxigenic Clostridia. One anaerobe, Cl. difficile, was recovered from the 10 blood cultures run. Among 7 autopsied cases, 25 specimens yielded 5 strains of Clostridia. These included Cl. sporogenes, Cl. bifermentans, Cl. capitovale and two strains of Cl. multifermentans. No toxigenic strains were recovered from autopsies.
The aerobic flora in war wounds is numerically predominant, and undoubtedly plays a prominent role in the course of recovery. Three hundred fifty-seven aerobic strains were recovered from the 81 tissue specimens cultured as described above. Results are summarized in Table 3. Prominent major groups were Streptococci (49.1 per cent of total), Proteus species (10 per cent) Bacillus species (14 per cent) and Staphylococci (9.6 per cent of strains). The totals of Streptococci were similar for clinical cases and autopsies. On the other hand, beta-hemolytic Streptococci were more numerous from autopsies than from clinical material (23.6 percent contrasted with 16.2 per cent). The Streptococcus strains included 7.3 per cent of anaerobic or microaerophilic forms.
Among autopsy specimens, 76 strains were recovered. An increase in proportion of Proteus strains, and a slight decrease in Staphylococci, were the principal differences apparent between tissue specimens and autopsy samples.
The flora of the blood stream included that observed in numerous instances during artificial kidney dialysis of patients. The bacterial content of the dialysis bath was very high, and it is possible that the blood stream flora originated there. Streptococci were the most frequently encountered form, representing 34.4 per cent of total strains recovered from blood. Proteusspecies and Aerobacter aerogenes were next common in frequency.
160
Table 3. Aerobic Flora of Wounds and Blood and From Autopsies of25 Anuric Patients
|
| Sources |
| |||||
Tissues and Wounds | Autopsies | Blood | |||||
No. of Strains | Per Cent of Strains | No. of Strains | Per Cent of Strains | No. of Strains | Per Cent of Strains | ||
|
Staphylococcus, hemolytic |
13 |
3.7 |
2 |
2.7 |
1 |
4.4 |
16 |
|
Staphylococcus, nonhemolytic |
21 |
5.9 |
2 |
2.7 |
1 |
4.4 |
24 |
|
Streptococcus, beta-hemolytic |
58 |
16.2 |
18 |
23.6 |
3 |
13.0 |
79 |
|
Streptococcus, alpha-hemolytic |
23 |
6.5 |
1 |
1.3 |
1 |
4.4 |
25 |
|
Streptococcus, nonhemolytic |
68 |
19.1 |
17 |
22.4 |
3 |
13.0 |
88 |
|
Streptococcus, anaerobic, beta-hemolytic |
16 |
4.5 |
1 |
1.3 |
1 |
4.0 |
18 |
|
Streptococcus, anaerobic, alpha-hemolytic |
6 |
1.7 |
1 |
1.3 |
0 |
0 |
7 |
|
Streptococcus, anaerobic, nonhemolytic |
4 |
1.1 |
0 |
0 |
0 |
0 |
4 |
|
Bacillus species |
50 |
14.0 |
10 |
13.2 |
2 |
8.7 |
62 |
|
Corynebacterium species |
12 |
3.3 |
0 |
0 |
0 |
0 |
12 |
|
Pseudomonas species |
10 |
2.8 |
0 |
0 |
1 |
4.4 |
11 |
|
Proteus species |
36 |
10.0 |
18 |
23.6 |
5 |
21.7 |
59 |
|
Aerobacter aerogenes |
6 |
1.7 |
1 |
1.3 |
3 |
13.0 |
10 |
|
Escherichia coli |
12 |
3.3 |
0 |
0 |
0 |
0 |
12 |
|
Paracolobactrum species |
6 |
1.7 |
2 |
2.7 |
2 |
8.6 |
10 |
|
Paracolobactrum aerogenes |
0 |
0 |
1 |
1.3 |
0 |
0 |
1 |
|
Alcaligenes species |
10 |
2.8 |
1 |
1.3 |
0 |
0 |
11 |
|
Serratia marcescens |
1 |
0.3 |
0 |
0 |
0 |
0 |
1 |
|
Gram-negative rods, unclassified |
5 |
1.4 |
1 |
1.3 |
0 |
0 |
6 |
|
|
|
|
|
|
|
|
|
161
A short summary of the total aerobic flora is presented to clarify the relative distribution of bacterial forms in Table 4. The striking predominance of Streptococci is conspicuous. Representative samples of strains of Streptococci were typed; over 90 per cent were Group D, with some group H and K noted.
Table 4. Combined Aerobic Flora of Wounds and Blood and From Autopsies of 31 Anuric Patients
Organism |
| No. of Strains | Per Cent of Total |
|
Staphylococcus |
40 |
|
8.8 |
|
Hemolytic |
|
16 |
|
|
Nonhemolytic |
|
24 |
|
|
Streptococcus |
221 |
|
48.4 |
|
Alpha-hemolytic |
|
25 |
|
|
Beta-hemolytic |
|
79 |
|
|
Nonhemolytic |
|
88 |
|
|
Anaerobic, alpha-hemolytic |
|
7 |
|
|
Anaerobic, beta-hemolytic |
|
18 |
|
|
Anaerobic, nonhemolytic |
|
4 |
|
|
Bacillus species |
62 |
|
13.6 |
|
Pseudomonas species |
11 |
|
2.4 |
|
Proteus species |
59 |
|
13.0 |
|
Corynebacterium species |
12 |
|
2.6 |
|
Coliform |
50 |
|
11.0 |
|
Aerobacter aerogenes |
|
10 |
|
|
Escherichia coli |
|
12 |
|
|
Paracolon sp. |
|
10 |
|
|
Alcaligenes sp. |
|
11 |
|
|
Paracolobactrum aerogenes |
|
1 |
|
|
Gram-negative rods, unclass. |
|
6 |
|
|
Serratia marcescens |
1 |
|
0.2 |
|
Total |
456 |
|
|
Antibiotic Sensitivity of Flora of Wounds in Anuric Patients
The clostridial flora of the wounds studied was sampled for antibiotic sensitivity to penicillin, aureomycin, terramycin and chloramphenicol. Antibiotic concentrations inhibiting over 80 per cent of each predominant species tested were determined; a small number of strains in each species were resistant to higher levels of antibiotic. In the case of chloramphenicol, this resistance was extremely high; the proportion of resistant strains was lower in the case of penicillin, aureomycin and terramycin.
The sensitivity values observed are summarized in Table 5. The inhibitory levels of three principal species of Clostridia from these
162
Table 5. Antibiotic Sensitivity of Principal Clostridial Species in Wounds of Anuric Patients
Clostridium Species | No. of Strains Tested | Amount of Antibiotic Inhibiting over 80 Per Cent of Strains Tested | |||
Penicillin | Terramycin | Aureomycin | Chloromycetin | ||
|
|
| units/ml. | mcg./ml. | mcg./ml. | mcg./ml. |
|
Cl. perfringens |
12 |
0.7 |
1.0 |
0.5 |
12.5 |
|
Cl. sporogenes |
15 |
0.6 |
0.7 |
0.5 |
10.0 |
|
Cl. novyi |
10 |
0.4 |
0.25 |
0.4 |
10.0 |
|
Cl. bifermentans |
4 |
0.35 |
0.35 |
0.3 |
7.5 |
|
Cl. multifermentans |
4 |
0.25 |
0.45 |
0.3 |
7.5 |
Table 6. Antibiotic-resistant Clostridia From Wounds of Anuric Patients
Species | Antibiotic | No. Tested | No Resistant | Maximum Level Tolerated |
|
Cl. perfringens |
Penicillin |
12 |
2 |
5 u./ml. |
|
|
Aureomycin |
12 |
1 |
1 mcg./ml. |
|
|
Terramycin |
12 |
1 |
5 mcg./ml. |
|
|
Chloramphenicol |
12 |
2 |
20 mcg./ml. |
|
Cl. sporogenes |
Penicillin |
15 |
3 |
5 u./ml. |
|
|
Aureomycin |
15 |
3 |
1 mcg./ml. |
|
|
Terramycin |
15 |
2 |
5 mcg./ml. |
|
|
Chloramphenicol |
15 |
1 |
20 mcg./ml. |
|
Cl. novyi |
Penicillin |
10 |
1 |
5 u./ml. |
|
|
Aureomycin |
10 |
1 |
1 mcg./ml. |
|
|
Terramycin |
10 |
1 |
5 mcg./ml. |
|
|
Chloramphenicol |
10 |
2 |
20 mcg./ml. |
wounds were comparable with those seen in other extensive series of wound Clostridia studied; no essential increase in resistance was noted. The antibiotic-resistant strains observed included two Cl. perfringens, three Cl. Sporogenes and two Cl. novyi. These grew in concentrations of 5 units/ml. for penicillin, 1 mcg./ml. for aureomycin, 5 mcg./ml. for terramycin and 20 mcg./ml. for chloramphenicol. The reactions of these resistant strains are shown in Table 6.
Sensitivity of aerobic flora was set up with antibiotic-impregnated disks on blood agar plates. The levels of antibiotic indicated are the amounts present in the disks. Streptomycin was included in this series. Table 7 shows the results of these tests. The predominance of
163
Table 7. Sensitivity of Aerobic Microbial Flora From 25 Anuric Patients to 5 Antibiotics
|
| No. of Strains Tested | Penicillin (Unit) 1 unit | Aureomycin | Terramycin | Streptomycin | Chloromycetin (12.5 mcg.) | |||||
S* | R* | S | R | S | R | S | R | S | R | ||
|
Staphylococcus, hemolytic |
12 |
2 |
10 |
9 |
3 |
7 |
3 |
1 |
11 |
7 |
1 |
|
Staphylococcus, nonhemolytic |
12 |
5 |
7 |
11 |
1 |
11 |
1 |
6 |
7 |
4 |
|
|
Streptococcus, nonhemolytic |
14 |
4 |
10 |
7 |
7 |
6 |
8 |
5 |
9 |
1 |
1 |
|
Streptococcus, beta-hemolytic |
12 |
4 |
7 |
2 |
10 |
2 |
10 |
1 |
11 |
2 |
1 |
|
Streptococcus, alpha-hemolytic |
10 |
2 |
8 |
1 |
9 |
2 |
8 |
1 |
8 |
|
1 |
|
Aerobacter aerogenes |
5 |
|
5 |
3 |
2 |
1 |
4 |
1 |
4 |
2 |
2 |
|
Proteus species |
24 |
1 |
23 |
3 |
20 |
2 |
21 |
1 |
21 |
|
13 |
|
Paracolon species |
13 |
1 |
12 |
3 |
11 |
2 |
12 |
3 |
10 |
|
1 |
|
Escherichia coli |
5 |
|
5 |
2 |
3 |
2 |
3 |
4 |
1 |
|
|
|
Alcaligenes species |
4 |
|
4 |
2 |
2 |
2 |
2 |
|
4 |
|
|
|
Pseudomonas aeruginosa |
4 |
1 |
3 |
2 |
2 |
3 |
1 |
1 |
3 |
|
|
|
Totals |
115 |
20 |
94 |
45 |
70 |
40 |
73 |
24 |
89 |
16 |
20 |
*S=Sensitive.
R=Resistant.
164
penicillin-resistant forms is a conspicuous feature of this series, although other flora also show a markedly resistant trend.
Discussion
The bacterial flora of the wounds of patients with anuria showed no significant departures from that which might have been observed in another series of seriously wounded men. The totals for Clostridia are somewhat higher in terms of number of patients involved than were those of a series of patients not classified as anuric, but if a selection had been made of severe wounds only, the flora would have been similar in distribution.
Antibiotic-resistant forms of Clostridia were slightly more numerous than were those observed in a series of strains selected at random from recently wounded men. This increase in proportion of antibiotic-resistant strains is a not surprising result of the exposure of wound flora to antibiotic. There was not a large preponderance of resistant strains.
The aerobic flora was similar to that observed in recent wounds of patients not suffering with anuria. The predominant flora were cocci, with Streptococci the major part of the whole. The beta-hemolytic strains were chiefly D (enterococci). Proteus strains were relatively common.
The blood stream aerobic flora was profuse, but this must be interpreted in light of the fact that many of these cultures were drawn on men undergoing dialysis. Invasion through the dialyzing membrane could have accounted for the apparent transient bacteremia these men experienced. Some selectivity was shown in the preponderance of Proteus strains surviving in the blood; the antibiotic employed may have reduced the number of cocci found.
Summary
A series of 31 anuric patients were studied by means of wound and blood cultures. A profuse anaerobic and aerobic flora was found, but there was no indication that a specific alteration of the nature of wound flora occurred. Antibiotic-resistant strains were found in increased numbers over those recovered from more recent wounds, and an extensive antibiotic-resistant aerobic flora was present. There was no specific indication of an infection pattern typical of the anuric patient.
