Caring for patients with burn injuries is challenging secondary to the acute disease process, chronic comorbidities, and underrepresentation in evidence-based literature. Much current practice relies on extrapolation of guidance from different patient populations and wide variations in universal practices. Identifying infections or sepsis in this hypermetabolic population is imperfect and often leads to overprescribing of antimicrobials, suboptimal dosing, and multidrug resistance. An understanding of pharmacokinetics and pharmacodynamics may aid optimization of dosing regimens to better attain treatment targets. This article provides an overview of the current status of burn infection and attempts recommendations for consideration to improve universally accepted care.
Key points
- •
Caring for patients with burn injuries is challenging secondary to the acute disease process, chronic comorbidities, and underrepresentation in evidence-based literature.
- •
Much of current practice relies on extrapolation of guidance from different patient populations and wide variations in universal practices.
- •
Identifying infections or sepsis in this hypermetabolic population is imperfect and often leads to overprescribing of antimicrobials, suboptimal dosing, and multidrug resistance.
- •
An understanding of pharmacokinetics and pharmacodynamics may aid optimization of dosing regimens to better attain treatment targets.
Introduction
Over the past century, the armamentarium available to the burn practitioner has dramatically increased. Mortality from a burn injury has been dramatically reduced with the discovery of new technology and medications, development of new surgical philosophies, and the continual expansion of burn-specific literature. Parallel to the expanding repertoire of available antimicrobials is the occurrence of multidrug-resistant (MDR) organisms. The increasing prevalence of MDR organisms is unfortunate, as they are associated with increased morbidity and mortality. Proper selection and use of antimicrobials is imperative for reducing unneeded exposure, cost, resistance, and mortality. Knowledge of existing literature and an understanding of pharmacokinetics and pharmacodynamics will aid appropriate antimicrobial prescribing and optimize patient outcomes.
Introduction
Over the past century, the armamentarium available to the burn practitioner has dramatically increased. Mortality from a burn injury has been dramatically reduced with the discovery of new technology and medications, development of new surgical philosophies, and the continual expansion of burn-specific literature. Parallel to the expanding repertoire of available antimicrobials is the occurrence of multidrug-resistant (MDR) organisms. The increasing prevalence of MDR organisms is unfortunate, as they are associated with increased morbidity and mortality. Proper selection and use of antimicrobials is imperative for reducing unneeded exposure, cost, resistance, and mortality. Knowledge of existing literature and an understanding of pharmacokinetics and pharmacodynamics will aid appropriate antimicrobial prescribing and optimize patient outcomes.
Defining infection and sepsis
Defining infection in a patient with burn injuries can be challenging in light of the hyperdynamic, hypermetabolic, and proinflammatory presentation. As a result, burn injury is often an exclusion criteria for studies of sepsis identification, treatment, and outcomes. Unfortunately, infection is a frequent accompaniment to acute illness and even more so in patients with burn injuries with a compromised primary immunologic barrier. However, the identification of early signs that distinguish infection from acute burn injury physiology is key for prompt and appropriately targeted intervention.
The diagnosis of pneumonia in a patient in the intensive care unit is controversial. Patients with acute burn injury are uniquely challenging in that they are prone to pulmonary dysfunction due to multiple noninfectious mechanisms, such as inhalational injury, pulmonary edema, dysregulated systemic inflammation, and the acute respiratory distress syndrome. Because of these and other noninfectious sources of pulmonary dysfunction, patients with burn injuries often exhibit the clinical signs and symptoms associated with infectious pneumonia. Even in the absence of bacterial pneumonia, many patients with burn injuries will have fever, purulent sputum, leukocytosis (or leukopenia), abnormal gas exchange, and infiltrates on chest imaging. There have been numerous clinical scoring systems created to aid in the diagnosis of pneumonia, such as the clinical pulmonary infection score. Such tools have not been adopted or validated in patients with burn injuries because they use many of the same clinical signs seen in patients with burn injuries without pneumonia. Most burn centers depend on bronchoscopic sampling (protected specimen brush or bronchoalveolar lavage [BAL]) or nonbronchoscopic BAL and quantitative cultures of the samples obtained to make a diagnosis of pneumonia. Pneumonia should be diagnosed if clinically suspected and BAL results in quantitative culture of ≥10 4 colony-forming units/mL.
Burn wound infections do not occur with the frequency that they did several years ago because of a more aggressive surgical approach, topical antimicrobials, and the appropriate use of systemic antibiotics. The appearance of the burn wound often holds the key to early diagnosis of the infection and thus optimal care. Therefore, it is imperative that constant wound surveillance be performed. It is stated that early eschar separation is indicative of burn wound infections, but this is rarely seen today due to early wound excision. Color changes within the wound are often the first subtle signs of infection. Conversion of partial-thickness wounds to full thickness and the loss of grafts are indicative of localized wound infections. Pseudomonas colonization may be a yellow/green exudate in the wound bed, whereas black violent areas suggest invasive infection. Typically, fungal infections are insidious. Candida infections may be more purulent in appearance, whereas Aspergillus may be gray-brown and Mucor appear as black-staining growths on the wound bed itself. Herpetic infections will appear more like punched-out lesions and often occur in healed second-degree burns. With changes occurring subtly, vigilant visual surveillance is vital for survival of tissue and sometimes the person. As Krizek and Robson have noted: “Having preceded man on earth, bacteria continue to exert a ‘territorial imperative’ and the interaction between man, his environment and his defense system is either a symbiotic relationship or one that is leading to the path of infection.”
Systemic Inflammatory Response Syndrome (SIRS) criteria have been repeatedly documented as having poor correlation with infection in patients with burn injuries, with up to 98% of patients fulfilling criteria regardless of clinical stability or infection status. Burn injury is traditionally classified in 2 phases: “ebb” and “flow.” The first 24 to 48 hours after burn injury is termed the ebb phase and is characterized by the initiation of the inflammatory process. Inflammatory mediators surge to produce local vasodilation and augment vascular permeability. The resulting albumin and fluid shifts into the interstitial space transiently produce a low cardiac output, increase systemic vascular resistance, and potential for reduced organ perfusion. After adequate resuscitation, the flow phase is characterized by the hyperdynamic response to the insult with increased cardiac contractility and output plus a reduced systemic vascular resistance.
SIRS has traditionally been considered a trigger for the initial suspicion of an infectious process in patients without burn injuries. Danger exists when extrapolating definitions and treatment protocols for sepsis validated only in patients without burn injuries, as they can lead to overtreatment with resuscitation volumes and antimicrobials. Early goal-directed therapy has improved outcomes in nonburn septic patients; however, use of recommended resuscitation volumes to reach end hemodynamic targets may lead to new unwanted issues in a fragile and often overresuscitated population. To be discussed later, overexposure to antimicrobials also must be avoided in a population with an expected prolonged hospital stay and heightened risk for MDR and fungal pathogens.
Recognizing the irrelevance in application of sepsis criteria to patients with burn injuries, recommendations, largely based on expert opinion, were published in 2007 by the American Burn Association (ABA) to serve as a catalyst. Since the consensus conference, several studies have attempted to either validate or improve the criteria for identifying sepsis in the patient with burn injuries. Mann-Salinas and colleagues suggested the model most predictive of sepsis from bacteremia included 6 variables: heart rate >130 beats per minute, mean arterial pressure <60 mm Hg, base deficit less than −6, temperature <36°C, vasoactive medications, and serum glucose >150 mg/dL. Other investigators disagree on which parameters are most predictive. Hypothermia (<36°C) is generally recognized as a predictor of poor prognosis, but disagreement exists for what febrile threshold should trigger suspicion for infection. Despite evident baseline heart rate elevations in patients with burn injuries, tachycardia appears to be an important distinguishing factor for an infectious process, provided appropriate volume repletion. It also seems clear the SIRS threshold of 90 beats per minute should not be used as a trigger in patients with burn injuries. Some studies have validated the ABA threshold, whereas greater than 130 beats per minute also has been reported as most predictive. Threshold definitions for respiratory rate greater than 25 breaths per minute, thrombocytopenia, and enteral nutrition intolerance also seem unreliable. Hyperglycemia and particularly glucose variability also appear to have strong correlation with infection.
Biomarkers
With white blood cell count and C-reactive protein as unreliable predictors, several clinicians have searched for other biomarkers that would serve as a reliable early indicator of sepsis. Procalcitonin (PCT) is a propeptide for calcitonin with cleavage occurring in the thyroid and storage found in several tissues. Additionally, monocytes and, inherently, macrophages have been shown to produce PCT. PCT’s proposed roles for immunologic modulation are to increase monocyte motility, inhibit nitric oxide synthase in vascular smooth muscle, and inactivate certain cytokines. Minimally detected in plasma under normal circumstances, PCT has been shown to release within 3 hours of endotoxin exposure and peak at approximately 14 hours with a half-life of 22.5 hours. In patients with severe renal dysfunction (creatinine clearance <30 mL/min), PCT half-life could extend to 1.5 times longer than nonrenal impairment. Plasma clearance of PCT shows weak, but significant, correlation with creatinine clearance (CrCl), as renal elimination accounts for a little more than one-third of PCT plasma elimination. Some nonseptic conditions may present with marked elevations in PCT, including trauma, burns, prolonged cardiogenic shock, heat stroke, rhabdomyolysis, liver transplantation, and severe pancreatitis.
Elevated baseline values may raise red flags for the utility of PCT in patients with burn injuries. Some investigators have shown PCT as a quality marker for early sepsis identification, whereas others disagree. Additionally, there is disagreement whether larger total body surface area (TBSA) correlates with augmented admission PCT concentrations. Due to elevated baseline values of PCT in patients with burn injuries, it may be best to raise septic thresholds to greater than 1.5 ng/mL to maximize sensitivity and specificity. Furthermore, most PCT studies have been conducted with bacteremic sources and thresholds may be variably dependent on sources with some less systemically invasive infections having lower cutoffs. If most PCT release into the plasma is a result of systemic symptoms from infection, it stands to reason, severe infections would have higher values and may be best to differentiate sepsis from local infection. PCT also has shown utility as a prognostic indicator for mortality and response to antimicrobial therapy. PCT may best serve as a daily monitored laboratory marker and incorporated as an algorithm component to indicate appropriate response to treatment and reduce duration of antimicrobial therapy. It is worth remembering that reduced elimination is expected in renal dysfunction and may reflect as a false-negative response to therapy.
One important factor may be to consider the high interpatient variability that is exhibited in the burn population. Perhaps the best avenue may be to not use a set threshold for any particular clinical or laboratory parameter, but consider each patient individually as his or her own control and monitor for acute changes. Additionally, it must be stressed to not rely on 1 single variable to reliably raise alarm for suspected infection. For the best balance of predictive values, consideration should be given to multiple factors of the complex clinical picture that is the patient with burn injury.
Prophylactic antimicrobial selection
Topicals
Beyond prevention techniques like separate bed enclosures, the shift from immersion to showering hydrotherapy, universal barrier precautions, removal of colonized invasive devices, and tight glycemic control lies the use of prophylactic antimicrobial therapy. Topical antimicrobials, like honey, have been used for wound care dating back several millennia. Today’s topical arsenal contains a wide range of agents targeted at reducing the incidence of wound infection by controlling microbial contamination at the wound surface. A potential advantage to local antimicrobial therapy is the ability to get high concentrations of the active agent at the site. Deeper thermal injuries may result in damaged blood vessels and impede delivery of systemic agents. Additionally, local therapy offers the theoretic advantage of less systemic toxicities. Most of the data relating to the efficacy of topical antimicrobials are quite dated. Recent literature has focused on disadvantages, like skin reactions, metabolic acidosis, delayed wound healing, toxicities, systemic absorption, and reduced sensitivities. Specifics regarding pharmacodynamics is a focus later in this article, but must be a consideration when selecting the most appropriate agent. Eschar penetration, safety profile, desired spectrum of activity, patient tolerance, and projected length of therapy must be balanced.
Honey offers many unique properties that have proven ideal for wound healing. Honey’s broad spectrum of activity against many bacteria and fungi is believed to be mainly due to hydrogen peroxide activity. Additional proposed properties are stimulation of tissue growth, reduced inflammatory changes, and debriding action. Unfortunately, honey may not be suitable for deep burns because of poor eschar penetration.
Sodium hypochlorite is a very potent traditional agent with a comprehensive antimicrobial profile. Unfortunately, its potency is not selective and its deleterious effect on human cells can impair wound healing. A concentration of 0.25% is frequently used; however, a dilution down to 0.00025% may provide the best efficacy/toxicity profile. Due to a favorable cytotoxic profile, some burn centers have switched product selection to more pH-balanced hypochlorous acid agents. Reduced efficacy against MDR organisms also has been reported.
Silver is another long-standing staple for the treatment of wounds. Several products exist with silver as the active moiety. Silver nitrate, silver sulfadiazine, and silver-based dressings are the most commonly used agents. Silver works quickly and may have one of the broadest spectra of activity that includes gram-positive, gram-negative, and fungal organisms. Interactions with thiol groups, release of intracellular potassium, direct cellular membrane disruption, and microbial DNA interaction are mechanisms attributed to silver’s successful antimicrobial properties. Recent data suggest the emergence of resistance and reduced efficacy to some MDR pathogens. Conflicting studies suggest that silver sulfadiazine may delay wound healing, react in sulfa-allergic patients, complicate wound evaluation secondary to the formation of a pseudoeschar, systemically absorb in larger TBSA burns, cause leukopenia, and have poor eschar penetration. Silver-based dressings offer the antimicrobial benefits of topicals with less frequent dressing changes and better patient tolerability.
First introduced as a 10% ointment, mafenide is a topical sulfonamide with broad gram-positive and gram-negative activity, particularly Pseudomonas spp. Mafenide offers the added benefit of eschar penetration for deep burns. Although rare, adverse reactions in sulfa-allergic patients, painful application, and metabolic acidosis from carbonic anhydrase inhibition have been reported. Side effects and hospital costs are minimized and efficacy retained with dilutions down to 5.0% and 2.5%. Mixed reports exist of mafenide selecting out yeast, yet retaining in vitro activity against some mold.
Systemic
Patients with burn injuries often have protracted hospital stays with multiple episodes of infection and, subsequently, spend more than half of their admission on several courses of antibiotics. The use of prophylactic antibiotics on admission for contaminated skin is sometimes used to reduce future infection, but efficacy data are inconclusive and the practice discouraged. Patients with a delay in time to presentation after burn injury may offer the best scenario for initiating antibiotics on admission secondary to higher wound-infection rates, but careful assessment of the wounds may prove the term prophylactic missplaced. As previously discussed, antimicrobial exposure can lead to resistance, need for “last resort” or combination antimicrobials, or death. Efficacy also is in question, secondary to poor perfusion in deeper wounds with extensive vessel damage. Easily overlooked, antimicrobial use and resultant changes in flora extend selection effects to the burn unit ecosystem and other patients. Prophylactic antimicrobials on admission may seem like an attractive short-term option, but may lead to a difficult abiding battle.
Burn wound manipulation extending from routine daily cleansing to more invasive burn wound excision carries a risk of bacteremia with rates increasing with invasiveness, TBSA, and time to excision. It is debated whether such bacteremia results in clinically significant sepsis or is just transient. Additionally, frequency of bacteremia has been scrutinized owing to advances in techniques and the evolution of burn care. As is often the case, current guidance for antimicrobial prophylaxis in surgery require extrapolation for application to patients with burn injuries. Current guidelines recommend using preoperative antibiotics for plastics procedures in patients with risk factors, despite lack of supporting evidence from randomized controlled trials. Considering existing evidence, it is difficult to recommend perioperative prophylaxis, especially for early excision and TBSA less than 40%. If systemic antimicrobials are used, care must be taken for appropriate agent selection, based on the local antibiogram, and frequency of dosing. Improper selection of agents and prolonged exposure may lead to emergence of selected bacteria or resistance. Consideration must be given to the differences presented by the patient with burn injuries with regard to pharmacokinetics and pharmacodynamics, as some agents may require redosing depending on length of procedure and antibiotic half-life.
Empiric
When choosing empirical antimicrobials, it is critical to consider local flora over national pathogen incidence. It is recommended that each unit keep an annual antibiogram that is unit specific and separate from that of the broader institution. Technology hurdles and low pathogen yield are frequent preclusions to annual analysis. However, a multiyear format can be used for units with low incidence of certain pathogens of concern. Knowing incidence and sensitivities to unit-specific pathogens will allow more accurate targeting of empirically prescribed antimicrobials ( Table 1 ). Exposure to broad-spectrum antibiotics can hasten development of antimicrobial resistance and increases the risk for undesirable adverse drug effects. Empiric antimicrobials should be narrowed and tailored to culture and sensitivity results as soon as they are available.
| 2015 Burn Antibiogram | Gram-Positive Percent Strains Susceptible | No. Tested | Oxacillin (%) | Vancomycin (%) | Linezolid (%) | Ampicillin (%) | Gentamicin (Synergy) (%) | Tetracycline (%) | Trimethoprim/Sulfamethoxazole (%) | Clindamycin (%) | Synercid (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Data Collected 1/15–12/15 (Hospital name) Departments of Pharmacy, Laboratory, & Infection Control Duplicate organisms excluded Blanks indicate susceptibility was not tested, not clinically applicable, or ≤45% susceptible Page ___________________ with questions | Enterococcus faecalis | 8 | — | 88 | 100 | 100 | 88 | — | — | — | — |
| Enterococcus faecium | 6 | — | 14 | 75 | 0 | 86 | — | — | — | 100 | |
| Staphylococcus aureus | 109 | 41 | 100 | 100 | — | 98 | 94 | 99 | 81 | 100 | |
| Oxacillin-Susceptible S aureus (MSSA) 41% | 45 | 100 | 100 | 100 | — | 100 | 93 | 96 | 93 | 100 | |
| Oxacillin-Resistant S aureus (MRSA) 59% | 64 | — | 100 | 100 | — | 97 | 95 | 100 | 73 | — |
| Gram-Negative Percent Strains Susceptible | No. Tested | Ampicillin/Sulbactam (%) | Piperacillin/Tazobactam (%) | Cefazolin (%) | Ceftriaxone (%) | Ceftazidime (%) | Cefepime (%) | Colistin (%) | Imipenem (%) | Aztreonam (%) | Trimethoprim/Sulfamethoxazole (%) | Ciprofloxacin (%) | Gentamicin (%) | Tobramycin (%) | Amikacin (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Pseudomonas aeruginosa | 59 | — | 96 | — | — | 87 | 89 | 100 | 83 | 93 | — | 93 | 80 | 83 | 100 |
| Enterobacter sp a | 42 | — | 100 | — | 97 | 97 | 100 | — | 100 | 97 | 97 | 100 | 100 | 100 | 100 |
| Acinetobacter sp | 33 | 61 | — | — | — | — | — | 100 | 54 | — | — | — | — | — | 84 |
| Escherichia coli a , b | 32 | 45 | 93 | 80 | 97 | 97 | 97 | — | 100 | 100 | 73 | 67 | 90 | 90 | 100 |
| Klebsiella sp a , b | 30 | 79 | 93 | 86 | 93 | 90 | 93 | — | 97 | 89 | 93 | 93 | 97 | 93 | 96 |
| Proteus sp | 20 | 95 | 100 | 68 | 90 | 94 | 95 | — | 100 | 100 | 100 | 95 | 100 | 90 | 94 |
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
