The Table shows the results of the respective serology assays performed. For each pathogen (with the exception of CCHFV, for which no seroprevalence was reported), 2 categories of a seropositive volunteer were identified: those who showed detectable levels of antibody before and after deployment; and those who only showed positivity on return, with no detectable antibody in predeployment samples. During deployment, the troops showed the highest seroconversion rates for sandfly fever virus (3.1%) and rickettsiae (2.7%). However, seroconversions for hantavirus and C. burnetii (1.3% and 1.7%, respectively) also occurred.

The group of volunteers who were positive for each pathogen before deployment showed that there was a background level of seroprevalence for these pathogens within the UK military. The group of initially antibody-negative personnel seroconverted to the pathogens while on deployment, which suggests that military personnel are being exposed to such diseases while on operations in Afghanistan and that these diseases are being transmitted in the region.

Figure 1 shows the prevalence of volunteers who seroconverted during each deployment period. This comparison enabled us to plot prevalence as a function of season: winter deployments (October–March) compared with summer deployments (March–October). This analysis showed a higher prevalence of seroconversion in the Northern Hemisphere summer months, when the deployment area is hotter and drier than over the winter. Sandfly fever virus showed the greatest seasonal variation, with most cases in the summer months, and C. burnetii showed the least variations, findings that are consistent with these pathogens’ known modes of transmission.

Many seroconversions for the studied pathogens appeared to be asymptomatic. All volunteers took part in a return questionnaire when the postdeployment sample was taken. This questionnaire was used to determine whether those who seroconverted after deployment experienced a “flu-like” illness during deployment. Figure 2 shows that, for all 4 pathogens acquired (Rickettsia spp., C. burnetii, sandfly fever virus, and hantavirus), a proportion of volunteers showed seroconversion but did not report feeling ill. As expected, sandfly fever virus and rickettsial infections showed the highest proportion of asymptomatic cases; just 6.5% and 7.4%, respectively, of those who seroconverted reporting feeling ill (i.e., 93.5% of sandfly fever virus exposures and 92.6% of Rickettsia spp. exposures were asymptomatic). The less common but often more clinically significant pathogens hantavirus and C. burnetii resulted in more reports of illness; 64.7% of C. burnetii seroconversions and 30.8% of hantavirus seroconversions were asymptomatic.

Postdeployment questionnaires showed that nearly one fifth (90/467, 19.3%) of all volunteers reported experiencing “flu-like” symptoms while on operations in Helmand Province. A total of 56.7% of those who recorded >1 episodes of feeling unwell reported experiencing fever only, whereas 12.2% had no additional/specific symptoms; the remaining 31.2% reported either diarrhea and vomiting (15.6%) or “other” symptoms (Figure 3).

UK military personnel consist of ≈237,190 persons (estimate of combined regular and volunteer reserve military personnel in UK, 2010 [UK Ministry of Defence, unpub. data]). Our results show that military personnel are being exposed to, at minimum, Rickettsia spp., C. burnetii, sandfly fever virus, and hantaviruses while on operations in Afghanistan, a region where these agents and resulting diseases are endemic. These results suggest that human or animal (for the zoonotic pathogens) reservoirs likely exist in the local populations of the Helmand region for these diseases.

Troops often patrol through vegetation and farm land where both reservoir and vector populations are high, which can easily result in exposure to pathogens and further supports the hypothesis that the nature of their work makes armed forces personnel a group with an increased incidence of exposure to such vector-borne or zoonotic pathogens. However, this group may also act as a good sentinel cohort for other populations of nonindigenous workers from governmental and nongovernmental organizations.

UK troops do sometimes stop over at British military bases in Cyprus during their return to the United Kingdom. However, these stopovers are typically very short in duration (2–3 days), and they do not occur in an environment where exposure is likely to occur, with the possible exception of sandfly bites. In addition, for each pathogen studied, a proportion of troops showed antibody positivity before deployment, suggesting that they may have been exposed to these pathogens on previous deployments to Afghanistan (before being recruited to the study) or on other exercises and operations around the world.

Overall, seroprevalence for these pathogens appears to be low, although this is hard to determine without a comprehensive study of nondeploying troops. These results may reflect good discipline or education of UK troops in the use of personal protection measures such as the use of DEET-containing insect repellent and mosquito nets, both of which are actively promoted by the UK military. A higher proportion of those deployed over the summer months showed seroconversion, corresponding with an increase in the numbers of biting vectors such as the Phlebotomus sandfly and ticks (rickettsial vectors). Further study of seasonality might show whether these increases correlate with increased military activity (e.g., military patrolling and combat operations increase during the summer months), environmental/entomologic changes (e.g., decreases in rain and/or increases in temperature conditions favorable to agent, reservoir, or host density), or possible lapses in discipline regarding personal protective measures.

At first glance, summer seasonality may seem odd for an increase in hantavirus seroconversion, for which incidence would be more likely to correlate with closer proximity of humans with the rodent reservoir. However, UK summer deployments extend to late September and in some cases the first week or two of October; thus, these deployment continue into early autumn, when rodents are actively moving into buildings. In addition, the apparent lack of rickettsial infections after summer 2010 raises questions of whether this might correlate with a change in vector life-cycle, abundance of reservoir species, changes in insect bite prevention regimens, or changes to local livestock management regimens.

The increased incidence of rickettsial and sandfly fever virus infections that we found among deployed troops does not necessarily mean that these vector-borne pathogens are more prevalent in the environment and local population than the zoonotic pathogens C. burnetii or hantavirus. Rather, these results may mean that mode of transmission and incidence of encounters with the vectors and reservoirs differ. However, our results do indicate that these pathogens may pose the greatest risk to UK troops.

Although most of the diseases we detected are relatively self-limiting after the initial acute infection, our results suggest ways to improve control measures to reduce the rate of transmission. UK military personnel are already given medical advice and education before deployment, emphasizing the prevention of insect bites while deployed by use of arthropod repellent, insecticide-impregnated clothing, and mosquito nets. The observation that rickettsial infections and Q fever might account for a sizeable proportion of cases of undifferentiated febrile illness seen in military field hospitals has led to the empirical use of doxycycline in such cases because Rickettsia and Coxiella species are sensitive to this drug. This practice justifies further research on use of doxycycline for chemoprophylaxis in these instances.

More than half (56.7%) of volunteers who reported feeling unwell specifically reported a fever without other specific symptoms, which is consistent with the “undifferentiated febrile illness” that the pathogens we investigated can cause. The 15.6% of volunteers reporting illness who reported diarrhea and vomiting could have had gastrointestinal infections; localized outbreaks of norovirus and similar infections are common in military operating bases) (15). Of the many volunteers who experienced fever but did not show seroconversion for the 5 pathogens tested here, exposure to influenza virus or other respiratory infections should be considered. In 2011, Eick et al. reported that deployed US troops experienced 30.1% seroconversion for influenza and 6 other respiratory infections (16). However, uptake of influenza vaccination, which is offered to UK military personnel before winter deployments, has increased for each of the past 4 years (2). Data from the militaries of other countries, which might have different deployment patterns and protective measures, should be compared only with caution. In addition, although some volunteers who seroconverted for all 4 pathogens reported flu-like symptoms while deployed, we cannot ascertain whether these symptoms were a result of exposure to the pathogen of interest.

Some volunteers who seroconverted for the pathogens tested were asymptomatic for the duration of their deployments. Clinical disease probably did not develop in these patients, despite evidence of an exposure and an antibody response. Our finding that 65% of acute C. burnetii infections were asymptomatic is consistent with previous reports (17,18), but even in asymptomatic persons infected with this pathogen, long-term complications such as chronic Q fever and Q fever fatigue syndrome may develop. Overall, ≈5% of acute cases progress to chronic Q fever, and complications can include endocarditis requiring prolonged antimicrobial drug treatment and possibly heart valve surgery (19). In addition, Q fever fatigue syndrome may develop in ≈20% of those infected (17), which is generally incompatible with a military career and has substantial effects on patients’ quality of life.

Royal et al. (20) recently reported C. burnetii seroprevalence in US troops deployed to the Al Asad region of Iraq in 2005; a known Q fever outbreak occurred in this region at this time. This smaller study (n = 136) reported a 7.2% prevalence of C. burnetii infection among troops located in that area at the time of the outbreak. The report further supports (in addition to original work by Bailey et al. [13]) the C. burnetii seroprevalence reported here. Although we did not see as high an incidence of seroconversion, we were not specifically looking in an area with a known Q fever outbreak, merely an area in which Q fever is believed to be endemic.

Although this study did not find seroconversion for the viral hemorrhagic fever agent CCHFV, military and public health reports demonstrated that the virus is circulating in the region (21). In 2009, a US Army soldier in the neighboring Kandahar Province died of CCHFV infection (22), and in 2012, a UK citizen returning from the northwest of the country also died (23,24).

In conclusion, this study highlights and confirms the potential for vector-borne and zoonotic diseases that are endemic in southern Afghanistan to emerge or reemerge to pose a substantial public health threats as the country rebuilds its public health infrastructure. A study of this type cannot give a specific indication of prevalence for these pathogens in the local population, but this surveillance can provide a valuable way of exploring emerging disease epidemiology, particularly of vector-borne and zoonotic infections, in areas with poor public health reporting and infrastructure. Our findings of seroconversion for 4 of these pathogens among deployed UK troops reinforce the need for continued surveillance and continued education of health care providers so that, should military operations or environmental factors change in such a way that these modest incidence numbers increase, costly outbreaks can be avoided. This study also highlights the need for rapid, field-capable, point-of-care diagnostics in regions or situations for which full laboratory diagnostic facilities are not practical or available.