Rheumatology 2004;43:479–485 Advance Access publication 3 February 2004
doi:10.1093/rheumatology/keh071
Detection of Pneumocystis carinii by DNA amplification in patients with connective tissue diseases: re-evaluation of clinical features of P. carinii pneumonia in rheumatic diseases K. Saito, S. Nakayamada, K. Nakano, M. Tokunaga, S. Tsujimura, K. Nakatsuka, T. Adachi and Y. Tanaka Objectives. We evaluated the polymerase chain reaction (PCR) detection of Pneumocystis carinii DNA in induced sputum of patients with connective tissue diseases and assessed the clinical features of patients positive for P. carinii. Methods. Sputum was induced by nebulization in 29 in-patients with various connective tissue diseases who presented with symptoms suggestive of P. carinii pneumonia (PCP), and was examined by PCR. Results. Detection of P. carinii DNA by PCR was significantly more sensitive than cytology; 54.5% patients were positive by PCR and only 4.5% by cytology. The prevalence of PCP was higher than previously considered and was especially high in patients receiving >30 mg/day prednisolone with or without other immunosuppressants. P. carinii-positive patients had significant lymphocytopenia and a low serum IgG level compared with P. carinii-negative patients. P. carinii disappeared within 7–10 days after therapy with trimethoprim/sulfamethoxazole. Conclusion. We propose that the use of PCR for detection of P. carinii using induced sputum is a useful and non-invasive method that has high sensitivity and specificity for the early diagnosis of PCP. KEY WORDS: Pneumocystis carinii pneumonia, Polymerase chain reaction, Rheumatic diseases, Steroids, Immunosuppressants, -D-glucan, Trimethoprim/sulfamethoxazole.
Pneumocystis carinii pneumonia (PCP) is one of the most predominant opportunistic infectious diseases in exclusively immunocompromised patients [1]. Host defence against P. carinii depends mainly on cellular immunity, and the depletion of CD4 T lymphocytes that is observed in patients with acquired immunodeficiency syndrome (AIDS) results in fatal pneumonia. More recently, an increasing number of reports have described the occurrence of PCP in patients with connective tissue diseases such as rheumatoid arthritis (RA), polymyositis (PM), systemic lupus erythematosus (SLE) and Wegener granulomatosis (WG) [2]. Several factors seem to be involved in the manifestation of PCP in patients with connective tissue diseases. These include impaired humoral and cellular immunity, which has been reported in autoimmune diseases, such as SLE, both by our laboratories and other investigators [3–5]. Furthermore, the use of immunosuppressive agents for the treatment of connective tissue diseases is closely related to the occurrence of PCP. At present, immunosuppressive agents may be used more commonly, in higher doses or in combinations, which may thus increase the susceptibility of patients to PCP. P. carinii infection in immunocompromised hosts sometimes results in rapid and fatal pneumonia, and thus early and precise diagnosis is not easy. PCP patients usually present with nonproductive cough and P. carinii is considered to adhere tightly to the alveolar wall with its pseudopodia. It is therefore difficult to obtain high-quality sputum for precise diagnosis. However, cytological examination of samples from the respiratory tract has been the mainstay of the laboratory diagnosis of PCP. P. carinii
organisms have not yet been successfully cultured, and diagnostic identification of the organism depends on direct microscopy after Giemsa or silver staining of bronchoscope lavage material or induced sputum [6]. The development of assays that use monoclonal antibodies against P. carinii has improved specificity but the sensitivity remains uncertain [7]. The majority of studies in this area have been conducted in the context of HIV-induced immunosuppression, but the clinical presentation of PCP in patients with connective tissue diseases who are on immunosuppressants is different. In particular, disease is more rapid in onset and is often more severe, and there are generally far fewer organisms in induced sputum or bronchoalveolar lavage fluid [8]. Thus, there is a need for a diagnostic method that has high sensitivity and specificity and which is clinically applicable for the early diagnosis of PCP. Recently, oligonucleotide primers and probes were used in a polymerase chain reaction (PCR) method designed to amplify P. carinii-specific DNA sequences from alveolar lavage samples and induced sputum. PCR is a highly sensitive and specific technique compared with conventional microscopic examination for the detection of P. carinii [9]. To our knowledge, however, there are no studies that have used PCR for the diagnosis of PCP in patients with connective tissue diseases, and to date the features of PCP have been evaluated mainly by conventional microscopic examination, which might result in the underevaluation of PCP or delay in its treatment. In the present study, we used PCR to detect DNA of P. carinii in patients with connective tissue diseases suspected of having PCP
First Department of Internal Medicine, University of Occupational and Environmental Health, School of Medicine, Fukuoka, Japan. Submitted 29 May 2003; revised version accepted 13 October 2003. Correspondence to: Y. Tanaka, First Department of Internal Medicine, University of Occupational & Environmental Health, School of Medicine, 1-1 Iseigaoka, Yahatanishi, Kitakyushu 807-8555, Japan. E-mail:
[email protected] 479 Rheumatology Vol. 43 No. 4 ß British Society for Rheumatology 2004; all rights reserved
480
K. Saito et al.
based on the presence of progressive arterial hypoxaemia and interstitial pneumonitis on plain chest X-ray or computed tomography. Then we evaluated the clinical features of PCP in these patients. This is the first report that provides clinical assessment of patients with PCP complicating connective tissue diseases, and evaluates the risk factors for PCP, such as the daily dose of prednisolone, laboratory data, the clinical course and the prognosis after treatment based on DNA diagnosis using induced sputum.
Statistical analysis Data are expressed as mean S.D. for the number of patients indicated. Differences between pairs of groups were examined for statistical significance with the Mann–Whitney U-test. The correlation coefficient was determined with Fisher’s exact probability test. A P-value less than 0.05 denoted a statistically significant difference.
Results Patients and methods
PCR analysis of P. carinii in connective tissue diseases
Clinical specimens
PCR was performed for detection of P. carinii in induced sputum samples from 29 patients with connective tissue diseases. Table 1 summarizes the clinical characteristics of patients with suspected diagnosis of PCP, including diagnosis, recent involvement within the preceding 4 weeks and current treatment. Almost all patients with connective tissue diseases, except those with RA, were being treated with more than 25 mg/day of prednisolone with or without other immunosuppressants. All patients presented with suspicious clinical signs of PCP, such as non-productive cough, shortness of breath and progressive hypoxia, as well as ground glass appearance on plain chest X-ray or CT. They had no previous manifestation of PCP and no prophylactic treatment for PCP. DNA was extracted from the sputum obtained after nebulization with distilled water, and PCR using specific oligonucleotides for P. carinii was carried out. The results showed a 124 bp single band in 3% agarose gel containing ethidium bromide (Fig. 1). Based on the results of PCR, 19 out of 29 patients (65.5%) were confirmed to have PCP. Among the connective tissue diseases, five out of seven patients with SLE, five of five with PM/DM, two of three with MCTD, four of nine with RA, two of four with MPA, and one of one with SS were PCR-positive for P. carinii. However, 10 sputum samples were also examined from patients with idiopathic interstitial pneumonitis and healthy controls, all of whom were negative for P. carinii.
Sputum samples were obtained from hospitalized patients with connective tissue diseases who presented with a provisional diagnosis of PCP between September 1998 and February 2003 at the University Hospital of Occupational and Environmental Health. PCP was suspected in these patients because of the presence of all of the following parameters: (i) typical radiographic features of the lungs on plain chest X-ray or computed tomography (CT); and (ii) typical clinical signs and symptoms, such as non-productive cough, shortness of breath, and progressive hypoxia. In almost all patients, sputum was induced by nebulization with 10 ml of distilled water and then collected for examination. Then Grocott staining was performed as described in a previous paper [6]. Samples were obtained from 29 patients, comprising seven with SLE, five with PM/dermatomyositis (DM), nine with rheumatoid arthritis (RA), four with microscopic polyangiitis (MPA), three with mixed connective tissue diseases (MCTD) and one with Sjo¨gren syndrome (SS). The samples were examined for P. carinii DNA by PCR. Each connective tissue disease was diagnosed based on previously proposed classification criteria for RA [10], SLE [11], PM/DM [12], SS [13], MCTD [14] and MPA [15]. Ten sputum samples from patients with idiopathic interstitial pneumonitis without connective tissue diseases were also examined. The study protocol was approved by the Human Ethics Review Committee of University of Occupational and Environmental Health, School of Medicine, and a signed consent form was obtained from each subject.
DNA amplification and detection of P. carinii The sputum sample was diluted with distilled water (1:1), mixed with dithiothreitol to a final concentration of 5 mM and then vortexed vigorously. After 10 min incubation at 37 C, samples were centrifuged and pellets were washed three times with to phosphate-buffered saline (PBS). Then the pallets were resuspended in 1 ml of PBS, 50 l of lysis buffer (10 mM Tris–HCl pH 8.4, 50 mM KCl, 1.5 mM MgCl2, 0.1 mg/ml of gelatin, 0.25% Tween-20, and 0.25% NP-40) and 5 l of 20 mg/ml of proteinase K (Gibco BRL, MD, USA). The mixture was incubated for 6 h at 60 C and DNA was extracted with phenol/chloroform followed by ethanol precipitation. Finally DNA was dissolved in 20 l of TE buffer, pH 8.0, and 5 l of DNA was used as a template. Oligonucleotide primers (sense, 50 -AGTTACGGCCATACCTCAGA-30 ; antisense, 50 -AAAGCTACAGCACGTCGTAT-30 ) were used at 100 pmol in 50 l of amplification reaction mixture, with denaturation at 94 C for 1 min, annealing at 50 C for 1 min and extension at 72 C for 90 s for 35 cycles using an iCycler Thermal cycler (Bio-Rad, Richmond, CA, USA) [16]. Negative controls with no added template and positive control DNA from a sample of definite PCP were included. The products (20 l) were subjected to electrophoresis in 3% agarose gels and the presence of 124 base pair (bp) specific bands was confirmed.
Comparison of PCR with Grocott staining detection for P. carinii using the same specimens Both PCR and conventional microscopic examination by Grocott staining were simultaneously performed on 22 samples. As shown in Table 2, positive results were noted in 54.5% of the samples by PCR and in only 4.5% by cytology. Eleven samples negative by cytology were positive by PCR. In contrast, none of the samples that were positive by cytology were PCR-negative. Detection of P. carinii DNA by PCR was significantly more sensitive than cytology.
Clinical background of PCR-positive patients with P. carinii Because PCP is one of the most well-known opportunistic infections, we evaluated the relation between the occurrence of PCP and immunosuppressive agents used by our patients. As shown in Figure 2, the oral dose of steroids, converted into prednisolone, correlated closely with PCP. The mean dose of prednisolone was significantly higher in patients with PCP (48.7 mg/day) than in PCP-negative patients (15.6 mg/day). The frequency of PCP was markedly higher in patients who were taking >30 mg prednisolone daily (16 out of 19; 84.2%). Furthermore, 15 out of 20 patients (excluding those with RA) were also receiving oral or intravenous (i.v.) cyclophosphamide (CY), azathioprine (AZA) or cyclosporin A (CSA) because of disease severity. Even in patients on
