Jury still out on β-glucan diagnosis of pneumonia
By Brian G. Blackburn, MD, Clinical Assistant Professor of Medicine, Division of Infectious Diseases and Geographic Medicine, Stanford University School of Medicine
Synopsis: Among a cohort of AIDS patients with opportunistic infections, the sensitivity of plasma β-glucan for the diagnosis of Pneumocystis jirovecii pneumonia (PCP) was 92% and the specificity was 65%. Although an intriguing alternative, β-glucan did not perform well enough in this study to supplant sputum/BAL examination as the primary laboratory means of diagnosing PCP.
Source: Sax PE, et al. Blood (1->3)-β-D-glucan as a diagnostic test for HIV-related Pneumocystis jirovecii pneumonia. Clin Infect Dis 2011;53:197-202.
Pneumocystis jirovecii pneumonia (PCP) is a common opportunistic infection (OI) in AIDS patients. Laboratory diagnosis of this life-threatening infection is based primarily upon identifying P. jirovecii cysts in respiratory secretions, a technique that is variably sensitive and requires adequate patient effort (for induced sputum examination) or an invasive test (bronchoscopy); both are quite operator dependent.1
β-glucan is a component of the cell wall of many fungi, including P. jirovecii. Earlier studies have suggested that measurement of β-glucan in the blood may have utility in the diagnosis of PCP.2,3 The authors therefore undertook a study to evaluate the usefulness of this assay for diagnosing PCP in a cohort of AIDS patients with various OIs.
HIV-infected patients were recruited on the basis of having a suspected acute OI; patients with tuberculosis and some other OIs were excluded. Plasma samples from each patient were tested by the Fungitell® β-glucan assay (Associates of Cape Cod). The "gold standard" for the diagnosis of PCP in the study was a combination of clinical, radiologic, and laboratory parameters, as adjudicated by the study investigators. Both probable and confirmed cases were included in the study, and for confirmed cases, the case definition included direct observation of Pneumocystis in respiratory secretions.
Two hundred fifty-two persons with a valid β-glucan result were included in the study. Their median CD4+ count was 26 cells/µL; 69% had PCP, 14% had cryptococcosis, 9% had bacterial pneumonia, 6% had a Mycobacterial infection, and 3% had histoplasmosis. Although by itself not an inclusionary criterion, 44% had oral/esophageal candidiasis in addition to another OI.
Median β-glucan levels were significantly higher in patients with PCP than in those without PCP (408 vs. 37 pg/mL; P < 0.001). Using a cutoff of 80 pg/mL, significantly more patients with PCP had a positive β-glucan result than those without PCP (92% vs. 35%; P < 0.001). Conversely, significantly fewer patients with PCP had a negative β-glucan result than those without PCP (8% vs. 65%; P < 0.001). Detection of β-glucan was not affected by antimicrobial treatment and did not correlate with disease severity (as measured by use of concomitant corticosteroids). The sensitivity of the test was 92%, specificity 65%, positive predictive value (PPV) 85%, and negative predictive value (NPV) 80%. β-glucan is not specific for Pneumocystis, and many patients with oral/esophageal candidiasis and histoplasmosis had positive β-glucan results.
At first glance, this report of the association between high β-glucan levels and AIDS-related PCP might seem a major step forward in the diagnosis of this infectious disease, particularly given the limitations of the induced sputum examination. Indeed, having a standardized, reliable serologic test would greatly simplify the diagnosis of PCP. However, several methodological problems hinder the interpretation of the study results. Given that only HIV-infected patients were enrolled, the results of this study are not generalizable to other immunocompromised populations. In addition, the lack of a true gold standard for the diagnosis of PCP means that interpretation of the sensitivity and specificity data in this study are subject to uncertainty. Data are not provided regarding the relative proportion of probable vs. confirmed cases, and the performance of β-glucan was assessed based upon a case definition that included only clinical and radiologic criteria for probable cases. Even among laboratory-confirmed cases (for which a positive direct examination of respiratory secretions was required), use of the very test against which β-glucan will be considered in the clinical setting as part of the gold standard definition is problematic.
Because β-glucan is a component of the cell wall of many fungi, it is inherently nonspecific for PCP. As the authors note, other fungi, hemodialysis, intravenous immunoglobulin, and even certain antimicrobials can cause false-positive results. Many patients in the study who had PCP also had candidiasis, further confounding interpretation of the results. Despite a specificity of only 65%, the PPV was a surprisingly high 85% (in part because of the high prevalence of PCP in the study population). Unfortunately, this PPV is still suboptimal for clinical practice, and may actually be lower in many real-world settings. Therefore, the true value of β-glucan testing may be instead to rule out PCP when negative, analogous to the use of other nonspecific (but highly sensitive) tests such as the D-dimer for pulmonary embolism, or the sedimentation rate and C-reactive protein for bone and joint infections.4,5 Unfortunately, even with the relatively high sensitivity of 92%, the NPV of β-glucan was only 80% in this study, insufficiently low for this purpose. Perhaps adjusting the positive/negative cutoff to a lower β-glucan value, which would increase sensitivity (and thus NPV) at the expense of specificity, would maximize the use of β-glucan in this manner, rendering the diagnosis unlikely when negative, but leaving a positive result to be of less certain value.
At this point, PCP remains largely a clinical diagnosis, supported by direct visualization of respiratory secretions. β-glucan may have an emerging, supportive role, but at this point does not appear to be the Holy Grail we might have hoped for in the diagnosis of PCP.
- Krajicek BJ, et al. Pneumocystis pneumonia: Current concepts in pathogenesis, diagnosis, and treatment. Clin Chest Med 2009;30:265-278.
- Desmet S, et al. Serum (1-3)-β-D-glucan as a tool for diagnosis of Pneumocystis jirovecii pneumonia in patients with human immunodeficiency virus infection or hematological malignancy. J Clin Microbiol2009;47:3871-3874.
- Watanabe T, et al. Serum (1->3)-β-D-glucan as a noninvasive adjunct marker for the diagnosis of Pneumocystis pneumonia in patients with AIDS. Clin Infect Dis 2009;49:1128-1131.
- Agnelli G, et al. Acute pulmonary embolism. N Engl J Med 2010;363:266-274.
- Spangehl MJ, et al. Prospective analysis of preoperative and intraoperative investigations for the diagnosis of infection at the sites of two hundred and two revision total hip arthroplasties. J Bone Joint Surg Am1999;81:672-683.