Is there any evidence to support the use of empiric antifungals in the setting of esophageal perforation?

Comment by InpharmD Researcher

There is a lack of consensus on routine use of empiric antifungals in the setting of esophageal perforation. Available data, primarily retrospective studies and case reports, remain conflicting regarding mortality benefits. Despite limited evidence, some literature suggests antifungal agents may be beneficial in certain circumstances, particularly for patients with a history of an immunocompromised state, esophageal lesions or infections, prolonged proton pump inhibitor use, or other factors predisposing to fungal infection.

Background

A recent review discusses management of esophageal perforation in the emergency department based on available evidence. Given the risk of gastrointestinal microorganism spillage into the finite potential spaces within the neck, mediastinum, and peritoneum, broad-spectrum antibiotics should be administered upon suspicion of esophageal perforation to ensure adequate coverage for gram positive, negative, and anaerobic bacteria. On the other hand, empiric antifungal coverage is recommended for high-risk patient populations, including those with a history of an immunocompromised state, esophageal lesions or infections, prolonged proton pump inhibitor use, or other factors associated with fungal infection. In such cases, use of intravenous antifungal, such as fluconazole or caspofungin, may be considered along with infectious disease consultation. Instead of being given as a universal empiric regimen, antifungal agents may be used as adjunctive therapies for esophageal perforation in select patients. [1]

A 2012 retrospective review investigated outcomes of patients admitted with esophageal perforation at a single institution. A total of 27 patients were admitted with an isolated perforated esophagus over a period of 6 years. All patients were treated with broad spectrum antibiotics on admission and positive bacterial cultures were obtained in about 75% of patients. Fungal organisms were cultured in pleural or blood samples in 14 patients within the first 7 days, most commonly Candida albicans. Of these patients, 12 received antifungal treatment. Overall mortality in the cohort was 19% (5 of 27), with all 5 non-survivors having yeast cultures from one or more specimens during their hospital stay. Each of these patients received antifungal therapy after diagnosis of infection until time of death. No deaths were reported among patients without yeast cultures; conversely, a mortality rate of 31% (5 of 16) was observed in patients with cultured fungal infection (p<0.001). A positive fungal culture was also associated with increased preoperative requirement of organ support (p= 0.001), ICU stay (p= 0.03), and inpatient hospital stay (p= 0.02). Based on findings from this review, the authors discussed benefits of empirical antimycotic therapy. Development of local candidal sepsis was thought to occur soon after esophageal perforation. Additionally, Candida colonization is considered to be significantly higher in patients with benign esophageal disease, thought to be an inherent risk factor for perforation. Systemic fungal infection confers extra risk in patients with existing complications, and addition of an empiric antifungal may help limit the chances of further systemic infection. Based on this experience as well as other applicable studies finding improved mortality with antifungal prophylaxis, the authors concluded esophageal perforation patients are ideal candidates for empirical antifungal therapy and implement this therapy as routine care in their hospital. [2]

References:

[1] DeVivo A, Sheng AY, Koyfman A, Long B. High risk and low prevalence diseases: Esophageal perforation. Am J Emerg Med. 2022;53:29-36. doi:10.1016/j.ajem.2021.12.017
[2] Elsayed H, Shaker H, Whittle I, Hussein S. The impact of systemic fungal infection in patients with perforated oesophagus. Ann R Coll Surg Engl. 2012;94(8):579-584. doi:10.1308/003588412X13373405388095
Literature Review

A search of the published medical literature revealed 2 studies investigating the researchable question:

Is there any evidence to support the use of empiric antifungals in the setting of esophageal perforation?

Level of evidence

C - Multiple studies with limitations or conflicting results  Read more→


Please see Tables 1-2 for your response.

 

Outcome and management of invasive candidiasis following oesophageal perforation

Design

Retrospective cohort study 

N= 80

Objective

To report clinical outcomes of patients with invasive candidiasis following esophageal perforation 

Study Groups

Participants with microbiological results (N= 68)

Invasive candidiasis (n= 24)

Without invasive candidiasis (n= 44)

Inclusion Criteria

Consecutive patients who were treated as a result of esophageal perforation, positive results for Candida 

Exclusion Criteria

Aged < 16 years, perforations associated with anastomotic leakage after esophageal resection

Methods

Patients' medical records at the study clinic were retrospectively reviewed to collect relevant microbiological results (based on deep biopsies of the mediastinum, pleural secretions, broncho-alveolar secretions [BAL], urine, and blood) and clinical outcomes. Patients with invasive candidiasis were compared to those without invasive candidiasis in regards to mortality rates and use of empiric antifungal agents (2 doses of intravenous [IV] fluconazole 200 mg). 

Two patients with Candida detection in the BAL fluid were excluded from the analysis, and 3 patients who were colonized with an Aspergillus species were included in the analysis in the group of patients without invasive candidiasis.

Duration

Between 1986 and 2010

Outcome Measures

Microbiological results of the overall patient cohort and clinical course of select patients 

Baseline Characteristics

  

All participants (N= 80)

 

  

Etiology of esophageal perforation

Endoscopy⁄endoscopic interventions

Spontaneous perforations

Benign or malignant esophageal lesions

Mediastinoscopy

 

44 (55%)

12 (15%)

22 (27%)

2 (2.5%)

    

Management of esophageal perforation

Watchful waitinga

Interventional therapyb

Surgical therapy 

 

13 (17%)

27 (33%)

40 (50%)

    

aWatchful waiting: Nothing by mouth, IV antibiotic therapy, parenteral nutrition, gastric tube

bInterventional therapy: Esophageal stenting, large bore pleural catheters, endoluminal sealing with fibrin

Results

Endpoint

Positive microbiological results for a Candida species (n= 18)

Without Candida species
(n= 30)

 p-value

Survival in those receiving empiric antifungal (n= 48)

 10/18 (55.6%) 10/30 (33.3%) 0.131
 

Detections overall

 Detections 2009 ⁄ 2010   

Candida species*

Candida albicans

Candida glabrata

Candida parapsilosis

Candida tropicalis

Candida krusei

 

76%

12%

6%

4%

2%

 

53.8%

23.1%

7.7%

7.7%

7.7%

  
  Patients with an invasive mycosis (n= 22)

Patients without positive results for Candida species (n= 44)

 p-value

Mortality

 9/22 (41%) 10/44 (23%) 0.124

Median length of intensive-care therapy, d

 31.2 12.6 for overall cohort  0.039

Overall, 19 patients died because of mediastinitis and/or pneumonia (mediastinitis n= 1, pneumonia n= 7, mediastinitis, and pneumonia n= 11).

The time between perforation and initiation of any therapy (surgical⁄non-surgical ± antibiotics⁄antimycotics) was 24 h in the remaining 42% of the patients. Mortality was significantly lower if any therapy was initiated within 24 h (p= 0.049).

*Together with the shift towards non-albicans species, the proportion of resistant or less susceptible strains to fluconazole was no more than 5% during the first 10 years of the study, whereas it was 26% in the years 2009 and 2010.

Study Author Conclusions

Esophageal perforation is a life-threatening condition. Surgical source control was crucial after the development of mediastinitis and pleural empyema. Deep biopsies support direct monitoring of antifungal therapy. With the growing prevalence of fluconazole resistance and non-albicans Candida strains, an echinocandin is recommended as the first line of therapy. 

InpharmD Researcher Critique

While the study detected no significant difference in survival in those in the patient cohort receiving empiric fluconazole with and without a Candida detection, the results should be interpreted with caution as the study provided no further details on baseline patients characteristics and included a sicker sample population. 



References:

Hoffmann M, Kujath P, Vogt FM, et al. Outcome and management of invasive candidiasis following oesophageal perforation. Mycoses. 2013;56(2):173-178. doi:10.1111/j.1439-0507.2012.02229.x

 

Esophageal perforation secondary to angio-invasive Candida glabrata following hemopoietic stem cell transplantation

Design

Case report 

Case presentation

A 40-year-old female with chronic eosinophilic leukemia, complicated by transformation to precursor B-cell acute lymphoblastic leukemia, underwent induction chemotherapy in July 2001 and subsequently matched unrelated donor hemopoietic peripheral blood stem cell transplant after achieving complete remission in October 2001 with intensive chemoregimen. Severe mucositis developed on day 2, requiring total parenteral nutrition and intravenous narcotic analgesia until day 14. Fever occurred on day 34 after the transplant, while blood cultures remained negative. Various broad-spectrum antibiotics were progressively introduced, and prophylactic intravenous fluconazole (400 mg daily) given from day 1 was stopped on day 10 and replaced by liposomal amphotericin B (3 mg/kg/day) until day 24. Intravenous antivirals were also given for herpes simplex virus and cytomegalovirus prophylaxis.

A chest X-ray and high-resolution computerized tomography (CT) of the chest performed on day 10 demonstrated thickening of the wall of the middle and distal esophagus, with dilatation of the proximal esophagus. As a repeat CT scan performed on day 22 demonstrated persisting, but unchanged abnormality of the esophageal wall, with an increase in the size of the pleural effusions, a diagnostic and therapeutic pleurocentesis was performed on day 34.
 
With recurrent fever, followed by respiratory distress on day 36 with clinical and radiologic features of a right tension pneumothorax, microscopy from pleural fluid showed squamous cells, vegetable material of dietary (esophageal) origin, abundant acute inflammatory cells, red blood cells, and fungal elements; fungal cultures grew C. glabrata in pure culture. A CT performed on day 37 following ingestion of oral contrast demonstrated extravasation of the contrast into the mediastinum and right pleural space, consistent with a ruptured esophagus. Subsequently, therapy with liposomal amphotericin B (3 mg/kg/day) was given for 2 weeks, followed by intravenous voriconazole (3 mg/kg b.d.) for 4 weeks, then ongoing liposomal amphotericin B (1.5 mg/kg three times per week). The patient's condition steadily improved, enabling discharge on day 85. 

Study Author Conclusions

The general approach to management in the majority of patients should be surgery, especially when perforation is recognized late or if septic complications are evident. Antibiotics are used adjunctively following surgery. Empirical antibiotics should cover anaerobes, as well as aerobic Gram-positive and Gram-negative organisms until the results of cultures are available. In neutropenic, immunosuppressed patients, antifungal cover is advocated. Infection with non-albicans Candida species less susceptible to fluconazole should be considered in those receiving fluconazole prophylaxis. Intravenous amphotericin B or other amphotericin products constitute appropriate therapy in this context. Newer agents such as voriconazole, a potent second-generation triazole with broad antifungal activity against yeasts and molds, have been used to treat esophageal candidiasis in immunocompromised patients. 

Esophageal perforation should be considered in patients with pre-existing mucositis, unexplained fever, pleuritic chest pain, and pleural effusion. The described experience suggests that ischemic necrosis secondary to angio-invasion of Candida species may be a possible cause in the appropriate clinical context. Early recognition, surgical intervention, and aggressive antimicrobial therapy are essential in favorably influencing the outcome.
References:

Tran HA, Vincent JM, Slavin MA, Grigg A. Esophageal perforation secondary to angio-invasive Candida glabrata following hemopoietic stem cell transplantation. Clin Microbiol Infect. 2003;9(12):1215-1218. doi:10.1111/j.1469-0691.2003.00762.x