In this Primer, we review the epidemiology of listeriosis, from both a bacterial perspective (reservoirs, global distribution and population structure including hypovirulent and hypervirulent strains) and a host perspective (risk factors and disease outcomes), its pathophysiology, its clinical and microbiological diagnosis and management, and strategies for its treatment and prevention.
The ability of L. monocytogenes to grow at temperatures ranging from 0 °C to 45 °C, as well as to tolerate a wide range of pH (3.0-9.0) and osmolarity levels (up to 10% salt concentration), explains its ubiquitousness in the environment. Livestock, especially cattle, can be contaminated, either as asymptomatic carriers or by developing listeriosis, and contribute to the persistence, amplification and spread of L. monocytogenes in dairy products and the farm environment. Additionally, domestic and wild mammals and birds can spread L. monocytogenes in their faeces, contaminating the environment or animal feed. L. monocytogenes enters the food chain through contaminated raw materials, animal products such as raw milk and environmental surfaces in food-processing environments. Cross-contamination during food processing, inadequate sanitation and poor temperature control also facilitate its spread. As L. monocytogenes can grow slowly at refrigeration temperatures and at a wide range of pH and salt concentrations, it can persist in food production chains and be found in many types of food, particularly ready-to-eat (RTE) products such as deli meats, dairy products including soft cheeses, pâtés and smoked fish. Fruits and vegetables can also be contaminated during food processing and distribution, or in domestic refrigerators, as can vegan substitutes for milk and cheese.
The consumption of contaminated food is almost the only source of human infection with L. monocytogenes. The only known cases of inter-host transmission of L. monocytogenes include: vertical from pregnant women to their fetus via the placenta; very rare cases of contamination in farmers and veterinarians through direct contact with tissues of infected animals, leading to cutaneous infection; and horizontal faecal-oral transmission in hospitals, which has been reported in neonates, due to their immature microbiota.
In the European Union (EU) and the USA, reported positive samples ranged from 0% to 3% across most food products. A 2019 meta-analysis of international data indicated a prevalence of 2.9% in deli meat, 2.4% in soft cheese and 2.0% in packaged salad, with 2% (salad) to 25% (soft cheese) of positive samples above the regulatory food safety limit criterion of 100 colony-forming units (CFU) per gram. Although transient faecal or asymptomatic carriage of L. monocytogenes in humans is not uncommon, with L. monocytogenes prevalence in faecal culture-based studies ranging between 0.2% and 5%, and L. monocytogenes nucleic acids detectable in faeces in up to 10% of the general population, human cases of listeriosis are rare and associated with predisposing conditions. Indeed, despite the relatively high number of contaminated food portions (mass of RTE food ingested per meal, estimated at 55 million above 100 CFU/g consumed by the population of those over 75 years of age in the EU each year), the incidence of confirmed cases remains low. Listeriosis is thought to occur worldwide (Fig. 1) but its prevalence and incidence are only reliably known in countries with mandatory reporting systems. In 2023, there were 0.66 cases of listeriosis per 100,000 people across the EU (2,952 cases, of which 96.5% resulted in hospitalization) and 0.31 cases per 100,000 people in the USA, and the worldwide incidence was estimated at 0.337 per 100,000 people in 2010 (ref. ). In countries with mandatory reporting systems, most cases of listeriosis are sporadic. Indeed, epidemiological surveillance and whole-genome sequencing-based microbiological surveillance allow clusters of cases to be detected, contamination sources to be identified, and large listeriosis outbreaks to be prevented. Studies aimed at providing a global view of L. monocytogenes clonal diversity have shown its worldwide distribution and the existence of a few predominant and globally distributed clones. For example, lineage I clonal complexes (CC) CC1, CC4 and CC6 strains are the most frequently associated with clinical isolates in Western countries. Sequence type (ST) 87 (ST87) is the most common L. monocytogenes clinical ST in East Asia and ST328 is the most common in India, with these strains rarely reported in Europe and North America. These epidemiological findings highlight the need for high-quality surveillance systems for listeriosis, especially in parts of the world where little is currently known about L. monocytogenes prevalence and listeriosis incidence.
It is essential to implement mandatory reporting of all invasive listeriosis cases (septicaemia, CNS infection and maternal-fetal infection), along with submitting the isolate to a reference laboratory. Microbiological surveillance of at-risk food production and distribution sites is also essential. The public health cornerstone of listeriosis surveillance is the early detection of clusters of cases associated with an isolate, and the identification of the source of contamination using genomic typing, in order to avoid large outbreaks (see below) (Box 1).
Listeriosis occurs mainly in older adults and/or immunocompromised patients. Pregnancy is also a major risk factor for listeriosis (relative risk 30-110 (refs. )), with newborns being a population at increased risk of infection. Newborns are infected prenatally, via the transplacental route, but in rare cases can become infected in the first days of life, due to their immature microbiota which leads to low neonatal gut resistance to L. monocytogenes colonization. A higher prevalence has been reported in pregnant women in England and Wales in those with low socioeconomic status than in those with high socioeconomic status, as well as among groups with specific dietary preferences, such as Hispanic women in the USA who frequently consume unpasteurized Mexican-style cheese.
Other common risk factors for listeriosis include: male sex; older age (over 65 years of age, with much increased prevalence in those over 80 years of age), which may reflect immunosenescence and a higher frequency of comorbidities compared with younger individuals; acquired cellular and/or innate immunodeficiency; solid organ or bone marrow transplantation; HIV infection; haematological malignancies (especially lymphoproliferative haemopathies, such as chronic lymphocytic leukaemia, lymphoma or multiple myeloma); solid organ cancer; diabetes; chronic renal failure or dialysis; or cirrhosis. The use of immunosuppressive drugs such as corticosteroids, chemotherapeutic agents and other immunosuppressive biotherapies, such as anti-TNF monoclonal antibodies, are also associated with increased risk of listeriosis. The most common immunosuppressive comorbidities in the French MONALISA national prospective cohort study (818 total patients between 2009 and 2013) were solid organ cancer (in 31%) and diabetes (in 22%). A French retrospective cohort study including 1,959 patients between 2001 and 2008 demonstrated a 1,139-fold increase in listeriosis in patients with chronic lymphocytic leukaemia and a 350-fold increase in patients with multiple myeloma. This study also suggested that the risk of listeriosis, compared with a control population under 65 years of age, was 361 times higher in patients undergoing dialysis, 356 times higher in patients with giant cell arteritis and an oral corticosteroids dose above 0.5 mg/kg/day, 78 times higher in patients with a solid tumour, and 20 times higher in patients over 74 years of age. Finally, the existence of a subset of patients (4-10%) who develop neurolisteriosis without any known risk factor raises questions about the possibility of yet-unidentified host susceptibility factors, which are currently being investigated (NCT03357536).
Invasive listeriosis is associated with poor outcomes. A meta-analysis estimated that listeriosis caused 23,150 illnesses (95% credible interval 6,061-91,247), 5,463 deaths (1,401-21,497) and 172,823 disability-adjusted life-years (DALYs) (44,079-676,465) worldwide in 2010 (ref. ) (Fig. 1a). Isolated septicaemia (with no other clinically apparent infection location) is associated with a 3-month mortality of up to 46%, despite appropriate antibiotic therapy, and neurolisteriosis with a 3-month mortality of 13% to 40% depending on the absence or presence of concomitant bacteraemia, respectively. This might also reflect, at least in part, the severity of patient comorbidities including malignancy and immunosuppression. If left untreated, neurolisteriosis is fatal. Importantly, the mortality associated with other focal invasive L. monocytogenes infections is also very high, as exemplified by the 3-month mortality of L. monocytogenes-associated spontaneous bacterial ascites (52%) or L. monocytogenes-associated endocarditis (41%). Parameters independently associated with increased 3-month mortality include monocytopenia, ongoing neoplasia, concomitant multi-organ failure or worsening of a pre-existing condition. In addition, persistent neurological impairment is reported in 44% of patients recovering from neurolisteriosis. Such impairments include (1) persistent focal motor deficits, sensory loss, seizures, altered consciousness or memory loss and (2) the number of neurological signs at baseline, which are two parameters that are independently associated with persistent neurological impairment (OR 21.65 (95% CI 2.58-181.59) and 1.37 (95% CI 1.11-1.69), respectively).
In a 2022 prospective study of the outcomes of neonatal listeriosis, in-hospital mortality was 5%, which is much lower than previously reported in the 1990s and 2000s. Improved survival rates are likely to reflect major improvements in neonatal intensive care, particularly for premature infants. A study of long-term neurological outcomes showed that 66% of children that survived neonatal listeriosis had persistent neurological impairment at a median age of 5 years, with 18% having severe impairment. Impairments include cognitive deficiencies, reduced executive function, and sensory or motor impairments. Rather than listeriosis itself, gestational age at birth (that is, prematurity due to infection), seemed to be the main determinant of neurological impairment. Indeed, the neurological and neurodevelopmental outcomes of children with neonatal listeriosis did not differ from those of gestational age-matched control children without infection from a contemporary national cohort. These data support the implementation of systematic long-term screening for this vulnerable population, and the provision of tailored education and support.
L. monocytogenes strains were first subdivided into 14 serovars by serotyping based on Listeria somatic and flagellar antigens. The L. monocytogenes species was next divided into four evolutionary lineages based on multilocus enzyme electrophoresis (lineages I and II) and multilocus genotyping (lineages III and IV). A multilocus sequence typing (MLST) approach based on the allelic variation of seven housekeeping genes allowed identification of sequence types and the definition of CCs that have at least six alleles in common. A core genome MLST (cgMLST) scheme, based on the sequencing of 1,748 core loci, is now widely used internationally to group strains into sublineages (with up to 150 allelic differences) and cgMLST type (CT) (with up to seven allelic differences). An alternative cgMLST scheme based on 1,701 core loci with a CT threshold of up to ten allelic differences is used for surveillance in Germany and Austria.
Lineage I (primarily serotypes 1/2b and 4b) and lineage II (primarily serotypes 1/2a and 1/2c) account for most isolates, but there is an uneven distribution at the level of CCs and sublineages. Most human listeriosis cases and outbreaks worldwide are associated with lineage I isolates. A study based on 6,633 strains prospectively collected in France, where listeriosis is a notifiable disease, showed that lineage I, in particular CC1, CC4 and CC6, was strongly associated with a clinical origin (isolated in human samples), whereas lineage II, in particular CC9 and CC121, was the most common lineage isolated from food (Fig. 2). In the USA, CC1, CC4 and CC6 are also the most common in clinical samples, whereas the lineage I CC87 is the most frequent clonal complex in clinical samples from China and Taiwan. Analyses of clinical and biological data collected at the National Reference Centre in France from 818 patients with listeriosis enrolled in the MONALISA prospective cohort study on Listeria and listeriosis further showed that food-associated clones CC9 and CC121 were more frequently isolated from highly immunocompromised patients, whereas CC1, CC4 and CC6 were more common in patients with few or no immunosuppressive comorbidities. These findings led to the hypothesis that CC1, CC4 and CC6 might be hypervirulent, and that CC9 and CC121 might be hypovirulent, which was confirmed in a humanized mouse model of L. monocytogenes infection. Lineage III and IV strains are rare and tend to be associated with listeriosis in animals.