Serological response to "Rochalimaea henselae" antigen in suspected cat-scratch disease

The histologic spectrum of hepatic cat scratch disease. A series of six cases with confirmed Bartonella henselae infection

Cat scratch disease (CSD), a common cause of regional lymphadenitis, has been linked to Bartonella henselae infection. Although rare, dissemination with hepatic involvement has been documented. Six cases of hepatic CSD were retrieved and probed for B. henselae DNA. Hematoxylin-eosin, trichrome, methenamine silver, Gram, Ziehl-Neelsen, and Warthin-Starry stained slides were reviewed by all authors, and the histologic findings were recorded. B. henselae infection was confirmed in all cases using combined polymerase chain reaction and Southern blot methods. All biopsies contained multiple granulomas, many with characteristic stellate abscesses surrounded by three distinct zones: an inner layer of palisading histiocytes, an intermediate lymphocytic rim, and an outermost zone of fibrosis. One case had noncaseating granulomas with prominent giant cells. Warthin-Starry stains were positive in two. The surrounding parenchyma in all cases contained lymphocytic infiltrates within expanded fibrotic portal triads and adjacent dilated sinusoids. There is a distinct spectrum of histologic changes in hepatic CSD. Although the unit lesion is granulomatous, the organization varies markedly. All cases show predictable nonspecific surrounding parenchymal changes. When multiple hepatic lesions are found in patients with constitutional symptoms suggesting an infectious illness, CSD should be included in the differential diagnosis.

Will the real agent of cat-scratch disease please stand up?

Cat-scratch disease has been recognized since 1889 in association with the oculoglandular syndrome of Parinaud. The epidemiologic association with cats was first made in 1931 and further substantiated throughout the years, refining the interaction predominantly to kittens. Putative infectious agents have included numerous species of bacteria, chlamydiae, and viruses. The cultivation of Afipia spp. in the late 1980s appeared to answer the mystery of the identity of the agent. However, even more recent analysis, which has combined traditional microbiology, molecular methods, and additional epidemiology, has demonstrated that Bartonella (Rochalimaea) henselae is the definitive agent of cat-scratch disease. Our understanding of the pathogenesis of cat-scratch disease and other diseases caused by Bartonella species is incomplete and the spectrum of diseases continues to emerge. We review historic and modern efforts to understand the etiology of cat-scratch disease and related syndromes.

Does coinfection of Bartonella henselae and FIV induce clinical disorders in cats?

It was found that Bartonella henselae (B. henselae) may induce clinical disorders in cats in natural conditions from a comparison of the serological status for B. henselae with the serostatus for feline immunodeficiency virus (FIV) and several clinical characteristics in 170 domestic cats. Seropositivity for B. henselae was not significantly different between FIV antibody-positive and -negative cats (18.4% vs 16.0%). The incidence of clinical characteristics were compared among four cat groups distinguished by the reactivity of sera against B. henselae and FIV. The incidence of lymph node swelling was lower in only FIV antibody-positive cats (3.0%), but higher in B. henselae antibody-positive cats (13.6%) and significantly higher in both B. henselae and FIV antibody-positive cats (42.9%) compared with the incidence of lymph node swelling in cats which were negative for both antibodies (5.5%). The same relation was also observed for the incidence of gingivitis among the 4 cat groups, suggesting that coinfection of B. henselae and FIV may be associated with gingivitis and lymphadenopathy in cats.

Serological study of Bartonella henselae in cat scratch disease in Japan

It has become clear that Bartonella henselae is a common cause of cat scratch disease (CSD). The indirect fluorescence antibody (IFA) test for detection of IgG and IgM antibodies to B. henselae concerning CSD showed that 5 (50%) of 10 patients with CSD had a serum IgG antibody titer of 1:128 or more and that 2 (20%) patients had a serum IgM antibody titer of 1:20 or more. One of 7 asymptomatic members of patients' families (14%) had IgG antibody to B. henselae at a titer of 1:256. IgM antibody to B. henselae was not detected in sera from the patients' families. Both IgG and IgM antibodies to B. henselae were not detected in sera from the healthy control group. These data suggest that B. henselae may be a cause of CSD in Japan.

Experimental transmission of Bartonella henselae by the cat flea

Bartonella henselae is an emerging bacterial pathogen, causing cat scratch disease and bacillary angiomatosis. Cats bacteremic with B. henselae constitute a large reservoir from which humans become infected. Prevention of human infection depends on elucidation of the natural history and means of feline infection. We studied 47 cattery cats in a private home for 12 months to determine the longitudinal prevalence of B. henselae bacteremia, the prevalence of B. henselae in the fleas infesting these cats, and whether B. henselae is transmitted experimentally to cats via fleas. Vector-mediated transmission of B.henselae isolates was evaluated by removing fleas from the naturally bacteremic, flea-infested cattery cats and transferring these fleas to specific-pathogen-free (SPF) kittens housed in a controlled, arthropod-free University Animal Facility. B. henselae bacteremia was detected in 89% of the 47 naturally infected cattery cats. A total of 132 fleas were removed from cats whose blood was simultaneously cultured during different seasons and were tested individually for the presence of B. henselae DNA by PCR. B. henselae DNA was detected in 34% of 132 fleas, with seasonal variation, but without an association between the presence or the level of bacteremia in the corresponding cat. Cat fleas removed from bacteremic cattery cats transmitted B. henselae to five SPF kittens in two separate experiments; however, control SPF kittens housed with highly bacteremic kittens in the absence of fleas did not become infected. These data demonstrate that the cat flea readily transmits B. henselae to cats. Control of feline infestation with this arthropod vector may provide an important strategy for the prevention of infection of both humans and cats.

Cytologic features of necrotizing granulomatous inflammation consistent with cat-scratch disease

Approximately 2,000 cases of cat-scratch disease are reported annually. It is an uncommon cause of unilateral lymphadenopathy in children and adults. We present the cytologic features of necrotizing granulomatous lesions consistent with cat-scratch disease from various sites. Eleven cases from 10 patients with a mean age of 16 yr (range, 2-33 yr) were biopsied by fine-needle aspiration. All gave a history of previous exposure to cats. Two children had atypical presentations; one with multiple hypodense areas within the liver and spleen, and the other with features of encephalitis. Other locations included lymph nodes from axilla, inguinal, preauricular, and cervical areas, and other body sites such as the scapula and the parotid gland. Histologic staging of the inflammatory process was correlated with cytologic features. There were 5 cases of early features, 3 cases in the middle phase, and 3 cases in late stage. The differential diagnosis of each stage is discussed. The 2 children with atypical presentations are described in detail.

Prevalence of Bartonella henselae bacteremia, the causative agent of cat scratch disease, in an Australian cat population

In order to determine the prevalence of Bartonella henselae becteremia in an Australian cat population we examined blood cultures on a group of Sydney cats. Cats referred to the Concord Animal Hospital for euthanasia were selected randomly for blood culture and serum sampling. Blood samples were lysed and centrifuged and then cultured for up to five weeks. Suspicious colonies were identified biochemically as probable B. henselae. Selected isolates were confirmed as B. henselae using the polymerase chain reaction. Of the cats accrued throughout Sydney, 27/77 (35%) were culture positive for B. henselae, of these 24/59 (40%) were feral cats and 3/18 (16%) were domestic. Most cats in the study were younger than one year (mean 9.9 months). Our study demonstrates that bacteremia with B. henselae is common in the metropolitan cat population and suggests that it is particularly prevalent among feral animals. By contrast Cat Scratch Disease (CSD) is a relatively uncommon clinical diagnosis in the Australian population. Explanations for this discrepancy may include poor transmission, low bacterial virulence and underdiagnosis. It is possible that feral animals are a greater potential risk source for this infection than domestic cats.

Bartonella (Rochalimaea) quintana infections

Bartonella (formerly Rochalimaea) quintana is the etiological agent of trench fever, a disease extensively reported during the World Wars. Recent molecular biology approaches have allowed dramatic extension of the spectrum of Bartonella infections. B. quintana is now also recognized as an etiological agent of fever and bacteremia, endocarditis, bacillary angiomatosis, and chronic lymphadenopathy. Human immunodeficiency virus-infected patients and/or homeless people are the most vulnerable to infection. Poverty and louse infestation were the main epidemiological factors associated with B. quintana infections during wartime. Although poverty and chronic alcoholism have been associated with modern cases of trench fever and bacteremia due to B. quintana in Europe and the United States, vectors for B. quintana have not been clearly identified and B. quintana has not been isolated from modern-day lice. Microscopic bacillary angiomatosis lesions are characterized by tumor-like capillary lobules, with proliferating endothelial cells. In vitro experiments have shown that B. quintana survives within endothelial cells and stimulates cell proliferation. These observations, together with the finding that lesions may regress when antibiotic therapy is administered, strongly suggest that B. quintana itself stimulates angiogenesis. Bartonella infections are characterized by a high frequency of relapses after brief courses of antibiotic therapy. It is to be noted that in vitro, although Bartonella species are highly susceptible to antibiotics, only the aminoglycosides have proved to be bactericidal. However, the most effective antibiotic regimen for Bartonella infections remains to be established.

Optimum treatment of intracellular infection

The intracellular location of some micro-organisms has been early recognised as a critical point to explain failure of antibiotic therapy to eradicate such pathogens from infected hosts. Most often parasites invade 'professional' phagocytic cells, including neutrophils, monocytes and macrophages, by resisting the intracellular bactericidal phagolysosomal pathway. Alternatively, they may invade 'non-professional' phagocytic cells (cells with fewer phagocytic and bactericidal abilities) such as endothelial cells, or even cells without lysosomes such as erythrocytes. The intracellular activity of an antibiotic depends on several factors including its ability to reach the eukaryotic cell membrane, its subcellular localisation as compared to that of the parasite, the possibility that the intracellular milieu may partially inactivate its activity, and the susceptibility of the intracellular form of the parasite. In vitro and animal models have been developed to investigate antibiotic activity against intracellular pathogens. However, it should be emphasised that only data obtained from patients give reliable information to define the optimum antibiotic regimen.

Cat scratch disease and acquired immunodeficiency disease: diagnosis by transmission electron microscopy

A 33-year-old, homosexual, cat-owning, African-American man with human immunodeficiency virus infection by positive serologic tests and acquired immunodeficiency syndrome by CD4 lymphocyte count alone (39 cells/mL) presented with a one-year history of intermittent fever, weight loss, and generalized lymphadenopathy. A malignant lymphoma was suspected clinically. Light microscopic study of a left inguinal lymph node biopsy specimen revealed effacement of the lymph node architecture by a diffuse infiltrate of large, atypical reticulum cells, loose, patchy granulomatous inflammation, diffuse hyaline fibrosis, diffusely proliferated blood vessels, and multifocal degeneration and necrosis. Lymph follicles were absent and lymphocytes were moderately depleted. Microorganisms were not seen in lymph node sections stained with special histochemical stains (including the Warthin-Starry stain). These light microscopic changes were considered suggestive of a malignant lymphoma, especially Hodgkin's disease. The diagnosis of cat scratch disease (CSD) became apparent only after transmission electron microscopic study of the lymph node revealed clusters of small, pleomorphic bacteria in degenerated collagenous tissue and in blood vessel walls. This case illustrates the value of transmission electron microscopy in making the diagnosis of CSD, especially when light microscopic changes are superimposed on those of late human immunodeficiency virus infection of the lymph node.

Comparative seroreactivity to Bartonella henselae and Bartonella quintana among cats from Israel and North Carolina

Bartonella henselae, the predominant cause of cat scratch disease, and Bartonella quintana, the cause of trench fever, are closely related Bartonella species that induce cross-reactivity when cat or human sera are tested using an indirect immunofluorescence antibody (IFA) test. Cats are the natural reservoir for B. henselae, whereas a mammalian reservoir host for B. quintana has not been identified. Serum samples from 114 cats from Israel and 114 cats from North Carolina were tested by IFA for seroreactivity to B. henselae and B. quintana antigens. Similar numbers of cats from Israel [45 (39.5%)] and from North Carolina [46(40.4%)] were seroreactive to both antigens, however, as compared to cats from North Carolina [8 (7%)], a significantly (P = 0.001) larger number of cats from Israel were seroreactive to B. quintana antigen only [23 (20.2%)]. In addition, mean antibody titers were lower to B. henselae than to B. quintana (P = 0.0001) in the cats from Israel, whereas similar mean titers to both antigens were identified in cats from North Carolina. Absorption of serum using whole B. henselae organisms resulted in a significantly greater (P = 0.0001) decrease in antibody titer to B. henselae between absorbed and non-absorbed sera, as compared to the decrease in antibody titer following absorption with whole B. quintana organisms. There was a similar decrease in antibody titer in sera from cats experimentally infected with B. henselae and in cats naturally exposed to Bartonella species from Israel and North Carolina. Our results indicate that absorption of serum will, in most instances, distinguish species-specific reactivity by IFA to B. henselae from cross-reactivity to B. quintana in cats experimentally infected with B. henselae. The data support the conclusion that B. henselae is the principal Bartonella species responsible for seroreactivity against B. henselae and B. quintana in naturally exposed cats from Israel or North Carolina. It also suggests that in Israel, cats are exposed to one or more antigenically different Bartonella species, sub-species or strains, that seroreact by IFA more intensely with B. quintana antigen.

[Cat scratch disease in immunocompetent children]

First cause of chronic benign lymphadenopathy, cat scratch disease occurs mainly in children and young adults. The nature of its bacterial agents is not definitely established, but Atlanta's CDC actually considers Afipia felis and Rochalimaea hensaela as the agents of the disease. It usually presents as chronic adenitis but there are severe forms of the disease with systemic manifestation. In such forms, new techniques such as serology and molecular biology are particularly useful for the diagnosis. The prognosis is always excellent and an antibiotic treatment is only recommended for systemic forms to reduce the duration of the disease.

Bartonelloses

The genus Bartonella now includes four species which may infect humans : B. bacilliformis, B. quintana, B. henselae, and B. elizabethae. B. bacilliformis, the agent of Carrion's disease, was the only species of the genus since 1993 when Rochalimaea species were removed from the genus Rochalimaea and included in the genus Bartonella, within the family Bartonellaceae. B. quintana is the etiologic agent of trench fever, bacillary angiomatosis, septicemia, endocarditis, and chronic lymphadenopathy. B. henselae is responsible for bacillary angiomatosis, peliosis of the liver or the spleen, septicemia, endocarditis, and cat scratch disease. There is a single isolate of B. elizabethae, which was recovered from the blood of a patient involved with endocarditis. The spectrum of clinical manifestations related to Bartonella species has extended since 1990, partly because of newly available molecular biological techniques. However, some aspects of Bartonella-related diseases remain unsettled, including epidemiology, physiopathology, and optimum therapy to be administered.

Diagnostic value of the indirect immunofluorescence assay in cat scratch disease with Bartonella henselae and Afipia felis antigens

Serum samples from 35 cat scratch disease (CSD) patients, 180 control patients (123 without lymph node enlargement and 57 with lymph node enlargement not evoking CSD), and 102 nonpatient subjects (35 with cat contact and 67 without cat contact) were tested by semiquantitative indirect immunofluorescence assay for the presence of antibodies directed to Afipia felis (ATCC 53690T) or Bartonella henselae (ATCC 49882T). The CSD group had statistically higher antibody titers against B. henselae than the control groups (P < 10(-5)), whereas no difference in A. felis antibody titers was evidenced among all groups tested. Among the 317 serum samples studied, the three with high A. felis antibody titers ( > or = 64) also had high antibody titers against other alpha-2 proteobacteria. The value of the indirect immunofluorescence assay with B. henselae antigen for the diagnosis of CSD was as follows: for a cutoff of 32, sensitivity was 0.80, specificity was 0.85, and the likelihood ratio was 5.1; for a cutoff of 64, the likelihood ratio was 12.1. In summary, in France, CSD is associated with high antibody titers against B. henselae, as previously described in the United States. However, the causes for B. henselae seronegativity in CSD patients and those for high antibody titers outside the typical nosological frame of CSD still have to be identified.

New serotype of Bartonella henselae in endocarditis and cat-scratch disease

BACKGROUND

The fastidious nature of Bartonella henselae is such that demonstration of clinical infection relies mainly on serological or molecular biological methods. Only five isolates have been obtained from patients with cat-scratch disease (CSD) and none from endocarditis.

METHODS

We isolated B henselae from a CSD patient and, for the first time, from a patient with endocarditis. The isolates were characterised on the basis of morphology, biochemistry, cell-wall fatty-acid analysis, PCR-restriction fragment length polymorphism analysis of the 16-23S intragenic spacer region, and 16S rRNA gene sequences. Characterisation of these isolates indicated them to belong to a new serogroup, which we have called "Marseille", and to a new genotype based on the 16S rRNA gene sequence. The new variant was incorporated into an immuno-fluorescence antibody test (IFAT), which was used to reassess serum samples from 113 CSD patients who were seronegative with the conventional IFAT.

FINDINGS

18 (16%) of these apparently seronegative patients yielded significantly raised titres. 20 CSD patients who were seropositive as judged by the conventional IFAT remained seropositive with the new IFAT.

INTERPRETATION

Antigenic variability within the species is one possible reason for inconsistent results in the serological diagnosis of CSD.

Predominance of two Bartonella henselae variants among cat-scratch disease patients in the Netherlands

Restriction endonuclease analysis of the PCR-amplified 16S-23S rRNA gene spacer region was used to investigate the prevalence of Bartonella henselae variants in samples from cat-scratch disease (CSD) patients. Analysis of spacer PCR fragments from 27 Bartonella DNA-positive samples from Dutch patients with CSD with AluI revealed two restriction fragment length polymorphism (RFLP) patterns, patterns A and B. Twenty samples yielded B. henselae pattern A, and 7 samples yielded B. henselae pattern B. Three samples from North American patients with CSD were shown to contain B. henselae with RFLP pattern B. To be able to detect and differentiate Bartonella DNA in clinical material more sensitively and faster, two B. henselae PCRs which amplify part of the 16S rRNA gene and which can discriminate between two B. henselae variants were developed. Thirty-two of 41 Bartonella DNA-positive samples from Dutch patients with CSD contained type I B. henselae, 7 samples contained type II B. henselae, and two samples were negative in both type-specific PCRs. Two samples from North American patients with CSD both contained type II B. henselae. A 100% correlation was found between the AluI spacer RFLP pattern and the 16S rRNA PCR type. We have shown that Dutch patients with CSD contain a limited number of B. henselae variants, suggesting that, in contrast to systemic bartonellosis, CSD in immunocompetent patients is caused by a limited number of B. henselae variants.

Molecular approaches to the identification of unculturable infectious agents

New molecular biologic techniques, particularly representational difference analysis, consensus sequence-based polymerase chain reaction, and complementary DNA library screening, have led to the identification of several previously unculturable infectious agents. New agents have been found in tissues from patients with Kaposi's sarcoma, non-A, non-B hepatitis, hantavirus pulmonary syndrome, bacillary angiomatosis, and Whipple's disease by using these techniques without direct culture. The new methods rely on identifying subgenomic fragments from the suspected agent. After a unique nucleic acid fragment belonging to an agent is isolated from diseased tissues, the fragment can be sequenced and used as a probe to identify additional infected tissues or obtain extended portions of the agent's genome. For agents that cannot be cultured by standard techniques, these approaches have proved invaluable for identification and characterization studies. Applying these techniques to other human diseases of suspected infectious etiology may rapidly elucidate novel candidate pathogens.

Bartonella (Rochalimaea) quintana causing fever and bacteremia in an immunocompromised patient with non-Hodgkin's lymphoma

A 48-yr-old man with stage IV non-Hodgkin's lymphoma, became neutropenic following chemotherapy and developed a fever. His blood cultures were processed to enhance the yield of fastidious bacteria. A slow-growing, capnophilic Gram-negative rod was isolated. The febrile episode was treated with cefotaxime, imipenem and vancomycin and resolved. The bacterial isolate was identified as Bartonella (Rochalimaea) quintana by 16S-rDNA gene sequencing. The isolate showed 99.8% sequence homology with the type strain. This is the first isolation of Bartonella (Rochalimaea) quintana from a bacteremic patient in Australia. This bacterium is a fastidious Gram-negative rod requiring prolonged culture for its isolation. Patients with culture-negative pyrexia, especially immunocompromised patients, may need to be investigated for infection with this agent.

Evaluation of serological response to Bartonella henselae, Bartonella quintana and Afipia felis antigens in 64 patients with suspected cat-scratch disease

The serological response to Bartonella henselae, B. quintana, and Afipia felis was assessed by an indirect fluorescence antibody test (IFAT) in 64 patients with suspected cat-scratch disease (CSD) recruited from the Bordeaux area in France. Blood samples were collected from 57 patients with chronic lymphadenopathy who underwent lymph-node biopsy with suggestive histopathologic features of CSD, and from an additional 7 patients with suspected CSD who underwent surgical incision and drainage because of lymph-node tenderness. Of the patients, 31 were male and 33 female, with a median age of 27 years (range 2-89). 69.8% reported cat and/or dog contact. Of the 26/64 (40.6%) patients, serum samples were positive at a titer of 1:100 or more for immunoglobulin G (IgG) antibodies (17 only to B. henselae, 1 only to B. quintana, 3 only to Afipia felis, and 5 to both B. henselae and B. quintana). IgM or IgA antibodies were also detected in 10 patients with IgG antibodies to B. henselae. 11 (17.2%) of the 64 patient serum samples were positive at a low titer of 1:50. These data suggested that serological response assessed by standard IFAT is not enough to confirm a CSD diagnosis.

Prolonged Bartonella bacteremia in cats associated with cat-scratch disease patients

Recent evidence supports a causal relationship between Bartonella (Rochalimaea) henselae, cat-scratch disease (CSD), and bacillary angiomatosis. Cats appear to be the primary reservoir. Blood from 19 cats owned by 14 patients diagnosed with CSD was cultured. Blood samples from cats owned by veterinary students (n = 25) having no association with CSD or bacillary angiomatosis were cultured as controls. Eighty-nine percent (17 of 19) of cats associated with CSD patients and 28% (7 of 25) of controls were bacteremic with Bartonella species (chi-square = 16.47; P < 0.001). Twenty-three isolates were characterized as B. henselae, while one isolate from the cat of a CSD patient appeared to be a new Bartonella species. Thirteen cats remained culture positive during the ensuing 12-month period. Our results support the conclusion that B. henselae is the predominant species involved in CSD and is transmitted by cats. The incidence of Bartonella bacteremia in control cats suggests that B. henselae bacteremia is prevalent among the domestic cat population in the United States.

MICs of 28 antibiotic compounds for 14 Bartonella (formerly Rochalimaea) isolates

We assessed in vitro the antibiotic susceptibilities of 14 Bartonella isolates of the species B. quintana, B. vinsonii, B. henselae, and B. elizabethae. Columbia agar base supplemented with 5% horse blood was used as the antibiotic assay medium. Bacterial growth could be evaluated within 5 days after incubation of the plates at 37 degrees C in a 5% carbon dioxide atmosphere. The MICs at which 90% of isolates are inhibited (MIC90s) were 0.06 microgram/ml for penicillin G and amoxicillin and 0.25 microgram/ml for ticarcillin and cefotaxime. The MIC90s of oxacillin and cephalothin were 4 and 16 micrograms/ml, respectively. The MIC90s ranged from 1 to 4 micrograms/ml for aminoglycosides. Erythromycin, doxycycline, and rifampin displayed MIC90s of 0.12, 0.12, and 0.25 microgram/ml, respectively. MIC90s were 1 and 5 micrograms/ml for trimethoprim-and sulfamethoxazole, respectively, 64 micrograms/ml for fosfomycin, and 16 micrograms/ml for colistin and vancomycin. The study confirms the high levels of in vitro susceptibility of Bartonella agents to antibiotics.

Intracellular location of Bartonella henselae cocultivated with Vero cells and used for an indirect fluorescent-antibody test

Bartonella henselae, the major causative agent of cat scratch disease, was cocultivated with Vero cells on chamber slides and visualized by indirect immunofluorescence by using a patient serum containing specific antibodies. Confocal microscopy localized the granular B. henselae-specific fluorescence mainly around the nuclei of Vero cells. By transmission electron microscopy, these granules were identified as clusters of multiple intracellular organisms. Fixed slides with the monolayers of Vero cells with intracellular B. henselae were used for an indirect fluorescent-antibody test to investigate the seroprevalence of specific immunoglobulin G in 100 serum samples from blood donors. Seventy-four serum samples were negative; 19, 3, and 4 were positive at dilutions of 1:64, 1:128, and 1:256, respectively. In our population, a serum titer of 1:256 or greater should stimulate further investigations. Moreover, elucidation of the mechanism by which B. henselae enters the cells may help to understand the pathogenesis of cat scratch disease.

Bartonella henselae is a causative agent of cat scratch disease in Australia

We report the first isolation of Bartonella henselae from the blood and fleas of a cat of a patient with cat scratch disease (CSD) in Australia. A 49-year-old man presented with a history that 3 weeks after he had removed fleas from his cat he had developed fever, lethargy and anorexia for 3 days. This was followed by the appearance of axillary lymphadenopathy. There was no history of a bite or scratch and no primary lesion on the skin. Two fine needle aspirates of the axillary lymph node showed granulomatous lymphadenitis with no organisms seen by Warthin-Starry silver staining or electron microscopy. No organism was cultured from the patient's lymph node aspirates or blood cultures processed by lysis centrifugation. However, the polymerase chain reaction (PCR) using p24E and p12B primers gave a 280 bp band indistinguishable from Bartonella henselae when using DNA extracted from the lymph node aspirates and the patient's blood leucocytes. DNA sequencing of the PCR product from the patient's blood showed that the DNA was from Bartonella henselae. The patient's serum had a titre of 1024 in an indirect immunofluorescence antibody test for Bartonella henselae. Bartonella henselae was subsequently cultured from fleas and blood taken from the patient's cat. This case provides evidence that Bartonella henselae is a causative agent of CSD in Australia and supports a possible role for fleas in transmission of the disease.

Serology to Bartonella (Rochalimaea) henselae may replace traditional diagnostic criteria for cat-scratch disease

Bartonella (Rochalimaea) henselae has been identified as causative agent of cat-scratch disease (CSD). Employing an indirect fluorescence antibody test we found that 20 (100%) out of 20 children with suspected CSD had serum titres to B. henselae of > or = 512. By contrast, in all but one of the controls (n = 332) including subjects exposed to the same cats, patients with diseases other than CSD, and blood donors the antibody titres were < or = 256 (P < 0.001). However, significant regional differences in seroprevalence were noted. CONCLUSION. Serology to B. henselae may supplant traditional criteria for the diagnosis of CSD and prevent patients from unnecessary surgery.

Characterization of a 17-kilodalton antigen of Bartonella henselae reactive with sera from patients with cat scratch disease

A library of Bartonella (Rochalimaea) henselae DNA was constructed in the cloning vector lambda ZAPII and screened for expression of antigenic proteins by using a pool of sera from patients who had been diagnosed with cat scratch disease (CSD) and had antibodies to Bartonella spp., as determined by indirect fluorescent-antibody (IFA) assay. Ten immunoreactive phages were subcloned as recombinant plasmids by in vivo excision. All 10 recombinants expressed a protein of approximately 17 kDa when they were examined by immunoblot with the pool of human sera. Restriction endonuclease digestion of each recombinant plasmid indicated seven profiles, suggesting that cloning bias was not the reason for repeated isolation of clones expressing the 17-kDa antigen. The gene coding for the 17-kDa antigen was sequenced and shown to code for an open reading frame of 148 amino acids with a predicted molecular mass of 16,893 Da. The amino terminus of the deduced amino acid sequence was hydrophobic in nature and similar in size and composition to signal peptides found in gram-negative bacteria. The remainder of the deduced amino acid sequence was more hydrophilic and may represent surface-exposed epitopes. Further subcloning of the 17-kDa antigen as a biotinylated fusion protein in the expression vector PinPoint Xa-2 resulted in a 30-kDa protein that was highly reactive on immunoblots with individual serum samples from patients with CSD. The agreement between reactivity with the 30-kDa fusion protein on immunoblot analysis and the results obtained by IFA assay was 92% for IFA-positive sera and 88% for IFA-negative sera. The recombinant-expressed 17-kDa protein should be of value as an antigen for serologic diagnosis of CSD and Bartonella infections and warrants further study in attempts to develop a subunit vaccine to prevent long-term Bartonella infection in cats and the potential for further spread of these organisms to humans.

Bartonella henselae prevalence in domestic cats in California: risk factors and association between bacteremia and antibody titers

The isolation of Bartonella henselae, the agent of cat scratch disease, from the blood of naturally infected domestic cats and the demonstration that cats remain bacteremic for several months suggest that cats play a major role as a reservoir for this bacterium. A convenience sample of 205 cats from northern California was selected between 1992 and 1994 to evaluate the B. henselae antibody and bacteremia prevalences and to determine the risk factors and associations between bacteremia and antibody titers. B. henselae was isolated from the blood of 81 cats (39.5%). Forty-two (52%) of these bacteremic cats were found to be infected with > or = 1,000 CFU/ml of blood. Impounded or former stray cats were 2.86 (95% confidence interval [CI] = 1.94, 4.22) times more likely to be bacteremic than the pet cats. Young cats ( < 1 year old) were more likely than adult cats to be bacteremic (relative risk = 1.64; (95% CI = 1.19, 2.28). Bacteremic cats were more likely than nonbacteremic cats to be infested with fleas (relative risk = 1.64; 95% CI = 1.38, 1.96). No association between B. henselae infection and feline immunodeficiency virus antibody prevalence was observed. Eighty-one percent of the cats (166 of 205) tested positive for B. henselae antibodies, and titers were higher in bacteremic than in nonbacteremic cats. Multiple logistic regression analysis indicated that younger age and seropositivity for B. henselae antibodies were associated with bacteremia. Serological screening for Bartonella antibodies may not be useful for the identification of bacteremic cats (positive predictive value = 46.4%), but the lack of antibodies to B. henselae was highly predictive of the absence of bacteremia (negative predictive value = 89.7%). Seronegative cats may be more appropriate pets for immunocompromised individuals who are at increased risk for developing severe B. henselae disease.

Strategy to detect and identify Bartonella species in routine clinical laboratory yields Bartonella henselae from human immunodeficiency virus-positive patient and unique Bartonella strain from his cat

We wished to develop a cost-effective, rapid strategy to detect and identify Bartonella species in the clinical laboratory and to determine the prevalence of Bartonella infection in the Houston veteran population. Bartonella colonies were identified by colony morphology, Gram stain, RapID ANA, repetitive extragenic palindromic-PCR (REP-PCR) and whole-cell fatty acid (CFA) analysis, and these methods were compared for their usefulness. A new test order for "Rochalimaea culture" (the genus Bartonella was previously known as the genus Rochalimaea) was instituted, and in addition, all blood specimens submitted for fungal culture (obtained in an isolator tube) were processed for Bartonella culture. Over a 16-month period we isolated Bartonella henselae from only 0.4% (2 of 533) of total cultures but from 1% (2 of 204) of human immunodeficiency virus-positive patients. After sufficient growth, identification of the Bartonella isolates to the species level could be obtained in 2 days. The REP-PCR allowed discrimination of all known species, whereas CFA analysis distinguished all except B. henselae and Bartonella quintana. The RapID ANA results failed to differentiate between B. henselae and B. quintana, and results for other species differed by only one or two tests. Blood obtained from a kitten which had been introduced into the household of one patient 2 months before the onset of fever yielded a Bartonella strain which was shown to be different from the strain from the patient and distinct from other Bartonella species by a combination of REP-PCR, CFA, and growth characteristics. Subsequent analysis of the citrate synthase gene sequence showed only an 86% similarity with any of the other known Bartonella species, suggesting that this isolate represents a distinct, previously uncharacterized species of Bartonella.

Cutaneous manifestations of opportunistic infections in patients infected with human immunodeficiency virus

Bacillary angiomatosis (BA) presents most commonly as a cutaneous disease and is caused by two organisms. Bartonella (Rochalimaea) henselae and Bartonella (Rochalimaea) quintana. Biopsy confirmation of cutaneous BA is essential because lesions can mimic nodular Kaposi's sarcoma in appearance. Although the vast majority of human immunodeficiency virus (HIV)-infected patients with BA have CD4 lymphocyte counts of less than 100 cells per mm3, the disease responds well to antimicrobial therapy. Staphylococcus aureus is the most common bacterial skin pathogen affecting HIV-infected patients. The prevalence of skin disease due to S. aureus may be explained by high nasal carriage rates for the organism ( > or = 50%) and altered immune function in conjunction with an impaired cutaneous barrier. Herpes simplex virus causes mucocutaneous disease early in the course HIV infection and ulcerative lesions at any site in advanced HIV infection. Herpes zoster is common early in the course of HIV infection; recurrent and disseminated herpes zoster infections are characteristic of patients with advanced HIV disease. Acyclovir resistance is usually seen in patients with large, untreated, ulcerative lesions of herpes simplex virus and in patients with chronic, verrucous lesions of varicella-zoster virus. Cutaneous cryptococcosis, histoplasmosis, and coccidiomycosis are markers of disseminated disease and require biopsy confirmation. Scabies is easily diagnosed but may be atypical in presentation and difficult to eradicate in advanced HIV disease.

Cat-scratch disease in Hawaii: etiology and seroepidemiology

OBJECTIVE

To study the etiology and seroepidemiology of cat-scratch disease (CSD) in Hawaii.

METHODS

Blood and fine-needle aspirate (FNA) from the lymph nodes of 39 consecutive patients with clinical CSD were cultured for Bartonella henselae, and blood samples from index cats, stray cats, and dogs were cultured and their sera were tested by indirect fluorescence antibody test for antibodies to B. henselae and Afipia felis. Sera from age- and sex-matched human subjects without cat exposure served as controls.

RESULTS

Warthin-Starry staining showed positive results in only 4 of 32 FNAs, and B. henselae was isolated from only one FNA specimen. All of 38 patients who had two or more sera tested had elevated titers of antibody to B. henselae. Only 1 of 48 human control sera had antibody to B. henselae. Of 31 kittens, 21 had positive blood culture results and elevated antibody titers to B. henselae. Of three adult cats, all had negative blood culture results, but they had serologic evidence of past infection. Of 23 adult stray cats, 18 had elevated titers of antibody to B. henselae, but in only one was the blood culture result positive. Results of IFA tests were marginally positive for A. felis in 1 of 29 patients with CSD and in one adult stray cat and one dog.

CONCLUSIONS

This study shows that the B. henselae IFA test is both highly sensitive and specific for the detection of infection caused by B. henselae and for the laboratory diagnosis of CSD, and that FNA is seldom helpful in confirming the diagnosis. We further demonstrated that CSD in Hawaii is due to B. henselae and that infection is directly linked to the scratch or bite of a kitten. Older cats seldom have bacteremia but often have serologic evidence of past infection. Our study fails to implicate dogs in the epidemiology of CSD in Hawaii, and A. felis was not etiologically implicated in CSD in the human subjects and animals we studied.

Differentiation of Bartonella-like isolates at the species level by PCR-restriction fragment length polymorphism in the citrate synthase gene

The citrate synthase gene (gltA) of Bartonella henselae was cloned and sequenced to compare genetic divergence among alpha and gamma branches of the class Proteobacteria and to develop enhanced genotypic reagents for B. henselae identification. B. henselae gltA is 1,293 nucleotides in length and 63 to 66% homologous with corresponding gene sequences of Rickettsia prowazekii, Escherichia coli, and Coxiella burnetii. The observed genetic variability suggests that gltA sequences can provide a useful means for studying moderate divergence among related bacteria. Oligonucleotides specific for B. henselae gltA were evaluated for the ability to prime PCR amplification within the alpha and gamma branches of the proteobacteria. Under the conditions used, only B. henselae, Bartonella quintana, and R. prowazekii template DNAs yielded amplification products (approximately 380 bp). DNAs from 28 Bartonella-like isolates of feline origin were amplified by B. henselae primers and analyzed for restriction fragment length polymorphism. The resulting patterns for all 28 isolates were similar or identical to that of the recognized B. henselae strain. Current studies are aimed at optimization of PCR conditions for specificity and sensitivity of amplification of Bartonella sequences from clinical isolates.

In vitro evaluation of the role of humoral immunity against Bartonella henselae

The contribution of humoral immunity against Bartonella henselae was evaluated by examining the in vitro bactericidal activity of sera and the ability of these microorganisms to activate complement and stimulate phagocytosis and an oxidative burst in polymorphonuclear leukocytes. The organism was killed by complement-mediated cytolysis. Complement activation preferentially proceeded by the alternative pathway. The presence of specific antibodies did not increase the serum bactericidal activity or complement activation. However, phagocytosis and the subsequent production of oxygen radicals, evaluated by flow cytometry, were significantly enhanced in the presence of bacteria previously opsonized with immune sera.

Inter- and intraspecies identification of Bartonella (Rochalimaea) species

Species of the genus Rochalimaea, recently renamed Bartonella, are of a growing medical interest. Bartonella quintana was reported as the cause of trench fever, endocarditis, and bacillary angiomatosis. B. henselae has been implicated in symptoms and infections of human immunodeficiency virus-infected patients, such as fever, endocarditis, and bacillary angiomatosis, and is involved in the etiology of cat scratch disease. Such a wide spectrum of infections makes it necessary to obtain an intraspecies identification tool in order to perform epidemiological studies. B. vinsonii, B. elizabethae, seven isolates of B. quintana, and four isolates of B. henselae were studied by pulsed-field gel electrophoresis (PFGE) after restriction with the infrequently cutting endonucleases NotI, EagI, and SmaI. Specific profiles were obtained for each of the four Bartonella species. Comparison of genomic fingerprints of isolates of the same species showed polymorphism in DNA restriction patterns, and a specific profile was obtained for each isolate. A phylogenetic analysis of the B. quintana isolates was obtained by using the Dice coefficient, UPGMA (unweighted pair-group method of arithmetic averages), and Package Philip programming. Amplification by PCR and subsequent sequencing using an automated laser fluorescent DNA sequencer (Pharmacia) was performed on the intergenic spacer region (ITS) between the 16 and 23S rRNA genes. It was found that each B. henselae isolate had a specific sequence, while the B. quintana isolates fell into only two groups. When endonuclease restriction analysis of the ITS PCR product was done, three enzymes, TaqI, HindIII, and HaeIII, allowed species identification of Bartonella spp. Restriction fragment length polymorphism after PCR amplification of the 16S-23S rRNA gene ITS may be useful for rapid species identification, and PFGE could be an efficient method for isolate identification.

Bacillary Angiomatosis: A Unique Cutaneous Complication of Hiv Infection

As the prevalence of human immunodeficiency virus (HIV) infection increases, so does the recognition of unusual manifestations of this syndrome. This study describes two patients who presented with a unique, rare, cutaneous manifestion of HIV infection. One patient developed an aggressive atypical cellulitis and ascending lymphangitis of the hand, which failed to improve following multiple courses of several different antibiotics. The other patient presented with multiple, vascular, nodular lesions, clinically resembling Kaposi's sarcoma. Routine biopsies and wound cultures were not helpful for diagnosis. A definitive diagnosis of bacillary angiomatosis (BA) was made using a special silver-staining (Warthin-Starry) histological technique and electron microscopy. Both patients responded completely to oral erythromycin therapy. BA is a newly recognized bacterial infection caused by bacteria of the genus Bartonella. It is seen primarily in patients with acquired immunodeficiency syndrome (AIDS), and less commonly in patients with other immunosuppressed conditions. The term BA describes the lesion's pathogenesis (infective-bacillary) and clinico-histology (angiomatosis). The causal organism cannot be cultured reliably and is resistant to most antibiotics. This study reviews these two patients, the differential diagnosis of BA, specific confirmatory tests used in diagnosis, and the treatment outcome of this condition. It is important for the plastic surgeon to understand BA, because it can resemble other common presentations, but it can also be associated with internal involvement leading to mortality. However, when recognized, it can usually be cured by oral erythromycin therapy (500 mg qid) for several weeks.

Seroepidemiological survey of Bartonella (Rochalimaea) henselae in domestic cats in Japan

A total of 199 domestic cat serum samples from 3 geographical areas (northeastern, central and southwestern) of Japan collected between 1992 and 1994 were examined for serum antibody against Bartonella henselae using an immunofluorescent assay. The antibody prevalence was 15.1% (30/199). A significant difference in the prevalence of B. henselae antibody was observed between the northeastern area (6.3%:3/48) and the central area (22.0:13/59) in Japan. There was no significant difference between the average age of seropositive cats (4.39 +/- 3.26 years) and that of seronegative cats (4.03 +/- 3.84 years), and also between the frequency of seropositive male cats (16.5%:15/91) and that of seropositive female cats (11.8:9/76). This is the first report of B. henselae antibodies in cats in Japan

Genomic fingerprinting of Bartonella species by repetitive element PCR for distinguishing species and isolates

Repetitive-element PCR (rep-PCR) with primers based on repetitive extragenic palindromic (REP) and enterobacterial repetitive intergenic consensus (ERIC) repeated DNA sequences was used for genomic finger-printing of Bartonella species. This technique was applied by using either extracted genomic DNA or preparations of whole bacterial cells directly. PCR fingerprints with either the REP-based primers (REP-PCR) or primers based on the ERIC repeat (ERIC-PCR) revealed species-specific band patterns for the various Bartonella isolates. DNA fingerprints obtained from rep-PCR of extracted genomic DNA or from preparations of whole cells yielded comparable patterns. ERIC-PCR banding patterns were less complex than those obtained by REP-PCR but allowed better discrimination between strains within species. By combining results of REP-PCR and ERIC-PCR, five different fingerprint profiles were identified among 17 isolates of Bartonella henselae, but only one profile was identified among the five isolates of Bartonella quintana. Other Bartonella species yielded distinct rep-PCR fingerprints. rep-PCR is a useful technique for identification of Bartonella organisms to the species level and offers the advantage of ease of performance, with only small quantities of cells needed for the whole-cell procedure. This technique also appears to be useful for subtyping B. henselae isolates.

A chemically defined liquid medium that supports primary isolation of Rochalimaea (Bartonella) henselae from blood and tissue specimens

Rochalimaea (Bartonella) henselae is a fastidious, slowly growing, gram-negative bacillus that is an etiologic agent of bacillary angiomatosis, cat scratch disease, and related syndromes. Accumulation of direct microbiologic evidence of the relationship between the organism and the syndromes compatible with cat scratch disease has been hindered by the difficulties in the primary isolation of the organism from infected tissue specimens. A chemically defined liquid medium was developed to support the growth of Rochalimaea species to facilitate study of the organism. This medium was also used successfully to isolate R. henselae from clinical specimens from infected patients and a domestic cat. Recovery of R. henselae in this was more successful than when recovery was attempted on solid agar. This cell-free, extract-free, defined medium additionally supported the growth of Rochalimaea quintana and Afipia felis.

Investigations of culture and properties of Afipia spp

The culture conditions of Afipia felis, A. broomeae, A. clevelandensis and three unnamed Afipia genospecies were investigated on BCY agar supplemented with different substances known as growth factors of Legionella spp. and, furthermore, with sodium chloride and other salts. The organisms were found to be susceptible to a certain degree to byproducts of the autoclaving which are scavenged by activated by charcoal. Growth was weakly enhanced by ferric pyrophosphate, cystein.HCl, and alpha-ketoglutarate. These substances are no obligatory growth factors. The optimal pH value was about 6.8. Afipia spp. showed a strong susceptibility to NaCl and other salts. They possess phosphatase, phosphoamidase, phosphodiesterase, a weak sulfatase, glycine aminopeptidase, and L-lysine aminopeptidase. The strains differed with regard to other proteases and aminopeptidases. The decimal reduction times of A. felis at 55 degrees C and 60 degrees C were 11 min, < 1 min, respectively.

Bacillary angiomatosis with cutaneous and oral lesions in an HIV-infected patient from the U.K

We describe an HIV positive patient with histologically confirmed cutaneous and oral lesions of bacillary angiomatosis, an unusual clinical presentation. As far as we are aware, this is the first reported case of cutaneous bacillary angiomatosis in the U.K.

An analysis of lymph node DNA for possible bacterial agents of cat-scratch disease

Recent investigations have implicated Afipia felis and Rochalimaea henselae as possible agents of cat-scratch disease (CSD). We studied lymph nodes with necrotizing granulomas characteristic of CSD for A. felis and R. henselae DNA so that the relationship of these organisms to lymph nodes with necrotizing granulomas of unknown etiology might be better defined. We examined formalin-fixed, paraffin-embedded lymph node biopsies with necrotizing granulomas suggestive of CSD from 28 children obtained over the last 10 years. None had identifiable bacteria, fungi, or acid-fast organisms on routine staining. Pleomorphic bacillary structures consistent with the CSD bacillus were seen with the Steiner stain in 17 cases. We performed the polymerase chain reaction (PCR) on the extracted lymph node DNA with DNA primers for these organisms after demonstrating the presence of amplifiable DNA with c-K-Ras primers. R. henselae was identified in two samples. A. felis DNA was found in just one specimen. These putative CSD bacteria are infrequently associated with necrotizing granulomas using standard PCR techniques. It is possible that some of the patients did not have clinical CSD. The preservation of DNA or numbers of bacteria in the extracted sections may be inadequate for demonstration by DNA amplification methods. These bacilli may be responsible for a small proportion of these characteristic lesions of unknown etiology, or the typical CSD histology, including the presence of pleomorphic bacillary structures, may be nonspecific.

Endocarditis in a dog due to infection with a novel Bartonella subspecies

Vegetative valvular endocarditis involving the aortic and, to a lesser extent, mitral valves was diagnosed echocardiographically in a 3-year-old spayed female Labrador retriever. Historically, the dog had been treated with tetracycline hydrochloride and prednisolone for positive seroreactivity to Ehrlichia canis and antinuclear antigens. Although three aerobic and anaerobic blood cultures failed to grow bacteria, blood cultured simultaneously by the lysis centrifugation technique grew a fastidious, gram-negative organism. Despite an initial therapeutic response, the owner elected euthanasia 17 days later. Necropsy confirmed aortic and mitral valvular endocarditis. Bacteria phenotypically similar to Bartonella species were visualized in the heart valve by light and electron microscopy, and Bartonella DNA from a frozen heart valve was amplified by PCR. Subsequent phenotypic and genotypic characterization of the isolate, including biochemical testing, cellular fatty acid analysis, DNA hybridization, and sequencing of the 16S rRNA gene indicated that this organism, which can induce endocarditis in dogs, is a novel Bartonella subspecies containing an insertion sequence unique among currently recognized Bartonella species. The name Bartonella vinsonii subsp. berkoffii subsp. nov. will be proposed for this organism.