* Context.-Cat scratch disease (CSD) commonly occurs secondary to Bartonella henselae infection, and the diagnosis has traditionally been made by microscopic findings, the identification of organisms by cytochemistry, and clinical history. However, cytochemical analysis tends to be very difficult to interpret, and histology alone may be insufficient to establish a definitive diagnosis of CSD.
Objective.-To demonstrate the presence of B henselae in tissue suspected of involvement by CSD, using a novel polymerase chain reaction (PCR) assay.
Design.-Isolates of 8 henselae (American Tissue Culture Collection 49793) and Afipia felis (American Tissue Culture Collection 49714) were cultured on blood agar and buffered charcoal yeast extract agar, respectively. DNA was isolated from these organisms and from formalin-fixed, paraffin-embedded tissue sections with involvement by CSD (8 patients). Negative controls included water, human placental tissue, and lymph node specimens from 6 patients with reactive lymphoid hyperplasia and from 2 patients with granulomatous lymphadenitis. A primer complementary to B henselae citrate synthase gItA gene se
quence was designed to perform a seminested PCR amplification. For restriction fragment length polymorphism analysis, PCR products were digested by Taql restriction enzyme and analyzed by gel electrophoresis.
Results.-Seminested PCR analysis of the cultured isolates of B henselae, but not of A felis, showed specific amplification. However, nonnested PCR did not provide consistently positive results in tissue sections with CSD. Therefore, we used a seminested PCR, which revealed positivity in all of the cases with clinicopathologic diagnoses of CSD. None of the negative controls showed positivity. Restriction enzyme provided confirmation of the specific PCR amplification of the B henselae sequence.
Conclusions.-Since the amplification product has a low molecular size (
Cat scratch disease (CSD) has been a recognized entity for approximately a century,1 since the initial identification of the uncommon manifestation of Parinaud oculoglandular syndrome. Since that time, the etiologic agent has been under continuous investigation, with various suspected etiologic agents coming briefly to the fore, including fungi,2 viruses,3 and atypical mycobacteria.45 In the 1980s, bacteria were identified in patient samples by Warthin-Starry staining.6 Bacteria from a patient with CSD were initially cultivated and identified as Afipia fells.6-8 However, recent studies using molecular techniques established B henselae as the etiologic agent.9
Despite this progress, diagnosis in suspected cases has been problematic because the culture of B henselae is known to be difficult and unreliable. 10,11 Direct detection of bacteria in tissue specimens by Warthin-Starry stain has traditionally been used; however, this stain is unreliable and lacks species specificity.11-14 Therefore, until recently, standard diagnosis of CSD has been based on the combination of multiple clinical features and laboratory information, including a history of animal exposure (often feline), lymphadenopathy, skin testing, and antibody testing.14,15 Recent efforts to diagnose CSD have relied on polymerase chain reaction (PCR), particularly amplification of the 16S ribosomal RNA (rRNA) and other molecular techniques to demonstrate B henselae infection in these patients, and this approach has significantly increased our ability to diagnose CSD.16-25
Polymerase chain reaction amplification of 16S rRNA is a useful technique for demonstration of B henselae; however, this method has not gained wide clinical application. Recently, PCR using primers against the citrate synthase gene, gItA, was developed for demonstration of B henselae; however, we had limited success using tissue fixed in formalin with this assay.24,26 To improve the clinical utility of PCR amplification of B henselae and diagnosis of CSD, we successfully developed a seminested PCR assay using tissue samples from patients with clinical evidence of CSD.
Clinical samples from 7 patients with clinically suspected CSD were evaluated retrospectively, and fresh tissue samples were analyzed prospectively from 1 patient (case 1). The latter was an 18-year-old man with recent multiple kitten scratches who presented with right inguinal lymphadenopathy. Combined with the clinical history, histologic findings supported the diagnosis of CSD. Both fresh and paraffin-embedded lymph node tissues were available for molecular assays. DNA from this sample and cultured organisms were used for the initial analysis and development of the PCR assay. In addition, 7 archival formalin-fixed, paraffin-embedded samples from 7 patients with the clinical diagnosis of CSD were used for PCR assays. Negative controls were water, normal placenta, and 8 lymph nodes from 6 patients with reactive lymphoid hyperplasia and 2 patients with nonnecrotizing granulomatous lymphadenitis (unknown etiology not related to CSD).
Bacterial Strains and Cultures
An isolate of B henselae obtained from the American Tissue Culture Collection (ATCC), Manassas, Va (ATCC 49793) was cultured on blood agar plates at 37C. An isolate of A felis (ATCC #49714) was grown on buffered charcoal yeast extract agar (humidified) at 30 deg C.
The 18-year-old man with recent history of multiple kitten scratches presented with lymphadenopathy. He was initially evaluated by fine-needle aspiration biopsy, and the cytologic examination showed a polymorphous lymphoid aggregate. Owing to marked cervical lymphadenopathy, subsequent excisional biopsy of the lymph node was performed, which showed necrotizing granulomatous lymphadenitis with epithelioid histiocytes surrounding central abscesses (Figure 1). Warthin-Starry stain was attempted; however, this staining failed to show unequivocal evidence of organisms because of high background silver staining. Combined with the clinical history, the histologic findings supported the diagnosis of CSD. Both fresh and paraffin-embedded lymph node tissues were available for molecular assays. DNA from this sample and cultured organisms were used for development of PCR assays. In addition, 7 archival formalin-fixed, paraffin-embedded samples from 7 patients with clinical diagnoses of CSD were used for PCR assays. Clinical features and histologic findings of patients analyzed in this study are shown in Table 2. Negative controls were 8 lymph nodes from 6 patients with reactive lymphoid hyperplasia and 2 patients with granulomatous lymphadenitis, normal placenta, and water.
Initial analysis of the PCR assay was performed with DNA isolated from cultures of B henselae and A felis. Our early attempts to amplify a portion of the 16S rRNA gene or using primers against the citrate synthase gene originally developed by Norman et a126 had inconsistent amplification or nonspecific bands (data not shown).24 Since these assays did not produce positive results consistently, we developed a seminested PCR assay.
Seminested PCR analysis using DNA isolated from cultures of B henselae and A felis showed specific amplification of B henselae (Figure 2). The primers against the gitA gene did not show amplification of DNA isolated from A felis, while seminested PCR amplification showed clear bands with the DNA isolated from the cultured B henselae organism. The PCR assay described in this study consistently showed positive results in all 8 samples with clinicopathologic diagnosis of CSD (Figure 2). None of the patients with lymphadenopathy unrelated to CSD, including 6 patients with reactive hyperplasia and 2 with granulomatous lymphadenitis, and no DNA isolated from normal placenta or water control samples showed any evidence of PCR amplification. Restriction fragment length polymorphism (RFLP) analysis with TaqI was used to assure the specific amplification of B henselae. All cases analyzed by this restriction enzyme yielded bands that were consistent with the RFLP profile of B henselae (Figure 2). Following restriction cutting, there are 2 products expected to be generated since the TaqI recognition site (TCGA) lies on the gene sequence at base pairs 918-921. Restriction digestion was performed using 1 unit of TaqI, which generates 2 fragments with sizes of 82 and 57 bp, respectively.
Cat scratch disease is usually a self-limiting disease and does not require therapy.28 The most common manifestation of CSD is lymphadenitis subsequent to scratches from cats or kittens; however, patients may rarely present with Parinaud oculoglandular disease, encephalopathy, osteomyelitis, endocarditis, or hepatosplenic infection." Some patients with multisystem involvement may benefit from antibiotic administration; therefore, it is critical that rapid identification of the organism be made by clinical laboratory assays.28,29 Culturing organisms from affected tissue is difficult, and it usually requires 2 to 3 weeks.30,31 Fresh material is not usually available because the differential diagnosis of CSD is often brought to the primary physician's attention after the review of histologic material. The skin testing used in the past for diagnosis also has limited clinical utility. Novel assays for detection of CSD in clinical samples is becoming more crucial, especially in light of the difficulties inherent in culturing the organisms that cause CSD.
To improve diagnosis of CSD, recent studies relied on PCR amplification. There have been 2 main molecular approaches to the diagnosis of CSD by PCR. Some of the earlier assays targeted amplification of 16S rRNA gene16-18 or the adjacent intergenic region between the 16S rRNA and 23S rRNA.23 These regions are present in all bacteria with species polymorphism, and conserved segments present within these regions allow specific identification of various organisms. We initially wanted to develop a robust clinical PCR-based detection and used PCR assays based on 16S rRNA amplification. However, as in our past experience and review of previously published data, this amplification was usually associated with presence of nonspecific bands.18
Molecular assays using the 16S rRNA gene generally require further analysis for the specific species of bacteria targeted for amplification. These analyses usually use hybridization with a species-specific probe and subsequent detection by radioactivity, enzyme activity, cloning, or subsequent sequencing.18,19,21-23,32 In the case of the 16S-23S rRNA intergenic region, the product can be identified by size alone. However, this technique has not been pursued in human tissue.23 Recently, the use of a gene specific to B henselae, the citrate synthase gene (gltA),24,26,33 appeared to provide a better opportunity for quick confirmation of B henselae. This gene was originally cloned by Norman et al26 and has been amplified by PCR in cultured clinical isolates from bacteremic cats. Added confirmation of the identity of the PCR product was performed by RFLP analysis using TaqI restriction endonuclease.
We initially used primers against the citrate synthase gene originally described by Norman et al.26 However, PCR amplification using the original primers against the citrate synthase gene gave inconsistent results on samples isolated from formalin-fixed, paraffin-embedded tissues, similar to our previous experience with the amplification of 16S rRNA gene.18,26 Therefore, we developed a sensitive seminested PCR assay for the diagnosis of CSD. To improve the efficient amplification of DNA that are usually fragmented during fixation in formaldehyde, we targeted amplification fragment size less than 200 bp and designed yet another primer allowing seminested PCR amplification. We used the same target gene, gItA, used by Avidor et al;33 however, we used a nested-PCR assay with a shorter amplicon size. With this technique we demonstrated successful amplification of B henselae from fresh tissues, paraffin-embedded archival material, and unstained histologic slide samples. Use of the seminested assay optimized PCR amplification and provided reliable identification of a single, unique band in each CSD patient and in B henselae DNA derived from bacterial culture. Since the amplification product still contains the unique restriction site present in the citrate-synthase gene recognized by the TaqI restriction enzyme, it could be used for confirmation of the amplification product by RFLP, eliminating the need for probe hybridization.
We emphasize that precautions are necessary because carryover contamination could be a significant source of false-positive results. Therefore, routine procedures for detection of carryover contamination are essential and should be complemented by prevention with special contamination-control procedures (eg, designated room for PCR amplification and special attention for prevention of microaerosol contamination when small aliquots are taken for the second PCR amplification).
The technique described here lends itself to a rapid timeframe and avoids more cumbersome clinical assays, such as bacterial culture and serology. Seminested PCR as described in this study is reliable for the identification of B henselae and provides laboratory support for diagnosis of CSD in a relatively short amount of time. Thus, both in specificity and rapidity, this technique provides a clear clinical advantage for patients needing a diagnostic evaluation for CSD.
We conclude that the PCR is a reliable assay for identification of B henselae, thus providing useful clinical information in a relatively short amount of time. Given the difficulties of culture and uncertainty in clinical diagnosis, this technique should be useful in arriving at a precise diagnosis of CSD.
1. Parinaud H. Conjonctivite infectieuse transmise parles animaux. Ann Ocul (Paris). 1889; 101 Ol :252-253.
2. Verhoeff FH. Parinaud's conjunctivitis: a mycotic disease due to a hitherto undescribed filamentous organism. Arch OphthalmoL 1913;42:344-351.
3. Turner W, Bigley NJ, Dodd MC, Anderson G. Hemagglutinating virus isolated from cat scratch disease. I Bacteriol. 1960;80:430-435.
4. Boyd GL, Craig G. Etiology of cat-scratch fever. J Pediatr. 1961;59:313-317.
5. Warwick W. The cat-scratch syndrome: many diseases or one disease? Prog Med Virol. 1967;9:256-301.
6. Wear DJ, Margileth AM, Hadfield TL, Fischer GW, Schlagel CJ, King FM. Cat scratch disease: a bacterial infection. Science. 1983;221:1403-1404.
7. English CK, Wear DJ, Margileth AM, Lissner CR, Walsh GP. Cat scratch disease: isolation and culture of the bacterial agent. JAMA. 1988;259:1347-1352.
8. Brenner DJ, Hollis DG, Moss CW, et al. Proposal of Afipia gen. nov. with Afipia felis sp. nov. (formerly the cat scratch disease bacillus), Afipia clevelandensis sp. nov. (formerly the Cleveland Clinic Foundation Strain), Afipia broomeae sp. nov., and three unnamed genospecies. J Clin Microbiol. 1991;29:24502460.
9. Giladi M, Avidor B, Kletter Y, et al. Cat scratch disease: the rare role of Afipia felis. J Clin Microbiol. 1998;36:2499-2502.
10. La Scola B, Raoult D. Culture of Bartonella quintana and Bartonella henselae from human samples: a 5-year experience (1993 to 1998). J Clin Microbiol. 1999;37:1899-1905.
11. Carithers HA. Cat scratch disease: an overview based on a study of 1200 patients. Am J Dis Child. 1985;139:1124-1133.
12. Margileth AM, Wear Di, Hadfield TL, Schlagel Cl, Spigel GT, Muhlbauer JE. Cat-scratch disease: bacteria in skin at the primary inoculation site. JAMA. 1984;252:928-931.
13. Min KW, Reed JA, Welch DF, Slater LN. Morphologically variable bacilli of cat scratch disease are identified by immunocytochemical labeling with antibodies to Rochalimaea henselae. Am I Clin Pathol. 1994;101:607-610.
14. Demers DM, Bass JW, Vincent JM, et al. Cat-scratch disease in Hawaii: etiology and seroepidemiology. J Pediatr. 1995;127:23-26.
15. Regnery RL, Tappero ). Unraveling mysteries associated with cat-scratch disease, bacillary angiomatosis, and related syndromes. Emerg Infect Dis. 1995; 1:16-21.
16. Medlin L, Elwood HJ, Stickel S, Sogin ML. The characterization of enzymatically amplified eukaryotic 16S-like rRNA-coding regions. Gene. 1988;71: 491-499.
17. Wilson KH, Blitchington RB, Greene RC. Amplification of bacterial 16S ribosomal DNA with polymerase chain reaction. Clin Microbiol. 1990;28:19421946.
18. Alkan S, Morgan MB, Sandin RL, Moscinski LC, Ross CW. Dual role for Afipia felis and Rochalimaea henselae in cat scratch disease. Lancet. 1995;345: 385.
19. Anderson BE, Sims K, Refinery RL, et al. Detection of Rochalimaea henselae DNA in specimens from cat scratch disease patients by PCR. J Clin Microbiol. 1994;32:942-948.
20, Sander A, Penno S. Semiquantitative species-specific detection of Bartonella henselae and Bartonella quintana by PCR-enzyme immunoassay. J Clin Microbiol. 1999;37:3097-3101.
21. O'Connor SP, Dorsch M, Steigerwalt AG, Brenner DJ, Stackebrandt E. 16S rRNA sequences of Bartonella bacilliformis and cat scratch disease bacillus reveal phylogenetic relationships with the alpha-2 subgroup of the class Proteobacteria. J Clin Microbiol. 1991;29:2144-2150.
22. Relman DA, Loutit JS, Schmidt TM, Falkow S, Tompkins LS. The agent of
bacil lary angiomatosis: an approach to the identification of uncultured pathogens. N Engl Jl Med. 1990;323:1573-1580.
23. Jensen WA, Fall MZ, Rooney J, Kordick DL, Breitschwerdt EB. Rapid identification and differentiation of Bartonella species using a single-step PCR assay. J Clin Microbiol. 2000;38:1717-1722.
24. Avidor B, Kletter Y, Abulafia S, Golan Y, Ephros M, Giladi M. Molecular diagnosis of cat scratch disease: a two-step approach. J Clin Microbiol. 1997;35: 1924-1930.
25. Zeiter Z, Liang Z, Raoult D. Genetic classification and differentiation of Bartonella species based on comparison of partial ftsZ gene sequences. I Clin Microbiol. 2002;40:3641-3647.
26. Norman AF, Regnery RL, Jameson P, Greene C, Krause DC. Differentiation of Bartonella-like isolates at the species level by PCR-restriction fragment length polymorphism in the citrate synthase gene. J Clin Microbiol. 1995;33:1797-1803.
27. Alkan S, Lehman C, Sarago C, Sidawy MK, Karcher DS, Garrett CT. Polymerase chain reaction detection of immunoglobulin gene rearrangement and bcl2 translocation in archival glass slides of cytologic material. Diagn Mol Pathol. 1995;4:25-31.
28. Conrad DA. Treatment of cat-scratch disease. Curr Opin Pediatr. 2001;13: 56-59.
29. Bass JW, Freitas BC, Freitas AD, et al. Prospective randomized double blind placebo-controlled evaluation of azithromycin for treatment of cat-scratch disease. Pediatr Infect Dis. 1998; 17:447-452.
30. Welch DF, Pickett DA, Slater LN, Steigerwalt AG, Brenner DJ. Rochalimaea henselae, sp. nov., a cause of septicemia, bacillary angiomatosis, and parenchymal bacillary peliosis. J Clin Microbiol. 1992;30:275-280.
31. Welch DF, Slater LN. Bartonella. In: Murray PR, Barron EJ, Haller MA, Tenover FC, Yolken RH, eds. Manual of Clinical Microbiology. 6th ed. Washington, DC: American Society of Microbiology; 1995:690-695.
32. Sander A, Posselt M, Bohm N, Ruess M, Altwegg M. Detection of Bartonella henselae DNA by two different PCR assays and determination of the genotypes of strains involved in histologically defined cat scratch disease. J Clin Microbiol. 1999;37:993-997.
33. Avidor B, Varon M, Marmor S, et al. DNA amplification for the diagnosis of cat-scratch disease in small-quantity clinical specimens. Am j Clin Pathol. 2001;115:900-909.
Ben Margolis, MD; Isinsu Kuzu, MD; Marille Herrmann, MD, PhD; Michele D. Raible, MD; Eric Hsi, MD; Serhan Alkan, MD
Accepted for publication January 24, 2003.
From the Department of Pathology, Loyola University Medical Center, Maywood, Ill (Drs Margolis, Kuzu, Herrmann, Raible, and Alkan); and Cleveland Clinic Foundation, Cleveland, Ohio (Dr Hsi). Dr Margolis is now with Kishwaukee Community Hospital, DeKalb, Ill; Dr Kuzu is now with the University of Ankara, Ankara, Turkey; Dr Herrmann is now with the Armed Forces Institute of Pathology, Washington, DC; and Dr Raible is now with The University of Illinois at Chicago, Chicago. Ill.
Reprints: Serhan Alkan, MD, Department of Pathology, Loyola University Medical Center, 2160 S First Ave, Building 110, Room 2230, Maywood, IL 60153 (e-mail: Balkan@lumc.edu).
Copyright College of American Pathologists Jun 2003
Provided by ProQuest Information and Learning Company. All rights Reserved