|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
Chikungunya infection is generally described as a febrile illness of sudden onset, often accompanied by headache, myalgia, rash, and the characteristic arthralgia that led to its name. Chikungunya means "that which bends up" in the language of the inhabitants of the Tanzanian Makonde plateau. Other less frequent clinical manifestations include nausea, vomiting, and hemorrhagic symptoms such as epistaxis or gum bleeding. Most patients recover quickly without sequelae and acquire immunity against the virus.18 However, chronic forms are sometimes observed, which are characterized by persistent arthralgia that may be incapacitating.19 Neurologic and sensory complications have been reported but are poorly documented.20
On March 16, 2005, the World Health Organization (WHO) sent an alert to the Global Alert and Response Network about an epidemic of chikungunya fever in the Comoro Islands. This was the first known emergence of the virus in the southwestern Indian Ocean region. By mid-April, suspected cases imported from Grande-Comore were reported in Mayotte, and the epidemic reached Mauritius at the end of the month. In Réunion, a serologically confirmed case of chikungunya virus infection imported from Grande-Comore was reported on April 29, and three local suspected clinical cases were reported on May 3. A field survey conducted by the vector control team in the neighborhood of the cases already reported identified numerous additional suspected cases, most of them autochthonous, and three more confirmed cases were reported. These findings were consistent with the presence of viral circulation on Réunion Island, where it persisted until the beginning of year 2007, although with a sporadic pattern. The present work describes the course of this epidemic, one of the largest so far observed.
| METHODS |
|
|
|---|
|
Biologic confirmation methods. Serologic confirmation was based on IgM antibody-capture enzyme-linked immunosorbent assay using anti-human IgM as the capture antibody, chikungunya antigen obtained from cellular culture and hyperimmune mouse ascitic fluids detected with peroxidase-conjugated anti-mouse antibodies. Results were considered positive when the optical density of the assay exceeded the mean optical density of negative controls plus 3 standard deviations.16 The RT-PCR was performed using a SuperscriptTM One-Step RT-PCR with a Platinum Taq kit. (Invitrogen, Carlsbad, CA).21 Physicians were encouraged to seek biologic confirmation of every atypical case.
Community surveillance. After the international WHO alert of March 2005, the local health authorities set up an island-wide operational epidemiologic surveillance system for chikungunya infections as soon as the first cases were identified and reported in April 2005. The objectives of this surveillance system were to monitor epidemic trends, characterize cases, and detect new transmission clusters to provide orientations for prevention and vector control. Until December 2005, it relied on vector control teams, which conducted active and retrospective case-finding around the cases reported by a sentinel physician network, medical laboratories, private practitioners, and patients themselves. After validation, data were seized and duplicates were eliminated in an EpiData database managed by the team of the Cellule Interrégionale dEpidémiologie. However, by December 2005, the number of cases exceeded the capacity of this surveillance system, which was no longer able to follow the epidemic trends. Surveillance was then entirely based on the sentinel network. The validity of the epidemic curve was assessed by comparison with a variety of other indicators, including reports from physicians outside the sentinel network, hospital activity, data from health insurance funds, and self-reports by the population to a toll-free telephone line. To extrapolate the total number of cases from the sentinel network data, a coefficient was calculated after confirmation of the correlation between the number of suspected chikungunya cases reported by the 31 network physicians and the number of suspected symptomatic cases identified by the vector control teams during the first 40 weeks of the epidemic. It was thus determined that in Réunion, one case reported by a physician belonging to the sentinel network corresponded to 67 symptomatic chikungunya cases (95% confidence interval = 0–150). From week 51 of 2005 onwards, this central value was applied to the weekly consolidated number of cases reported by the network to estimate the total number of cases.
Mortality and hospital-based surveillance. Mortality data were recorded from the death certificates on which physicians mentioned chikungunya among the causes of death. An active search for emerging severe forms of the disease and for materno-neonatal transmission was also implemented in the islands hospitals from patients medical files.
| RESULTS |
|
|
|---|
|
2 = 2,518, degrees of freedom [df] = 9, P < 0,001). During the subsequent epidemic period, the age distribution of cases reported by the sentinel network was closer to that of the general population. However, significant differences were still noted (
2 = 89, df = 9, P < 0,001): persons more than 79 years of age, young adults 20–29 years of age, and children less than 10 years of age were the least affected age groups. Women were over-represented in both epidemic periods (
2 = 217, df = 1, P < 0,001 and
2 = 11, df = 1, P < 0,001) and in all age groups except those less than 20 years of age.
|
Laboratory results. Over the entire study period (January 2005–May 2006), 16,050 cases were laboratory confirmed, 991 of them by RT-PCR only, 14,727 by serologic analysis only, and 332 by both methods.
Emerging severe disease.
By April 16, 2006, 123 cases with emerging severe disease suspected by physicians to be associated with biologically confirmed chikungunya infection had been recorded among patients more than 9 days of age. The most frequent complications associated with these confirmed cases were respiratory failure (n = 19), cardiovascular decompensation (n = 18), meningoencephalitis (n = 16) or another central nervous system problem (n = 7), severe acute hepatitis (n = 11), severe cutaneous effects (n = 10), and kidney failure (n = 7). More than half of these serious cases (65 [53%]) involved patients
65 years of age, and more than one-third (43 [35%]) died.
Also, by April 16, 2006, 55 cases of materno-neonatal transmission were reported in infants less than 10 days of age; 41 of these cases were confirmed and included 10 infants with meningoencephalitis and 1 death.
Deaths.
By April 16, 2006, 203 death certificates reporting chikungunya infection had been received at the Regional Health and Welfare Office; 121 of these certificates (60%) mentioned the infection among the diseases leading directly to death, whereas the others listed it as a morbid condition that may have contributed to death. According to the 203 certificates, the overall mortality rate associated with chikungunya virus infection was 0.3/1,000 persons, and age-specific mortality rates increased markedly with age. The median age at death was 79 years (Figure 3
).
|
| DISCUSSION |
|
|
|---|
The epidemic reported here has affected every locality on Réunion Island, both sexes, and all age groups. The two epidemic peaks identified were separated by the austral winter, during which the number of cases decreased considerably, but viral transmission did not cease completely. Several factors may have contributed to the exceptionally dynamic characteristics of the epidemic: the immune status of the population relative to this emerging virus, housing density, vector density and other entomologic characteristics, knowledge of and behavior of the population towards individual protection, and control of larval habitats.
As already observed in Indonesia between 2001 and 2003, children less than nine years of age were one of the least affected age groups, especially during the first epidemic period.13 Similarly, over-representation of women among the cases has also been reported.7,13 One explanation is that women may be more exposed to the vector because of greater home and gardening activities because backyards are especially liable to contain vectorial breeding sites. Another possible explanation might be the differences between clothing of men and women regarding exposure of the skin, especially the ankles and legs. Similar differences in sex-related attack rates were observed during a limited dengue outbreak that occurred on Réunion Island in 2004. Further studies are under way to determine whether the change in age and sex distribution of cases observed when the epidemic became massive may be explained by a change in the epidemiologic pattern or by the modification of the surveillance system.
Until December 2005, the surveillance system established for this epidemic of chikungunya virus infection was based on active case-finding conducted by vector control teams. Transmission areas could thus be identified even when the incidence was low and geographically heterogeneous. However, this surveillance system and its subsequent adaptation when the epidemic reached massive proportions were based on symptomatic suspected cases. The real number of infected persons and the resulting herd immunity may therefore have been underestimated by the omission of asymptomatic forms of the disease. Furthermore, the ratio of the total number of cases to the number reported by sentinel network physicians may have changed during the course of the epidemic. Despite these limitations, the data obtained made it possible to describe the epidemiologic trends, target disease control measures, and generate research hypotheses.
The most frequent clinical symptoms, sudden onset of high fever and joint pains, fitted the definition of suspected cases. Approximately 3% of the reported cases did not meet this definition, perhaps in part because of errors in reporting, but mainly because laboratory testing confirmed some atypical cases because physicians were encouraged to seek laboratory confirmation if in doubt. Thus, it cannot be excluded that some of these confirmed cases were consulting for unrelated problems and happened to have had an asymptomatic chikungunya virus infection in the recent past.
Only one study has dealt with neurologic effects in patients infected with chikungunya virus.20 During the epidemic on Réunion Island, other severe forms of the disease were recorded in addition to neurologic forms. These forms include respiratory, cardiovascular, hepatic, cutaneous, and renal syndromes, although cause-and-effect relationship has not been ascertained. Whatever relationship they might have with the virus, these forms only account for a tiny fraction of the total number of estimated cases. Clinical studies are under way to determine the physiopathologic processes involved. At this stage, however, health professionals dealing with a large outbreak of infection with chikungunya virus should be aware that such forms might emerge, as occurred on Réunion Island, where they greatly affected the work of intensive care units.
Conversely, before this epidemic, mortality linked to infection with chikungunya virus had never been reported. In January 2006, numerous death certificates, mainly concerning the elderly, began to mention chikungunya, either among the diseases that led directly to death or as an associated morbid event. It is nonetheless difficult to determine to what extent these deaths can be attributed to chikungunya virus in the absence of autopsies that might have confirmed a direct link between death and infection. Studies of global mortality were conducted to compare the number of deaths recorded on Réunion Island during this epidemic with the number of deaths expected, which were estimated from mortality data for previous years that included demographic information. These studies confirmed significant excess mortality that coincided with the epidemic incidence peak in 2006.23
Vector control of Aedes albopictus, which was identified as the main epidemic vector, continues throughout the island and combines measures against both adult mosquitoes and larvae. Deltamethrin is the insecticide currently used against adult mosquitoes, and has replaced fenitrothion, which was used initially. Larval control is mainly based on mechanical destruction of breeding sites; when this not possible, Bacillus thuringiensis israelensis, a biologic larvicide that has replaced temephos, is used. Fenitrothion and temephos were replaced in February 2006 to reduce hazards for non-professional users and possible damage to the environment when vector control activities became massive. These activities, combined with individual measures of protection and community action in the management of solid waste and of containers that may serve as breeding sites in a tropical climate, are the fundamental measures of control of chikungunya virus on Réunion Island. Their impact on the course of the epidemic remains to be evaluated. More rigorous entomologic surveillance and vector control appear to be critical for the reduction of the health risks linked not only to chikungunya virus but also to other arboviruses likely to emerge on Réunion Island because this island is accessible to their introduction.
The scale of the epidemic on Réunion Island has required the adaptation of the surveillance system. Further adaptation will be necessary when the number of cases decreases. An early detection system for cases of infection with chikungunya virus will be an important issue once the incidence on Réunion Island decreases to a low level. With this in view, an automated system of teletransmission is under study to facilitate and accelerate the communication of reports by local physicians.
At the same time, research is under way to improve understanding of the characteristics of the epidemic and monitor its repercussions, especially the course of chronic forms of infection and of children born to infected mothers. At the end of the epidemic, a seroprevalence survey will be conducted to estimate the proportion and characteristics of the population infected by chikungunya virus, including its asymptomatic forms.
Received August 2, 2006. Accepted for publication April 16, 2007.
Acknowledgments: We thank the sentinel physicians, biologists, hospital physicians, and healthcare professionals for caring for their patients and participating in the surveillance system by reporting cases; Hervé Zeller, Isabelle Schuffenecker, and the National Center for Arboviruses for biologic expertise; the vector control teams for vector control efforts and active case-finding in the field; and the Institut de Veille Sanitaire, especially Brigitte Helynck, Florence Kermarec, and Christophe Paquet, for sustained assistance and support on site; and Assimoula Economopoulo, Morgane Dominguez, and Charlotte Renaudat for their intensive work on cases with severe disease. We also thank Intute Publishers (Oxford, United Kingdom) for permission to use Figure 1
.
* Address correspondence to Philippe Renault, Cire Réunion-Mayotte, BP 50, 97408 Saint-Denis Cedex 9, Réunion, France. E-mail: philippe.renault{at}sante.gouv.fr ![]()
Authors addresses: Philippe Renault, Jean-Louis Solet, Daouda Sissoko, Elsa Balleydier, and Vincent Pierre, Cire Réunion-Mayotte, BP 50, 97408 Saint-Denis Cedex 9, France. Sophie Larrieu and Laurent Filleul, Cire Aquitaine, Espace Rodesse, 103 bis Rue Belleville, BP 952, 33063 Bordeaux Cedex, France. Christian Lassalle and Julien Thiria, Regional Health and Welfare Office de la Réunion, BP 50, 97408 Saint-Denis Cedex 9, France. Emmanuelle Rachou, Observatoire Régional de la Santé de la Réunion, 12 Rue Colbert, 97400 Saint-Denis, France. Henriette de Valk, Daniele Ilef, Martine Ledrans, and Isabelle Quatresous, Institut de Veille Sanitaire, 12 Rue du Val dOsne, 94441 Saint-Maurice Cedex, France. Philippe Quenel, Cire Antilles-Guyane, Centre dAffaires Agora, ZAC de lEtang ZAbricot, Pointe des Grives, BP 658, 97261 Fort-de-France Cedex, France.
| REFERENCES |
|
|
|---|
This article has been cited by other articles:
![]() |
C.-K. Lim, T. Nishibori, K. Watanabe, M. Ito, A. Kotaki, K. Tanaka, I. Kurane, and T. Takasaki Chikungunya Virus Isolated from a Returnee to Japan from Sri Lanka: Isolation of Two Sub-Strains with Different Characteristics Am J Trop Med Hyg, November 1, 2009; 81(5): 865 - 868. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Desvars, T. Duval, C. Punelle, M. Pascal, and G. Vourc'h The Flying Fox Pteropus seychellensis of Mayotte (Comoros): Method of Capture and Blood Sampling J. Wildl. Dis., July 1, 2009; 45(3): 870 - 873. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Weissmann Gore's Fever and Dante's Inferno: Chikungunya Reaches Ravenna FASEB J, March 1, 2008; 22(3): 635 - 638. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |