Distribution, diversity, and host specificity of Bartonella in rodents from the Southeastern United States

A number of Bartonella isolates were obtained from seven species of rodents sampled from 12 geographic sites representing the major biotic communities of the southeastern United States. Bartonella were isolated from the blood of 42.2% of 279 tested rodents. The highest prevalence of infection typically occurred among the most commonly captured species in the rodent community. Four phylogenetic groups, uniting 14 genotypic variants of Bartonella, were identified by sequence analysis of the citrate synthase gene. The level of sequence homology between genotypic groups varied from 88.8% to 96.4%, and the degree of homology among variants within groups was > or = 97%. Cotton rats (Sigmodon hispidus) harbored up to three phylogenetic groups of Bartonella at a single site, and Bartonella of two phylogenetic groups were isolated from a single rodent. All the Bartonella isolated from three species of Peromyscus clustered in a single distinct phylogenetic group, suggesting some host specificity may occur. Mouse ascitic fluids produced in BALB/c mice inoculated with Bartonella of three phylogenetic groups demonstrated high indirect fluorescent antibody (IFA) titers to homologous antigens. However, use of eight Bartonella antigens in an IFA test with sera from 394 wild-caught rodents resulted in either little or extremely low titers of antibody.

Isolation and characterization of Ehrlichia chaffeensis strains from patients with fatal ehrlichiosis

Two new isolates of Ehrlichia chaffeensis (designated Jax and St. Vincent) were obtained from patients with fatal ehrlichial infections. Patients developed characteristic manifestations of severe disease due to E. chaffeensis, including marked thrombocytopenia, pulmonary insufficiency, and encephalopathy. Primary isolation was achieved in DH82 cells; the Jax and St. Vincent isolates were detected within 19 and 8 days postinoculation, respectively. The isolates were characterized by molecular evaluation of the 16S rRNA gene, the groESL heat shock operon, a 120-kDa immunodominant protein gene, and an incompletely characterized repetitive-motif sequence (variable-length PCR target [VLPT]). The sequences were compared with those of the corresponding molecular regions in the type isolate (Arkansas). St. Vincent contained one fewer repeat unit in both the 120-kDa protein gene and the VLPT compared with corresponding sequences of the Jax and Arkansas isolates. 16S rRNA gene sequences from the two new isolates had 100% identity to the corresponding sequences of the 91HE17 and Sapulpa isolates of E. chaffeensis, and to the corrected 16S rRNA gene sequence of the Arkansas isolate. The Jax isolate grew more slowly than the St. Vincent isolate in DH82 cells, and both of the new isolates grew more slowly than the extensively passaged Arkansas isolate. Although specific associations between ehrlichial pathogenicity and genotype were not identified from these comparisons, recovery of this organism from a spectrum of clinical presentations remains an integral step in understanding mechanisms of disease caused by E. chaffeensis.

A survey for antibodies to Lassa virus among health workers in Nigeria

A study was conducted among 552 health workers at 6 health facilities in Nigeria. Lassa virus immunoglobulin (Ig) G antibody was detected in 12.3%, using an enzyme-linked immunosorbent assay. Antibody prevalence in the 6 health centres ranged from 1.2% to 27.3%. Prevalences were higher in primary and secondary health facilities than in tertiary centres. Seroprevalences ranged from 1.7% to 23.7% among different occupational groups of health workers; the highest observed antibody prevalence was among ward aids. Lassa virus IgM antibody, indicating recent infection, was present in 6 of the health workers, 5 of whom were ward aids and one was a nurse. All of the health workers with specific IgM came from a single facility in Lafia, sampled during an outbreak of Lassa fever.

Community structure and prevalence of hantavirus infection in rodents: a geographic division of the enzootic area in far eastern Russia

The results of an extensive rodent trapping effort throughout the southern part of far eastern Russia and hantavirus antigen screening of tissues were used to develop a multifaceted approach for the geographic division of the enzootic territory of hantavirus. Four species of rodents (Apodemus agrarius, Apodemus peninsulae, Microtus fortis, and Clethrionomys rufocanus) comprised 88.5 percent of 10,595 captured rodents and 94.1 percent of 996 antigen-positive animals. Rodent fauna and the prevalence and distribution of hantavirus antigen-positive animals were compared among major biotic communities in the region. The species composition of the rodent communities and the predominant hantavirus reservoir species were used as criteria to define zones with similar enzootic characteristics.

An indirect immunofluorescence assay using a cell culture-derived antigen for detection of antibodies to the agent of human granulocytic ehrlichiosis

An indirect immunofluorescence assay for the detection of human antibodies to the agent of human granulocytic ehrlichiosis (HGE) was developed and standardized. Antigen was prepared from a human promyelocytic leukemia cell line (HL-60) infected with a tick-derived isolate of the HGE agent (USG3). Suitable antigen presentation and preservation of cellular morphology were obtained when infected cells were applied and cultured on the slide, excess medium was removed, and cells were fixed with acetone. Use of a buffer containing bovine serum albumin and goat serum reduced background fluorescence, and use of an immunoglobulin G (gamma-specific) conjugate reduced nonspecific binding. The assay readily detected specific antibody from HGE patients and did not detect antibody from healthy individuals. No significant reactivity was noted in sera from patients with high titers of antibodies to other rickettsial species. We were able to identify antibodies reactive to USG3 antigen in samples from areas where HGE is endemic that had tested negative to other rickettsial agents. Animal sera reactive against Ehrlichia equi or Ehrlichia phagocytophila bound to the HGE antigen, indicating that the assay may be useful for veterinary use. Comparability between two different laboratories was assessed by using coded human sera exchanged between laboratories. Results from the two laboratories were similar, indicating that the assay can be easily integrated into use for routine testing for HGE. The assay was then compared to an assay using horse neutrophils infected with ehrlichiae. The two assays gave comparable results, indicating that the cell culture-derived antigen can be used for testing samples that have been previously tested with E. equi as an antigen. The new assay offers several advantages over other immunofluorescence methods that use animal-derived antigen and is suitable for use in testing for human antibodies to the HGE agent.

Experimental evaluation of rodent exclusion methods to reduce hantavirus transmission to humans in rural housing

An experimental assessment of methods to reduce rodent infestations in rural housing was conducted in Yosemite National Park, California, Sequoia/Kings Canyon National Parks, California, and Shenandoah National Park, Virginia. During pretreatment surveys, nearly all (63 of 68) selected units had past or ongoing rodent activity inside. Active infestations were found in 58.8% of the units. Peromyscus spp. represented 91.2% of all animals caught inside housing units. Despite little harborage, rodent activity was common near housing (290 animals/2,254 trap nights). The most common species present was Peromyscus maniculatus (43-50% of all captures). This species was especially frequent (49-87% of Peromyscus captures) around the foundations of housing units. Habitat had little effect on captures. There were 1.8 Peromyscus caught per unit along the foundations of housing in modified rural settings with grass lawns compared with 1.2 Peromyscus caught per unit in sites located in mature woodlands. During autumn of 1994, randomly selected housing units were rodent proofed by sealing openings associated with chases, roof eaves, and attics with insulation and wire mesh. Housing was examined and the fauna was resampled in the spring-summer of 1995. Rodent-proofed houses were infested significantly less often (3 of 28) than control houses (13 of 36) (P = 0.02) and the intensity of infestation was lower in experimental houses (6 versus 23 mice/treatment). More than 25% of the mice trapped inside the houses had been marked outside the houses during the three-day surveys, demonstrating movement of mice adjacent to the buildings into not rodent-proofed housing. As in the previous autumn, most of the animals captured in (98.9%) and along the foundations of the houses (77.5%) were Peromyscus spp. These results demonstrate that Peromyscus frequently invade rural housing but rodent-proofing effectively eliminates or substantially reduces rodent activity.

Patterns of association with host and habitat: antibody reactive with Sin Nombre virus in small mammals in the major biotic communities of the southwestern United States

The distribution and prevalence of antibody reactive with Sin Nombre virus were determined in mammals in biotic communities of the southwestern United States. Small mammals (n = 3,069) of 69 species were trapped in nine communities from lower Sonoran desert to alpine tundra. Antibody was found in rodents from all communities (overall prevalence = 6.3%); prevalence was lowest at the altitudinal and climatic extremes (0.4% in desert and 2.0% in alpine tundra). Antibody occurred in 11% of 928 deer mice, 20% of 355 brush mice, 23% of 35 western harvest mice, and 12% of 24 Mexican voles. No infected deer mice were found in desert habitat; prevalence varied from 4% in chaparral to 17% in pinyon-juniper. Brush mice were frequently infected in chaparral and montane forest (25%). Seropositivity was higher in males and in heavier animals, suggesting horizontal transmission among adult males. Decreasing prevalence with age among the youngest deer mice suggests that infected dams confer passive immunity to pups.

Survey of veterinary professionals and other veterinary conference attendees for antibodies to Bartonella henselae and B quintana

OBJECTIVE

To determine serologic and epidemiologic characteristics of an occupational group potentially at risk for Bartonella sp infection.

DESIGN

Epidemiologic survey.

SAMPLE POPULATION

351 veterinarians, veterinary technicians, and other individuals attending a veterinary conference in Ohio.

PROCEDURE

A serum sample was obtained from each individual and tested for antibodies to Bartonella henselae or B quintana. A 24-question survey also was administered regarding demographic, occupational, and exposure information.

RESULTS

25 (7.1%) individuals were seropositive for B henselae or B quintana. Forty-seven, of whom 5 were seropositive, reported a history of illness consistent with cat-scratch disease and 18, of whom 3 were seropositive, reported a previous diagnosis of cat-scratch disease. Of the variables analyzed, only years of experience with cats was correlated with seropositivity.

CLINICAL IMPLICATIONS

The overall seroprevalence for 2 species of Bartonella in this occupational group was only slightly higher than that reported from other surveys. Seroprevalences among veterinarians, veterinary technicians, hospital staff, and others were essentially identical. Small sample groups, high percentage of cat ownership among participants, unknown duration of seropositivity, and unknown prevalence of infection among cats were potential confounders.

Surveillance and spatiotemporal associations of rabies in rodents and lagomorphs in the United States, 1985-1994

Between 1985 and 1994, 368 cases of rabies in rodents (95% of reports) and lagomorphs (5%) were reported to the Centers for Disease Control and Prevention, Atlanta, Georgia (USA), from 22 states. This was a 354% increase from the period 1971 to 1984. Most reports were cases of rabies in woodchucks (Marmota monax) (n = 317), primarily from the eastern United States, which has been recently experiencing an epizootic of raccoon (Procyon lotor) rabies. Cases of rabies in woodchucks were temporally and spatially associated with reports of raccoon rabies. Antigenic or genetic characterization of variants of rabies viruses from rodents and woodchucks corresponded to the variants associated with the major terrestrial wildlife reservoir within the geographic region of specimen origin. Although rodents and lagomorphs are infrequently infected with rabies and human contact with these animals rarely requires postexposure treatment, appropriate health authorities need to evaluate individual circumstances surrounding potential exposures.

Rabies surveillance in the United States during 1995

In 1995, 49 states, the District of Columbia, and Puerto Rico reported 7,877 cases of rabies in nonhuman animals and 4 cases in human beings to the Centers for Disease Control and Prevention. Nearly 92% (7,247 cases) were wild animals, whereas 8% (630 cases) were domestic species. The total number of reported cases decreased 4.2% from that of 1994 (8,230 cases). Most of the decline was the result of 17.1% fewer reported cases of rabies in raccoons in areas of the Northeast, where rabies is now enzootic rather than epizootic. Exceptions to this decline were detected in states where the virus has only recently entered raccoon populations or where ongoing epizootics persist. States experiencing increasing epizootic activity associated with this variant include Maine (3 cases in 1993 to 101 cases in 1995), North Carolina (9 cases in 1990 to 466 cases in 1995), Rhode Island (1 case in 1993 to 324 cases in 1995), and Vermont (45 cases in 1993 to 179 cases in 1995). The raccoon variant of the rabies virus is now present in Alabama, Pennsylvania, Vermont, West Virginia, and all Atlantic Seaboard states from Florida to Maine. In Ohio, this variant, last detected in 1992 as a single case, was again detected in 1996. Epizootics of rabies in foxes in west central Texas and in dogs and coyotes in southern Texas attributable to canine variants continue, with this state reporting 137 rabid foxes, 55 rabid dogs, and 80 of the 83 cases in coyotes during 1995. The number of rabid bats (787) increased by almost 25%, with cases reported by 47 of the 48 contiguous states. Nationally, reported cases of rabies in cattle (136) and cats (288) increased by 22.5 and 7.9%, respectively, whereas cases in dogs (146) decreased by 4.6%. Cats continued to be the domestic animal most frequently reported rabid. The cases of rabies reported in human beings were all caused by viral variants associated with bats. Eighteen states and Puerto Rico reported decreases in rabies in animals in 1995, compared with 28 states and the District of Columbia in 1994. Hawaii was the only state that did not report a case of rabies in 1995.

Antibodies to Bartonella species in inner-city intravenous drug users in Baltimore, Md

BACKGROUND

Bartonella quintana has recently been associated with homeless alcoholic men. Both B quintana and Bartonella henselae have been shown to be opportunistic pathogens of people with acquired immunodeficiency syndrome. The reservoirs and modes of transmission of these infections are incompletely known.

OBJECTIVES

To examine serum samples that were taken from inner-city intravenous (IV) drug users for antibodies to Bartonella organisms to determine whether there is an urban transmission cycle for Bartonella species and to examine the demographic and behavioral characteristics of IV drug users to identify possible risk factors for infection with any of these agents.

MATERIALS AND METHODS

A serologic survey was conducted, using a convenience sample of serum specimens collected during a study of IV drug use and human immunodeficiency virus infection among 630 inner-city residents in Baltimore, Md. A detailed questionnaire was administered at the initial collection of serum, and additional serum collections and questionnaire updates were made at 6-month intervals. The most recent available serum sample was tested for Bartonella antibody titer by using the indirect immunofluorescent antibody test with 3 antigens: Bartonella elizabethae, B henselae, and B quintana. Univariate and multivariate analyses of selected potential demographic and behavioral risk factors were conducted.

RESULTS

Antibodies to Bartonella were highly prevalent in this group; more than 37% of all samples reacted with at least 1 antigen. Overall seroprevalence of antibodies to B elizabethae, B henselae, and B quintana was 33%, 11%, and 10%, respectively. Current IV drug use, frequency of injection, and seronegative human immunodeficiency virus status were significantly associated with Bartonella antibody presence, but these associations varied by analysis. There was a significant inverse association of antibody prevalence to B henselae and B quintana by using CD4+ cell counts among human immunodeficiency virus-seropositive individuals.

CONCLUSIONS

Intravenous drug users have an elevated prevalence of antibodies to Bartonella organisms and may be at significant risk of becoming infected. Current IV drug use, high frequency of injection, and seronegative human immunodeficiency virus status are significant risk factors for an increased prevalence of Bartonella antibodies. The current natural histories of Bartonella species are rapidly changing, and mechanisms of transmission remain unknown.

Mass human exposure to rabies in New Hampshire: exposures, treatment, and cost

OBJECTIVES

This study assessed the rabies exposure and treatment that at least 665 persons in Concord, NH, received as a result of one proven rabid pet-store kitten in October 1994.

METHODS

All treatment recipients were interviewed by person or phone.

RESULTS

The median age of the treatment recipients was 14 years; 58% were female. The most common exposures were low risk (e.g., picking up, petting, nuzzling, or being scratched by a potentially rabid kitten). Local reactions to vaccine or immune globulin were reported by 76.5% of recipients, while 48.8% reported at least one systemic reaction. Cost for the biologicals was estimated at more than $1.1 million.

CONCLUSIONS

Because of the inadequacy of pet store records, the inconsistent application of treatment guidelines, and other factors, many people received postexposure treatment as a result of contacts that were unlikely to transmit rabies. The rates of local and systemic adverse reactions experienced were consistent with previous reports.

Prevalence of antibodies to arenaviruses in rodents from the southern and western United States: evidence for an arenavirus associated with the genus Neotoma

The objectives of this study were to extend our knowledge of the geographic distribution and rodent host range of arenaviruses in North America. Sera from wild rodents collected from the southern and western United States were tested for antibody against Tamiami, Pichinde, Junin, and lymphocytic choriomeningitis viruses, using an indirect fluorescent antibody test. Antibody to at least one arenavirus was found in 220 (3.1%) of 7,106 rodents tested. The antibody-positive animals included Mus musculus from Florida and Texas; Neotoma albigula from Arizona, Colorado, and New Mexico; N. fuscipes and N. lepida from California: N. mexicana from Arizona, New Mexico, and Utah; N. stephensi from Arizona and New Mexico; and Oryzomys palustris and Sigmodon hispidus from Florida. Sigmodon hispidus seropositive for Tamiami virus were found only in Florida (156 [27.0%] of 578 tested), although 463 hispid cotton rats from outside that state were examined. High-titered antibodies to Tamiami virus were present in sera from S. hispidus, (geometric mean antibody titer [GMAT] of 1:792), whereas sera from Neotoma spp. reacted at high titer to both Tamiami (GMAT = 1:905) and Pichinde (GMAT = 1:433) viruses. The results suggest that arenaviruses are widely distributed in the southern United States and that one or more indigenous arenaviruses are associated with Neotoma spp. in North America.

Evaluation of a canine rabies vaccination campaign and characterization of owned-dog populations in the Philippines

A mass canine rabies vaccination campaign in Sorsogon Province, the Republic of the Philippines, was conducted in April and May 1993. From 1 to 14 days following visits by vaccination teams to 30 selected villages (barangays), survey teams revisited the barangays to assess vaccine coverage. Modified cluster survey methods were used to gather information about vaccine coverage in the owned-dog population (210 households) and about characteristics of owned-dogs and factors influencing owner willingness to participate in the campaign. Vaccinated dogs were identified by asking owners about receipt of certificates given by the vaccinating teams and examining each animal for a special collar or paint mark placed on the animal at the time of vaccination. Survey results indicated that 73% (178/243) of eligible dogs were vaccinated and 82% of vaccinated dogs were marked with a collar or paint. Dogs were owned by 69% of households and ranged in age from 3 days to 13 years (median = 1 year), and the ratio of male to females dogs was 1:1. The dog-to-human ratio was 1:3.8, with an average of 1.4 dogs per household or 2.1 dogs per dog-owning household. Most dogs were kept as guards (83%) and most were free-ranging (85%). The most common reasons dogs were not vaccinated included they could not be restrained (11/64), the owner was not home (10/64), and fear of injury resulting from vaccination (10/64). The owners of 20% of vaccinated dogs reported some adverse reaction in their pet. Improved vaccine coverage was significantly associated with restrained dogs kept primarily for guard functions by owners who received information about the vaccination campaign from multiple sources. Vaccine coverage was sufficiently high to potentially control rabies transmission among dogs through herd immunity and indicated a successful vaccine campaign.

Characterization of Oliveros virus, a new member of the Tacaribe complex (Arenaviridae: Arenavirus)

Oliveros virus is an agent isolated in cell culture from Bolomys obscurus (Rodentia, Muridae, Sigmodontinae) captured on the central Argentine pampa. Oliveros virus was shown to be related to members of the Tacaribe complex of the family Arenaviridae by immunofluorescent antibody (IFA) tests, electrophoretic pattern of viral proteins, and morphology as observed by electron microscopy. It was distinct from 12 other arenaviruses by a combination of plaque-reduction neutralization tests, comparison of endpoint titers among cross-IFA tests, and comparison of viral RNA sequence data. This agent is the third new arenavirus from South America described within the last three years.

Current issues in rabies prevention in the United States health dilemmas. Public coffers, private interests

OVER THE LAST 100 years, rabies in the United States has changed dramatically. More than 90% of all animal rabies cases reported annually to the CDC now occur in wildlife, whereas before 1960 the majority were in domestic animals. The principal rabies hosts today are wild carnivores and bats infected with several viral variants. Annual human deaths have fallen from more than a hundred at the turn of the century to one to two per year despite major outbreaks of animal rabies in several geographic areas. Modern day prophylaxis has proven nearly 100% successful; most human fatalities now occur in people who fail to seek medical treatment, usually because they do not recognize a risk in the animal contact leading to the infection. Although these human rabies deaths are rare, the estimated public health costs associated with disease detection, prevention, and control have risen, exceeding millions of dollars each year. Cost considerations must be weighed along with other factors in addressing issues such as the appropriate handling of nontraditional and exotic pets, future guidelines for rabies prophylaxis, and novel methods of disease prevention.

Rabies surveillance in the United States during 1994

In 1994, 48 states, the District of Columbia, and Puerto Rico reported 8,224 cases of rabies in nonhuman animals and 6 cases in human beings to the Centers for Disease Control and Prevention. Nearly 93% (7,632 cases) were wild animals, whereas 7% (592 cases) were domestic species. The total number of reported cases decreased 13.4% from that of 1993 (9,498 cases), with most of the decline resulting from 19.2% fewer cases of rabies in raccoons. Two previously described epizootics of rabies involving the raccoon variant of the rabies virus have converged in North Carolina, and the resulting region is now continuous from Alabama and Florida in the South to Maine in the North. Epizootics of rabies in foxes in west central Texas and in dogs and coyotes in southern Texas continue to expand, with this state reporting 144 rabid foxes, 53 rabid dogs, and 77 of the 85 cases in coyotes during 1994. Maine and New Hampshire reported cases of rabies in foxes (6 and 9, respectively) for the first time in 10 years. Nationally, reported cases of rabies in dogs (153) increased by 17.7%, whereas cases in cattle (111) and cats (267) decreased by 14.6 and 8.3%, respectively. Cats continued to be the domestic animal most frequently reported rabid. Twenty-eight states and the District of Columbia reported decreases in rabies in animals in 1994, compared with 22 states, the District of Columbia, and Puerto Rico in 1993. Hawaii and Nebraska were the only states that did not report cases of rabies in 1994.

Cluster of five children with acute encephalopathy associated with cat-scratch disease in south Florida

Between August 12 and September 27, 1994, five children in South Florida were hospitalized at a single hospital because of encephalopathy, presenting as status epilepticus, associated with cat-scratch disease (CSD). Diagnoses were confirmed by using an indirect fluorescent antibody test to detect antibody to Bartonella henselae, the causative agent of CSD. These cases represent the first cluster of CSD encephalopathy cases to be recognized in the United States. The patients lived within 7 miles of each other and all reported contact with pet or stray cats before developing regional lymphadenopathy and encephalopathy. All recovered fully. A high proportion of 124 cats from the local area were seropositive (62%) or bacteremic (22%). This study suggests that B. henselae can be associated with geographically focal clusters of CSD encephalitis and should be considered in the evaluation of children with acute encephalopathy.

Ehrlichiosis in a golf-oriented retirement community

BACKGROUND

Ehrlichiosis due to Ehrlichia chaffeensis usually occurs sporadically or in small clusters, with an annual incidence estimated at 3 to 5 cases per 100,000 population in areas of endemic disease. The putative principal vector is the Lone Star tick (Amblyomma americanum). We investigated an outbreak of ehrlichiosis that occurred in June 1993 among members of a golf-oriented retirement community (community A) in Tennessee. The community is densely wooded and borders a wildlife-management area where deer are numerous.

METHODS

We conducted a case-control study, using medical-history reviews, serologic testing, and testing with the polymerase chain reaction for E. chaffeensis infection. We also surveyed a sample of 10 percent of the households in community A and in another golf-oriented community (community B) more than 20 miles (32 km) from the wildlife-management area. Survey participants completed a questionnaire and provided specimens for serologic testing. In both communities, searches for ticks were undertaken.

RESULTS

Eleven cases of symptomatic ehrlichiosis were identified in the case-control study, 10 of which were in community A (attack rate, 330 per 100,000). Of 311 surveyed residents of community A, 12.5 percent had serologic evidence of past E. chaffeensis infection, as compared with 3.3 percent of 92 in community B (relative risk in community A as compared with community B, 3.9; 95 percent confidence interval, 1.2 to 12.2). The risk of infection was associated with tick bites, exposure to wildlife, golfing, and among golfers, retrieving lost golf balls from the rough. Persons who never used insect repellent were more likely to have had infection than persons who did. In community A, thousands of Lone Star ticks were found; in community B, only three ticks were found.

CONCLUSIONS

The high rate of E. chaffeensis infection in community A resulted from its proximity to a wildlife reserve. When outdoor recreational activities are common and concentrations of ticks are high, outbreaks of arthropod-borne zoonoses can be anticipated.

Use of Bartonella antigens for serologic diagnosis of cat-scratch disease at a national referral center

BACKGROUND

Bartonella henselae (formerly the genus Rochalimaea) has recently been isolated from patients with cat-scratch disease and their cats, and since September 1992 the Centers for Disease Control and Prevention has offered an indirect fluorescent antibody assay for Bartonella-specific antibody.

METHODS

Physicians submitted serum samples from patients suspected of having cat-scratch disease or other Bartonella-associated illness and completed a questionnaire that recorded clinical information. Indirect fluorescent antibody assay was performed with the use of antigen derived from three Bartonella species: B henselae, Bartonella quintana, and Bartonella elizabethae.

RESULTS

During 16 months, 3088 serum samples were received. The largest numbers of specimens and the highest percentages positive (titer, > or = 64) were observed in the fall and winter. Clinical histories of the first 600 patients for whom serum samples and completed information forms were received were examined in detail; seropositivity was significantly associated with cat contact, cat age of less than 1 year, cat scratch, presence of an inoculation papule, and regional adenopathy. Of 91 patients whose illness met a strict clinical definition of cat-scratch disease, 86 (95%) had titers of 64 or greater to either B henselae or B quintana. A fourfold rise or fall in titer was observed in 87 of 132 patients with paired serum samples.

CONCLUSIONS

The indirect fluorescent antibody assay for Bartonella-specific antibody is sensitive for the diagnosis of cat-scratch disease. Redefinition of cat-scratch disease on the basis of cause and use of this assay as a diagnostic criterion is recommended.

Infections associated with Bartonella species in persons infected with human immunodeficiency virus

Two members of the genus Bartonella, Bartonella quintana (formerly Rochalimaea quintana) and Bartonella henselae (formerly Rochalimaea henselae), have recently been recognized as agents of severe or fatal disease in patients infected with human immunodeficiency virus (HIV). The development of infection with B. henselae in HIV-infected individuals has been associated with traumatic contact with cats (scratches or bites), and domestic cats have been identified as a major reservoir for this organism. Specific information regarding the transmission of B. henselae to humans is not yet available, but common-sense precautions that minimize exposure to cat-associated organisms are appropriate. Preliminary accounts suggest that B. quintana infections are more common than B. henselae infections among HIV-infected individuals in San Francisco. The source of infection with B. quintana and the mechanism of its transmission remain unknown.

Analysis of Borrelia burgdorferi sequentially isolated from Peromyscus leucopus captured at a Lyme disease enzootic site

Thirty isolates of Borrelia burgdorferi sequentially cultured from 15 naturally infected white-footed mice (Peromyscus leucopus) were examined for variability in protein and plasmid profiles. Heterogeneity was detected in OspB and OspC and in proteins between 18.0 and 28.0 kDa by PAGE. Plasmid profiles were heterogeneous in the first isolate from 11 mice (73%) and between the first and last sequential isolate from 13 mice (87%). Comparison of the first and last isolates showed increased expression of OspC in 6 mice (40%) and was associated in each case with a shift in mobility of a 16.0-kb plasmid, suggesting that regulatory elements of ospC may reside on this plasmid. Hybridization studies suggested that individual mice may have been infected by a heterogeneous population of spirochetes and that changes in the protein and plasmid profiles between the first and last sequential isolates from some mice may have been the result of clonal selection.

Ear biopsy location influences detection of Borrelia burgdorferi by PCR, but not by culture in naturally infected Peromyscus leucopus

We determined if the ear biopsy location affected detection of Borrelia burgdorferi when either culture or the polymerase chain reaction (PCR) was used among 50 white-footed mice (Peromyscus leucopus), live-captured in a Lyme disease enzootic area in Maryland (USA) between March and October of 1991 and 1992. The infection status of individual mice was determined by organ culture; ear biopsy samples were obtained from the peripheral and central part of the ear for detection of B. burgdorferi by culture and by PCR. Overall, B. burgdorferi was cultured from one or more tissue samples in 33 (66%) of 50 captured mice. Among infected mice, B. burgdorferi was detected by culture in 29 (88%) of 33 peripheral and 28 (85%) of 33 central ear biopsy samples. By PCR it was detected in 24 (73%) of 33 peripheral and all 33 central samples (P = 0.002). Detection of B. burgdorferi by culture was independent of the ear biopsy location; however, the organism was detected by PCR with greater frequency in central ear biopsy samples as compared to peripheral samples. Agreement between culture and PCR was moderate (Kappa = 0.64) on peripheral ear samples and excellent (Kappa = 0.79) on central samples. We propose that when ear biopsy samples are used to detect B. burgdorferi by PCR in wild-caught P. leucopus, removal of biopsy samples from the central part of the ear will achieve maximum sensitivity and will achieve the highest concordance between assays when both culture and PCR of ear biopsy samples are conducted in parallel.