|
|
||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
In the fall of 2005, Texas health officials reported a 31-year old woman from Cameron County, in southern Texas, with autochthonously acquired DHF.8 This was only the second case of locally acquired DHF ever reported in Texas and the first in a native of the Texas–Mexico border region.9 By November 2005, two additional autochthonous cases of dengue fever were reported in Cameron County, along with 22 persons with dengue fever who had traveled to Mexico (J. Schuermann and others, unpublished data). This was the largest outbreak of dengue reported among Texas residents since 1999.
The 2005 South Texas outbreak was linked to a dengue epidemic just south of the Texas–Mexico border, in the Mexican state of Tamaulipas, which has a population of 3.2 million (Figure 1
). In 2005, 7,062 dengue cases were reported in Tamaulipas, including 1,832 (26%) cases classified as DHF. DENV-2 was the predominant serotype in Tamaulipas (i.e., 27 of 28 viral isolates were DENV-2, 1 was DENV-1; Panorama Epidemiológico del Dengue y Dengue Hemorrhágico, http://www.cenave.gob.mx/dengue/panorama/Panoramasemana52.pdf) The DENV-2 virus was a southeast Asia strain previously associated with DHF in the Americas.4,10
|
| MATERIALS AND METHODS |
|
|
|---|
Binational teams, each consisting of a nurse or physician, an entomologist, and a bilingual interviewer, conducted the survey. At each participating household, all residents present who were at least 5 years of age were asked to provide a blood sample and to respond to a short questionnaire, which included information on sociodemographics, travel history, and mosquito avoidance practices. A questionnaire for the head of household solicited general household information including additional sociodemographic data, the presence of air conditioning, whether windows and doors had screens, and knowledge of dengue prevention. Lot size and distance to next residence were recorded.
Entomology. Yards and patios of participating residences were inspected for evidence of Aedes aegypti and Aedes albopictus larval and pupal habitats. Larvae and pupae were collected and sent to the Texas Department of State Health Services for speciation. We recorded the types of artificial containers (tires, buckets, etc.) that were found infested with mosquito larvae and pupae. We calculated the Breteau index (the number of infested containers per 100 residences inspected) for each city.
Laboratory.
Serum samples were tested for IgM antibodies to dengue using an IgM antibody-capture enzyme-linked immunosorbent assay (MAC-ELISA).12 A quantitative IgG ELISA was performed to detect IgG antibodies.13 Serum samples that tested positive for IgM antibodies were tested by a duplex microsphere-based immunologic assay to rule out cross-reactivity with IgM antibodies to other flaviviruses: West Nile virus (WNV) and St. Louis encephalitis virus (SLEV).14 ELISA format microneutralization testing was performed on samples that tested positive for IgG antibody titers
1:2,560 to test for neutralizing IgG antibodies to DENV serotypes and to WNV and SLE.15 Those with IgG titers > 1:2,560 were excluded from microneutralization testing because of greater potential for cross-reactivity among the flaviviruses.
Serologic evidence of recent dengue infection was defined by IgM antibodies
0.2 optical density (OD) or the presence of high-titered IgG antibodies > 1:40,960, consistent with a recent secondary infection13,16 Evidence of any past dengue infection either recent or remote, was defined by the presence of IgM antibodies to dengue (
0.2 OD) or IgG antibodies
1:40.12,13
Analysis.
Data were weighted to reflect probability of selection, taking into account the population and number of households per census track and size of household. Analysis was performed using SAS v.9.1 (SAS Institute, Cary, NC) software. Frequencies and crude and adjusted odds ratios (ORs) and their 95% confidence intervals (CIs) were calculated. Variables found to be associated with dengue infection on univariate analysis (P
0.10), as well as covariates age, sex, and socioeconomic status (SES) as measured by census tract income data from the 2000 US and Mexico census, were included in the multivariate logistic regression models.
Ethical review. The investigation protocol was reviewed by the Human Subjects Coordinator at the National Center for Infectious Diseases, Centers for Disease Control and Prevention, and by the Centro Nacional de Vigilancia Epidemiológica y Control de Enfermedades of Mexico and determined to be an outbreak study and public health response that did not require further human subjects review. Blood samples (~4 mL) were collected from consenting participants. Informed consent of a parent or guardian was obtained before taking blood from a minor.
| RESULTS |
|
|
|---|
One third (29%) of homes surveyed in Matamoros had air conditioning; in contrast, most (85%) homes surveyed in Brownsville had air conditioning (P < 0.05). Nearly two thirds of homes (65% in Matamoros and 61% in Brownsville) had screens on windows and doors, per the residents report. The mean lot size was smaller in Matamoros (307 m2) than in Brownsville (1070 m2; P < 0.05), reflecting its more dense urban environment. Forty-six percent of participants from each city reported having crossed the US–Mexico border in the preceding 3 months. Few Matamoros residents (13%) or Brownsville residents (21%) reported using insect repellent "often" or "always" when outdoors in the 3 months before the survey.
Entomology.
During the residential surveys in Matamoros, we found larvae and pupae of Ae. aegypti mosquitoes; we did not find Ae. albopictus. In contrast, both Ae. aegypti and Ae. albopictus immature forms were collected in Brownsville. The Breteau index was 28 in Matamoros (Ae. aegypti only) versus 11 (Ae. aegypti only) and 16 (combined Aedes species) in Brownsville. Tires represented one third (32%) of the infested containers seen in Matamoros and a quarter (26%) of the infested containers identified in Brownsville, making up the largest single category of infested container in either site (Figure 2
). Buckets were the second leading category of infested container for both cities.
|
|
|
|
Autochthonous dengue—Brownsville. Twenty-four Brownsville survey participants were born in the United States and reported never having traveled outside the United States. Of these 24 people, 6 (25%; 95% CI, 12–45%) were seropositive for antibodies to dengue, indicating past infection. One person had IgM antibodies only, four had IgG antibodies only, and one person had both IgG and IgM antibodies. Their ages ranged from 24 to 85 years and all had lived in Brownsville for at least the past 20 years. Seropositive and seronegative residents had similar durations of residence in Brownsville (mean, 40 versus 31 years, P > 0.2).
| DISCUSSION |
|
|
|---|
Previous investigators have proposed that various infrastructure factors, such as the widespread use of air conditioning, limit dengue transmission in Texas by reducing human contact with disease vectors.17 Our data are consistent with this hypothesis. In Matamoros, where dengue incidence and seroprevalence were both significantly higher in Matamoros than in Brownsville, residential lot sizes are smaller and air conditioning was less available than in Brownsville. Both small lot size and lack of air conditioning could potentially increase human contact with the peridomestic Aedes mosquito. Furthermore, we found that lack of air conditioning was a risk factor for past dengue infection among Matamoros residents, and smaller lot size was a risk factor for recent dengue infection among Brownsville residents.
Our survey findings also indicated that the 2005 dengue epidemic infected many more South Texas residents than the few that were reported to Texas health officials. Although only a small proportion of the population was affected, in absolute terms when extrapolated to the Brownsville population (167,493), 4% (95% CI, 1–6%) represents from 1,675 to 10,050 residents with recent dengue infections. The 39% (95% CI, 28–50%) seroprevalence of antibodies to dengue that we observed among Brownsville residents is consistent with findings from a 2004 Brownsville serosurvey.18 These data indicate that, despite living conditions at home that include the availability of air conditioning, substantial numbers of Brownsville residents acquired dengue infection. Many residents likely acquired infection while across the border in Mexico, where dengue is endemic. Almost one half of Brownsville residents surveyed reported crossing the Texas–Mexico border in the 3 months preceding the survey. However, evidence of autochthonous transmission also indicates that dengue may now be endemic in Brownsville. Therefore, infections could have been acquired in the United States.
The current epidemiology of dengue and DHF in south Texas reflects changes that have already occurred in Mexico.19–21 Since 1995, when all four dengue virus serotypes were first identified in Mexico and in the border state of Tamaulipas, an increasing percentage of reported dengue cases in both Mexico and Tamaulipas have been DHF.20,22 The percentage of DHF cases in Tamaulipas increased from 3.7% between 2000 and 2004 to 25.9% in 2005 and 23.4% in 2006 (Boletín Epidemiolgía, http://www.dgepi.salud.gob/mx/boletin). All four serotypes have circulated locally in South Texas,23–25 and locally (US) acquired DHF has been recently reported.8,9 Furthermore, additional case finding efforts in late 2005 in Cameron County, TX, suggest that more DHF cases occurred in 2005 in south Texas than initially recognized.8
Our finding that source reduction awareness was associated with dengue incidence in Matamoros was unexpected. We had hypothesized that knowledge of source reduction as an important component in the control of dengue would be associated with lower dengue infection risk. Our finding of a positive association between prevention knowledge and dengue infection could reflect increased awareness after infection or after educational campaigns in high-risk areas. Our data suggest the possibility of prevention messages being communicated, but not in a timely enough manner to prevent dengue infections. Also, knowledge of source reduction as important to control dengue does not necessarily mean preventive actions were taken. Our cross-sectional study design makes these results difficult to interpret: we do not know if awareness of source reduction to prevent dengue preceded infection or vice versa.
Lack of air conditioning and not using insect repellent were both found to be associated with dengue infection in Matamoros. This protective effect of air conditioning is consistent with a 1999 study in the Laredo/Nuevo Laredo border area.17 We found that insect repellents were used "often or always" by 13% of Matamoros participants and had a protective effect against dengue infection. These data suggest an unexpected prevention opportunity, because earlier focus group data from border regions had suggested that repellent use would be very limited in Mexico (E. Zielinski-Gutierrez, unpublished data).
Discarded waste tires made up the single largest category of infested containers found in either city. In the subtropical climate of the region, the water holding capacity of tires, their insulating qualities that protect against weather extremes, and their dark color make tires an ideal oviposition habitat for Aedes mosquitoes.
The only factor found to be associated with recent dengue infection in Brownsville was smaller lot size. This finding could reflect denser concentration of people or could be related to socioeconomic factors.18,26 Past dengue infection among Brownsville residents was associated with birth outside the United States, as expected given the high level of dengue transmission in Mexico.
Limitations. Short winter days limited available daylight hours in which to do surveying. Low temperatures may have hampered our ability to fully detect all mosquito larvae and pupae on premises. The mosquito populations detected during December may not have reflected those during the peak dengue transmission period. Small sample size limited our ability to detect statistically significant associations. Last, there may be recall bias in that data were collected retrospectively.
Conclusions. Dengue and, more recently, DHF are growing public health problems in the Texas–Mexico border area. The entomologic, serologic, and virologic conditions in South Texas could support the continued development of locally (US) acquired DHF. Abundant Aedes species mosquitoes were detected in December on both sides of the subtropical Texas–Mexico border. Because nearly 40% of Brownsville residents have been infected with DENV, a substantial proportion of the population is likely to be at increased risk for DHF should they acquire a second DENV infection, especially one with a virulent strain.27
We recommend strengthening dengue surveillance in the region to include more virologic testing. Actively monitoring for circulating DENV serotypes may provide early warning of outbreaks. Clinicians practicing in South Texas and the general public should be aware of the possibility of DHF and dengue fever in the region. Clinicians should be trained in recognizing and managing DHF. Last, prevention measures including repellent use and source reduction efforts involving waste tire disposal and proper storage of buckets should be encouraged by public health officials along both sides of the border.
Received October 16, 2007. Accepted for publication December 18, 2007.
Acknowledgments: The authors thank Carlos A. Carrillo, Oscar Velasquez-Monroy, CENAVECE, Secretaria de Salud de Tamaulipas, Jurisdiccion Sanitaria No. III, Matamoros, Cameron County Department of Health and Human Services, City of Brownsville Health Department, Texas Department of State Health Services, Jim Schuermann, Chester Moore, Roberto Barrera, Joshua Smith, Carmen Perez, Elizabeth Hunsperger, and Nadonnia Jones.
* Address correspondence to Mary M. Ramos, Department of Pediatrics, University of New Mexico, 300 San Mateo Boulevard, NE, Suite 902, Albuquerque, NM 87108. E-mail: mramos{at}salud.unm.edu ![]()
Other members of The Dengue Serosurvey Working Group include Carlos Moya-Rabelly (Mexico Section of the US–Mexico Border Health Commission), Carlos Álvarez-Lucas and Cuahtemoc Mancha (Centro Nacional de Vigilancia Epidemiológica y Control de Enfermedades), Luis Fernando Garza Frausto, Ernesto Lavin Hernandez, and Norma Alicia Villarreal Reyes (Servicios de Salud de Tamaulipas), Victor Garcia Fuentes and Oscar Ramirez Contreras (Jurisdicción Sanitaria No. III de Matamoros), Joshua Ramirez (City of Brownsville Public Health Department), Mark Beatty (Pediatric Dengue Vaccine Initiative), Rafael Moreno-Sanchez (University of Colorado), Iris Sosa, Sophie Wenzel, Brad Biggerstaff, and Miguel Escobedo (Centers for Disease Control and Prevention). ![]()
Authors addresses: Mary M. Ramos, Department of Pediatrics, University of New Mexico, 300 San Mateo NE, Suite 902, Albuquerque, NM 87108, Tel: 505-222-8684, Fax: 505-222-8675, E-mail: mramos{at}salud.unm.edu. Hamish Mohammed and Jorge L. Muñoz, Dengue Branch, Centers for Disease Control and Prevention, 1324 Calle Cañada, San Juan, Puerto Rico 00920-3860, Tel: (787) 706-2399, Fax: (787) 706-2496, E-mails: HMohammed{at}cdc.gov and ckq2{at}cdc.gov. Emily Zielinski-Gutierrez, Division of Vector Borne Infectious Diseases, Centers for Disease Control and Prevention, 3150 Rampart Road, Foothills Campus, Fort Collins, CO 80521, Tel: 970-221-6477, Fax: 970-266-3502, E-mail: ebz0{at}cdc.gov. Mary H. Hayden, National Center for Atmospheric Research, PO Box 3000, Boulder, CO 80309, Tel: 303-497-8116, Fax: 303-497-8125, E-mail: mhayden{at}ucar.edu. Jose Luis Robles Lopez, Francisco Sarabia 153 Col. Mexico Agrario C.P. 87440, H. Matamoros Tamaulipas Mexico, Tel: (868) 8192565 and (868) 8140280, Fax: (868) 8174930, E-mails: jlrobles{at}salud.gob.mx and drjlrobles{at}hotmail.com. Marta Fournier and Brian Smith, Tx DSHS, Health Service Region 11, 601 W. Sesame, Harlingen, TX 78550, Tel: (956) 444-3227, (956) 444-3202, Fax: (956) 444-3299, E-mails: Marta.Fournier{at}dshs.state.tx.us and Brian.Smith{at}dshs.state.tx.us. Alfredo Rodríguez Trujillo, Palacio Federal 3r. Piso, Servicios de Salud de Tamaulipas, Cd. Victoria, Tamaulipas, Tel: (011-52-834) 315-68-83, Fax: (011-52-834) 315-68-83, E-mail: alfredordz{at}salud.gob.mx. Roy Burton, TX DSHS, PSQA Environmental Health, 8407 Wall St., Austin, TX 78754, Tel: (512) 834-6773, x2302, Fax: (512) 834-6706, E-mail: Roy.Burton{at}dshs.state.tx.us. Joan M. Brunkard, Epidemic Intelligence Service Program, Centers for Disease Control and Prevention, 3101 West Napoleon Ave., Metairie, LA 70001, Tel: 504-219-4732, Cell: 504-875-8584, E-mail: jbrunkard{at}cdc.gov. Luis Anaya-Lopez, Director de Servicios y Apoyo Técnico, CENAVECE/InDRE, Carpio No. 470, Col. Sto Tomás, Deleg. Miguel Hidalgo, Mex. D.F., 11340, Tel: (011-52-55) 5396-4986, 5342 7550 X.204 o 303, Fax: (011-52-55) 341 32 64, E-mail: anayaluis{at}hotmail.com. Allison Abell Banicki, Office of Border Health, Texas Dept. of State Health Services, 1100 W. 49th St., Austin, TX 78756, Tel: 512-458-7111 ext 6705, E-mail: allison.abell{at}dshs.state.tx.us. Pablo Kuri Morales, Francisco de P. Miranda No. 177, 4° Piso, Colonia Lomas de Plateros, 01480 México, Distrito Federal (DF), Tel: (011-52-55) 5337-1664 al 66, Fax: (011-52-55) 5337 1667, E-mail: pkuri{at}dgepi.salud.gob.mx. Stephen H. Waterman, San Diego Quarantine and Border Health Services, Division of Global Migration and Quarantine, Centers for Disease Control and Prevention, 3851 Rosecrans Street, PO Box 85524, San Diego, CA 92138, Tel: (619) 692 5659, E-mail: shw2{at}cdc.gov.
| REFERENCES |
|
|
|---|
This article has been cited by other articles:
![]() |
G. G. Clark Dengue and Dengue Hemorrhagic Fever in Northern Mexico and South Texas: Do They Really Respect the Border? Am J Trop Med Hyg, March 1, 2008; 78(3): 361 - 362. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |