In the acute phase of burn injury, thermal damage to the human skin barrier alongside local and systemic immune dysfunction, predisposes patients to potentially serious fungal wound infections (Church et al, 2006). The nature of burn injuries often result in complex intensive care and exposure to multiple risk factors for fungal infection e.g. prolonged ICU stay, central lines, TPN feed and systemic antibiotics (Ballard et al, 2008).
Significantly increased mortality is associated with Aspergillus infection in comparison to other fungi due to the angioinvasive nature of disease. Increasing age, burn size and number of culture sites are all independently associated with increased mortality (Ballard et al, 2008; Murray et al, 2008).
Significant pulmonary infiltration has also been demonstrated, particularly in patients with inhalational injury (Ballard et al, 2008; Murray et al, 2008).
The incidence may have risen due to the suppression of bacterial infections with use of silver sulphadiazine and prompt surgical excision. Aspergillosis affects 0.4-7% of hospitalised patients with burns.
Aspergillus infection of the burn wound is due to airborne contamination of the burned skin/fascia (Becker et al, 1991). Open dressings are associated with an increased incidence of infection than occlusive dressing (Mouza et al, 1999) as are those with extensive burns (>50%), full thickness burns, prolonged hospital stay and the use of broad spectrum antibiotics. High quality air filtration may reduce the number of cases (Levenson et al, 1991; Mouza et al, 1999). Bandages may be a source of infection (Bryce et al, 1996).
Aspergillus has also been isolated in cadaveric allografts, despite adherence to the American Association of Tissue Banks guidelines (Obeng et al, 2001). The use of prophylactic antifungal agents has been discouraged by several trials due to potential resistance. However, this is disputed by other studies, arguing the strong association between total burn size and fungal wound infection warrants consideration of empiric or early prophylactic treatment (Horvath, 2007).
Early wound closure and surgical excision may be the most significant preventative measure (Capoor et al, 2010; Hovarth et al, 2007).
Clinical Features and diagnosis
Aspergillosis of burn wounds usually occurs 2-8 weeks after injury (Stone et al, 1979; Becker et al, 1991, Caetano 2008). Clinically the burn changes in character becoming dark and typically full thickness, although early features can be subtle often making it difficult to diagnose (Bruck, 1971; Nash et al, 1971; Salisbury et al, 1974; Pruitt et al, 1984). Often the fungus will sporulate on the wound (Panke et al, 1978). Species implicated include A. flavus,A. fumigatus, A. niger, A. terreus (amphotericin B resistant) and A. ustus (Chakrabarti et al, 1998).
Microscopy from direct specimens or frozen sections of the abnormal areas will often show hyphae and reveal the genus of fungus involved. Cultures of tissue have a higher yield than swabs (Mouza, 1999). Cultures are usually positive rapidly and affected areas often spread quickly.
Diagnosis of Aspergillus burn wound infection can be difficult due to colonisation, co-presence of bacteria and empiric treatment with antifungal drugs (Morace, 2010). No clear consensus is reported regarding the method of distinguishing colonisation from infection; Hovarth et al defined infection as the presence of fungus in viable tissue whereas fungal elements confined to the eschar were defined as colonisation (Hovarth et al, 2007).
The sensitivity of culture is limited, and it can take at least 48 hours; but diagnosis of species by this method is valuable for guidance of appropriate anti-fungal therapy (Morace, 2010). Correlation between histopathological diagnoses and culture appears inconsistent (Schofield, 2007). Other methods may be used for rapid diagnoses such as detection of galactomannan (GM), a polysaccharide on the outer wall of Aspergillus and PCR; these methods have variable sensitivity and specificity (Wheat 2006).
The utility of GM is limited in patients receiving systemic β-lactam antimicrobial therapy due to false positives (Bart-Delabeese et al, 2005). The use of GM and other diagnostic biomarkers has been widely investigated for diagnosis of invasive aspergillosis infection in immuno-compromised patients, however, the effectiveness specifically in burn wound infection is not reported in the literature.
PCR offers a rapid diagnostic approach for skin or fascia samples, or blood and is a more sensitive alternative to traditional culture, but no experience has been published in burn would infections (Morace 2010), although fresh tissue can be successfully processed (Lass-Flörl, 2011).
Treatment
Aggressive therapy is warranted in burns patients with a diagnosis of Aspergillus infection due to the potentially invasive nature of the disease; studies have demonstrated a 12-fold increase in mortality with the isolation of Aspergillus in comparison to other fungal wound infections (Ballard et al, 2008).
Surgical Treatment
Radical debridement of an Aspergillus infection of a burn wound is mandatory and even then may fail to control the infection. The affected areas should be inspected and redressed daily after debridement and further debridement carried out as required from visual inspection.
At the limits of infection, biopsies examined histologically will indicate whether debridement is extensive enough. Amputation may be necessary if only distal involvement is present. Skin grafting should be delayed until infection has resolved and negative swabs obtained (Capoor et al, 2010; Hovarth et al, 2007).
Medical Therapy
Voriconazole is recommended for the systemic treatment of Aspergillus wound infections due to its broad spectrum; the innate resistance of A. terreus and side effect profile of amphotericin B have seen it replaced by voriconazole therapy as first line therapy (Walsh 2008; Capoor et al, 2010).
Voriconazole has variable pharmacokinetics and requires stringent monitoring. Interestingly, higher doses are often required in paediatric patients with significant inter-patient variability demonstrated (Pasqualotto et al, 2007). Caspofungin (or another echinocandin) may be useful in renal failure (Capoor, 2011).
A topical alternative approach was described by one group using high concentration nystatin powder (6,000,000u/g) (Pharma-Tek Inc., Huntingdon, NY) after early fascial excision, followed by auto or homografting. When Aspergillus or Fusarium infection occurred, and failed further excision and systemic amphotericin B, the authors applied 4-6 grams of nystatin powder daily, followed by wet and dry dressings, with success (Barrett et al, 1999).
An in vitro study investigating the effect of silver on Aspergillus spores demonstrated some success with use of silver coated dressings; however the use of silver sulfadiazine was not effective for Aspergillus burn wound infection (Wright 2005; Wright et al, 1999).
The role of adequate hospital infection control has been postulated by Capoor et al. who demonstrated the presence of differing Aspergillus species in the environment in comparison to colonisation from the patient in 25 severely burnt patients. There is currently no evidence regarding the efficacy of routine surveillance wound swabs.
The overall mortality was about 50% (30-78%) despite surgical debridement and antifungal therapy in the 1980’s but appears to be falling to ~25% currently (Ballard, 2008). The mortality is highest in those with extensive burns.
Helena O’Flynn and David W. Denning
National Aspergillosis Centre
University Hospital of South Manchester
Manchester UK.
September 2011
helenaoflynn@doctors.org.uk
ddenning@manchester.ac.uk
| Images of aspergillosis in burned patients |