U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content







AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window






Chapter 19

Battle Casualties in Korea: Studies of the Surgical Research Team Volume III

The Bacterial Flora of Healing Wounds:  A Study of the Korean Battle Casualty*

1st Lieutenant Joseph G. Strawitz, MC, USAR
Theodore F. Wetzler, B. S.
Captain John D. Marshall, MSC, USA
Lieutenant Colonel Robert B. Lindberg, MSC, USA
Captain John M. Howard, MC, USAR
Lieutenant Colonel Curtis P. Artz, MC, USA

Several studies have described the bacterial flora of battle wounds at the time of initial débridement. In one study during World War I, Stoddard found Clostridium  perfringens present in 80 percent of the wounds.1  Working in the Middle East and later in Tripolitania and Tunisia during World War II, MacLennan found that between 45 percent and 50 percent of the wounds contained clostridia.2  3  In the Pacific area, Neel and Cole found that 48 percent of 160 wounds which were cultured shortly after injury contained clostridia.4  Except for differences in species distribution, the pattern of simple contamination was similar in wounds of the Korean casualties. At the time of primary débridement, 46 percent were found to contain clostridia.5

This study was undertaken to explore the fate of microorganisms in the open, débrided wound during the earlier stages of healing.

Material and Methods

Patients received their earliest surgical treatment at hospitals located within 10 miles of the main line of resistance.

At battalion aid stations, external hemorrhage was controlled; the wounds were dressed; fluid replacement therapy was initiated; and penicillin (300,000-600,000 units) and tetanus toxoid were given routinely.

When resuscitation was considered optimum, the patient was anesthetized and the area surrounding the wound was cleansed with tincture of green soap and sterile water. The surrounding skin was prepared with tincture of merthiolate, and the surgical field was draped with sterile towels.

During March and April of 1953, 11 patients were selected for study. Selection was based on the accessibility of the wound for serial biopsies

 *Previously published in Surgery 37: 400, 1955.


and on its location somewhat removed from any focus of subsequent contamination. In the majority of patients, the wound of entrance in the skin was small as compared with the extent of destruction of the underlying subcutaneous tissue and muscle. Thus, when a perforating wound was incised, a large cavity was sometimes found in the deeper layers.

The initial surgery of these wounds consisted of débridement of skin, subcutaneous tissue, and muscle. The wounds were then covered with fine-mesh gauze and a thick, sterile dressing. Each patient was given penicillin (300,000 units) and streptomycin (.05 gm.) intramuscularly every 12 hours.

At the beginning of débridement, tissue blocks were taken from healthy muscle in the base of each wound. In addition, blocks of devitalized muscle were taken from six of the wounds. Subsequently, tissue blocks from typical areas of healing muscle were taken at 2-day intervals. For these secondary biopsies, the patients were returned to the operating room where sterile technic was used. After the last tissue block was taken and wound closure accomplished, patients were held from 1 to 3 days before being evacuated.

Each tissue block was placed in a medium of cooked meat broth (modified Robertson's) ; the tube was sealed with Vaspar and sent by air to the 406th Medical General Laboratory in Tokyo. Twenty control specimens, which were held and shipped in a similar manner, maintained viable

Isolation and Identification

The incoming specimens were incubated for 24 hours at 37o C. Then 1 ml. was transferred to thioglycollate broth and, in turn, it was incubated for an additional 18 hours at 37o C. A yeast- extract, blood-agar, plate and a chloral-hydrate, sodium-azide plate were streaked with a 4 ml. loop from the thioglycollate subculture and incubated anaerobically for 40 hours. A representative of each morphologic colony was selected from each plate and transferred to thioglycoflate broth. After an 18-hour incubation period, gram stains were performed. The gram-positive, sporeforming bacilli were plated on blood-agar base and incubated aerobically.  Those that showed no growth on the aerobic plates were replated on blood-agar base and incubated anaerobically for 48 hours to determine the uniformity of colony type. A representative colony was transferred to thioglycollate broth to be used as a seed culture for biochemical identification.

Identification of the aerobic flora was effected by standard laboratory technics. The clostridia were identified by using a modification of the Reed and Orr classification, utilizing 10 carbohydrates, hy-


drogen sulfide production, litmus milk reaction, gelatine liquefaction, indole production, acrolein production, nitrate reduction, and pathogenicity for guinea pigs.6


Anaerobes.  Nine of the 11 wounds studied contained one or more species of clostridia at the time of débridement (Table 1). One wound did not demonstrate clostridia on initial culture (Patient 1) but showed one species on the fifth day. Clostridia were never isolated from the remaining wound (Patient 10). Two of the wounds (Patients 3 and 7) were clinically infected, but neither showed gas gangrene. Clostridia were absent from two-thirds of the wounds when studied following initial surgery.

Table 1.  The Bacterial Flora of Eleven Wounds During the First Week of Healing

Aerobes.  All wounds contained a rich flora of gram-positive and gram-negative aerobes on initial culture (Table 1). Without exception, these organisms persisted in the wounds during the first week of healing. However, there appeared to be considerable variation in the number of organisms and species from one culture to the next.

Pathogenicity of Clostridia   By using the standard guinea-pig inoculating procedure, the following species of clostridia were found to be pathogenic in this laboratory: Cl. perfringens, Cl. fallax, Cl. carnis, Cl. novyi, Cl. sordelli, Cl. histolyticum, Cl. difficile, Cl. tetani, and Cl. parabotulinum.  The findings in this study compared favorably with pathogenicity of clostridia described previously.


Although this series is small, the consistency of the observations permits valid conclusions. The rich aerobic and anaerobic flora demonstrated in most wounds confirmed the results of similar studies since World War I.

The absence of clostridia in two-thirds of cultures taken subsequent to initial surgery suggests that the present methods of débridement devitalized tissue are important factors in reducing the bacterial flora.

The two wounds in which adequate débridement was not performed (Patients 3 and 7) showed clinical evidence of infection; they contained rich clostridial and aerobic flora throughout. the first week of healing. These findings are consistent with the belief that nonviable tissue is one of the necessary requirements for maintaining clostridia. The persistently rich, aerobic flora in all cultures is not surprising, since these organisms are found extensively in skin, air, soil, and in the respiratory tracts of medical personnel.


The changing character of both the aerobic and the clostridial flora from one hospital day to another was observed in all wounds.  Further, contamination at wound dressing could account for the appearance of many new aerobic organisms. Antibiotic therapy may have influenced both the aerobic and the anaerobic flora, but it did not infuence appreciably the flora in the presence of necrotic tissue.  Culturing a single, small-tissue block from a large wound probably did not give a complete picture of the total bacterial flora.  The consistency of the findings, however, suggests that a representative index of the baterial flora was obtained.

Histological studies of the same wounds are reported in another paper.8 In the absence of  necrotic tissue, the bacterial flora did not appear to effect the healing of the open wound.


1.  Pathogenic and nonpathogenic clostridia were isolated in 9 of 11 wounds at the time of initial surgery.  All wounds contained a rich aerobic flora.

2.  Only three of the four wounds showing clostridia at the time of initial surgery contained these organisms in subsequent cultures.  This would appear to indicate that present methods of debridement and irrigation are effective in reducing the bacterial population in wounds.

3.  At the time of secondary closure, pathogenic aerobic organisms are often present in the wound.  In the absence of other deterrents, however, they show no clinical evidence of interfering with wound healing.


1.  Stoddard, J.L.:  The Occurrence and Significance of B. welchi in Certain Wounds.  J.A.M.A. 71:  1400, 1918.
2.  MacLennan, J.D.:  Anaerobic Infections of War Wounds in the Middle East.  Lancet, 245 123, 1943.
3.  MacLennan, J.D.:  Anaerobic Infections in Tripolitania and Tunisia.  Lancet 246: 203, 1944
4.  Neel, H.B., and Cole, J.P.:  Gas Gangrene in Amphibious Warfare in the Pacific Area. Am. J. Surg. 66: 290, 1944.
5. Lindberg, R.B.; Wetzler, T.F.; Marshall, J.D.; Newton, A.; Strawitz, J.G.; and Howard, J.M.: The Bacterial Flora of Battle Wounds at the Time of Primary Debridement:  A Study of Korean Battle Casualties.  Ann. Surg. 141: 369, 1955.  (Chapter 18 of this volume.)
6. Reed, C.B., and Orr, J.H.:  Rapid Identification of Gas Gangrene Anaerobes.  War Med, 1: 493, 1941.
7. Breed, R.S.; Murray, E.G.D.; and Hitchens, A.P.: Bergey's Manual of Determinative Bacteriology, 6th edition.  Williams and Wilkins Company, Baltimore, Maryland, 1948
8.  Scully, R.E.; Artz, C.P.; and Sako, Y.:  The Criteria for Determining the Viability of Muscle in War Wounds.  (Chapter 13 of this volume.)