Pneumocystis carinii

Since Pneumocystis carinii cannot be effectively cultured in vitro, the current laboratory diagnosis relies on direct microscopic demonstration of the pathogen. This is usually performed in broncho-alveolar lavage (BAL) by histochemical or immunohistochemical staining (Montaner and Zala 1995). Wakefield et al. (1990) presented a clinical application of a PCR assay, based on sequences coding for mitochondrial rRNA (mt rRNA) of P. carinii. Several other gene targets of the P. carinii genome have been used in PCR assays (Table 34.1). In a comparative study, Lu et al. (1995) found this assay to be the most sensitive single PCR for detection of P. carinii in BAL samples. The nested PCR methods, based on other P. carinii gene targets (Lu et al. 1995), performed better, as expected, than single PCR methods with regard to sensitivity and specificity.

Although the sensitivity of histochemical stains for detection of P. carinii in BAL specimen is high, up to 98% (Lipschik et al. 1992; Tamburrini et al. 1993; Roux et al. 1994), patients with Pneumocystis carinii pneumonia (PCP) may be severely ill and unable to undergo invasive diagnostic procedures. Less invasive sampling for the diagnosis of PCP is needed. The advantages with PCR concerns these less invasive sampling methods, mainly induced sputum samples, which in general contain fewer infectious organisms (Wakefield et al. 1991; Lipschik et al. 1992; Olsson et al. 1993; Cartwright et al. 1994; Roux et al. 1994; Leibovitz et al. 1995) (Figure 34.1). A forbearing alternative sampling procedure to BAL and sputum is oropharyngeal washes, which can be performed on patients with tendencies of bleeding and respiratory failure. The sampling technique does not require special equipment and no specially trained staff are required. The method has been used in HIV-infected patients with a reported sensitivity of 78% (Wakefield et al. 1993) and in a prospective study using the mt RNA single round PCR. The latter study showed a sensitivity of 72%, a specificity of 96%, a positive predictive value of 95% and a negative predictive value of 77% in diagnosing PCP among 49 HIV infected patients (Lundgren et al. 1996). Recently, the method has proven useful in diagnosing PCP among patients with haemotological malignancies (Helweg-Larssen et al. 1997).

PCR on serum or blood samples has shown conflicting results (Lipschik et al. 1992; Schluger et al. 1992; Contini et al. 1993; Roux et al. 1994; Atzori et al. 1995; Evans et al. 1995; Tamburrini et al. 1996) (see also Table 34.2). Thus, the use of blood products as a non-invasive specimen in the diagnosis of PCP remains to be established, although in rare cases of disseminated P. carinii infection, PCR on blood samples may consistently show P. carinii DNA (Lipschik et al. 1992; Roux et al. 1994).

Table 34.1 Gene targets for diagnostic detection of P. carinii with PCR amplification

Single copy genes

Reference

Enzymes

Thymidylate synthase (TS) Dihydrofolate (DFHR) synthase Nuclear genes

Internal trancribed spacers (ITS)

18S RNA

18S RNA (Kitada et al. 1991)

Multi-copy genes

Mitochondrial LSU rRNA (mt LSU rRNA)_(Wakefield et al. 1990)

Figure 34.1 Sensitivity and specificity of PCR on induced sputum for diagnosing P. carinii pneumonia

Table 34.2 Serum/blood as non-invasive specimen for diagnosing PCP

Specimen

Target gene

Sensitivity (%)

Reference

Serum

DHFR

86

(Schluger et al. 1992)

Blood

18sRNA

10

(Lipschik et al. 1992)

PBMC

18sRNA

95

(Contini et al. 1995)

Blood

mt LSU rRNA

0

(Roux et al. 1994)

Serum

PBMC

Serum

ITS

100

(Atzori et al. 1995)

DHFR

10

Serum

mt LSU rRNA

0

(Tamburini et al. 1996)

PBMC

7

PMNC

33

The clinical significance of PCR for detection of P. carinii has been questioned since positive findings of P. carinii in immunosuppressed patients not always progressed to PCP (Leigh et al. 1992; Lipschik et al. 1992; Tamburrini et al. 1993; Weig et al. 1997). This has been interpreted as temporarly colonization or subclinical infection of P. carinii in susceptible individuals. On the contrary, Elvin et al. (1996) showed a predictive value of PCR in a study, where asymptomatic HIV infected patients without primary prophylaxis were followed during 3 years. Similar results have also been shown by others, mainly in HIV positive patients (Wakefield et al. 1991; Leigh et al. 1992, 1993; Lipschik et al. 1992).

Figure 34.1 Sensitivity and specificity of PCR on induced sputum for diagnosing P. carinii pneumonia

In summary, laboratory diagnosis of P. carinii still relies on morphological identification of the organism in specimens obtained from the lung, although PCR is recommended as a diagnostic complement on negative stained induced sputum or other less invasive samples.

Page 336

0 0

Post a comment