Article | Open

Bed bug aggregation on dirty laundry: a mechanism for passive dispersal

  • Scientific Reports 7, Article number: 11668 (2017)
  • doi:10.1038/s41598-017-11850-5
  • Download Citation
Published online:


Bed bugs have shown a recent and rapid global expansion that has been suggested to be caused by cheap air travel. How a small, flightless and anachoretic insect that hides within its host’s sleeping area manages to travel long distances is not yet clear. Bed bugs are attracted to the odour of sleeping humans and we suggest that soiled clothing may present a similarly attractive cue, allowing bed bugs to ‘hitch-hike’ around the world after aggregating in the laundry bags of travellers. We show that (1) soiled clothing is significantly more attractive than clean clothing to active bed bugs moving within a bedroom sized arena and (2) elevation of CO2 to a level that simulates human occupancy in the same arena appears to initiate search behaviour rather than direct it. Our results show, for the first time, how leaving worn clothing exposed in sleeping areas when travelling can be exploited by bed bugs to facilitate passive dispersal.


The common bed bug Cimex lectularius L. has recently undergone a global resurgence1 which has been partly attributed to the increase in low cost international travel2. How a small, flightless, anachoretic (living in crevices or holes3) insect that prefers to hide in the sleeping area of its host disperses over long distances is unclear. We know bed bugs are able to actively disperse on a local scale4 because, for example, infestations in the same multi-occupancy building usually consist of closely related populations established from a single founding event5, 6. One possible mechanism facilitating long-range dispersal is that the insects find their way into clothing and/or luggage – the ‘vehicle’ – that then allows them to accompany the host to a new refuge2. The ability of, and preference for, bed bugs to use established hiding places is well documented1, 7 and is an important phenomenon in the problems associated with their control. However, it appears to be at odds with an ability to sequester and hide themselves in ‘novel’ hiding places such as suitcases and clothes in order to facilitate passive dispersal.

In contrast to their dispersal mechanisms, the host-seeking behaviour of the bed bug (and other haematophagous insects8) has received considerable empirical attention9 and is known to involve a combination of cues including host thermal, olfactory and visual signals10,11,12 with CO2 playing an important modulatory role12. In mosquitos, elevated CO2 stimulates foraging behaviour13 by activating host-seeking and directing flight towards the CO2 source14. However, when human odour is also present, the mosquito ignores the CO2 plume and navigates towards the odour source15, suggesting CO2 functions as a cue that activates host searching rather than directing it. Similar effects may operate in bed bugs11. Human odour is thought to attract bed bugs since it elicits both electrophysiological16 and behavioural responses11. Potential ‘vehicles’ for passive dispersal, such as luggage, are likely to contain recently worn clothes (i.e. those soiled with sweat and volatiles) that release human odour, especially since travellers tend to take home their dirty laundry. Odours from soiled clothing (or luggage containing soiled clothing) may therefore influence host-searching behaviour in bed bugs and consequently facilitate the passive dispersal of bed bugs via long-distance transport networks. We used a bedroom-scale experimental arena to determine whether (a) bed bugs would leave their refugia, (b) bed bugs would aggregate on soiled clothing, and (c) whether elevated CO2 (to simulate a human host in the room) modulated bed bug behaviours.


Bed bugs were most likely to be on/in bags containing soiled clothes than on/in bags containing clean clothes (Table 1a, Fig. 1). Elevated CO2 had no effect on this result (Fig. 1).

Table 1: Generalised Mixed Effect Model results summary for (a) bed bug distribution in relation to their location in the arenas and the presence of a CO2 source. Of the bed bugs found on clothes, (b) bed bug distribution in relation to their gender, human odour treatment and the presence of a CO2 source. Interaction and main effects were evaluated using log-likelihood ratio tests. Entries in bold are statistically significant, only significant interactions shown, n = 6.
Figure 1
Figure 1

Number (mean ± S.E.) of bed bugs found in or on bags containing clothes with or without human odour, n = 6.

However, there was a significant interaction between CO2 treatment and the distribution of bed bugs within an arena (Fig. 2). In the presence of elevated CO2 more bed bugs left the refuge (Table 1b, Fig. 2) with only 1.6 ± 1.6% of the bed bugs remaining in the refuge. By contrast, significantly more (25 ± 9.9%) bugs were in the refuge at the end of the ambient CO2 trial (Table 1b, Fig. 2). The extra dispersing individuals under elevated CO2 were not found aggregating on clothing (Fig. 2), but instead, were found on the floor within the arena such as adjacent to the arena perimeter or, on three occasions, the open floor area. There was no difference in the number of males and females that dispersed within the room (Table 1a). The position of bed bugs in the room was not influenced by the location of the CO2 source (χ2(11) = 10.01, p = 0.53).

Figure 2
Figure 2

Distribution (mean ± S.E.) of bed bugs within the experimental arena without (white) and with (grey bars) a CO2 source, n = 6. Original refuge refers to corrugated filter paper on which the bed bugs were introduced into the arena. Floor is any location within the arena that is not clothing or original refuge.


In the absence of a human host, bed bugs were twice as likely to aggregate on bags containing soiled clothes compared to bags containing clean clothes. Contrary to our predictions, elevated CO2 did not affect this result. We did not observe bed bugs aggregating on the side of the room with the CO2 source, but instead found an increased likelihood of bed bugs leaving their refuge. Our data are most consistent with the notion that elevated CO2 initiates host-seeking behaviour. This is similar to findings in Aedes aegypti and Anopheles gambiae where very small increases in CO2 concentration stimulate host-seeking behaviour19, 20. The position of the CO2 source in a closed room above and away from the bed bug release point means directional cues as to the source of CO2 may have been difficult for bed bugs to discern and so it is not possible to conclude whether or not the lack of aggregation on that side of the room suggests a lack of attraction to CO2. Many commercial bed bug monitors use CO2 in conjunction with thermal and/or chemical cues, and claim higher catch rates than monitors without CO2. Our results suggest that any increase in capture rate in such devices10, 21 might be caused by increased activity rather than, or in addition to, increased attractiveness.

The smallest bottle-neck through which a new bed bug population can pass, and therefore the most likely passive (i.e. long-range) dispersal phenotype, is a single mated adult female5. There is some experimental evidence for a female ‘disperser’: work using small arenas and shorter time scales found that females were more active than males22 and more likely to disperse from aggregations with a 1:1 sex ratio23. By contrast, our study showed there was no significant sex difference in the number of insects found on bags with, or without, human smell (Table 1a). Since dispersal from the refuge may be initiated by hunger, we would expect females (with their higher metabolic rate24) to start host-seeking before males and potentially to do so more often. However, over longer periods, such as that used in this study, males and females are both likely to experience hunger and disperse from the refuge. Our design does not enable us to resolve this issue.

Our results do show that bags containing soiled clothes were significantly more attractive to bed bugs than identical bags containing clean clothes. Human odour is a known cue for host-seeking bed bugs11, but until now studies have focussed on how these cues determine the attractiveness of a sleeping host. Our results show that bed bugs will move to, and aggregate on, soiled clothing in the absence of a host. Bed bugs can sense 104 different volatiles16 found in the odorant profile of human skin25. Many of these volatiles are likely to evaporate from ‘soiled’ clothes for some time after they have been worn, removed and/or stored. We propose that residual human odour on soiled clothes acts as an elicitor of host-seeking or host-seeking-like behaviours, and that cues such as heat9 or elevated CO2 may facilitate the activation and sensitivity of such host-seeking behaviour. Consequently, soiled clothing left in an open suitcase, or left on the floor, of an infested room is likely to attract bed bugs. When packed into the suitcase, they will accompany their host back home.

Our results show that over a period of several days bed bugs are attracted to, and remain on, soiled clothing: this provides a biologically realistic mechanism that underpins passive, long-range dispersal in bed bugs. The advent of relatively frequent, short-stay holidays in locations long distances from the hosts’ residence will facilitate the proposed mechanism of dispersal. Careful management of holiday clothing may be an important strategy in the prevention of bringing home bed bugs.


Experimental production of soiled clothing

Four volunteers were chosen at random from a pool of eight for each run of the experiment. They washed using a non-perfumed soap (Simple® soap, Unilever, UK) and then wore a white, 100% cotton, t-shirt and socks (Sportee range, Decathlon, UK), for three hours during normal daily activity between 12:00 and 18:00. The clothes were immediately sealed in an airtight bag (Ziploc™, 18 cm × 19 cm) and were used in the experiment within 24 hours of being worn. Clothes and cotton bags (see below) were washed (90 °C) with a non-perfumed detergent (Surcare™ Non-Biological laundry liquid – ingredients include 5–15% non-ionic surfactants, anionic surfactants, soap, < 5% phosphonate) between trials. Clothes were placed in clean cotton tote-bags (38 × 43 cm, Clever Baggers Ltd, UK) and were washed after every use.

Experimental rooms

Experiments were conducted simultaneously in two identical, temperature-controlled rooms (4.8 m (L) × 4.3 m (W) × 2.4 m (H)). Each room was maintained at 22 °C ± 0.5 °C, 55% ± 10% relative humidity and 12 h L:D with a 30-minute ramping period to simulate dawn and dusk. Ventilation within each room provided six complete air changes per hour. During each experiment one room was designated the ‘elevated CO2 treatment’ room and received an increase in CO2 concentration in the centre of the arena at ground level by 28 ± 1 ppm (Fig. S1) to simulate a situation where a human is breathing away from the clothing, (a single inactive adult sitting in the same room −1.5 m from the centre - increased the CO2 concentration by ca. 14 ± 1 ppm at the centre). CO2 was generated by placing approx. 2 kg dry ice in a polystyrene container (25 × 25 × 23 cm) 1.5 m from the centre of the room. Dry ice was replaced daily during the light-phase. The other room was designated the ‘ambient CO2 treatment’ and had a polystyrene container with no dry ice.

Four clean cotton tote bags, each containing one t-shirt and one pair of socks (two bags with soiled clothes, two with clean clothes), were placed 1 m from the room centre at 90 degrees from each other (Fig. S1) alternating between clean and soiled contents. Therefore, bags containing the same treatment were always opposite each other. The cross-shape formation of the bags (Fig. S1) was maintained at all times, but rotated around a virtual clock face. The position of the first bag was chosen at random from numbers 1–12 using a random number generator (Fig. S1), which then in turn determined the position of the other bags. Between each experimental run the floor was cleaned using bleach (Mexcel®, SLS, UK) and remained empty for 24 hours before the next experimental run.

Experimental procedure

Bed bugs were fed to satiation on a human blood meal 24 hours prior to being placed in the arena. A standard laboratory rearing container for bed bugs (a 60 ml clear acrylic pot with 5 cm × 20 cm corrugated filter paper – henceforth ‘the refuge’) containing 5 male and 5 female adult bed bugs was placed under a 10 × 10 × 10 cm clear plastic box in the centre of the room for 48 hours (to prevent the bugs dispersing as a potential escape response). Four clean tote-bags (containing the clean and soiled clothes) were then introduced into the room. Twenty-four hours later the clear plastic box over the refuge was removed (giving the bugs unhindered physical and sensory access to the room). No interim measurements were taken and the arena was observed via a viewing window. After a further 96 hours, the number and location of each bed bug was recorded. Location was categorised as remaining in the original refuge, within/on clothing bag, or on the floor of the arena. The experiment was repeated six runs, alternating which room was designated to ambient or elevated CO2.

Statistical analysis

Analysis was performed with R version 3.3.1 (R Core Team 2016). Using a GLMM17 with a binomial error distribution, the response variable bed bug count, was generated as a proportion of the number of available bed bugs – number found at different locations within the room. This method accounted for any individual that died (5 over the six runs). To account for uncontrolled variation within the rooms (e.g. slight variation in lighting or ventilation), room and experimental run were used as nested random terms within the model. Minimum adequate models were determined from stepwise exclusion of parameters from a full model using log-likelihood tests18.

Data availability

The data that supports the findings of this study are available via The University of Sheffield Online Research Data Archive.

Additional Information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    Reinhardt, K. & Siva-Jothy, M. T. Biology of the bed bugs (Cimicidae). Annu. Rev. Entomol. 52, 351–374 (2007).

  2. 2.

    Fountain, T., Duvaux, L., Horsburgh, G., Reinhardt, K. & Butlin, R. K. Human-facilitated metapopulation dynamics in an emerging pest species. Cimex lectularius. Mol. Ecol. 23, 1071–1084 (2014).

  3. 3.

    Edmundo, M. Defence in animals a survey of anti-predador defences (ed. Edmundo, M). (New York, Longman, (1974).

  4. 4.

    Cooper, R., Wang, C. L. & Singh, N. Mark-release-recapture reveals extensive movement of Bed bugs (Cimex lectularius L.) within and between apartments. Plos One 10 (2015).

  5. 5.

    Akhoundi, M. et al. Spatial genetic structure and restricted gene flow in bed bugs (Cimex lectularius) populations in France. Infect. Genet. Evol. 34, 236–243 (2015).

  6. 6.

    Raab, R. W. et al. New introductions, spread of existing matrilines, and high rates of pyrethroid resistance result in chronic infestations of Bed bugs (Cimex lectularius L.) in lower-income housing. PLoS ONE 11, e0117805 (2016).

  7. 7.

    Johnson, C. G. The ecology of the bed-bug, Cimex lectularius L., in Britain: Report on Research, 1935–40. J. Hyg. – Lond. 41, 345–461 (1941).

  8. 8.

    Lazzari, C. R. Orientation towards hosts in haematophagous insects: an integrative perspective. Adv. Insect Physiol. 37, 1–58 (2009).

  9. 9.

    Legrand, P., Verheggen, F., Haubruge, E. & Francis, F. Host-seeking behavior in the bed bug (Cimex lectularius) and applications in integrated pest management. A review Description of the subject. Biotechnol. Agron. Soc. Environ. 20, 195–202 (2016).

  10. 10.

    Anderson, J. F., Ferrandino, F. J., McKnight, S., Nolen, J. & Miller, J. A carbon dioxide, heat and chemical lure trap for the bedbug. Cimex lectularius. Med. Vet. Entomol. 23, 99–105 (2009).

  11. 11.

    Harraca, V., Ryne, C., Birgersson, G. & Ignell, R. Smelling your way to food: can bed bugs use our odour? J. Exp. Biol. 215, 623–629 (2012).

  12. 12.

    Singh, N., Wang, C. L. & Cooper, R. Effectiveness of a sugar-yeast monitor and a chemical lure for detecting Bed bugs. J. Econ. Entomol. 108, 1298–1303 (2015).

  13. 13.

    Ray, A. Reception of odors and repellents in mosquitoes. Curr. Opin. Neurobiol. 34, 158–164 (2015).

  14. 14.

    Webster, B. & Cardé, R. T. Use of habitat odour by host-seeking insects. Biol. Rev. (2016).

  15. 15.

    Lacey, E. S., Ray, A. & Cardé, R. T. Close encounters: contributions of carbon dioxide and human skin odour to finding and landing on a host in Aedes aegypti. Physiol. Entomol. 39, 60–68 (2014).

  16. 16.

    Liu, F. & Liu, N. N. Human odorant reception in the common Bed bug, Cimex lectularius. Scientific Reports 5, 15558 (2015).

  17. 17.

    Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting linear mixed-effects modelts using lme4. J. Stat. Softw. 67, 1–48 (2015).

  18. 18.

    Lewis, F., Butler, A. & Gilbert, L. A unified approach to model selection using the likelihood ratio test. Methods Ecol. Evol. 2, 155–162 (2011).

  19. 19.

    Dekker, T., Geier, M. & Cardé, R. T. Carbon dioxide instantly sensitizes female yellow fever mosquitoes to human skin odours. J. Exp. Biol. 208, 2963–2972 (2005).

  20. 20.

    Webster, B., Lacey, E. F., Carde, R. T. & Carde, R. T. Waiting with bated breath: opportunistic orientation to human odor in the malaria mosquito, Anopheles gambiae, is modulated by minute changes in carbon dioxide concentration. J. Chem. Ecol. 41, 59–66 (2015).

  21. 21.

    Springer, Y. P., Taylor, J. R. & Travers, P. D. Variation in baiting intensity among CO2-baited traps used to collect hematophagous arthropods. J. Insect Sci. 15, 92 (2015).

  22. 22.

    Aak, A., Rukke, B. A., Soleng, A. & Rosnes, M. K. Questing activity in bed bug populations: male and female responses to host signals. Physiol. Entomol. 39, 199–207 (2014).

  23. 23.

    Pfiester, M., Koehler, P. G. & Pereira, R. M. Effect of population structure and size on aggregation behavior of Cimex lectularius (Hemiptera: Cimicidae) J. Med. Entomol. 46, 1015–1020 (2009).

  24. 24.

    DeVries, Z. C., Kells, S. A. & Appel, A. G. Standard metabolic rate of the bed bug, Cimex lectularius: Effects of temperature, mass, and life stage. J. Insect Physiol. 59, 1133–1139 (2013).

  25. 25.

    Dormont, L., Bessiere, J. M. & Cohuet, A. Human skin volatiles: a review. J. Chem. Ecol. 39, 569–578 (2013).

Download references


Funding for this project provided by a grant from the Dept. of Animal & Plant Sciences, Sheffield University. We would like to thank Rebecca Wilson, Catherine Oke and Laura Bulson for their help in data collection.

Author information


  1. The Department of Animal and Plant Sciences, The University of Sheffield, Sheffield, UK

    • William T. Hentley
    • , Ben Webster
    • , Sophie E. F. Evison
    •  & Michael T. Siva-Jothy


  1. Search for William T. Hentley in:

  2. Search for Ben Webster in:

  3. Search for Sophie E. F. Evison in:

  4. Search for Michael T. Siva-Jothy in:


The experiment was conceived and designed by W.H., B.W., M.S.-J., was delivered and analysed by W.H. with assistance from S.E. The manuscript was written by W.H. and M.S.-J.: all authors critically discussed the experimental results / implications and commented on the manuscript at all stages.

Competing Interests

The authors declare that they have no competing interests.

Corresponding author

Correspondence to William T. Hentley.

Electronic supplementary material


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Creative Commons BY

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit