Evaluation of community-based serologic screening for identification of chronic Salmonella typhi carriers in Vietnam.

OBJECTIVES: To determine the utility of screening anti-Vi antibodies to detect chronic Salmonella Typhi carriers in an endemic community. METHODS: We conducted a community-based serologic survey for anti-Vi antibodies to identify chronic Salmonella Typhi carriers in a typhoid endemic region in Vietnam. RESULTS: We tested sera from 3209 (67.2%) of 4772 eligible adults. The median age was 37 years (range 20-92), 57.3% were female, 4.6% reported a history of typhoid fever and 0.3% reported typhoid vaccination. Anti-Vi antibody titers tested in Vietnam were < 1:40 in 2759 (86.0%), 1:40 in 194 (6.0%), 1:80 in 168 (5.2%), 1:160 in 57 (1.8%), and > or = 1:320 in 31 (1.0%). On re-testing in the USA, an additional 19 sera with titers > or = 1:160 were identified. We collected 589 rectal swabs from 103 (96.3%) of 107 persons with Vi antibody titers > or = 1:160 and 183 swabs from 33 persons with antibody titers < 1:80. No Salmonella Typhi was isolated. CONCLUSIONS: Community-based serologic screening is a feasible, but impractical method for identifying chronic Salmonella Typhi carriers. Background levels of anti-Vi antibody titers in this endemic area may be high despite a low prevalence of chronic carriers.

Hemolytic-Uremic Syndrome and Escherichia coli O121 at a Lake in Connecticut, 1999.

OBJECTIVE: Non-O157 Shiga toxin-producing Escherichia coli (STEC) have emerged as an important public health problem. Outbreaks attributed to non-O157 STEC rarely are reported. In 1999, follow-up of routine surveillance reports of children with hemolytic- uremic syndrome (HUS) identified a small cluster of 3 cases of HUS, all of whom had spent overlapping time in a Connecticut lake community in the week before onset of symptoms. We conducted an investigation to determine the magnitude and source of the outbreak and to determine risk factors associated with the transmission of illness. METHODS: We conducted a cohort study and an environmental investigation. The study population included all people who were at the lake in a defined geographic area during July 16-25, 1999. This time and area were chosen on the basis of interviews with the 3 HUS case-patients. A case was defined as diarrhea (>/=3 loose stools/d for >/=3 days) in a person who was at the lake during July 16-25, 1999. Stool samples were requested from any lake resident with diarrheal illness. Stools were cultured for Salmonella, Shigella, Campylobacter, and E coli O157. Broth cultures of stools were tested for Shiga toxin. Case-patients were asked to submit a serum specimen for antibody testing to lipopolysaccharides of selected STEC. Environmental samples from sediment, drinking water, lake water, and ice were obtained and cultured for E coli and tested for Shiga toxin. An environmental evaluation of the lake was conducted to identify any septic, water supply system, or other environmental condition that could be related to the outbreak. RESULTS: Information was obtained for 436 people from 165 (78%) households. Eleven (2.5%) people had illnesses that met the case definition, including the 3 children with HUS. The attack rate was highest among those who were younger than 10 years and who swam in the lake on July 17 or 18 (12%; relative risk [RR]: 7.3). Illness was associated with swimming (RR = 8.3) and with swallowing water while swimming (RR = 7.0) on these days. No person who swam only after July 18 developed illness. Clinical characteristics of case-patients included fever (27%), bloody diarrhea (27%), and severe abdominal cramping (73%). Only the 3 children with HUS required hospitalization. No bacterial pathogen was isolated from the stool of any case-patient. Among lake residents outside the study area, E coli O121:H19 was obtained from a Shiga toxin-producing isolate from a toddler who swam in the lake. Serum was obtained from 7 of 11 case-patients. Six of 7 case-patients had E coli O121 antibody titers that ranged from 1:320 to >1:20 480. E coli indicative of fecal contamination was identified from sediment and water samples taken from a storm drain that emptied into the beach area and from a stream bed located between 2 houses, but no Shiga toxin-producing strain was identified. CONCLUSIONS: Our findings are consistent with a transient local beach contamination in mid-July, probably with E coli O121:H19, which seems to be able to cause severe illness. Without HUS surveillance, this outbreak may have gone undetected by public health officials. This outbreak might have been detected sooner if Shiga toxin screening had been conducted routinely in HUS cases. Laboratory testing that relies solely on the inability of an isolate to ferment sorbitol will miss non-O157 STEC, such as E coli O121. Serologic testing can be used as an adjunct in the diagnosis of STEC infections. Lake-specific recommendations included education, frequent water sampling, and alternative means for toddlers to use lake facilities.

The United States National Prospective Hemolytic Uremic Syndrome Study: microbiologic, serologic, clinical, and epidemiologic findings.

The frequency of Shiga toxin-producing Escherichia coli (STEC) serotypes associated with postdiarrheal hemolytic uremic syndrome (HUS) cases among children and adults in the United States and the proportion with IgM or IgG lipopolysaccharide antibodies to E. coli O157 were determined by use of a nationwide sample from January 1987 through December 1991. Among 83 patients, STEC were isolated from 30 (43%) of 70 whose stool cultures yielded bacterial growth (25 E. coli O157 isolates and 5 non-O157 STEC isolates). Fifty-three (80%) of 66 patients with serum samples had positive O157 lipopolysaccharide antibody titers. Of the 83 patients, 60 (72%) had evidence of STEC infection, including 6 of 8 adults whose illnesses also met criteria for thrombotic thrombocytopenic purpura. Data from a subset of patients suggest that E. coli O157 was the cause of > or = 80% of the STEC infections. All 3 women who were postpartum had evidence of E. coli O157 infection. STEC infection should be considered the likely cause for all persons with postdiarrheal HUS.

Restaurant-associated outbreak of Salmonella typhi in Nauru: an epidemiological and cost analysis.

Typhoid fever is endemic in the South Pacific. We investigated an outbreak in Nauru. Through interviews and medical records, we identified 50 persons with onset between 1 October 1998 and 10 May 1999, of fever lasting > or = 3 days and one other symptom. Salmonella Typhi was isolated from 19 (38%) cases. Thirty-two (64%) patients were school-aged children, and 17 (34%) were in four households. Case-control studies of (a) culture-confirmed cases and age- and neighbourhood-matched controls; and (b) household index cases and randomly selected age-matched controls implicated two restaurants: Restaurant M (matched OR [MOR] = 11, 95% confidence interval [CI] = 1.3-96) and Restaurant I (MOR = 5.8, 95% CI = 1.2-29). Food-handlers at both restaurants had elevated anti-Vi antibody titres indicative of carrier state. The annual incidence was 5.0/1000 persons. Outbreak-associated costs were $46,000. Routine or emergency immunization campaigns targeting school-aged children may help prevent or control outbreaks of typhoid fever in endemic disease areas.

Escherichia coli O157:H7 outbreak associated with an improperly chlorinated swimming pool.

A cluster of gastrointestinal illnesses, including one case of hemolytic-uremic syndrome and one culture-confirmed Escherichia coli O157:H7 infection, followed a trailer park pool party. We interviewed a cohort of party attendees and park residents. A primary case was defined as the first gastrointestinal illness within a household between 5 July and 20 July in which the titer of IgG antibody to E. coli O157 (if determined) was elevated. Of 51 party attendees and trailer park residents, 18 developed a gastrointestinal illness, including 10 who met the definition of a primary case. Swimming in the pool significantly increased the risk of primary illness (relative risk = 6.3; 95% confidence interval = 1.8-18.9). No other exposure was significantly associated with primary illness, after pool exposure was controlled for. The implicated pool had little to no chlorine added during the period of 4-10 July. This outbreak provides new evidence of the importance of proper pool maintenance in controlling the spread of E. coli O157:H7.

An outbreak of Escherichia coli O157:H7 infections associated with leaf lettuce consumption.

In July 1995, 40 Montana residents were identified with laboratory-confirmed Escherichia coli O157:H7 infection; 52 residents had bloody diarrhea without laboratory confirmation. The median age of those with laboratory-confirmed cases was 42 years (range, 4- 86); 58% were female. Thirteen patients were hospitalized, and 1 developed hemolytic-uremic syndrome. A case-control study showed that 19 (70%) of 27 patients but only 8 (17%) of 46 controls reported eating purchased (not home-grown) leaf lettuce before illness (matched odds ratio, 25.3; 95% confidence interval, 3.9-1065.6). Pulsed-field gel electrophoresis identified a common strain among 22 of 23 isolates tested. Implicated lettuce was traced to two sources: a local Montana farm and six farms in Washington State that shipped under the same label. This outbreak highlights the increasing importance of fresh produce as a vehicle in foodborne illness. Sanitary growing and handling procedures are necessary to prevent these infections.

An outbreak of gastroenteritis and fever due to Listeria monocytogenes in milk.

BACKGROUND: After an outbreak of gastroenteritis and fever among persons who attended a picnic in Illinois, chocolate milk served at the picnic was found to be contaminated with Listeria monocytogenes. METHODS: In investigating this outbreak, we interviewed the people who attended the picnic about what they ate and their symptoms. Surveillance for invasive listeriosis was initiated in the states that receive milk from the implicated dairy. Stool and milk samples were cultured for L. monocytogenes. Serum samples were tested for IgG antibody to listeriolysin O. RESULTS: Forty-five persons had symptoms that met the case definition for illness due to L. monocytogenes, and cultures of stool from 11 persons yielded the organism. Illness in the week after the picnic was associated with the consumption of chocolate milk. The most common symptoms were diarrhea (present in 79 percent of the cases) and fever (72 percent). Four persons were hospitalized. The median incubation period for infection was 20 hours (range, 9 to 32), and persons who became ill had elevated levels of antibody to listeriolysin O. Isolates from stool specimens from patients who became ill after the picnic, from sterile sites in three additional patients identified by surveillance, from the implicated chocolate milk, and from a tank drain at the dairy were all serotype 1/2b and were indistinguishable on multilocus enzyme electrophoresis, ribotyping, and DNA macrorestriction analysis. CONCLUSIONS: L. monocytogenes is a cause of gastroenteritis with fever, and sporadic cases of invasive listeriosis may be due to unrecognized outbreaks caused by contaminated food.

A foodborne outbreak of gastroenteritis involving Listeria monocytogenes.

An outbreak of gastroenteritis occurred in Italy among 39 persons who had attended a private supper. All guests were previously healthy, young, non-pregnant adults; 18 (46%) had symptoms, mostly gastrointestinal (78%), with a short incubation period. Four were hospitalized with acute febrile gastroenteritis, two of whom had blood cultures positive for Listeria monocytogenes. No other microorganisms were recovered from the hospitalized patients' specimens. Epidemiological investigation identified rice salad as the most likely vehicle of the food-borne outbreak. L. monocytogenes was isolated from three leftover foods, the kitchen freezer and blender. Isolates from the patients, the foods and the freezer were indistinguishable: serotype 1/2b, same phage type and multilocus enzyme electrophoretic type. Eight (36%) of 22 guests tested were found to have antibodies against L. monocytogenes, compared with none of 11 controls from the general population. This point source outbreak was probably caused by infection with L. monocytogenes. Unusual features included the high attack rate among immunocompetent adults and the predominance of gastrointestinal symptoms.

Multilocus enzyme electrophoresis for characterization of Listeria monocytogenes isolates: results of an international comparative study.

Multilocus enzyme electrophoresis (MEE) is a standard technique that is used to elucidate the epidemiology of a variety of bacterial species. Recently, the method has been employed by several laboratories for investigations of clinical and foodborne isolates of Listeria monocytogenes. To assess the sensitivity and reproducibility of MEE in characterising L. monocytogenes isolates for epidemiological purposes and, ultimately, to agree on a standard protocol, seven laboratories participated in a blinded study of 80 strains. The strain collection included both epidemiologically related and unrelated isolates. Each laboratory used its own protocol for MEE. The number of enzymes that were assayed by the laboratories ranged from 8 to 23, and the total number of identified electrophoretic types (ETs) varied between 14 and 25. Of the II pairs of duplicate strains, the number of pairs recognised as identical by the seven laboratories ranged from 3 to 10 (median = 8). From 10 to 18 (median = 15) of the 22 groups of epidemiological related strains were recognised as homogeneous by the different laboratories. The discriminatory power of the method, calculated using Simpson's index of diversity for 69 strains (80 strains minus the 11 duplicates), ranged from 0.827 to 0.925. This relatively low discriminatory power is a consequence of a somewhat low genetic diversity of L. monocytogenes compared to other bacterial species. Efforts should be pursued to standardise the method in order to improve the intra- and inter-laboratory reproducibility.

Molecular subtyping of Neisseria meningitidis serogroup B: comparison of five methods.

In order to compare methods for subtyping Neisseria meningitidis serogroup B isolates, 96 isolates obtained from various locations in the United States and northwestern Europe were subtyped by five methods: monoclonal antibody (MAb)-based serotyping and serosubtyping, DNA macrorestriction analysis by pulsed-field gel electrophoresis (PFGE), multilocus enzyme electrophoresis (MEE), ribotyping, and PCR-restriction fragment length polymorphism of the internally transcribed spacer region of the rRNA operon (ITS PCR-RFLP). All N. meningitidis serogroup B isolates were typeable by PFGE, MEE, ribotyping, and ITS PCR-RFLP. Only 44.8% of the isolates were completely typeable (both serotype and serosubtype determination) by MAb-based serotyping and serosubtyping. 60.4% of the isolates could be serotyped but not serosubtyped, and 90.6% of the isolates could be either serotyped or serosubtyped. Simpson's discrimination indices of diversity for the methods were as follows: PFGE, 99.7%; MEE, 99.4%; ribotyping, 98.8%; MAb serotyping, 75.8%; MAb serotyping and/or serosubtyping 97.5%; and ITS PCR-RFLP, 84.2%. The high degree of diversity observed by PFGE, MEE, and ribotyping can be explained by the fact that isolates were collected from different geographic locations at various times. PFGE, MEE, and ribotyping showed greater discriminatory abilities than MAb-based serotyping and serosubtyping or ITS PCR-RFLP.

Meat grinders and molecular epidemiology: two supermarket outbreaks of Escherichia coli O157:H7 infection.

Between 23 June and 15 July 1994, 21 cases (19 primary and 2 secondary) of Escherichia coli O157:H7 infection were identified in the Bethel, Connecticut, area. Three pulsed-field gel electrophoresis (PFGE) patterns from 15 isolates (I, n = 13; II, n = 2; and III, n = 1) were observed. A case-control study that excluded secondary cases and patients with PFGE II and III patterns (n = 16) demonstrated that consumption of food from one supermarket was associated with illness (15/16 cases vs. 31/47 geographically matched controls, odds ratio [OR] undefined, lower 95% confidence interval OR = 1.45, P = .018). No one food was associated with illness. Inspection of the supermarket revealed deficiencies in hygiene and meat handling practices. The 2 cases with PFGE II ate raw beef and raw lamb from a second supermarket. These outbreaks demonstrate the value of PFGE in supporting epidemiologic investigations and the potential for outbreaks arising from retail outlets.

A rapid dot immunoassay for detecting the Brazilian purpuric fever clone of Haemophilus influenzae biogroup aegyptius with a "flow through" device.

Brazilian purpuric fever (BPF) is a highly fatal pediatric disease that may follow an episode of purulent conjunctivitis caused by a virulent clone of Haemophilus influenzae biogroup aegyptius (Hae). Oral rifampin prophylaxis, by eliminating carriage of the BPF clone in children with conjunctivitis, may prevent onset of the systemic disease. A test to detect the BPF clone directly from eye swabs could identify those in need of prophylaxis. This is a preliminary report of a rapid dot immunoassay performed on a "flow-through" cartridge that was developed for use under field conditions. The test is based upon recognition of a unique epitope of the 25-kDa pilin protein on the surface of BPF clone cells by a monoclonal antibody. With 36 laboratory-maintained cultures of Hae (15 clone isolates and 21 others), sensitivity of the assay was 67% and specificity was 95%. When fimbrial-enriched (25-kDa+) phenotypes of five false-negative clone strains were prepared for use as test antigens, sensitivity rose to 100%. Evaluation of the immunoassay under field conditions is necessary to prove its efficacy.

Serotypes and subtypes of Neisseria meningitidis: results of an international study comparing sensitivities and specificities of monoclonal antibodies.

An international study supported by the World Health Organization comparing monoclonal antibodies for serotyping and serosubtyping of Neisseria meningitidis strains was performed and the results were assessed in 1992. A collection of 6 serotype-specific (1, 2a, 2b, 4, 14, and 15) and 12 serosubtype-specific (P1.1, P1.2, P1.4, P1.5, P1.6, P1.7, P1.9, P1.10, P1.12, P1.14, P1.15, and P1.16) monoclonal antibodies was provided to 11 participating laboratories throughout the world. Monoclonal antibodies were tested on 85 Neisseria meningitidis strains with known reference results. Whole-cell enzyme-linked immunosorbent assay was used for analysis in 10 of 11 laboratories. The sensitivities and specificities of individual serotype- and subtype-specific monoclonal antibodies were evaluated. Differences in individual laboratories and with individual monoclonal antibodies were assessed. Relatively large differences in sensitivities were achieved in individual laboratories. On the contrary, the specificities remained at high levels in all laboratories. The sensitivities of serotype-specific monoclonal antibodies ranged from 72.0 to 100%. Individual serosubtype-specific monoclonal antibodies showed sensitivities ranging from 64.1 to 98.1%. The most frequent reason for the incorrect results obtained with the monoclonal antibodies were false-negative results. The collaborative study demonstrated that some monoclonal antibodies are not very sensitive. Another study to define the most suitable monoclonal antibodies is planned.

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

There are no generally accepted diagnostic tests for cat-scratch disease (CSD), the cause of which is unknown. During the development of an indirect fluorescence antibody (IFA) test for detection of antibodies to "Rochalimaea henselae", sera from patients with CSD were found to have high titres to R henselae antigens. Further tests with this assay showed that 36 (88%) of 41 patients with suspected CSD had serum titres of 64 or more to R henselae antigen, that there was a low prevalence (3%) of substantial titres to R henselae in healthy controls (n = 107), and that the immune responses to R henselae antigens were specific. These data suggest that the R henselae IFA test, or other serological assays based on R henselae, may be useful for diagnosis of CSD.

Immunoaffinity purification, stabilization and comparative characterization of listeriolysin O from Listeria monocytogenes serotypes 1/2a and 4b.

We developed a simple and highly effective procedure for stabilizing the haemolytic activity of listeriolysin O (LLO) from Listeria monocytogenes after immunoaffinity purification. The haemolytic activity of LLO was stabilized by eluting it directly into tubes containing an alkaline buffer (5 mM lysine, 140 mM KCl, 50% ethylene glycol, pH 11.5). The purified LLO retained 100% of its haemolytic activity after 6 weeks of storage at -20 degrees C. LLO purified from a strain of L. monocytogenes serotype 1/2a (ATCC 43249) and LLO purified from a strain of L. monocytogenes serotype 4b (F 2365) isolated from a Mexican-style cheese, showed no significant differences in pH and temperature stability. When incubated in buffers at pH values from 4 to 12 at 4 degrees C and 25 degrees C, LLO from serotypes 1/2a and 4b retained maximal haemolytic activity at pH 8 after 4 h of incubation. LLO from both serotypes lost their haemolytic activity after incubation at 50 degrees C for 25 min.

Characterization of Neisseria meningitidis serogroup C by multilocus enzyme electrophoresis and ribosomal DNA restriction profiles (ribotyping).

We compared multilocus enzyme electrophoresis (MEE) and ribosomal DNA fingerprinting (ribotyping) for subtyping 44 strains of Neisseria meningitidis serogroup C that were isolated in Los Angeles County, California, between December 1985 and July 1986. The isolates were divided into six enzyme types (ETs) by MEE, but 36 of the isolates were clustered in one ET, 3. The same isolates were divided into 17 ribotypes by use of restriction endonucleases ClaI, EcoRI, and XhoI. Twenty of the 36 ET 3 isolates were divided into 17 ribotypes by use of restriction endonucleases ClaI, EcoRI, and XhoI. Twenty of the 36 ET 3 isolates were grouped in a single ribotype, J. The rate of infection with ribotype J strains was higher in the southern part of the study area than in the northern part. Isolates from each of eight pairs (each isolate pair was cultured from the same patient from the same or different sites) were found identical by MEE, but ribotyping revealed a difference in one pair. In this study, ribotyping showed a greater discriminating capacity than MEE for subtyping N. meningitidis serogroup C, but the epidemiologic relevance of this increased sensitivity needs further assessment.

Multicenter comparison of levels of antibody to the Neisseria meningitidis group A capsular polysaccharide measured by using an enzyme-linked immunosorbent assay.

There is no standard immunoassay for evaluating immune responses to meningococcal vaccines. We developed an enzyme-linked immunosorbent assay to measure total levels of antibody to Neisseria meningitidis group A capsular polysaccharide. Five laboratories measured the antibody levels in six paired pre- and postvaccination serum samples by using the enzyme-linked immunosorbent assay. Methylated human serum albumin was used to bind native group A polysaccharide to microtiter plate surfaces. The between-laboratory coefficients of variation for pre- and postvaccination sera had ranges of 31 to 91 and 17 to 31, respectively. The mean laboratory coefficients of variation for pre- and postvaccination sera, respectively, were 17 and 11 (Molecular Biology Laboratory, Centers for Disease Control), 12 and 15 (Immunodiagnostic Methods Laboratory, Centers for Disease Control), 22 and 19 (Dana-Farber Cancer Institute), 38 and 38 (Bacterial Polysaccharide Laboratory, U.S. Food and Drug Administration), and 11 and 10 (Praxis Biologics, Inc.). Standardization of this enzyme-linked immunosorbent assay should allow interlaboratory comparison of meningococcal vaccine immunogenicity, thus providing a laboratory-based assessment tool for evaluating meningococcal vaccines.

Proposal of Afipia gen. nov., with Afipia felis sp. nov. (formerly the cat scratch disease bacillus), Afipia clevelandensis sp. nov. (formerly the Cleveland Clinic Foundation strain), Afipia broomeae sp. nov., and three unnamed genospecies.

On the basis of phenotypic characterization and DNA relatedness determinations, the genus Afipia gen. nov., which contains six species, is described. The type species is Afipia felis sp. nov. (the cat scratch disease bacillus). Afipia clevelandensis sp. nov., Afipia broomeae sp. nov., and three unnamed not associated with cat-borne disease. All but one strain (Afipia genospecies 3) were isolated from human wound and respiratory sources. All Afipia species are gram-negative, oxidase-positive, nonfermentative rods in the alpha-2 subgroup of the class Proteobacteria. They are motile by means of a single flagellum. They grow on buffered charcoal-yeast extract agar and nutrient broth, but rarely on MacConkey agar, at 25 and 30 degrees C. They are urease positive; but they are negative in reactions for hemolysis, indole production, H2S production (triple sugar iron agar), gelatin hydrolysis, esculin hydrolysis, and peptonization of litmus milk. They do not produce acid oxidatively from D-glucose, lactose, maltose, or sucrose. The major cell wall fatty acids are 11-methyloctadec-12-enoic (CBr19:1), cis-octadec-11-enoic (C18:1omega7c), and generally, 9,10-methylenehexadecanote and 11,12-methyleneoctadecanoate; and there are only trace amounts of hydroxy acids. The guanineplus-cytosine content is 61.5 to 69 mol%. A. felis is positive for nitrate reduction and is delayed positive for acid production from D-xylose, but it is catalase negative. A. clevelandensis is negative in all of these tests. A. broomeae is weakly positive for catalase production and acid production from D-xylose, but it is negative for nitrate reduction.

Meningococcal disease in the United States--1986. Meningococcal Disease Study Group.

Active surveillance for invasive meningococcal disease was conducted during 1986 and 1987 in six areas of the United States with a total population of approximately 34 million persons. The incidence of meningococcal disease was 1.3:10(5). The highest incidence of disease among the surveillance areas was in Los Angeles County (1.65:10(5). Neisseria meningitidis serogroups B and C caused about equal amounts of disease, which reflects a recent increase in the incidence of group C disease. Group C caused more than half of the cases of meningococcal disease in Los Angeles and Tennessee but less than one-third of the cases in Missouri and Oklahoma. Multilocus enzyme electrophoresis demonstrated that a group of closely related isolates of N. meningitidis was prevalent in Los Angeles during the surveillance period and was associated with an increased incidence of meningococcal disease there.

The epidemiology of listeriosis in the United States--1986. Listeriosis Study Group.

To determine the morbidity and mortality due to listeriosis in the United States, the authors undertook an active surveillance project in 1986 to identify all cases in which Listeria monocytogenes was isolated from cultures of ordinarily sterile sites in a population of 34 million persons. The authors estimated that at least 1,700 cases of listeriosis and 450 deaths occurred in the United States in 1986; 27% of these cases occurred in pregnant women, with 22% of perinatal cases resulting in stillbirths or neonatal deaths. The risk of listeriosis in adults (0.5 per 100,000 population) was similar in all regions studied; the incidence of perinatal listeriosis was three times higher in Los Angeles County, California, than in the other areas (24.3/100,000 live births vs. 7.8/100,000 live births). Geographic variation may have resulted from underdiagnosis of perinatal listeriosis in five of the study areas. Multilocus electrophoretic enzyme typing was useful for elucidating the molecular epidemiology of L. monocytogenes; perinatal listeriosis was significantly associated with one group of related strains. Multilocus electrophoretic enzyme typing also identified three clusters representing possible common-source outbreaks. These findings document the substantial morbidity due to listeriosis in the United States; to the extent that sporadic listeriosis is foodborne, this morbidity could be reduced by appropriate preventive measures, particularly in persons known to be at increased risk of infection.