Case of the Month

August 2011

Drs Riaan Writes and Madaleen Jansen van Vuuren – PathCare

Case history

A 51 year old female patient with chronic epigastric pain and reflux was admitted for elective gastric surgery. She was hypertensive on treatment and a smoker. A laparotomy was performed with a Nissen fundoplication, Roux-en-Y with reconstruction, vagotomy and antrectomy. A splenectomy had to be performed secondary to an iatrogenic splenic injury. After initial uneventful postoperative convalescence and discharge from ICU to a general surgical ward on day 5, the patient was readmitted into ICU on day 11 after barium swallow showed total gastric obstruction, anastomotic breakdown and gastric perforation. A second laparotomy was performed to repair the anastomosis, gastric perforation and an oesophageal tear. No mediastinitis was evident. Cefoxitin was used as intra-operative prophylaxis (both operations).

CRP remained high and the patient was started on imipenem and metronidazole postoperatively. She also received TPN. Metronidazole was discontinued after 5 days. Fever spikes were noted from day 5 of ICU admission. Teicoplanin was added on day 6. The patient was intubated on day 9 due to desaturation and possible LRTI. Fluconazole was started on day 10. Low dose IV hydrocortisone was added. Pseudomonas aeruginosa was cultured from tracheal aspirate. The patient was intubated and ventilated for a total of 11 days. Intermittent spiking fevers continued despite the patient being extubated on day 20 and more than 3 weeks of treatment with imipenem, teicoplanin and fluconazole. Several blood cultures taken during this period were negative. The patient developed clinical wound sepsis and fistulas at drain sites. Pus from the fistula isolated Citrobacter freundii, Klebsiella pneumoniae (non-ESBL) & Enterococcus faecalis. Candida was cultured from wound/pus swabs on several occasions. The patient received 2u packed cells (Hb 9) but never had neutropaenia. Repeated contrast visualization studies and abdominal CT-scans revealed no further evidence of perforations, anastomotic breakdown or intra-abdominal fluid collections.

The patient was discharged from ICU to general ward for wound care and resolution of the fistula after 34 days. Documentation of intermittent fever spikes continued. Saccharomyces cerevisiae was isolated from a blood culture taken on day 30 of readmission to ICU. On enquiry it was found that a Probiotic (INTEFLORA® 250) was added to the patient’s prescription on day 4 of readmission to ICU. Inteflora contains Saccharomyces boulardii. The probiotic was discontinued after which the fever subsided. The patient was discharged after spending 10 weeks in hospital and is currently progressing well and gaining weight.

Question 1: How common are non-Candida invasive yeast infections?

Answer to Q1

The epidemiology of invasive yeast infections is changing due to increasing numbers of immunosuppressed patients resulting in more frequent diagnosis of invasive infections by unusual yeasts. Use of fluconazole is also considered a reason for a change in prevalence of Candidaemia due to non-albicans Candida species. The ARTEMIS Global Antifungal Surveilance program found C. albicans to be the most common cause of invasive fungal infection (63 – 70%), followed by C. glabrata (44%), C. tropical (6%) and C. parapsilosis (5%). C. krusei, C. guillermondii, C. lusitanaie, C. kefyr, C. inconspicua, C. famata and C. rugosa constitute the remainder of the ten most frequently isolated Candida species. Geographic variations do occur.

Non-Candida yeasts, apart from Cryptococcus neoformans, are still rare (4.1% of total yeast isolates). Little is known about epidemiological traits of these organisms, they are often difficult to identify phenotypically and may show variability in antifungal susceptibility. Saccharomyces spp. is the second most common of the non-Candida yeasts isolated (11.7%), Trichosporon spp. third (10.6%), followed by Rhodotorula spp. (4.1%). Other yeasts uncommonly isolated include Geotrichum spp., Hanensula and Pichia spp. and Malassezia spp.

Question 2: What are the risk factors for invasive yeast infections?

Answer to Q2

Invasive candidiasis is the most common invasive yeast infection. Emerging invasive yeast infections share the risk factors for invasive candidiasis. Risk factors can be divided into three groups.

  1. Increased colonization by Candida spp. (or other yeast). This can be the result of endogenous or exogenous factors. Broad spectrum antibiotics that suppress endogenous flora and allow overgrowth of Candida at mucosal sites. Prolonged hospital stay increases the risk for acquisition of Candida strains from the environment and health care workers. Length of stay in intensive care is especially important. The rate of infection rise rapidly after 7-10 days.
  2. Compromised integrity of mucosal barriers. Damage to the gut mucosa leading to increased translocation of organisms. This is seen with surgery, TPN, malnutrition, severe burns, chemotherapy induced mucositis and graft-versus-host disease. The presence of a central venous catheter can also provide a port of entry.
  3. Immunosuppression. Conditions that suppress T cell and phagocytic immunity predispose to invasive fungal infections and include prematurity, severe burns, haemodialysis, TPN, malignancy (esp. haematological), neutropaenia, AIDS, immunosuppressive therapy with corticosteroids, chemotherapy and transplant patients (bone marrow and solid organ).

Risk factors associated with Saccharomyces infection is similar to the risk factors reported for invasive candidiasis and also include treatment with a probiotic containing S. cerevisiae subtype boulardii.

Malassezia spp. is the only other yeast with a specific risk factor. Fungemia is often related to lipid infusions.

The patient discussed in this case had a splenectomy. Individuals who are asplenic or have impaired splenic function are at increased risk of developing life-threatening infections, especially due to encapsulated bacteria. This risk is higher in children, but adults can also develop fulminant infection or "post splenectomy sepsis" (PSS). There is a paucity of data in the literature and a lack of detailed case reports regarding splenectomy as a risk factor for invasive yeast infections. A single case report of disseminated C. neoformans infection in an asplenic patient was found. In a review of Cryptococcal infection in HIV-negative patients, splenectomy was reported to be a risk factor for infection in only 3% of cases.

Question 3: Describe the laboratory diagnosis of invasive yeast infections?

Answer to Q3

Diagnosis of an invasive yeast infection is often based on a combination of clinical and laboratory considerations but mostly still rely on culture of organisms from blood and other sterile sites.

Direct microscopy

Direct microscopy of clinical specimens from tissue sections and normally sterile body fluids for budding yeast cells and hyphae provides a rapid result, but is less sensitive than culture. However, demonstration of tissue invasion is necessary when interpreting results from certain body sites where colonization by Candida and other yeasts that may serve as opportunistic pathogens occur.


Detection of fungemia is useful in the definitive diagnosis of invasive yeast infection and blood cultures should be performed in all cases of suspected invasive fungal infection. Blood culture may be negative despite disseminated disease. It is positive in only 50% of cases of disseminated infection with Candida spp. and results may be delayed. Culture of an adequate volume of blood is critical for optimal detection and at least 20 – 30 mL is required.

Identification of yeasts in most laboratories is based on carbohydrate assimilation and fermentation. Morphological features on specialized media that incorporate chromogenic substrates to detect species specific enzyme activity is used by some, but is restricted to identification of a few Candida species.


The insensitivity of culture based methods to detect invasive fungal infection has led to a number of screening strategies involving serological methods to assist in earlier diagnosis. Detection of (1,3)-β-D-glucan (BG) is one such assay that is more widely available. Glucans are cell wall components of most pathogenic fungi and can be detected in serum at levels as low as 1 pg/mL. The FDA approved Fungitell ® BG assay, is a colorimetric assay that indirectly determine the concentration of BG in serum, based on the activation of a BG-sensitive proteolytic coagulation cascade using components purified from the horseshoe crab.

(1,3)-β-D-glucan is a broad spectrum marker and detects invasive infections with Candida spp (less sensitive for C. parapsilosis), Trichosporon and Saccharomyces yeasts as well as several moulds including Aspergillus, Fusarium, Scedosporium, Acremonium, but cannot distinguish between fungal species. Zygomycetes and Cryptococcus is not detected.

Obtaining multiple samples increases the sensitivity, specificity and positive predictive value of the assay. At a cut-off value of 60 pg/mL, the sensitivity and negative predictive value of twice weekly sampling is 100%. An overall sensitivity of 70 – 100% and specificity of 87 – 96% to detect probable invasive fungal infections was found in studies. Positive results must be further evaluated with radiological and microbiological methods.

False positive results is common and causes include Gram positive bacteraemia, haemolysis of serum, haemodialysis with cellulose membranes, therapy with intravenous immunoglobulins, albumin, coagulation factors, plasma protein factor and use of glucan containing gauze products. A single positive result must be confirmed by testing a follow-up specimen. High concentration of bilirubin and trigliserides can inhibit the assay and lead to false negative results.

Detection of fungal DNA

Several molecular methods for the detection of fungal DNA in critically ill patients have been evaluated in recent years, offering the potential for rapid diagnosis. Target sequences vary widely, including genus and species specific variable regions as well as highly conserved regions of the fungal genome. Single and multicopy gene targets have been studied. Use of multicopy ribosomal targets offers potentially sensitive panfungal markers, followed by identification at genus and species levels. Most reports in the literature indicate sensitivity of PCR based methods to be equal to or better than other currently used methods. However, routine use of PCR for the detection of fungal DNA is not yet a reality because of lack of a standardised and validated commercial method. Real time PCR techniques may change this in the near future.

Question 4: Describe the epidemiology, clinical characteristics and treatment of invasive Saccharomyces infection.

Answer to Q4

Saccharomyces is an Ascomycetous yeast represented by S. cerevisiae, also known as “baker’s or brewer’s yeast”. It is widespread in nature and found on plants, fruit and in soil. Saccharomyces cerevisiae subtype boulardii is also used in probiotic preparations. S. cerevisiae colonize mucosal surfaces and is part of the normal flora of the gastrointestinal tract, respiratory tract and vagina. It is not known if S. cerevisiae is a persistent commensal or only transiently present after food ingestion.

Invasive infections with S. cerevisiae occur in immunosuppressed patients, critically ill patients, but also in relatively healthy individuals. S. cerevisiae accounts for 0.1 – 3.6% of fungemia episodes in some population based studies. Eighty percent of cases of S. cerevisiae infections described in the literature occur after 1990. S. cerevisiae subtype boulardii accounts for 40% infections and 51% of fungemia cases. Eighty-six percent of these patients had a history of probiotic use or evidence of nosocomial acquisition in an environment where probiotics was used.

Infection is acquired through translocation of organisms from the enteral or oral mucosa and contamination of intravenous catheters. Outbreaks in ICU’s have been described. Opening of probiotic capsules for nasogastric administration can disperse viable yeasts through aerial transmission up to 1m. Yeasts persist on room surfaces for 2 hours and hands of health care workers despite hand washing.

Saccharomyces infection is clinically indistinguishable from invasive candidiasis. Fever is present in 75% of patients. Chorioretinitis and esophagitis can be present.

Other clinical syndromes reported include pneumonia, empyema, liver abscess, peritonitis, esophagitis, urinary tract infection, vaginitis, cellulitis, fever and septic shock.

Treatment of invasive S. cerevisiae infection includes discontinuing probiotics, removal of central venous catheters and administration of antifungals.

S. cerevisiae is consistently susceptible to Amphotericin B (MIC range 0.032 – 4 mg/L). Most isolates are moderately susceptible to fluconazole and itraconazole. There seems to be no difference in the rates of favourable outcome between patients treated with Amphotericin B and fluconazole. However, resistance to itraconazole and fluconazole has been described with treatment failure. Voriconazole has been effective in the patients with itraconazole and fluconazole failure. Posaconazole and caspofungin is also considered to have good activity although few isolates have been tested.

Saccharomyces infection occurs mostly in patients with a number of comorbidities and morbidity data is difficult to evaluate. Reviews of cases in the literature found a favourable outcome in 63% of patients with invasive Saccharomyces infection. Outcome did not differ significantly between immunocompromised and immunocompetent patients.

Question 5: What are the indications and evidence for use of Saccharomyces in probiotics?

Answer to Q5

Saccharomyces cerevisiae subtype boulardii is the strain mostly used in probiotics. Other Saccharomyces cerevisiae strains have only infrequently been investigated for probiotic properties. Saccharomyces boulardii was discovered by a French microbiologist, Henri Boulard in 1920 in IndoChina. He visited during a cholera outbreak and noticed that some people drinking a special tea made from cooking the outer skin of a tropical fruit did not develop cholera. He succeeded in isolating the agent responsible and named the special strain of yeast Saccharomyces boulardii.

S. boulardii was classified as a subtype of S. cerevisiae after rRNA sequencing methods reported S. boulardii indistinguishable from other strains of S. cerevisiae. Microsatelite polymorphism analysis however shows a unique clustering in S. boulardii, different from other strains of S. cerevisiae. S. boulardii also differ physiologically from other strains of S. cerevisiae in that the optimum growth temperature is 37 °C (vs 30 - 33 °C), it is resistant to low pH and tolerant to bile acids.

S. boulardii probiotic action includes a number of different mechanisms. In the gut lumen the organisms exhibit anti-toxinic effects by blocking pathogen toxin receptor sites, acting as decoy receptor and destroying pathogenic toxins through the action of proteases and phosphatases. C. difficile toxin A and B, Cholera toxin and endotoxin of pathogenic E. coli is destroyed by enzymes from S. boulardii. S. boulardii interfere with attachment of pathogenic organisms to intestinal receptor sites and enhance the integrity of tight junctions between enterocytes. S. boulardii increases colonic short chain fatty acids that are otherwise depressed during disease. S. boulardii has a trophic effect on enterocytes by releasing polyamines that favour maturation and increasing brush border disaccharidase levels. S. boulardii also regulate immune responses by increasing sIgA levels in the gut and decreasing synthesis of inflammatory cytokines.

S. boulardii is available as probiotics either as lyophilized or heat-dried powders in capsules, or as one of several strains in a probiotic mixture in capsules or liquid beverages. Most products contain 1 x 109 S. boulardii organisms/mg. Great variation in the quality of products occurs and many products contain fewer organisms than stated on the label. The concentration of probiotic in the colon is dose dependent. An oral dose of 1-2 x 1010 organisms /day leads to colonic concentrations in healthy individuals of 2 x 108 organisms /g of stool. Steady state concentrations are reached within 3 days and the organisms is usually cleared within 3-5 days after discontinuation. Colonic concentration and clearance is affected by disturbed microflora.

The efficacy of probiotics is both strain and disease specific. S. boulardii probiotics have been tested in several types of acute and chronic intestinal diseases for clinical efficacy. There is strong evidence for clinical efficacy of S.boulardii containing probiotics in the prevention of antibiotic associated diarrhoea, prevention of traveller’s diarrhoea, enteral nutrition related diarrhoea, symptoms related to H. pylori and acute infectious diarrhoea in children.

A meta-analysis of the role of S. boulardii in preventing antibiotic associated diarrhoea that included 10 randomised controlled trails (RCTs) in adults, found that S. boulardii was significantly protective with a pooled RR of 0.47. The number to treat to prevent one case was 10.2. These results were confirmed in studies in children as well.

Meta-analysis of 12 RCTs of various probiotics found a significant reduction in traveller’s diarrhoea. S. boulardii was studied in 2 RCTs and found to reduce the incidence of diarrhoea (32 vs 43% and 29 vs 39%). Probiotics appears to be more effective in preventing traveller’s diarrhoea than treating it once symptomatic.

Enteral tube feeding is commonly complicated by diarrhoea resulting in loss of nutrition. Three RCTs assessing the ability of S. boulardii to reduce diarrhoea in patients receiving enteral feeding found that fewer patients receiving S. boulardii had diarrhoea for fewer days than patients receiving placebo (8.7%, 1.5% and 7.7% vs 16.9%, 9.1% and 12.7% days of diarrhoea). No adverse effects were reported in these trails.

S. boulardii is not effective in eradicating H. pylori, but it is effective in reducing symptoms of dyspepsia, epigastric distress and side-effects of triple therapy, especially antibiotic associated diarrhoea.

A meta-analysis of 5 RCTs evaluating the efficacy of S. boulardii in treating acute infectious diarrhoea in children showed that S. boulardii significantly reduced the duration of diarrhoea compared to the control group with a mean of 1.1 days. Subsequently, S. boulardii is recommended, together with Lactobacillus GG, as an adjunct to rehydration therapy in children with acute gastroenteritis.

There are conflicting opinions regarding the role of probiotics in C. difficile diarrhoea. S. boulardii in combination with high dose vancomycin was effective in one trial in preventing recurrent C. difficile diarrhoea. Systematic reviews however, found little evidence to support routine use of probiotics in C. difficile infection. A meta-analysis of 6 RCTs however found an overall significant efficacy of probiotics in preventing recurrence of C. difficile disease.

Evidence of clinical efficacy to recommend use of S. boulardii probiotics in the treatment of acute adult diarrhoea, inflammatory bowel disease, irritable bowel syndrome, Giardiasis and chronic HIV-related diarrhoea is insufficient and further randomized controlled trials are needed.

The only adverse effects to S. boulardii probiotics described in clinical trials were constipation and thirst in patients with C. difficile diarrhoea. No cases of fungemia were described. Fungemia seemed to be a sporadic occurrence in certain patient groups and probiotics containing S. boulardii is therefore considered safe.


  1. Pfaller et al. Results from ARTEMIS DISK Global Antifungal Surveilance Study 1997 – 2007. J Clin Microbiol 2009;47(1): 117-123 and J Clin Microbiol 2010;48(4):1366-1377
  2. Miceli, MH, Diaz JA, Lee SA. Emerging opportunistic yeast infections. Lancet Infect Dis 2011;11:142-51
  3. Anaissie EJ. Clinical Mycology. 2nd Ed. Churchill-Livingstone
  4. Qazzafi Z, Thiruchunapalli D, Birkenhead D, Bell D, Sandoe JA. Invasive Cryptococcus neoformans infection in an asplenic patient. Journal of Infection, Volume 43, Issue 3, October 2001, Pages 182-186
  5. Perman J, Zaragoza R. Current diagnostic approaches to invasive Candidiasis in critical care settings. Mycoses 2009; 53:424-433
  6. Freydiere AM, Guinet R, Boiron P. Yeast identification in the clinical microbiology laboratory: phenotypical methods. Med Mycol 2001;39:9-33
  7. Budhavari S. What’s new in diagnostics? Fungitell®: 1,3 beta-D Glucan assay. South Afr J Epidemiol Infect 2009;24(1):37-38
  8. Enache-Angoulvant A, Hennequin C. Invasive Saccharomyces Infection: A comprehensive review. Clin Infect Dis 2005; 41:559-68
  9. Munoz P et al. Saccharomyces cerevisiae fungemia: An emerging infectious disease. Clin Infect Dis 2005;40:1625-34
  10. Riquelme AJ et al. SaccharomyceAs cerevisiae fungemia after Saccharomyces boulardii treatment in immunocompromised patients. J Clin Gastroenterol 2003;36(1):41-43
  11. McFarland LV. Systematic review and meta-analysis of Saccharomyces boulardii in adult patients. World J Gastroenterol 2010;16(18):2202-222
  12. Vandenplas Y, Brunser O, Szajewska H. Saccharomyces boulardii in childhood. Eur J Pediatr 2009;168:253-265
  13. Whelan K, Myers C. Safety of probiotics in patients receiving nutritional support. Am J Clin Nutr 2010;91:687-703
  14. FLoch MH et al. Recommendations for probiotic use. J Clin Gastroenterol 2008;42(Suppl 2):S104-8
  15. Pham M, Lemberg DA, Day AS. Probiotics: sorting the evidence from themyths. MJA 2008;188:304-308
Lessons Learnt

Saccharomyces is an emerging cause of invasive fungal infections. Fungemia with Saccharomyces is infrequent, but use of Saccharomyces containing probiotics may be a risk factor for invasive infection in severely ill patients in intensive care. Close follow up of these patients for unexplained episodes of fever is necessary. Saccharomyces containing probiotics should be used with caution in immunosuppressed patients and those with central venous catheters.

Earn CPD Points


FIDSSA Members can earn CPD points by logging into the secure section of the website and visiting the MyCPD section.