Epidemiological insights from a large-scale investigation of intestinal helminths in Medieval Europe

Abstract

Helminth infections are among the World Health Organization's top neglected diseases with significant impact in many Less Economically Developed Countries. Despite no longer being endemic in Europe, the widespread presence of helminth eggs in archaeological deposits indicates that helminths represented a considerable burden in past European populations. Prevalence of infection is a key epidemiological feature that would influence the elimination of endemic intestinal helminths, for example, low prevalence rates may have made it easier to eliminate these infections in Europe without the use of modern anthelminthic drugs. To determine historical prevalence rates we analysed 589 grave samples from 7 European sites dated between 680 and 1700 CE, identifying two soil transmitted nematodes (Ascaris spp. and Trichuris trichiura) at all locations, and two food derived cestodes (Diphyllobothrium latum and Taenia spp.) at 4 sites. The rates of nematode infection in the medieval populations (1.5 to 25.6% for T. trichiura; 9.3-42.9% for Ascaris spp.) were comparable to those reported within modern endemically infected populations. There was some evidence of higher levels of nematode infection in younger individuals but not at all sites. The genetic diversity of T. trichiura ITS-1 in single graves was variable but much lower than with communal medieval latrine deposits. The prevalence of food derived cestodes was much lower (1.0-9.9%) than the prevalence of nematodes. Interestingly, sites that contained Taenia spp. eggs also contained D. latum which may reflect local culinary practices. These data demonstrate the importance of helminth infections in Medieval Europe and provide a baseline for studies on the epidemiology of infection in historical and modern contexts. Since the prevalence of medieval STH infections mirror those in modern endemic countries the factors affecting STH decline in Europe may also inform modern intervention campaigns.

Fig 1. The sites, samples and parasites.

A map showing the location of sampled sites (A); Summary table providing background information on each site and overall percentages of each parasite (B). Representative photomicrographs of the eggs detected in this study (C: Ascaris, D: Trichuris, E: Taenia and F: Diphyllobothrium, scale bar: 20 μm). The map represented in A was modified from the NASA SEDAC centre https://sedac.ciesin.columbia.edu/maps/gallery/search?facets=region:europe&facets=theme:water.

Fig 2. Comparison of prevalence rates of Ascaris and Trichuris between modern and medieval populations.

Prevalence rates of Ascaris (A) and Trichuris (B) identified in archaeological sites and comparison with regional prevalence rates published in meta-analyses by Chan 1994 [8], de Silva 2003 [9] and Pullan 2014 [10]. Orange dots indicate prevalence rates in different endemic regions. Blue dots indicate prevalence rates in each of the archaeological sites within Europe. Significant differences between groups indicated by * p<0.05, ** p<0.005.

Fig 3. The prevalence of helminth infection in males and females.

The ratio of infected male to female individuals for each of the four parasites, Ascaris, Trichuris, Taenia and Diphyllobothrium. Each of the sites indicated by a different symbol. The actual numbers of males and females infected with each parasite are given in the Table below the graph as is the cumulative total of all sites and the total numbers of males and females identified at each site. The bars represent the 95% confidence interval of all sites and no significant differences were identified according to sex.

Fig 4. The prevalence of nematode infections in different age groups.

The prevalence of Ascaris (A) and Trichuris (B) in different age groups at five medieval sites. Infant 0–5 years, Child 6–18 years, Younger Adult 19–40 years, Older adult 40+years. Each site is depicted with different symbols and colours according to the key. Significant differences in the overall rates of infection are indicated by horizontal bars * p<0.05, ** p<0.005.

Fig 5. The influence of population size on parasite prevalence.

The population size of the different locations was estimated from historical records and plotted against the prevalence of Ascaris (A and C) or Trichuris (B and D). Panels A and C represent the whole data set. Panels B and D represent data from adult groups (>18 years or identified as adult). Each site is identified with a different symbol according to the insert. The numbers of samples considered with each site are given in the insert. The error bars represent the range of estimated population size for each group.

Fig 6. The prevalence of helminth infections in Ellwangen over time.

The prevalence rates for Ascaris (A), Trichuris (B), Taenia (C) and Diphyllobothrium (D) infection in Ellwangen were segregated according to 6 time periods from the 7^th-8^thc to the 17^th-18^thc as indicated. The number of samples in each time period is also identified. Bars represent the proportion of infected individuals and error bars represent 95% confidence intervals.

Fig 7. The diversity of Trichuris trichiura ITS1 fragment sequences within single grave samples and contemporary communal deposits.

A stacked bar chart (A) indicating the proportion of different sequences in each of 6 single grave samples (5 from Pohansko, Po and one from Ipswich, Ip) and 13 communal deposits (4 from Bristol, Br1-4; and 9 from Lübeck, Lu1-9) where greater than 100 sequences/sample were obtained. The colours represent sequences that were identified in multiple samples with solid colours representing the widespread group 1 sequences (identified in reference 7) and the striped bars group 2 sequences that were common in Lübeck, rare in Bristol and not found elsewhere (7). Segments with no colour represent sequences identified in a single sample. The species richness of samples (B and C) segregated by sample type (single grave, blue) and communal deposit (black). The diversity of all samples are represented in panel B with the mean species richness ± Standard Error of the mean depicted at different depths of sub-sampling between 50 and 250 sequences. Higher depths of subsampling are presented in panel C to reveal the plateau of the sampled diversity, the mean diversity (line) with 95% confidence intervals (shaded region) are depicted.