Healthcare-Associated Infections in Burn Patients

Hayley Elisabeth Cunningham

David van Duin

Anne Monica Lachiewicz

Despite notable advances in burn prevention and wound care, burns remain a substantial cause of morbidity and mortality. Worldwide, 180 000 people die annually from flame burns alone, with scald, electrical, and chemical burns contributing to additional death and disability.¹ Developing nations are disproportionately affected due to limited prevention measures and access to high-quality wound care, with child mortality in these countries seven times higher than in high-income nations.¹ Even with advanced prevention and treatment modalities, over 486 000 burn injuries were medically treated in the United States (US) in 2016, including 40 000 necessitating hospital admission and 3275 resulting in death.² Mortality from burn injuries reported in the United States between 2009 and 2018 was 3.0% overall. Burn types in the United States reflect those worldwide, with flame and scald injury predominating at 40.6% and 31.4% of all burns, respectively.³ As such, burn wound literature and guidelines primarily pertain to flame, and to a lesser degree scald, types.

While total body surface (TBSA) of the burn and inhalation injury have the greatest impact on mortality risk, a number of additional factors increase burn mortality risk, including advanced age and preexisting comorbidities.⁴ Most risk factors increase patients’ susceptibility to infectious complications,⁴^,⁵ which are responsible for 42%-65% of burn deaths⁶^,⁷^,⁸ and are associated with mortality rates three to six times higher than those without infection.⁹^,¹⁰ Pneumonia occurs in 20% of those with inhalation injury and in 65% who require mechanical ventilation.¹¹ Urinary tract infection (UTI) and cellulitis, the second and third most prevalent complications, are relatively less common but substantial contributors to morbidity and mortality.³

Infection prevention, diagnosis, treatment, and control in burn patients present unique challenges given the resulting state of immune compromise. Virulent pathogens and modes of transmission abound in the hospital setting, particularly in burn intensive care units (BICUs), as many patients require lengthy hospital stays, frequent dressing changes, Foley catheters, peripheral or central intravenous (IV) catheters, and mechanical ventilation.⁶^,¹² In addition to destroying the skin’s physical barrier, burns trigger metabolic and immunologic responses¹³^,¹⁴^,¹⁵^,¹⁶ that predispose patients to infection and cause signs and symptoms that mimic infection.¹⁷^,¹⁸^,¹⁹ Such responses can also alter antimicrobial pharmacokinetics,²⁰^,²¹ further complicating management. Overuse and underdosing of antimicrobials contribute to the growing problem of antimicrobial resistance,⁶ with rates of multidrug resistance for some pathogens reaching over 50%.⁹^,²²^,²³ It is important to understand the unique characteristics of burn wounds and associated infections to provide appropriate patient care while practicing antimicrobial stewardship to control the emergence of multidrug-resistant (MDR) pathogens.⁶

PHYSIOLOGIC RESPONSE TO BURN INJURY

At the site of burn wound injury, dead tissue constituting the eschar is surrounded by a tenuous area of vascular stasis and hypoperfusion within a wider zone of hyperemia in response to injury.²⁴ Hypercoagulability and oxidative stress lead to thrombosis and endothelial cell damage, respectively, while increased vascular permeability produces edema. As blood flow to and from the area is compromised, toxic cytokines and free radicals accumulate,¹³ killing initially viable tissues and fueling an inflammatory and hypermetabolic response. Prompt surgical excision of the burn eschar has become the standard of care, as this decreases mortality by removing much of the inflammatory stimulus and nidus for infection.²⁵^,²⁶^,²⁷

Systemic inflammation in response to burn injury can be profound, leading to multiorgan failure or shock, and is closely followed by a compensatory anti-inflammatory response.⁷^,²⁴ The resulting immune dysregulation produces a state of immunocompromise and complicates the diagnosis of systemic infection.⁷^,²⁴ Hyperinflammation can impair physiological barriers in multiple organ systems, allowing inflammatory cells, reactive oxygen species, and other physiologic insults to contribute to multiorgan dysfunction or failure.¹⁵ For example, postburn ileus and changes in intestinal permeability enable bacterial translocation that can lead to sepsis, particularly in burns associated with concomitant alcohol use.²⁴^,²⁸

A multitude of immune cells and inflammatory mediators contribute to postburn immune dysregulation. Severe burns impair the chemotaxis and function of monocytes, macrophages, and neutrophils responsible for clearing dead tissue, bacteria, and cytotoxins.²⁹^,³⁰ In addition to immune dysregulation, burn injury induces hypermetabolism
correlating with the extent of burn injury that contributes to mortality¹⁴ and can persist for up to 3 years.³¹ This hypermetabolic response contributes to multiorgan dysfunction by increasing systemic inflammation, promoting skeletal and cardiac muscle catabolism, and increasing the risk for sepsis.³²^,³³ Catecholamines, stress hormones, and cytokines play a significant role by inducing insulin resistance and alterations in glycolysis, proteolysis, and lipolysis; the result is loss of body mass, impaired wound healing and immune function, and greater risk for infection.³²^,³³

PATHOGENESIS AND EPIDEMIOLOGY OF BURN-ASSOCIATED INFECTIONS

Despite surgical excision of the eschar, burn wounds are colonized by Gram-positive skin flora within 48 hours of injury, by Gram-negative bacteria within 1 week, and lastly by fungi after roughly 2 weeks postburn.³⁴^,³⁵^,³⁶^,³⁷ While Staphylococcus aureus predominates in the first week following injury, it is surpassed by Pseudomonas aeruginosa by week 3.³⁴ Median time from admission to isolation of Streptococcus spp., S aureus, and Enterococcus spp. from burn patients may be 2, 3, and 9 days, respectively, compared with 11.5, 18, and 26 days for isolation of Enterobacteriaceae, Pseudomonas spp., and Acinetobacter spp., respectively.³⁸ P aeruginosa is rarely found within 7 days of admission but may colonize over half of patients after 1 month of hospitalization.³⁹ Colonization progresses to infection in many cases, as patients’ compromised immune systems are challenged by nosocomial pathogens, frequent wound dressing changes, IV and bladder catheters, and mechanical ventilation.¹² Hospital-acquired infection is common among burn patients, but reported rates vary widely, likely reflecting variation in the severity of burns treated at different institutions. Among all patients hospitalized for flame burns, rates of pneumonia, UTI, and cellulitis are ˜4.1%, 2.4%, and 1.7%, respectively.³ In severely burned patients, wound, skin, or soft tissue infections may develop in 19%-56% during hospitalization, while the rates of pneumonia, bloodstream infections, and UTIs may reach 10.5%-40%, 12%-25%, and 3%-22%, respectively.⁵^,⁹^,¹⁰^,²³^,³⁸^,⁴⁰ Burn wound, skin, and soft tissue infections generally occur within the first week, while UTIs and pneumonia tend to develop at least 30 days after hospitalization.³⁸

The elements required for colonization and subsequent infection include a source or reservoir, a means of transmission to the wound, and risk factors that enable colonization and invasion.⁴¹ Sources include patients’ skin, upper respiratory and gastrointestinal (GI) tracts, fomites in the hospital environment, and other patients.²⁴^,⁴¹ Since surgical excision and antimicrobial dressings have become standard of care, burn wounds are less frequently the reservoir for infection.²³ Vehicles for transmission include the hands and clothing of healthcare personnel,⁴²^,⁴³^,⁴⁴ mattresses and other environmental surfaces,⁴²^,⁴³ and medical equipment including nonsterile exam gloves⁴⁵ and hydrotherapy equipment.⁴³^,⁴⁶ Wound care performed in patients’ rooms has largely replaced submersion hydrotherapy in response to multiple outbreaks related to this practice.²⁴ Patients themselves can serve as modes of transmission when their hands contact contaminated surfaces or other patients and transfer pathogens to their burn wound, IV lines, respiratory mucosa, or other entry sites.⁴¹ Additionally, GI colonization with virulent Gram-negative pathogens, such as P aeruginosa and Acinetobacter baumannii complex, can lead to infection via bowel translocation or transmission of fecal material to any of the aforementioned entry sites.⁴⁷^,⁴⁸^,⁴⁹^,⁵⁰

Risk factors for infection are common in hospitalized burn patients. TBSA plays a prominent role in patient outcomes, as the degree of burn injury largely determines the level of immunocompromise and the need for lengthy hospital stays and invasive supportive measures.⁵^,⁹^,²²^,⁵¹^,⁵²^,⁵³ Compared with patients with <5% TBSA, the risk for infection in those with 10%-20% TBSA and >20% TBSA is at least 6 and 10 times higher, respectively.⁵³ Mortality may reach 24% when burns cover 40% TBSA, and over half of patients with >70% TBSA burns will die.³ Greater TBSA is often associated with inhalation injury and need for mechanical ventilation, which independently significantly increase infection risk.³^,⁵^,⁶^,⁵²^,⁵³ Other invasive interventions required in patients with extensive burns, including blood transfusions,⁴³ IV catheters,⁶^,²³ and Foley catheters⁶^,²³ also increase infectious risk.

TBSA of 40%-45% in adult burn patients significantly increases the risk for adverse outcomes,⁵¹ but the effect varies with age. For example, the TBSA that predicts whether a patient will develop sepsis is 50% for adults vs 80% for children. Similarly, 35% TBSA predicts pneumonia in adults, compared with 65% in children.⁵¹ While each percent of TBSA confers 1 day in the ICU for adults, children tend to require only 0.5 days for every percent TBSA.⁵¹

The physiologic response to burns evolves with age, placing infants and older adults at the greatest risk for poor outcomes.³^,⁵ Although infants’ naive immune systems place them at greater risk for mortality,³ children experience better outcomes and are less likely to develop severe infection than adults despite being more prone to burn wound contamination and infection.⁵¹ The elderly tend to have worse outcomes due to their more exaggerated immune response to burn injury, with a mortality rate over seven times higher than younger patients.⁵⁴ Baseline comorbidities such as congestive heart failure, prior myocardial infarction, peripheral vascular disease, and renal disease,⁵³ as well as underlying hyperglycemia and insulin resistance,¹⁴^,⁵³ also increase the risk for hospital-acquired infection.

Infection is both a cause and effect of increased hospital length of stay (LOS).²² The average LOS for patients who develop infections is two to three times longer than those who do not.⁹^,¹⁰ The timeline of infectious etiology resembles that of colonization, with Gram-positive bacteria causing most infections within the first few weeks of hospitalization and Gram-negative bacteria and fungi increasingly identified thereafter.³⁸ P aeruginosa, Acinetobacter spp., and S aureus are the most common causes of infection worldwide.²³ While S aureus remains a leading cause of burn wound infection, Gram-negative organisms have become the most common cause of infection overall⁹^,²²^,²³^,⁴⁰^,⁵⁵^,⁵⁶ and represent the greatest contributors to mortality.⁵⁷ Although fungi may be cultured in 6.3% of patients⁵⁸ and cause burn wound infection in only about 2%,⁵⁹ infection is highly associated with increased mortality.⁵⁸^,⁵⁹

Hospital LOS is also associated with resistant infections as patients are colonized with more virulent pathogens
from the hospital environment.²²^,³⁸^,³⁹^,⁵⁵^,⁶⁰ One study found colonization with resistant organisms jumped from 6% within 1 week of hospitalization to 44% after 28 days.³⁹ LOS >7 days may increase methicillin-resistant Staphylococcus aureus (MRSA) risk 12-fold compared with shorter hospitalizations.⁶¹ MDR organisms, particularly P aeruginosa and A baumannii complex,⁹^,²²^,⁴⁰^,⁵⁶ have also become increasingly common and are characterized by acquired nonsusceptibility to one or more antimicrobials in at least three categories.⁶² While the average time to an initial non-MDR isolate is around 11 days, a much longer LOS of almost 40 days may be required for MDR organism colonization or infection.³⁸ In addition to hospital LOS,⁶³ many other risk factors for infection also apply to MDR organisms specifically. Patients with inhalation injury⁶¹ and longer duration of mechanical ventilation⁴³ are at increased risk, as are those requiring Foley catheters⁶⁴ or blood products.⁴³ Prior hospitalizations,⁶⁵ previous antibiotic exposure,⁶³^,⁶⁴ duration of antimicrobial treatment,⁶⁵ and number of antimicrobials received⁶⁶ also increase the risk for infection with MDR organisms. Evidence is conflicting regarding the effect of age⁶¹^,⁶⁴ and TBSA²²^,⁵⁵^,⁶¹ on the development of MDR pathogen infection.

Infection with MDR bacteria is becoming more common and may constitute over 65% of positive cultures in burn patients.²² Extensively drug resistance (XDR), defined as nonsusceptibility to one or more antimicrobials in all but one to two categories,⁶² is also a growing concern. In one large tertiary burn center, 44% of bacterial isolates met criteria for MDR and 23% exhibited XDR.³⁸ The magnitude of this problem is exemplified by the resistance profile of one US BICU, in which MRSA and vancomycin-resistant Enterococcus spp. (VRE) constituted 59.5% and 13.0% of S aureus and Enterococcus spp., respectively, and criteria for MDR were met by 90.8% of Acinetobacter spp., 33.8% of P aeruginosa, 21.1% of Stenotrophomonas maltophilia, 18.8% of Serratia marcescens, and 7.7% of Escherichia coli isolates.²³ Similarly high rates of resistance have been reported elsewhere.⁹^,²²^,⁵⁶ Finally, burn units are a common location for outbreaks of MDR bacteria, most frequently with S aureus and A baumannii complex.⁴²

DEFINITIONS AND DIAGNOSIS OF BURN WOUND INFECTION

Surgical excision of the eschar significantly reduces the risk of burn wound infection,²⁵^,²⁶^,²⁷ yet timely excision is not always undertaken, as in resource-limited settings, when patients present late to care or when complications or comorbidities make patients unable to safely undergo debridement. Clinical signs of burn wound infection include early eschar separation, brown or black discoloration, purulence beneath the eschar, and hemorrhagic pustules or necrotic ulcers characteristic of ecthyma gangrenosum.¹¹ Local infection is characterized by proliferation of microorganisms confined within burned tissue to a concentration >10⁵ colony-forming units (CFU)/g with a local immune response.¹¹ This cutoff was chosen because higher concentrations are associated with graft or wound closure failure and sepsis.⁶⁷^,⁶⁸^,⁶⁹^,⁷⁰^,⁷¹ When microorganisms invade surrounding unburned tissues at concentrations >10⁵ CFU/g and cause local and systemic signs of infection, the diagnosis of invasive infection can be made. Cellulitis can also occur, causing progressive erythema, induration, swelling, warmth, and tenderness around the burn that may be associated with systemic symptoms or sepsis.¹¹ In contrast, burn wound impetigo, characterized by loss of epithelium from a reepithelialized area, generally represents colonization rather than infection, with microbial concentrations lower than 10⁵ CFU/g and no changes in surrounding unburned tissue.⁷

Given that the physiologic burn response can closely mimic systemic infection,¹⁷^,¹⁸^,¹⁹ clinical signs alone cannot reliably distinguish between local and invasive or systemic infection. While the Centers for Disease Control and Prevention (CDC) National Healthcare Safety Network (NHSN) criteria for burn wound infection require both a change in wound appearance and identification of microorganisms from diagnostic, rather than surveillance, blood samples,⁷² biopsies have proven more sensitive than swab or blood cultures for diagnosing infection.⁶⁹^,⁷¹^,⁷³ The gold standard for burn wound infection diagnosis involves both quantitative tissue culture and histologic examination of a biopsy specimen that includes both burned as well as deep, unburned tissue,¹¹ although there is no gold standard method for biopsy collection and processing.⁶⁹ Both culture and histology are needed to definitively diagnose and characterize bacterial burn wound infections, and dedicated fungal cultures and stains should be added if fungal infection is suspected.

Gold standard diagnostic methods are not routinely followed. Quantitative culture can take 1-3 days to result, preventing early diagnosis of sepsis from invasive burn infection. Additionally, because early wound debridement has markedly reduced the risk of burn wound infection²⁵^,²⁶^,²⁷ and quantitative culture is relatively laborious and expensive, it is not routinely performed.¹¹ Many institutions utilize semiquantitative or qualitative cultures for surveillance and diagnosis, while a clinical diagnosis may be all that is possible in resource-limited settings. Qualitative cultures identify potential pathogens and antimicrobial susceptibilities, while semiquantitative cultures also provide information regarding pathogens’ predominance on specific growth mediums.¹¹

Burn wound colonization is a risk factor for infection, and identification of organisms colonizing burn wounds can help guide empiric antimicrobial therapy if infection develops. Burn wound infection surveillance is performed using surface swabs after dressings are removed and the wound is cleaned with an alcohol-based solution.¹¹ Guidelines recommend that surveillance swabs be obtained at regular intervals or when infection is suspected clinically. Although routine surveillance cultures have been used to more accurately predict MDR infection and select appropriate empiric antibiotics,⁵⁰^,⁷⁴^,⁷⁵ the absence of definitive evidence that routine surveillance cultures improve patient outcomes has led some hospital systems to resist this practice. Moreover, positive swab cultures along do not indicate infection or warrant systemic antimicrobials, as quantitative culture counts between swabs and tissue biopsies can be discordant due to bacterial load variation within the burn wound.⁷⁰^,⁷¹^,⁷⁶^,⁷⁷^,⁷⁸ More information on infection surveillance is provided in Chapter 2.

OTHER BURN-ASSOCIATED INFECTIONS

As stated previously, severely burned patients are at increased risk for a number of associated infections due to immune compromise and lengthy hospital stays with frequent dressing changes, IV and Foley catheters, and mechanical ventilation.⁶^,¹² Information on sepsis and ventilator-associated pneumonia (VAP) in burn patients is presented below, while infections related to IV devices, bloodstream infections, and UTIs are covered in Chapters 12, 13, and 14, respectively.

Sepsis

The leading cause of burn death in burn patients is multiorgan failure, which is most often triggered by sepsis⁷⁹^,⁸⁰^,⁸¹ and may be responsible for 65%-70% of deaths.⁷⁹^,⁸⁰ Sepsis may occur in 25%-35% with severe burns,⁸²^,⁸³ progressing to multiorgan failure in 65%⁸³ and death in 35%⁸² of septic patients. Sepsis and multiorgan failure are more likely to occur in patients with greater TBSA and inhalation injury,⁸⁰^,⁸³^,⁸⁴ with sepsis rates as high as 81% among patients with inhalation injury.⁸⁴ Risk persists until the skin barrier is restored, with sepsis often developing weeks or even months postburn.¹⁸ The pathogens most often identified are P aeruginosa, A baumannii, S aureus, and Candida spp.,⁵⁶^,⁷⁹^,⁸⁰^,⁸⁴ and antimicrobial resistance is more prevalent in BICU infections than infections on the common burn ward.⁵⁶ Although many studies describing infections in burn patients do not describe sepsis specifically or distinguish between infection types, MDR Gramnegative bacteria have been identified as the leading cause of death from infection overall.⁵⁷

Many criteria used to define sepsis in other settings do not apply to burn patients, who often develop tachycardia, tachypnea, leukocytosis, and low-grade fever reflecting their physiologic response to burn.¹⁹^,⁸⁵ Burns >15%-20% TBSA are associated with the systemic inflammatory response syndrome (SIRS), which can persist for months after burns have healed.⁸⁵ To address this, the American Burn Association (ABA) developed diagnostic criteria for sepsis in burn patients,¹⁹ which require documented infection identified via positive cultures, pathologic tissue source, or clinical response to antimicrobials, as well as three or more of the following:

Fever >39°C or hypothermia <36.5°C
Progressive tachycardia >110 beats per minute
Progressive tachypnea >25 breaths per minute or minute ventilation >12 L/min
Thrombocytopenia <100 000/µL (does not apply until postburn day 3)
Hyperglycemia without preexisting diabetes, defined as untreated plasma glucose >200 mg/dL, IV insulin requirement >7 units/h, or >25% increase in insulin requirements over 24 hours
Feeding intolerance >24 hours, characterized by abdominal distension, enteral feeding residuals twice the feeding rate, or diarrhea >2500 mL/d

The ABA adjusted these parameters for pediatric patients to account for normal child physiology, with heart rate, respiratory rate, and thrombocytopenia cutoffs set two standard deviations from age-specific norms. Feeding intolerance was set at >150 mL/h and diarrhea at >400 mL/d. Additional signs of sepsis in burned children may include those of decreased perfusion, urine output <1 mL/kg, decreased peripheral pulses, altered mental status, capillary refill >2 seconds, and cool or mottled extremities.¹⁹

Sepsis and multiorgan failure are not always distinguished in the literature, with one listed as the cause of death while the other may have also been present. One retrospective study of 92 mortalities in a UK BICU over 7 years did identify both multiorgan failure and sepsis, attributing 70% of burn deaths to multiorgan failure and naming sepsis as the primary cause of multiorgan failure in about 54% of cases.⁸⁰ The study authors also noted that only 26.5% of septic patients had positive blood cultures, with most pathogens identified by wound culture. More than half of patients with sepsis may have negative blood cultures⁸⁰^,⁸⁴^,⁸⁶ including 36%-70% of patients who die from sepsis.⁷³^,⁸⁷ Studies are conflicting regarding the correlation between blood cultures and sepsis or mortality.⁶⁹^,⁷⁸^,⁸⁸^,⁸⁹

Biopsies are more sensitive than swab or blood cultures for diagnosing sepsis,⁶⁹^,⁷¹^,⁷³^,⁷⁸^,⁸⁷ and a microbial concentration of 10⁸ CFU/g or greater is associated with sepsis and increased mortality.⁷¹^,⁸⁷ There is also limited evidence suggesting that swabs can be used to accurately identify the causative organism, with culture counts >10⁶ CFU or visualization of bacteria on surface swab gram stain correlating with biopsy cultures of 10⁵ CFU/g.⁹⁰

The need for more sensitive and specific methods to identify sepsis in burn patients has spurred a search for diagnostic biomarkers.¹⁵ Procalcitonin (PCT) has emerged as the most promising.⁹¹ PCT is a precursor of calcitonin released by the thalamus to regulate calcium metabolism. In the setting of systemic bacterial, and to a lesser extent fungal, infection, inflammatory mediators trigger the liver, lungs, kidneys, and adipose tissue to release PCT. While barely detectable in uninfected patients, levels rise significantly with infection and quickly fall when it is adequately controlled or treated.⁹²^,⁹³ Patients with SIRS can display many of the ABA sepsis criteria regardless of infection status, but septic patients have significantly higher PCT levels than noninfected patients.¹⁷ Preliminary evidence suggests using PCT to guide antibiotic therapy in the BICU may reduce antibiotic exposure without increasing mortality or other negative outcomes.⁹⁴ However, as with individual ABA criteria, PCT should not be used alone to diagnose sepsis; the optimum cutoff of 1.0 ng/mL has only 52% sensitivity and 77% specificity with a 38% positive predictive value and 86% negative predictive value.⁹⁵

Pneumonia

Burns place patients at increased risk for pneumonia, the most common complication associated with burn wounds.³ Smoke inhalation often plays a significant role, causing thermal and chemical epithelial injury that leads to sloughing of respiratory mucosa, thereby hampering mucociliary clearance and obstructing moderate diameter airways. Microvascular injury and the release of oxygen free radicals and inflammatory mediators result in pulmonary edema, with exudate filling small airways via damaged endo- and epithelium. Patients with truncal burns may exhibit shallow breathing due to pain or restrictive
eschars, further promoting atelectasis, microbial growth, and resulting pneumonia.²⁴ Additionally, inhalation injury appears to induce immune hyporesponsiveness, with greater reductions in white blood cells and pulmonary immunomodulators associated with mortality.⁹⁶ One mechanism may involve higher local levels of anti-inflammatory IL-10, which suppresses the pulmonary immune response and increases risk for respiratory infection with Gram-negative organisms.⁹⁷

Even in the absence of inhalation injury, immune dysregulation places burn patients at increased risk for respiratory infections. This risk persists long after discharge, as demonstrated by a retrospective study of 14 893 hospitalized burn patients without inhalation injury or need for mechanical ventilation. Admissions for respiratory tract infection remained elevated at least 15 years postburn, with admission rates for influenza and viral pneumonia increased by 73% and bacterial pneumonia and other respiratory infections more than double rates for matched unburned patients.⁹⁸

Pneumonia occurs in 4.1% of all flame injuries requiring hospital admission and more than one-third to onehalf of mechanically ventilated patients.⁹⁹ Inhalation injury and pneumonia alone increase mortality by 20% and 40%, respectively, while the coexistence of both may increase mortality by 60%.¹⁰⁰ The risk for pneumonia is significantly higher with severe burns than in other critically ill patients, with BICUs accruing the most VAP days per ventilator days compared with other critical care units.¹⁰¹ For example, one retrospective analysis of 1772 admissions to a US BICU reported a VAP rate of 4.22/1000 ventilator days compared with 2.60/1000 ventilator days in other ICUs within the same hospital.¹⁰² Pneumonia risk is highest in those requiring more than 4 days of mechanical ventilation³ and greater TBSA.¹⁰³ Preexisting conditions including diabetes, cerebrovascular disease, ischemic heart disease, and liver disease individually increase risk more than 2.5-fold.¹⁰⁴ Age is also a significant risk factor, with burn patients 65 years of age or older at almost 10-fold risk for pneumonia compared with those 18-40 years old.¹⁰⁴

A clinical diagnosis of pneumonia in burn patients may be made if two of the following conditions are met¹¹^,¹⁹^,¹⁰⁵:

Chest x-ray shows new, persistent infiltrate, consolidation, or cavitation.
The patient meets criteria for sepsis (as defined in the section on Sepsis).
Sputum is purulent or has recently changed.

If patients meet clinical criteria and a pathogen is isolated, the diagnosis of pneumonia is confirmed. Pneumonia is probable if clinical criteria are met without microbial confirmation and is possible in patients with an abnormal chest x-ray with low to moderate clinical suspicion in the setting of an identified pathogen or positive microbiology. Microbiology is considered positive when tracheal aspirate contains at least 10⁵ organisms, bronchoalveolar lavage contains at least 10⁴ organisms, and protected bronchial brush contains at least 10³ organisms.¹¹

In 2013, the National Health and Safety Network (NHSN) changed the adult surveillance VAP algorithm to capture a variety of ventilator-associated events (VAE).¹⁰⁶ According to this algorithm, prior to being defined as a possible VAP, a case must meet criteria for VAE, ventilator-associated condition (VAC), and infection-related ventilator (IVAC) complication, including worsening oxygenation and need for new antibiotic therapy. Radiographic evidence of pneumonia is no longer a criterion for possible VAP cases in the VAE algorithm. One study of a single burn ICU compared the incidence of VAE-possible VAP to incidence of VAP using the pre-2013 VAP definition over 1.5 years and identified considerably fewer VAP cases using the pre-2013 definition (0.55 vs 4.96 events/1000 ventilator days).¹⁰⁷

In contrast to the clinical VAP definition used by IDSA/ATS guidelines or the surveillance definitions for health-care-associated infections (HAIs) outlined by NHSN, the ABA National Burn Repository has no strict criteria to define pneumonia, depending instead on provider assessment for prevalence estimates. As the inclusion criteria for respiratory infection may vary greatly from study to study, the definition of pneumonia and tracheobronchitis should be taken into consideration as one reads the literature on pneumonia and burn.

The most common causes of respiratory infection in BICUs are P aeruginosa, enteric Gram-negative bacilli, S aureus, Acinetobacter spp., and S maltophilia.¹⁰² The proportion of MDR pathogens in BICUs is much higher than in other ICUs, with rates of 41% and 14%, respectively, reported in one study.¹⁰² The same study reported high rates of multidrug resistance among pathogens causing pneumonia, including 36% of P aeruginosa, 17% of enteric Gram-negative bacilli, 57% of S aureus, 100% of Acinetobacter spp., and 18% of S maltophilia. Similarly, another study, in which 86.8% of burn patients with inhalation injury developed VAP, identified MDR pathogens in 32.9% of infections, most commonly P aeruginosa and Enterobacter spp.⁷⁴ In this study, surveillance cultures diagnosed MDR pathogen infection with 83.0% sensitivity and 96.2% specificity, with a positive and negative predictive value of 87.0% and 95.0%, respectively.⁷⁴ Additional studies of ventilated patients report that weekly or twice-weekly surveillance with BAL or tracheal aspirate may be used to predict the presence or absence of resistant bacteria,⁷⁵^,¹⁰⁸ thereby offering a means to improve the accuracy of antibiotic selection when pneumonia develops.⁷⁴ However, not all studies have found surveillance cultures to be helpful,¹⁰⁹ and providers may be hesitant to let them guide antibiotic selection.¹¹⁰ More information on the diagnosis and treatment of healthcare-associated pneumonia and VAE may be found in Chapter 16.

CAUSATIVE PATHOGENS IN BURN-ASSOCIATED INFECTION

As described in preceding sections, infections in burn patients generally progress along a timeline beginning with Gram-positive bacteria, followed by Gram-negative bacteria and fungi.³⁴^,³⁵^,³⁶^,³⁸ Drug resistant infections are also more likely to emerge with longer hospital stays²²^,³⁸^,³⁹^,⁵⁵^,⁶⁰ and are becoming more common among the leading causes of infection, including P aeruginosa, A baumannii complex,
and S aureus.⁹^,²²^,²³^,⁴⁰^,⁵⁵^,⁵⁶^,¹⁰²^,¹¹¹^,¹¹²^,¹¹³ Anaerobic bacteria rarely cause invasive burn infections, but infection with Bacteroides or Fusobacterium spp. can occur in compartmental myonecrosis caused by electrical burns.⁷ Unsurprisingly, Clostridioides difficile infection is also common in burn centers given the widespread use of antibiotics; however, this pathogen is covered in detail in Chapter 17. Viral and fungal infections are less common in burn patients but should be considered in impaired wound healing that fails to respond to antibiotics as expected.¹¹⁴ The most common pathogens infecting burn patients are described briefly below and in Table 27-1.

Common Gram-Positive Pathogens

S aureus, a Gram-positive coccus, is both a contributor to normal skin flora and a potential pathogen in the setting of immunocompromise and skin disruption in burn patients. This bacterium is easily transmitted in the hospital setting, the leading cause of burn wound infection, and a major contributor to skin graft loss, pneumonia, and sepsis in burn patients.⁷ In contrast Enterococcus spp. are indolent gut flora capable of gut translocation in the setting of immunocompromise and microbiome disruption associated with burn wounds. Enterococcus spp. are typically not considered pathogenic in the lungs and rarely
cause burn wound infection. However, some species such as Enterococcus faecium can be extremely difficult to treat due to antimicrobial resistance.⁶³^,¹²⁷

TABLE 27-1 Clinically Important Bacterial Pathogens in Hospitalized Burn Patients

	*Prevalence*	*Reservoirs and transmission*	*Clinical significance*
Staphylococcus aureus	MRSA colonization rates are generally higher in developing countries, 4%-8.3% upon admission to U.S. burn centers^61,115 vs 50% in Ghana¹¹⁶ Responsible for 10%-40% of infections^a,^{5,23,38,84,117,118,119,120,121,122} Cause infection in 10%-13%^84,116	Ubiquitous in the community and hospital environment with inducible genes allowing for both rapid¹²³ and gradual¹²⁴ development of resistance following antibiotic exposure	Common cause of skin graft loss, pneumonia, and sepsis in burn patients⁷ Fatal in 17%-35% of cases^119,125 Over half of patients with nasal MRSA colonization at admission develop MRSA wound colonization or infection¹²⁶
Enterococcus spp.	Colonize GI tract of 16% upon admission⁶³ Responsible for 6%-15% of infections^23,118,119	Generally indolent gut flora that can become pathogenic with increased intestinal permeability, disruption of the microbiome, and immunocompromise seen in burn patients	A less common cause of bacteremia and sepsis that rarely causes burn wound infection Responsible for 25% of sepsis deaths in the 1990s vs 2% in the 2000s, likely due to decreased incidence with routine vancomycin use⁷ Mortality as high as 58%¹²⁵ Intrinsic resistance to many antibiotics, including beta-lactams and aminoglycosides⁶³ VRE may colonize 7% after 28 d of hospitalization⁶³ and VRE rates may reach as high as 13%-28%^23,127
Pseudomonas spp.	Colonize 25%-43%^{49,128,129,130} Cause 13%-43% of infections^{5,23,38,47,84,118,119,120,121,122,131}	Thrive in moist environments including sink taps, aerators, and respiratory and hydrotherapy equipment and can be transmitted under fingernails of healthcare workers and on raw fruits and vegetables.¹³² Can persist weeks to months following colonization or infection despite antibiotic treatment, causing recurrent infections¹³³	Leading cause of pneumonia, sepsis, and burn-related mortality in many burn centers,^7,79,80 with mortality rates of 25%-41%^119,125 Colonization may increase infection risk to 39% vs 3.4% in uncolonized patients⁴⁹ 75%-89% of isolates may be carbapenem-resistant^111,127 MDR organism rates range from 15% to 36% in the United States to ˜70% in Iran and Brazil^134,135,136
Acinetobacter spp.	Colonize 8.5%-30%^{43,50,128,137,138} Cause 3%-65% of infections^{5,23,84,118,119,120,121,122,131}	Ability to survive on wet and dry surfaces¹³⁹ contributes to rapid dissemination through burn units via fomites^43,128	A leading cause of pneumonia, bacteremia, and burn wound infection,^{7,50,56,84,140} with mortality rates of 14%-56%^43,119,125 Infection may occur in 56.7% of colonized burn patients vs 4.9% of uncolonized patients⁵⁰ Carbapenem resistance reaches 80%-100% in the United States,¹⁰² Iran,^{112,138,141,142} Bulgaria,¹²⁰ and Turkey⁹ MDR in 61% of patients with >30% TBSA and in 75% with LOS 15-30 d.⁵⁵
Klebsiella spp.	Among community-dwelling adults, GI colonization increases from 6% to 19% with recent healthcare contact⁴⁸ Responsible for 3%-22% of infections^{23,118,119,120,121,122}	Common in soil, water, and human intestine.¹⁴³ Hospital reservoirs include sinks, showers, and wastewater systems¹⁴⁴	Most commonly causes pneumonia, followed by wound infection, UTI, and bacteremia^7,48 Mortality varies widely from 8% in Austria¹¹⁹ to 50% in Brazil.¹²⁵ Gut colonization may increase infection risk to 16% vs 3% in uncolonized patients⁴⁸ MDR organism rates vary widely worldwide, reaching as high as 54% in one Indian burn unit¹²¹
^aAll statistics refer to hospitalized burn patients unless stated otherwise.
MRSA, methicillin-resistant Staphylococcus aureus; GI, gastrointestinal; VRE, vancomycin-resistant enterococcus; MDR, multidrug resistance; U.S., United States; TBSA, total body surface area; LOS, length of stay.