JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 1998, p. 2346–2348 0095-1137/98/$04.0010 Copyright © 1998, American Society for Microbiology. All Rights Reserved.
Vol. 36, No. 8
Enzyme Immunoassay Detecting Teichoic and Lipoteichoic Acids versus Cerebrospinal Fluid Culture and Latex Agglutination for Diagnosis of Streptococcus pneumoniae Meningitis KRISTIN STUERTZ,1 IMKE MERX,1 HELMUT EIFFERT,2 ERICH SCHMUTZHARD,3 ¨ DER,1 AND ROLAND NAU1* MICHAEL MA ¨ttingen, Go ¨ttingen, Germany, Departments of Neurology1 and Medical Microbiology,2 University of Go and Department of Neurology, University of Innsbruck, Innsbruck, Austria3 Received 1 October 1997/Returned for modification 9 January 1998/Accepted 5 May 1998
A newly developed enzyme immunoassay (EIA) was used to detect the presence of pneumococcal teichoic and lipoteichoic acids in cerebrospinal fluid (CSF) from patients with Streptococcus pneumoniae meningitis who were being treated with antibiotics. All initial CSF samples, which on culture grew S. pneumoniae, were positive in the EIA. A total of 14 subsequent culture-negative samples gave clear signals in the EIA up to day 15 after the onset of antibiotic treatment. For 11 CSF specimens, culture, microscopy, and latex agglutination were negative while the EIA detected pneumococcal antigens. The EIA did not react either with CSF of patients with meningitis caused by bacteria other than S. pneumoniae or by viral pathogens. In conclusion, this EIA can be a valuable tool for the diagnosis of S. pneumoniae meningitis from CSF samples in cases in which prior antimicrobial therapy minimizes the usefulness of culture or other antigen detection tests. this study. Prior to admission, three patients received oral antibiotic treatment (patient B, norfloxacin; patient C, ciprofloxacin; and patient G, cefixime), and patient J received two 2-g doses of amoxicillin and two 0.5-g doses of clavulanate prior to lumbar puncture. In the other patients, intravenous antibiotic therapy was initiated immediately after lumbar puncture. For empiric therapy of meningitis, the patients were treated as follows: patients A and B received intravenous amoxicillin; patients C, D, E, G, and H were administered amoxicillin plus cefotaxime; patient F was treated with penicillin G and ceftriaxone; patient I was given amoxicillin and clavulanate; and patient J received penicillin G alone. The repeat lumbar punctures were considered necessary by the physicians in charge of the patients. The authors of this study were not involved in the timing of repeat lumbar punctures. Patient G had an external ventriculostomy for treatment of acute occlusive hydrocephalus. After withdrawal, the CSF was cultured, CSF leukocytes were counted, and cytocentrifuge preparations were examined microscopically. The CSF was then centrifuged for 10 min at 3,000 3 g, and the supernatant was stored at 4°C for approximately 1 day to allow the completion of routine analyses. Thereafter, CSF which was not required for routine analyses was frozen at 270°C. EIA. LTA was prepared from S. pneumoniae R6 (1). Polyclonal antibodies were raised in two New Zealand White rabbits immunized subcutaneously with 500 mg of LTA mixed with an equal volume of incomplete Freund’s adjuvant. Immunization was repeated every 4 weeks until high titers ($1:32,000) were obtained. The sera were preserved at 220°C. The EIA used the commercially available monoclonal immunoglobulin A (IgA) antiphosphorylcholine antibody TEPC-15 (Sigma, Deisenhofen, Germany) as the capture antibody and the polyclonal antiserum raised against LTA as the detector antibody. Ninety-six-well microtiter plates were coated with TEPC-15 monoclonal antibody at a concentration of 5 mg/ml. Blocking, as well as the subsequent steps, was carried out with 5% fetal
The management of patients suspected of having bacterial meningitis depends on their state of consciousness. Lumbar puncture is performed immediately after admission if the patient is awake. Antibiotic treatment is started directly thereafter. Patients presenting with coma or focal neurological deficits may have severe edema or mass lesions. In this situation, lumbar puncture may be harmful because of the potential for cerebral herniation. For this reason, lumbar puncture is delayed for these patients. Instead, antibiotic treatment is started immediately after samples are taken for blood cultures (10). Cranial computed tomography is then performed. Cerebrospinal fluid (CSF) is not obtained by lumbar puncture if the cranial computed tomograph shows severe brain swelling or mass lesions. In these cases, CSF is not available for microbiological cultivation prior to antibiotic treatment, and hence a sensitive test for the diagnosis of Streptococcus pneumoniae meningitis after pretreatment of the patient with antibiotics for several days is highly desirable. We therefore investigated the specificity and sensitivity of a newly established enzyme immunoassay (EIA) that recognizes the phosphorylcholine moieties of pneumococcal teichoic acids (TA) and lipoteichoic acids (LTA) in patients with cultureproven cases of S. pneumoniae meningitis. The results were compared with those obtained by microbiological cultivation, CSF Gram stain, and a latex agglutination assay. (The data herein were presented in part as a poster (no. D-160) at the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Ontario, Canada, 28 September to 1 October 1997.) Clinical data. Ten unselected patients treated at our institutions (Departments of Neurology, Universities of Go ¨ttingen and Innsbruck) for S. pneumoniae meningitis were included in * Corresponding author. Mailing address: Dept. of Neurology, University of Go ¨ttingen, Robert-Koch-Str. 40, D-37075 Go ¨ttingen, Germany. Phone: 49-551-398455 or 396684. Fax: 49-551-398405. E-mail:
[email protected]. 2346
VOL. 36, 1998
calf serum in phosphate-buffered saline at room temperature. CSF samples diluted 1:2 or higher in blocking buffer were incubated for 2 h; this was followed by a 2-h incubation with the polyclonal antiserum (1:4,000). Bound rabbit antibodies were detected with peroxidase-conjugated goat anti-rabbit IgG antibodies (1:8,000; Dianova, Hamburg, Germany). Enzyme activity was determined with a 1-mg/ml ABTS solution [2,29azino-di(3-ethylbenzthiazoline sulfonate) in 50 mM phosphate-citrate buffer (pH 4.4) containing 3 mM sodium perborate]. The absorbance at 405 nm was determined after 30 min of incubation. The assay was calibrated with a standard LTA preparation (1) at concentrations ranging from 0.8 to 50 ng/ml and did not differentiate between TA and LTA. The quantification limit of the assay for LTA-spiked CSF from patients without evidence of bacterial infection was 3.1 ng/ml. At room temperature and at 4°C, levels of LTA in the spiked CSF were stable for 14 days. After 70 days at 4°C, the LTA concentrations in these CSF samples still exceeded 50% of the initial levels. No reduction of the LTA concentrations in CSF was observed after several months of storage at 220°C. Routine CSF analysis. Standard procedures were used for Gram staining, cultivation of microbes from CSF and blood, and typing of bacteria. The commercially available S. pneumoniae-specific Wellcogen latex agglutination test (Murex Diagnostika GmbH, Burgwedel, Germany) was performed as recommended by the manufacturer. In Table 1, the results of microbial cultivation, CSF Gram stain, and detection of pneumococcal antigens, by EIA and latex agglutination, in the CSFs of 10 patients with cultureproven S. pneumoniae meningitis are shown. For the majority of the patients, the first lumbar puncture was performed before the initiation of antibiotic treatment, and S. pneumoniae could be cultured from CSF in all of these patients and from blood in five cases. In 9 of 10 patients, the second CSF specimen obtained was sterile, although gram-positive cocci were still visible in the CSFs of 3 patients up to the 4th day of antibiotic treatment. The CSF culture for patient J still was positive for S. pneumoniae 1 day after initiation of treatment. The samples obtained on all subsequent days were culture negative. On the first day, the latex agglutination assay detected pneumococcal antigen in 8 of the 10 culture-positive CSF specimens. Agglutination of the latex particles was observed in the sterile CSFs of two patients (E and F) from the 3rd day of antibiotic therapy. The latex agglutination assay was negative for three culture-positive CSF samples from two patients (C and J). All initial CSF samples, which on culture grew S. pneumoniae, were positive in the EIA. A total of 14 subsequent culture-negative CSF samples from seven patients gave clear signals in the EIA up to day 15 after the onset of antibiotic treatment. In 11 CSF specimens drawn during antibiotic therapy, all of which were negative by culture, microscopy, and latex agglutination, the EIA detected pneumococcal antigens. The time course of LTA and TA concentrations in the CSFs of two patients who underwent repeated lumbar puncture is shown in Fig. 1. Control CSF samples from patients with viral meningitis (n 5 6) or bacterial meningitis caused by Borrelia burgdorferi (n 5 3), Corynebacterium spp. (n 5 1), Enterococcus faecalis (n 5 1), Listeria monocytogenes (n 5 3), Mycobacterium tuberculosis (n 5 2), Neisseria meningitidis (n 5 1), Staphylococcus aureus (n 5 2), or Streptococcus agalactiae (n 5 1) did not cross-react in the EIA. Similarly, no cross-reactions were observed with supernatants from cultures of Corynebacterium xerosis, Enterobacter cloacae, Enterobacter sakazakii, Enterococcus faecalis, Escherichia coli, Haemophilus influenzae, Listeria
NOTES
2347
TABLE 1. Detection of S. pneumoniae by culture, EIA, and latex agglutination in consecutive CSF specimens from 10 patients with culture-proven S. pneumoniae meningitis Result fora: Patient
A B C
D
E F
G H I J
Day of antibiotic treatment
1 12 1 3 1 3 5 7 12 21 1 4 5 10 1 3 9 1 3 4 6 7 14 17 1 6 17 1 15 27 1 7 1 2 10
Bacterial culture of: CSF
Blood
1 2 1 2 1 2 2 2 2 2 1 2 2 2 1 2 2 1 2 2 2 2 2 2 1 2 2 1 2 2 1 2 1 1 2
1 2 2 2 2 NDb ND ND 2 ND 2 ND ND 2 1 ND ND 2 ND ND ND ND ND ND 1 2 ND 1 2 ND 1 ND 2 ND ND
EIA
Latex agglutination
Microscopy
1 2 1 1 1 1 1 1 2 2 1 2 2 2 1 1 1 1 1 1 1 1 1 2 1 2 2 1 1 2 1 1 1 1 1
1 2 1 2 2 2 2 2 2 2 1 2 2 2 1 1 2 1 1 2 2 2 2 2 1 2 2 1 2 2 1 2 2 2 2
1 2 1 2 1 2 2 2 2 2 1 1 2 2 1 1 2 1 1 1 2 2 2 2 1 2 2 1 2 2 2 2 1 2 2
a 1, positive result (S. pneumoniae detected); 2, negative result (S. pneumoniae not detected). b ND, not determined.
monocytogenes, Proteus sp., Pseudomonas aeruginosa, Salmonella enteritidis, Staphylococcus epidermidis, and viridans streptococci killed by exposure to bactericidal antibiotics. In broth, the EIA reacted with protein A released by Staphylococcus aureus; this reaction could easily be eliminated by adding solutions containing human IgG (e.g., CSF). Cultivation of S. pneumoniae from a CSF specimen is sometimes compromised by spontaneous autolysis or antibiotic treatment which decreases the concentration of living bacteria below the critical threshold for culturing (2, 3). Moreover, bacterial identification by culture takes at least 24 h. Thus, a rapid and reliable alternative means of diagnosis, like capsular antigen detection, is of clinical importance. Several approaches to diagnosing S. pneumoniae meningitis based on antigen detection have been pursued. Latex agglutination proved to be more sensitive than counterimmunoelectrophoresis or coagglutination (6, 7, 9). Latex agglutination is very simple to carry out and rapid, and special equipment is not required. Several latex agglutination kits are commercially available. In this study, a newly developed EIA specific for the com-
2348
NOTES
FIG. 1. Concentrations of LTA and TA detected by EIA in repeat CSF samples from two patients with pneumococcal meningitis. ,, EIA, latex agglutination, and culture positive; 1, EIA and latex agglutination positive.
mon pneumococcal antigens TA and LTA was compared with microbiological cultivation of bacteria and with the Wellcogen latex agglutination test. In contrast to the other methods, the EIA allowed the detection of antigens from S. pneumoniae up to 15 days after the onset of antibiotic therapy. For 11 CSF specimens drawn during antibiotic therapy, only the EIA was able to establish the diagnosis of S. pneumoniae meningitis. Holloway et al. (5) reported that bacterial concentrations of 106 to 107 CFU/ml were necessary to consistently detect pneumococcal antigen by latex agglutination. Yolken et al. (11) found a limiting concentration of 103 CFU/ml to be necessary for a positive response in their EIA. These authors (11) used a polyclonal antibody raised against the C polysaccharide (i.e., TA covalently linked to peptidoglycan) as the capture reagent and a polyclonal antibody against capsular polysaccharides from all 83 known serotypes of S. pneumoniae (Omniserum; Statens Serum Institut, Copenhagen, Denmark) as the detector antibody. Therefore, this assay detects large cell wall fragments containing TAs together with capsular polysaccharide linked by peptidoglycan (11). In addition to recognizing larger pneumococcal fragments, the EIA developed by us also recognizes free TA and LTA released by bacteria treated with antibiotics (4). This enlarges the spectrum of detectable antigens. The fact that small soluble antigens probably are not phagocytosed qualifies our assay to recognize S. pneumoniae antigens for several days after initiation of antibiotic treatment. The EIA reacted with all samples containing culturable S. pneumoniae, whereas three false-negative results were obtained with the latex agglutination assay. Furthermore, as a quantitative assay, this EIA can be used for scientific purposes, e.g., to correlate the CSF LTA and TA concentrations with the clinical outcome. The EIA is, however, more time-consuming and expensive than latex agglutination. Due to its greater demand for equipment compared with culture and latex agglutination, the EIA
J. CLIN. MICROBIOL.
will not replace these methods in routine clinical circumstances. Nevertheless, its use is indicated when the routine methods fail to identify the causative agent of meningitis. When quantification of the concentration of LTA and TA is not necessary, the EIA can be performed with a reduced standard curve: blank wells lacking LTA, wells containing LTA at 3.1 ng/ml (i.e., the quantification limit), and wells containing LTA at 50 ng/ml. This strategy reduces the demands for time, reagents, and equipment. Sørensen et al. (8) found that components of several Streptococcus mitis strains cross-reacted in the coagglutination assay. A cross-reaction with S. mitis was also found by us when culture supernatants of S. mitis killed by heat or by antibiotics were tested in the EIA. However, bacteria that typically cause meningitis did not produce false-positive results in this EIA. In conclusion, due to its high level of sensitivity and rapidity of performance, this EIA can be a valuable tool for the diagnosis of S. pneumoniae meningitis (i) from CSF specimens of patients with severe brain edema, on whom lumbar puncture cannot be performed during the first few days of treatment, or (ii) from CSF samples in cases for which prior antimicrobial therapy minimizes or negates the usefulness of culture or other antigen detection tests. This work was supported by the Deutsche Forschungsgemeinschaft (Na 165/2-2). REFERENCES 1. Behr, T., W. Fischer, J. Peter-Katalinic, and H. Egge. 1992. The structure of pneumococcal lipoteichoic acid. Improved preparation, chemical and mass spectrometric studies. Eur. J. Biochem. 207:1063–1075. 2. Fischer, G. W., R. Longfield, V. G. Hemming, A. Valdes-Dapena, and L. P. Smith. 1982. Pneumococcal sepsis with false-negative blood cultures. Am. J. Clin. Pathol. 78:348–350. 3. Fischer, G. W., L. P. Smith, V. G. Hemming, R. Longfield, A. A. ValdesDapena, and J. O. Lopreiato. 1984. Avoidance of false-negative blood culture results by rapid detection of pneumococcal antigen. JAMA 252:1742– 1743. 4. Fischer, H., and A. Tomasz. 1984. Production and release of peptidoglycan and wall teichoic acid polymers in pneumococci treated with beta-lactam antibiotics. J. Bacteriol. 157:507–513. 5. Holloway, Y., W. G. Boersma, H. Kuttschru ¨tter, and J. A. M. Snijder. 1992. Minimum number of pneumococci required for capsular antigen to be detectable by latex agglutination. J. Clin. Microbiol. 30:517–519. 6. Ingram, D. L., A. W. Pearson, and A. R. Occhiuti. 1983. Detection of bacterial antigens in body fluids with the Wellcogen Haemophilus influenzae b, Streptococcus pneumoniae, and Neisseria meningitidis (ACYW135) latex agglutination tests. J. Clin. Microbiol. 18:1119–1121. 7. Sippel, J. E., P. A. Hider, G. Controni, K. D. Eisenach, H. R. Hill, M. W. Rytel, and B. L. Wasilauskas. 1984. Use of the Directigen latex agglutination test for detection of Haemophilus influenzae, Streptococcus pneumoniae, and Neisseria meningitidis antigens in cerebrospinal fluid from meningitis patients. J. Clin. Microbiol. 20:884–886. 8. Sørensen, U. B. S., and J. Henrichsen. 1987. Cross-reactions between pneumococci and other streptococci due to C polysaccharide and F antigen. J. Clin. Microbiol. 25:1854–1859. 9. Tilton, R. C., F. Dias, and R. W. Ryan. 1984. Comparative evaluation of three commercial products and counterimmunoelectrophoresis for the detection of antigens in cerebrospinal fluid. J. Clin. Microbiol. 20:231–234. 10. Tunkel, A. R., and W. M. Scheld. 1995. Acute bacterial meningitis. Lancet 346:1675–1680. 11. Yolken, R. H., D. Davies, J. Winkelstein, H. Russell, and J. E. Sippel. 1984. Enzyme immunoassay for detection of pneumococcal antigen in cerebrospinal fluid. J. Clin. Microbiol. 20:802–805.
