Role of Empiric Anti-Fungal Therapy in the Treatment of Perforated Peptic Ulcer Disease: Review of the Evidence and Future Directions

Abstract

Background:

Peptic ulcer disease (PUD) affects four million people worldwide. Perforated peptic ulcer (PPU) occurs in less than 15% of cases but is associated with significant morbidity and mortality rates. Administration of antibiotics is standard treatment for gastrointestinal perforations, including PPU. Although fungal growth is common in peritoneal fluid cultures from patients with PPU, current data suggest empiric anti-fungal therapy fails to improve outcomes. To examine the role of anti-fungal agents in the treatment of PPU, the Surgical Infection Society hosted an Update Symposium at its 37^th Annual Meeting. Here, we provide a synopsis of the symposium's findings and a brief review of prospective and retrospective reports on the subject.

Methods:

A search of Pubmed/MEDLINE, EMBASE, and the Cochrane Library was performed between January 1, 2000, and November 1, 2018, comparing outcomes of PPU following empiric anti-fungal treatment versus no anti-fungal therapy. We used the search terms “perforated peptic ulcer,” “gastroduodenal ulcer,” “anti-fungal,” and “perforated” or “perforation.”

Results:

There are no randomized clinical trials comparing outcomes specifically for patients with PPU treated with or without empiric anti-fungal therapy. We identified one randomized multi-center trial evaluating outcomes for patients with intra-abdominal perforations, including PPU, that were treated with or without empiric anti-fungal therapy. We identified one single-center prospective series and three additional retrospective studies comparing outcomes for patients with PPU treated with or without empiric anti-fungal therapy.

Conclusion:

The current evidence reviewed here does not demonstrate efficacy of anti-fungal agents in improving outcomes in patients with PPU. As such, we caution against the routine use of empiric anti-fungal agents in these patients. Further studies should help identify specific subpopulations of patients who might derive benefit from anti-fungal therapy and help define appropriate treatment regimens and durations that minimize the risk of resistance, adverse events, and cost.

Peptic ulcer disease (PUD) is a common gastrointestinal disorder affecting as much as 3% of the population [1]. It results from an imbalance between exposure to acid and pepsin in the stomach and duodenum versus epithelial defenses comprising the mucous layer, bicarbonate production, and adequate mucosal blood flow [1,2]. A major contributor to ulcer development is infection with Helicobacter pylori [2]. Colonization with H. pylori causes chronic mucosal inflammation that exacerbates damage from acid exposure and impairs bicarbonate secretion [3]. Recognition of the causative role of H. pylori infection and acid overproduction resulted in recommendations to eradicate the bacterium and administer proton pump inhibitors (PPIs) and histamine-2 receptor antagonists (H₂ blockers). As a result of these measures, recurrence rates and hospitalizations for ulcer complications have decreased significantly over the past 20–30 years [4,5].

The most common complications of PUD are bleeding, obstruction, and perforation [6]. The lifetime prevalence of perforation is approximately 5% [1]. Risk factors include advanced age, multiple medical co-morbidities, smoking, H. pylori infection, and use of alcohol, steroids, and NSAIDs [1,2]. Although bleeding complications are seven times more common than perforation, PPU causes nearly 40% of all ulcer-related deaths [7]. Poor prognostic factors include older age, active co-morbidities, and delays in surgery [7,8]. As such, the Boey and Peptic Ulcer Perforation (PULP) scores predict worse outcomes after PPU in patients with pre-operative shock, concurrent major medical illness, and perforation lasting longer than 24 hours before surgery [9,10].

Sepsis is a leading cause of death following PPU [2,11]. Initial management should adhere to recommendations put forth by the Surviving Sepsis Campaign and the Surgical Infection Society [12,13]. Prompt intravenous (IV) fluid resuscitation and administration of broad-spectrum antibiotics are critical, as are nasogastric tube decompression, pain medication, and IV proton pump inhibitors [12 –14]. The final objective is obtaining source control of the infection, which requires urgent surgical intervention in most patients [15]. Operative treatment of PPU is one of the most common emergency general surgery procedures. The gold standard approach is omental pedicle patch overlay with or without suture repair of the perforation, by either a minimally invasive or a traditional open approach [16,17].

At the time of surgery for PPU, peritoneal fluid cultures show a high incidence of fungal species, predominantly Candida albicans [18]. Several studies indicate that fungal growth in peritoneal fluid is a risk factor for worse outcome [19 –21]. Interestingly, studies do not demonstrate clinical benefits of empiric anti-fungal treatment in these patients [22]. In comparison, delay in treatment of fungal blood stream infections is an independent predictor of death [23]. This apparent discrepancy may represent inadequate study design or underpowered analysis. Considering that Candida spp. are common gastrointestinal flora, it also is possible that fungal growth in peritoneal fluid represents colonization rather than true infection, which may explain the poor correlation between treatment and outcome [24].

In the current environment of increasing anti-bacterial and anti-fungal stewardship aimed at improving clinical outcomes, minimizing toxicity and costs, and curtailing resistance, the Surgical Infection Society hosted an Update Symposium at its 37^th Annual Meeting to define better the role of empiric anti-fungal therapy in the setting of PPU. Here, we summarize the meeting's findings and recommendations and review the relevant literature on the subject.

Clinical Outcomes

We identified five studies published in the defined search period comparing patient outcomes for PPU treated with or without empiric anti-fungal therapy (Table 1). Horn et al. retrospectively queried a single center, prospectively maintained, Acute and Critical Care Surgery database for patients with PPU of the stomach or duodenum with or without obstruction [25]. Those investigators compared the outcomes of patients receiving peri-operative anti-fungal drugs both in the entire cohort and in patients with intra-operative peritoneal fluid cultures positive for Candida spp. Of the 118 patients, 107 underwent surgery for PPU, and 25.2% received pre-operative anti-fungal therapy consisting of fluconazole (85%), micafungin (11%), or anidulafungin (4%). Patients receiving pre-operative anti-fungal therapy were significantly more likely to be taking H₂ blockers. Intra-operative fluid cultures were performed in 33 cases (30.8%), and fungal growth was detected in 17 (51.5%). Fungal isolates included C. albicans (53%), unspeciated yeast (35%), C. glabrata (6%), C. krusei (6%), and C. tropicalis (6%). There were no other significant differences in demographics between the patient groups.

Table 1.

Summary of Perforated Peptic Ulcer Features, Treatment, and Outcome

Series	Type	Inclusion criteria	Exclusion criteria	Study arms	Anti-fungal agents	Distribution of Candida spp.	Follow-up period	Primary outcomes	Secondary outcomes
Horn et al. (2018)	Retrospective review of prospective database	PPU of stomach or duodenum with or without obstruction	N/A	Pre-operative antifungal therapy vs. no therapy	Fluconazole (85%) Micafungin (11%) Anidulafungin (4%)	C. albicans (53%) Unspeciated yeast (35%) C. glabrata (6%) C. krusei (6%) C. tropicalis (6%)	30 d	In-hospital death, mean LOS, mean ICU LOS, mean ventilator days, 30-day re-admission, intra-abdominal abscess, post-operative intra-abdominal fungal abscess, fungemia	N/A
Li et al. (2017)	Retrospective review	Communtity-acquired PPU-associated peritonitis with Candida species in ascites	Age <18, PPU diagnosed after 48 h of arrival, no growth of Candida species	Empiric anti-fungal therapy after surgery vs. no therapy	Fluconazole Micafungin Anidulafungin Caspofungin	Unknown	30 d	30-day all-cause mortality rate	Intra-abdominal abscess or anastomotic leakage, ventilator use for 14 or more days after surgery, ICU stay for 14 or more days after surgery
Vergidis et al. (2016)	Retrospective review	Isolation of Candida from intra-abdominal source and clinical evidence of initial intra-abdominal infection	Non-adults, samples collected from drains in place >24 h	Early anti-fungal treatment within 5 d of positive Candida culture vs. late treatment	Fluconazole alone (50%) Caspofungin (7%) Voriconazole (0.6%) Multitherapy (19%)	C. albicans (57%) C. glabrata (25%) C. parapsilosis (10%) C. tropicalis (5%) C. krusei (3%) C. rugosa (<1%) C. utilis (<1%)	100 d	100-d mortality rate	N/A
Lee et al. (2014)	Prospective cohort	Age ≥21 y, recent abdominal surgery for GI perforation or obstruction, malignancy, pancreatitis, severe peritonitis, or liver transplant, post-operative SIRS despite antibiotics with sepsis from intra-abdominal infection	Age <21 y, received anti-fungal agents before SICU admission, existing positive fungal cultures before ICU admission	Early empiric antifungal treatment within 2 d of sepsis onset vs. culture-directed therapy	Fluconazole (67%) Caspofungin (25%) Anidulafungin (6%) Amphotericin (2%)	Candida unspeciated (42.9%), C. albicans (26.8%), C. tropicalis (12.5%), C. glabrata (10.7%)	Death or discharge from hospital	30-d overall mortality rate	Fungal-related death, SICU LOS, hospital LOS, days to clinical improvement
Sandven et al. (2002)	Multicenter, randomized	Confirmed intra-abdominal perforation with intraoperative specimen for microbiological culture from abdominal cavity	Receiving antifungal treatment when perforation occurred	Single fluconazole dose + antimicrobial regimen vs. placebo + antimicrobial regimen	Fluconazole (100%)	C. albicans (76%) C. glabrata (15%) C. norvegensis (3%) Unidentified yeast (6%)	90 d	Death within 3 mos of surgery	Mechanical ventilation for ≥5 d, ICU LOS ≥10 d, use of central venous catheter ≥10 d

GI = gastrointestinal; ICU = intensive care unit; LOS = length of stay; N/A = not available; SICU = surgical ICU; SIRS = systemic inflammatory response syndrome.

There was no difference in the outcomes of patients who did and did not test positive for fungal growth in their peritoneal fluid. Compared with no treatment, there were no differences in overall length of stay (LOS), intensive care unit LOS, ventilator days, 30-day re-admission rate, or rate of intra-abdominal abscess or fungemia in patients receiving anti-fungal therapy. Lastly, administration of anti-fungal therapy did not improve outcomes in patients who presented with >24 hours of symptoms and established intra-abdominal infection. The authors concluded that pre-operative anti-fungal drugs are unnecessary in patients undergoing surgery for PPU.

In 2017, Li et al. conducted a retrospective analysis of the effect of anti-fungal therapy on the outcomes of patients with community-acquired PPU-associated peritonitis at a single tertiary referral center in Taiwan [26]. Patients aged ≥18 years hospitalized with PPU diagnosed within 48 hours and subsequent growth of Candida spp. from ascites fluid at the time of surgery were included. All patients underwent surgery within 12 hours of diagnosis and received broad-spectrum antibiotic treatment. Physician discretion determined whether fluconazole or an echinocandin (micafungin, anidulafungin, or caspofungin) was administered for at least three days. Of note, all recent isolates of Candida spp. at the study hospital were susceptible to all of these agents. The primary endpoint was 30-day all-cause death, and the secondary endpoints were the presence of intra-abdominal abscess or anastomotic leak found by abdominal computed tomography (CT) or imaging during re-operation and ventilator use or ICU stay for 14 or more days after surgery.

Among the 133 patients, 42.8% received anti-fungal therapy. The distribution of fluconazole versus echinocandin use was not discussed. No significant differences in demographics, laboratory data, or percentage of patients receiving sub-optimal antibiotic therapy were identified. However, greater clinical severity was observed in patients receiving anti-fungal therapy, characterized by higher proportions of fever, tachycardia, shock, acute kidney injury, and high Acute Physiology and Chronic Health Evaluation (APACHE) II scores. On multivariable analysis, shock and high APACHE II scores were independent risk factors for death within 30 days. To account for the independent association of clinical severity with within–30-day death, the authors matched 40 patients from each group using propensity scores. No significant differences were found between treatment groups on any of the primary or secondary endpoints. The authors concluded that anti-fungal therapy does not benefit patients with community-acquired peritonitis in which a Candida spp. is isolated after PPU.

To describe treatment and outcomes and to identify risk factors for death in patients with intra-abdominal candidiasis (IAC), Vergidis et al. performed a retrospective study of adults found to have IAC over a two-year period at a tertiary academic medical center [27]. Intra-abdominal candidiasis was defined in patients with clinical evidence of an initial episode of intra-abdominal infection and isolation of Candida spp. from an intra-abdominal site collected under sterile conditions. Patients were excluded if samples were collected from drains in place longer than 24 hours. The primary study endpoint was death within 100 days. A series of 163 adults with IAC were identified, with 53% having undergone abdominal surgery in the preceding 12 months. The most common causes of IAC were intra-abdominal abscesses (55%), secondary peritonitis (33%), primary peritonitis (5%), infected pancreatic necrosis (5%), and cholecystitis/cholangitis (3%). Secondary peritonitis resulted from gastrointestinal sources in 83% of cases, of which 19% were either gastric or duodenal. The majority of fungal isolates (57%) were C. albicans, followed by C. glabrata (25%), C. parapsilosis (10%), C. tropicalis (5%), C. krusei (3%), and less than 1% each for C. rugosa and C. utilis. Bacterial co-infection occurred in 67% of patients.

In terms of treatments and outcomes, 96% of patients underwent source control via drainage or surgical repair. Early interventions, within five days of positive culture results, occurred in 72%. Secondary peritonitis was treated with surgery in 87% of cases. Anti-bacterial and anti-fungal therapies were given to 99% and 77% of cases, respectively. Early anti-fungal therapy, within five days, was given to 72% of patients, and the median duration of treatment was 14 days. Fluconazole alone was administered to 50% of subjects, whereas caspofungin and voriconazole were given to 7% and 0.6%, respectively. Nineteen percent of patients received multiple agents. The overall 100-day mortality rate was 28% and was highest among patients with primary peritonitis (88%) and secondary peritonitis attributable to hepatobiliary/pancreatic (75%) or gastrointestinal tract (34%) sources. Early source control, the presence of an abscess, and younger age were associated independently with survival in patients with IAC. In patients with secondary peritonitis or abscesses from GI tract sources, younger age, the presence of an abscess, and early anti-fungal therapy were independently associated with a better survival rate. The authors concluded that clinicians are unable to identify patients with IAC who will do well without anti-fungal therapy. Therefore, they recommended immediate anti-fungal treatment for all patients with intra-abdominal infections where Candida spp. are found on culture.

In 2014, Lee et al. conducted a prospective, single-center, one-year observational cohort study on patients in the surgical intensive care unit (SICU) at a tertiary hospital who were admitted after surgery for gastrointestinal (GI) perforation, bowel obstruction or ischemia, malignancy, or anastomotic leak [28]. The objective was to compare the mortality and clinical improvement rates of patients who received anti-fungal treatment within two days of sepsis onset versus those who received anti-fungal therapy only after positive culture results. Patients aged ≥21 years with recent abdominal surgery and persistent post-operative systemic inflammatory response syndrome (SIRS) despite receiving broad-spectrum antibiotics for intra-abdominal infection (peritonitis, colitis, abscess) were enrolled. Patients aged <21 years who either had received anti-fungal agents or had positive fungal cultures before SICU admission were excluded.

The primary endpoint was death within 30 days. Secondary endpoints were the length of the SICU stay and hospitalization; days to clinical improvement as measured by normalization or a decreasing trend in temperature, leukocyte count, hemodynamic status with either wound healing or improved gut function; and fungal-related death. This was defined as on-going sepsis with Candida spp. isolated at any site (except sputum) at death or continued sepsis despite adequate antibiotic coverage for any accompanying bacterial infection and source control in the absence of positive fungal cultures. The only significant differences between the study groups in demographics and clinical presentation were more male patients in the empiric treatment group and more patients with GI perforations in the culture-directed group.

As expected, the median days-to-initiation of therapy was significantly shorter, by four days, in the empiric treatment group. Two-thirds of patients received fluconazole and one-third were given echinocandins. There were significantly more patients with positive fungal cultures in the culture-directed group (100% vs. 53%; p = 0.003). Patients who received empiric anti-fungal therapy were significantly less likely to die within 30 days (odds ratio [OR] 0.25; 95% confidence interval [CI] 0.069–0.905; p = 0.03). Empiric anti-fungal treatment also significantly reduced fungal-related deaths. There were no significant differences in median length of SICU or hospital stay or in the proportion of patients demonstrating clinical improvement. Nevertheless, the number of days to improvement was significantly shorter in the empiric treatment group. Finally, multivariable logistic regression found that early empiric anti-fungal therapy was significantly associated with a lower 30-day mortality rate. The authors concluded that the use of empiric anti-fungal treatment should be reserved for patients at highest risk of invasive candidiasis where the benefit of treatment outweighs the risk.

We identified one randomized trial examining the effect of anti-fungal therapy in patients with PPU. Sandven et al. published a randomized, double-blind, multi-center study in Norway comparing the effect of a single 400-mg intra-operative dose of fluconazole on clinical outcome in patients with intra-abdominal perforations [29]. The study also aimed to determine the significance of recovering Candida spp. from intra-operative abdominal cultures. Intra-abdominal perforations could be spontaneous (primary) or result from a prior surgical procedure (secondary). Patients receiving anti-fungal therapy at the time of the perforation were excluded. The 109 patients were randomized into two groups: 53 received one intra-operative infusion of fluconazole and an antimicrobial regimen against aerobic and anaerobic bacteria, whereas 56 received placebo infusion plus the antimicrobial regimen. The median ages of the groups were 68 and 60 years, respectively. There were no significant differences between the study groups in demographics, clinical characteristics, or presentation. The primary endpoint was death within three months of surgery. Secondary endpoints were the duration of mechanical ventilation, days in the ICU, and days with a central venous catheter.

Of the 109 patients, 73 had a spontaneous perforation, and 36 had a secondary perforation. The overall mortality rate was 11% and was highest for proximal GI perforations. Patients with positive intra-abdominal Candida spp. cultures had a higher risk of death (OR 11.5300; CI 2.3–58.6; p = 0.003). Single-dose fluconazole prophylaxis at the time of surgery did not significantly reduce death (7.5% vs. 14.3%; OR 0.21; CI 0.04–1.06; p = 0.059). Fluconazole prophylaxis also did not reduce the probability of finding Candida spp. post-operatively from abdominal drainage fluid. On the basis of these results, the authors concluded that single-dose fluconazole prophylaxis does not significantly improve survival in patients with intra-abdominal perforations, whereas positive operative cultures were associated with higher morbidity and mortality rates. The authors also cautioned that the findings should not lead to administration of anti-fungal agents to all patients with GI perforations.

Critical Appraisal of Trials

There are important limitations to consider regarding the aforementioned studies. The study by Li et al. is retrospective and thus prone to inherent selection biases, including missing data and confounding related to nonrandomized patient allocation [26]. Moreover, it is a single-site study, which limits reproducibility. The study enrolled a small number of patients, thus increasing the probability of a Type II error. The authors included only patients with community-acquired PPU-associated peritonitis, which limits the external validity to hospitalized patients who develop PPU. There were no significant differences in the demographics or laboratory data, but there was a higher degree of disease severity in the anti-fungal therapy group and a trend toward higher positive bacterial growth in the ascites cultures (p = 0.07). The authors did not identify all of the species of Candida isolated from the peritoneal cultures. Although all recent C. albicans isolates at the study hospital were susceptible to the anti-fungal regimens, it is possible treatment was suboptimal if current Candida spp. isolates developed resistance to the chosen drugs.

The study by Horn et al. is prone to similar biases, as it is retrospective and included a small number of patients from a single center [25]. The authors reported that patients receiving pre-operative anti-fungal therapy were more likely to be taking H₂ blockers (p = 0.02), although the role of these drugs in peritonitis is unclear. There were no other differences in the demographics of the patients. The decision to use pre-operative anti-fungal drugs was attending physician dependent. The factors determining when to administer anti-fungal treatment are unknown. Therefore, it is possible that anti-fungal agents were given to sicker patients. The anti-fungal regimen was not uniform, as 15% of subjects did not receive fluconazole. The susceptibility of the fungal isolates to the prescribed drugs was not discussed. Nearly half of the isolates (47%) were confirmed as either non-C. albicans or unspeciated. Considering that both classes of organisms have a lower probability of susceptibility to fluconazole, which was administered in 85% of cases, this raises some doubt as to the appropriateness of the anti-fungal therapy.

Aside from the retrospective biases discussed above, the subjects in the study by Vergidis et al. had significant heterogeneity in the types of intra-abdominal candidiasis, ranging from abscesses to primary and secondary peritonitis and infected pancreatic necrosis [27]. The sites of origin for secondary peritonitis included the stomach, small intestine, colon, liver, and pancreas. Thus, the study was not powered to allow definitive conclusions regarding specific treatment for PPU. The authors showed that patients who received an infectious disease consultation were more likely to receive an anti-fungal agent. Although consultation itself was not associated with better outcomes, there may be biases in that consults were obtained for cases that are more difficult. The authors also reported that anti-fungal susceptibility testing was requested on only 13% of Candida spp. isolates. Considering that more than 30% of the isolates from patients with secondary peritonitis grew C. glabrata, and this species demonstrates resistance to multiple drugs, it is unclear whether this affected the results. Lastly, compared with non-survivors, there was a trend for survivors to have a source control intervention within five days (p = 0.09), which is known to improve outcomes [12,13].

The prospective observational study by Lee et al. has important limitations as well [28]. The sociodemographics, co-morbidities, and severity of illness were similar in the groups, except there were more males in the empiric therapy group; and there were more patients with gastrointestinal perforations in the culture-directed arm. Moreover, the target sample size was not achieved because of slow patient accrual. Non-randomized allocation could have resulted in treatment bias, as patients who received empiric anti-fungal therapy may have had more aggressive treatment overall. The culture-driven treatment group had higher numbers of positive fungal cultures and significant (accompanying signs and symptoms of infection) fungal cultures. This finding confounds the results because positive fungal cultures have been shown to portend a worse prognosis in certain circumstances [19 –21]. It also is possible that a substantial number of patients in the empiric-treatment group never had a fungal infection. Lastly, measures of clinical improvement (other signs indicative of sepsis) were subjective and thus prone to information bias.

The randomized, double-blind, multi-center study by Sandven et al. was the most robust study of the group [29]. There were adequate descriptions of the allocation method and concealment, attrition, and exclusion of patients after randomization; predefined primary and secondary outcomes; and power calculations for the group sizes. Notwithstanding these measures, several limitations still exist. The study was not designed to look specifically at patients with PPU. It was therefore not adequately powered to discern an effect of anti-fungal therapy on this population. Only 25% of perforations occurred in the stomach and duodenum, and nearly 25% of these lesions were found post-operatively, which adds to the heterogeneity of the potential study population. The target population of 120 was not reached because of slow patient recruitment. Although the intention was for all study hospitals to contribute a minimum of five patients to the analysis, one hospital contributed only one patient. The antibiotic regimens differed between sites, and durations of treatment were not reported. If deemed necessary, the clinician in charge could change the antibiotic regimen independently of the anti-fungal regimen, but the rationale was not described further. Lastly, the external validity of the findings may be limited by differences in host patient demographics, Candida spp. prevalence, disease virulence, or other unrecognized treatment variables in Norway versus the United States.

Antimicrobial Stewardship

In view of the overutilization of antibacterial compared with anti-fungal agents and the high prevalence of multi–drug-resistant bacteria, antimicrobial stewardship has traditionally focused on antibacterial agents [30 –32]. With greater use of anti-fungal agents for both prophylaxis and treatment of immunosuppressed patients, the focus of antimicrobial stewardship has shifted to include these agents as well [30]. Moreover, poor outcomes associated with fungal infections have led to expanded use of anti-fungal drugs. As with antibacterial agents, the potential untoward effects of indiscriminate anti-fungal administration mandate a careful reassessment of the appropriateness of these agents across multiple clinical settings. These agents are associated with serious adverse drug effects (e.g., renal, hepatic, and hematologic toxicities), selection for resistance, and high associated cost [33].

Candida spp. display variable susceptibility patterns to anti-fungal drugs. Historically, fluconazole demonstrated excellent in vitro activity against C. albicans. In contrast, the product has limited, dose-dependent, activity against C. glabrata; and C. krusei demonstrates intrinsic resistance [33]. Alarmingly, cases of fluconazole resistance in C. albicans have been reported [34]. Frequent or prolonged use of fluconazole may lead to selection of more resistant Candida isolates [35, 36]. Compounding this problem is the emergence of healthcare-associated, multi–drug-resistant species, as illustrated by recent outbreaks of C. auris [37].

The study by Vergidis et al. found 30% of the isolates to be non-albicans species [27]. This highlights the idea that a one-size-fits-all strategy of providing fluconazole empirically might provide insufficient coverage. When given empirically or prophylactically, fluconazole may select more resistant species, which then reduces its effectiveness and that of other anti-fungal agents. This has led to greater utilization of broader-spectrum agents, such as echinocandins and amphotericin B, for the empiric and definitive treatment of invasive candidiasis in critically ill patients with intra-abdominal infections [13,38]. This shift to broader-spectrum agents, as seen with antibacterial agents, likely will provide only a temporary solution. Indeed, resistance to echinocandins in Aspergillus spp. is well described, and some Candida spp. have intrinsically higher minimum inhibitory concentrations to this class. More judicious use of these agents clearly is warranted [34,39].

Unlike the usual situation in bacterial infections, candidiasis therapy did not trigger a need for anti-fungal susceptibility testing. New guidelines call for augmented testing strategies [38]. To this end, the Clinical and Laboratory Standards Institute provides susceptibility breakpoints for select species and anti-fungal drugs [40]. Anti-fungal susceptibly testing will continue to help guide effective culture-guided treatment, monitor for resistance development, and possibly enhance future studies in the context of previous studies that were confounded by assumed susceptibility [41].

Anti-fungal therapy represents a spectrum of approaches ranging from pre-operative prophylaxis to extended empiric courses to culture-guided initiation with or without selective use for septic patients. As demonstrated in the study by Lee et al., empiric strategies decrease the time to effective therapy relative to culture-guided strategies [28]. Considering that isolation of fungal species may be delayed relative to that of their bacterial counterparts, utilization of rapid diagnostic methods provides the opportunity to improve time to detection of invasive fungal infections [41]. The role for these modalities in the management of intra-abdominal infections, in particular those resulting from perforations, remains unclear. These modalities may identify fungal species more rapidly but fail to discriminate between colonization and true infection because of limited specificity [31].

The studies reviewed here implemented variable durations of anti-fungal therapy, from one dose pre-operatively to more than two weeks of administration [25 –29]. As noted, prolonged use may select for resistance and increase adverse events. The most appropriate anti-fungal drug duration remains unknown. Notwithstanding, a post-hoc analysis of the STOP-IT trial suggested that abbreviated treatment results in similar outcomes if utilized in tandem with source control [42].

In summary, as the prevalence of fungal infections increases, healthcare providers must remain strict stewards of anti-fungal agents by administering them only when indicated and for the shortest possible time. Through careful and appropriate use of anti-fungal agents, practitioners can prevent unnecessary exposure, reduce toxicity, and decrease selection pressure for resistant isolates.

Future Directions

Multiple unresolved topics of PPU treatment require additional studies. The efficacy of empiric anti-fungal therapy in this patient population remains unproved. Differences in community-acquired versus hospital-acquired PPU disease require further elucidation. Management strategies for patients with significant co-morbidities or higher severity of disease need additional clarification. Geographic differences between patient populations and disease pathogenesis deserve further investigation. With heightened awareness of the risks of anti-fungal therapy, a cost–benefit analysis can define their utility better. Improved tracking of adverse drug events, emerging resistance patterns, and dysbiosis may help as well.

Conclusions

Patients with PPU undergoing surgery often demonstrate positive intra-abdominal fungal cultures, which are associated with worse outcomes. It has been suggested that empiric administration of anti-fungal agents be included in the management of these cases. On the basis of evidence presented at the Surgical Infection Society Update Symposium and recent literature reviewed here that does not clearly support the therapeutic efficacy of this approach, we caution against the use of empiric anti-fungal agents in these circumstances. Additional studies are needed to identify better the patient populations in which the addition of anti-fungal therapy would yield clinical benefits.

Footnotes

Author Disclosure Statement

No competing financial interests exist for any of the authors.

References

Chung

, Shelat

. Perforated peptic ulcer: An update. W J Gastrointest Surg, 2017; 9:1–12.

Soreide

, Thorsen

, Harrison

, et al. Perforated peptic ulcer. Lancet, 2015; 386:1288–1298.

Sung

, Kuipers

, El-Serag

. Systematic review: The global incidence and prevalence of peptic ulcer disease. Aliment Pharmacol Ther, 2009; 29:938–946.

Hermansson

, Ekedahl

, Ranstam

, et al. Decreasing incidence of peptic ulcer complications after the introduction of the proton pump inhibitors: A study of the Swedish population from 1974–2002. BMC Gastroenterol, 2009; 9:25.

Bashinskaya

, Nahed

, Redjal

, et al. Trends in peptic ulcer disease and the identification of Helicobacter pylori as a causative organism: Population-based estimates from the US Nationwide Inpatient Sample. J Glob Infect Dis, 2011; 3:366–370.

Malfertheiner

, Chan

, McColl

. Peptic ulcer disease. Lancet, 2009; 374:1449–1461.

Thorsen

, Soreide

, Kvaloy

, et al. Epidemiology of perforated peptic ulcer: Age and gender adjusted analysis of incidence and mortality. World J Gastroenterol, 2013; 19:347–354.

Moller

, Adamsen

, Thomsen

, et al. Preoperative prognostic factors for mortality in peptic ulcer perforation: A systematic review. Scand J Gastroenterol, 2010; 45:785–805.

Boey

, Choi

, Poon

, et al. Risk stratification in perforated duodenal ulcers: A prospective validation of predictive factors. Ann Surg, 1987; 205:22–26.

10.

Moller

, Engebjerg

, Adamsen

, et al. The Peptic Ulcer Perforation (PULP) score: A predictor of mortality following peptic ulcer perforation. A cohort study. Acta Anaesthesiol Scand, 2012; 56:655–662.

11.

Thorsen

, Soreide

. Long-term mortality in patients operated for perforated peptic ulcer: Factors limiting longevity are dominated by older age, comorbidity burden and severe postoperative complications. World J Surg, 2017; 41:410–418.

12.

Dellinger

, Levy

, Rhodes

, et al. Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock. Crit Care Med, 2013; 41:580–637.

13.

Mazuski

, Tessier

, May

, et al. The Surgical Infection Society Revised Guidelines on the Management of Intra-Abdominal Infection. Surg Infect, 2017; 18:1–56.

14.

Sturkenboom

MCJM

, Goettsch

, Picelli

, et al. Inappropriate initial treatment of secondary intra-abdominal infections leads to increased risk of clinical failure and costs. Br J Clin Pharmacol, 2005; 60:438–443.

15.

Thorsen

, Soreide

. What is the best predictor of mortality in perforated peptic ulcer disease? A population-based, multivariable regression analysis including three clinical scoring systems. J Gastrointest Surg, 2014; 18:1261–1268.

16.

Soreide

, Thorsen

, Soreide

. Strategies to improve the outcome of emergency surgery for perforated peptic ulcer. BJS, 2014; 101:e51–e64.

17.

Thorsen

, Glomsaker

, von Meer

, et al. Trends in diagnosis and surgical management of patients with perforated peptic ulcer. J Gastrointest Surg, 2011; 15:1329–1335.

18.

van Ruler

, Kiewiet

JJS

, van Ketel

, Boermeester

. Initial microbial spectrum in severe secondary peritonitis and relevance for treatment. Eur J Clin Microbiol Infect Dis, 2012; 31:671–682.

19.

Shan

, Hsu

, Hsieh

, et al. Significance of intraoperative peritoneal culture of fungus in perforated peptic ulcer. Br J Surg, 2003; 90:1215–1219.

20.

Montravers

, Dupont

, Gauzit

, et al. Candida as a risk factor for mortality in peritonitis. Crit Care Med, 2006; 34:646–652.

21.

Prakash

, Sharma

, Saxena

, et al. Effect of Candida infection on outcome in patients with perforation peritonitis. Indian J Gastroenterol, 2008; 27:107–109.

22.

Montravers

, Mira

, Gangneux

, et al. A multicenter study of anti-fungal strategies and outcome of Candida spp. peritonitis in intensive-care units. Clin Microbiol Infect, 2011; 17:1061–1067.

23.

Garey

, Rege

, Pai

, et al. Time to initiation of fluconazole therapy impacts mortality in patients with candidemia: A multi-institutional study. CID, 2006; 43:25–31.

24.

Bassetti

, Marchetti

, Chakrabarti

, et al. A research agenda on the management of intra-abdominal candidiasis: Results from a consensus of multinational experts. Int Care Med, 2013; 39:2092–2106.

25.

Horn

, Coleoglou Centeno

, Rasane

, et al. Pre-operative anti-fungal therapy does not improve outcomes in perforated peptic ulcers. Surg Infect, 2018; 19:587–592.

26.

, Lee

, Liu

. Anti-fungal therapy did not improve outcomes including 30-day all-cause mortality in patients suffering community-acquired perforated peptic ulcer-associated peritonitis with Candida species isolated from their peritoneal fluid. J Microbiol Immunol Infect, 2017; 50:370–376.

27.

Vergidis

, Clancy

, Shields

, et al. Intra-abdominal candidiasis: The importance of early source control and anti-fungal treatment. PLoS One, 2016; 10;e0153247.

28.

Lee

, Liew

, Chlebicki

, et al. An observational study on early empiric versus culture-directed anti-fungal therapy in critically ill with intra-abdominal sepsis. Crit Care Res Pract, 2014; 2014:479413.

29.

Sandven

, Qvist

, Skovlund

, et al. Significance of Candida recovered from intraoperative specimens in patients with intra-abdominal perforations. Crit Care Med, 2002; 30:541–547.

30.

Urbancic

, Thursky

, Kong

DCM

, et al. Anti-fungal stewardship: Developments in the field. Curr Opin Infect Dis, 2018; 31:490–498.

31.

Ananda-Rajah

, Slavin

, Thursky

. The case for anti-fungal stewardship. Curr Opin Infect Dis, 2012; 25:107–115.

32.

Dellit

, Owens

, McGowan

Jr , et al. Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing an institutional program to enhance antimicrobial stewardship. Clin Infect Dis, 2007; 44:159–177.

33.

Lewis

. Current concepts in anti-fungal pharmacology. Mayo Clin Proc, 2011; 86:805–817.

34.

Whaley

, Berkow

, Rybak

, et al. Azole anti-fungal resistance in Candida albicans and emerging non-albicans Candida species. Front Microbiol, 2016; 7:2173.

35.

Slavin

, Sorrell

, Marriott

, et al. Candidaemia in adult cancer patients: Risks for fluconazole-resistant isolates and death. J Antimicrob Chemother, 2010; 65:1042–1051.

36.

Lortholary

, Desnos-Ollivier

, Sitbon

, et al. Recent exposure to caspofungin or fluconazole influences the epidemiology of candidemia: A prospective multicenter study involving 2,441 patients. Antimicrob Agents Chemother, 2011; 55:532–538.

37.

Spivak

, Hanson

. Candida auris: An emerging fungal pathogen. J Clin Microbiol, 2018; 56.

38.

Pappas

, Kauffman

, Andes

, et al. Clinical practice guideline for the management of candidiasis: 2016 update by the Infectious Diseases Society of America. Clin Infect Dis, 2016; 62:e1–e50.

39.

Wiederhold

. Anti-fungal resistance: Current trends and future strategies to combat. Infect Drug Resist, 2017; 10:249–259.

40.

Clinical Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing for Yeasts. 1^st ed. CLSI Supplement M60. Wayne, PA: The Institute; 2017.

41.

Albataineh

, Sutton

, Fothergill

, Wiederhold

. Update from the laboratory: Clinical identification and susceptibility testing of fungi and trends in anti-fungal resistance. Infect Dis Clin North Am, 2016; 30:13–35.

42.

Elwood

, Guidry

, Duane

, et al. Short-course antimicrobial therapy does not increase treatment failure rate in patients with intra-abdominal infection involving fungal organisms. Surg Infect, 2018; 19:376–381.