Secondary Logo

Journal Logo

Transmission, Pathogenesis, and Epidemiology of Listeria Monocytogenes

Buchholz, Udo; Mascola, Laurene

Infectious Diseases in Clinical Practice: January 2001 - Volume 10 - Issue 1 - p 34-41
Mini Review
Free

MORE THAN 70 years after the discovery of the organism in 1929, Listeria monocytogenes (LM) still presents a challenge for clinicians, microbiologists, and epidemiologists. Although important advances have allowed the reduction of the incidence of LM, this organism alone accounts for an estimated 38% of deaths caused nationwide by food-borne bacterial pathogens [1]. This article elaborates on the important role clinical care providers and public health workers have in addition to patient treatment. The routine provision of food safety education to patients along with reporting of listeriosis cases to the health department are equally critical. Current data from the Los Angeles County surveillance system for listeriosis are used to illustrate the epidemiology of LM known today.

Back to Top | Article Outline

The Organism and Its Reservoir

Listeria monocytogenes (LM) is a gram-positive, facultatively anaerobe, non-spore-forming rod. The most common serotypes are 1/2a, 1/2b, and 4b which account for 95% of all isolates [2]. Characteristically, listeria survives and even proliferates at refrigerator temperature. LM is destroyed through pasteurization. The organism is ubiquitous and can be found in the inanimate world in soil, water, sewage, and silage as well as in a wide variety of plants and animals commonly used for human consumption, such as cattle, pigs, chicken, and seafood. A Dutch study found listeria in 21% of households in common home items such as dishcloths and bathrooms [3].

Back to Top | Article Outline

Transmission

Various modes of transmission have been described. The most important route is through the ingestion of LM in food. Given the ubiquity of LM, it is probable that food-borne exposure to listeria is a routine event for all humans. LM can also be transmitted vertically from mother to fetus both in utero and peripartum. Other, less frequent possible modes include animal-to-person transmission, resulting for example in cutaneous lesions among veterinarians, or nosocomial transmission, as documented in an outbreak in a newborn nursery through the use of contaminated mineral oil [4].

Back to Top | Article Outline

Clinical and Public Health Course of a Listeria Infection

Overview

After the ingestion of LM, the following scenarios can occur. The majority of times, the organism is soon cleared from the gastrointestinal tract (Fig. 1). Alternatively, LM may cause illness after an incubation period of 11–71 days (median 31 days) [5] or the organism may remain intracellular. Which occurs in an individual depends on the combined effects of the classical three epidemiologic variables—vector, agent, and host, or inoculum size in contaminated food, virulence factors of the organism, and host risk factors.

FIGURE 1.

FIGURE 1.

Disease attributed to LM is traditionally divided into perinatal and nonperinatal listeriosis. Perinatal listeriosis is usually defined as infection in either a fetus/newborn, mother, or both. From 1986 through 1999, the 446 (61%) of 728 Los Angeles County (LAC) cases of listeriosis were nonperinatal and 39% were perinatal. Disease syndromes for nonperinatal and perinatal listeriosis along with public health follow-up and interventions are explained below.

Back to Top | Article Outline

The vector: foods

Organisms 109 are estimated to cause infection. Although normally listeriosis is a disease of patients with T-cell mediated immunocompromise, sufficiently contaminated food can cause illness also in immunocompetent individuals [6,7].

The food responsible for the largest North American outbreak was Mexican-style cheese. In May 1985, an infection-control practitioner of a Los Angeles County Hospital notified the health department (HD) about a high number of infections with LM among Hispanic pregnant women. A rapid and vigorous investigation implicated Mexican-style cheese [5].

The second largest listeriosis outbreak in the United States occurred in 1998/1999 [8]. After a review of national surveillance data showed that four states had a listeriosis case count elevated above baseline, subsequent molecular fingerprinting, known as pulsed field gel electrophoresis (PFGE), discovered that 20 of 41 strains of listeriosis cases were similar. A multistate case-control study implicated the consumption of hot dogs within the month before illness resulting in a recall of 35 million pounds of hot dogs. At the end, 101 cases of listeriosis from 22 states were counted with 15 adult deaths and six fetal demises.

At least two outbreaks implicated vegetable sources. The first one occurred in Nova Scotia, Canada, in 1981 caused by coleslaw [6]. The second outbreak involving vegetables was reported from the Piedmont region in Italy, the largest listeriosis outbreak to date [7]. In 1997, 1566 persons became ill with febrile gastroenteritis, an unusual clinical manifestation of listeriosis, from eating canned corn and tuna salad.

Additional foods associated with outbreaks included ready-to-eat meat products, such as pork tongue, dairy products, including soft cheese, chocolate and regular milk, butter, and smoked rainbow trout [3].

In the United States, the Food and Drug Administration (FDA) monitors the presence of LM in ready-to-eat foods. Between 1986 and 1997, the FDA recalled approximately 2000 products for an average of about 160 products per year. The average annual number of companies with a recall attributed to LM in the above time period was 21. Each of these recalled contaminated foods could have potentially caused outbreaks. In those 12 years, 26 companies issued a recall in more than 1 year. Foods most frequently recalled were salad, sandwiches, sliced fruits or vegetables, followed by ice cream products, dairy, and seafood. Since 1990, the United States has adopted a “zero-tolerance” policy for the occurrence of listeria in food. Other countries have different policies dealing with LM contamination. For example, in Canada, ready-to-eat foods that have not been associated with an outbreak and do not allow growth of LM during a 10-day period of refrigerated storage may contain up to 100 LM organisms per gram [9]. Denmark has six classes of foods that have to meet various criteria for LM. In raw ready-to-eat foods, for example, two of five samples can contain between 10 and 100 organisms per gram and no sample can exceed 100 organisms per gram. LAC surveillance data since 1986 show that a decrease of the incidence of listeriosis coincides with the implementation of both public health food avoidance guidelines along with the zero-tolerance policy (Table 1). For 1999, the incidence of listeriosis in LAC was reported as eight perinatal cases per 100,000 live births and 0.2 nonperinatal cases per 100,000 population (Figure 2).

TABLE 1

TABLE 1

FIGURE 2.

FIGURE 2.

Back to Top | Article Outline

Listeriosis through “activation” of intracellular organisms

Although most outbreaks were associated with exposure to contaminated food, often additional factors must be present to cause illness due to LM. Carriage studies have recovered LM in up to 10% of stool specimens of healthy individuals [10], indicating that the presence of LM in the gastrointestinal tract alone (as well as in vaginal flora) does not necessarily cause illness. Another interesting study examined the refrigerator contents of patients with sporadic listeriosis. Although at least one food specimen grew LM in almost ⅔ of refrigerators of all listeriosis patients, in only ⅓ of listeriosis patients was it possible to identify the same strain as in the patient [11]. Although this was significantly higher than expected by chance alone, it is consistent with the possibility that a proportion of patients with listeriosis become colonized with LM months or longer before disease onset, explaining why no specific food item or no food with a LM strain identical to the patient’s was identified.

An outbreak of listeriosis in Philadelphia in 1986/87 provides additional insight into the pathogenesis of listeriosis: the epidemiologic investigation was unable to confirm the same outbreak strain among case-patients or to implicate a single food. However, the investigators speculated that gastrointestinal coinfections may have precipitated invasion of commensal LM [12]. Finally, numerous published and unpublished case reports describe the occurrence of invasive listeriosis after events that disrupt the gastrointestinal micro-environment, such as colonoscopy, shigellosis, colon cancer, or onset of ulcerative colitis. Because of its intracellular status, LM organisms survive the toxic effects of the host cells’ phagolysosomes for a prolonged period of time, until either systemic or local factors contribute to its subsequent invasion. Therefore, pinpointing an exact exposure date or food vehicle for the occurrence of sporadic LM remains problematic.

Back to Top | Article Outline

The host: risk factors for perinatal and nonperinatal listeriosis

In cases of sporadic listeriosis, risk factors play a major role, with a recognized risk found in 80% to 90% of all cases. During pregnancy, suppression of cell-mediated immunity permits the presence and growth of the genetically foreign fetus while also increasing maternal susceptibility for LM. Furthermore, LM has a predilection for the placenta; therefore, pregnant women with multiple gestations seem to be at higher risk for listeriosis than pregnant women with singletons [13].

Risk factors for nonperinatal listeriosis can be divided into two categories: systemic and local (Fig. 1). Systemic risks are those that suppress immunity. Listeriosis incidence directly increases with age, rising dramatically after 60 years of age [14]. Other risk factors include steroid or other immunosuppressive therapy, malignancies and their treatment, human immunodeficiency virus infection, diabetes, renal failure, congestive heart failure, organ transplantation, liver cirrhosis, alcoholism, autoimmune diseases, status post gastrectomy or medication with antacids, histamine-2 antagonists, or proton pump inhibitors. Intravenous drug users are at increased risk if the heroin is cut and adulterated with plants or earth. Local risk factors include situations that disrupt the local gastrointestinal flora, as mentioned above.

Both systemic and local risk factors may work additively and can promote development of listeriosis in immunocompetent patients at any age. We recently noted a 15-year-old patient who presented with bloody diarrhea, with two negative sets of routine stool cultures. Further medical work-up and interventions included a colonoscopy (diagnosing ulcerative colitis) with subsequent treatment with corticosteroids and cyclosporin. One month after diarrheal onset, the patient became septic and a blood culture yielded LM.

Back to Top | Article Outline

Syndromes and presentations of perinatal and nonperinatal listeriosis

Perinatal listeriosis has various presentations. The mother, if affected at all, is usually only mildly ill, exhibiting only flu-like illness, fever, headache, and myalgia. Diagnosis of infected pregnant women is critical because appropriate antibiotic treatment for listeriosis can save the fetus. Thus, febrile pregnant women should always have blood cultures performed. Infection in the fetus or neonate is mostly severe and can result in abortion, stillbirth, and if born alive, in early-or late-onset neonatal listeriosis. Intrauterine infection of the fetus leads to early-onset listeriosis and presents either as septicemia, or less frequently as granulomatosis infantiseptica. In contrast, late-onset listeriosis develops after the neonate is exposed to LM during labor and delivery. The initially healthy newborn develops meningitis several days to weeks after delivery. In LAC, we found that among all perinatal cases with known outcome (n = 365), 71% of neonates were born alive and survived, whereas 6% resulted in abortion, 12% in stillbirth, and 11% in neonatal death. Since 1986, prognosis has not improved for the fetus. All mothers survived.

The typical presentation of nonperinatal listeriosis is either meningitis or nonspecific septicemia in an elderly or immunosuppressed patient with some significant comorbidity, such as diabetes or cancer. Prodromic symptoms, such as flu-like illness, back pain, myalgia, nausea, and diarrhea may occur. The presentation of gastrointestinal symptoms, such as diarrhea, may be part of the listerial prodromic spectrum or a gastrointestinal coinfection, such as a viral or bacterial infection, which facilitated subsequent disease.

For clinical purposes, the simultaneous infection of the meninges and brain parenchyma as well as subcortical brain abscesses should prompt strong consideration of LM as etiologic agent [14]. Additional clinical pictures associated with listeriosis include rhombencephalitis, endocarditis, arterial infections (at susceptible sites, such as aneurysms), hepatitis and liver abscesses, peritonitis (either spontaneous or in continuous ambulatory peritoneal dialysis patients), osteomyelitis and septic arthritis, pleuropulmonary infections, and wound and skin infections (i.e., arteriovenous shunt sites of dialysis patients).

One clinical manifestation of listeriosis that has recently gained more attention is that of febrile gastrointestinal illness. The Italian listeria gastroenteritis outbreak affected 1566 persons; 292 (19%) of them were hospitalized [7] Almost all of the cases were immunocompetent. The most frequent symptoms were headache (88%), abdominal pain (72%), and fever (68%). The median incubation time was measured as 24 hours. National statistics show that 65% of all food-borne outbreaks reported to the Centers for Disease Control and Prevention (CDC) with an incubation period of greater than 6 hours (Michelle Young, CDC, personal communication) have unidentified etiologies. Therefore, testing of stool samples for LM in food-borne outbreaks of febrile gastroenteritis, where routine cultures fail to yield a pathogen, may substantially increase the number of identified LM gastroenteritis outbreaks. However, special transport and media requirements have to be applied when obtaining and culturing feces (see below).

In LAC, since 1986, the overall case-fatality among patients with nonperinatal listeriosis with known outcome was 30% (128/429), but case fatality has improved from 40% in 1986 to 10% in 1999.

Back to Top | Article Outline

Laboratory diagnosis of listeriosis

Typically, LM is easily diagnosed from sterile site specimens, such as blood or cerebral spinal fluid. Serologic tests are considered either insensitive or not specific enough to be useful in clinical practice. If LM is suspected in cases of gastroenteritis, it can be isolated from stool, but the specimen needs to be transported under cool conditions and should arrive in the laboratory within 4 hours of collection. If transport is delayed, it can be cooled for up to 48 hours at 4°C (Joan Sturgeon, personal communication, 2000). The stool sample also needs to be submitted either “unpreserved” or in special enrichment broths. Because selective media used for stool cultures normally suppress the growth of listeria, lithium chloride-phenylethanol-moxalactum media must be used.

Back to Top | Article Outline

Treatment of listeriosis

Because of the intracellular existence, the recommended treatment for listeriosis is a 3-week course of ampicillin (Tables 2 and 3). Although the value of the addition of an aminoglycoside, such as gentamicin, is debated, it is frequently used in addition. During epidemics of listeria, the CDC does not recommend that persons, even those in a high-risk group, obtain prophylaxis or specific laboratory testing if asymptomatic. Those individuals in a high-risk group who have eaten a known contaminated product, and who become ill with fever or signs of illness, are advised to contact their physician and inform him or her about this exposure [17]. Recommendations for treatment of healthy patients with febrile listeria gastroenteritis are nonexistent. One reviewer recently stated that it might be prudent to consider treatment in patients with known immunocompromise to prevent invasive disease [18].

TABLE 2

TABLE 2

TABLE 3

TABLE 3

Back to Top | Article Outline

Reporting of cases of listeriosis to the HD and public health follow-up

In California and other states, cases of listeriosis are reportable to the HD, which receives reports by telephone, as written confidential morbidity reports, or via fax. A recent evaluation of the LAC listeria surveillance system demonstrated that completeness of reporting is approximately 80%, typically with the hospital infection-control practitioner reporting the case. However, physicians are legally responsible for disease reporting. Clinicians may under-appreciate the important role of infectious-disease reporting and the detailed chain of events that the report triggers (Fig. 1). For reported cases of listeriosis, a public health response may consist of the following: (1) interview of the patient by a public health nurse, (2) statistical case analysis, (3) outbreak investigations, and (4) molecular analysis of human isolates. The following paragraphs illustrate potential lifesaving consequences that the report of even a single case of listeriosis can have.

Back to Top | Article Outline

Interview of the patient by a public health nurse

The HD assigns the case to a public health nurse who contacts the patient and conducts an in-depth interview using a standardized investigation form. The public health nurse gathers demographic information (place of residence, age, gender), the type of listeriosis (perinatal/nonperinatal), clinical details, and risk factors (food eaten in the incubation period, medical risk factors [preexisting illnesses, prior surgeries, medications]). The listeria isolate is requested to be sent to the Public Health Laboratory.

Back to Top | Article Outline

Ongoing and periodic review and statistical evaluation of listeriosis cases

HDs often have dedicated listeriosis surveillance officers who review the data routinely to discover potential disease outbreaks. However, frequently it is the astute health care professional who is the first to notice and report a clustering of cases.

Another important function of listeriosis surveillance data may be performed by local, state, or national surveillance officers. For example, epidemiologic data presented throughout this article can only be derived from a systematic and ongoing collection of listeriosis cases and may assist in the development of preventive guidelines, clinical recommendations, and the evaluation of public health interventions, such as the FDA “zero-tolerance” policy for food.

Back to Top | Article Outline

Outbreak investigations

If a cluster of listeriosis cases warrants closer investigation, an epidemiologic outbreak investigation will confirm the existence of an outbreak, ascertain cases, and develop hypotheses. The implication of a specific food usually through a case-control study ideally leads to a recall of the product and the removal of potentially contaminated products already on the shelf.

Back to Top | Article Outline

Fingerprinting of recovered LM strains

In California, LM human isolates are sent to the Public Health Laboratory for PFGE analysis. Often, successful outbreak investigations are able to match the PFGE DNA fingerprint pattern of patient isolates with those recovered from implicated food (Fig. 1; 4a) [7,8]. A comparison of these strains with a strain recovered from a food item epidemiologically implicated in the outbreak investigation (Fig. 1; 4b) may result in a recall of contaminated food. It also serves as an excellent discriminatory tool to distinguish or match LM strains with each other in non-outbreak situations.

An award-winning initiative by the CDC called PulseNet, enables an increasing number of state and selected county health laboratories to submit electronically standardized molecular fingerprinting of food-borne pathogens, such as LM, to detect similarities and discover, for example, multistate outbreaks. Molecular typing of strains of LM isolated from recalled food is also submitted routinely. The power of this new tool was demonstrated recently when the comparison of LM strains at CDC found that a single patient’s strain of LM submitted from LAC (Fig. 1; 5a), matched the strain from a recalled smoked fish product (Fig. 1; 5b) in the Eastern United States. The company had been cited previously but without actual product linkage to a human illness. A revised interview of the LAC patient revealed that she was able to remember marketing details of the fish product that matched with the recalled product.

Back to Top | Article Outline

Summary

Although the public health importance of listeriosis may seem insignificant because of its low incidence and long incubation time, this article illustrates how critical it is to take LM into account both clinically and epidemiologically. First, every pregnant woman with a febrile illness may be ill with listeriosis with potential deleterious consequences for the unborn baby. These can be avoided with prompt treatment of the mother with adequate antibiotics. Second, the high casefatality of listeriosis merits every effort to diagnose LM swiftly. Third, epidemiologists and laboratorians demonstrate how powerful the alliance of clinicians with public health professionals can be. Several outbreak investigations have shown that the prompt recognition of outbreaks can effectively reduce the number of cases. Finally, in the time of DNA fingerprinting, occasionally only a single reported case of listeriosis may suffice to identify foods with proven human hazard which can facilitate the effective regulation of contaminated food.

Back to Top | Article Outline

Acknowledgments

We thank Tina Rouse, FDA, for her assistance to make data on food recalls available to us, Joan Sturgeon, LAC Department of Health Services, for her microbiological advice, and Anchalee Chillanond for her secretarial assistance.

Back to Top | Article Outline

References

1. Mead PS, Slutsker L, Dietz V, McCaig LF, Bresee JS, Shapiro C, et al. Food-related illness and death in the United States. Emerg Infect Dis 1999; 5:607–25.
2. Broome CV, Pinner R, Schuchat A. Listeria monocytogenes. In: Gorbach SL, Bartlett JG, Blacklow NR, editors. Infectious diseases. 2nd ed. Philadelphia: WB Saunders; 1998. p. 1751–5.
3. Janda JM, Abbott SL. Unusual food-borne pathogens:Listeria monocytogenes, Aeromonas, Plesiomonas, and Edwardsiella species. Clin Lab Med 1999; 19:553–82.
4. Schuchat A, Lizano C, Broome CV, Swaminathan B, Kim C, Winn K. Outbreak of neonatal listeriosis associated with mineral oil. Pediatr Infect Dis J 1991; 10:183–9.
5. Linnan MJ, Mascola L, Lou XD, Goulet V, May S, Salminen C, et al. Epidemic listeriosis associated with Mexican-style cheese. N Engl J Med 1988; 319:823–8.
6. Schlech III, WF Lavigne PM, Bortolussi RA, Allen AC, Haldane EV, Wort AJ, et al. Epidemic listeriosis: evidence for transmission by food. N Engl J Med 1983; 308:203–6.
7. Aureli P, Fiorucci GC, Caroli D, Marchiaro G, Novara O, Leone L, et al. An outbreak of febrile gastroenteritis associated with corn contaminated by Listeria monocytogenes. N Engl J Med 2000; 342:1236–41.
8. Update: multistate outbreak of listeriosis–United States, 1998–1999. Morb Mortal Wkly Rep 1999;47:–8.
9. Farber JM, Ross WH, Harwig L. Health risk assessment of Listeria monocytogenes in Canada. Int J Food Microbiol 1996; 30:145–56
10. Mascola L, Sorvillo F, Goulet V, Hall B, Weaver R, Linnan M. Fecal carriage of Listeria monocytogenes: observations during a community-wide, common-source outbreak [letter]. Clin Infect Dis 1992; 15:557–8.
11. Pinner RW, Schuchat A, Swaminathan B, Hayes PS, Deaver KA, Weaver RE, et al. Role of foods in sporadic listeriosis. II. Microbiologic and epidemiologic investigation. JAMA 1992; 267:2046–50.
12. Schwartz B, Hexter D, Broome CV, Hightower AW, Hirschhorn RB, Porter JD, et al. Investigation of an outbreak of listeriosis: new hypotheses for the etiology of epidemic Listeria monocytogenes infections. J Infect Dis 1989; 159:680–5.
13. Mascola L, Ewert DP, Eller A. Listeriosis: a previously unreported medical complication in women with multiple gestations. Am J Obstet Gynecol 1994; 170:1328–32.
14. Lorber B. Listeriosis. Clin Infec Dis 1997; 24:1–11.
15. Silver HM. Listeriosis during pregnancy. Obstet Gynecol Surv 1998; 53:737–40.
16. Bortolussi R, Kennedy W. Bacteria: aerobic gram-positive bacilli. In: Armstrong D, Cohen J, editors. Infectious diseases. Philadelphia: Mosby; 1999. p. 6–7.
    17. Cited October 17, 2000. Available at http://www.cdc.gov/ ncidod/dbmd/diseaseinfo/listeriosis_g.htm.
    18. Davis B. Medscape Infectious Diseases Journal Scan. vol. 3, no. 5. Available at http://www.medscape.com/medscape/fe.../JournalScan/ID2000/js-id0305.html. An outbreak of febrile gastroenteritis associated with corn contaminated with Listeria monocytogenes. Cited October 17, 2000.
    © 2001 Lippincott Williams & Wilkins, Inc.