Abstract
Background:
Viral infections after burns are less common than bacterial infections but usually occur in the more severely burned patients and have been associated with poor outcomes.
Methods:
Retrospective reviews and case series were examined to provide an overview of the management of viral infections in the burn patient.
Results:
The most common viral pathogens in these patients are the herpesviruses, which include herpes simplex, varicella zoster, cytomegalovirus, and human herpesvirus 6. Established viral infections that may complicate patient management include human immunodeficiency virus, hepatitis B and C, and, more recently, the novel coronavirus SARS-CoV-2. Herpesvirus infections can occur as primary or nosocomial pathogens but clinical manifestations most commonly are re-activation of latent viral infection. Because of the paucity of data in the burn population, much of the evidence for specific treatments is extrapolated from patients with severe immunosuppression or critical illness. Antiviral therapy is employed for the burn patient with herpesvirus infections. This is an area of active study, and further research is needed to better understand the risks, clinical manifestations, and attributable morbidity and mortality of viral infections.
Conclusions:
Major burn injury results in immunosuppression and viral infection in a small number of patients. Recognition and antiviral therapy are employed, but additional studies are necessary to improve outcomes in these patients.
Viral infections in burn patients range from asymptomatic shedding (of unclear clinical significance) to wound and graft involvement or lower respiratory tract infection (LRTI) associated with a high mortality rate [1-4]. In addition to the physiologic disturbances that accompany severe burns, significant immunologic dysfunction puts these patients at risk for reactivation of latent viruses [5-9]. The critical care literature has provided a better understanding of risk factors and outcomes associated with asymptomatic viral shedding, and much is extrapolated from these data [3,4,10].
Incidence and morbidity attributable to viral infections in burn patients are difficult to tease out, not only because of the nature of retrospective data, but perhaps more importantly because of the difficulty in diagnosis [6]. Some studies from the 1980s suggest that serologic evidence of herpes simplex virus (HSV) infection in burns may be as high as 25%. However, few of these studies used viral isolation, and when they did, rates of HSV were closer to 10% [6]. In a 12-year autopsy review of 97 patients from a single burn center, only 5.2% died from viral infections, all of which were LRTIs. In this study, four of the five patients had HSV LRTIs, and one patient had cytomegalovirus (CMV) [3]. In contrast, studies evaluating viral wound infections have shown an association with delayed healing but have yet to show an impact on the mortality rate [11]. This illustrates a crucial distinction: Anatomic sites of viral infection are an important marker of prognosis.
The next section of this review briefly outlines broad concepts important to the complex interplay between immunologic dysfunction after burns and control of viruses. These changes are key to understanding how viral infections can cause disease in these patients. We start by reviewing the immunologic effects of burns, then examine the mechanisms of viral infections in burn patients and review the most common viruses and their management, including the role asymptomatic shedding may have in burn patients based on available data from the critical care literature. Finally, we briefly review the literature on human immunodeficiency virus (HIV) and hepatitis viruses in burn patients and the recently described severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).
Burn-Related Immunosuppression
Burn injuries result in immunosuppression as well as a dysregulated inflammatory response [12]. Hypotheses about the mechanism of this phenomenon have differed over the years [7,13-17]. Early experiments argued that this abnormal response was related to formation of inhibitory cytokines from the burn wound itself [6]. Animal models have provided evidence for a variety of post-burn immunosuppressive effects, including murine models demonstrating populations of cells produced in burn wounds capable of causing immunosuppression when transferred into healthy mice. There was dysregulated opsonization, which correlates with the described predominance of bacterial over viral infections [5]. There also was a prolonged increase in suppressor/cytotoxic T-cells and a decrease in helper/inducer T-cells even in mice with total body surface area (TBSA) involvement as low as 15%–20% [14].
Human studies show similar findings. In patients with more than 20% TBSA burns, the “genomic storm” affects both the innate and the adaptive immune response—with the former undergoing up-regulation and the latter down-regulation [18]. Researchers demonstrated that several genes responsible for regulation of Toll-like receptors important in inflammatory signaling were down-regulated in patients with >30% TBSA burns, with downstream effects on cytokine production, particularly of interleukin (IL)-6 and IL-12 [17]. Changes in these cytokines, and others, result in decreased antigen presentation and T-cell proliferation [12]. Ultimately, this effectively downregulates the body's ability to respond to both new and latent viruses. Immunologic studies in burn patients also have demonstrated wide-ranging down-regulation of the immune system to include decreased T-cell counts, prolonged decrease in natural killer (NK) cell cytotoxicity (broadly important in the response to viral infection), and decreased antibody-dependent cellular cytotoxicity [6,15].
Typically, viral infections are delayed, presenting after the first week of injury [6]. Burn injuries result first in a pro-inflammatory state, which is followed by an anti-inflammatory response. This latter response is thought to result in greater susceptibility to both bacterial and viral infections. The so-called “compensatory anti-inflammatory response” in critically ill patients, which has been described elsewhere, is posited to result in reduction in lymphocytes, cytokine response, and antigen presentation by T cells, potentially explaining the kinetics of these infections [13]. This evidence highlights several important points: Burn injuries result in a patient who has innate and adaptive (including humoral and cellular) immune dysfunction [7,9,11,13,14,17,18]. This dysfunction places them at risk for a variety of infections during a particularly high-risk period of their recovery. In the case of viruses, with their potential for latency and reactivation, this immune dysregulation (especially that of cellular immunity) increases the risk of clinical disease [1,2,6,11,13,19,20].
Mechanism of Infection
There are three main mechanisms of infection that have been postulated to play a role in viral infections in burns: Re-activation of latent virus, primary infection, and nosocomial re-infection [13]. Reactivation of latent virus can range from asymptomatic viral shedding to severe clinical disease [1,11,13,21]. Primary viral infections follow initial exposure when there is no immunity to these viruses [11,22]. An important example of this in burn patients is varicella zoster virus (VZV), as primary infection (chickenpox) in the post-burn period has been associated with increased morbidity and mortality rates [21]. Other examples of primary infection occur when CMV-seronegative adults receive CMV-positive blood products or CMV donor-positive allografts, potentially resulting in clinical disease such as hepatitis or pneumonitis [22]. Lastly, nosocomial reinfection (often referred to elsewhere as “exogenous reinfection”) refers to reinfection in a host with established immunity or disease [13,23]. This is particularly challenging to study, as it requires sophisticated molecular or serologic testing to differentiate between re-infection and re-activation of latent virus [23,24].
Herpesviruses in Burn Patients
Herpesviruses are ubiquitous double-stranded DNA viruses that have the ability to establish latent infection in human hosts [25]. Of the viruses that complicate burn patients' recovery, the Herpesviruses family are the most common [3,6,9,11]. And while this family of viruses includes HSV-1, HSV-2, VZV, Epstein-Barr virus (EBV), CMV, human herpesvirus (HHV)-6, HHV-7, and HHV-8, the burn literature to date typically focuses on HSV-1, HSV-2, VZV, and CMV (3,6,11,13). These viruses can have a variety of clinical presentations, from asymptomatic shedding to disseminated disease. Differentiating between them can be challenging but can be important for treatment and prognostication. Although outbreaks of less common viruses have been documented in burn patients (for example, the poxvirus Orf), most of the literature and clinical experience is dominated by the Herpesvirus family [3,6,13,26]. The following summarizes the key features of the main Herpesviruses affecting burn patients.
Herpes simplex virus
Herpes simplex infections in burns range from asymptomatic viral shedding to fulminant end-organ disease, which usually is discovered at autopsy [3]. The development of erosions or vesicles, particularly near sites of regeneration on the face, neck, or nasolabial folds, should clue clinicians into considering HSV [1,2,6,19]. Although historically, HSV-1 has been associated with orolabial disease and HSV-2 with genital disease, recent serologic surveys have argued that this distinction is less valid [27,28]. Other manifestations of HSV disease can include keratoconjunctivitis, pneumonitis, and hepatitis [6,11,28]..
Epidemiologic studies of burn patients with cutaneous HSV infections are limited by overall numbers of patients; however, a few common themes emerge. These patients are more likely to be younger, male, have >20% TBSA burns, and have face and neck burns [6]. Viral typing argued that most of these are HSV-1, and early papers hypothesized this was attributable to oropharyngeal re-activation [6]. Classically, infection tends to occur 1–3 weeks after the burn injury and involve healing partial-thickness burns with areas of regeneration rather than full-thickness burns [1,6,11,19].
Many research groups have attempted to show a relation between HSV re-activation after burns and various markers of morbidity, including prolonged hospital stay and death. In 2012, a retrospective review of 71 patients with at least 30% TBSA burns revealed an association between HSV infection (defined as documented lesions with culture of HSV) and increased length of stay and ventilator days and isolation of Acinetobacter baumannii from the respiratory tract, but not a higher mortality rate [9]. Other groups have shown an increased length of stay and duration of mechanical ventilation. However, only D'Avignon et al. found an association between HSV infection and death, which was driven by HSV LRTI [3,9,29].
Asymptomatic HSV shedding (e.g., recovery of HSV via polymerase chain reaction [PCR] or culture) is common in critically ill patients, including burn patients [30-34]. This makes interpretation of testing in the appropriate clinical context critical, as well as challenging. The Tzanck smear is a rapid test that can be used to look for multi-nucleated giant cells of suspicious cutaneous lesions; however, it cannot discriminate between viruses [19]. Viral culture or PCR is needed to diagnose HSV definitively. In practice, the viral culture yield is much lower than PCR. Demonstration of HSV DNA in lesion scrapings by PCR is the gold standard for diagnosis of cutaneous disease, and most laboratories have turned to this modality [35].
Questions regarding the diagnosis of HSV (along with CMV and VZV) disease in end organs can be challenging to resolve. Herpes pneumonitis is a good example of this problem. Recovery of HSV from bronchoalveolar lavage (BAL) viral culture alone is not sufficient to make the diagnosis. Unfortunately, histopathologic evidence can be difficult both to obtain and to interpret. Evaluation of lung tissue biopsies from burn patients is helpful in addressing this question. One group studied 54 patients who had lung tissue evaluated at autopsy. The mean survival rate in both HSV-positive and -negative cases was similar (35.4% and 39.2%). None of these patients had viral inclusion bodies on hematoxylin and eosin stain, but 50% of them were positive by immunohistochemical staining. Interestingly, 81% of the patients with acute respiratory distress syndrome (ARDS) were HSV positive compared with 37% of those with pneumonia [36]. Herpes simplex virus reactivation in ARDS patients also has been demonstrated in other critically ill populations, for example, those requiring extra-corporeal membrane oxygenation (ECMO) [37]. It should be noted that the evidence thus far supports correlation, not causation, between HSV and ARDS and cannot as yet inform management decisions; however, it does suggest questions for further research regarding what pathophysiologic role HSV is playing in this setting. When evaluating for herpes pneumonitis, bronchoscopy with tissue biopsy for immunohistochemical staining, isolation of HSV (via either PCR or viral culture from BAL fluid), with a consistent clinical and radiographic picture, are needed [28].
No clinical trials exist in the burn patient population to evaluate the best route and formulation of antiviral management for HSV infections. Whereas the initial treatment of choice is acyclovir, the route of administration (intravenous [IV] versus oral) depends on the site of clinical infection [38]. For clinically severe HSV disease (e.g., HSV encephalitis), the treatment of choice is IV acyclovir, while topical acyclovir has been used with modest success for cutaneous disease in burn patients [19,39,40]. All HSV-active antiviral drugs have associated toxicities, and the risks and benefits of each need to be weighed (Table 1).
Common Antivirals, Treatment Indications, Evidence for Prophylaxis, and Toxicities [39, 65]
CMV = cytomegalovirus; CNS = central nervous system; HSV = herpes simplex virus; IV = intravenous; PO = oral; VZV = varicella zoster virus.
Modified from: Luyt C-E, et al. Acyclovir for mechanically ventilated patients with herpes simplex virus oropharyngeal reactivation: A randomized clinical trial. JAMA Intern Med 2019;10:1001/jamainternmed.2019.5713; Frange P, Leruez-Ville M. Maribavir, brincidofovir and letermovir: Efficacy and safety of new antiviral drugs for treating cytomegalovirus infections. Med Malad Infect 2018;48495–48502; Tuxen DV, Wilson JW, Cade JF. Prevention of lower respiratory herpes simplex virus infection with acyclovir in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1987;136:402–405; Andres Bran JKO. In: Andrej Spec GE, Chrisler C, Davies B, editors. Comprehensive Review of Infectious Diseases. Philadelphia: Elsevier, 2020; Marty FM, et al. A randomized, double-blind, placebo-controlled phase 3 trial of oral brincidofovir for cytomegalovirus prophylaxis in allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant 2019;25:369–381; Tuxen DV, Wilson JW, Cade JF. Prevention of lower respiratory herpes simplex virus infection with acyclovir in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1987;136:402–405; Emmet JE, Blaser MJ, Editors. Mandell, Douglas, and Bennet's Principles and Practice of Infectious Diseases. 8th edition. Philadelphhia: WB Saunders, 2015.
Cytomegalovirus
Cytomegalovirus infection in burn patients also ranges from asymptomatic viral shedding, to a limited febrile illness, and end-organ disease including pneumonitis and hepatitis [42,43]. Unexplained fever and hepatitis in burn patients has been linked to primary CMV infection, with pediatric burn patients at particular risk [40–43]. A combination of PCR, viral culture, histopathology examination, and immunohistochemical staining of involved tissues (in the appropriate clinical context) are used for diagnosis [42]. As with HSV, diagnosing CMV end-organ involvement (i.e., pneumonia, hepatitis, etc.) can be challenging. Outside of the burn literature, studies evaluating for CMV pneumonia have shown that in many cases, serum CMV DNA concentrations do not correlate with BAL concentrations. For instance, in one study of lung transplant patients, only 63% of those with CMV pneumonia had positive serum CMV DNA tests [45]. This means that the diagnosis typically is categorized by either probable (positive CMV BAL, signs or symptoms, and radiographic findings consistent with the diagnosis) or proven CMV pneumonia (symptoms and transbronchial biopsy positive for CMV by immunohistochemical staining) [45].
Notably, interventions required in burn management (e.g., blood transfusions and allograft placement) place patients at risk for CMV infection [22,41]. Murine burn models have demonstrated that CMV transmission can occur via infected donor allografts and that CMV infection increases the risk of bacterial infections [41]. Lifetime CMV seroprevalence is approximately 50% [11]. Early studies were hampered by either unreliable diagnostics or imprecise definitions (e.g., four-fold rise in anti-CMV IgG titer), which resulted in difficulty differentiating between reactivation and passive antibody transfer (e.g., via blood products) [41].
Prospective studies with small numbers of enrolled burn patients demonstrated that although CMV viremia was associated with longer mechanical ventilation and ICU stays, there were no differences in the mortality rate [46]. Prospective observational work by Bordes et al. showed that a serum CMV concentration of >1,000 copies/mL was associated with more major infections and red blood cell transfusions and longer mechanical ventilation. Others have identified such CMV viremia risk factors as the amount of blood products transfused, use of skin allografts from CMV-seropositive donors, relative and active immunosuppression post-burn, size of burn, and degree of burn [10,22,41,42,46].
There are no randomized controlled trials evaluating the use of anti-CMV agents (e.g., valgancyclovir, ganciclovir, cidofovir, foscarnet) in burn patients. These drugs are associated with high rates of toxicities such as myelosuppression and renal injury [47,48]. Letermovir or other drugs that have been used to prevent CMV in transplant patients (e.g., brincidofovir or marabivir), with their better toxicity profile, might offer a new avenue for evaluating this research gap in the future (Table 1) [47]. Attempts to use immunoglobulin in burn patients to avoid antiviral-associated toxicities were either unsuccessful or underpowered [11,49].
Varicella zoster virus
Varicella zoster virus, manifested as chickenpox in primary infections (most commonly in children) and shingles when it reactivates (more common in adults), occurs less frequently than the other Herpesviruses post-burn. Many investigators hypothesize that this could be attributable to increasingly widespread immunity from childhood disease and vaccination programs, but there are no conclusive data to show this yet [11,21,50]. In the few reported cases of VZV re-activation in burn patients, it was more likely to be found in re-epithelializing skin and associated with a higher mortality rate [8,50].
The majority of case reports of primary VZV in burn patients come from children without existing exposure or immunity. The incidence reported in pediatric burn centers is low; however, the morbidity in disseminated forms of the disease can be devastating. One center had 12 pediatric cases over 15 years and one death from VZV pneumonitis [21]. Given the use of vaccination for varicella in the early 2000s, this is likely to continue to change the epidemiology. In the United States, all 50 states have school age requirements for varicella prevention (Immunize.org). Outside the United States, including Europe, debate continues about universal varicella vaccination [51]. These geographic considerations can play a role in adjusting pre-test probability for disease.
Other Viruses (Hepatitis, HIV, SARS-CoV-2)
The HIV epidemic changed the face of medicine when it began in the 1980s. With advances in anti-retroviral therapy (ART), the prognosis today for patients living with HIV with access to healthcare and ART is excellent. Human immunodeficiency virus may complicate burn care in a number of ways: Existing cell-mediated immunosuppression, pharmacokinetic and pharmacodynamic considerations of ART and other prophylactic medications, and widening differential diagnoses of infections in the post-burn period. Complications of HIV infection in burn patients have been studied only retrospectively, and interpretation is difficult given the wide variety of HIV severity and the point in the epidemic at which these studies were conducted [51-54]. A study of burn patients living with HIV in Malawi demonstrated a higher probability of death if sepsis developed [52]. In patients living with HIV in Iran, the virus was associated with longer hospital stays [53]. Further research is needed to characterize the role HIV plays in post-burn care and how to optimize management.
Hepatitis viruses, in particular hepatitis B (HBV) and hepatitis C, typically are screened for prior to cadaveric allograft collection [22]. As we have seen with other viruses, re-activation can occur in patients with severe burns and chronic hepatitis B. One study found that patients who were HBV surface antigen positive at burn admission with sepsis were more likely to develop hepatic damage and HBV viremia [55].
Lastly, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a novel coronavirus that emerged from Wuhan, China, in 2019, progressed rapidly to a pandemic [56]. This virus and the infectious syndrome it causes (COVID-19) threatens hospital-system capabilities worldwide. The American Burn Association has issued crisis standards of care for burn patients during this pandemic. This includes recommendations for meticulous attention to hand hygiene, social distancing, and assessing capability to flex critical care skills often used to manage burn patients to COVID-19 patients. Surgical plans will need to minimize aerosolizing procedures where possible. Other unique considerations include expanding the use of telemedicine, rapid transition to long-term dressings in order to minimize staff exposure, and preservation of the personal protective equipment supply [57]. As trials are enrolling patients to further guide diagnosis, prognostication, and management of COVID-19 (with no data yet in burn patients), further discussion is outside the scope of this review. The reader is directed to the Centers for Disease Control and Prevention and the World Health Organization websites for further information.
Related Areas of Research and Further Questions
The clinical significance of asymptomatic viremia in burn patients currently is unclear and, as noted earlier, extrapolated from work done in the ICU population [10,33,42]. Various scholars have hypothesized that this might be a marker of severity of illness (the “bystander phenomenon”), related to the relative immunosuppression of the host, rather than evidence of clinical disease [10]. Work that has been done on herpesvirus viremia in critically ill patients serves as a good starting point. In 2017, Ong et al. evaluated this effect in previously immunocompetent patients with septic shock [33]. Interestingly, nearly 70% developed herpes viremia while in the ICU, and around 30% developed viremia with more than one herpesvirus. Only viremia with both CMV and EBV concurrently was an independent risk factor for death. Data to clarify antiviral use (prophylactic, preemptive, or strictly for clinically significant disease) are likely to come from the critical care literature first.
Prospective trials studying the role and management of HSV have been focused on critically ill patients with asymptomatic viral shedding. Tuxen et al. showed in ARDS patients that prophylactic administration of acyclovir prevented HSV disease but did not have a significant impact on the mortality rate [58]. The role of preemptively treating oropharyngeal re-activation of HSV has been evaluated recently in non-burn patients. A double-blind, placebo-controlled randomized clinical trial enrolled critically ill, ventilated patients based on PCR detection of HSV from oropharyngeal swabs to either acyclovir or placebo. Ultimately, there was no difference between the groups in their primary outcome of ventilator-free days or secondary outcomes, including the 60-day mortality rate [32]. Studies such as these may indicate that asymptomatic shedding is a marker of severity of underlying illness rather than a predictor of HSV-related death. More work is needed, and in particular in burn patients to help tease out the role of re-activation of latent disease and its management.
One theme in managing burn patients is prominent: Clinical presentations coupled with diagnostics often can be confounding. This makes diagnosing viral infections difficult. There has been a concerted effort over the years to describe the various biomarkers in burn patients to try to alleviate these diagnostic challenges. These biomarkers have included various interleukins such as IL-8, procalcitonin, and pro-B-type natriuretic peptide. Interleukin-2, IL-8, and IL-10 have been studied outside the burn population and seem to correlate with viral infection [60]. So far, there have been mixed results in these studies, and no clear superior biomarker has emerged and been used widely in clinical practice [60]. This is an area where further research could prove vital.
Lastly, as new diagnostic techniques, particularly whole genome sequencing, become more available, we predict that more knowledge about the viral microbiome will have the potential to influence how we approach these infections in burns. Some authors have argued that certain viruses (e.g., human papilloma virus) are commensals in healthy skin [61]. Others, describing the skin's microvirome, have discovered evidence of multiple viruses colonizing the skin of healthy volunteers [62]. This work has been continued in combat extremity wound infections and will be helpful in characterizing further the role that viruses play, not only in infectious complications of burns, but also patterns of microbiology found in burn wounds [63]. These techniques may help clarify the evolution of the microbiome of burn wounds over time (including viruses that have been difficult to study), its clinical significance, and subsequent management.
Footnotes
Author Disclosure Statement
No competing financial interests exist. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Department of the Army, the Defense Health Agency, the Department of Defense, or the U.S. Government. This work was prepared as part of the authors' official duties; and, as such, there is no copyright to be transferred.