Heartworm disease in dogs is caused by Dirofilaria immitis, which is transmitted by mosquitoes.1,2 Cases of heartworm infection have been reported from every state in the United States since at least 2011, with high densities in the Mississippi Delta region and southeastern United States and > 136,000 new cases reported from across the United States in 2017.3 Although canids, particularly dogs and coyotes, are the primary, definitive hosts, D immitis can infect other animals, such as domestic cats, which can develop serious complications from heartworm-associated respiratory disease in response to as few as 1 to 3 adult D immitis worms.
Despite the wide availability of heartworm preventives, heartworm infection remains a major veterinary concern. In addition, evidence that some D immitis strains have become resistant to macrocyclic lactones has necessitated discontinuation of the so-called slow-kill method for treating micro-filaremia in dogs, which involved monthly administration of heartworm preventive.4–6
Other filarial diseases (eg, onchocerciasis and lymphatic filariasis) have been controlled well through integrated programs that use a combination of mitigation strategies (eg, mass treatment of at-risk human populations, vector management, and reduction of vector-host contact) to break the cycle of disease transmission.7,8 However, D immitis is transmitted by a broad range of mosquito species (≥ 17 species in Florida alone and ≥ 25 species in the United States as a whole) and can cycle between several species of definitive hosts. This broad diversity of potential vectors suggests that all mammal-biting mosquitoes could be candidate vectors.9–11 Because of differences in vector feeding patterns, breeding habitats, and geographic ranges,9 the key vectors of D immitis can vary from region to region along with local host and ecological conditions.
Various biological and sociobehavioral components of the risk that dogs will acquire heartworm infection have been proposed, including economic status of owners,6 proximity to vector habitat,12,13 emerging vector mosquito species combined with drug resistance among D immitis strains,9,14 and climatic factors.15,16 However, to the authors' knowledge, no comprehensive study of these relationships has been conducted in communities where heartworm infection is endemic. The purpose of the study reported here was to assess knowledge, attitudes, and practices regarding D immitis and mosquito vectors among residents (dog owners and non-dog owners) in 2 Florida communities—1 with a high prevalence of dogs with heartworm infection and 1 with a low prevalence—and to perform entomological surveys of mosquito species in these neighborhoods and identify mosquito species infected with D immitis.
Materials and Methods
The 2 study sites were selected according to county-level data on the prevalence of dogs with heartworm infection as determined by the Companion Animal Parasite Council from tests performed by Idexx Laboratories and Antech Diagnostics by Florida veterinary clinics in 2012.3 A neighborhood in Lake City, Fla, and a neighborhood in St Augustine South, Fla, were selected. Lake City is in Columbia County, which reportedly had a high prevalence (9% [1/11]) of dogs with heartworm infection in 2012; St Augustine South is in St John's County, which reportedly had a low prevalence (1.30% [195/15,024]) of dogs with heartworm infection in 2012. Both communities had active mosquito control programs at the time of the study and were chosen because of the clustering of residences within walking distance of each other, which allowed for efficient door-to-door progression of questionnaire administration. County mosquito control districts were notified of our study dates, affiliations, and intention to request residents' participation in questionnaires and larval mosquito surveys of their property.
Questionnaire and survey strategy
Methods of Tuiten et al17 were generally followed for surveys of community residents and larval mosquitos. A 32-item questionnaire was developed to gather information on residents' knowledge, attitudes, and practices regarding heartworms (Supplementary Appendix SI, available at avmajournals.avma.org/doi/suppl/10.2460/javma.254.1.93). This questionnaire was submitted to the Cornell University Institutional Review Board and deemed exempt from institutional oversight requirements.
The questionnaire was orally administered to participants by trained interviewers from June 19, 2013, to July 24, 2013. Residences in each neighborhood were included in the study by convenience sampling as research teams (2 people/team) progressed door-to-door. An effort was made to include several places throughout each neighborhood for even coverage. Because of time and personnel constraints, the aim was to include approximately 50 residences in each community.
Questionnaires were administered only to consenting adults, and participants were assured of the confidentiality of their responses. If no consenting adult was present at the initial visit, the team returned for another attempt.
Larval and adult mosquito collections
After the questionnaire was administered, permission was obtained to perform a larval mosquito survey of each respondent's property. Containers of standing water were counted, diameter and height were measured, and the number of larvae, if present, was estimated and recorded. Ground depressions, water pooling in plant leaves (phytotelmata), and standing pools of water were included. Subsamples of larvae and pupae (when possible) were collected with an aquarium net, turkey baster, or pipette and placed in labeled 50-mL conical tubes. Larvae were transported back to the laboratory and examined under a dissecting microscope to confirm species. Pupae were held until adult eclosion and then identified.
Adult mosquitoes were collected from July 1, 2013, to July 26, 2013, in each community from surrounding wooded areas and residences of consenting survey participants. Collection methods targeted resting and host-seeking female mosquitoes. Adult mosquito collections were conducted with vegetation aspirators,a Aedes-specific traps,b CO2-baited traps,c and resting boxes.18 Half (4/8) of the traps in Lake City and three-quarters (6/8) of traps in St Augustine South were placed on residents' properties. Other traps were placed in natural areas or parks.
Vegetation aspirators fitted with 3.8-L paint strainer bagsd were used to capture resting mosquitoes. Collection time and duration were recorded for each vegetation aspirator collection for calculation of yield per collection hour. Collection bags and containers from all 4 collection methods were stored in a cooler on ice for transportation back to the laboratory.
Mosquito processing and identification
Female mosquitoes were identified to species following published keys.10,19 A small number of mosquitoes too damaged for morphological identification were discarded and not counted. Blood-fed mosquitoes were processed individually rather than pooled for screening tests, and their abdomens were included in heartworm screening. Nonfed mosquitoes identified from the same collection were pooled by species and placed in aliquots of up to 20 mosquitoes/tube. All mosquitoes were stored at −20°C prior to DNA extraction.
Molecular detection of infective-stage D immitis
DNA extraction—All DNA extractions were performed by means of a modified DNA extraction protocol.e Mosquitoes were dissected, and heads and thoraces were pooled for heartworm screening apart from 1 group (n = 127 pools) extracted as whole mosquitoes. Each mosquito pool was homogenized in cell lysis solutione and incubated at 65°C for 15 minutes. Lysate was chilled and protein precipitation solution was added. After centrifugation, supernatant containing DNA was transferred to 100% isopropanol to produce a final solution of 70% isopropanol. Samples were centrifuged, supernatant was decanted, and the DNA pellet was resuspended in fresh 70% ethanol solution. Samples were centrifuged and ethanol solution was poured off before storage at −20°C.
Screening mosquito pools for D immitis—Briefly, DNA was extracted as described previously from pools of no greater than 20 mosquito heads and thoraces. We screened mosquitoes for D immitis using primers specific to a 203-base pair region of D immitis cytochrome oxidase I (forward primer, 5′-AGT GTA GAG GGT CAG CCT GAG TTA-3′; reverse primer, 5′-ACA GGC ACT GAC AAT ACC AAT-3′20). Amplification was performed with a modified protocol.20 Products of the PCR assay were separated on an ethidium bromide-stained 1% agarose gel before being visualized and digitally photographed via a digital imaging system.f
In-house validation of this method consistently detected single L3 stage D immitis larvae in pools of heads and thoraces of up to 20 laboratory-reared, heartworm-free mosquitoes. Each screening group was run alongside negative controls of heartworm-free pools of mosquito heads and thoraces, positive controls of 6 L3 stage D immitis larvae, and 20 mosquito heads and thoraces spiked with 1 L3. The 127 pools that were extracted as whole bodies were run alongside L3 stage larvae-only and L3-spiked mosquito whole-body positive controls.
Samples producing a band in the 203-base pair region were prepared with a PCR product cleanup agentg before submission for Sanger sequence confirmation of amplification along with positive and negative controls. Sample sequence results were compared with the reference D immitis cytochrome oxidase I sequences in a bioinformatic search tool.h
Statistical analysis
Resident responses were coded, removing personally identifiable information to ensure confidentiality. Questionnaire responses on knowledge, attitudes, and practices regarding heartworms were tested for independence from city, highest education, and dog ownership by χ2 analysis of proportions with statistical software.i Adult mosquito collection data were standardized by collection hours and trap nights before inclusion in models. Minimum infection rates/1,000 mosquitoes were calculated for D immitis-positive mosquito pools with computational software.21,j
Results
A total of 96 questionnaires were administered (St Augustine South, n = 46; Lake City, 50). Response totals < 96 reflect missing data, no answer for a question, or questions that were not applicable to all respondents. For response categories that were not mutually exclusive, sample sizes represent counts of answer selections. For some questions of knowledge or attitudes, non-pet owners' answers reflect their decisions if they were to hypothetically own a dog.
Demographics and dog ownership
Despite substantial differences between communities in median income (US Census Bureau records from 2012 indicated that St Augustine South residents' median income [$61,288] was about twice that of Lake City residents [$36,542]), the highest level of education for respondents from St Augustine South and Lake City was high school or lower for 22% (10/45) and 41% (20/49), respectively, and some college or more for 78% (35/45) and 59% (29/49), respectively (P = 0.053). Most respondents in both communities owned pets; 53% (51/96) owned at least 1 dog, and 30% (29/96) owned at least 1 cat (Figure 1), and proportions of pet owners did not differ significantly (P = 0.053) between communities. Education level also did not differ significantly (P = 0.33) between dog owners and non-dog owners.
Pet ownership status among Florida residents of St Augustine South (in a county with low prevalence of heartworm infection; gray bars; n = 46) and Lake City (in a county with a high prevalence of heartworm infection; black bars; 50) who participated in a survey in 2013.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Pet ownership status among Florida residents of St Augustine South (in a county with low prevalence of heartworm infection; gray bars; n = 46) and Lake City (in a county with a high prevalence of heartworm infection; black bars; 50) who participated in a survey in 2013.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Pet ownership status among Florida residents of St Augustine South (in a county with low prevalence of heartworm infection; gray bars; n = 46) and Lake City (in a county with a high prevalence of heartworm infection; black bars; 50) who participated in a survey in 2013.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Knowledge and concern about D immitis and other vector-borne infections
Residents who completed at least some college were more likely to know that dogs are at risk of heartworm infection than residents whose education did not extend beyond high school (P = 0.045). Overall, 41% (39/96) of residents knew that mosquitoes transmit heartworms. Dog owners were more likely than non-dog owners to know that mosquitoes transmit heartworms and that standing water is an important breeding site for mosquitoes (Table 1). Dog owners were also more likely to perceive high mosquito activity levels on their property. Non-dog owners were more likely to know that mosquitoes are active at dawn. Lake City residents were more likely than St Augustine South residents to be aware of diurnal mosquito activity, and St Augustine South residents were more likely to perceive a high mosquito activity level on their property (Table 2).
Comparison of the number (%) of dog owners and non-dog owners with various answers to questions about their knowledge, attitudes, and perceptions of heartworms in a survey of Florida residents of St Augustine South (in a county with a low prevalence of heartworm infection) and Lake City (in a county with a high prevalence of heartworm infection) in 2013.
| Question topic | Non-dog owners | Dog owners | P value |
|---|---|---|---|
| Level of concern about heartworms | |||
| None | 13 (29) | 4 (8) | 0.007 |
| Low | 1 (2) | 3 (6) | 0.37 |
| Moderate | 5 (11) | 6 (12) | 0.92 |
| High | 24 (53) | 38 (75) | 0.03 |
| Don't know or can't rate | 2 (4) | 0 (0) | 0.13 |
| Knowledge of dogs as heartworm hosts | |||
| Dogs are not hosts | 9 (20) | 9 (18) | 0.77 |
| Dogs are hosts | 30 (67) | 40 (78) | 0.20 |
| Don't know | 6 (13) | 2 (4) | 0.10 |
| Mode of heartworm transmission | |||
| Not transmitted by mosquitoes | 37 (82) | 20 (39) | < 0.001 |
| Mosquito transmission | 8 (18) | 31 (61) | < 0.001 |
| Estimated cost of heartworm treatment | |||
| < $1,000 | 21 (64) | 18 (35) | 0.01 |
| $1,000–2,000 | 2 (6) | 15 (29) | 0.009 |
| > $2,000 | 1 (3) | 3 (6) | 0.55 |
| Other or don't know | 9 (27) | 15 (29) | 0.83 |
| Highest cost willing to pay for heartworm treatment | |||
| < $100 or not willing | 4 (12) | 0 (0) | 0.01 |
| Up to $100 | 8 (24) | 9 (18) | 0.46 |
| Up to $500 | 13 (39) | 15 (29) | 0.34 |
| Up to $1,000 | 2 (6) | 5 (10) | 0.54 |
| > $1,500 | 6 (18) | 22 (43) | 0.02 |
| Perceived level of mosquito activity near home | |||
| None | 2 (5) | 1 (2) | 0.46 |
| Low | 13 (30) | 16 (31) | 0.91 |
| Medium | 19 (44) | 13 (25) | 0.06 |
| High | 9 (21) | 21 (41) | 0.04 |
| Peak time of mosquito activity | |||
| Dawn | 33 (73) | 27 (53) | 0.04 |
| Day | 3 (7) | 3 (6) | 0.87 |
| Dusk | 35 (78) | 42 (82) | 0.57 |
| Night | 10 (22) | 12 (24) | 0.88 |
| Mosquito breeding sites | |||
| Standing water | 30 (67) | 46 (90) | 0.005 |
| Standing water not mentioned | 15 (33) | 5 (10) | — |
— = Not applicable.
Values of P < 0.05 were considered significant.
Comparison of the number (%) of respondents from St Augustine South and Lake City, Fla, with various answers to questions about their knowledge, attitudes, and perceptions of heartworms in the survey of Table 1.
| Question topic | St Augustine South | Lake City | P value |
|---|---|---|---|
| Species susceptible to heartworm infection | |||
| Dogs | 34 (57) | 36 (58) | 0.91 |
| Cats | 23 (38) | 21 (34) | 0.52 |
| Don't know | 3 (5) | 5 (8) | — |
| Mode of heartworm transmission | |||
| Mosquitoes | 23 (50) | 16 (32) | 0.07 |
| Mosquitoes not mentioned | 23 (50) | 34 (68) | |
| Seriousness of heartworm infection | |||
| Moderately serious | 1 (2) | 1 (2) | 0.96 |
| Highly serious | 39 (95) | 42 (98) | 0.96 |
| Don't know | 1 (2) | 0 (0) | |
| Peak time of mosquito activity | |||
| Dawn | 18 (39) | 18 (36) | 0.75 |
| Day | 0 (0) | 6 (12) | 0.02 |
| Dusk | 38 (83) | 39 (78) | 0.57 |
| Night | 8 (17) | 14 (28) | 0.22 |
| Mosquito breeding sites | |||
| Standing water | 39 (85) | 37 (74) | 0.19 |
| Standing water not mentioned | 7 (15) | 13 (26) | — |
| Frequency that respondent eliminates standing water | |||
| Once a week or more | 28 (62) | 23 (47) | 0.14 |
| Never | 11 (24) | 16 (33) | 0.38 |
| Once a month | 4 (9) | 2 (4) | 0.34 |
| Don't know | 2 (4) | 8 (16) | 0.06 |
| Highest cost willing to pay for heartworm treatment | |||
| < $l00 or not willing | 2 (5) | 2 (5) | 0.96 |
| Up to $100 | 8 (20) | 9 (21) | 0.87 |
| Up to $500 | 13 (32) | 15 (35) | 0.76 |
| Up to $l,000 | 4 (10) | 3 (7) | 0.64 |
| > $1,500 | 14 (34) | 14 (33) | 0.88 |
| Perceived level of mosquito activity near home | |||
| None | 1 (2) | 2 (4) | 0.61 |
| Low | 13 (29) | 16 (33) | 0.69 |
| Medium | 11 (24) | 21 (43) | 0.06 |
| High | 20 (44) | 10 (20) | 0.01 |
| Frequency of veterinary visits for dog owners | |||
| Once a year | 11 (41) | 10 (43) | 0.70 |
| More than once a year | 12 (44) | 10 (43) | 0.50 |
| Only when necessary or less than once a year | 3 (11) | 3 (13) | 0.87 |
| Don't know | 1 (4) | 0 (0) | — |
| Frequency of veterinary visits for cat owners | |||
| Once a year | 7 (47) | 4 (29) | 0.32 |
| More than once a year | 4 (27) | 3 (21) | 0.74 |
| Only when necessary or less than once a year | 4 (27) | 4 (29) | 0.91 |
| Never | 0 (0) | 3 (21) | 0.06 |
| Veterinarian has informed respondent about heartworms | |||
| Yes | 30 (91) | 26 (79) | 0.17 |
| No | 3 (9) | 7 (21) | — |
| Heartworm preventive administration | |||
| All year | 26 (96) | 15 (79) | 0.06 |
| Less than all year | 1 (34) | 4 (21) | — |
| Pet's heartworm preventive status | |||
| Not on preventive treatment | 8 (23) | 11 (35) | 0.26 |
| On preventive treatment | 27 (77) | 20 (65) | — |
See Table 1 for key.
Residents were asked to name the months during which they believed animals are at risk of being infected with heartworms, and in a different part of the questionnaire, they were also asked to name months during which they believed mosquitoes are active. These responses followed a similar distribution: June, July, and August were most commonly mentioned as active mosquito months; the least frequently mentioned months for mosquito activity were November through February. There was a general overlap in the range of months for mosquito activity, compared with range of months for risk of heartworm infection cited (Figure 2). The correct responses are that mosquito activity and heartworm infection are a year-round risk in this area; however, only 28% (27/96) of respondents stated that mosquitoes were active all year, and only 41% (29/70) of respondents indicated heartworm infection as a year-round risk.
Risk of heartworm infection (gray bars; n = 84) and mosquito activity (black bars; 92) per month as perceived by the respondents to the survey of Figure 1.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Risk of heartworm infection (gray bars; n = 84) and mosquito activity (black bars; 92) per month as perceived by the respondents to the survey of Figure 1.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Risk of heartworm infection (gray bars; n = 84) and mosquito activity (black bars; 92) per month as perceived by the respondents to the survey of Figure 1.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Respondents' estimates of heartworm treatment cost were compared with the authors' estimate ($1,000 to $2,000) that included hospitalization fees and treatment cost as suggested by the American Animal Hospital Association.22 A significantly (P = 0.03) greater proportion of Lake City residents versus St Augustine South residents underestimated the cost of heartworm treatment, and a significantly (P = 0.01) greater proportion of St Augustine South residents correctly estimated the treatment cost (Figure 3). As expected, dog owners were more likely to correctly estimate treatment cost (29% [15/51] vs 6% [2/33]; P = 0.009), whereas non-dog owners were more likely to underestimate costs (64% [21/33] vs 35% [18/51]; P = 0.01). Of pet-owning respondents (n = 70), 67% (47) received heartworm information from their veterinarian, and 9% (6) each received this information from a friend or family member, internet sources, or other media or had no source of heartworm information.
Cost of treating heartworm-infected dogs as estimated by the respondents to the survey in Figure 1 living in St Augustine South (gray bars; n = 41) and Lake City (black bars; 43). *Percentage of respondents differs significantly (P > 0.05) between communities.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Cost of treating heartworm-infected dogs as estimated by the respondents to the survey in Figure 1 living in St Augustine South (gray bars; n = 41) and Lake City (black bars; 43). *Percentage of respondents differs significantly (P > 0.05) between communities.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Cost of treating heartworm-infected dogs as estimated by the respondents to the survey in Figure 1 living in St Augustine South (gray bars; n = 41) and Lake City (black bars; 43). *Percentage of respondents differs significantly (P > 0.05) between communities.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Levels of concern and knowledge about mosquito-borne disease differed greatly among diseases (Figure 4). Respondents were most familiar with West Nile virus and D immitis infection. St Augustine South residents were more likely to be highly concerned about West Nile virus infection than Lake City residents (59% [27/46] vs 38% [19/50]; P = 0.04). Many respondents (42% [40/96]) had not heard of eastern equine encephalitis virus infection, despite detection of the virus in sentinel animals in St John's County before and during the survey period.23 Most respondents had not heard of Chikungunya, Saint Louis encephalitis virus infection, or dengue fever. Residents lacked familiarity with these infections, even though each of these diseases has been reported in people from the state of Florida.
Level of concern of the respondents to the survey of Figure 1 regarding various mosquito-borne diseases. CHK = Chikungunya fever. DEN = Dengue fever. DK = Don't know. EEE = Eastern equine encephalitis. H = High. L = Low. M = Moderate. N = None. SLE = St Louis encephalitis. WNV = West Nile virus infection. See Figure 1 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Level of concern of the respondents to the survey of Figure 1 regarding various mosquito-borne diseases. CHK = Chikungunya fever. DEN = Dengue fever. DK = Don't know. EEE = Eastern equine encephalitis. H = High. L = Low. M = Moderate. N = None. SLE = St Louis encephalitis. WNV = West Nile virus infection. See Figure 1 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Level of concern of the respondents to the survey of Figure 1 regarding various mosquito-borne diseases. CHK = Chikungunya fever. DEN = Dengue fever. DK = Don't know. EEE = Eastern equine encephalitis. H = High. L = Low. M = Moderate. N = None. SLE = St Louis encephalitis. WNV = West Nile virus infection. See Figure 1 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Respondents who completed some college were more likely to be concerned about Saint Louis encephalitis virus (20% [6/30]) than residents whose education did not extend beyond high school (6% [4/64]; P = 0.001). No other concern differed by education level.
Attitudes toward heartworm infection
Dog owners were more concerned about heartworm infection than non-dog owners (Table 1). Dog owners were more likely to be willing to pay more than $1,500 for treatment than non-dog owners, and non-dog owners were more likely to only be willing to pay up to $100 for treatment if they were to own a dog.
Practices and attitudes related to heartworm infection
No difference was identified in the frequency with which pet owners in Lake City and St Augustine South took their animals to their veterinarian (Figure 5); however, most dog and cat owners (77% [61/79]) took their animals to the veterinarian at least once a year. Most pet owners (71% [47/66]) administered heartworm preventives to their dogs, whereas only 1 administered them to their cat. Of pet owners not administering heartworm preventives (n = 19), the top 3 reasons were that they did not believe their pet was at risk, they had never considered that their pet could be infected, or they did not know why they were not administering the drugs. Cost was the least common reason for not administering heartworm preventives.
Frequency of veterinary visits for dog-owning (n = 50) and cat-owning (29) respondents to the survey of Figure l. LC = Lake City. SA = St Augustine South.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Frequency of veterinary visits for dog-owning (n = 50) and cat-owning (29) respondents to the survey of Figure l. LC = Lake City. SA = St Augustine South.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Frequency of veterinary visits for dog-owning (n = 50) and cat-owning (29) respondents to the survey of Figure l. LC = Lake City. SA = St Augustine South.
Citation: Journal of the American Veterinary Medical Association 254, 1; 10.2460/javma.254.1.93
Larval mosquito surveys
Three mosquito species were collected as larvae and pupae in peridomestic containers at both locations. Aedes albopictus was the most abundant species collected, followed by Culex quinquefasciatus (Table 3). Toxorhynchites rutilus septentrionalis (1/823) was collected; however, this species does not feed on blood and is not considered an important disease vector. Aedes albopictus and C quinquefasciatus were the only potential vector species breeding in standing water and artificial containers on residents' property despite adult collections of Aedes aegypti near the same locations in St Augustine South.
Results of larval mosquito surveys in St Augustine South and Lake City, Fla, in 2013.
| Variable | St Augustine South | Lake City |
|---|---|---|
| Total No. of containers collected | 140 | 147 |
| Total No. of houses surveyed | 52 | 52 |
| Container index* | 56 | 37 |
| House index† | 65 | 48 |
| Total No. of larvae collected | 351 | 472 |
| No. (%) of mosquito species | ||
| Aedes albopictus | 292 (83.2) | 434 (91.9) |
| Culex quinquefasciatus | 59 (16.8) | 37 (7.8) |
| Toxorhynchites rutilus septentrionalis | 0 (0) | 1 (0.2) |
Percentage of containers containing larvae.
Percentage of houses with containers containing larvae.
Adult mosquito collections
A total of 28 adult mosquito species were collected (Tables 4 and 5). Collection efforts in Lake City and St Augustine South were comparable. Lake City trap nights for CO2-baited traps, Aedes-specific traps, and resting boxes were 24, 24, and 20, respectively. St Augustine South trap nights totaled 31 (CO2-baited traps), 28 (Aedes-specific traps), and 32 (resting boxes). Collection hours with the vegetation aspirator totaled 6.8 in St Augustine South and 3.5 in Lake City. Aedes albopictus represented most of the collection and were trapped most effectively with Aedes-specific and CO2-baited traps. After standardizing by collection effort, the most abundant species in Lake City were A albopictus and Anopheles quadrimaculatus; A albopictus was the most abundant mosquito species in St Augustine South collections. Anopheles quadrimaculatus were captured most efficiently by CO2-baited traps, and the majority of Anopheles bradleyi and Anopheles crucians were from CO2-baited trap collections. Anopheline species were almost absent from St Augustine collections. Although species represented by both collections were similar—27 species in Lake City and 26 in St Augustine South—abundance was more evenly distributed among 6 species in Lake City, in contrast to the predominance of A albopictus in St Augustine South collections. Aedes aegypti were collected from only 3 sites in St Augustine South.
Results for adult mosquito collections performed in Lake City, Fla, in 20l3.
| Mosquito species | No. of specimens | No. of pools | No. (%) of positive pools | Minimum infection rate (95% CI) |
|---|---|---|---|---|
| Anopheles quadrimaculatus | 381 | 67 | 1 (1)* | 2.6 (0.0–7.8) |
| A albopictus | 361 | 62 | 1 (2)*† | 2.8 (0.0–8.2) |
| Culex erraticus | 274 | 44 | 1 (2) | 3.7 (0.0–10.8) |
| Aedes infirmatus | 185 | 45 | 0 | NA |
| Mansonia titillans | 159 | 48 | 0 | NA |
| Anopheles bradleyi and Anopheles crucians | 122 | 36 | 0 | NA |
| Aedes atlanticus | 102 | 31 | 0 | NA |
| Coquillettidia perturbans | 64 | 35 | 1 (3)‡ | 15.6 (0.0–46.0) |
| C quinquefasciatus | 26 | 14 | 0 | NA |
| Psorophora ferox | 25 | 13 | 0 | NA |
| Psorophora ciliata | 22 | 18 | 0 | NA |
| Culex nigripalpus | 19 | 12 | 1 (8)* | 52.6 (0.0–153.0) |
| Mansonia dyari | 19 | 9 | 0 | NA |
| Culiseta melanura | 11 | 7 | 0 | NA |
| Culiseta inornata | 10 | 9 | 1 (11)‡ | 100.0 (0.0–285.9) |
| Aedes fulvus pallens | 6 | 5 | 0 | NA |
| Culex salinarius | 6 | 6 | 0 | NA |
| Aedes triseriatus | 5 | 3 | 0 | NA |
| Anopheles punctipennis | 4 | 2 | 0 | NA |
| Culex coronator | 3 | 3 | 0 | NA |
| Aedes canadensis | 2 | 2 | 0 | NA |
| Aedes vexans | 2 | 2 | 0 | NA |
| Anopheles barberi | 2 | 2 | 0 | NA |
| Psorophora columbiae | 2 | 2 | 0 | NA |
| Aedes taeniorhynchus | 1 | 1 | 0 | NA |
| Culex restuans | 1 | 1 | 0 | NA |
| Culex territans | 1 | 1 | 0 | NA |
Minimum infection rates represent the number of infected mosquitoes/1,000 mosquitoes collected.
Confirmed positive by DNA sequencing.
Specimen had positive results for its whole body.
Specimen had positive results for its abdomen, but its head and thorax had negative results.
NA = Not applicable.
Results for adult mosquito collections performed in St Augustine South, Fla, in 20l3.
| Mosquito species | No. of specimens | No. of pools | No. (%) of positive pools | Minimum infection rate (95% CI) |
|---|---|---|---|---|
| A albopictus | 470 | 71 | 1 (1) | 2.1 (0.0–6.3) |
| A infirmatus | 67 | 12 | 0 | NA |
| C erraticus | 44 | 16 | 1 (6) | 22.7 (0.0–66.8) |
| C quinquefasciatus | 39 | 19 | 0 | NA |
| C nigripalpus | 24 | 10 | 0 | NA |
| A quadrimaculatus | 22 | 6 | 0 | NA |
| A atlanticus | 21 | 9 | 0 | NA |
| Aedes aegypti | 11 | 6 | 1 (17) | 90.9 (0.0–260.8) |
| C perturbans | 11 | 6 | 0 | NA |
| A taeniorhynchus | 7 | 5 | 0 | NA |
| C coronator | 7 | 6 | 0 | NA |
| P columbiae | 6 | 4 | 0 | NA |
| C salinarius | 4 | 4 | 0 | NA |
| A vexans | 3 | 2 | 0 | NA |
| M titillans | 3 | 3 | 0 | NA |
| A triseriatus | 2 | 2 | 0 | NA |
| A bradleyi and A crucians | 2 | 2 | 0 | NA |
| C inornata | 2 | 1 | 0 | NA |
| C restuans | 2 | 2 | 0 | NA |
| M dyari | 2 | 2 | 0 | NA |
| P ferox | 2 | 2 | 0 | NA |
| A canadensis | 1 | 1 | 0 | NA |
| A fulvus pallens | 1 | 1 | 0 | NA |
| Aedes sollicitans | 1 | 1 | 0 | NA |
| Culex peccator | 1 | 1 | 0 | NA |
| P ciliata | 1 | 1 | 0 | NA |
See Table 4 for key.
D immitis screening of mosquito pools
Of 676 mosquito pools tested, 6 (0.9%) head and thorax pools had positive results of D immitis screening. The abdomens of 1 blood-fed Coquillettidia perturbans and 1 blood-fed Culiseta inornata had positive results, whereas the respective heads and thoraces had negative results. One mosquito pool of A albopictus (n = 12) whole bodies had positive results of D immitis screening. The highest point estimates of minimum infection rate obtained from head and thorax screening were for A aegypti (90.9/1,000 mosquitoes) from St Augustine South and Culex nigripalpus*** (52.6/1,000 mosquitoes) from Lake City; however, all 95% CIs calculated on minimum infection rates included 0.
Discussion
The understanding of the biology and risk of heartworm transmission by the surveyed Florida residents in the present study was most often incorrect. A lack of knowledge and awareness was prevalent: many residents did not identify mosquitoes as transmission routes for heartworms and did not report heartworm transmission risk and mosquito activity as year-round concerns in their area. Most owners provided their dogs with heartworm preventives, but only 1 cat owner did so, despite the risk and severity of heartworm disease in cats. Unexpectedly, cost was the least common self-reported reason for failure to provide such prophylactic treatment. Lack of risk awareness and misperceptions of risk were the top reasons for noncompliance, highlighting an area for improved veterinarian-client communication.
The differences between St Augustine South and Lake City in mosquito species distributions emphasized the importance of evidence-based, ecological approaches to vector management. Peridomestic larval collections were overwhelmingly dominated by A albopictus in both St Augustine South and Lake City; this was reflected in the adult mosquito collections from St Augustine South. When integrated with positive results of molecular D immitis screening for A albopictus and A aegypti pools from St Augustine South, comprehensive heartworm control efforts in that community should prioritize reduction of mosquito breeding in peridomestic containers.
In contrast, abundances of adult mosquito species in Lake City were more evenly distributed. Aedes albopictus and A quadrimaculatus together represented equal parts of most of the collection when standardized by collection time. Aedes infirmatus, Mansonia titillans, and A bradleyi and A crucians were collected at approximately equal frequency. These species depend on a variety of ecological niches such as woodland areas, lakes, and temporary pools with submerged vegetation for successful breeding and larval development. Consequently, a multimodal approach to larval control would be ideal, but would not be feasible without intensive, invasive management of natural areas. Preliminary, more cost-effective efforts at heartworm control in Lake City should stress the avoidance of natural areas during peak mosquito activity times, guide residents in the use of repellant and physical barriers against mosquitoes entering their living spaces, and improve messaging regarding the risks and consequences of heartworm infection for both indoor and outdoor pets. Application of insecticides to mosquito districts can help reduce the adult population during times of increased mosquito burden. Additionally, a D immitis-positive pool of A albopictus was found in Lake City, making reduction of mosquito breeding in peridomestic containers important to control this known heartworm vector in this community.
To the authors' knowledge, the present study was the first to confirm vector competence of US populations of A aegypti naturally infected with D immitis in the environment.9 At the time of our study in 2013, A aegypti was only infrequently detected in St Augustine South, whereas A albopictus predominated our collections from both communities; however, A aegypti should not be discounted as a heartworm vector in locations where abundance is high.
Anopheles quadrimaculatus was an abundant species in our collection, particularly in Lake City, where we also detected a D immitis-positive pool of this species. Anopheles quadrimaculatus mosquitoes in Florida have been shown to be a competent vector of heartworm in laboratory settings, and naturally infected mosquito pools have also been found.24–26 Its abundance and mammal-biting preferences as well as the detection of a naturally infected pool in the present study make A quadrimaculatus a likely key vector of heartworm in Lake City. Culex nigripalpus, similarly, was an expected heartworm vector in the surveyed communities. This species has been found to be heartworm infected in environmental mosquito collections26,27 and is also a vector of Saint Louis encephalitis virus and eastern equine encephalitis virus in Florida.28
Two D immitis-positive pools of Culex erraticus were identified in the study reported here. This species has been found in collections of naturally infected mosquitoes in Arkansas, and there is recent evidence that C erraticus is more of an opportunistic feeder than previously understood.29,k Our findings supported this hypothesis, and more investigation is required to determine the full importance of this species in suburban and rural disease transmission cycles. An abdomen of a blood-fed C inornata was positive on D immitis screening, whereas its corresponding head and thorax were negative. Culiseta inornata has been found in collections of naturally infected mosquitoes from Arkansas30 and California31 but appears resistant to experimental infection.32 The importance of detecting this 1 D immitis-positive C inornata remains unclear. This result was most likely due to detection of recently ingested microfilariae or developing stages of D immitis in the mosquito's Malpighian tubules, which would not confirm vector competence.
Similarly, an abdomen of a blood-fed C perturbans was positive on D immitis screening; the head and thorax of the same mosquito were negative, suggesting that this specimen had likely taken a microfilaremic blood meal but did not support D immitis development to maturity. No Coquillettidia perturbans specimens have been collected that were found to be naturally infected with D immitis larvae, and this mosquito species is reportedly a poor D immitis vector in laboratory settings.33 It typically prefers avian hosts, although it has been implicated as a bridge vector for eastern equine encephalitis virus from avian transmission cycles to dead-end human and animal hosts.34 Confirmation of the vector capacity of C perturbans would need to be investigated via mosquitoes collected from the same populations, given that D immitis vector competence can vary between mosquito populations.14 Even if these C perturbans populations allow for complete D immitis development, the typical host preferences of C perturbans make it an unlikely key vector in the evaluated habitats.
The present study was limited in scope and duration for logistic reasons. Participants were approached by convenience sampling to facilitate door-to-door enrollment; many personal, sensitive demographic factors were not examined, and only 2 neighborhoods were surveyed over a 6-week period. With these limitations in mind, our aim was to provide an exploratory collection of data from 2 communities that we suspected would yield a wide range of responses to questions regarding their knowledge, attitudes, and practices regarding heartworms. Although sample sizes of surveyed residents were too small to be statistically representative of each population and therefore care must be taken in generalizing our findings, the responses provided valuable information about gaps in many residents' knowledge and highlighted opportunities for further evaluation and improvement.
The low mosquito infection rates in the present study may have been the result of low sample sizes in areas where heartworm transmission may not be as high as estimated, and our 6-week collection period may have failed to capture the true peak of transmission season. The identified key vectors may change as relative species abundances shift throughout mosquito population cycles. The methods used were based on the assumption that the positive results of molecular screening of mosquito heads and thoraces resulted from L3-stage D immitis within these anatomic sites prior to transmission through the mosquito's next blood feeding event. It was possible, although extremely unlikely, that a positive result could have been obtained when a mosquito that had recently ingested microfilaremic blood and retained microfilarial DNA was present in the mouthparts and upper digestive tract. This complication was more likely to occur for pooled samples involving whole bodies or abdomens; however, positive whole-body or abdomen pools are still valuable as evidence that those mosquito species (ie, A albopictus, C perturbans, and C inornata) are encountering and feeding on microfilaremic hosts. We confirmed PCR amplification with DNA sequencing on a subset of our positive results and are confident of our assay's specificity for D immitis sequencing. These primers were developed to amplify D immitis DNA to the exclusion of even closely related Dirofilaria repens,20 which is not endemic to the United States.
Most pet-owning respondents in the present study claimed that their veterinarian was their source for heartworm information, and some pet owners reported no source of information. Access to veterinary care is likely the limiting factor for spreading information about heartworms, and opportunities exist to improve messaging throughout the year through other sources such as the internet, television, and pamphlets; partnerships with mosquito control district officials could also be pursued as a means of disseminating veterinary recommendations pertaining to this heartworm infection.
High numbers of mosquitoes were found around residences in the present study, reinforcing the need for veterinarians to convey to owners that an indoor lifestyle does not prevent their animals from heartworm exposure. One way to portray community-specific risks and encourage use of heartworm preventives may be to provide regional heartworm incidence rates for pets or vectors. Public messaging could be improved to emphasize the importance of mosquitoes as disease vectors and to motivate mosquito-reduction efforts. Although most respondents knew that standing water was an important breeding site for mosquitoes, many residents who claimed that they had no standing water on their property did not realize that artificial containers such as garden ornaments, plant pots, and children's toys can also serve as mosquito-breeding sites. On surveying yards for mosquito larvae, we also learned that many residents did not know what mosquito larvae look like. Mosquito control messaging could be improved to include pictures of mosquito larvae and garden objects that could serve as mosquito development containers.
Year-round heartworm prevention is important in regions with year-round mosquito exposure, and even a few lapses in prevention provide the opportunity for D immitis larvae to continue their life despite rigorous mosquito control programs. Striving for and maintaining community-wide, correctly administered, year-round heartworm prevention remain an important aspect of any heartworm control program.
Acknowledgments
This research was supported by NIH Training Grant T32RR018269 and by the USDA National Institute of Food and Agriculture, multistate project NYC-139835/NE-1443.
The authors declare that there were no conflicts of interest.
Presented in part as an oral presentation at the 6th International Congress of the Society for Vector Ecology, La Quinta, Calif, September 2013; as a poster at the 62nd Annual Meeting of the American Society of Tropical Medicine and Hygiene, Washington, DC, November 2013; and as a poster at the Cornell College of Veterinary Medicine 2nd Annual DVM Research Poster Session, Ithaca, NY, April 2014.
Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the National Institute of Food and Agriculture or the USDA.
The authors thank Pam Davis from Columbia County Mosquito Control Program for her assistance and Sylvie Pitcher, who assisted with all aspects of mosquito collections and analysis.
ABBREVIATIONS
| CI | Confidence interval |
Footnotes
Cornell large vegetation aspirator, Harrington Laboratory, Ithaca, NY.
BG-Sentinel trap, Biogents AG, Regensburg, Germany.
CDC Miniature Light Trap (light removed), model 512, John W. Hock Co, Gainesville, Fla.
Paint strainer bag, Master Craft Manufacturing Co, South El Monte, Calif.
Gentra PureGene Blood Kit, Gentra Systems, Minneapolis, Minn.
Multi Doc-It Digital Imaging System, UVP Inc, Upland, Calif.
ExoSAP-IT PCR Product Cleanup, USB Products, Affymetrix Inc, Cleveland, Ohio.
BLAST, National Center for Biotechnology Information, National Institutes of Health, Bethesda, Md. Available at: blast.ncbi.nlm.nih.gov/. Accessed May 16, 2016.
IBM SPSS Statistics for Windows, version 20.0, IBM Corp, Armonk, NY.
PooledInfRate software, version 4.0, Division of Vector-Borne Diseases, CDC, Fort Collins, Colo.
Coon BR. Field and laboratory studies of Culex erraticus (Diptera: Culicidae) ability to detect hosts, habitat identification and attempts at colonization. PhD dissertation, Entomology and Nematology Department, University of Florida, Gainesville, Fla, 2006.
References
1. Phillips J. Studies on the transmission of Dirofilaria immitis in Massachusetts. Am J Epidemiol 1939;29:121–129.
2. Bowman DD, Atkins CE. Heartworm biology, treatment, and control. Vet Clin North Am Small Anim Pract 2009;39:1127–1158.
3. Companion Animal Parasite Council. Heartworm. Available at: www.capcvet.org/maps/#2017/all/heartworm-canine/dog/united-states/. Accessed Sep 30, 2018.
4. Geary TG, Bourguinat C, Prichard RK. Evidence for macrocyclic lactone anthelmintic resistance in Dirofilaria immitis. Top Companion Anim Med 2011;26:186–192.
5. Bourguinat C, Keller K, Bhan A, et al. Macrocyclic lactone resistance in Dirofilaria immitis. Vet Parasitol 2011;181:388–392.
6. Brown HE, Harrington LC, Kaufman PE, et al. Key factors influencing canine heartworm, Dirofilaria immitis, in the United States. Parasit Vectors 2012;5:245.
7. World Health Organization. Progress report 2000–2009 and strategic plan 2010–2020 of the global programme to eliminate lymphatic filariasis: halfway towards eliminating lymphatic filariasis. Geneva: World Health Organization, 2010.
8. Cupp EW, Sauerbrey M, Richards F. Elimination of human onchocerciasis: history of progress and current feasibility using ivermectin (Mectizan) monotherapy. Acta Trop 2011;120(suppl 1):S100–S108.
9. Ledesma N, Harrington L. Mosquito vectors of dog heartworm in the United States: vector status and factors influencing transmission efficiency. Top Companion Anim Med 2011;26:178–185.
10. Darsie RJ, Ward R. Identification and geographical distribution of the mosquitoes of North America, north of Mexico. Gainesville, Fla: University of Florida Press, 2005.
11. Kulasekera VL, Kramer L, Nasci RS, et al. West Nile virus infection in mosquitoes, birds, horses, and humans, Staten Island, New York, 2000. Emerg Infect Dis 2001;7:722–725.
12. Sacks BN, Chomel BB, Kasten RW. Modeling the distribution and abundance of the non-native parasite, canine heartworm, in California coyotes. Oikos 2004;105:415–425.
13. Sacks BN, Woodward DL, Colwell AE. A long-term study of non-native-heartworm transmission among coyotes in a Mediterranean ecosystem. Oikos 2003;102:478–490.
14. Tiawsirisup S. The potential for Aedes albopictus (skuse) (Diptera: Culicidae) to be a competent vector for canine heartworm, Dirofilaria immitis (leidy). Southeast Asian J Trop Med Public Health 2007;38(suppl 1):208–214.
15. Fortin JF, Slocombe JOD. Temperature requirements for the development of Dirofilaria immitis in Aedes triseriatus and Ae vexans. Mosq News 1981;41:623–633.
16. Knight DH, Lok JB. Seasonality of heartworm infection and implications for chemoprophylaxis. Clin Tech Small Anim Pract 1998;13:77–82.
17. Tuiten W, Koenraadt CJM, McComas K, et al. The effect of West Nile virus perceptions and knowledge on protective behavior and mosquito breeding in residential yards in upstate New York. EcoHealth 2009;6:42–51.
18. Edman JD, Evans FDS, Williams JA. Development of a diurnal resting box to collect Culiseta melanura (COQ). Am J Trop Med Hyg 1968;17:451–456.
19. Darsie RF, Morris CD. Keys to the adult females and fourth instar larvae of the mosquitoes of Florida (Diptera, Culicidae). Gainesville, Fla: Florida Mosquito Control Association, 2003;1:159.
20. Rishniw M, Barr SC, Simpson KW, et al. Discrimination between six species of canine microfilariae by a single polymerase chain reaction. Vet Parasitol 2006;135:303–314.
21. Biggerstaff BJ. PooledlnfRate, version 4.0: a Microsoft Office Excel Add-In to compute prevalence estimates from pooled samples. Fort Collins: Division of Vector-Borne Diseases, CDC, 2009.
22. Townzen JS, Brower AVZ, Judd DD. Identification of mosquito bloodmeals using mitochondrial cytochrome oxidase sub-unit I and cytochrome b gene sequences. Med Vet Entomol 2008;22:386–393.
23. CDC. Non-human disease cases. Reported to CDC ArboNET by county of residence. Available at: wwwn.cdc.gov/arbonet/maps/ADB_Diseases_Map/index.html. Accessed Sep 30, 2018.
24. Kartman L. Factors influencing infection of the mosquito with Dirofilaria immitis (leidy, 1856). Exp Parasitol 1953;2:27–78.
25. Scoles GA. Vectors of canine heartworm in the United States: a review of the literature including new data from Indiana, Florida, and Louisiana, in Proceedings. Heartworm Symp Am Heartworm Soc 1998;21–36.
26. Watts KJ, Reddy GR, Holmes RA, et al. Seasonal prevalence of third-stage larvae of Dirofilaria immitis in mosquitoes from Florida and Louisiana. J Parasitol 2001;87:322–329.
27. Sauerman DM Jr, Nayar JK. A survey for natural potential vectors of Dirofilaria immitis in Vero Beach, Florida. Mosq News 1983;43:222–225.
28. Day JF, Stark LM. Transmission patterns of St. Louis encephalitis and eastern equine encephalitis viruses in Florida: 1978–1993. J Med Entomol 1996;33:132–139.
29. Mendenhall IH, Tello SA, Neira LA, et al. Host preference of the arbovirus vector Culex erraticus (Diptera: Culicidae) at Sonso Lake, Cauca Valley department, Colombia. J Med Entomol 2012;49:1092–1102.
30. Mckay T, Bianco T, Rhodes L, et al. Prevalence of Dirofilaria immitis (Nematoda: Filarioidea) in mosquitoes from northeast Arkansas, the United States. J Med Entomol 2013;50:871–878.
31. Huang S, Smith DJ, Molaei G, et al. Prevalence of Dirofilaria immitis (Spirurida: Onchocercidae) infection in Aedes, Culex, and Culiseta mosquitoes from north San Joaquin Valley, CA. J Med Entomol 2013;50:1315–1323.
32. Frimeth JP, Arai HP. Some potential mosquito vectors of the canine heartworm, Dirofilaria immitis, in the Calgary region of southern Alberta. Can J Zool 1983;61:1156–1158.
33. Todaro WS, Morris CD, Heacock NA. Dirofilaria immitis and its potential mosquito vectors in central New York state. Am J Vet Res 1977;38:1197–1200.
34. Nasci RS, Mitchell CJ. Arbovirus titer variation in field-collected mosquitoes. J Am Mosq Control Assoc 1996;12:167–171.
