Corresponding Author Lu Ran, Division of Infectious Disease Prevention and Control, Chinese Center for Disease Control and Prevention. No.155, Changbai Road, Changping, Beijing. 102206, China. Tel.: +86 10 58900508; Fax +86 10 58900567; E-mail: firstname.lastname@example.org
Listeria is an important foodborne pathogen with severe manifestations and high case-fatality rate. However, listeriosis is not yet a notifiable disease in China, and there is no national monitoring system for cases. We conducted a systematic review to better understand the clinical and epidemiologic features of listeriosis in China.
Both electronic and manual retrieval systems were used to search Chinese literature for cases and isolates of human listeriosis reported between 1964 and 2010. We recorded and analysed demographic, clinical and laboratory information available for reported cases.
A total of 147 clinical cases, 479 Listeria isolates and 82 outbreak-related cases were reported in 28 (90%) provinces in China from January 1964 to December 2010. Of the clinical cases, 45 (31%) were central nervous system infections, 68 (46%) were septicaemia and 34 (23%) were focal infections or gastroenteritis. The overall case-fatality rate was 26% (34/130) among clinical cases with known outcomes and 46% (21/46) among neonatal cases.
Listeriosis cases occurred in China throughout the study period between 1964 and 2010. Case-fatality was similar to published data from other countries. China should consider requiring notification of listeriosis cases to improve estimates of incidence, identification of risk factors and design of preventive measures.
Listeria est un pathogène important d'origine alimentaire avec des manifestations graves et un taux de létalité élevé. Cependant, la listériose n'est pas toujours pas une maladie à déclaration obligatoire en Chine et il n'existe pas de système national de surveillance des cas. Nous avons effectué une revue systématique pour mieux comprendre les caractéristiques cliniques et épidémiologiques de la listériose en Chine.
Les systèmes de recherche électroniques et manuels ont été tous deux utilisés pour rechercher la littérature chinoise pour les cas et isolats de listériose humaine signalés entre 1964 et 2010. Nous avons enregistré et analysé les informations démographiques, cliniques et de laboratoires disponibles pour les cas rapportés.
Au total 147 cas cliniques, 479 isolats de Listeria et 82 cas liés à une épidémie ont été signalés dans 28 (90%) provinces en Chine entre janvier 1964 et décembre 2010. Parmi les cas cliniques, 45 (31%) étaient des infections du système nerveux central, 68 (46%) étaient des septicémies et 34 (23%) étaient des infections focales ou des gastro-entérites. Le taux global de létalité était de 26% (34/130) parmi les cas cliniques avec des résultats connus et 46% (21/46) parmi les cas néonataux.
Les cas de listériose sont survenus en Chine tout au long de la période d’étude entre 1964 et 2010. La létalité était similaire à celle des données publiées dans d'autres pays. La Chine devrait envisager d'exiger la notification des cas de listériose afin d'améliorer les estimations de l'incidence, l'identification des facteurs de risque et la conception de mesures préventives.
La Listeria es un importante patógeno transmitido por alimentos con manifestaciones graves y una alta tasa de mortalidad. Sin embargo, la listeriosis aún no es una enfermedad de notificación obligatoria en China, y no hay un sistema nacional de monitorización para los casos. Hemos realizado una revisión sistemática para entender mejor las características clínicas y epidemiológicas de la listeriosis in China.
Se utilizaron sistemas de recuperación tanto electrónicos como manuales para buscar literatura sobre casos y cepas aisladas en China de listeriosis humana, reportados entre 1964 y 2010. Recogimos y analizamos la información disponible sobre datos demográficos, clínicos y de laboratorio de casos reportados.
Un total de 147 casos clínicos, 479 cepas de Listeria y 82 casos relacionados con un brote epidémico fueron reportados en 28 (90%) provincias de China entre Enero de 1964 y Diciembre 2010. De los casos clínicos 45 (31%) eran infecciones del sistema nervioso central, 68 (46%) eran septicemia y 34 (23%) eran infecciones localizadas o gastroenteritis. La tasa de letalidad era del 26% (34/130) entre los casos clínicos con resultado conocido y del 46% (21/46) entre los casos de neonatos.
A lo largo de todo el periodo de estudio, entre 1964 y 2010, se dieron casos de listeriosis en China. La tasa de letalidad era similar a los datos publicados para otros países. China debería considerar implantar la notificación obligatoria para los casos de listeriosis con el fin de mejorar los cálculos de incidencia, identificación de factores de riesgo y el diseño de medidas de prevención.
Listeriosis is a rare but serious foodborne disease caused by the bacterium Listeria monocytogenes (LM), affecting mostly pregnant women, neonates and immune-compromised hosts (Low & Donachie 1997; Posfay-Barbe & Wald 2009; Rebagliati et al. 2009). Because it can be associated with severe diseases — such as septicaemia, central nervous system (CNS) infections, severe focal infections/gastroenteritis and maternal–foetal infections including abortion, stillbirth, preterm birth and neonatal listeriosis — and a high case-fatality rate (20–50%), listeriosis is considered an important threat to human health (Doganay 2003; Swaminathan & Gerner-Smidt 2007). Moreover, because Listeria is mainly transmitted to humans through food, human listeriosis is often preventable. The first documentation of listeriosis in China was a case of neonatal Listeria meningitis, reported by Ding et al. (1964). Since then, Chinese medical journals have published hundreds of reports of listeriosis. Although the number of published articles about listeriosis has increased, China does not require notification of listeriosis cases to public health authorities (Feng et al. 2010). To better understand clinical and epidemiologic features of Listeria infections in China, we systematically reviewed the Chinese medical literature to identify reports of listeriosis cases or isolates from 1964 to 2010 in China.
Materials and methods
Literature retrieval and inclusion
The genus Listeria comprises ten species: LM, L. innocua, L. welshimeri, L. seeligeri, L. grayi, L. rocourtiae, L. marthii, L. fleischmannii, L. weihenstephaniensis and L. ivanovii, the latter comprising L. ivanovii subsp. Ivanovii and L. ivanovii subsp. londoniensis (Leclercq et al. 2010; Bertsch et al. 2012). Although species of the genus Listeria are environmental bacteria, two species (LM and L. ivanovii) are pathogenic for animals and humans. Almost all cases of human listeriosis are due to LM (Hof & Hefner 1988; Schmid et al. 2005). As a result of its ubiquitous nature, LM can occur in most non-sterile sites of humans (Rocourt 1996). Invasive listeriosis, therefore, is generally defined as isolation of LM from a normally sterile site (US CDC 1999; Denny & McLauchlin 2008). In China, there is no diagnosis guideline for listeriosis. As Listeria isolates were reported only to the level of genus in a large proportion of reports, we defined any patient with isolation of the genus Listeria from any body site, either sterile or non-sterile, as a listeriosis case.
We searched for and then included articles published in Chinese medical journals from January 1964 to December 2010. A flow chart of our search is shown in Figure 1. Different search strategies were employed for two different periods: from January 1964 to December 1989 and from January 1990 to December 2010.
For January 1964 to December 1989, articles were not documented in most electronic databases. Therefore, we manually reviewed the Chinese Science and Technology Information Directories (Medicine and Health), the leading retrieval tool for medical information run by the National Institute of Medical Information (Wu & Zhang 2000). We identified 27 articles with titles relevant to Listeria in humans. After full-text review, we identified 9 with sufficient data to warrant abstraction.
For January 1990 to December 2010, we initially retrieved the full text of 2920 articles reporting Listeria by directly searching three primary academic electronic databases in China: China National Knowledge Infrastructure (CNKI), Wanfang (Wanfang Data full text database) and VIP (VIP Database for Chinese Technical Periodicals) (Du et al. 2006), using all terms of Listeria in Chinese (李斯特菌、利斯特菌、李斯忒菌) as key words. We subsequently identified and deleted 1407 duplicate articles among the three academic electronic databases according to title, publication date and author. We also excluded articles not relevant to humans or without description of listeriosis cases or isolates in China. Finally, we included 172 articles for further case study or isolate study from January 1990 to December 2010. Therefore, combined with the articles identified in the earlier period, we included a total of 181 articles for analysis.
Extraction criteria for listeriosis cases or isolates from the literature
Based on the contents, the 181 articles, as shown in Figure 1, were categorised into three groups: 80 clinical case reports, 100 clinical laboratory sample reports involving at least the number and type of Listeria isolates and one outbreak report.
For the 80 clinical case reports, we collected data about patient demographics (e.g. sex, age), hospital (e.g. name, address, size), clinical course (e.g. underlying disease, onset time, symptom, therapy and outcome) and laboratory information (e.g. sample source that yielded Listeria). For the 100 laboratory sample reports, we collected date and location of collection, number of isolates, type of illness and sample, specific species, isolation rate, characteristic of isolates (e.g. serotype, antimicrobial resistance). For the outbreak report, we collected information on the outbreak process, the investigation of suspected sources and characteristics of related cases (e.g. onset time, symptom and outcome).
We used the extracted criteria above to identify and delete duplicate cases in clinical case reports and duplicate isolates in laboratory sample reports, respectively. In the end, as shown in Figure 1, a total of 147 listeriosis cases, 479 Listeria isolates, and 82 outbreak-related cases were included in the analysis.
Definitions used in this study
We defined a case of listeriosis as growth of Listeria from any body site of a person with clinical illness. We categorised the patient's clinical syndrome into three groups: CNS infection, where either Listeria was isolated from cerebrospinal fluid (CSF) or a clinical diagnosis of CNS infection with Listeria isolated from blood; septicaemia, where Listeria was isolated from blood but CNS involvement was not reported; focal infection/gastroenteritis, where Listeria was isolated from at least one body site (or stool), which was neither CSF nor blood.
According to clinical characteristics of listeriosis, patients from the clinical case reports were categorised into two broad population groups: perinatal patients, including pregnant women or neonates (<28 days old) and non-perinatal patients, including all other patients. Non-perinatal patients were further categorised into two subgroups because of the different risks of infection: immune-compromised patients and immune-competent patients. For perinatal cases, a mother and her live birth were counted as separate cases. A perinatal case was only classified as a maternal one when a mother had an abortion or stillbirth due to Listeria infection.
An immune-compromised patient in this study referred to a patient who was more susceptible to listeriosis as a result of a special clinical status, including suffering from haematological malignancy, solid malignancy, autoimmune disease, HIV infection, chronic renal failure, chronic liver disease, diabetes mellitus, transplantation and alcoholism or use of immune-suppressive therapies, such as corticosteroids.
Statistical analysis was performed using SAS Version 10.0 (SAS Institute, Cary, NC, USA). Pooled t-test analysis was performed to compare the median ages between immune-compromised and immune-competent patients; chi-squared analysis was performed for other comparisons. A P-value <0.05 was defined as statistically significant.
Cases from clinical case reports
Reporting periods of all clinical listeriosis cases for each population subgroup studied were shown in Table 1. More than half (69%) of the cases were reported during 2001–2010. Only 13 of 147 cases were reported during 1964–1990. The 147 patients reported by clinical case reports comprised 70 (48%) non-perinatal patients and 77 (52%) pregnancy-related cases. Characteristics of the patients are listed in Table 2. These cases had a median age of 26 years (ranging 0–74 years). The age distribution of each population subgroup is shown in Figure 2. The age of 22 pregnant women remained unknown, but they could most likely fall into the group of 26–35 years. The most common clinical manifestation was septicaemia (46%), followed by CNS infection (31%) and focal infection/gastroenteritis (23%). The overall case-fatality rate was 26% (34/130), highest among neonatal cases (46%) and lowest among pregnant cases (4%).
Table 1. Reporting periods of clinical listeriosis cases for each subgroup in China, 1964–2010
No. of cases
Table 2. Population-specific demographics, manifestations and case-fatality rates of listeriosis cases reported in China, 1964–2010
As shown in Table 2, among non-perinatal patients, immune-compromised patients were much older than the immune-competent (P < 0.05). Among immune-competent cases, there were more than twice as many males as females; among immune-compromised cases, the ratio was reversed (0.3:1 male: female) (P < 0.05). Immune-compromised patients most frequently had CNS infection (62%), while immune-competent patients most frequently had septicaemia (41%). A total of 12 patients died; two deaths were due to septicaemia and the remainder to CNS infection. The case-fatality rate among immune-compromised patients (26%) was higher, but not significantly so, than that of immune-competent patients (17%). Only two surviving patients, both of whom were immune-competent, reported long-term sequelae: one patient with CNS infection had exotropia in combination with blunted responsiveness; the other with septicaemia had hypoevolutism in motor development.
Among the 30 pregnant women, the median gestational age was 28 weeks (range 16–42 weeks), and there were no immune-compromising conditions reported. Of these, 23 (77%) became ill during the third trimester and 7 (23%) in the second trimester. Only 4 (13%) pregnancies, all with Listeria isolated from cervical secretions, had no symptoms during the infection, while the remaining 26 pregnancies reported multiple clinical symptoms. The most common symptom was fever (24 cases, 80%), followed by abdominal pain (nine cases, 30%) and gastrointestinal symptoms such as nausea, vomiting and diarrhoea (four cases, 13%). Only two pregnant women, both of whom had Listeria only isolated from cervical secretions, had comorbidities: one suffered from gestational diabetes; the other developed gestational hypertension. Both pregnant women recovered from the infections satisfactorily, and delivered healthy infants without Listeria infections. One (4%) pregnant woman died from listeriosis. She presented with fever and abdominal pain at 32 weeks gestation and had Listeria isolated from blood, CSF and bone marrow; her two foetuses were delivered stillborn.
Clinical and laboratory information about offspring was available for 28 pregnant patients. One of the 28 patients was pregnant with twins. Types of mother–infant infection and positive Listeria specimens from these 28 pregnancies and their offspring are shown in Table 3. Among the 16 infected newborns, 15 (94%) had either CNS infection or septicaemia, and the case-fatality rate was 50% (8/16). Eight of the infected neonates died from the infection, and five foetuses died in utero. Thus, the case-fatality rate in the 29 offspring of 28 infected pregnancies was 45% (13/29).
Table 3. Types of mother-infant infection, positive samples of LM in 28 perinatal cases of listeriosis reported in China, 1964–2010
Except for the 16 neonatal cases mentioned in the pregnant cases above, there were 31 neonatal cases without maternal information. For the 47 total neonates, 68% were preterm (gestation age <37 weeks). Three neonates presented with infection on days 27, 15 and 10 of life, respectively; the remaining 44 neonates, however, all were infected within 1 week of life, of whom 72% presented with respiratory distress as the initial symptom, and 68% manifested as septicaemia. The case-fatality rate was higher, but not significantly, in preterm neonates (47%, 15/32) than in term ones (43%, 6/14). One surviving neonate, who suffered Listeria-associated pneumonia, was found to have sub-normal intelligence and hypoevolutism in motor development because of listeriosis.
Details about antibiotic therapy were unavailable for most (81%) patients. However, almost all patients with known information on antibiotic therapy were treated with penicillin or ampicillin, sometimes in combination with aminoglycosides, such as gentamicin, amikacin, and netilmicin.
Samples from clinical laboratory reports
A total of 479 Listeria isolates were obtained from clinical laboratory samples, including blood, CSF, marrow, umbilical swab, nasopharyngeal swab and secretion from inflamed sites, such as perforated tympanic membrane, cervical tract and diabetic feet.
Five reports provided isolation rates of Listeria for patients suffering from different clinical conditions (Table 4) (Chen 1998, 2004; Huang et al. 2000; Li et al. 2002; Rao et al. 2007). The isolation rates varied from 0.36% to 1.98%; the highest isolation rate was in neonates with pneumonia. Only one report studied serotypes of LM. This study reported that 79% (46/58) of LM isolates were 4b, 14% (8/58) 1/2a, and 5% (3/58) 1/2c (Zhou et al. 2009).
Table 4. Positivity of Listeria in patients with different diseases reported in China, 1964–2010
The only listeriosis outbreak report obtained in the study described an outbreak among students aged 8–12 years in an elementary school of Taizhou, Zhejiang Province in October 2003. Epidemiological investigation indicated that 82 of 150 students who ate a meal in school got sick and were diagnosed with listeriosis. No deaths occurred. The incubation period for this outbreak was 8–10 h, and the symptoms were primarily chills, headache, dizziness, nausea and vomiting. Four cases had neurological manifestations, including confusion, delirium, meningeal irritability and coma. A vacuum-sealed product of cooked, unshelled eggs served to children in school was suspected to be the source. Laboratory testing isolated LM from students' vomit and leftover eggs.
Seasonal variation of sporadic cases
Overall, the month of infection was recorded in 51 sporadic cases from clinical case reports. Infection occurred throughout the year, but most often during spring and fall (Figure 3).
Geographic distribution of sporadic cases and positive samples
Listeriosis cases included in this study were reported from 28 of 31 mainland provinces of China, except for Heilongjiang, Tibet, and Hainan. The top five provinces with the largest total number of cases were Sichuan (n = 117), Jiangsu (n = 88), Guangdong (n = 73), Beijing (n = 70) and Shanghai (n = 56); the total number of cases was regarded as the number of cases from clinical case reports plus the number of positive samples from laboratory summary reports where the number of positive samples was considered equal to the number of listeriosis cases.
The 80 clinical case reports and 100 laboratory sample reports were from a total of 137 hospitals throughout China, including 92 (67%) tertiary hospitals with >500 beds and 24 (18%) secondary hospitals with 100–500 beds.
This is the first systematic review of human Listeria infections in China. We found that clinical manifestations, clinical outcomes and serotypes were similar to those reported in other countries (Siegman-Igra et al. 2002; Gerner-Smidt et al. 2005; Denny & McLauchlin 2008). Our study design, however, did not permit us to measure incidence or overall disease burden.
In China, clinical manifestations of non-perinatal and perinatal listeriosis are similar to those reported elsewhere (Swaminathan & Gerner-Smidt 2007; Allerberger & Wagner 2010; Lamont et al. 2011). We found that, among non-perinatal cases, patients with serious immune-compromising conditions may suffer more serious forms of listeriosis (CNS infections) than immune-competent patients (McLauchlin 1990). It is widely known that, although listeriosis during pregnancy is life-threatening to foetuses, severe clinical manifestations and fatal outcomes are extremely rare among pregnant women themselves (Lamont et al. 2011). However, we found one fatal case of maternal listeriosis with Listeria isolated from blood, CSF and bone marrow. It is uncertain whether comorbidities could worsen the clinical outcomes of pregnant listeriosis cases, but among the cases studied here, neither the two women's treatment nor their infants' health were negatively affected by the pregnancy comorbidities.
Our study showed a case-fatality rate of 26% for listeriosis, in accordance with published data of 20–30% elsewhere (Swaminathan & Gerner-Smidt 2007; Allerberger & Wagner 2010). Similar to other studies (Schlech 2000; Rocourt et al. 2000), our data showed that higher case-fatality might be associated with presence of CNS infection, immune-compromising conditions and premature birth. However, the number of cases in this study was too small to make a firm conclusion. Previous studies (Siegman-Igra et al. 2002; Gerner-Smidt et al. 2005) found that there was no correlation between death and age for the whole population. We, too, note that all patients aged >60 years survived Listeria infection.
Outside of China, contaminated foods implicated in listeriosis outbreaks included contaminated milk, meat, fish and vegetable products (Rocourt et al. 2000; Lundén et al. 2004; Swaminathan & Gerner-Smidt 2007). In our study, the one outbreak involving 82 cases was caused by a contaminated commercial egg product. However, it is unlikely that the eggs themselves were originally contaminated; they were probably contaminated somewhere during preparation and packaging (Carpentier & Cerf 2011). The vehicle of sporadic infections has rarely been established because Listeria is ubiquitous in the environment and has a long incubation period. Two case–control studies (Schuchat et al. 1992; Varma et al. 2007) in the US reported that sporadic listeriosis cases were more likely to have eaten melons, hummus prepared in a commercial establishment and soft cheeses or food purchased from delicatessen counters, indicating that commercial foods were at higher risk than home-made foods. In our study, only five patients were suspected of having consumed contaminated food: four pregnant women likely were infected through rare roast meat; a dairy product; Chinese hot pot and seafood; and preserved vegetables and salted fish, respectively. The remaining case was an immune-competent man, and the suspected source of infection was seafood. However, as there was no available information about the time of food exposure for all five cases, we could not determine how long their incubation times were.
Listeriosis cases were mainly reported in eastern costal and south-western areas of China. More studies are needed to determine whether this was due to reporting bias or represents the true geographic distribution of listeriosis in China. Interestingly, most cases with known onset time of infection occurred in spring and fall. As the average incubation period for listeriosis is 30 days (range 1–91 days), primary exposure to Listeria leading to infection in a high proportion of cases may occur in mid-summer and in mid-fall. A similar seasonal variation has been reported in other studies (McLauchlin 1990), but the reason is unknown. In recent years, an unexplained increase in incidence of listeriosis among people older than 60 years has been reported in several European countries (Goulet et al. 2008; Allerberger & Wagner 2010; Mook et al. 2011), but this is difficult to detect in our study because of the paucity and dispersion of cases among the elderly.
Our report is subject to important limitations. First, our study is biased by reporting issues. Not all diagnosed cases of listeriosis are published, and cases with more interesting findings (e.g., association with pregnancy; fatal outcome; drug resistance; unusual serotypes) are more likely to be reported. Therefore, our results might not be generalisable to the whole population of China. Second, data on each case or positive sample of listeriosis studied here were far from complete; thus, only the information closely associated with the main purpose of the reports was available. Third, isolates from laboratory sample reports might have been counted repeatedly, as the methods used in this study were unable to fully distinguish whether Listeria isolates were reported back and forth among different laboratory sample reports or between clinical case reports and laboratory sample reports, especially when they were shared by a same hospital or a same author. Finally, there might be some bias because of the different searching methods used for 1964–1989 and 1990–2010. However, these were the only available searching tools for the study periods.
In conclusion, listeriosis cases were reported in many regions of China over the past 47 years, with a high case-fatality rate, particularly in neonates. Listeria has been isolated from various kinds of Chinese food recently (Chen et al. 2009), and, as China develops rapidly, it is likely that the Chinese will increase their purchase of commercially prepared food, potentially increasing their risk of exposure to Listeria. China has recently made strong efforts to improve food safety, including enactment of strict laws related to inspection of food. We believe that China should consider adding listeriosis to the list of notifiable diseases to improve understanding of this condition in China, including measuring incidence, detecting outbreaks and facilitating epidemiologic investigations into and recall of potentially contaminated foods.
This study was supported by WHO Global Foodborne Infections Network program and the China – US Collaborative Program on Emerging and Re-emerging Infectious Diseases.