• View in gallery

    Map of study area. The smaller upper map identifies Mexico. The main map represents the Yucatán Peninsula, with the states of Campeche, Yucatán, and Quintana Roo. Dots indicate the location of the villages from which field data were taken.

  • View in gallery

    Collections of T. dimidiata before and after the hurricane Isidore. The total number of adults (A) and the respective larval stages (B) are shown per trimester for pre- (left bars) and posthurricane (right bars) collections. *Indicates a significant difference (P < 0.05).

  • View in gallery

    Comparison of T. dimidiata collections before and after Hurricane Isidore. The total number of bugs collected in each village is shown for the indicated trimesters (A–D) for pre- and posthurricane collections. The diagonal dotted line indicates no changes in bug collections between the two periods. Only villages for which pre- and posthurricane data are available are shown. (A) N = 23 villages, (B and C) N = 34 villages, (D) N = 21 villages. Note the log scale.

  • View in gallery

    Maps of T. dimidiata domestic abundance before and after Hurricane Isidore. The total number of bugs collected per village and for each trimester was interpolated for the entire Yucatán Peninsula and color-coded as indicated. Seasonal variations in bug abundance are shown for pre- (A–D) and posthurricane years (E–H). The track of Hurricane Isidore is superimposed on the maps corresponding to the year 2003 with the symbols indicating the position of the hurricane center at 6-hour intervals, starting on September 21st at 1800 UTC.

  • View in gallery

    Relationships between the changes in bug collections and Hurricane Isidore. (A) Relationship between changes in bug collection before and after the hurricane during the months of January–March and the distance from the track of the hurricane (N = 23 villages). The line corresponds to the fitted equation: change in bug number = 16.9 – 3.1Log(distance from the hurricane). (B) Relationship between absolute changes in bug collection during the months of July–September and the distance from the track of the hurricane (N = 34 villages). The line corresponds to the fitted equation: |change in bug number|0.5 = 2.64 – 0.01(distance from the hurricane). (C) Relationship between absolute changes in bug collection during the months of July–September and the cumulative number of fires around the villages for the months of January–July (N = 34 villages). The line corresponds to the fitted equation: |change in bug number| = −0.81 + 0.01(number of fires).

  • 1

    Committe on Climate, Infectious Disease and Human Health, 2001. Under the Weather: Climate, Ecosystems, and Infectious Disease. Washington DC: National Academy Press.

  • 2

    Sanders EJ, Rigau-Perez JG, Smits HL, Deseda CC, Vorndam VA, Aye T, Spiegel RA, Weyant RS, Bragg SL, 1999. Increase of leptospirosis in dengue-negative patients after a hurricane in Puerto Rico in 1996. Am J Trop Med Hyg 61 :399–404.

    • Search Google Scholar
    • Export Citation
  • 3

    Anonymous, 1999. Principales enfermedades infecciosas en Centroamérica duante 1998, antes y despés de Mitch. Rev Panam Salud Publica 6 :440–443.

    • Search Google Scholar
    • Export Citation
  • 4

    Nasci RS, Moore CG, 1998. Vector-borne disease surveillance and natural disasters. Emerg Infect Dis 4 :333–334.

  • 5

    Campanella N, 1999. Infectious diseases and natural disasters: the effects of Hurricane Mitch over Villanueva municipal area, Nicaragua. Public Health Rev 27 :311–319.

    • Search Google Scholar
    • Export Citation
  • 6

    Bissell RA, 1983. Delayed-impact infectious disease after a natural disaster. J Emerg Med 1 :59–66.

  • 7

    Shaman J, Stieglitz M, Stark C, Le Blancq S, Cane M, 2002. Using a dynamic hydrology model to predict mosquito abundances in flood and swamp water. Emerg Infect Dis 8 :6–13.

    • Search Google Scholar
    • Export Citation
  • 8

    Ramsey JM, Ordonez R, Cruz-Celis A, Alvear AL, Chavez V, Lopez R, Pintor JR, Gama F, Carrillo S, 2000. Distribution of domestic triatominae and stratification of Chagas disease transmission in Oaxaca, Mexico. Med Vet Entomol 14 :19–30.

    • Search Google Scholar
    • Export Citation
  • 9

    Zeledon R, Ugalde JA, Paniagua LA, 2001. Entomological and ecological aspects of six sylvatic species of triatomines (Hemiptera, Reduviidae) from the collection of the National Biodiversity Institute of Costa Rica, Central America. Mem Inst Oswaldo Cruz 96 :757–764.

    • Search Google Scholar
    • Export Citation
  • 10

    Galvao C, Jurberg J, Carcavallo RU, Segura CA, Galindez Giron I, Curto de Casas SI, 1998. Distribuição geográfica e dispersão alti-latitudinal de aluns gêneros e espécies da tribo Triatomini Jeannel, 1919 (Hemiptera, Reduviidae, Triatominae). Mem Inst Oswaldo Cruz 93 :33–37.

    • Search Google Scholar
    • Export Citation
  • 11

    Costa J, Peterson AT, Beard CB, 2002. Ecologic niche modeling and differentiation of populations of Triatoma brasiliensis neiva, 1911, the most important Chagas’ disease vector in northeastern Brazil (Hemiptera, Reduviidae, Triatominae). Am J Trop Med Hyg 67 :516–520.

    • Search Google Scholar
    • Export Citation
  • 12

    Gorla DE, 2002. Variables ambientales registradas por sensores remotos como indicadores de la distribución geográfica de Triatoma infestans. Ecología Austral 12 :117–127.

    • Search Google Scholar
    • Export Citation
  • 13

    Dumonteil E, Gourbiere S, Barrera-Perez M, Rodriguez-Felix E, Ruiz-Piña H, Baños-Lopez O, Ramirez-Sierra MJ, Menu F, Rabinovich JE, 2002. Geographic distribution of Triatoma dimidiata and transmission dynamics of Trypanosoma cruzi in the Yucatán peninsula of Mexico. Am J Trop Med Hyg 67 :176–183.

    • Search Google Scholar
    • Export Citation
  • 14

    Dumonteil E, Gourbiere S, 2004. Prediction of Triatoma dimidiata vector abundance and infection rate: a risk map for Trypanosoma cruzi natural transmission in the Yucatan peninsula of Mexico. Am J Trop Med Hyg 70 :514–519.

    • Search Google Scholar
    • Export Citation
  • 15

    Güemez-Pineda M, Quintal-Aviles EF, 2003. Repercusiones del huracán “Isidoro” en la población maya-yucateca. A un año del huracán: Repercusiones del huracán Isidoro en la población yucateca. Merida, Universidad Automa de Yucatán, Yucatán, Mexico.

  • 16

    Dorn PL, Engelke D, Rodas A, Rosales R, Melgar S, Brahney B, Flores J, Monroy C, 1999. Utility of the polymerase chain reaction in detection of Trypanosoma cruzi in Guatemalan Chagas’ disease vectors. Am J Trop Med Hyg 60 :740–745.

    • Search Google Scholar
    • Export Citation
  • 17

    Levine N, 2002. CrimeStat: a spatial statistics program for the analysis of crime incident locations (v. 2.0). Houston, TX, and Washington, DC: National Institute of Justice.

  • 18

    Kulldorff M, 1997. A spatial scan statistic. Comm Stat Theory Meth 26 :1481–1496.

  • 19

    Williams P, 1970. Phlebotomine sandflies and leishmaniasis in British Honduras (Belize). Trans R Soc Trop Med Hyg 64 :317–364.

  • 20

    Sgambatti de Andrade AL, Zicker F, Silva IG, Souza JM, Martelli CM, 1995. Risk factors for Trypanosoma cruzi infection among children in central Brazil: a case-control study in vector control settings. Am J Trop Med Hyg 52 :183–187.

    • Search Google Scholar
    • Export Citation
  • 21

    Catala S, Crocco LB, Morales GF, 1997. Trypanosoma cruzi transmission risk index (TcTRI): an entomological indicator of Chagas disease vectorial transmission to humans. Acta Trop 68 :285–295.

    • Search Google Scholar
    • Export Citation
  • 22

    Rabinovich JE, Wisnivesky-Colli C, Solarz ND, Gürtler RE, 1990. Probability of transmission of Chagas’ disease by Triatoma infestans (Hemiptera: Reduviidae) in an endemic area of Santiago del Estero, Argentina. Bull World Health Org 68 :737–746.

    • Search Google Scholar
    • Export Citation
  • 23

    Boose ER, Foster DR, Barker Plotkin A, Hall BS, 2003. Geographical and historical variation in hurricanes across the Yucatán peninsula. Gómez-Pompa A, Allen MF, Fedick SL, Jiménez-Osornio JJ, eds. The lowland Maya Area: Three Millennia at the Human-Wildland Interface. New York: The Haworth Press Inc., 495–516.

 

 

 

 

EFFECT OF HURRICANE ISIDORE ON TRIATOMA DIMIDIATA DISTRIBUTION AND CHAGAS DISEASE TRANSMISSION RISK IN THE YUCATÁN PENINSULA OF MEXICO

View More View Less
  • 1 Laboratorio de Parasitología, Centro de Investigaciones Regionales “Dr. Hideyo Noguchi,” Universidad Autónoma de Yucatán, Mérida, Mexico

Hurricanes can have devastating effects on health and may directly modulate vector-borne diseases. Chagas disease is a zoonosis caused by the protozoan parasite Trypanosoma cruzi and transmitted by triatomine bugs, and the effect of hurricanes on these bugs is largely unknown. We thus performed a detailed study of the changes in Triatoma dimidiata geographic distribution and infection rates after Hurricane Isidore devastated the Yucatán Peninsula in September 2002. Bugs were collected in 34 villages from the entire peninsula, during a year, starting 3 months after the hurricane. Pre- and posthurricane bug collections were compared to assess changes. The most notable effect was a large increase in domestic abundance of T. dimidiata during the 6 months after the hurricane. This increase was maximum along the path of the hurricane. These results suggest that vector control programs should be implemented along the path of hurricanes to prevent an increase in Chagas disease transmission risk in the ensuing months.

INTRODUCTION

It is well established that climate can have a profound impact on the epidemiology of infectious diseases.1 In particular, natural disasters such as hurricanes or floods may lead to infectious disease outbreaks. The risks associated with water-borne diseases are fairly well understood, and outbreaks of cholera and leptospirosis have been associated with disruptions in public health and sanitation services after natural disasters.2,3

With vector-borne diseases, the situation is much more speculative.4 Hurricanes and floods may affect vector populations in several complex ways that can either increase or decrease disease transmission. However, there are very few studies documenting vector-borne disease outbreaks associated with disasters.2,3,5 Also, most of these studies only cover a few weeks after a disaster, so that long-term effects cannot be assessed.6 Nonetheless, a good understanding of these phenomena is required to ensure optimal public health disaster response measures. For example, a recent modeling study of mosquito population based on a dynamic hydrology model showed that it is possible to predict mosquito abundance and particularly major emergence of mosquitoes after flooding7 with sufficient anticipation to allow prevention of mosquito-borne diseases after such disaster.

Triatomine bugs and Chagas disease have also been found to be influenced by climatic factors.812 Particularly, in the Yucatán Peninsula, we described an important heterogeneity and seasonality in the geographic distribution of Triatoma dimidiata, the main vector of Chagas disease in the region.13 We further showed that several climatic factors such as temperature, humidity, and rainfall were important predictors of house infestation by T. dimidiata as well as of their infection rate by Trypanosoma cruzi parasites. This allowed the elaboration of the first transmission risk map as a decision-making tool for vector control in the region.14

Hurricane Isidore was a category III hurricane on Saffir-Simpson scale and devastated the Yucatán Peninsula in September 2002. It caused more than US$16 million in property damage, in particular to the agricultural and fishing sectors. More than 30,000 head of livestock and 100,000 chickens and turkeys were lost, and there was an estimated total of 8 million cadavers from domestic and sylvatic animals.15 In addition, the accumulation of organic material from fallen trees and broken branches led to an important increase in the number of fires during the dry season, 7–9 months after the hurricane. It is likely that such fires may have generated further ecological perturbations. We hypothesized in previous studies that house infestation by T. dimidiata is mostly due to dispersing bugs from around the dwellings.13,14 Thus, we suspected that ecological perturbations after such a hurricane would translate into significant changes in house infestation by triatomines and in Chagas disease transmission risk. We thus investigated in this study the potential changes in T. dimidiata distribution after Hurricane Isidore, to provide a framework to assess its possible effects on Chagas disease transmission in the region.

MATERIALS AND METHODS

Study area and bug collection.

The Yucatán Peninsula is located in the southeast of Mexico, between longitude 86–92° W and latitude 17–22° N. It includes the Mexican states of Campeche, Yucatán, and Quintana Roo.

Prehurricane baseline entomologic data were obtained from previous studies.13,14 These corresponded to T. dimidiata collections and infection rates in 115 houses from 23 villages studied between October 1999 and September 200013 and an additional 55 houses from 11 villages studied from January to December 2001,14 giving a total of 170 houses from 34 villages monitored each trimester during 1 year (Figure 1). This composite data set for prehurricane house infestation thus allowed integration of some yearly variations in our baseline data. Posthurricane bug collections were initiated in January 2003, 3 months after the hurricane, and lasted until December 2003. A major concern was to achieve a capture effort that would be consistent and comparable with our pre-hurricane collections, to minimize changes due to sampling strategy. Insect collections were thus conducted in the same georeferenced villages, using the same methodology as our prehurricane collections.13 Briefly, after a short interview, households were provided with plastic vials labeled for domicile and peridomestic areas and were requested that any triatomine found in each area be collected in their respective vials. Households were then visited every 3 months to collect the insects captured during the previous 3-month period and to provide them with new vials.

One hundred eight of 170 households (63%) had participated in our previous studies, and the remaining 62 (37%) participated for the first time. All the participants already knew about triatomine bugs and were able to identify them at the beginning of the study. New and previous participants collaborated equally to the collections (see “Results”). As in our previous work,13,14 interviews carried out at each visit revealed that some participants developed increased interest during the study, whereas others showed decreased interest. However, their random distribution resulted in a homogenous and overall constant collection effort among villages and during the year, so that variations due to sampling would be minimal and randomly distributed in space and time.

Diagnosis of Trypanosoma cruzi infection.

The presence of T. cruzi in triatomine feces was detected by direct microscopic observation for live insects and by PCR for dead insects, as previously described.13,16

GIS database elaboration.

Data on Hurricane Isidore were obtained from the National Hurricane Center, Miami, Florida (http://www.nhc.noaa.gov/). Data on fire occurrence for each trimester of years 2000, 2001, and 2003 were obtained from the Web Fire Mapper (http://maps.geog.umd.edu/default .asp). The number of fires in a 7.5 × 7.5 km area centered on each village was calculated. Entomological data were imported into a raster-based GIS (MacGIS 3.0, University of Oregon, Eugene, OR). Point data from bug collections were interpolated to provide continuous data for the whole study area. All maps were set at a pixel resolution of 13 × 13 km.

Data analysis.

Changes in domestic bug abundance were assessed by comparing pre- and posthurricane collections for each village during each trimester, with Wilcoxon nonparametric test for paired groups. Changes in T. cruzi infection rates were assessed by Fisher exact tests. Spatial clustering of changes in bug abundance was assessed by Moran I statistic using CrimeStat 2.017 and Bernoulli spatial scan test using SaTScan 5.0.18 The relationships between changes in bug numbers before and after the hurricane and the distance of the villages from the hurricane center or fire density were assessed by regression analysis. All nonspatial statistical analyses were carried out using JMP 5.0 software (Cary, NC).

RESULTS

Changes in Triatoma dimidiata abundance and infection rates after Hurricane Isidore.

Because the effects of ecological perturbations after hurricanes on triatomine bugs and Chagas disease transmission risk are largely unknown, we first evaluated if changes in T. dimidiata abundance had occurred after Hurricane Isidore in the Yucatán Peninsula. Overall, 80 of 170 (47%) of the houses were infested by triatomines before the hurricane, and 92 of 170 (54%) after the hurricane (P = 0.19). House infestation rates were identical for new and previous participants (33 of 62, 53% versus 58 of 108, 54%), confirming that all participants were equally efficient at collecting bugs. A total of 740 bugs were collected from the 34 villages before the hurricane, and 931 after, but this overall increase in T. dimidiata abundance in the domiciles was not significant (Z = 0.72; P = 0.46). We then compared bug collections for each trimester (Figure 2), as we previously established that house infestation by T. dimidiata is markedly seasonal in the region.13 As in previous years, domestic bug abundance varied greatly during the year after the hurricane. However, we observed a significantly higher bug abundance during the months of January–March compared with the prehurricane level (Figure 2, Z = −2.02; P = 0.04). This higher abundance was due to a fourfold increase in adult bugs, while larval stages abundance remain unchanged, suggesting that these were newly arriving bugs. For the remainder of the year, total bug collections and the proportion of larval stages and adults showed little pre- and posthurricane differences.

We then examined in more detail pre- and posthurricane bug collections for the different villages (Figure 3). For the period of January–March, the increase in bug collection after the hurricane was not distributed evenly among the villages (Figure 3A). Indeed, there was no change in 8 villages, whereas in other villages there was an increase in bug collections that reached 10- to 15-fold of the previous collections for the same period. Bugs were also collected in villages in which no bugs had been collected before the hurricane. This higher house infestation suggested a possible increase in T. cruzi transmission risk 4–6 months after the hurricane.

Bug collections during April–June were practically identical before and after the hurricane (Figure 3B, Z = 0.40; P = 0.39), with the exception of 2 villages that showed major increases (Kikil and Suma). During July–September, T. dimidiata abundance appeared more variable, being higher in some villages, and lower in others, but these changes did not reach statistical significance (Figure 3C, Z = 0.42; P = 0.67). No significant changes were observed during October–December (Figure 3D, Z = 0.06; P = 0.95), possibly due to the very low number of bugs usually collected during this period.

We also compared T. cruzi infection rates in bugs collected before and after the hurricane in the villages in which bug abundance had changed the most. Because of the low number of bugs collected in some villages during some of the trimesters, data were pooled for the entire year. T. dimidiata infection rates by T. cruzi varied from 0 to 46%, depending on the village. Table 1 shows that there were no significant differences between pre- and posthurricane infection rates for T. cruzi for any of these villages nor for the total analyzed population (Fisher exact test, P = 0.48). After the hurricane, infection rates remained high in villages where they were previously high and low where they were previously low. Infection rates thus did not contribute to any changes in transmission risk.

We then focused only on bug abundance and looked in more detail at the geographic distribution of the changes in house infestation. We compared maps of T. dimidiata abundance for each trimester before and after the hurricane. Figures 4A–4D shows the normal seasonal variations of domestic bug abundance that were described previously, with an important but transient increase in house infestation by triatomines in the northern part of the Yucatán Peninsula during the months of April–June.13 Figures 3E–3H shows the seasonal geographic distribution of bugs during the year after the hurricane. The important house infestation observed in January–March 2003 seemed to have occurred mainly along the path of Hurricane Isidore, in areas where very few or no bugs were usually collected during this period (Figures 4A and 4E). Moran I statistics suggested a very weak but significant spatial clustering of the changes in T. dimidiata abundance during this period (I = 0.16, P = 0.01), and spatial scan analysis identified the villages located in close proximity with the path of the hurricane (Dzidzilche, Baca, Tetiz, Abala, Bolonchen, Dzibalchen, Catmis, and Presumida) as being part of a significant spatial cluster (P = 0.001). The highest increase in bug abundance was located exactly at the point of landfall of the hurricane on the coast of Yucatán.

During the period April–June, the geographic distribution of bug abundance was strikingly unchanged after the hurricane compared with its previous distribution (Figures 4B and 4F), suggesting that house infestation had mostly returned to normal levels 7–9 months after the hurricane. The small changes observed were randomly distributed (I = −0.09, P = 0.08). During the months of July–September, some changes in T. dimidiata distribution seemed to have occurred 10–12 months after the hurricane (Figures 4C and 4G), suggesting some secondary perturbation, but no spatial clustering of changes were observed (I = −0.03, P = 0.2). Finally, T. dimidiata distribution appeared unchanged in the months of October–December 2003, 13–15 months after the hurricane (I = 0.04, P = 0.13).

Relationships between changes in T. dimidiata abundance and the hurricane.

We then further evaluated the relationships between the changes in T. dimidiata house infestation observed during the months of January–March and July–September and the hurricane. Correlation and regression analysis indicated that the distance between the villages and the track of the hurricane explained a significant part of the changes observed during the months of January–March 2003 (r2 = 0.30; P = 0.007). Villages located within 75 km from the track of the hurricane presented increases in the abundance of T. dimidiata in the houses, compared with prehurricane collections, whereas villages located further away appeared unaffected (Figure 5A). This strengthened the interpretation that these changes were neither due to year-to-year variations nor to our sampling strategy but indeed related to the hurricane.

During the months of July–September, there was no significant relationship between the changes in bug abundance in the houses and the distance of the villages from the hurricane (data not shown, P = 0.42), possibly because changes did not occur in the same direction in all the villages, with an increase in some villages and a decrease in others. However, there was a significant relationship between the absolute changes, indicating overall perturbation, and the distance between the villages and the track of the hurricane (r2 = 0.31; P = 0.0008; Figure 5B). Taken together, these results confirmed that part of the changes in domestic T. dimidiata abundance observed during these two periods was directly associated with the hurricane.

We also evaluated if the observed changes in T. dimidiata abundance were associated with the increase in forest fires that occurred several months after the hurricane. The changes in T. dimidiata abundance observed during either period January–March or July–September were not associated with fire density around the villages (data not shown; P = 0.58 and P = 0.28, respectively). However, the absolute changes in T. dimidiata abundance during the months of July–September were significantly associated with the cumulative fire density during the preceding months; that is, January–July (r2 = 0.26; P = 0.003; Figure 5C). These results suggested that these fires might have contributed to the observed perturbations in T. dimidiata abundance during this period.

DISCUSSION

We present here a unique evaluation of the effects of a major hurricane on Chagas disease vectors and their implications for disease transmission risk in the Yucatán Peninsula. The main observation from this study is that the hurricane was associated with medium- and long-term perturbations in house infestation by T. dimidiata that need to be taken into account for a better prevention of Chagas disease transmission after such a disaster.

Some changes in infection rates could have been expected after the hurricane, due to the possible death of mammalian reservoirs hosts. Indeed, a decrease in rodents reservoirs (of Leishmania mexicana) has been attributed to a major hurricane in Belize in 1961.19 However, T. dimidiata infection by T. cruzi appeared unaffected by the hurricane, so that infection rates remained high in areas where they were previously high, and conversely low where they were previously low. This suggests that infection rates did not contribute to major changes in Chagas disease transmission risk.

On the other hand, the most important changes were observed in bug abundance within 6 months after the hurricane, with a major increase (up to 10- to 15-fold) in domestic bug numbers that was strongly associated with the distance of the villages from the track of the hurricane. Although this relationship clearly associates these changes with the hurricane, the causes leading to them remain unclear. We may speculate that reduced bloodmeal sources after the hurricane due to the death of wild animals15 led to an increased dispersal of the bugs seeking alternative feeding sources.

Most importantly, because transmission risk is directly proportional to the abundance of infected bugs in the houses,14,2022 the increase we observed translated into a comparable increase in transmission risk during this period of the year, which is normally associated with a relatively low risk of transmission due to the near absence of bugs.13 Because we were unable to gather entomological data in the 3 months immediately after the hurricane, we do not know exactly when the increase in transmission risk began, and we may thus be observing the end of a larger perturbation. On the other hand, there are no epidemiologic data available due to the lack of official reporting of cases, so it is not possible to evaluate if the increase in transmission risk we detected resulted in increased human cases. Nonetheless, in a previous work, we found that transmission risk defined by the abundance of infected bugs correlated well with reported human cases.14

After the initial impact of the hurricane on T. dimidiata populations, bug collections corresponding to the months of April–June resulted strikingly identical before and 7–9 months after the hurricane for most of the villages. These observations strongly suggest that T. dimidiata populations had fully recovered from the hurricane and were back to their normal equilibrium. This conclusion seems to agree with previous work indicating that even though forest damage may appear catastrophic, the ecosystem is only modestly affected and recovers very rapidly after disturbance.23 In addition, these data confirm the reliability and reproducibility of our bug collection method and suggest that there are little inter-annual variations in house infestation by T. dimidiata.

Nonetheless, we observed that after this apparent recovery; some secondary changes in T. dimidiata populations seemed to have occurred 10–12 months after the hurricane. The absolute changes, reflecting the overall perturbation, were significantly associated with the distance from the hurricane and fire density around the villages. It is however difficult to reach a clear interpretation, likely because of the very complex network of interactions involved. For example, forest fires appear to sometimes increase, sometimes decrease domestic bug abundance, possibly due to differences in the magnitude of the fires, a factor that was not taken into account. Other studies suggest that fires may actually have more severe ecological impacts than the hurricanes,23 but additional variables should be taken into account to assess them.

In conclusion, given the current historical trends in the Yucatán Peninsula suggesting an increase in hurricane strength and impact over the years,23 it is important to increase our understanding of the short- and long-term impact of hurricanes, so that the most appropriate measures may be established to reduce their effects on human health. Our observations of an important increase in Chagas disease transmission risk lasting for about 6 months after hurricane Isidore imply that vector control measures as well as specific epidemiologic surveillance should be applied in the months after a similar disaster, particularly along its path, to avoid this transient increase in Chagas disease transmission risk and possible outbreaks. Our results also provide a framework for further studies to evaluate in more detail the impact of hurricanes on Chagas disease and other vector-borne diseases.

Table 1

Triatoma dimidiata infection rates by Trypanosoma cruzi

VillageBeforeAfterP
Data are presented as number of T. cruzi positive bugs/number of analyzed bugs (%).
Dzidzilche19/41 (46%)4/10 (40%)0.50
Tetiz0/32 (0%)0/19 (0%)0.75
Suma de Hidalgo1/17 (6%)3/16 (19%)0.32
Chacksinkin3/14 (21%)4/10 (40%)0.28
Bolonchen2/6 (33%)2/4 (50%)0.59
Baca6/15 (40%)1/9 (11%)0.25
Abala0/2 (0%)0/3 (0%)0.50
Total31/127 (24%)14/71 (20%)0.48
Figure 1.
Figure 1.

Map of study area. The smaller upper map identifies Mexico. The main map represents the Yucatán Peninsula, with the states of Campeche, Yucatán, and Quintana Roo. Dots indicate the location of the villages from which field data were taken.

Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 73, 6; 10.4269/ajtmh.2005.73.1019

Figure 2.
Figure 2.

Collections of T. dimidiata before and after the hurricane Isidore. The total number of adults (A) and the respective larval stages (B) are shown per trimester for pre- (left bars) and posthurricane (right bars) collections. *Indicates a significant difference (P < 0.05).

Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 73, 6; 10.4269/ajtmh.2005.73.1019

Figure 3.
Figure 3.

Comparison of T. dimidiata collections before and after Hurricane Isidore. The total number of bugs collected in each village is shown for the indicated trimesters (A–D) for pre- and posthurricane collections. The diagonal dotted line indicates no changes in bug collections between the two periods. Only villages for which pre- and posthurricane data are available are shown. (A) N = 23 villages, (B and C) N = 34 villages, (D) N = 21 villages. Note the log scale.

Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 73, 6; 10.4269/ajtmh.2005.73.1019

Figure 4.
Figure 4.

Maps of T. dimidiata domestic abundance before and after Hurricane Isidore. The total number of bugs collected per village and for each trimester was interpolated for the entire Yucatán Peninsula and color-coded as indicated. Seasonal variations in bug abundance are shown for pre- (A–D) and posthurricane years (E–H). The track of Hurricane Isidore is superimposed on the maps corresponding to the year 2003 with the symbols indicating the position of the hurricane center at 6-hour intervals, starting on September 21st at 1800 UTC.

Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 73, 6; 10.4269/ajtmh.2005.73.1019

Figure 5.
Figure 5.

Relationships between the changes in bug collections and Hurricane Isidore. (A) Relationship between changes in bug collection before and after the hurricane during the months of January–March and the distance from the track of the hurricane (N = 23 villages). The line corresponds to the fitted equation: change in bug number = 16.9 – 3.1Log(distance from the hurricane). (B) Relationship between absolute changes in bug collection during the months of July–September and the distance from the track of the hurricane (N = 34 villages). The line corresponds to the fitted equation: |change in bug number|0.5 = 2.64 – 0.01(distance from the hurricane). (C) Relationship between absolute changes in bug collection during the months of July–September and the cumulative number of fires around the villages for the months of January–July (N = 34 villages). The line corresponds to the fitted equation: |change in bug number| = −0.81 + 0.01(number of fires).

Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 73, 6; 10.4269/ajtmh.2005.73.1019

*

Address correspondence to Eric Dumonteil, Laboratorio de Parasitología, Centro de Investigaciones Regionales “Dr. Hideyo Noguchi,” Universidad Autónoma de Yucatán, Ave. Itzaes #490 x 50, 97000, Mérida, Yucatán, Mexico. E-mail: oliver@tunku.uady.mx.

Authors’ addresses: Yadira Guzman-Tapia, Maria Jesus Ramirez-Sierra, Javier Escobedo-Ortegon, and Eric Dumonteil, Laboratorio de Parasitología, Centro de Investigaciones Regionales “Dr. Hideyo Noguchi,” Universidad Autónoma de Yucatán, Ave. Itzaes #490 x 50, 97000, Mérida, Yucatán, Mexico, Telephone: 52-999-924-5910, Fax: 52-999-923-6120, E-mail: oliver@tunku.uady.mx.

Acknowledgments: The authors thank all the families who participated in this study for their interest and commitment in spite of the adverse circumstances they faced. We also acknowledge the helpful discussions and comments from F. Menu, S. Gourbière, J. F. Cornu, and E. Rebollar-Tellez.

Financial support: The study was funded by grant no. 20020404 from SISIERRA/CONACYT to E. D. Y. G. T. received an M.Sc. scholarship from CONACYT.

REFERENCES

  • 1

    Committe on Climate, Infectious Disease and Human Health, 2001. Under the Weather: Climate, Ecosystems, and Infectious Disease. Washington DC: National Academy Press.

  • 2

    Sanders EJ, Rigau-Perez JG, Smits HL, Deseda CC, Vorndam VA, Aye T, Spiegel RA, Weyant RS, Bragg SL, 1999. Increase of leptospirosis in dengue-negative patients after a hurricane in Puerto Rico in 1996. Am J Trop Med Hyg 61 :399–404.

    • Search Google Scholar
    • Export Citation
  • 3

    Anonymous, 1999. Principales enfermedades infecciosas en Centroamérica duante 1998, antes y despés de Mitch. Rev Panam Salud Publica 6 :440–443.

    • Search Google Scholar
    • Export Citation
  • 4

    Nasci RS, Moore CG, 1998. Vector-borne disease surveillance and natural disasters. Emerg Infect Dis 4 :333–334.

  • 5

    Campanella N, 1999. Infectious diseases and natural disasters: the effects of Hurricane Mitch over Villanueva municipal area, Nicaragua. Public Health Rev 27 :311–319.

    • Search Google Scholar
    • Export Citation
  • 6

    Bissell RA, 1983. Delayed-impact infectious disease after a natural disaster. J Emerg Med 1 :59–66.

  • 7

    Shaman J, Stieglitz M, Stark C, Le Blancq S, Cane M, 2002. Using a dynamic hydrology model to predict mosquito abundances in flood and swamp water. Emerg Infect Dis 8 :6–13.

    • Search Google Scholar
    • Export Citation
  • 8

    Ramsey JM, Ordonez R, Cruz-Celis A, Alvear AL, Chavez V, Lopez R, Pintor JR, Gama F, Carrillo S, 2000. Distribution of domestic triatominae and stratification of Chagas disease transmission in Oaxaca, Mexico. Med Vet Entomol 14 :19–30.

    • Search Google Scholar
    • Export Citation
  • 9

    Zeledon R, Ugalde JA, Paniagua LA, 2001. Entomological and ecological aspects of six sylvatic species of triatomines (Hemiptera, Reduviidae) from the collection of the National Biodiversity Institute of Costa Rica, Central America. Mem Inst Oswaldo Cruz 96 :757–764.

    • Search Google Scholar
    • Export Citation
  • 10

    Galvao C, Jurberg J, Carcavallo RU, Segura CA, Galindez Giron I, Curto de Casas SI, 1998. Distribuição geográfica e dispersão alti-latitudinal de aluns gêneros e espécies da tribo Triatomini Jeannel, 1919 (Hemiptera, Reduviidae, Triatominae). Mem Inst Oswaldo Cruz 93 :33–37.

    • Search Google Scholar
    • Export Citation
  • 11

    Costa J, Peterson AT, Beard CB, 2002. Ecologic niche modeling and differentiation of populations of Triatoma brasiliensis neiva, 1911, the most important Chagas’ disease vector in northeastern Brazil (Hemiptera, Reduviidae, Triatominae). Am J Trop Med Hyg 67 :516–520.

    • Search Google Scholar
    • Export Citation
  • 12

    Gorla DE, 2002. Variables ambientales registradas por sensores remotos como indicadores de la distribución geográfica de Triatoma infestans. Ecología Austral 12 :117–127.

    • Search Google Scholar
    • Export Citation
  • 13

    Dumonteil E, Gourbiere S, Barrera-Perez M, Rodriguez-Felix E, Ruiz-Piña H, Baños-Lopez O, Ramirez-Sierra MJ, Menu F, Rabinovich JE, 2002. Geographic distribution of Triatoma dimidiata and transmission dynamics of Trypanosoma cruzi in the Yucatán peninsula of Mexico. Am J Trop Med Hyg 67 :176–183.

    • Search Google Scholar
    • Export Citation
  • 14

    Dumonteil E, Gourbiere S, 2004. Prediction of Triatoma dimidiata vector abundance and infection rate: a risk map for Trypanosoma cruzi natural transmission in the Yucatan peninsula of Mexico. Am J Trop Med Hyg 70 :514–519.

    • Search Google Scholar
    • Export Citation
  • 15

    Güemez-Pineda M, Quintal-Aviles EF, 2003. Repercusiones del huracán “Isidoro” en la población maya-yucateca. A un año del huracán: Repercusiones del huracán Isidoro en la población yucateca. Merida, Universidad Automa de Yucatán, Yucatán, Mexico.

  • 16

    Dorn PL, Engelke D, Rodas A, Rosales R, Melgar S, Brahney B, Flores J, Monroy C, 1999. Utility of the polymerase chain reaction in detection of Trypanosoma cruzi in Guatemalan Chagas’ disease vectors. Am J Trop Med Hyg 60 :740–745.

    • Search Google Scholar
    • Export Citation
  • 17

    Levine N, 2002. CrimeStat: a spatial statistics program for the analysis of crime incident locations (v. 2.0). Houston, TX, and Washington, DC: National Institute of Justice.

  • 18

    Kulldorff M, 1997. A spatial scan statistic. Comm Stat Theory Meth 26 :1481–1496.

  • 19

    Williams P, 1970. Phlebotomine sandflies and leishmaniasis in British Honduras (Belize). Trans R Soc Trop Med Hyg 64 :317–364.

  • 20

    Sgambatti de Andrade AL, Zicker F, Silva IG, Souza JM, Martelli CM, 1995. Risk factors for Trypanosoma cruzi infection among children in central Brazil: a case-control study in vector control settings. Am J Trop Med Hyg 52 :183–187.

    • Search Google Scholar
    • Export Citation
  • 21

    Catala S, Crocco LB, Morales GF, 1997. Trypanosoma cruzi transmission risk index (TcTRI): an entomological indicator of Chagas disease vectorial transmission to humans. Acta Trop 68 :285–295.

    • Search Google Scholar
    • Export Citation
  • 22

    Rabinovich JE, Wisnivesky-Colli C, Solarz ND, Gürtler RE, 1990. Probability of transmission of Chagas’ disease by Triatoma infestans (Hemiptera: Reduviidae) in an endemic area of Santiago del Estero, Argentina. Bull World Health Org 68 :737–746.

    • Search Google Scholar
    • Export Citation
  • 23

    Boose ER, Foster DR, Barker Plotkin A, Hall BS, 2003. Geographical and historical variation in hurricanes across the Yucatán peninsula. Gómez-Pompa A, Allen MF, Fedick SL, Jiménez-Osornio JJ, eds. The lowland Maya Area: Three Millennia at the Human-Wildland Interface. New York: The Haworth Press Inc., 495–516.

Author Notes

Reprint requests: Eric Dumonteil, Laboratorio de Parasitología, Centro de Investigaciones Regionales “Dr. Hideyo Noguchi,” Universidad Autónoma de Yucatán, Ave. Itzaes #490 x 50, 97000, Mérida, Yucatán, Mexico, Telephone: 52-999-924-5910, Fax: 52-999-923-6120, E-mail: oliver@tunku.uady.mx.
Save