SAN FRANCISCO — A 60-second oropharyngeal saline wash coupled with a quantitative polymerase chain reaction assay is 88% sensitive and 85% specific in the diagnosis of Pneumocystis pneumonia, Laurence Huang, M.D., reported at a meeting on HIV management sponsored by the University of California, San Francisco.
In the blinded study, investigators compared quantitative touch-down polymerase chain reaction (QTD PCR) results with bronchoscopy with bronchoalveolar lavage (BAL), which is widely considered to be the standard for the diagnosis of Pneumocystis pneumonia (PCP).
A total of 113 cases of PCP among 108 patients were included in the study (J. Infect. Dis. 2004;189:1679–83).
The use of QTD PCR testing is a way to increase the specificity of the DNA amplification procedure.
In recent years, the sensitivity of bronchoscopy with BAL has proved to be at least 98%, with a specificity of 100%, said Dr. Huang of UCSF and San Francisco General Hospital. The problem is that bronchoscopy with BAL is a relatively lengthy, uncomfortable, and invasive procedure.
Sputum induction takes only 30 minutes, and studies early in the HIV epidemic pegged its sensitivity at somewhere between 74% and 83%. But for unknown reasons, the sensitivity of sputum induction has been decreasing, reaching a low of 57% among 63 cases at San Francisco General in 2003 and 2004. Its specificity remains close to 100%, however.
In the 60-second gargle test, the collection procedure is “as low tech as you can get,” Dr. Huang said, requiring only a sterile specimen cup and 10 mL of sterile saline. It's therefore suitable for use in remote areas or resource-limited settings such as those in developing countries.
The low-tech collection procedure is coupled with a very high-tech detection procedure, namely PCR amplification, that's capable of spotting the DNA from a single organism of Pneumocystis jiroveci.
The test's sensitivity increased to 92% in patients who had received empiric treatment for PCP for 1 day or less, but its specificity declined to 75%.
Compared with the 100% specificity of bronchoscopy with BAL and sputum induction, the 75%–85% specificity of the gargle test is its main drawback. Dr. Huang suggested that people without PCP who nevertheless test positive by the gargle test may be nonsymptomatic carriers who are colonized with P. jiroveci.
He noted that, according to some accounts, as many as 40% of health workers who are exposed to HIV-infected patients may be colonized with the organism.
Dr. Huang declared that he has no financial relationship with a commercial entity.