WOUND INFECTION

0039-6109/97 $0.00

WOUND HEALING

+ .20

WOUND INFECTION A Failure of Wound Healing Caused by an Imbalance of Bacteria Martin C. Robson, MD

Wound healing can no longer be thought of as a generic term. Wounds heal by various processes such as coagulation, inflammation, matrix synthesis and deposition, angiogenesis, fibroplasia, epithelialization, contraction, and r e m ~ d e l i n g The . ~ ~ Wound Healing Society has stated that when wounds proceed through these processes in an orderly and timely manner and achieve sustained anatomic and functional integrity, they are considered acute wound^.'^ When they either do not proceed in an orderly and timely fashion or do so without achieving sustained anatomic and functional integrity, they are considered chronic. Tarnuzzer and SchultzMhave suggested that repeated trauma, ischemia, and infection are leading causes of the pathobiology leading to wound chronicity. WOUND HEALING PROCESSES AFFECTED BY BACTERIA

Most individuals expect that healing is an inevitable outcome; wound healing is taken for granted. However, although healing is perceived as inevitable, it can in fact be fraught with problems and altered at many points.8 If one dkbrides the wound of nonviable tissue and repairs it in a physiologic manner, the normal phases of wound

From the Department of Surgery, Divisions of Surgical Research and Plastic Surgery, University of South Florida, Tampa; and the Department of Veterans Affairs Medical Center, Bay Pines, Florida

SURGICAL CLINICS OF NORTH AMERICA VOLUME 77 NUMBER 3 *JUNE 1997

637

638

ROBSON

healing-reaction, regeneration, and remodeling-should proceed without difficulty.26However, bacteria and bacterial products (e.g., endotoxins, metalloproteinases) can cause disturbances of this orderly scheme and affect each of the processes of healing4 The modifying effect of inflammation on wound healing was noted by Hippocrates, and Carrel and Hartmann6reported that repair was arrested when an aseptic surgical wound became infected. Carrel5 also found that an abscess far removed from the site of healing can cause a delay in the healing process. Each of the processes involved in the wound healing scheme is specifically modulated by the presence of bacteria in the tissue. For many of the processes, the response to bacteria appears to be like the yin-yang effect of pharmaceutical agents.41Low levels of bacteria seem to accentuate some processes such as fibroplasia, whereas higher tissue bacterial levels severely inhibit the process. Modulation of the individual processes, when added together, results in demonstrable clinical effects. INFECTION: AN IMBALANCE OF BACTERIA AND HOST RESISTANCE

Infection in clinical surgical practice has been defined as the product of the entrance, growth, metabolic activities, and resultant pathophysiologic effects of microorganisms in the tissues of the ~ a t i e n t Because .~ man’s biologic state is not germ-free even in the absence of clinical infection, a delicate balance must exist that allows him to survive in the presence of a great many species of bacteria, all with the potential to cause infection.38This balance is an equilibrium between the factors of host resistance and the actions of the bacteria when no infection is present. If the equilibrium remains stable, host resistance factors eventually overcome any contamination of the wound.28Once the equilibrium is upset, either by an impairment in the host defense mechanisms or by an increase in the bacterial inoculum, clinical infection may result.29 A major advance in the prevention and management of wound infection has been the understanding that the mere presence of organisms in a wound is less important than the level of bacterial growth. A wealth of clinical and experimental data has shown that a level of bacterial growth of greater than 100,000 organisms per gram of tissue is necessary to cause wound infection and the potential for invasive sepsis 38 for most species of bacteria.29, Although the concept that the numerical level of bacteria is of clinical importance for determining soft tissue and wound infection was suggested by French Army surgeons during World War the emphasis on wound bacterial levels disappeared following the war as attention shifted to the development of specific antimicrobials. Thirty to 40 years later, several unrelated observations re-established the significance of bacterial numbers. Elek demonstrated that it requires an average of 7.5 X lo6 staphylococci to produce a pustule in normal human skin and that this number could be reduced 10,000-fold in the presence of a single

WOUND INFECTION

639

silk s ~ t u r eKass . ~ reported a quantitative relationship between bacteria in urine and symptoms.14Patients with pyelonephritis had greater than 100,000 organisms per milliliter of urine. If fewer than 100,000 organisms per milliliter were present, asymptomatic bacteriuria existed. Liedburg et alZ0found that skin grafts were destroyed on rabbits when applied to beds inoculated with bacteria in concentrations greater than lo5, or 100,000 organisms per milliliter. Further evidence of a critical tissue level of bacteria above which infection occurs has been demonstrated in various situations.I2Lindsey et a121 found that goats whose experimental wounds were inoculated with Clostridium died when the clostridia level reached lo6 organisms per milliliter. Bendy et a13 showed that significant healing of decubitus ulcers occurred only when bacterial counts were less than 106/mL.They found that despite the healthy appearance of a wound, healing did not occur if the bacterial level was greater than 106/mL.Many studies from the United States Army Institute of Surgical Research have made invasive burn wound sepsis synonymous with a bacterial level of greater than lo5 organisms per gram of tissue. The author has reported several clinical scenarios in which wounds that are not in bacterial balance do not progress to successful healing. Just as Liedburg had demonstrated for experimental skin grafts, Krizek et all8 demonstrated the quantitative relationship between bacteria and skin graft survival in humans. In 50 granulating wounds, they performed quantitative bacterial cultures while preparing the wounds for grafting. Although all wounds were grafted purely on clinical grounds, when the bacterial counts were reviewed, the average graft survival was found to be 94% when the bacterial count was lo5 or fewer bacteria per gram of tissue and only 19% when the bacterial count was greater than lo5. Similar data have been reported for wounds undergoing delayed closure.39In these wounds, various topical antibacterial creams were evaluated for controlling bacteria in the wounds. The evaluation was performed using quantitative bacterial tissue cultures. The wounds were closed on clinical criteria alone, without knowledge of the bacterial counts. In the initial study, there were 40 wounds. Review of the bacterial counts performed at the time of delayed wound closure revealed that 28 of 30 wounds that contained lo5 or fewer bacteria per gram of tissue progressed to uncomplicated healing, whereas none of the 10 wound closures performed on wounds with greater than lo5 organisms per gram of tissue were successful.39This study was followed by one using quantitative bacteriology in a prospective manner. In that study, it was found that 89 of 93 wounds closed when the bacterial count was lo5 or fewer bacteria per gram of tissue progressed to rapid uncomplicated healingm When used to determine the feasibility of reclosing incisional abscesses, the quantitative estimates resulted in an average 14.3-day decrease in hospitalization. Experimentally, successful closure of wounds by pedicled flaps also depends on the bacterial load in the wound at the time of closure." In

640

ROBSON

heavily contaminated wounds containing lo6bacteria per gram of tissue, neither a random nor a musculocutaneous flap was able to prevent bacterial proliferation, and all flaps dehisced. In minimally contaminated wounds containing lo4 or fewer bacteria, both the random and musculocutaneous flaps achieved wound healing and decreased the bacterial level in the wound. However, in an intermediate group containing lo5 bacteria per gram of tissue, musculocutaneous flaps lowered the bacterial count and allowed wound closure, whereas the random flaps did not control the bacterial growth and failed.24 It is apparent from the preceding information that bacteria present in a wound at high levels can inhibit the normal wound healing processes and prevent wound closure by either direct approximation, skin graft, pedicled flap, or even spontaneous contraction and epithelialization. Wounds can result from several sources. They can be intentional wounds from a surgical incision, accidental wounds due to trauma, or the result of a disease process such as a chronic ulcer on the foot of a diabetic. The role of the surgeon in managing any soft tissue wound is first to evaluate whether the patient's balance is in equilibrium or upset in favor of the bacteria.28If equilibrium exists, all efforts must be expended to maintain this status and prevent an ensuing infection. If equilibrium does not exist, infection is present, and the management of the infection is directed at re-establishing the e q u i l i b r i ~ mAlthough .~~ the specifics may differ slightly for the various causes, the principles of diagnosis and treatment are the same. WOUND INFECTION DIAGNOSIS BY BACTERIAL ANALYSES

Definitive diagnosis of infection is made by the presence of purulent material draining from the wound or spreading inflammation greater than the normal inflammation of healingz8It can also be made from the number of bacteria per gram of wound tissue.%Quantitative cultures of tissue biopsies that reveal greater than lo5bacteria per gram of tissue or the presence of beta-hemolytic streptococci suggest clinical infection. Quantitative bacteriologic cultures can help to differentiate between the contaminated and the grossly infected wound. A quantitative tissue culture can be obtained by cleansing the wound surface, removing a tissue specimen, and aseptically weighing, flaming, and homogenizing it after 10-fold dilution with thioglycolate. Serial tube dilutions and pour plates or backplating can then yield an accurate colony count per gram of Such detailed analyses are of little value in determining immediate wound management because they require 24 to 36 hours for completion. However, the important information derived from this technique can be obtained in 15 minutes with the rapid slide method.'l A specimen is removed from the wound, weighed, diluted, and homogenized as be-

WOUND INFECTION

641

fore, but a 0.02-mL aliquot of the suspension is then placed on a glass slide and stained with Gram stain. The slide can then be examined under a microscope. If even a single organism is seen on the slide, the bacterial count in the original biopsy is greater than lo5 organisms per gram of tissue. If no organisms are seen on the slide, the bacterial count is lo5 or fewer per gram of tissue. Qualitative bacteriology is also important. When a tissue biopsy specimen is removed, a portion of tissue should be qualitatively analyzed under both aerobic and anaerobic conditions. This allows for identification of species and antibacterial sensitivities. The antibacterial sensitivity studies should be performed for both systemic and topical antimicrobials. One important reason for qualitative analysis is to identify wounds that are contaminated or infected by the beta-hemolytic streptococci.28These organisms can cause infection and prevent satisfactory wound management if they are present in soft tissue at any level. The specific vagaries of these organisms have been described elsewhere.37 Unfortunately, the precision provided by quantitative bacteriologic techniques has not been used in most reported clinical series. Positive qualitative bacterial cultures may be obtained from fresh operative wounds or postoperative wound drainage and may be followed by normal uncomplicated healing. Similarly, signs of inflammation and even purulent drainage may occur in a wound from which no viable bacteria can be recovered by standard culture techniques. Therefore, a workable definition of infection must be based on uniform criteria that can be easily applied to the clinical circumstan~e.~~ A wound can be said to be uninfected if it heals primarily without discharge. It can be assumed to be infected if discharge of purulent material occurs, even if no bacteria can be recovered by culture. Such drainage may represent dead bacteria, white cells, or other debris that yield negative However, quantitative biopsy cultures of the wound tissue itself might well yield quantitatively significant levels of bacteria if such cultures were routinely obtained. It is clinically necessary to include wounds that are inflamed but do not drain or in which culture-positive drainage but no actual pus occurs. These wounds are referred to as ”possibly infected.”’ An exception to these categories is the “stitch abscess.” Purulent drainage from suture sites is not considered to be infection if the inflammation and drainage are minimal and confined to the suture site. It is also important that the wound heals itself primarily and the suture sites clear within 72 hours after removal of the sutures.l,28 SOURCES OF BACTERIA CAUSING INCISIONAL WOUND INFECTION

All wounds, either elective or traumatic, are contaminated to a degree.29The exogenous inoculum of bacteria delivered to the wound

642

ROBSON

must be minimized. The contributions of Lister, Semmelweis, and von Bergmann established the standards of asepsis and antisepsis to maintain the environmental bacteria at a minimal level. Because of their contributions, a c1e.m operation such as a thyroidectomy or herniorrhaphy can be performed with only a negligible chance of a postoperative wound infection. A cooperative study by the National Research Council provided ample evidence that postoperative wound infections are not primarily due to exogenous airborne bacteria entering the wound at the time of the 0peration.l Despite a 31% to 74% reduction of airborne bacteria by use of ultraviolet light in the operating rooms, the overall postoperative infection rate was not reduced. In only one category of wound, the “refined clean” wound, did postoperative wound infection significantly decrease. Even though it is known that 30,000 to 60,000 bacteria can fall into a 3- to 4-sq m operating room field each hour, little correlation was found among the bacteria isolated from the air, the wounds at the time of operation, and resultant postoperative wound infections.’ However, the rate of infection increases linearly with the duration of the operation. Even when adjusted for type of operation, the size of incision, and the greater amount of tissue trauma, this increased infection rate persists, ranging from 3.6% for operations lasting less than 30 minutes to 16.4% for operations lasting more than 5 hours.’ Although evidence is lacking to support the exogenous source of bacteria as the usual causative agent for wound infection, overwhelming evidence documents endogenous sources.29Operative incisional infection can arise from endogenous bacteria by two separate mechanisms. The first, described by Miles et alB as “primary lodgement,” occurs when bacteria from within the body come into contact with the edges of the incision. Incisions appear to be susceptible by this route for a period of 3 to 5 hours. Local and systemic factors affecting host resistance have been shown to be capable of enhancing this primary lodgement.29 Examples implicating endogenous bacteria in subsequent incisional wound infections are replete in the literature. The increasing incidence rates of postoperative wound infections in operative categories progressing from refined clean to dirty suggest an endogenous mechanism.’ The author has reported two prospective clinical series implicating endogenous bacteria in subsequent surgical wound infections. In 100 patients who had undergone appendectomy, 19 of 21 patients in whom significant levels of incisional bacteria were found postoperatively had previously demonstrated bacteria in the peritoneal fluid at the time of the appende~tomy.~~ Similarly, 65 patients undergoing cholecystectomy were studied in a prospective manner by bacterial analyses on bile and on the gallbladder walls.3oPositive biliary tract cultures were found in 8 of 10 patients with incisional abscesses. Both of these series support the concept of an endogenous source of ”primary lodgement.” Patients with an established infection somewhere in the body re-

WOUND INFECTION

643

mote from the wound, such as in a skin carbuncle or in the urinary bladder, are at an increased risk for wound sepsis by a second mechanism of endogenous infection. Such a patient periodically experiences a bacterial invasion of the bloodstream and lymphatic system. The wound serves as a locus minoris resistentiae, and bacteria from a distant site enter the wound, resulting in an i n f e ~ t i o n . ~ ~ Krizek and DavW demonstrated that rat wounds could become infected from infection elsewhere in the body. Distant subcutaneous inoculations of Staphylococcus auyeus were made simultaneously with the time of wounding or 24 hours prior to wounding. Bacteria at levels of lo8 organisms per gram of tissue in the subcutaneous site can result in incisional wound levels of bacteria of greater than lo6 bacteria per gram of tissue. Remote infection clinically has been associated with a threefold increase in wound infection rates over those in patients in whom no remote infection existed.l ACUTELY CONTAMINATED TRAUMATIC WOUND

The above sources of bacteria occur in intentional operative wound incisions made with aseptic techniques. Traumatic wounds have obvious sources of bacteria, both exogenous and endogenous. All traumatic wounds are contaminated at least to the extent that bacteria can always be identified by cultures performed on tissue biopsies of specimens.2 Normally, the bacterial count of skin is quite high. These bacteria reside both on the surface of the skin and deep in the hair follicles and sweat glands. The amount of bacteria normally present in the recesses is 1000 organisms per gram of tissue.38 Pulaski et al,25 in 1941, found on culturing 200 fresh traumatic wounds aerobically and anaerobically that all contained organisms, the dirty wounds more than the clean. Even clean-appearing wounds may harbor organisms with sufficient frequency to suspect their presence in every case. No one can tell which of these wounds is going to develop infection. In a series of 80 emergency department wounds, 20% yielded at least lo5 organisms per gram of tissue.3I Kovaric et al,I5 studying combat wounds in Viet Nam, found the level of this contamination to be lo3 organisms per milliliter in 70% of the wounds.15A level of greater than lo5 would have resulted in clinical infection. Time is another important factor in predicting a wound’s bacterial imbalance.28In the emergency department study previously cited, the mean time from injury of patients with less than lo2 bacteria per gram The mean time from injury of tissue in their wounds was 2.2 increased to 3 hours in patients who had lo2 to lo5 bacteria per gram of wound tissue. In patients who presented to the emergency department with greater than lo5 organisms per gram of tissue even before being treated, the mean time from wounding was 5.17 hours. More importantly, only those in the last group, with more than lo5bacteria per gram of tissue, developed clinical infection that prevented primary healing.

644

ROBSON

Surgeons have always believed that wounds have a ”period of grace” or “golden period” during which time they could be closed with lesser fear of infection. From these data, this period of grace appears to be the time required for a given inoculum in the wound to reach the apparently critical level of lo5 bacteria per gram of tissue.28,31 CHRONICALLY CONTAMINATED WOUND

Chronically contaminated wounds all contain a tissue bacterial flora. Examples are thermal burns, venous stasis ulcers, diabetic foot ulcers, and traumatic wounds that were not closed acutely. The characteristic of these wounds is granulation tissue. Granulation tissue does not occur in the absence of bacteria. It is not found beneath the surface of a successfully closed wound. It has been likened to a pyogenic granuloma, and successful closure is based on the surgeon’s ability to control the level of bacterial Tarnuzzer and Schultz4 have suggested that chronicity begins with bacteria. This persistent tissue level of bacteria results in prolonged elevation of proinflammatory cytokines such as interleukin-1 and tumor necrosis factor-a. This condition in turn causes increased levels of matrix metalloproteinases, a decreased level of tissue inhibitors to the metalloproteinases (TIMP), and a decreased production of growth factors. TREATMENT OF WOUND INFECTION: RE-ESTABLISHMENT OF THE BACTERIAL BALANCE AND WOUND CLOSURE

The various types of wound infections discussed are all considered ”surgical” infections. They differ from medical infections such as pneumonia or pyelonephritis in that they require some type of operative intervention in their treatment. For medical infections, systemic antibiotic therapy is clearly the treatment of choice. However, wound infections require surgical principles such as drainage of an abscess, dkbridement of necrotic tissue, and well-timed wound closure.29All surgical therapy in the infected wound must be directed toward establishing the balance and rendering the local environment and bacterial flora suitable for closure.29Therefore, in addition to drainage, surgical dkbridement must remove residual necrotic tissue, hematoma, and foreign bodies, including sutures. The specific wound infections require somewhat different techniques to re-establish equilibrium. Treatment of the Infected Incision

Compared with the other types of wound infection to be discussed, the operative incision is one that responds best to preventive measures.

WOUND INFECTION

645

/’Clean,” ”clean-contaminated,” and even ”contaminated” wounds are usually in bacterial balance.28 Tissues from ”contaminated” wounds rarely yield bacterial counts in excess of lo5 bacteria per gram of tissue. Only in ”dirty” wounds would one routinely expect large numbers of bacteria. If the bacterial inoculum is not excessive, as would be expected in the majority of incisions, infection can be prevented by maintaining the host defenses at peak efficiency. The effect of the interaction between the bacteria and the host, although under systemic influence, is ultimately determined by the local factors in the Among these are necrotic tissue, decreased local wound perfusion, foreign body, hematoma, and dead space.” Surgeons can do little to change systemic host defense factors such as age, obesity, malnutrition, chronic steroid administration, and specific immune defects. However, the surgeon plays a significant role in eliminating the local deterrents to effective host defense in the If local wound factors are not controlled, they upset the equilibrium. Circumstances are then produced in which a normally subinfectious inoculum of bacteria may multiply to levels sufficient to produce infection. The most important factor introduced in the past half century to alter the balance between host resistance and bacteria has been the use of antibacterial agents.2yThe administration of these agents at the proper time, by the proper route, and in the proper dosage to be therapeutically effective should theoretically be able to prevent or treat infection from organisms susceptible to the agent administered?*However, the proliferating use of antibiotics has failed to eliminate infection. In fact, it has led to many more abuses than effective uses. One cannot and should not desire to create a “germ-free” patient. Systemic antibiotics have practical and potential value only if a therapeutic blood level or, more importantly, tissue level is achieved within the first 4 hours after wounding. Burke4has shown that bacterial lodgement is not influenced after that time. When antibiotics are begun after the time required for bacterial lodgement, infection rates are higher than when no antibiotics are used. A review of the principles of prophylactic antibiotic usage is beyond the scope of this paper, so the reader is referred to the excellent discussion of the subject by Meakins.22 When prevention is not successful, an incisional wound infection occurs. A grossly infected wound is one in which purulent material is obtained from the wound or one with surrounding cellulitis and inflammation beyond that appropriate for normal wound healing. The treatment of these two clinical presentations is different. For the wound with spreading cellulitis and no loculation of pus or drainage, systemic antibiotics can be usefuLZ8If the wound is not opened, extreme vigilance is required so as not to miss an inapparent loculation. If the wound quickly responds to the systemic antibiotics, the original closure may be maintained. If the wound drains purulent material or a loculation is suspected, the wound should be opened. It is important not to get too concerned

646

ROBSON

about the purulent material within the wound. It should be irrigated from the wound, and material for a tissue biopsy should be obtained from the wall of the abscess. In reported studies, the wound biopsy revealed the causative organism in 87% of cases.36However, cultures of the purulent exudate often reveal multiple organisms and not those responsible for tissue invasion. A Gram stain of any purulent material may be of use to identify clostridia when they are present. Culture of the purulent material is not as useful as culture of tissue from the wound 29 The fact that tissue biopsies usually yield a single bacterial species is very helpful if systemic antibiotics are chosen in the treatment of the Once the wound has been prepared and a tissue biopsy has identified the causative organism,. the goal is to re-establish the bacterial balance in the wound. The techniques for this are the same as those for a chronically contaminated wound, to be discussed later. Topical antimicrobials and temporary biologic dressings are useful. When the wound is in bacterial balance (lo5 bacteria per gram of tissue or fewer), delayed closure can be performed. HoweI3 has stated that once a contaminated wound has been converted to a "clean" wound, it can be treated as any other wound and surgically closed. In a series of delayed closures based on bacterial counts reported by the author, incisional abscesses were reclosed with a 96% success rate.4O In localized incisional abscesses that have been drained and are being treated as chronically contaminated wounds, systemic antibiotics are not used. They are reserved only for those wounds in which infection appears to be spreading through the subcutaneous tissue even after drainage of the wound and dkbridement of the infected soft tissue.28 Treatment of the Acutely Contaminated Wound

As previously discussed, all traumatic wounds should be considered contaminated. However, not all are infected and present with greater than lo5 bacteria per gram of tissue. The history of wounding, the wounding agent, wound location, and timing can all help suggest the level of bacteria in a wound.& If the suspicion is high or if the patient presents more than 5 hours after injury, a tissue biopsy and rapid slide evaluation are Management of the acutely contaminated wound is undertaken with the goal of re-establishing the bacterial balance, if indeed it is weighted in favor of the bacteria, and managing the wound as if it were a clean one.28The mainstay of management is sharp dbbridement, which indeed is the sine qua non of contaminated wound management. All debris and nonvitalized tissue should be excised. The edges of the wound should be freshened and, when possible, made perpendicular to facilitate closure. Irrigation is an important adjunct to sharp dbbridement. Low-pressure wound irrigation removes little but surface contamination, even with voluminous amounts of solutions. Therefore, irrigation should be

WOUND INFECTION

647

high pressure and intermittent.28In an experimental animal system, the pulsating jet lavage was shown to remove significantly more bacteria and result in fewer wound infections than the more conventional lowpressure irrigation.l0If the wound is small and has minimal contamination, a modification of the pulsating jet lavage can be constructed using a syringe with a 19-gauge needle.43Again, irrigation is most efficiently done with an intermittent-pressure technique so as to allow the normal elasticity of soft tissue to assist in removal of bacteria and foreign bodies.28 Systemic antibiotics are of little use in treatment of the acutely contaminated wound. They have practical and potential value only if a tissue level can be achieved within the first 4 hours after wounding, which is almost impossible in the acutely contaminated traumatic wound. When antibiotics are begun after the time required for bacterial lodgement, infection rates may be higher than when no antibiotics are used.', 28 Surgical debridement and freshening of the wound may render this restriction less rigid and make antibiotics more effective. Also, recent data about the role of inflammatory mediators suggest that once vasoconstricting mediators are present, wound penetration of antibiotics may be less efficient.35Therefore, debridement and adjunctive irrigation remain the mainstays of treatment of the acutely contaminated wound. Once bacterial balance has been re-established, wound closure can proceed as for a clean wound. If circumstances suggest the need for a tissue biopsy, either a rapid slide evaluation or full bacterial quantitative and qualitative analyses can be performed. When lo5 or fewer bacteria per gram of tissue are present, wound closure can proceed. However, if greater than lo5 bacteria per gram of tissue are present, the wound should be temporarily left open for delayed closure. The principles learned in military surgery are equally operative in potentially contaminated civilian wounds; that is, when in doubt, it is safer to leave a wound open.28It has been repeatedly shown that delayed closures can be performed as easily as those for acute clean wounds if the level of bacteria has been sufficiently reduced to re-establish the bacterial balance. Treatment of the Chronically Contaminated Wound

As defined earlier, when granulation tissue is present and a wound is not proceeding through a timely and orderly healing trajectory, it is a chronic wound.lg These wounds are frequently not in bacterial balance. Not only is the inoculum great but often the bacterial host defenses are impaired, as in the diabetic foot ulcer. It has been shown in blinded studies that it is clinically impossible to predict bacterial balance in a chronic granulating wound.I8 In some wounds, creeping epithelium from the wound edges has been a useful sign in predicting when a wound can be successfully closed. However, statistically, the best method has been the use of quantitative bacteriol-

648

ROBSON

ogy on a tissue biopsy specimen.18,

39, 40 When wounds have greater than lo5 bacteria per gram of tissue or beta-hemolytic streptococci are present, they cannot be successfully closed. Therefore, the goal of management in these chronically contaminated wounds with granulation tissue is to decrease the bacterial count in the granulation tissue to lo5 or fewer bacteria per gram of tissue. Management of these wounds consists of various techniques to decrease the bacterial count. Reduction of the bacterial flora is best accomplished by meticulous attention to surgical detail. Frequent dkbridement and surgical cleansing are critical. Enzymatic dkbridement may be of some value, but it tends to allow simultaneous rapid bacterial pr01iferation.l~This is a potential danger, particularly when large areas are involved, such as in thermal burns. Systemic antibiotics have been demonstrated not to reach adequate tissue levels in chronic granulation tissue and to have no effect on the bacterial level in granulating Conversely, water-based topical antibacterial creams do penetrate the depths of such wounds and have a direct effect on bacterial growth. Because true bacterial control is achieved only by wound closure, the use of temporary biologic dressings for cleansing and temporary closure has been shown to be of value. In a randomized study of 100 delayed wound closures, the authors compared several methods to decrease the bacteria to below the critical Temporary biologic dressings proved to be the most effective. The biologic dressings probably adhered to the wound surface and affected a “biologic closure” that allowed the inflammatory tissue to function at peak efficiency and phagocytosis to proceed effectively.” However, true control of infection in a granulating wound is closure of the wound with autologous tissue. Techniques of closure in chronically contaminated wounds vary widely, depending on the circumstances. However, once the wound is in bacterial balance, with lo5 or fewer bacteria per gram of tissue and no beta-hemolytic streptococci in the wound, closure by direct approximation, skin graft, or flap can be predictably successful. Successful closure by one of these methods eradicates the remaining bacteria in the wound.

SUMMARY

Infection in a wound, like infection elsewhere in the body, is a manifestation of a disturbed host-bacteria equilibrium in favor of the bacteria. This not only elicits a systemic septic response but actually inhibits the multiple processes involved in the wound healing scheme. Each process involved in healing is affected when bacteria proliferate in a wound. Wound infection, whether in an intentional operative incision, an acute traumatic laceration, or a chronic pressure ulcer, results when bacteria indigenous to the patient or exogenous to the wound achieve dominance over the systemic and local factors of host resistance. To be able to prevent and manage wound infections requires an understanding

WOUND INFECTION

649

of how each prophylactic or therapeutic maneuver works to maintain or re-establish the bacteria-host balance. Only when this equilibrium is in balance can the normal processes of wound healing proceed to give a satisfactory healing trajectory.

References 1. Ad Hoc Committee of the Committee on Trauma, Division of Medical Sciences, National Research Council: Report: Postoperative wound infections; the influence of ultraviolet irradiation of the operating room and the influence of various other factors. AM Surg 16O(Suppl l):l, 1964 2. Altemeier WA, Gibbs EW: Bacterial flora of fresh accidental wounds. Surg Gynecol Obstet 78164, 1944 3. Bendy RH, Nuccio PA, Wolfe E, et a1 Relationship of quantitative wound bacterial counts to healing of decubiti. Effect of topical gentamicin. Antimicrob Agents Chemother 4:147, 1964 4. Burke J F The effective period of preventive antibiotic action in experimental incisions and dermal lesions. Surgery 50161, 1961 5. Carrel A Cicatrization of wounds. XII. Factors initiating regeneration. J Exp Med 34:425, 1921 6. Carrel A, Hartmann A: Cicatrization of wounds: The relation between the size of the wound and the rate of its cicatrization. J Exp Med 24:429, 1916 7. Committee on Control of Surgical Infections of the Committee on Pre and Postoperative Care of the American College of Surgeons: Manual on Control of Infection in Surgical Patients. Philadelphia, JB Lippincott, 1976 8. Cooper DM, Robson MC: Wound healing: Infection. J Anasth Intensivbehandlung 3S25, 1996 9. Elek S D Experimental staphylococcal infections in the skin of man. Ann NY Acad Sci 65:85, 1956 10. Hamer MI, Robson MC, Krizek TJ, et al: Quantitative bacterial analyses of comparative wound irrigations. Ann Surg 1812319, 1975 11. Heggers JP, Robson MC, Doran ET: The quantitative assessment of open wounds by a slide technique. Trans R SOCTrop Med Hyg 63:532, 1969 12. Heggers JP, Robson MC: Quantitative Bacteriology: Its Role in the Armamentarium of a Surgeon. Boca Raton, FL, CRC Press, 1991 13. Howe CW: The early closure of constantly contaminated infected wounds with the aid of urethane-penicillin mixtures. Surg Gynecol Obstet 87425, 1948 14. Kass E H Bacteriuria and the diagnosis of infections of the urinary tract. Arch Intern Med 100:709, 1957 15. Kovaric JT, Matsumoto T, Dobek AS, et al: Bacterial flora of one hundred and twelve combat wounds. Mil Med 133:622,1968 16. Krizek TJ, Davis JH: Endogenous wound infection. J Trauma 6:239, 1966 17. Krizek TJ, Robson MC, Groskin MG: Experimental burn wound sepsis: Evaluation of enzymatic debridement. J Surg Res 17219, 1974 18. Krizek TJ, Robson MC, Kho E: Bacterial growth and skin graft survival. Surg Forum 18:518, 1967 19. Lazarus GS, Cooper DM, Knighton DR, et al: Definitions and guidelines for assessment of wounds and evaluation of healing. Arch Dermatol 130489, 1994 20. Liedburg NCF, Reiss E, Artz CP: The effects of bacteria on the take of split-thickness skin grafts in rabbits. AM Surg 14292, 1955 21. Lindsay D, Wise HM, Knocht MS, et al: The role of clostridia in mortality following an experimental wound in the goat. I. Quantitative bacteriology. Surgery 45:602, 1959 22. Meakins J L Surgical Infections: Diagnosis and Treatment. New York, Scientific American, 1994

650

ROBSON

23. Miles AA, Miles EM, Burke J: The value and duration of defense reactions of the skin to the primary lodgement of bacteria. Br J Exp Pathol3879,1957 24. Murphy RC, Robson MC, Heggers JP, et al: The effect of microbial contamination on musculocutaneous and random flaps. J Surg Res 41:75, 1986 25. Pulaski EJ, Meleney FL, Spaeth WLC: Bacterial flora of acute traumatic wounds. Surg Gynecol Obstet 72982,1941 26. Robson MC: Disturbances in wound healing. Ann Emerg Med 171274, 1988 27. Robson MC: Exogenous growth factor application effect on human wound healing. Prog Dermatol301,1996 28. Robson MC: Infected soft tissue wounds. In Cameron: Current Therapy in Plastic and Reconstructive Surgery. Philadelphia, BC Decker, 1989 29. Robson MC: Infection in the surgical patient: An imbalance in the normal equilibrium. Clin Plast Surg 6:493, 1979 30. Robson MC, Bogart JN, Heggers JP: An endogenous source for wound infections based on quantitative bacteriology of the biliary tract. Surgery 86471,1970 31. Robson MC, Duke WF, Krizek TJ:-Rapidbacterial screening in the treatment of civilian wounds. J Surg Res 14:426, 1973 32. Robson MC, Edstrom LE, Krizek TJ, et al: The efficacy of systemic antibiotics in the treatment of granulating wounds. J Surg Res 16:299, 1974 33. Robson MC, Funderburk MS, Heggers JP, et al: Bacterial quantification of peritoneal exudates. Surg Gynecol Obstet 130267, 1970 34. Robson MC, Heggers JP: Bacterial quantification of open wounds. Mil Med 134:19,1969 35. Robson MC, Heggers JP: Quantitative bacteriology and inflammatory mediators in soft tissue. In Hunt TK, Heppenstall RB, Pines E, et a1 (eds): Soft and Hard Tissue Repair. New York, Praeger, 1984 36. Robson MC, Heggers J P Surgical infection I. Single bacterial species or polymicrobic in origin? Surgery 65:608, 1969 37. Robson MC, Heggers JP: Surgical infection II: The beta hemolytic streptococcus. J Surg Res 9:289, 1969 38. Robson MC, Krizek TJ, Heggers JP: Biology of surgical infection. In Ravitch MM (ed): Current Problems in Surgery. Chicago, Year Book Medical Publishers, 1973 39. Robson MC, Lea CE, Dalton JB, et al: Quantitative bacteriology and delayed wound closure. Surg Forum 19:501,1968 40. Robson MC, Shaw RC, Heggers JP: The reclosure of postoperative incisional abscesses based on bacterial quantification of the wound. Ann Surg 171:279, 1970 41. Robson MC, Stenberg BD, Heggers JP: Wound healing alterations caused by bacteria. Clin Plast Surg 17485, 1990 42. Robson MC, Zachary LS Repair of traumatic cutaneous injuries involving the skin and soft tissue. In Georgiade GS, Georgiade NG, Riefkohl R, et al: Textbook of Plastic, Maxillofacial, and Reconstructive Surgery, ed 2. Baltimore, Williams & Wilkins, 1992 43. Stevenson TR, Thacker JG, Rodeheaver GT, et al: Cleansing the traumatic wound by high pressure syringe irrigation. Journal of the American College of Experimental Pathology 517, 1976 44. Tamuzzer RW, Schultz G S Biochemical analysis of acute and chronic wound environments. Wound Repair and Regeneration 4321, 1996

Address reprint requests to Martin C. Robson, MD Surgical Service (112) Bay Pines VAMC 10000 Bay Pines Boulevard Bay Pines, FL 33744