What Is the Incubation Period for Listeriosis?

Véronique Goulet; Lisa A King; Véronique Vaillant; Henriette de Valk


BMC Infect Dis. 2013;13(11) 

In This Article

Results and Discussion

Among the 23 reports, we identified 15 reports with precise documented incubation periods for invasive listeriosis (Table 1). In total, a precise incubation period was documented for 37 invasive cases (10 CNS cases, 15 bacteraemia cases and 12 pregnancy-associated cases). For invasive listeriosis, the overall median incubation period was 8 days (range: 1–67 days) (Figure 1).

Figure 1.

Distribution of the incubation period (in days) for 37 invasive cases of listeriosis.

The incubation period differed significantly by clinical form of invasive listeriosis (Kruskall-wallis, p<0.0001) (Figure 2). A longer incubation period was observed for pregnancy-associated cases (median: 27.5 days; range: 17–67 days; 1st and 3rd quartiles: 20, 36 days) than for CNS cases (median: 9 days; range: 1–14 days; 1st and 3rd quartiles: 4,13 days) and for bacteraemia cases (median: 2 days; range: 1–12 days; 1st and 3rd quartiles: 1,5 days) (Figure 3).

Figure 2.

Distribution of the incubation period (in days) of 37 invasive cases of listeriosis by clinical form of disease.

Figure 3.

Distribution of incubation period for each clinical form of 37 invasive cases of listeriosis (box-plot). Line in the middle of boxes represents median of data. Boxes extend from the 25th percentile (X[25]) to 75th percentile (X[75]), representing interquartile range (IQR). Lines emerging from boxes extend to upper and lower adjacent values. The upper adjacent value is defined as the largest data point ≤X[75] + 1.5 x IQR. The lower adjacent value is defined as the smallest data point ≥X[25] – 1.5 x IQR. Dots are outliers (every point more than 1.5 x IQR from the end of a box).

Estimates of an approximate incubation period were available for 14 cases (2 CNS cases, 6 bacteraemia cases, 6 pregnancy-associated cases) (Table 2).

For gastroenteritis, 9 outbreaks were reviewed. For each outbreak, the median and range of incubation periods are shown in Table 3. For gastroenteritis cases, the median incubation period was 24 hours. Incubation periods varied from 6 to 240 hours.

The results of this study clearly demonstrate that the listeriosis incubation period is shorter than generally assumed and varies according to the clinical form of the disease. Not surprisingly the shortest incubation period is observed for listeria associated gastroenteritis (one day) with an incubation period quite similar in duration to other enteric bacteria such as Salmonella. The incubation period is also short for bacteraemia cases, with a median of 2 days and is longer for CNS cases, with a median of 9 days (p<0.05). The range of incubation periods for CNS cases was, however, wide (1 to 14 days) compared with bacteraemia cases (1–7 days) if we exclude one outlier bacteraemia case with a much longer incubation of 12 days (Figure 2). With a median of 27.5 days and a range of 17–67 days, pregnancy-associated forms have a much longer incubation period than other clinical forms. A likely explanation is that there is a delay between bacteraemia and infection of the foetus due to the time necessary for Lm to colonize the placenta and to induce a placentitis that is at the origin of the fetal infection. An experimental study on pregnant guinea pig supports the hypothesis that the placenta is relatively protected from infection.[35] Once colonized, the placenta acts as a nidus of infection for the mother resulting in massive reseeding of maternal organs, where Lm cannot be cleared until trafficking is interrupted by expulsion of the infected placental tissue. This hypothesis is consistent with the delay observed between ingestion of a contaminated food and foetal infection.

Situations in which an incubation period can be precisely documented are exceptional for listeriosis. Evidence of the link between a contaminated food and illness can be obtained in outbreak situations, but such evidence is rarely available for sporadic cases. The situation best adapted to documentation of a precise incubation period is an outbreak linked to a single meal, since the moment of consumption of the incriminated product coincides with the moment of contamination. When the outbreak is due to a product consumed over a longer period of time or on a regular basis, it is extremely difficult to identify the exact date of contamination. To be accurate, only single exposures should be used for the calculation of the incubation period. Our study of all documented incubation periods with a point source exposure during a 32 year period was able to identify 37 precise incubation periods. It is the most complete and comprehensive series of such cases analyzed to address this question.

Approximate incubation periods that we calculated without precise point-source contamination are consistent with the precise estimates (Table 2). One outbreak reported in this table suggests that the incubation period of invasive listeriosis may be longer than our estimates.[25] All the eight patients of this outbreak had a comorbidity impairing their immunity and received ready-made food delivered at home by the same catering company. As the only meal in common was prepared on April 14, 2009, the authors suggested that this meal was at the origin of the outbreak. If this was the case, the range of incubation periods for the 7 bacteraemia cases would be 21–27 days and 21 days for the only CNS case. We were reluctant to take this outbreak into account for the calculation of precise incubation periods since most of these patients had daily delivery of meals from this catering company. As Lm can easily colonize kitchen surfaces, different meals prepared on subsequent days may also have been contaminated by cross contamination or by a contaminated ingredient used in the preparation of several meals.

The wide range of incubation periods observed for each form may be related to varying levels of contamination of food, the quantity of contaminated food consumed, the virulence of the Lm strain or the immunological status of the patient. In our study, the majority of bacteraemia cases with documented incubation periods were observed during hospital outbreaks among persons with comorbidities that impair immunity. When Lm was identified by blood culture, they all received an antibiotic treatment that effectively reduced colonisation of the CNS system. Interestingly in an outbreak that occurred after a single meal in Austria, one 72 year old attendee of the meal fell ill with fever and diarrhoea, recovering within two days, but then developed CNS symptoms on day 14.[19] Diagnosis of listeriosis with CNS involvement was confirmed by Lm isolation in CSF. In this outbreak, another attendee hospitalized two days after the meal for fever and diarrhea, had blood cultures positive for Lm. He was subsequently treated and did not develop CNS symptoms. These observations suggest that CNS involvement occurs after transient bacteraemia and thus has a longer incubation period. In France, 75% of the bacteraemia forms of listeriosis are diagnosed in patients with comorbidity. Febrile patients with comorbidity are more likely to have blood drawn for culture and to be diagnosed in the case of bacteraemia. In contrast, blood cultures are uncommon for febrile persons without comorbidity. If these unrecognized bacteraemia are not treated, Lm can subsequently infect the CNS. The French surveillance data show that 69% of listeriosis cases in patients without comorbidity have CNS involvement.

The exposure windows considered when interviewing patients about their food consumption is a delicate issue. By taking a wide exposure window, one gains in terms of sensitivity by including a variety of foods consumed that are more or less food habits of the consumer. By taking a smaller exposure window, one reduces recall biases and gains in specificity by limiting the number of foods consumed. In a case-control study of sporadic Salmonella Enteritidis infections, Molbak compared food exposure data obtained for an exposure window corresponding with the maximum incubation period (7 days) to food exposure data for an exposure window corresponding to the most relevant incubation period (1 day).[36] The conclusion was that for common food exposures, exposure classification that corresponds to the most common period of incubation rather than the maximum period is more accurate. Our study suggests that, to be efficient, food interviews for listeriosis outbreak investigations should use different exposure windows according to the clinical form of the disease. We suggest that Listeria gastroenteritis cases should be interviewed about their exposures during the two days before their first symptoms and pregnancy-associated cases, 6 weeks prior to their first symptoms. For other forms, the most appropriate option would be to interview bacteraemia cases about the 7 days prior to symptom onset and CNS cases during 14 days prior to their first symptoms. However, in the context of routine surveillance, it is sometimes difficult to discriminate bacteraemia patients with mental confusion from cases with CNS involvement. Therefore it could be more pragmatic to use a unique exposure window of 14 days when interviewing these patients for surveillance purpose.

Another important outcome of a more precisely documented listeriosis incubation period relates to prevention messages issued by health authorities when a contaminated product is withdrawn from the market. In France, persons who consumed Lm contaminated products are advised to watch carefully for any symptoms of listeriosis during a period of 2 months after exposure. Based on our results, we recommend that these messages should be adapted to recommend 6 weeks of vigilance for pregnant women and 2 weeks for other exposed individuals.