Published online April 28, 2008
PEDIATRICS Vol. 121 No. 5 May 2008, pp. e1413-e1425 (doi:10.1542/peds.2007-1897)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via CrossRef
Google Scholar
Right arrow Articles by Florin, T. A.
Right arrow Articles by Zaoutis, L. B.
PubMed
Right arrow PubMed Citation
Right arrow Articles by Florin, T. A.
Right arrow Articles by Zaoutis, L. B.
Related Collections
Right arrow Infectious Disease & Immunity

STATE-OF-THE-ART REVIEW ARTICLE

Beyond Cat Scratch Disease: Widening Spectrum of Bartonella henselae Infection

Todd A. Florin, MDa,b, Theoklis E. Zaoutis, MD, MSCEb,c,d,e and Lisa B. Zaoutis, MDa,b

a Divisions of General Pediatrics
c Infectious Diseases, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania
b Departments of Pediatrics
d Epidemiology
e Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania

ABSTRACT

Bartonella henselae was discovered a quarter of a century ago as the causative agent of cat scratch disease, a clinical entity described in the literature for more than half a century. As diagnostic techniques improve, our knowledge of the spectrum of clinical disease resulting from infection with Bartonella is expanding. This review summarizes current knowledge regarding the microbiology, clinical manifestations, diagnostic techniques, and treatment of B henselae infection.


Key Words: Bartonella • cat scratch disease • lymphadenopathy • fever of unknown origin • hepatosplenic disease

Abbreviations: CSD—cat scratch disease • VEGF—vascular endothelial growth factor • FUO—fever of unknown origin • CT—computed tomography • CSF—cerebrospinal fluid • Ig—immunoglobulin • PCR—polymerase chain reaction • IFA—indirect fluorescence assay • EIA—enzyme immunoassay

Cat scratch disease (CSD) has been reported in the literature for more than half a century as a syndrome of regional lymphadenopathy and fever. However, it has been only a quarter of a century since Bartonella henselae was identified as the etiologic agent. As diagnostic techniques have improved, Bartonella has been found to be responsible for a broad range of clinical syndromes, particularly prolonged fever of unknown origin (FUO), hepatosplenic disease, encephalopathy, and ocular disease. Although other Bartonella species can cause human disease, this review will focus on those caused by B henselae.

HISTORICAL CONTEXT

The clinical syndrome of CSD was first reported in 1950 by Debré et al1, although Parinaud2 described similar symptoms in the context of oculoglandular syndrome in 1889. Despite numerous reports and studies of CSD, the causative agent eluded detection until 1983. At that time, Wear et al3 discovered a small, pleomorphic Gram-negative bacillus by using a Warthin-Starry silver stain in infected lymph nodes of patients with CSD. It was not until 5 years later that this organism was successfully isolated and cultured.4 In 1991, Brenner et al5 named the CSD bacillus Afipia felis, after the Armed Forces Institute of Pathology, where the organism was discovered. In 1992, Rochalimaea henselae was isolated from HIV-infected patients with bacillary angiomatosis, peliosis hepatis, and fever syndromes. In that report, Regnery et al6 noted that the majority of their patients with clinically suspected CSD had high serum titers to the R henselae antigen. Additional studies in the 1990s refuted the role of A felis in CSD, in favor of Rochalimaea species.7,8 In 1993, the genera Bartonella and Rochalimaea were united, with Bartonella having nomenclatural precedence over Rochalimaea.9 Thus, B henselae is currently recognized as the causative agent of CSD. Since that time, the most significant study of patients with B henselae infection has been undertaken by Hugh Carithers,10 who saw and reported >1200 cases of CSD in pediatric private practice.

MICROBIOLOGIC FEATURES AND PATHOGENESIS

The genus Bartonella includes 19 distinct species, of which at least 6 are responsible for human disease (B henselae, Bartonella bacilliformis, Bartonella quintana, Bartonella elizabethae, Bartonella vinsonii, Bartonella koehlerae). These species are small, fastidious, intracellular Gram-negative bacilli that are aerobic and oxidase-negative. The organisms are most easily visualized by using a Warthin-Starry silver impregnation stain (see Fig 1) or a Brown-Hopps tissue Gram-stain. Two main genogroups of B henselae have been identified in humans and cats: Houston-1 and Marseille (also known as genotype II).11 These 2 genogroups are further subdivided into 4 variants: Marseille, CAL-1, Houston-1, and ZF-1.12 In infected patients, the organisms are found most commonly in vessel walls, in macrophages lining the sinuses of lymph nodes, in nodal germinal centers, in nonnecrotic areas of inflammation, and in areas of expanding and suppurating necrosis.13,14 Electron microscopy of lymph node tissues of patients with CSD confirms that the bacilli have an affinity for the vascular endothelium, with organisms seen in clumps in vessel walls, intracellularly and free in necrotic debris.15


Figure 1
View larger version (135K):
[in this window]
[in a new window]

 
FIGURE 1 B henselae seen as short rods by using Warthin-Starry silver stain.137

 
Cats are the major reservoir for B henselae, with up to half of domestic cats having antibodies to B henselae, thus testing seropositive for the bacteria. Direct horizontal transfer of B henselae does not occur, but rather, spread of infection between cats depends on the arthropod vector Ctenocephalides felis, or the cat flea.16 After transmission, the number of bacteria reach high levels in the feline host as a result of being intraerythrocytic parasites, thereby evading the host immune response. Once transmitted to humans via cat saliva or the scratch of a cat, B henselae invades CD34+ hematopoietic progenitor cells instead of human erythrocytes directly.17 Bacterial infection does not affect erythroid differentiation of hematopoietic progenitor cells; thus, infection of these progenitor cells results in intracellular presence and replication of B henselae in erythroid differentiated cells.17

The response to infection with B henselae depends on the immune status of the infected host. In immunocompetent individuals, the response is granulomatous and suppurative, as compared with a vasoproliferative response in immunocompromised patients.14 Early in the course of infection in an immunocompetent patient, lymphoid hyperplasia, arteriolar proliferation, and widened arteriolar walls are seen in biopsied lymph nodes. This progresses to granulomatous disease, with central areas of necrosis and multinucleated giant cells. Bartonella infection causes an interferon-{gamma}-mediated T helper 1 cell response, resulting in macrophage recruitment and stimulation, ultimately producing granulomatous disease.18 Late in the disease, stellate microabscesses form with suppuration of affected lymph nodes.14 In individuals with an intact immune system, infection generally remains within the lymphatics, with a symptomatic immune response that lasts 2 to 4 months.19

Immunodeficient patients are at risk for bacillary angiomatosis, which manifests as cutaneous angiogenic lesions. These lesions consist of vascular proliferation composed of endothelial cells and a mixed inflammatory cell infiltrate. The mechanism by which B henselae induces angiogenesis is not fully understood. One hypothesis is that Bartonella modulates host or target cell cytokines and growth factors, which lead to angiogenesis. When Bartonella adheres to or is phagocytosed by macrophages, these cells secrete vascular endothelial growth factor (VEGF). It is thought that Bartonella adhesin A is crucial for the secretion of VEGF and other proangiogenic cytokines.20 VEGF is thought to act as an endothelial cell inducer, leading to proliferation of endothelial cells and angiogenesis.19 Another hypothesis involves Bartonella directly triggering proliferation and apoptosis of endothelial cells, resulting in increased angiogenesis.21

EPIDEMIOLOGICAL FEATURES OF B henselae

B henselae has a worldwide distribution, with cases of classic Bartonella infection reported in the United States,10 Europe,22 Japan,23 New Zealand,24 and Australia.25 In the United States, there seems to be a seasonal distribution, with the majority of cases occurring between the months of July and January.10,26 Peak hospitalizations for CSD in 2000 were in October, with most hospitalizations occurring between July and October.27 Some authors have attributed this seasonal variation to the temporal breeding patterns of domestic cats, the acquisition of kittens as family pets, and the peak temporal presence of the cat flea, the major mode of Bartonella transmission among cats.28 Seroprevalence of antibodies in humans to B henselae and B henselae bacteremia was found to be highest in regions with warm, humid climates.29 One study concluded that in the United States, incidence in humans is higher in the south and lower in the west compared with the nation as a whole.26 Among felines, kittens, outdoor cats, and cats infested with fleas are more likely to be seropositive to B henselae.3032 Overall seroprevalence in cats in the United States was found to be between 28% and 51%.33,34 Of note is that culture-positive felines rarely seem sick, and cannot be clinically distinguished from those without B henselae. Although more common among felines, it was recently discovered that 10.1% of healthy dogs and 27.2% of sick dogs in the southeastern United States were found to have antibodies to B henselae.35 It is unclear at this time if the presence of Bartonella in canines has clinical significance.

The true incidence of Bartonella infection is difficult to establish, because it is not a reportable disease in a majority of states in the United States. An analysis of 3 national databases found the incidence of patients discharged from the hospital with a diagnosis of CSD to be between 0.77 and 0.86 per 100 000 per year.26 Incidence in this analysis was defined by number of patients discharged from the hospital per year with a listed diagnosis of CSD. This finding likely underestimates the true incidence, as most cases of Bartonella infection are not recognized or are treated on an outpatient basis. National CSD hospitalization rates in 2000 were found to be 0.60 per 100 000 children younger than 18 years of age and 0.86 per 100 000 children younger than 5 years of age.27 These estimates are similar to earlier estimates, despite an increase in cat ownership in the United States by ~14%.27 Clustering of cases within families has coincided with the acquisition of new pet cats, with as many as 3 siblings having clinical CSD simultaneously.36 By using skin tests on family members of patients, Carithers10 noted that there is significant asymptomatic infection, and close contact with cats increased the prevalence of positive skin-test reactions. In these family contacts, positive skin reaction occurred in 18% of the overall sample, 19% of those who were fond of cats, and 1.5% of those who disliked felines.

Bartonella infection was thought to be largely a disease of children, with studies reporting between 54% and 87% of cases of CSD in patients under 18 years of age.10,26 Recent studies have suggested that CSD may be more common in adults than previously recognized, with some studies reporting >40% of their patients being older than 20 years of age.30

CLINICAL MANIFESTATIONS

The clinical manifestations of infection with B henselae are expanding with the improved ability to recognize the presence of the organism. Some forms of infection seem to be regional, but may be on a spectrum with more systemic forms. A list of various recognizable clinical forms of B henselae infection is provided in Table 1.


View this table:
[in this window]
[in a new window]

 
TABLE 1 Clinical Manifestations of B henselae Infection

 
Typical Cat Scratch Disease
For the purposes of this review, "typical CSD" will refer to the syndrome of isolated lymphadenopathy with fever and no other signs or symptoms. Typical CSD is the most commonly recognized manifestation of infection with B henselae. Carithers'10 original series noted typical CSD in ~95% of his 1200 patients. This is likely a slight overestimate of the prevalence of typical CSD, because many of the atypical presentations were not appreciated in 1985. There has not been an extensive prevalence study recently to elucidate recent prevalence data. The disease begins with an erythematous papule at the site of inoculation. The papule appears 3 to10 days after inoculation, and progresses through erythematous, vesicular, and papular crusted stages. The lesion persists for between 1 and 3 weeks.37 Regional lymphadenopathy occurs 1 to 3 weeks after inoculation (Fig 2). Lymphadenopathy is seen in all patients with typical CSD, and 85% of patients have only a single node involved. Lymphadenopathy occurs most frequently in the axillary and epitrochlear nodes (46%), head and neck (26%), and the groin (17.5%).10 The nodal distribution reflects the fact that feline contact occurs most often with the hands. On ultrasound, nodes are multiple, hypoechoic, and highly vascularized with increased echogenicity of the surrounding soft tissues.38 On biopsy, nodes reveal granulomas with multiple microabscesses (Fig 3). Approximately 10% of nodes will suppurate, thereby requiring drainage.39 Systemic illness is mild in the majority of patients, and can include fever, generalized aches, malaise, anorexia, nausea, and abdominal pain. Of note is that <10% of patients have a fever higher than 39°C, and one-third are without fever.10


Figure 2
View larger version (94K):
[in this window]
[in a new window]

 
FIGURE 2 Typical CSD in a 10-year-old boy demonstrating a papule on the dorsum of the right hand near the base of the thumb (A) and right axillary lymphadenopathy (B). (Copyright Children's Hospital of Philadelphia. Reproduced with permission from Stephen Ludwig, MD and Walter W. Tunnessen Pediatric Image Library of Children's Hospital of Philadelphia, Philadelphia, PA.)

 

Figure 3
View larger version (81K):
[in this window]
[in a new window]

 
FIGURE 3 A, Low-power view of lymph node with suppurative granulomatous lymphadenitis. Note the normal germinal center (arrow), and compare it with the large, ovoid pale-appearing granuloma (arrowhead). (Hematoxylin and eosin staining, original magnification: x50.) B, Higher-power view of granuloma with eosinophilic epithelioid histiocytes and small lymphocytes forming the rim. The necrotic central core contains numerous neutrophils. (Hematoxylin and eosin staining, original magnification: x200. Inset, Immunohistochemical stain using mouse monoclonal antibody for B henselae organism with positive, granular brown staining of the bacterium in the central region of the granuloma. (Original magnification: x630.) (Reproduced with permission from Linda M. Ernst, MD, MHS, Department of Pathology and Laboratory Medicine, Children's Hospital of Philadelphia/University of Pennsylvania School of Medicine, Philadelphia, PA.)

 
Prolonged Fever/FUO
Although several definitions of FUO exist, a commonly accepted definition is fever that lasts for >2 weeks with no diagnostic signs or symptoms of an obvious clinical disease. Infectious etiologies dominate the long differential diagnosis for prolonged FUO, and new agents are continuously being added to this list. With the improvements in diagnostic methods for detecting B henselae, this agent is increasingly being recognized as a cause of prolonged FUO in children.40 One recent study identified B henselae as the third leading infectious cause of FUO, after Epstein-Barr virus infection and osteomyelitis.41 In this series, Bartonella accounted for 5% of the total 146 children with FUO and prolonged fever, 8% of the subset of 84 children with a confirmed diagnosis for their fever, and 11% of the subset of 64 children found to have an infectious disease. This study and other case reports in the literature reveal that a history of exposure to cats is not uniformly found among patients diagnosed with Bartonella infections, which suggests that such infections should be considered in the initial evaluation of FUO, irrespective of exposure to cats. Approximately 30% of cases of FUO caused by B henselae had hepatosplenic involvement. There have also been reported cases of B henselae presenting with intraabdominal lymphadenopathy, fever, and abdominal pain with no hepatosplenic disease.42,43 Thus, B henselae infection should always be considered as a diagnostic possibility in patients with FUO and in patients with fever and abdominal pain.

Hepatosplenic Manifestations
Bartonella infection that involves the liver and/or spleen occurs more than previously acknowledged, and is being recognized more frequently as a result of improvements in serologic and imaging diagnostic modalities. Hepatosplenic Bartonella infection typically presents with systemic symptoms, such as prolonged fever, and microabscesses in the liver and/or spleen. Granulomatous disease in the spleen resulting from Bartonella can be severe enough to result in spontaneous splenic rupture.44 In various studies, >60% of patients with hepatosplenic infection presented with abdominal pain, usually described as episodic dull pain over the periumbilical and/or upper quadrant regions with high severity.45,46 Other presenting symptoms include weight loss, chills, headache and myalgias. More than half of all patients will present with hepatomegaly, splenomegaly, or hepatosplenomegaly on physical examination. Patients will typically have an elevated erythrocyte sedimentation rate and elevated titers of antibodies to B henselae. White blood cell and platelet counts are normal or slightly elevated in most cases. Liver enzymes are typically normal. In 1 study, all patients with hepatosplenic Bartonella infection had evidence of hepatosplenic disease by using abdominal imaging, with 68% percent of having microabscesses of both the liver and spleen.45 In 1 interesting case, ultrasound revealed thickening of the terminal ileum in addition to hypoechogenic lesions in the liver and spleen, suggesting inflammatory bowel disease resulting from B henselae infection.47

Abdominal imaging is an important diagnostic step in patients with suspected hepatosplenic disease or in cases of prolonged FUO, because it can often identify changes characteristic of hepatic or splenic Bartonella infection. On ultrasound, hepatic lesions seem hypoechoic. On computed tomographic (CT) scan, hepatic lesions seem either hypoattenuated relative to the liver, isoattenuated to the surrounding tissues, or only marginally enhanced.48 In patients who have had biopsies performed, the predominant lesion on histopathology was a necrotizing granuloma.49 In general, symptoms and visceral lesions regress within 6 months; however, there have been rare reports of residual calcification.49 It is interesting to note that only 55% of children with hepatosplenic disease had lymphadenopathy of any sort.46 Based on the portal pattern of granulomas, hepatic disease may be due to organisms transmitted via the hands by ingestion, thus explaining the low incidence of nodal involvement in abdominal disease.

Bacillary peliosis hepatis is a specific form of hepatosplenic Bartonella disease seen in the immunocompromised hosts. Patients present with gastrointestinal symptoms, fever, chills, and hepatosplenomegaly. The liver demonstrates characteristic dilated capillaries or blood-filled cavernous spaces. The typical duration of fever ranges from 1 week to 2 months.45

Ocular Manifestations
Parinaud oculoglandular syndrome, consisting of fever, regional lymphadenopathy, and follicular conjunctivitis, was first described in 1889 and is the most common ocular presentation of B henselae infection, affecting ~5% of patients with CSD.10 Only within the last decade was B henselae identified as the causative agent of this syndrome.50 Route of infection is thought to be direct conjunctival inoculation. Typical symptoms include foreign body sensation, unilateral eye redness, serous discharge, and increased tear production. On examination, patients present with a necrotic granuloma with ulceration of the conjunctival epithelium and regional lymphadenopathy that affects the preauricular, submandibular, or cervical lymph nodes.51 The granuloma typically disappears after several weeks without scarring.10

Neuroretinitis, a form of optic neuropathy with optic disk swelling and macular stellate exudate, is the most common posterior segment ocular complication of Bartonella infection.51 B henselae is the most common identified etiology of neuroretinitis, with approximately two thirds of patients with neuroretinitis demonstrating serologic evidence of previous B henselae infection.52 Symptoms include painless visual loss with abrupt onset that is typically unilateral.53 On MRI, unilateral enhancement at the optic nerve-globe junction is highly specific for B henselae infection as cause for optic neuropathy.54 Macular exudates may take months to resolve and, even after resolution, patients may experience abnormal color vision and evoked potentials, subnormal contrast sensitivity, residual disk pallor, afferent pupillary defects, retinal pigment changes, and mildly decreased visual acuity.55 There are reports of ocular Bartonella disease with optic disk edema and retinal detachment without the classic macular stellate exudate seen with neuroretinitis.56 Other posterior segment presentations of B henselae infection include panuveitis with diffuse choroidal thickening, retinal vasoproliferative lesions, macular hole, vitreal detachment, vitritis, branch retinal artery and venous occlusions, retinal white spots, and papillitis.5760 In HIV-positive patients, ocular B henselae infection presents as a subretinal mass associated with abnormal vascular network, which is best diagnosed by fluorescein angiography.61

Neurologic Manifestations
Neurologic complications of B henselae infection are rare, occurring in ~2% of infected patients. The most common presentation is encephalopathy, accounting for 90% of cases that affect the nervous system.10 Neurologic symptoms generally occur 2 to 3 weeks after the onset of lymphadenopathy. Symptoms include headaches and mental status changes. Seizures develop in 46% to 80% of patients with Bartonella encephalopathy, with some presenting with status epilepticus.6264 Combative behavior has been reported in as many as 40% of patients with Bartonella encephalopathy.63,65 In addition to mental status changes, patients with encephalopathy may present with a variety of neurologic findings, including weakness, alterations in tone, nuchal rigidity, extensor plantar responses, and hyporeflexia or hyperreflexia.66

Outside of specific identification of B henselae infection, laboratory evaluation of infected patients with encephalopathy generally yields variable results, and is not helpful in diagnosis. Cerebrospinal fluid (CSF) analysis typically yields normal results, although pleocytosis and elevated CSF protein have been reported.62,64 Electroencephalography performed during the acute phase of illness reveals generalized slowing in 80% of patients, with complete normalization on follow-up.63 Only 19% of patients have abnormal findings on CT scan or MRI of the brain, and these include lesions of the cerebral white matter, basal ganglia, thalamus, and gray matter.67 Prognosis is generally excellent for patients with encephalopathy, with >90% of patients having complete, spontaneous recovery with no sequelae.62,63 The published literature reveals only 1 report of fatal meningitis and encephalitis of an immunocompetent child as a result of B henselae infection.68

Less common neurologic complications include meningomyeloradiculopathy, manifesting with lower extremity paresthesias, weakness and sphincter dysfunction,69 facial nerve palsy,70,71 Guillain-Barre Syndrome,72 epilepsia partialis continua,73,74 acute hemiplegia,75 transverse myelitis,76 and cerebral arteritis.77

Dermatologic Manifestations
Skin lesions other than the papule seen at the site of inoculation are rare, occurring in ~5% of patients infected with B henselae. These consist of maculopapular and urticarial eruptions, granuloma annulare, erythema nodosum, erythema marginatum, and leukocytoclastic vascultis.78

Bacillary angiomatosis, once a common dermatologic condition of AIDS patients with Bartonella infection, is now diminishing in incidence as the use of prophylactic drugs increases in immunocompromised patients. Although this systemic disease can occur in various organ systems, skin lesions are most frequent, occurring in up to 90% of cases. Lesions are reddish-brown papules that are difficult to differentiate from Kaposi's sarcoma, epithelioid hemangioma, and pyogenic granuloma. An example of angioproliferation in immunocompromised individuals infected with B henselae, there is an accumulation of rounded blood vessels on biopsy, with plump epithelial cells and a mixed inflammatory infiltrate with neutrophil predominance.37

Hematologic Manifestations
Hematologic complications of B henselae are rare. Hemolytic anemia has been reported in both adults and children.79,80 In children, there have been several cases in the literature of Bartonella resulting in thrombocytopenic purpura.81 Bartonella has also been reported to be associated with development of lupus anticoagulant and prolongation of the activated partial thromboplastin time.82 Anecdotally, a case is known of a red blood cell enzyme deficiency with chronic hemolysis that worsened intensely with the development of systemic B henselae infection, requiring transfusion therapy. This patient improved, and no longer required transfusion therapy, within 24 hours after treatment with gentamicin.

Orthopedic Manifestations
Bone lesions are a rare complication of infection with B henselae. Often, these lesions are osteolytic, and occur as an osteomyelitis. Clinical manifestations of bony disease include pain and tenderness over the affected bone and lymphadenopathy. The lytic lesions frequently occur in the context of systemic manifestations of Bartonella infection. Lymphadenopathy frequently occurs distant from the site of osteomyelitis, suggesting that bony infection occurs by hematogenous or lymphatic spread.83 Abnormalities on radiograph include lytic lesions, with occasional sclerosis or periosteal reaction. Lesions are sometime subtle on plain radiograph, and may require an MRI or radionuclide bone scan for diagnosis.84,85 In most patients, osteolytic disease is isolated to 1 bone. Vertebral infection has been most commonly described; however, infection has been reported in the skull, sternum,85 vertebrae,83 clavicles, humerus,86 femur, tibia,85 acetabulum,87 metacarpals, and metatarsals.88 Despite the propensity for B henselae to cause isolated bony disease, a recent case series reports 2 cases of multifocal bone marrow infection with Bartonella, with foci of increased MRI T2 signal intensity in the marrow of the sacrum, iliums, and femurs, with lesions in the hepatic parenchyma.85 Biopsy reveals necrotizing granulomas of bone.86 Bony lesions have been associated with adjacent abscesses.8991 Patients with osteomyelitis resulting from B henselae infection generally have an excellent prognosis.

A recent study published in 200592 revealed that ~3% of cases of B henselae infection in Israel had rheumatoid factor-negative arthritis/arthragia. Female gender, age of >20 years, and erythema nodosum were factors significantly associated with arthropathy in patients infected with Bartonella. The most frequently affected joints were the knee, wrist, ankle, and elbow joints. Often, the disease is severe enough to incapacitate and limit activities of daily living. In most patients, arthropathy began within 1 week of the appearance of lymphadenopathy and persisted for greater duration than the lymphadenopathy (13 weeks vs 9 weeks, median).

Cardiac Manifestations
The most commonly reported cardiac manifestation of Bartonella infection is endocarditis. Typically, this presentation is seen in adult males; however, it can occur in children, especially those with previous valvular disease. Bartonella species account for ~3% of cases of endocarditis.93 Presentation is insidious and subacute, with fever, dyspnea, bibasilar rales, cardiac failure, and cardiac murmur as presenting signs and symptoms. The aortic valve is usually involved, and vegetations are found in 100% of patients.

Renal Manifestations
Renal complications of Bartonella infection are uncommon, with glomerulonephritis being the most frequently encountered. Glomerulonephritis secondary to B henselae presents with gross or microscopic hematuria, low-grade proteinuria, and cola-colored urine, often accompanied by fever and lymphadenopathy. The renal disease can present as immunoglobulin A (IgA) nephritis, acute postinfectious glomerulonephritis or necrotizing glomerulonephritis.9498 Affected patients have normal serum complement 3 levels, normal renal function, and renal biopsies may reveal mesangial hypercellularity, IgA deposition, interstitial infiltrate and/or complement 3 deposition consistent with acute glomerulonephritis.9496 In general, spontaneous recovery can be expected in patients with renal manifestations of B henselae infection.

Pulmonary Manifestations
In general, rare cases of pulmonary involvement in Bartonella infection take the form of pneumonia or pleural thickening and/or effusion.99 Pulmonary disease appears 1 to 5 weeks after the appearance of lymphadenopathy. Prognosis has been excellent, with complete recovery in a mean time of 2 months.

Pseudomalignancy
There have been increasing numbers of reports in the literature of B henselae infection mimicking various malignancies. Infection simulating lymphoma is one of the most frequently reported, especially with lymphadenopathy in the neck and abdomen.100,101 The clinical picture is most confusing when splenic involvement occurs in the context of the so-called "B symptoms" of lymphoma, such as weight loss, night sweats, and prolonged fever.102 Hepatosplenic lesions and intraabdominal lymphadenopathy have been noted to have an appearance on both ultrasound and contrast-enhanced CT scan consistent with lymphoma.103 An interesting case was reported of a patient with a history of T cell lymphoblastic lymphoma that presented with inguinal lymphadenopathy and had a positron emission tomography scan consistent with lymphoma relapse, but had negative pathology on nodal biopsy and positive B henselae titers.104

Recently, Bartonella has been reported to mimic posttransplant lymphoproliferative disease in children who have undergone renal transplantation. Infection presented with fever, lymphadenopathy, and/or organomegaly 2 to 4 years posttransplantation. In some of these patients, B henselae infection was associated with acute rejection episodes that were reversed with intravenous corticosteroid therapy.105

There are several reports in the literature in both adults and children of Bartonella infection presenting as a solitary mass in the breast.106108 Initial clinical manifestations consist of a firm, mobile, tender breast mass, often in the lower outer quadrant of the breast, and inflammatory axillary lymphadenopathy. Disease in the breast has also presented as mastitis with soreness and erythema of the breast.109 Characteristic features of B henselae infection of the breast are abscesses or granulomas in the breast parenchyma with bacteria in necrotic regions. Bartonella titers may be negative, but the bacteria may be detected on polymerase chain reaction (PCR) analysis of nodal aspirate.

Although not a malignant process, a recent case series also suggests an association of B henselae with Kikuchi's disease, or histiocytic necrotizing lymphadenitis, in children.110 Another unusual presentation includes a patient with a solitary soft tissue mass overlying a lytic skull lesion, which was suggestive of Histocytosis X.111 In adults, B henselae has presented similarly to pancreatic or biliary malignancy,112 pharyngeal cancer,113 and vascular neoplasms.114

DIAGNOSIS

Laboratory Findings
Other than tests targeting the identification of Bartonella, laboratory findings of Bartonella infection are often nonspecific. Infection may result in normal or mildly elevated white blood cell counts, and normal, elevated, or diminished platelet counts. As noted above, CSF examination typically yields normal results. Liver enzymes are usually normal. The erythrocyte sedimentation rate may be normal or elevated.

Diagnostic Testing
The evolution of diagnostic techniques over the past decade has made B henselae less elusive to clinicians and researchers. Diagnostic techniques have allowed clinicians to discover a multitude of clinical manifestations resulting from Bartonella infection compared with just 5 years ago. With this growing understanding of the wide range of clinical disease that can be caused by Bartonella infection, accurate diagnosis is necessary to rule out other diseases that it may mimic, including serious conditions that may require invasive, expensive, and expedient evaluations for serious pathology.

Isolation of Bartonella species in culture is difficult, requiring a 2- to 6-week incubation for primary isolation. In addition, isolating B henselae is usually unsuccessful, particularly if patients lack systemic disease.28 Nodal culture of the organism also offers poor yield, because lymphadenopathy is thought to be due to an aggravated immune response rather than direct invasion.

An early laboratory aid in detection of B henselae infection was the intradermal skin test, which relies on a delayed-type hypersensitivity reaction within 48 to 96 hours of inoculation with B henselae antigen. The test had a specificity of 99%, with minimal cross-reactivity with other organisms.10 The test was impractical, however, because different antigens had great variance in reactivity, there was concern over the safety of human-derived reagents, and there was a lack of generalized availability of the antigen. Other early diagnostic methods included histopathologic examination of affected lymph nodes. Pathology suggestive for B henselae infection includes granuloma formation, with microabscesses and follicular hyperplasia.10,28 The bacillus is difficult to see with conventional staining methods, and it was not until 1983 that the Warthin-Starry silver stain was used to identify a bacterium as the cause of CSD.3 It is now well accepted that the organism stains well with a Warthin-Starry silver impregnation stain. Despite this, histopathologic diagnosis remains impractical because of its invasive nature.

More recently, advanced diagnostic techniques such as serology and PCR have been applied to the detection of Bartonella. There have been 3 main approaches to using PCR to diagnose Bartonella infection: amplification of the 16S rRNA gene, amplification of the citrase synthase gene (gltA), and amplification of the htrA gene of B henselae. Specificity of PCR has been excellent (100% in 1 study); however, it has been lacking in sensitivity, ranging from 43% to 76%.115,116 The true comparison of various methods of PCR analysis is difficult because of differences in the PCR target, the sample type, and the clinical criteria used. PCR provides the advantages of high specificity and rapid identification.117 Pitfalls of the use of PCR include variable sensitivity and the need for highly specialized equipment and personnel.

A more practical means of laboratory diagnosis is serology for B henselae antibodies, because it avoids invasive sample collection, use of specialized equipment and techniques, and long incubation periods.28 Although the Warthin-Starry stain was the first evidence that CSD was caused by a bacterium, serology was the means of recognizing Bartonella species as the etiologic agent of CSD.6 The 2 major serologic diagnostic methods used are indirect fluorescence assay (IFA) and enzyme immunoassay (EIA). Sensitivities of these methods vary in different reports using different assays, depending on the antigen used, test procedures, and cutoff used, as noted in Table 2. The IFA is the most frequently used serologic method. The duration of serologic detection of antibodies is important in determining acute infection versus historical exposure to the bacterium. Positive IgM EIA indicates acute disease, with duration of detection of ≤3 months. The short duration of IgM antibodies makes them infrequently discovered on serology; thus, negative results do not exclude acute disease. IgG EIA titers also decrease with time, with only 25% of patients remaining seropositive after 1 year.118 In the early stages of the disease, titers to IgG and IgM may be low, requiring a second serum sample at a later date for diagnosis.119 In addition, because IgG antibodies persist for up to a year, it is difficult to diagnose active infection compared with previous infection. Disadvantages to serologic diagnosis include variable sensitivity and specificity, inability to distinguish between active versus prior infection, and lack of Bartonella species-specific antibody response, resulting in cross-reactivity.28 Despite this, serology remains the most practical diagnostic tool in the laboratory detection of B henselae infection.


View this table:
[in this window]
[in a new window]

 
TABLE 2 Summary of Serologic Testing Available for B henselae

 
Diagnostic Criteria
Ultimately, no single criterion should be considered the diagnostic gold standard, and diagnosis of B henselae infection must rely on the combination of epidemiological, serological, clinical, histologic, and bacteriologic criteria. Initial diagnosis was based on 4 primarily anamnestic and clinical criteria: contact with a cat, regional lymphadenopathy, a site of inoculation, and a positive skin test. Carithers10 developed the "Rule of Five" as a diagnostic tool in his original series. Points are given to each of the 4 criteria: 1 point for lymphadenopathy, 2 points for cat exposure, 2 points for the presence of an inoculation site, and 2 points for a positive skin test. Amassing 5 points strongly suggested CSD, 7 points made the diagnosis definite. Much of the diagnosis of B henselae infection is still considered a clinical diagnosis, with laboratory evaluation used to confirm initial suspicion. Updated criteria by Margileth in 2000120 are listed in Table 3.


View this table:
[in this window]
[in a new window]

 
TABLE 3 Diagnostic Criteria for B henselae Infection

 

TREATMENT

The therapeutic approach to Bartonella infection varies on the basis of the clinical manifestations and immune status of the patient. There is a paucity of data in the literature as to the most effective therapy in all cases of Bartonella infection, with most data presented as part of case series rather than randomized, controlled trials. There is a significant divide in the literature between in vitro efficacy of antibiotics and the ability to successfully treat in clinical practice. In vitro, Bartonella species have been found to be susceptible to a number of antimicrobial agents, including macrolides (azithromycin, clarithromycin, erythromycin), aminoglycosides, β-lactams (penicillin G, amoxicillin), expanded-spectrum cephalosporins (cefotetan, cefotaxime, ceftazidime, ceftriaxone), and trimethroprim-sulfamethoxazole, rifampin, and ciprofloxacin.121124 This broad spectrum of activity has failed to be borne out in clinical practice; for example, penicillin has a very low mean inhibitory concentration in vitro, but has no in vivo efficacy. All antibiotics tested in vitro had only bacteriostatic activity, except for aminoglycosides, which have demonstrated bactericidal activity against Bartonella in vitro.123,125 This lack of bactericidal activity and the lack of cell membrane penetration of many antibiotics are 2 hypotheses as to why these agents fail to reach the intracellular Bartonella bacillus.126

Typical CSD is a self-limited illness that resolves within 2 to 6 months, and usually does not respond to therapy. Most studies show no benefit to antibiotic therapy in CSD, but 2 studies have revealed some in vivo effect. A retrospective study by Margileth127 of 268 patients with CSD revealed that mean duration of illness was 14.5 weeks in patients not treated, or those treated with antibiotics found to be ineffective against CSD. Mean duration of illness after treatment was 2.8 weeks in patients treated with antibiotics that the study found effective: rifampin, ciprofloxacin, gentamicin, and trimethoprim-sulfamethoxazole, in order of increasing effectiveness. Efficacy for these antibiotics ranged from 58% to 87%. There has been a single randomized, control trial of antibiotic therapy in typical CSD that used azithromycin.128 This study revealed an 80% decrease in lymph node volume in 50% of the azithromycin-treated patients compared with 7% of the placebo-treated patients in the first 30 days. Aside from lymph node volume, there was no difference in clinical outcome between study groups, and there was no efficacy demonstrated for disseminated disease.128 Because of the natural history of uncomplicated CSD, and the risk of adverse effects of antibiotics and the evolution of resistant flora, antibiotics are not suggested for regional CSD. For mild-to-moderate infections in immunocompetent patients, management consists of reassurance, adequate follow-up and analgesics for pain. Nodes should be aspirated if they suppurate to relieve painful adenopathy; however, incision and drainage is not recommended because of the potential of chronic sinus tract formation.127 During aspiration, the needle should be moved around in several different locations, because coalesced microabscesses often exist in multiple septated pockets. For patients with significant lymphadenopathy, treatment with azithromycin at doses of 10 mg/kg on day 1 and 5 mg/kg per day on days 2 to 5 can be considered. Other antibiotic options anecdotally shown to be efficacious include rifampin (20 mg/kg per day divided in 2 doses for 2–3 weeks), ciprofloxacin (20–30 mg/kg per day in 2 daily doses for 2–3 weeks), or trimethoprim-sulfamethoxazole (10 mg trimethoprim/kg per day in 2–3 daily doses for 7–10 days).120

As the clinical spectrum of disease caused by B henselae expands, choosing the proper treatment of these conditions becomes more difficult. The current knowledge of the treatment of neuroretinitis, encephalopathy, hepatosplenic disease, endocarditis, and bacillary angiomatosis and other disease processes is derived from observational case studies rather than randomized trials. In many of these complicated Bartonella infections, antibiotic choice is based on the fact that immunocompromised individuals show a dramatic response to antibiotics compared with the minimal response of immunocompetent patients; thus, seriously ill immunocompetent individuals are treated with similar regimens despite the lack of data.51 The lack of data are even more prevalent in the pediatric population; thus, pediatricians must use available adult data and the individual clinical situation to tailor therapy to children with complicated B henselae infections, preferably in consultation with a pediatric infectious diseases specialist.

For neuroretinitis, doxycycline is the preferred drug because of its excellent intraocular and central nervous system penetration. For children <8 years of age in whom tooth discoloration is a concern, erythromycin may be substituted for doxycycline.51 When coupled with rifampin, these antibiotics seem to promote disease resolution, improve visual acuity, decrease optic disk edema, and decrease the duration of disease.55 Duration of treatment is at least 2 to 4 weeks in immunocompetent patients and 4 months for immunocompromised ones. Some authors suggest that Bartonella neuroretinitis is a self-limited disease with excellent prognosis for complete visual recovery; therefore, no antibiotic therapy is necessary and conservative management will suffice.129

There have been no randomized, controlled trials of antibiotics in Bartonella encephalopathy, and their efficacy is controversial; thus, conservative, symptomatic treatment is usually recommended.66 If antibiotics were to be used, the combination of doxycycline and rifampin is suggested because of their strong penetration into the central nervous system.126 For encephalitis with seizures, anticonvulsant therapy should be used to control seizure activity.

In hepatosplenic disease in the immunocompetent patient, gentamicin, trimethoprim-sulfamethoxazole, rifampin, and ciprofloxacin have anecdotally been shown to be effective.39,45 In patients treated with gentamicin, trimethoprim-sulfamethoxazole, or rifampin, defervescence occurred between 1 and 5 days after the start of therapy.45 Because of the variety of antibiotic regimens and study parameters, it is difficult to determine if any single antibiotic regimen is superior in the treatment of hepatosplenic disease. Arisoy et al46 recommend rifampin at 20 mg/kg per day divided every 12 hours for 14 days alone or in combination.45 Another regimen uses gentamicin (2.5 mg/kg per dose every 8 hours) until patient is afebrile, followed by trimethoprim-sulfamethoxazole (5 mg/kg per dose every 12 hours) and rifampin (10 mg/kg per dose every 12 hours) for 2 to 4 weeks.

Because of the high mortality rate of Bartonella endocarditis, this condition should be treated aggressively. Antibiotics, and frequently, surgery, are required to treat endocarditis, although no antibiotic regimen has been proven effective in the literature. Typical treatment is an aminoglycoside combined with doxycycline or ceftriaxone.93 In a recent retrospective study in adults, patients who received an aminoglycoside, such as gentamicin, were more likely to fully recover, and those treated with an aminoglycoside for at least 14 days were more likely to survive compared with those with a shorter duration of therapy.130 Several authors recommend treating with an aminoglycoside combined with a β-lactam agent, such as ceftriaxone, with or without doxycycline.126

Immunocompromised individuals may develop severe, disseminated disease; however, their response to antibiotics is usually significantly more dramatic than those with intact immune systems. Systemic Bartonella infection in these patients has been treated with a number of agents, including β-lactam agents.131 Generally, bacillary angiomatosis tends to affect only those with impaired immunity and has been successfully treated with erythromycin, doxycycline, isoniazid, azithromycin, and rifampin.131,132 Lesions often improve after 4 to 7 days, with complete resolution in ~1 month.126 The drug of choice for children is erythromycin ethylsuccinate (40 mg/kg per day in 4 divided doses, with a maximum of 2 g/day) for 3 months. In cases of severe disease, combination therapy with intravenous erythromycin and rifampin is recommended. Relapses are more frequent if antibiotics are given for <3 months; thus, therapy should be given for at least 3 months' duration.126

The use of corticosteroids has been reported anecdotally for use in Bartonella encephalitis, hepatosplenic disease, ocular disease, and systemic disease; however, some patients failed to respond to these drugs.127,133136 More research is needed to determine if corticosteroids help to lessen the severity of disease or improve outcomes. At the current time, we do not recommend using corticosteroids in the treatment of Bartonella infection.

FUTURE DIRECTIONS

Over the last few decades, much has been learned about the spectrum of disease resulting from Bartonella henselae infection. As our knowledge of the microbiologic, pathologic, and clinical spectrum expands, an increasing number of questions develop. There has been no updated broad epidemiologic exploration of the various manifestations of disease. As more varied presentations of Bartonella are discovered, updated information on their patterns of occurrence, frequency, and distribution are needed. There is a significant gap in our knowledge of effective therapy for more complicated sequelae of infection. More randomized, rigorous trials are required to base our therapeutic decisions on meticulous evidence. Finally, clinicians should continue to include Bartonella in the differential diagnosis of prolonged fever, abdominal pain, and the many other varied presentations caused by this elusive bacterium.

FOOTNOTES

Accepted Oct 8, 2007.

Address correspondence to Todd A. Florin, MD, Children's Hospital of Philadelphia, 34th Street and Civic Center Boulevard, Philadelphia, PA 19104. E-mail: florint{at}email.chop.edu

Financial Disclosure: The authors have indicated they have no financial relationships relevant to this article to disclose.

REFERENCES

  1. Debre R, Lamy M, Jammet M, Costil L, Mozzicona P. La maladie des griffes de chat. Bull Mem Soc Med Hop Paris. 1950;66 :76 –79
  2. Parinaud H. Conjunctivite infectieuse transmise par les animaux. Ann d'Oculistique 1889;101 :252 –253
  3. Wear DJ, Margileth AM, Hadfield TL, Fischer GW, Schlagel CJ, King FM. Cat-scratch disease: a bacterial infection. Science. 1983;221 (4618):1403 –1405[Abstract/Free Full Text]
  4. English CK, Wear DJ, Margileth AM, Lissner CR, Walsh GP. Cat-scratch disease: isolation and culture of the bacterial agent. JAMA. 1988;259 (9):1347 –1352[Abstract]
  5. Brenner DJ, Hollis DG, Moss CW, et al. Proposal of Afipia gen. nov., with Afipia felis sp. Nov. (formerly Cat-scratch disease bacillus), Afipia clevelandenisis sp. Nov. (formerly the Cleveland Clinic Foundation strain), Afipia broomeae sp. Nov., and three unnamed genospecies. J Clin Microbiol. 1991;29 (11):2450 –2460[Abstract/Free Full Text]
  6. Regnery RL, Olson JG, Perkins BA, Bibb W. Serological response to "Rochalimaea henselae" antigen in suspected cat-scratch disease. Lancet. 1992;339 (8807):1443 –1445[CrossRef][ISI][Medline]
  7. Perkins BA, Swaminathan B, Jackson LA et al. Case 22–1992: pathogenesis of cat-scratch disease. N Engl J Med. 1992;327 (22):1599 –1601[ISI][Medline]
  8. Dolan MJ, Wong MT, Regnery RL, et al. Syndrome of Rochalimaea henselae adenitis suggesting cat-scratch disease. Ann Intern Med. 1993;118 (5):331 –336[Abstract/Free Full Text]
  9. Brenner DJ, O'Connor SP, Winkler HJ, Steigerwalt AG. Proposals to unify the Genera Bartonella and Rochalimaea, with descriptions of Bartonella Quintana Comb. Nov., Bartonella Vinsonii Comb. Nov., Bartonella Henselae Comb. Nov., and Bartonella Elizabethe Comb. Nov; and to remove the family Bartonellaceae from the order Richettsiales. Int J Syst Bacteriol. 1993;43 (4):777 –786[Abstract/Free Full Text]
  10. Carithers HA. Cat-scratch disease: an overview based on a study of 1200 patients. Am J Dis Child. 1985;139 (11):1124 –1133[Abstract]
  11. La Scola B, Liang Z, Zeaiter Z, Houpikian P, Grimont PA, Raoult D. Genotypic characteristics of two serotypes of Bartonella henselae. J Clin Microbiol. 2002;40 (6):2002 –2008[Abstract/Free Full Text]
  12. Zeaiter Z, Fournier PE, Raoult D. Genomic variation of Bartonella henselae strains detected in lymph nodes of patients with cat scratch disease. J Clin Microbiol. 2002;40 (3):1023 –1030[Abstract/Free Full Text]
  13. Moriarty RA, Margileth AM. Cat scratch disease. Infect Dis Clin North Am. 1987;1 (3):575 –590[Medline]
  14. Bass JW, Vincent JM, Person DA. The expanding spectrum of Bartonella infections: II. cat-scratch disease. Pediatr Infect Dis J. 1997;16 (2):163 –179[CrossRef][ISI][Medline]
  15. Hadfield TL, Malaty RH, Van Dellen A, Wear DJ, Margileth AM. Electron microscopy of the bacilli causing cat-scratch disease. J Infect Dis. 1985;152 (3):643 –645[ISI][Medline]
  16. Chomel BB, Abbott RC, Kasten KA, et al. Experimental transmission of Bartonella henselae by the cat flea. J Clin Microbiol. 1996;34 (8):1952 –1956[Abstract]
  17. Mändle T, Einsele H, Schaller M, et al. Infection of human CD34+ progenitor cells with Bartonella henselae results in intraerythrocytic presence of B henselae. Blood. 2005;106 (4):1215 –1222[Abstract/Free Full Text]
  18. Schweyer S, Fayyazi A. Activation and apoptosis of macrophages in cat scratch disease. J Pathol. 2002;198 (4):534 –540[CrossRef][ISI][Medline]
  19. Resto-Ruiz S, Burgess A, Anderson BE. The role of the host immune response in pathogenesis of Bartonella henselae. DNA Cell Biol. 2003;22 (6):431 –440[CrossRef][ISI][Medline]
  20. Riess T, Andersson SG, Lupas A, et al. Bartonella adhesin A mediates a proangiogenic host cell response. J Exp Med. 2004;200 (10):1267 –1278[Abstract/Free Full Text]
  21. Dehio C. Recent progress in understanding Bartonella-induced vascular proliferation. Curr Opin Microbiol. 2003;6 (1):61 –65[CrossRef][ISI][Medline]
  22. Bergmans AM, Schellekens JF, van Embden JD, Schouls LM. Predominance of two Bartonella henselae variants among cat-scratch disease patients in the Netherlands. J Clin Microbiol. 1996;34 (2):254 –260[Abstract]
  23. Maruyama S, Izumikawa K, Miyashita M, et al. First isolation of Bartonella henselae type I from a cat-scratch disease patient in Japan and its molecular analysis. Microbiol Immunol. 2004;48 (2):103 –109[ISI][Medline]
  24. Kelly PJ, Meads N, Theobald A, Fournier PE, Raoult D. Rickettsia felis, Bartonella henselae, and B. clarridgeiae, New Zealand. Emerg Infect Dis. 2004;10 (5):967 –968[ISI][Medline]
  25. Fournier PE, Robson J, Zeaiter Z, McDougall R, Byrne S, Raoult D. Improved culture from lymph nodes of patients with cat scratch disease and genotypic characterization of Bartonella henselae isolates in Australia. J Clin Microbiol. 2002;40 (10):3620 –3624[Abstract/Free Full Text]
  26. Jackson LA, Perkins BA, Wenger JD. Cat scratch disesae in the United States: an analysis of three national databases. Am J Public Health. 1993;83 (12):1707 –1711[Abstract/Free Full Text]
  27. Reynolds MG, Holman RC, Curns AT, O'Reilly M, McQuiston JH, Steiner CA. Epidemiology of cat-scratch disease hospitalizations among children in the United States. Pediatr Infect Dis J. 2005;24 (8):700 –704[CrossRef][ISI][Medline]
  28. Anderson BE, Neuman MA. Bartonella spp. as emerging human pathogens. Clin Microbiol Rev. 1997;10 (2):203 –219[Abstract]
  29. Dalton MJ, Robinson LE, Cooper JJ, Regnery RL, Olson JG, Childs JE. Use of Bartonella antigens for serologic diagnosis of cat-scratch disease at a national referral center. Arch Intern Med. 1995;155 (15):1670 –1676[Abstract]
  30. Zangwill KM, Hamilton DH, Perkins BA, et al. Cat Scratch Disease in Connecticut: epidemiology, risk factors, and evaluation of a new diagnostic test. N Engl J Med. 1993;329 (1):8 –13[Abstract/Free Full Text]
  31. Chomel BB, Boulouis HJ, Breitschwerdt EB. Cat scratch disease and other zoonotic Bartonella infections. J Am Vet Med Assoc. 2004;224 (8):1270 –1279[CrossRef][ISI][Medline]
  32. Maruyama S, Kabeya H, Nakao R, et al. Seroprevalence of Bartonella henselae, Toxoplasma gondii, FIV and FeLV infections in domestic cats in Japan. Microbiol Immunol. 2003;47 (2):147 –153[ISI][Medline]
  33. Jameson P, Greene C, Regnery R, et al. Prevalence of Bartonella henselae antibodies in pet cats throughout regions of North America. J Infect Dis. 1995;172 (4):1145 –1149[ISI][Medline]
  34. Guptill L, Wu CC, HogenEsch H, et al. Prevalence, risk factors, and genetic diversity of Bartonella henselae infections in pet cats in four regions of the United States. J Clin Microbiol. 2004;42 (2):652 –659[Abstract/Free Full Text]
  35. Solano-Gallego L, Bradley J, Hegarty B, Sigmon B, Breitschwerdt E. Bartonella henselae IgG antibodies are prevalent in dogs from southeastern USA. Vet Res. 2004;35 (5):585 –595[CrossRef][ISI][Medline]
  36. Gonzalez BE, Correa AG, Kaplan SL. Cat-scratch disease occurring in three siblings simultaneously. Pediatr Infect Dis J. 2003;22 (5):467 –468[CrossRef][ISI][Medline]
  37. Chian CA, Arrese JE, Pierard GE. Skin manifestations of Bartonella infections. Int J Dermatol. 2002;41 (8):461 –466[CrossRef][ISI][Medline]
  38. García CJ, Varela C, Abarca K, Ferres M, Prado P, Vial PA. Regional lymphadenopathy in cat-scratch disease: ultrasonographic findings. Pediatr Radiol. 2000;30 (9):640 –643[CrossRef][ISI][Medline]
  39. Massei F, Gori L, Macchia P, Maggiore G. The expanded spectrum of bartonellosis in children. Infect Dis Clin North Am. 2005;19 (3):691 –711[CrossRef][ISI][Medline]
  40. Malatack JJ, Jaffe R. Granulomatous hepatitis in three children due to cat-scratch disease without peripheral adenopathy: an unrecognized cause of fever of unknown origin. Am J Dis Child. 1993;147 (9):949 –953[Abstract]
  41. Jacobs RF, Schutze GE. Bartonella henselae as a cause of prolonged fever and fever of unknown origin in children. Clin Infect Dis. 1998;26 (1):80 –84[ISI][Medline]
  42. Dzelalija B, Petrovec M, Avsic-Supanc T. Probable atypical cat scratch disease presenting as isolated posterior pancreatic duodenal lymphadenitis and abdominal pain. Clin Infect Dis. 2001;33 (6):912 –914[CrossRef][ISI][Medline]
  43. Losanoff JE, Sauter ER, Rider KD. Cat scratch disease presenting with abdominal pain and retroperitoneal lymphadenopathy. J Clin Gastroenterol. 2004;38 (3):300 –301[CrossRef][ISI][Medline]
  44. Daybell D, Paddock CD, Zaki SR, et al. Disseminated infection with Bartonella henselae as a cause of spontaneous splenic rupture. Clin Infect Dis. 2004;39 (3):e21 –e24[CrossRef][ISI][Medline]
  45. Arisoy ES, Correa AG, Wagner ML, Kaplan SL. Hepatosplenic cat-scratch disease in children: selected clinical features and treatment. Clin Infect Dis. 1999;28 (4):778 –784[ISI][Medline]
  46. Dunn MW, Berkowitz FE, Miller JJ, Snitzer JA. Hepatosplenic cat-scratch disease and abdominal pain. Pediatr Infect Dis J. 1997;16 (3):269 –272[CrossRef][ISI][Medline]
  47. Massei F, Massimetti M, Messina F, Macchia P, Maggiore G. Bartonella henselae and inflammatory bowel disease. Lancet. 2000;356 (9237):1245 –1246[CrossRef][ISI][Medline]
  48. Mortelé KJ, Segatto E, Ros PR. The infected liver: radiologic-pathologic correlation. Radiographics. 2004;24 (4):937 –955[Abstract/Free Full Text]
  49. Ventura A, Massei F, Not T, Massimetti M, Bussani R, Maggiore G. Systemic Bartonella henselae infection with hepatosplenic involvement. J Pediatr Gastroenterol Nutr. 1999;29 (1):52 –56[CrossRef][ISI][Medline]
  50. Grando D, Sullivan LJ, Flexman JP, Watson MW, Andrew JH. Bartonella henselae associated with Parinaud's oculoglandular syndrome. Clin Infect Dis. 1999;28 (5):1156 –1158[ISI][Medline]
  51. Cunningham ET, Koehler JE. Ocular bartonellosis. Am J Ophthalmol. 2000;130 (3):340 –349[CrossRef][ISI][Medline]
  52. Suhler EB, Lauer AK, Rosenbaum JT. Prevalence of serologic evidence of cat-scratch disease in patients with neuroretinitis. Ophthalmology. 2000;107 (5):871 –876[CrossRef][ISI][Medline]
  53. Hu V, Dong B, MacFarlane A. Visual loss after cat scratch. J R Soc Med. 2005;98 (1):28 –29[Free Full Text]
  54. Schmalfuss IM, Dean CW, Sistrom C, Bhatti MT. Optic neuropathy secondary to cat scratch disease: distinguishing MR imaging features from other types of optic neuropathies. AJNR Am J Neuroradiol. 2005;26 (6):1310 –1316[Abstract/Free Full Text]
  55. Reed JB, Scales DK, Wong MT, Lattuada CP Jr, Dolan MJ, Schwab IR. Bartonella henselae neuroretinitis in cat scratch disease: diagnosis, management and sequelae. Ophthalmology. 1998;105 (3):459 –466[CrossRef][ISI][Medline]
  56. Wade NK, Levi L, Jones MR, Bhisitkul R, Fine L, Cunningham ET. Optic disk edema associated with peripapillary serous retinal detachment: an early sign of systemic Bartonella henselae infection. Am J Ophthalmol. 2000;130 (3):327 –334[CrossRef][ISI][Medline]
  57. Khurana RN, Albini T, Green RL, Rao NA, Lim JI. Bartonella henselae infection presenting as unilateral panuveitis simulating Vogt-Koyanagi-Harada syndrome. Am J Ophthalmol. 2004;138 (6):1063 –1065[CrossRef][ISI][Medline]
  58. Chang AA, Zeldovich A, Sachdev NH, Ly C, Beaumont P. Papillary vasoproliferative changes in cat scratch disease. Br J Ophthalmol. 2005;89 (1):122 –123[Free Full Text]
  59. Albini TA, Lakhanpal RR, Foroozan R, Holz ER. Macular hole in cat scratch disease. Am J Opthalmol. 2005;140 (1):149 –151[CrossRef][ISI][Medline]
  60. Ormerod LD, Skolnick KA, Menosky MM, Pavan PR, Pon DM. Retinal and choroidal manifestations of cat scratch disease. Ophthalmology. 1998;105 (6):1024 –1031[CrossRef][ISI][Medline]
  61. Curi ALL, Machado DO, Heringer G, Campos WR, Orefice F. Ocular manifestations of cat-scratch disease in HIV-positive patients. Am J Ophthalmol. 2006;141 (2):400 –401[CrossRef][ISI][Medline]
  62. Lewis DW, Tucker SH. Central nervous system involvement in cat scratch disease. Pediatrics. 1986;77 (5):714 –721[Abstract/Free Full Text]
  63. Carithers HA, Margileth AM. Cat-scratch disease: acute encephalopathy and other neurologic manifestations. Am J Dis Child. 1991;145 (1):98 –101[Abstract]
  64. Armengol CE, Hendley JO. Cat-scratch encephalopathy: a cause of status epilepticus in school-aged children. J Pediatr. 1999;134 (5):635 –638[CrossRef][ISI][Medline]
  65. Harvey RA, Misselbeck WJA, Uphold RE. Cat-scratch disease: an unusual cause of combative behavior. Am J Emerg Med. 1991;9 (1):52 –53[CrossRef][ISI][Medline]
  66. Wheeler SW, Wolf SM, Steinberg EA. Cat-scratch encephalopathy. Neurology. 1997;49 (3):876 –878[Abstract/Free Full Text]
  67. Seah AB, Azran MS, Rucker JC, Biousse V, Martin DF, Newman NJ. Magnetic resonance imaging abnormalities in cat-scratch disease encephalopathy. J Neuroophthalmol. 2003;23 (1):16 –21[Medline]
  68. Gerber JE, Johnson JE, Scott MA, Madhusudhan KT. Fatal meningitis and encephalitis due to Bartonella henselae bacteria. J Forensic Sci. 2002;47 (3):640 –644[ISI][Medline]
  69. Hmaimess G, Kadhim H, Saint Martin C, Abu Serieh BA, Mousny M, Sebire G. Cat scratch disease presenting as meningomyeloradiculopathy. Arch Dis Child. 2004;89 (7):691 –693[Free Full Text]
  70. Walter RS, Eppes SC. Cat scratch disease presenting with peripheral facial nerve paralysis. Pediatrics. 1998;101 (5). Available at: www.pediatrics.org/cgi/content/full/101/5/e13
  71. Ganesan K, Mizen K. Cat scratch disease: an unusual cause of facial palsy and partial ptosis: case report. J Oral Maxillofac Surg. 2005;63 (6):869 –872[CrossRef][ISI][Medline]
  72. Massei F, Gori L, Taddeucci G, Macchia P, Maggiore G. Bartonella henselae associated with Guillain-Barre syndrome. Pediatr Infect Dis J. 2006;25 (1):90 –91[CrossRef][ISI][Medline]
  73. Nowakowski GS, Katz A. Epilepsia partialis continua as an atypical presentation of cat scratch disease in a young adult. Neurology. 2002;59 (11):1815 –1816[Free Full Text]
  74. Puligheddu M, Giagheddu A, Genugu F, Giagheddu M, Marrosu F. Epielpsia partialis continua in cat scratch disease [published correction appears in Seizure. 2006;15(5):357]. Seizure. 2004;13 (3):191 –195[CrossRef][ISI][Medline]
  75. Rocha JL, Pellegrino LN, Riella LV, Martins LT. Acute hemiplegia associated with cat-scratch disease. Braz J Infect Dis. 2004;8 (3):263 –266[Medline]
  76. Baylor P, Garoufi A, Karpathios T, Lutz J, Mogelof J, Moseley D. Transverse myelitis in 2 patients with Bartonella henselae infection (cat scratch disease). Clin Infect Dis. 2007;45 (4):e42 –e45[CrossRef][ISI][Medline]
  77. Selby G, Walker GL. Cerebral arteritis in cat scratch disease. Neurology. 1979;29 (10):1413 –1417[ISI][Medline]
  78. Landau M, Kletter Y, Avidor B, et al. Unusual eruption as a presenting symptom of cat scratch disease. J Am Acad Dermatol. 1999;41 (5 pt 2):833 –836[CrossRef][ISI][Medline]
  79. Van Audenhove A, Verhoef G, Peetermans WE, Boogaerts M, Vandenberghe P. Autoimmune hemolytic anemia triggered by Bartonella henselae infection: a case report. Br J Haematol. 2001;115 (4):924 –925[CrossRef][ISI][Medline]
  80. Greenbaum B, Nelson P, Marchildon M, Donaldson M. Hemolytic anemia and hepatosplenomegaly associated with cat scratch disease. J Pediatr. 1986;108 (3):428 –430[CrossRef][ISI][Medline]
  81. Borker A, Gardner R. Severe thrombocytopenic purpura as a complication of cat-scratch disease. Clin Pediatr (Phila). 2002;41 (2):117 –118[Free Full Text]
  82. Economou M, Lithoxopoulou M, Aivazis V, Tsakalidis C, Athanassiou-Metaxa M. Bartonella henselae: association with the development of transient lupus anticoagulant and asymptomatic prolongation of activated partial thromboplastin time. Scand J Infect Dis. 2003;35 (2):149[CrossRef][ISI][Medline]
  83. Hulzebos CV, Koetse HA, Kimpen JLL, Wolfs TFW. Vertebral osteomyelitis associated with cat-scratch disease. Clin Infect Dis. 1999;28 (6):1310 –1312