Case of the Month

February 2014

Frans Radebe, NICD/NHLS

Krishnee Moodley, Chetna Govind and Nelesh P. Govender Adapted from a published case report (open-access publication under a creative commons license): Moodley K, et al. S Afr J HIV Med 2013;14:138-140.

Case History

A 46-year-old HIV-infected man on antiretroviral treatment was admitted to hospital with a history of fever and malaise. The patient had an undetectable HIV-1 viral load on admission with a CD4+ T-lymphocyte count of 435 cells/µl. He was also on haemodialysis for chronic renal disease (secondary to hypertension and diabetes mellitus), and had a central venous catheter (CVC) in situ. On examination, he was febrile and had evidence of severe mitral valve regurgitation with bilateral basal crackles. His admission total leucocyte count was elevated, i.e. 11.4 x 109 cells/l (reference range: 4.0 – 10.0 x 109 cells/l).

Question 1: What is your differential diagnosis and what investigations would you perform?

Answer to Q1

The differential diagnosis should include:

  1. Infective endocarditis (IE) – fever, mitral regurgitation, crackles suggestive of pulmonary oedema, indwelling central venous catheter as the possible source of infection
  2. Nosocomial catheter-related sepsis – fever, chronic haemodialysis with indwelling central venous catheter as the possible source of infection
  3. Community-acquired pneumonia (including tuberculosis) – HIV-infected, fever, pulmonary crackles, underlying diabetes mellitus
  4. Fluid overload - crackles suggestive of pulmonary oedema

However, the diagnosis of IE fits best with the clinical history, examination and initial investigations. The aetiological causes of IE are numerous but broadly include bacterial and fungal pathogens. In order to confirm a diagnosis of IE, blood cultures and echocardiography should be performed immediately. Three sets of blood cultures should be obtained, preferably from different venepuncture sites and at different times (spaced at least 30 minutes apart).[1] This allows documentation of a continuous (rather than transient or intermittent) bloodstream infection which improves the specificity of an IE diagnosis. For adults, each blood culture bottle should contain at least 10 ml of blood to increase the yield. Prolonged incubation of blood cultures may be required for detection of fastidious or slow-growing pathogens (several weeks in some cases). In addition, a set of blood cultures obtained through the CVC could provide useful information. To minimise the chance of contamination of blood cultures, it is important to ensure proper skin antisepsis is performed before venepuncture; for blood obtained through the CVC, the port should first be disinfected with 70% alcohol. Other baseline investigations that would provide useful supporting evidence include a chest radiograph, full blood count with differential, C-reactive protein and serum 1,3 β-d glucan assay (this is helpful if a fungal cause of IE is likely because of the low sensitivity of blood cultures). The patient’s renal function would also need to be carefully monitored.

A single set of blood cultures cultured Candida parapsilosis.

The 1,3 β-d glucan levels were also markedly elevated (>523 pg/ml). Based on the pathogen identity and due to his impaired renal function, the patient was empirically started on intravenous fluconazole. On echocardiography, large, friable mitral valve vegetations were observed consistent with a diagnosis of fungal infective endocarditis (Figure 1). Blood cultures submitted 48 hours after commencing fluconazole, cultured the same organism.

Figure 1: Echocardiography showing mitral valve vegetation (Moodley K, et al. S Afr J HIV Med 2013;14:138-140.)

Question 2: What are possible explanations for persistence of the organism in the bloodstream?

Answer to Q2

  1. The source of the infection has not been definitively dealt with. Fungal IE requires prolonged antifungal treatment, preferably with a fungicidal agent to which the pathogen is susceptible, and urgent surgical intervention, i.e. within a week of diagnosis. Based on retrospective studies, either amphotericin B or an echinocandin combined with flucytosine is recommended for Candida IE. Flucytosine is not registered in South Africa and can only be used if an application for compassionate use is submitted to the Medicines Control Council. Individual case reports have demonstrated that successful medical treatment of Candida IE is possible using an echinocandin without surgical intervention [5, 6] but there is insufficient experience with these agents to make a recommendation for medical management alone. It is also important to review the patient for evidence of Candida endophthalmitis.
  2. The original source of infection in this patient was most likely the haemodialysis CVC; this needs to be removed or changed, if possible.
  3. Fluconazole is a fungistatic agent (i.e. inhibits growth but does not kill) which is not ideal for treatment of potentially life-threatening fungal IE.
  4. The organism may be resistant to fluconazole. It is important for the laboratory to generate minimum inhibitory concentrations (MICs) for antifungal agents and not report the susceptibility category for each agent only.
  5. The dose/levels of fluconazole may be inadequate to achieve a fungistatic concentration at the site of infection.
Case History

Low MICs were documented for this isolate and all tested antifungal agents including amphotericin B (no breakpoints), fluconazole (susceptible), voriconazole (susceptible) and caspofungin (susceptible). The haemodialysis CVC was removed and replaced. Culture of the CVC tip revealed C. parapsilosis with susceptibility results corresponding exactly to those of the blood culture isolate. The patient also underwent urgent mitral valve resection and replacement; excised valve tissue was submitted for microbiological and histological evaluation. An isolate identical to the admission blood culture and CVC tip culture was cultured from the valve tissue. Medical therapy with fluconazole was continued for six weeks. Blood cultures performed two weeks post-surgery were negative. However, repeat assays showed elevated 1,3 β-d glucan levels four weeks after therapy.

Question 3: What is the significance of a persistently elevated 1,3 β-d glucan assay in this clinical scenario?

Answer to Q3

1,3 β-d glucan, a cell wall component of many fungi, is detected by the Fungitell assay (Associates of Cape Cod Clinical Diagnostics, MA). The assay assists with early diagnosis of fungal bloodstream infections as it is often positive before traditional blood cultures. In one study, the 1,3 β-d glucan assay levels were found to be significantly higher in patients with C. parapsilosis, C. tropicalis and C. guilliermondii IE compared to those with Candida albicans endocarditis.[2]. In a meta-analysis published in 2011, the pooled sensitivity and specificity for diagnosis was 77% and 88% respectively.[3] This assay is also positive in patients with Pneumocystis pneumonia (PCP) but negative in patients with mucormycosis or cryptococcosis. β-D-glucan assay levels may be falsely elevated in patients with bacteraemia and in those on intravenous antibiotics (such as amoxicillin-clavulanic acid), immunoglobulin and haemodialysis where cellulose membranes have been utilized.[4]

The elevated follow-up 1,3 β-d glucan level in this patient was probably not clinically significant:

  1. Although the patient is HIV-infected, PCP was ruled out both clinically and radiologically.
  2. Dialysis was performed using synthetic polysulfone membranes which do not significantly elevate levels of 1,3 β-d glucan.
  3. Elevated levels post-commencement of treatment have been described previously: high levels which are observed during the early phase of the disease do not always return to baseline during the early stages of antifungal therapy.[2, 3]
  4. An important consideration in this patient is a persistent nidus of infection as he was treated with a fungistatic agent. However, blood cultures remained negative following surgical removal of the infected valve.

Question 4: What is the long-term management of this patient?

Answer to Q4

In this case, fluconazole was selected empirically and then continued, based on antimicrobial susceptibility results. Early aggressive surgical therapy combined with medical treatment resulted in a good outcome for the patient. Although combined medical and surgical therapy has decreased the mortality of patients with Candida IE, the mortality remains high. Long-term oral fluconazole therapy may be continued following intravenous therapy; however the duration of suppressive therapy is not clearly defined, with some patients requiring life-long therapy.[5]


In this patient, oral suppressive fluconazole therapy was continued. Two months following diagnosis, the patient remained clinically stable and continued with haemodialysis, fluconazole suppressive therapy and antiretroviral therapy.


  1. Ntusi N, A.L., Oliver S, Whitelaw A, Mendelson M, Guideline for the optimal use of blood cultures. S Afr Med J, 2010. 100(12): p. 839-43.
  2. Lefort, A., et al., Diagnosis, management and outcome of Candida endocarditis. Clin Microbiol Infect, 2012. 18(4): p. E99-E109.
  3. Mikulska, M., et al., Persistence of a positive (1,3)-beta-D-glucan test after clearance of candidemia in hematopoietic stem cell transplant recipients. Clin Vaccine Immunol, 2011. 18(3): p. 518-9.
  4. Kato, A., et al., Elevation of blood (1-->3)-beta-D-glucan concentrations in hemodialysis patients. Nephron, 2001. 89(1): p. 15-9.
  5. Gould, F.K., et al., Guidelines for the diagnosis and antibiotic treatment of endocarditis in adults: a report of the Working Party of the British Society for Antimicrobial Chemotherapy. J Antimicrob Chemother, 2012. 67(2): p. 269-89.

Earn CPD Points


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