We collected ectoparasites from the clothing of participants in inner city Winnipeg. We separated ectoparasites from the same person into pools based on instar and sex. We pooled ectoparasites from the first and second instars but tested those from the third and fourth instars separately. We tested ectoparasites positive for B. quintana from the fourth instar in separate pools of male and female parasites. We decontaminated ectoparasite pools by using 70% ethanol and homogenized them by using a copper clad bead beater. We then extracted DNA by using the DNeasy 96 kit (QIAGEN, https://www.qiagen.com). We identified vector species, louse morphotype, and pathogens by using real-time PCR (Appendix). We used cytochrome b genes to identify louse species and Phum_PHUM540560 genes to identify ecotype (8). We identified pathogens by using the following targets: ITS3, Bartonella genus; yopP and fabB, B. quintana; ompB, Rickettsia prowazekii; IS1111a, Coxiella burnetii; and rpoB, Acinetobacter spp. We conducted statistical analysis by using Mann-Whitney U and Kruskal-Wallis tests (Bonferroni correction, post-hoc Dunn test) to compare groups of Ct values (Appendix). We considered values of p<0.05 significant.

Figure 1

Figure 1. Body lice collected from a person experiencing homelessness in inner city Winnipeg, Manitoba, Canada. Not all ectoparasites from this person were analyzed.

Figure 2

Figure 2. Two female body lice, Pediculus humanus humanus, collected from a person experiencing homelessness in inner city Winnipeg, Manitoba, Canada (9).

Seven persons submitted ectoparasites, 2 in 2020, and 5 in 2021 (Appendix). We analyzed 556 ectoparasites. The range of ectoparasites tested per participant was 5–218 and per pool was 5–48. We confirmed all ectoparasite pools were P. humanus humanus lice by using PCR positivity on louse and body lice targets and morphology (9) (Figures 1, 2). All louse pools from 1 participant (1/7 = 14%, 218 lice) demonstrated positivity on all Bartonella and B. quintana targets (Table 1). Of the 7 louse pools positive for B. quintana, 4 also demonstrated molecular positivity for Acinetobacter spp. Ectoparasites from all participants were negative for R. prowazekii and C. burnetii.

When analyzing B. quintana–positive louse pools, we found Ct values were similar between ITS3, yopP, and fabB genes (test statistic H = 0.54; p = 0.76). The average ITS3 Ct values decreased from the first and second instar pools (34.6) to the third instar pools (28.9) by 5.7, and from the third instar pools to the fourth instar pool (21.8) by 7.1. Pools from female lice demonstrated lower ITS3 Ct values than male lice pools (p = 0.0214) (Table 2).

We collected body lice from a person experiencing homelessness in Winnipeg that were positive for B. quintana bacteria by using molecular testing. This finding complements the recent Manitoba cluster of B. quintana cases, suggesting a poorly described burden of infection (1,2,4). The hidden presence of B. quintana bacteria in Canada was recently highlighted in an outbreak of transplant derived B. quintana infection in cities that had not previously reported transmission: 5 cases of bacillary angiomatosis were linked to 3 deceased donors from 2 cities in Alberta (Health Canada, pers. comm., email, 2023 Nov 4). All cases were confirmed to be B. quintana bacteria with donors experiencing homelessness as the common risk factor (Health Canada, pers. comm., email, 2023 Nov 4).

Our study suggests a minority of body lice cases from Winnipeg are positive for pathogens, including B. quintana bacteria. We did not collect epidemiologic data for this study, but all participants were persons who experienced homelessness in inner city Winnipeg. Because of Winnipeg’s harsh winters and few homeless shelters, it is possible the participant with B. quintana–positive lice lives in close proximity to others and other persons with B. quintana infection remain undocumented. Only 1/7 persons with body lice had B. quintana–positive lice, which may be because of the small number of participants and that 3 participants submitted few ectoparasites. Nationwide body lice studies to compare B. quintana bacterial prevalence across different areas are needed to identify locations of infection.

The absence of other pathogens likely reflects differences in transmission dynamics and ecology (10,11). Unlike B. quintana bacteria, which does not alter louse survival, lice infected with R. prowazekii bacteria die within a week of infection, limiting transmission (11). The urban setting of our study diminishes the chance of replicating the occasional documentation of C. burnetii bacteria in lice. Whereas Acinetobacter spp. bacteria are commonly identified in body lice, no proven cases of Acinetobacter disease caused by body lice have been confirmed (11,12).

The lower Bartonella Ct values (stronger signal) with advancing louse instar and female sex may indicate larger blood meals of those subpopulations. B. quintana bacteria replicate in the louse intestine but are not known to be transmitted transovarially, indicating the person with B. quintana–positive lice from all instars likely had sustained bacteremia for at least 1 month (body lice lifespan). This study highlights the usefulness of identifying ectoparasites by using molecular methods when arthropod taxonomic expertise is not accessible.

B. quintana bacteria is excreted in louse feces continuously for weeks in quantities up to 10⁷ bacteria/louse each day (13,14). The explosive replication, coupled with B. quintana bacteria remaining infectious in biofilm-like structures for up to 1 year, suggests even a single case of B. quintana infection may indicate a hidden burden of infected persons (5,14).

Our study is limited by a small sample size, the heterogenous number of ectoparasites submitted per person, the focus on urban populations from 1 jurisdiction, and the lack of DNA quantity normalization. Active case finding, contact tracing, and public health engagement are needed to elucidate the epidemiology of B. quintana infection in Canada. Manitoba residents with body lice should be evaluated for B. quintana infection. Sampling of ectoparasites may provide an effective way to perform surveillance for emerging pathogens in marginalized settings.