Dengue Surveillance in the French Armed Forces: A Dengue Sentinel Surveillance System in Countries Without Efficient Local Epidemiological Surveillance


Corresponding Author: Franck de Laval, MD, CESPA, Camp de Sainte Marthe, BP 40026, F-13568 Marseille Cedex 02, France. E-mail:


Surveillance of travel-acquired dengue could improve dengue risk estimation in countries without ability. Surveillance in the French army in 2010 to 2011 highlighted 330 dengue cases, mainly in French West Indies and Guiana: DENV-1 circulated in Guadeloupe, Martinique, French Guiana, New Caledonia, Djibouti; DENV-3 in Mayotte and Djibouti; and DENV-4 in French Guiana.

Dengue is a worldwide public health problem for local populations of endemic areas, travelers, and expatriates.[1-4] Each year, 50 million dengue infections occur among the 2.5 billion people living in areas where dengue can be transmitted, 12,000 of which lead to death.[5]

Biological and epidemiological surveillance results are essential to identify the risk of dengue in a population (monitoring of virus circulation and serotype), and to issue public health emergency alerts (acute increase of the dengue incidence rate). The World Health Organization (WHO) is building a global dengue surveillance system, based on the collection of data from every country.[5] This makes it possible to plan preventive or mosquito control strategies. Nevertheless, the efficiency of epidemiological surveillance is uneven and varies between countries. Dengue circulation and incidence are sometimes underestimated, particularly in Africa.[6] Surveillance of travel-acquired dengue could improve dengue risk estimation in these countries. French soldiers can be considered travelers, since they carry out short missions or can be stationed in dengue endemic areas. Each year, 25,000 French soldiers spend time in an endemic area. Because dengue is a real threat for the French armed forces, this population is under constant epidemiological surveillance. This paper presents the results of dengue virus circulation and dengue incidence rates for all the areas where French armed forces were stationed in 2010 to 2011, which enabled the dengue risk in each area to be identified.


Epidemiological surveillance of dengue in the French armed forces consists of continuous and systematic collection, analysis, interpretation, and feedback of epidemiological data from all military physicians, wherever they are located. Each patient with dengue symptoms requires blood sample. In French overseas departments and territories, samples are analyzed in local civilian laboratories, otherwise samples are sent to the National Arbovirus Reference Center based at the Institute of Tropical Medicine at the Army Health Service, Marseille, France (tests used are in-house assay, Mac Elisa and direct IgG Elisa).[7] Virus culture and/or reverse transcription polymerase chain reaction (RT-PCR) are carried out if an early sample is available; otherwise, serology is performed. Complementary Ag NS1 could be performed directly in local laboratories.

A specific individual dengue case report form, containing administrative, geographical, clinical, and biological data, is also sent to the Institute of Tropical Medicine at the Army Health Service, Marseille, France.

Possible dengue was defined in an epidemic context of dengue as a fever higher than 38.5 °C associated with at least one of the following symptoms: headache, myalgia, retro-orbital pain, rash, hemorrhagic signs. Confirmed dengue was defined as any of the above symptoms with virological evidence (PCR, culture, NS1 antigenemia) or positive serology (IgM or IgG seroconversion). Here we report the results of analysis of the data obtained from specific dengue case report forms from January 1, 2010 to December 31, 2011. Indicators are expressed as annual incidence and annual incidence rate. The denominator for the incidence rate is the average number of soldiers present in each dengue endemic area in 2010 to 2011. Statistical analysis was performed using R software.


In 2010 to 2011, 208 possible dengue cases and 122 confirmed dengue cases occurred in the French armed forces. On average 25,458 French soldiers were present each year in dengue endemic territories (Table 1). The sex ratio was 9/1 (6/1 in the armed forces as a whole); median age was 33 years (range: 19–56).

Table 1. Dengue incidence and serotype in the French armed forces in 2010 to 2011 according to location, with diagnostic method
LocationNumber of French soldiersNumber of casesMethod of confirmationSerotypes evidenced (and number)Incidence rate

(per 1000


TotalConfirmedCultureRT-PCRAg NS1IgM Serology
  • *

    One-time ship stopover.

Guadeloupe174710029711291 (×3)8
French Guiana40124941333321 (×3) − 4 (×2)5
Haiti9744  31 21
Martinique238416423 13191 (×1)5
Central African Republic24000     0
Chad98300     0
Djibouti288244 2 21 (×1) − 3 (×1)1
Gabon89200     0
Ivory Coast97211   1 1
Mayotte82322 1 13 (×1)1
Uganda1200     0
Reunion Island364000     0
Senegal121700     0
Tanzania*11   1 
French Polynesia2467211    0
New Caledonia3090221  11 (×1)0
Indonesia*11  1  
Total25,4583301091285237 2

Symptoms and clinical signs were myalgia (95%), fever (94%), headache (90%), retro-orbital pain (56%), rash (25%), and digestive symptoms (21%). Twenty-five patients were hospitalized for observation, but their condition was not serious.

Surveillance results highlighted dengue circulation in the West Indies, French Polynesia, Africa (Djibouti, Ivory Coast, Mayotte, Tanzania), French Guiana, and Indonesia. More exactly, laboratory results enabled the serotype to be identified: DENV-1 in Guadeloupe, Martinique, French Guiana, New Caledonia, and Djibouti; DENV-3 in Mayotte and Djibouti; and DENV-4 in French Guiana.

Incidence rates of dengue according to location are presented in Table 1. The incidence rate was highest in the French West Indies, immediately followed by French Guiana (p < 10−9). The risk was high in the French West Indian islands where an outbreak occurred among the local population during the summer of 2010. No dengue cases occurred in the French military in the Central African Republic, Chad, Gabon, Uganda, Reunion Island, and Senegal.


The limits of epidemiological surveillance have to be taken into account when considering these results. The actual number of cases is usually underestimated, resulting from failure to declare cases:[8] In French overseas departments and territories, patients have access to civilian health care and can thus be missed by military surveillance, whereas when stationed in foreign countries, they do not have that choice, but diagnostic capabilities are not always available. To detect early warning signals for an outbreak, we chose to use a sensitive case definition.[9] That is why possible dengue cases (without biological confirmation but in an epidemic context) and serologically confirmed dengue cases were included. However, serology could create confusion with other flaviviruses due to cross-reactive antibodies. In fact, only confirmed cases using culture, RT-PCR, or Ag NS1 methods were actual dengue cases.

Locations where the French armed forces' epidemiological surveillance system identified dengue circulation in 2010 to 2011 (French West Indies, French Polynesia, French Guiana, Africa, and Indonesia) were well known for dengue virus circulation.[10] In the French West Indies, the serotype was not the same as during the previous outbreak in 2007.[11] DENV-1 and DENV-4 circulated in 2010, whereas DENV-2 circulated in 2007. This type of situation is usually responsible for intense virus circulation and therefore for outbreaks. Serotype identification is very important to highlight epidemic risk. Our circulation results were complementary to WHO global surveillance results, and could serve to improve knowledge about serotype circulation, that is, detection of DENV-1 circulation in New Caledonia,[12] and DENV-3 in Djibouti. Previous French military biological results in Africa from 1998 to 2010 already highlighted circulation of DENV-1 in Cameroon, Djibouti, Gabon, Mayotte and DENV-3 in Comoros (periods of both studies are successive).[6, 7]

The French armed forces epidemiological surveillance system also made it possible to identify epidemic transmission in the French West Indies since June 2010, and therefore to increase mosquito control measures. This outbreak enabled a comparison between incidence rates from the local civilian surveillance system and the French military system (respectively 10% and 6%, p < 10−9).[13] However, similar upward trends were observed. Civilian and military epidemic peaks occurred at the same time, in August 2010. Either military mosquito control measures protected soldiers from dengue infection, or the military surveillance system was less efficient or sensitive. In these French territories, many soldiers consulted civilian instead of military physicians. That is not the case in foreign territories. However, similar upward trends were observed, with the epidemic peak occurring at the same time. A new, very sensitive early warning system is now being deployed in the French armed forces and will enable detection of very low increases of dengue-like fevers.[14] Therefore, French soldiers could serve as sentinels within the local population, with military epidemiological surveillance making it possible to detect increased virus circulation, in particular in countries without epidemiological tools.


Military epidemiological surveillance systems can detect dengue circulation where soldiers are stationed. Therefore, these systems could be used to evaluate dengue risk in countries without a local epidemiological surveillance system.

Declaration of Interests

The authors state that they have no conflicts of interest.