Home page  |  About this library  |  Help  |  Clear       English  |  French  |  Spanish  
Expand Document
Expand Chapter
Full TOC
to previous section to next section

close this bookWar Wounds of Limbs - Surgical Management (ICRC; 1993; 116 pages)
View the documentForeword
View the documentProtection of health in war
View the documentPreface
View the documentAcknowledgements
close this folder1 Wounds
View the documentIntroduction
View the documentMissile wounds
View the documentExplosive wounds
View the documentWound infection
open this folder and view contents2 Management principles
open this folder and view contents3 Salvage of limb function
open this folder and view contents4 Amputation
open this folder and view contents5 Special subjects
View the documentAppendix 1 The Red Cross wound classification
View the documentAppendix 2 Figures from the Red Cross database
View the documentBack Cover

Wound infection

Wound contaminants

A variety of contaminants are found in wounds. Most commonly, streptococci, staphylococci and Gram-negative bacilli are cultured. With time, the number of organisms increases exponentially; the number of species of Gram-negative bacilli cultured also increases. However, such findings do not indicate which of these organisms are likely to cause subsequent wound infection, nor which cause dangerous infection. One must also consider clinical accounts and observations; these indicate which organisms it is most important to eliminate and which infections result from incomplete treatment. An example of the discrepancy between cultured contaminants and the clinical aspects of war wound infection is that provided by Clostridium welchii; this organism is rarely found as a contaminant but much of the war surgery practised in the 20th century has rightly been directed towards prevention and elimination of infection by it.

The factors that determine whether a wound becomes dangerously infected are summarized in Table 1.3 and an approximate time-scale for when the principal organisms cause clinical infection is shown in Table 1.4.

Table 1.3 Factors that determine progression of wound infection

Volume of dead tissue

Bone involvement

Volume and type of contamination influenced by:


Clothing worn


Skin microflora




Type of weapon

Presence of foreign material

Condition of the patient, e.g. multiple wounds

Time before administration of antibiotics

Adequacy of primary surgery (excision/decompression)

Table 1.4 A time-scale for when important pathogens are responsible for early wound infection. (This does not necessarily equate with organisms cultured from wounds on admission, nor at the times shown)



12 h to 3 days

Clostridium welchii


Streptococcus pyogenes

After 3 days

Staphylococcus aureus

After 5 days

Gram-negative bacilli, e.g.:


Pseudomonas aeruginosa


Escherichia coli


Serratia marcescens


Klebsiella pneumoniae

The natural history of untreated or incompletely treated wounds

Data concerning the evolution of the microbiology of war wounds without treatment are not available. However, patients with untreated wounds commonly present to International Committee of the Red Cross (ICRC) hospitals up to a month after injury; they provide an important insight into wound infection and spontaneous healing.

Some patients are seen within a few days of wounding in whom the diagnosis of clostridial myonecrosis (gas gangrene) is apparent; they have a high mortality and usually require proximal amputation. The incidence of this disease is low but varies with geography. More commonly, severe early infection without the clinical features of clostridial myonecrosis is seen; these are probably due to streptococci.

Many present with wounds over a week old. They have undergone a process of natural selection by surviving or avoiding clostridial or streptococcal infection. Their wounds are purulent and offensive but often show evidence of spontaneous healing (granulation tissue, callus formation and re-epithelialization) around the necrotic tissue and loose bone fragments. These patients may be in good condition; their only generalized sign is anaemia. The wounds are infected predominantly by staphylococci or Gram-negative organisms against which the body's healing processes can make headway.

Wound size is an important factor. Those that heal spontaneously are those with a smaller volume of culture medium, those in which clostridia and streptococci were not introduced at wounding and those which could discharge their culture medium. Large wounds contain a larger volume of culture medium with a correspondingly larger volume of ischaemic tissue. Such wounds are more likely to involve bone and resulting infection extends the culture medium and so, eventually, is more likely to overwhelm the patient if he does not receive treatment. Thus, the surgeon should ensure that he not only treats those with wounds that would naturally deteriorate, but also that he does not hinder spontaneous healing. This natural history is particularly important when considering the management of small wounds and of old or mistreated wounds.

Many deaths from limb wounds in World War I were due to clostridial myositis and streptococcal infection. This mortality was reputedly reduced by the operation of wound excision. In World War II it was found that streptococcal infection was a problem even at delayed closure. Since then it has been established that antibiotics, particularly penicillin, have a major role in the management of war wounds. More recently, there has been focus on Gram-negative bacilli, which are found in the wounds of patients who have already received penicillin or other antibiotics. These accounts stress the necessity of wound excision in the prevention of infective complications of the wound and so have done little to change the principles of management established in the first half of this century. This historical view indicates that Gram-negative bacilli are selected by the administration of penicillin and incomplete excision of dead tissue.

The contaminants, the natural history of wounds, penicillin and wound excision are interrelated. A nine-point interpretation of this relates wound infection to the role of the surgeon.


1 All wounds are contaminated by a mixture of organisms.

2 Infection by Clostridium welchii, C. oedematiens or Streptococcus pyogenes is rapid, dangerous and extends the culture medium.

3 Serious infection by all organisms is more likely with larger wounds.

4 Overall risk of serious infection is reduced by penicillin.

5 Overall risk of serious infection is reduced by wound excision.

6 The risk of infection by streptococci and clostridia is substantially reduced by penicillin.

7 The risk of later serious infection by Gram-negative organisms and staphylococci is substantially reduced by complete wound excision.



8 The presence of dead tissue combined with penicillin selects Gram-negative organisms and, especially with bone wounds, staphylococci.



9 Infection by Gram-negative organisms or staphylococci is only dangerous to patients with very large and incompletely excised wounds.

Specific clinical wound infections

Streptococcal infection and cellulitis

Streptococcus pyogenes is a Gram-positive coccus which secretes potent exotoxins. It is commonly cultured from wounds (unlike clostridia). The organism is responsible for the majority of early, severe wound infections.

The clinical features of streptococcal infection of a wound are not well-defined. Soon after injury, the patient has fever and pain around the wound. The wound exudes a little clear serum and has considerable surrounding swelling and erythema. Classical signs of spreading lymphangitis are rarely seen, although erythema spreads proximally (see Fig. 4.1). Because of skin blistering, it may be confused with clostridial myonecrosis.

Streptococci may be responsible also for inflammation and subsequent breakdown of a wound which was apparently clean and closed after a delay. Such inflammation responds readily to penicillin in the majority of cases. Likewise skin grafts which have taken and then are subsequently lost to infection are probably infected by S. pyogenes.

Gas gangrene

This early and dangerous infection is usually caused by anaerobic spore-forming clostridia. Clostridium welchii and C. oedematiens are said to be the species most frequently causing the disease. Their spores, when introduced into the wound, only cause disease if there is a large quantity of dead tissue. The local effects of the infection are myonecrosis with gas production (Fig. 1.5). Exotoxins may propagate the local effects and also exert dangerous systemic effects such as haemolysis and cardiac depression.

The disease may develop within 12 h of wounding and is most commonly seen within 3 days. Its general clinical features are fever, confusion, hypotension and anaemia. Specifically, the limb is swollen, mottled and tender; in the later stages there may be subcutaneous emphysema, a sweet, nauseating smell, skin blisters and, at operation, muscle discoloration. Oedema and gas track proximally around the neurovascular bundles.

Treatment must include excision of the affected muscle; this usually involves amputation (Fig. 1.5). Intravenous penicillin is essential.


Tetanus is more commonly seen in the wounded of African conflicts than in other parts of the world. The causative organism is Clostridium tetani, a strict anaerobe, which infects deep or large wounds. The toxin, tetanospasmin, spreads by an intraneural or vascular route to the motor nerve cells in the spinal cord and the brain; interference with the inhibitory elements of motor function gives the clinical signs.

Figure 1.5 (a) A 3-day-old shelling injury of the forearm with clinical gas gangrene (E30, X-, C-, F2, V0, M0; grade 3, type F). Surgical emphysema was palpable in the arm where discoloration and blistering are evident. The patient had a subcapital humeral amputation.

Figure 1.5 (b) The radiograph shows the fractures of the forearm bones and the presence of gas in the surrounding tissues.

The diagnosis often catches medical staff unawares. In less severe cases the signs may be manifest locally by tingling and spasms. Early and more serious signs are trismus, 'lock-jaw', neck rigidity, hyperpyrexia and difficulty of micturition. In the later stages, the patient exhibits generalized spasms triggered by minimal stimuli, convulsions, difficulty swallowing, opisthotonos and respiratory distress.

The early development of the signs of tetanus is associated with a poor prognosis.

Prophylaxis against the disease, in the patient without immunization, is achieved by wound excision, penicillin and antitetanus serum (500 iu).

Treatment comprises elimination of the organism and its culture medium (surgery and penicillin) and human antitetanus serum (6000 iu) which may have to be repeated after some days. The patient must also be supported whilst recovering from the effects of the toxin already bound in the central nervous system. This includes intravenous or intramuscular diazepam and placing the patient in quieter surroundings. More severe cases may require tracheostomy and ventilatory support.

The disease does not confer immunity and so a course of tetanus toxoid injections is commenced after recovery.

Staphylococcal infections

Most staphylococci isolated from wounds are Staphylococcus aureus. The most serious consequence of wound infection by this organism is abscess formation in closed wounds. Such abscesses are frequently associated with retained foreign material or bone fragments and are encapsulated without invasion of surrounding tissue.

Both Staphylococcus aureus and S. epidermidis can cause chronic infective osteitis in bone wounds and are difficult to eliminate.

Gram-negative infections

Patients with large, incompletely excised wounds and who have been given penicillin may develop signs of septic shock after some days; it is possible to culture Gram-negative organisms from both the wound and blood. However, their poor general condition is produced by a combination of the injury, blood transfusion, surgery and the infection and so it is difficult to know whether, in these circumstances, antibiotics directed towards Gram-negative organisms are of benefit.

Otherwise, Gram-negative organisms are found in most chronically open wounds without signs of sepsis; the patient has no fever, is generally well and the limb is not especially swollen. The wound may have started to heal despite copious purulent discharge. The organisms responsible for the infection are only present because of dead tissue; most such wounds respond readily to exploration and re-excision.

Synergistic infections

Tissue necrosis is occasionally seen within and around wounds in a pattern which does not fit any of the clinical descriptions above. Such infections are given a variety of names such as anaerobic cellulitis, synergistic gangrene or necrotizing fasciitis. It is probable that they represent a synergistic infection with an anaerobic element. The combination may be of either streptococci or clostridia and a Gram-negative organism.

The treatment involves excision of the necrotic tissue and antibiotics; it is not clear whether metronidazole should be added to the treatment regime for these patients.

Further Reading

Adams DB, Fackler ML (1990) Grenade fragmentation injuries. J Trauma (China) 6 (suppl):48-52.

Adams DB, Schwab CW (1988) Twenty one year experience with land mine injuries. J Trauma 28 (suppl):S159-62.

Berlin R, Janzon B, Liden E, Nordstrom B, Seeman T, Westling F (1988) Terminal behaviour of deforming bullets. J Trauma 28 (suppl):S58-62.

Cauderay GC, Doswald-Beck L (1990) ICRC concerns with regard to the effects of high energy transfer bullets and the methodology of research in this field. J Trauma (China) 6 (suppl):88-91.

Cheng XM, Lin YQ, Guo RF, Lian WK, Wang DT (1990) Analysis of wound ballistics in 2414 cases of battle casualties. J Trauma 6 (suppl): 169-72.

Cooper GJ, Ryan JM (1990) Interaction of penetrating missiles with tissues: some common misapprehensions and implications for wound management. Br J Surg 77:606-10.

Coupland RM (1991) The Red Cross Wound Classification. Geneva: The International Committee of the Red Cross.

Coupland RM, Howell PR (1988) An experience of war surgery and wounds presenting after 3 days on the border of Afghanistan. Injury 19:259-62.

Coupland RM, Korver A (1991) Injuries from antipersonnel mines: the experience of the International Committee of the Red Cross. Br Med J 303:1509-12.

Coupland RM, Hoikka V, Sjoeklint OG, Cuenod P, Cauderay GC, Doswald-Beck L (1992) Assessment of bullet disruption in armed conflicts. Lancet 339:35-7.

Dahlgren B, Berlin R, Brandberg A, Rybeck B, Scantz B, Seeman T (1982) Effects of benzyl penicillin on wound infection rate and on the effect of devitalised tissue, 12 hours after infliction of experimental missile trauma. Acta Chir Scand 148 (suppl):107-12.

Fackler ML (1986) Ballistic injury. Ann Emerg Med 15(12): 1451-5.

Fackler ML (1988) Wound ballistics: a review of common misconceptions. JAMA 259:2730.

Fackler ML, Surinchak JS, Malinowski JA, Bowen RE (1984) Bullet fragmentation: a major cause of tissue disruption. J Trauma 24:35-9.

Fackler ML, Bellamy RF, Malinowski JA (1988) The wound profile: illustration of the missile - tissue interaction. J Trauma 28 (suppl):S21-9.

Heggers JP, Barnes ST, Robson MC, Ristroph JD, Omer GE (1969) Microbial flora of orthopaedic war wounds. Milit Med 134:602-3.

International Committee of the Red Cross (1973) Weapons that may cause Unnecessary Suffering or have Indiscriminate Effects: Report on the Work of Experts. Geneva: ICRC.

International Committee of the Red Cross (1989) The International Law Concerning the Conduct of Hostilities. Geneva: ICRC.

Janzon B, Seeman T (1985) Muscle devitalisation in high energy missile wounds and its dependence on energy transfer. J Trauma 25:138-44.

Jones EL, Peters AF, Gasior RM (1968) Early management of battle casualties in Vietnam. Arch Surg 97:1-15.

Kennedy TL, Johnston GW (1975) Surgery of violence 1. Civilian bomb injuries. Br Med J 1:382-3.

Klein R, Berger S, Yekutiel P (1975) Wound infection during the Yom Kippur war. Ann Surg 182:15-21.

Mellor SG, Cooper GJ (1989) Analysis of 828 servicemen killed or injured by explosion in Northern Ireland 1970-84: the Hostile Action Casualty System. Br J Surg 76:1006-10.

Ogston A (1898) The wounds produced by modern small-bore bullets. Br Med J 2:813-15.

Ogston A (1899) The peace conference and the dum-dum bullet. Br Med J 2:278-81.

Owen Smith MS (1981) High Velocity Missile Injuries. London: Edward Arnold.

Owen Smith MS, Matheson JM (1968) Successful prophylaxis of gas gangrene of the high velocity missile wound in sheep. Br J Surg 55:36-9.

Ragsdale BD, Josselson A (1988) Experimental gunshot fractures. J Trauma 28 (suppl):S109-15.

Scott R (1982) High velocity and other missile injuries. In: Tubbs N, London PS (eds) Topical Reviews in Accident Surgery. Bristol: Wright, pp. 155-97.

Sellier K, Kneubuehl BP (1992) Wundballistik - und ihre ballistischen Grundlagen. Berlin: Springer Verlag.

Tong MJ (1972) Septic complications of war wounds. JAMA 219:1044-7.

Traverso LW, Fleming A, Johnson DE, Wongrukmitr B (1981) Combat casualties in northern Thailand: emphasis on land mine injuries and levels of amputation. Milit Med 146:682-5.

Trouwburst A, Weber BR, Dufour D (1987) Medical statistics of battlefield casualties. Injury 18:96-9.

Wang Z, Tang C, Chang X, Shi T (1988) Early pathologic characteristics of the wound track caused by fragments. J Trauma 28 (suppl):S89-95.

to previous section to next section

Please provide your feedback   English  |  French  |  Spanish