The FDA's Center for Veterinary Medicine monitors adverse events associated with approved products for heartworm prophylaxis.1 Heartworm medications that are licensed for prevention of heartworm infection in dogs are currently approved for 100% efficacy by the Center for Veterinary Medicine; however, surveillance data indicate treatment failures occur in all heartworm preventive categories.1 In a recent study,2 2 groups of dogs were administered 1 dose of ivermectin or milbemycin oxime by mouth 30 days after experimental infection with 50 L3-stage larvae of Dirofilaria immitis. At necropsy, 1 of 14 dogs in each treatment group had a single adult heartworm on day 123 after treatment. An in vitro microfilariae-based assay has also been developed to determine the susceptibility of D immitis to the macrocytic lactone class of preventives3; however, the results are yet to be published. Several studies4–9 have found macrocyclic lactones to be 100% effective in preventing heartworm infection. In addition, when considering the proportion of D immitis in refugia, together with the specific stage in the life cycle that is targeted, and a dosage that does not have an adulticidal effect, it is unlikely that resistance to macrocyclic lactones would be selected with current treatment strategies.10 Most failures of prophylaxis have been reported in heartworm-endemic states. At this time, it is unclear whether there has been a true increase in complaints of ineffectiveness and, if so, whether this increase is because of an increase in reporting, lack of compliance, or failure of 1 or more specific products.1
The AHS published guidelines in 2010 for the prevention and management of heartworm infection in dogs.11 These guidelines were revised from recommendations published in 2003 and 2005 to focus on newer research and additional clinical experience, particularly in the subjects of heartworm chemoprophylaxis and serologic testing and retesting for heartworm infection.
The Center for Veterinary Medicine has suggested that more comprehensive reporting in the future may provide evidence of patterns of emerging resistance to products used for heartworm prophylaxis and lead to an updated label claim of efficacy1 The odds of heartworm infection in dogs is reportedly associated with increasing age up to 7 to 8 years, time spent outdoors (particularly in the evening and at night), sexually intact males and large hound breeds, high dog or coyote population density, and neighborhoods of low socioeconomic status.12–15 To the authors' knowledge, there are no published reports of studies conducted to identify risk factors for failure of heartworm prophylaxis on the basis of practices in use by dog owners. The purpose of the study reported here was to compare practices used by a group of dog owners, dog trainers, and kennel owners to prevent heartworm infection in dogs in which the prophylactic measures were successful and those in which the measures failed, with the aim of identifying practices associated with failure.
Materials and Methods
Study subjects—A national hunting dog club was contacted for permission to conduct an online survey of heartworm prevention practices used by club members. Permission to conduct the survey was granted by club officers, and an invitation to participate was placed on the home page of the club website. Access to the online questionnaire was available to anyone accessing the website from May 1 to November 15, 2008. In addition, during the same period, dog owners attending 8 club-sponsored events and field trials were made aware of the online questionnaire via a general announcement and were also given the opportunity to complete and mail in a paper-based questionnaire identical to the online version.
Procedures—The survey instrument was designed to acquire the following information: respondent's experience with dogs, US state of residence, risk and duration of exposure of dogs to mosquitoes, method by which products used for heartworm prophylaxis were chosen, duration of prophylaxis, method of administration, frequency and timing of heartworm tests administered, and whether the respondent had ever had a dog with a positive heartworm test result after prophylaxis had been administered for ≥ 1 year or season (defined as prophylaxis failure). The questionnaire was pretested by a group of dog-owning staff members and students at the University of Tennessee College of Veterinary Medicine and edited on the basis of their responses.
Respondents who opted for the online questionnaire were directed to a secure server at the College of Veterinary Medicine. In addition, paper copies of the questionnaire were handed out at the club events, together with stamped envelopes preaddressed to the principal investigator (BWR).
Online questionnaire—Online survey data were converted to HTML format and stored in a relational database.a Data from respondents who filled out the paper questionnaire were added to the database as the questionnaires were received. At the termination of the data collection period, questionnaire responses were downloaded from the server into a spreadsheet programb and analyzed with a statistical software package.c
Heartworm transmission in the continental United States is predicted to occur for 6 months or less above the 37th parallel north of the earth's equatorial plane (ie, Virginia-North Carolina state line).16 Therefore, respondents were separated into 2 groups: those residing north (group A) and those residing south (group B) of the 37th parallel. States included in group B were North Carolina, Tennessee, South Carolina, Georgia, Florida, Arizona, Louisiana, Oklahoma, Texas, New Mexico, Arizona, Mississippi, and Alabama. All other states were classified as group A.
Statistical analysis—Summary statistics for continuous variables (eg, years of experience caring for dogs and number of hours dogs were kept outdoors at dusk, at dawn, or after dark during the mosquito season) are expressed as mean ± SD unless data were nonnormally distributed, in which situation median values and 25th and 75th percentiles (IQR) are provided. A statistical software package was used for all analyses.c Univariate comparisons of distributions for categorical variables (eg, location of residence or kennel relative to 37th parallel and presence or absence of risk factors within 100 yards of dogs housing) between respondents that experienced success of heartworm prophylaxis during the previous 12 months and those that experienced failure were made by use of a χ2 or Fisher exact test (when an expected value in the 2 × 2 contingency table was < 5). Values of continuous variables were compared between the 2 groups with the Student t test or a nonparametric equivalent, depending on the distribution of data. Fit of continuous data to a normal distribution was assessed with the Shapiro-Wilk method.
To fit a model to the data and increase the power to detect differences between prophylaxis success and failure in a multivariate analysis, respondents that ever experienced failure were included in a logistic regression model. The final multivariate model was identified by use of a selection procedure that displays the best 3 models by selecting those with the highest χ2 values for models with 3 to 9 variables The dependent variable was prophylaxis failure at any time in the past versus successful prophylaxis, and independent variables included those identified to have a P value ≤ 0.20 in the univariate analyses, together with all 2-way interactions of those variables. Values of P < 0.05 were considered significant for all final analyses.
Results
Study subjects—Seven hundred eight completed questionnaires were received (online, n = 669; paper, 39). Respondents identified themselves and were classified as dog owners (441/707 [62%]) or in a commercial group (266/707 [38%]) that included dog trainers and commercial kennel operators. One hundred ninety-nine of 682 (29%) respondents resided north of the 37th parallel, and 483 of 682 (71%) resided south of the line (26 respondents did not provide an address). The geographic distribution and density of respondent residences was similar to the reported geographic distribution of heart-worm disease in dogs in 2007 (Figures 1 and 2).
Incidence of heartworm disease in dogs in the United States in 2007. (Available at: www.heartwormsociety.org/veterinary-resources/slide.html. Accessed Mar 15, 2011. Reprinted with the permission of the AHS).
Citation: Journal of the American Veterinary Medical Association 238, 9; 10.2460/javma.238.9.1150
Incidence of heartworm disease in dogs in the United States in 2007. (Available at: www.heartwormsociety.org/veterinary-resources/slide.html. Accessed Mar 15, 2011. Reprinted with the permission of the AHS).
Citation: Journal of the American Veterinary Medical Association 238, 9; 10.2460/javma.238.9.1150
Incidence of heartworm disease in dogs in the United States in 2007. (Available at: www.heartwormsociety.org/veterinary-resources/slide.html. Accessed Mar 15, 2011. Reprinted with the permission of the AHS).
Citation: Journal of the American Veterinary Medical Association 238, 9; 10.2460/javma.238.9.1150
Distribution of respondents in a survey of heartworm control practices in dogs (n = 615). Each dot was generated from available GPS coordinates for the address for a single respondent (GPS coordinates and addresses were unavailable for some survey respondents).
Citation: Journal of the American Veterinary Medical Association 238, 9; 10.2460/javma.238.9.1150
Distribution of respondents in a survey of heartworm control practices in dogs (n = 615). Each dot was generated from available GPS coordinates for the address for a single respondent (GPS coordinates and addresses were unavailable for some survey respondents).
Citation: Journal of the American Veterinary Medical Association 238, 9; 10.2460/javma.238.9.1150
Distribution of respondents in a survey of heartworm control practices in dogs (n = 615). Each dot was generated from available GPS coordinates for the address for a single respondent (GPS coordinates and addresses were unavailable for some survey respondents).
Citation: Journal of the American Veterinary Medical Association 238, 9; 10.2460/javma.238.9.1150
Identification of risk factors for prophylaxis failure—Prophylaxis failure during the past year was reported by 65 of 708 (9%) respondents, and 56 (8%) reported failure at some time in the past but not in the past 12 months. When data from kennel owners and dog trainers were combined, these respondents were at lower odds of reporting prophylaxis failure than hunting dog owners (OR, 0.59; 95% CI, 0.33 to 1.04); however, the difference between groups was not significant (P = 0.07; Table 1). When the annual incidence of prophylaxis failure was calculated with the total number of dogs cared for during the previous 12 months as the denominator, dog owners were at significantly (P < 0.001) greater odds of prophylaxis failure (47/1,753 [3%]) than kennel owners and dog trainers combined (18/1,708 [1%]; OR, 2.6; 95% CI, 1.5 to 4.6). The odds of failure increased significantly as the number of dogs in a respondent's care decreased. Having ≥ 1 dog ≤ 3 months of age was associated with a reduced odds of prophylaxis failure (OR, 0.43; 95% CI, 0.23 to 0.80; P = 0.007), as was residence north of the 37th parallel (OR, 0.25; 95% CI, 0.11 to 0.55; P < 0.001).
Results of univariate analyses comparing characteristics of respondents (hunting dog owners, dog trainers, and kennel owners) to a survey of heartworm prevention practices who reported failure or success of heartworm prophylaxis in the past 12 months.
| Characteristic | Value for those reporting failure | Value for those reporting success | OR | 95% CI | P value |
|---|---|---|---|---|---|
| Relationship to dog (n = 649) | |||||
| Kennel owner or dog trainer | 18 (28) | 230 (39) | 0.59 | 0.33–1.04 | 0.07 |
| Dog owner* | 47 (72) | 354 (61) | — | — | — |
| Experience with dogs (y; n = 650)† | 17 (9–30) | 19 (10–28) | — | — | 0.85 |
| Median No. of dogs cared for in past 12 mo (n = 650)† | 2 (1–3) | 3 (2–5) | — | — | < 0.001 |
| 1 | 18 (28) | 76 (13) | 5.97 | 2.13–16.73 | < 0.001 |
| 2–5 | 42 (65) | 383 (65) | 2.76 | 1.07–7.14 | 0.01 |
| > 5* | 5 (8) | 126 (22) | 1.0 | — | — |
| 1 or more dogs ≤ 3 mo of age | 13 (20) | 216 (37) | 0.43 | 0.23–0.80 | 0.007 |
| Commercial† | 2 (1–3) | 4 (3–8) | — | — | 0.03 |
| Dog owner† | 2 (2–4) | 3 (2–4) | — | — | 0.009 |
| Location of residence relative to 37th parallel | |||||
| North (n = 194) | 7 (4) | 187 (96) | 0.25 | 0.11–0.55 | < 0.001 |
| South (n = 432) * | 57 (13) | 375 (87) | — | — | — |
Values are reported as number (percentage) unless indicated otherwise. A value of P < 0.05 was considered significant.
Referent group.
Values are reported as median (IQR) because data were not normally distributed.
— = Not applicable.
Mosquito activity was observed during the entire previous year by 6 of 65 (9%) respondents reporting prophylaxis failure, compared with 96 of 585 (16%) respondents reporting successful prophylaxis; however, this difference was not significant (P = 0.13). The proportion of respondents reporting mosquito activity from May to October was significantly (P = 0.005) greater among those reporting prophylaxis failure (13/65 [20%]) versus for those reporting success (52/584 [9%]). The proportions of respondents reporting mosquito activity prior to March and after November 2007 were similar between groups. When the data from the subset of respondents residing south of the 37th parallel were evaluated, 23 of 57 (40%) reporting prophylaxis failure also reported mosquito activity in November and December, compared with 167 of 375 (45%) reporting successful prophylaxis (P = 0.55).
One or more environmental factors known to provide a favorable habitat for mosquitoes were reportedly in proximity to 34 of 65 (52%) dogs cared for by owners reporting prophylaxis failure and 330 (56%) respondents reporting success (P = 0.53). The number of hours dogs spent outdoors during the mosquito season for their areas was significantly (P = 0.005) greater for those cared for by respondents reporting successful prophylaxis (median, 10 hours; IQR, 4 to 24 hours) than for those cared for by respondents reporting prophylaxis failure (median, 4 hours; IQR, 4 to 24 hours; Table 2). Dogs were more likely to be outdoors at dawn, at dusk, or after dark during mosquito season when they belonged to respondents that reported successful prophylaxis (464/585 [79%]) than when they belonged to respondents reporting failure (47/65 [72%]); however, this difference was not significant (P = 0.19).
Results of univariate analyses comparing characteristics of dogs' environments as described by respondents (hunting dog owners, trainers, and kennel owners) to a survey of heartworm prevention practices who reported failure (n = 65) or success (585) of heartworm prophylaxis for their dogs in the past 12 months.
| Characteristic | Value for those reporting failure | Value for those reporting success | OR (95% CI) | P value |
|---|---|---|---|---|
| Factors within 100 m of dog | ||||
| Debris (wood pile, old tires, etc) | 2 (3) | 21 (4) | 0.85 (0.20–3.72) | 1.00 |
| Low area with standing water | 11 (17) | 101 (17) | 0.98 (0.49–1.93) | 0.94 |
| Low area and debris | 3 (5) | 14 (2) | 1.97 (0.55–7.06) | 0.24 |
| Risk factors for presence of mosquitoes | ||||
| None | 31 (48) | 255 (44) | 1.18 (0.71–1.98) | 0.53 |
| Pond | 6 (9) | 86 (15) | 0.59 (0.25–1.41) | 0.23 |
| Pond and debris | 0 (0) | 4 (1) | NC | 1.0 |
| Pond and low area | 1 (2) | 32 (5) | 0.27 (0.04–2.01) | 0.24 |
| Pond, low area, and debris | 0 (0) | 11 (2) | NC | 0.61 |
| Stream | 6 (9) | 18 (3) | 3.20 (1.22–8.38) | 0.03 |
| Stream and debris | 0 (0) | 4 (1) | NC | 1.00 |
| Stream and low area | 1 (2) | 16 (3) | 0.56 (0.07–4.26) | 1.0 |
| Stream, low area, and debris | 1 (2) | 2 (< 1) | 4.55 (0.41–50.93) | 0.27 |
| Stream, pond, and debris | 0 (0) | 1 (< 1) | NC | 1.0 |
| Stream, pond, and low area | 1 (2) | 12 (2) | 0.75 (0.10–5.83) | 1.00 |
| Stream, pond, low area, and debris | 2 (3) | 8 (2) | 2.29 (0.48–11.02) | 0.26 |
| Duration dog was outdoors during mosquito season (h)* | 4 (2–24) | 10 (4–24) | 0.005 | |
| Dog outdoors at dusk | 21 (32) | 181 (31) | 1.07 (0.62–1.84) | 0.82 |
| Dog outdoors at dawn | 28 (43) | 206 (35) | 1.39 (0.83–2.34) | 0.21 |
| Dog outdoors after dark | 31 (48) | 287 (49) | 0.95 (0.57–1.58) | 0.83 |
Values are reported as number (percentage) unless indicated otherwise.
Values reported are median (IQR); number reporting failure was 65, and number reporting success was 584.
NC = Not calculable.
The referent group for all comparisons is successful prophylaxis. A value of P < 0.05 was considered significant.
Methods used to control exposure of dogs to mosquitoes were similar among respondents reporting prophylaxis failure or success (Table 3). None of the methods used to reduce exposure to mosquitoes were used by 15 of the 65 (23%) respondents reporting prophylaxis failure and by 162 of the 585 (28%) reporting success. The brand of heartworm preventive used at the time the questionnaire was completed had been used for a median of 60 months (IQR, 36 to 108 months) by respondents reporting prophylaxis failure and a median of 60 months (IQR, 24 to 120 months) by respondents reporting successful prophylaxis (P = 0.78).
Results of univariate analyses comparing distributions of methods used to control dog exposure to mosquitoes as described by respondents (hunting dog owners, trainers, and kennel owners) to a national survey of heartworm prevention practices who reported failure (n = 65) or success (585) of heartworm prophylaxis for their dogs in the past 12 months.
| Method | No.(%) reporting failure | No. (%) reporting success | OR (95% CI) | P value |
|---|---|---|---|---|
| Environmental chemical use | 5 (8) | 39 (7) | 1.17 (0.44–3.07) | 0.79 |
| Environmental and indoor chemical use | 2 (3) | 13 (2) | 1.40 (0.31–6.33) | 0.65 |
| Indoor chemical use | 15 (23) | 130 (22) | 1.05 (0.57–1.93) | 0.88 |
| Topical repellent | 17 (26) | 134 (23) | 1.19 (0.66–2.14) | 0.56 |
| Topical repellent and environmental chemical use | 1 (2) | 22 (4) | 0.40 (0.05–3.02) | 0.72 |
| Topical repellent and environmental and indoor chemical use | 1 (2) | 11 (2) | 0.82 (0.10–6.42) | 1.00 |
| Topical repellent and indoor chemical use | 8 (12) | 56 (10) | 1.33 (0.60–2.92) | 0.48 |
| No mosquito control | 15 (23) | 162 (28) | 0.78 (0.43–1.43) | 0.43 |
| Other* | 1 (2) | 18 (3) | 0.49 (0.06–3.75) | 0.71 |
Combination of window screens plus other methods.
Chemicals consisted of chemical products used to control mosquitoes in the environment.
Although the majority of respondents (533/650 [82%]) chose a product to prevent heartworm disease on the basis of advice from their veterinarian, 92% of those reporting prophylaxis failure did so, compared with 81% of those reporting success (P = 0.02; Table 4). Among respondents that did not provide year-round heartworm prophylaxis for their dogs (n = 93), approximately a third of respondents from both groups reported using advice from their veterinarian, presence of mosquito activity, and tradition (starting and stopping on the same dates each year) to determine the duration of prophylaxis.
Results of univariate analyses comparing the means by which decisions regarding prophylaxis timing, treatment, dose, and route of administration were made by respondents (hunting dog owners, trainers, and kennel owners) to a national survey of heartworm prevention practices who reported failure (n = 65) or success (585) of heartworm prophylaxis for their dogs in the past 12 months.
| Basis of decision | No. (%) reporting failure | No. (%) reporting success | OR (95% CI) | P value |
|---|---|---|---|---|
| Advice from veterinarian* | 60 (92) | 473 (81) | 2.8 (1.12–7.24) | 0.02 |
| Advice from nonveterinarian* | 25 (38) | 218 (37) | 1.05 (0.62–1.78) | 0.85 |
| Magazine advertisements* | 2 (3) | 16 (3) | 1.13 (0.25–5.02) | 0.70 |
| Other | 5 (8) | 55 (9) | 0.80 (0.31–2.08) | 0.65 |
| Cost | 1 (20) | 19 (35) | 0.47 (0.06–3.53) | 0.71 |
| Personal research | 2 (40) | 15 (27) | 1.21 (0.27–5.40) | 0.68 |
| Experience | 0 (0) | 12 (22) | NC | 0.62 |
| Multiple use | 0 (0) | 4 (7) | NC | 1.00 |
| Tradition | 1 (20) | 3 (5) | 3.03 (0.31–29.57) | 0.34 |
| Sales representative | 1 (20) | 1 (2) | 9.13 (0.56–147.65) | 0.19 |
| Free samples | 0 (0) | 1 (2) | NC | 1.00 |
Referent group.
NC = Not calculable.
Multiple responses were possible; therefore, the total percentage for any 1 characteristic may exceed 100. A value of P < 0.05 was considered significant.
Most respondents obtained advice on the dose of heartworm preventive from their veterinarian, and the dose was determined by weighing the dog; however, significantly (P = 0.04) more respondents (43/65 [66%]) reporting prophylaxis failure did so, compared with the proportion reporting successful prophylaxis (309/585 [53%]; Table 5). Respondents reporting failure were less likely to rely on memory when administering heartworm preventives and slightly more likely to record the date the preventive was given than were those reporting successful prophylaxis; however, these differences were not significant. Although the numbers were small, respondents reporting use of a topical heartworm preventive were significantly (P = 0.02) more likely to report failure than respondents that reported successful prophylaxis. The continuous administration of heartworm preventives was reported by 82% of respondents reporting prophylaxis failure and 86% of those reporting success. When data from only respondents residing south of the 37th parallel were evaluated, 44 of the 54 (81%) respondents reporting prophylaxis failure administered heartworm preventives year-round, versus 302 of the 351 (86%) respondents reporting success; however, the difference was not significant (P = 0.38). For respondents reporting seasonal administration of heartworm preventives, administration was for a mean ± SD duration of 204 ± 69 days (6.7 months) for those reporting failure versus 217 ± 66 days (7.3 months) for those reporting success (P = 0.64).
Results of univariate analyses comparing dosage determination and methods of administration of heartworm preventives as reported by respondents (hunting dog owners, trainers, and kennel owners) to a national survey of heartworm prevention practices who reported failure (n = 65) or success (585) of heartworm prophylaxis for their dogs in the past 12 months.
| Characteristic | No. (%) reporting failure | No. (%) reporting success | OR (95% CI) | P value |
|---|---|---|---|---|
| Respondent's method of dosage determination | ||||
| Read label and weigh all dogs separately | 26 (40) | 239 (41) | 0.97 (0.57–1.63) | 0.89 |
| Read label and estimate weight for each dog | 3 (5) | 50 (9) | 0.52 (0.16–1.71) | 0.27 |
| Read label and all dogs given same dose | 1 (2) | 7 (1) | 1.29 (0.16–10.65) | 0.57 |
| Veterinarian and weigh all dogs separately | 43 (66) | 309 (53) | 1.75 (1.02–3.00) | 0.04 |
| Veterinarian and all dogs given same dose | 2 (3) | 31 (5) | 0.57 (0.13–2.43) | 0.76 |
| Other | 1 (2) | 30 (5) | 0.29 (0.04–2.16) | 0.35 |
| Method for timing preventive administration* | ||||
| Same day each month | 59 (92) | 550 (95) | 0.63 (0.25–1.55) | 0.28 |
| Rely on memory to give preventive each month | 39 (60) | 407 (70) | 0.66 (0.39–1.11) | 0.11 |
| Record date when preventive administered | 14 (22) | 101 (17) | 1.32 (0.70–2.47) | 0.39 |
| Other† | 13 (20) | 117 (20) | 1.00 (0.53–0.90) | 1.0 |
| Method of preventive administration | ||||
| With food | 12 (18) | 126 (22) | 0.82 (0.43–1.59) | 0.57 |
| By mouth | 42 (65) | 385 (66) | 0.95 (0.55–1.62) | 0.85 |
| Other | 7 (11) | 28 (5) | 2.40 (1.00–5.74) | 0.07 |
| Nature of preventive product | ||||
| Chewable | 1 (2) | 14 (2) | 0.64 (0.08–4.93) | 1.0 |
| Injection | 0 (0) | 3 (0.5) | — | 1.0 |
| Topical | 6 (9) | 18 (3) | 3.2 (1.22–8.38) | 0.03 |
| Heartworm preventive administered year-round | 56 (86) | 510 (87) | 0.92 (0.43–1.93) | 0.82 |
| Residence north of 37th parallel | 6 (86) | 165 (88) | 0.80 (0.09–6.96) | 0.59 |
| Residence south of 37th parallel | 44 (81) | 302 (86) | 0.71 (0.34–1.51) | 0.38 |
| Dog observed to spit out pill | 10 (15) | 69 (12) | 1.34 (0.65–2.74) | 0.43 |
| No. of times dog observed to spit out pill‡ | 2 (1–5) | 2 (2–3) | — | 0.67 |
The number reporting failure was 64 and the number reporting success was 579.
Refers to other regular monthly occurring events.
Values reported are median (IQR) because data were not normally distributed.
— = Not applicable.
The referent group for all comparisons is successful prophylaxis. A value of P < 0.05 was considered significant.
For respondents residing south of the 37th parallel, 15 of 52 (29%) reporting prophylaxis failure and 118 of 307 (38%) reporting successful prophylaxis had their dogs tested for heartworm infection prior to April 1 during the previous 12 months (P = 0.19; Figure 3). Testing at least once a year was reported by 59 of 65 (91%) respondents reporting failure and 445 of 584 (76%) reporting successful prophylaxis (P = 0.008; Table 6). In addition, respondents reporting failure were more likely to report having a heartworm test performed in the past 12 months (97%) than respondents reporting successful prophylaxis (82%; P = 0.002). Of respondents that changed heartworm preventive products during the past year, 12 of 17 (71%) reporting prophylaxis failure had had their dogs tested for heartworm infection prior to changing products, compared with 42 of 105 (40%) reporting successful prophylaxis (P = 0.02). Only 47% of those reporting prophylaxis failure and 7% of those reporting successful prophylaxis reported having a second test performed after changing products (P = 0.001).
Distribution of respondents (hunting dog owners, dog trainers, and kennel owners) to a national survey of heartworm control practices in dogs by month of performance of a heartworm test in the previous 12 months. Respondents include only those that resided south of the 37th parallel north of the earth's equatorial plane and are separated according to whether they reported failure (n = 52) or success (307) of heartworm prophylaxis.
Citation: Journal of the American Veterinary Medical Association 238, 9; 10.2460/javma.238.9.1150
Distribution of respondents (hunting dog owners, dog trainers, and kennel owners) to a national survey of heartworm control practices in dogs by month of performance of a heartworm test in the previous 12 months. Respondents include only those that resided south of the 37th parallel north of the earth's equatorial plane and are separated according to whether they reported failure (n = 52) or success (307) of heartworm prophylaxis.
Citation: Journal of the American Veterinary Medical Association 238, 9; 10.2460/javma.238.9.1150
Distribution of respondents (hunting dog owners, dog trainers, and kennel owners) to a national survey of heartworm control practices in dogs by month of performance of a heartworm test in the previous 12 months. Respondents include only those that resided south of the 37th parallel north of the earth's equatorial plane and are separated according to whether they reported failure (n = 52) or success (307) of heartworm prophylaxis.
Citation: Journal of the American Veterinary Medical Association 238, 9; 10.2460/javma.238.9.1150
Results of univariate analyses comparing heartworm testing approaches as reported by respondents (hunting dog owners, trainers, and kennel owners) to a national survey of heartworm prevention practices who reported failure (n = 65) or success (585) of heartworm prophylaxis for their dogs in the past 12 months.
| Characteristic | No. (%) reporting failure | No. (%) reporting success | OR (95% CI) | P value |
|---|---|---|---|---|
| Frequency of heartworm testing* | ||||
| Annual | 39 (60) | 389 (67) | 0.76 (0.45–1.28) | 0.29 |
| Every other year | 2 (3) | 50 (9) | 0.34 (0.08–1.43) | 0.12 |
| No test | 2 (3) | 67 (11) | 0.25 (0.06–1.02) | 0.04 |
| Twice yearly | 20 (31) | 56 (10) | 4.20 (2.32–7.61) | < 0.001 |
| Other | 2 (3) | 22 (4) | 0.81 (0.19–3.53) | 1.00 |
| Heartworm test at least 1 time/y | 59 (91) | 445 (76) | 3.07 (1.30–7.27) | 0.008 |
| Dog tested on or prior to introduction to household or kennel | 45/65 (69) | 361/580 (62) | 1.37 (0.79–2.37) | 0.27 |
| Second test after introduction to household or kennel | 12/62 (19) | 70/578 (12) | 1.74 (0.88–3.43) | 0.10 |
| Interval to second test if second test performed (mo)† | 6 (5–12) | 6 (2–6) | — | 0.24 |
| Tested for infection in past 12 mo | 62/64 (97) | 478/583 (82) | 6.81 (1.64–28.28) | 0.002 |
| Of respondents that changed heartworm preventive in past 12 mo | ||||
| Test performed prior to change | 12/17 (71) | 42/105 (40) | 3.60 (1.18–10.97) | 0.02 |
| Second test performed after change | 7/15 (47) | 5/73 (7) | 11.90 (3.05–46.44) | < 0.001 |
| Interval to second test if second test performed (mo)† | 6 (5–12) | 6 (2–6) | — | 0.21 |
| Name of test used to detect heartworm infection in past 12 mo* | ||||
| Antigen and microfilaria | 27 (42) | 53 (9) | 7.13 (4.04–12.59) | < 0.001 |
| Antigen only | 4 (6) | 36 (6) | 1.00 (0.34–2.90) | 1.00 |
| Microfilaria only | 0 (0) | 6 (1) | NC | 1.00 |
| No test done | 0 (0) | 72 (12) | NC | 0.003 |
| Unknown | 33 (51) | 417 (71) | 0.42 (0.25–0.70) | < 0.001 |
For this variable, the number of respondents reporting failure was 65 and the number reporting success was 584.
Values reported are median (IQR) because data were not normally distributed.
— = Not applicable. NC = Not calculable.
Multivariate analysis revealed that the variables dog outdoors after sunset, before sunrise, or after dark and number of hours spent outdoors during the mosquito season were independently associated with successful heartworm prophylaxis (Table 7). However, when dogs that were both outdoors after sunset, before sunrise, or after dark and outdoors for a greater number of hours during the mosquito season were compared with all others, the odds of prophylaxis failure were significantly (P = 0.008) increased (OR, 1.36; 95% CI, 1.08 to 1.71). Residence north of the 37th parallel was associated with successful heartworm prophylaxis, and respondents whose prophylaxis strategy failed were more likely to know the name of the heartworm test that was used. Respondents that had their dogs tested for heartworm infection less than once a year and had any testing performed prior to April 1 were significantly (P = 0.018) more likely to report prophylaxis failure than were other respondents (OR, 10.6; 95% CI, 1.64 to 74.66).
Results of multivariate analyses of respondent (n = 570*; hunting dog owners, trainers, and kennel owners) practices associated with failure of heartworm prophylaxis at any time in the past.
| Factor | OR | 95% CI | P value |
|---|---|---|---|
| Heartworm test prior to April 1 | 0.55 | 0.40–2.24 | 0.24 |
| Residence north of 37th parallel | 0.24 | 0.12–0.46 | < 0.001 |
| Know name of heartworm test | 3.14 | 1.96–5.02 | < 0.001 |
| Dog outdoors at dusk, at dawn, or after dark | 0.29 | 0.11–0.73 | 0.009 |
| Duration dog spent outdoors during mosquito season (h) | 0.73 | 0.59–0.92 | 0.007 |
| Increased hours outdoors and out at dusk, dawn, or dark† | 1.36 | 1.08–1.71 | 0.008 |
| Heartworm test performed < 1 time/y | 0.29 | 0.06–1.29 | 0.90 |
| Heartworm test performed < 1 time/y and test performed prior to April† | 10.59 | 1.64–74.66 | 0.018 |
Failure at any time in the past (n = 108) and never experienced failure (462).
Represents the odds for the 2-way interaction between risk factors for dogs with both factors, compared with the odds for all other dogs.
Probability modeled is failure of heartworm prophylaxis at any time in the past. A value of P < 0.05 was considered significant.
Discussion
In the study reported here, heartworm prevention practices were reported by dog owners, dog trainers, and kennel owners who were members of a national hunting dog club or anyone who accessed the club website or attended events sponsored by the club. Kennel owners and professional dog trainers have a financial stake in preventing heartworm infection among dogs in their care and were responsible for a significantly greater number of dogs per respondent than were those who classified themselves as dog owners. This may have motivated them to seek information and to be more compliant with AHS recommendations for heartworm prevention.
Continuous administration of heartworm preventives was high among survey respondents. Among respondents residing south of the 37th parallel, those reporting failure of prophylaxis were less likely to report year-round administration of prophylaxis (81%) when compared with those reporting success (86%); however, this difference was not significant. In a previous study,16 heartworm transmission in the continental United States was found to be limited to 6 months or less above the 37th parallel. At the time of the present survey, the 2005 AHS guidelines stated that because of the longer transmission season south of the 37th parallel and to increase dog-owner compliance, heartworm prophylaxis should be administered year-round.17 The guidelines in 2010 state that to be maximally effective, heartworm prophylaxis should be given to all dogs residing in the continental United States year-round, but if seasonal treatment is chosen, administration should begin within 1 month of the anticipated start of the transmission season and the last dose should be given 1 month after transmission ceases.11 Although these data support the year-round administration of heartworm prophylaxis, additional studies that include a larger sample size are needed to confirm the usefulness of this practice. In contrast to our results, the necessity of year-round administration of heartworm preventives was questioned in a previous report16 that concluded it was not justified, other than for dogs in the most southern regions of Florida and Texas, on either a biological or economic basis.
In the present study, efforts to reduce dog exposure to mosquitoes and habitat favorable to mosquito activity were similar between the comparison groups; however, there was a significant increase in the number of hours that dogs of respondents reporting successful prophylaxis spent outdoors during mosquito season and no difference in the proportions of dogs kept outdoors at dusk, at dawn, or after dark between the comparison groups. It was only in the multivariable analysis, when dogs that spent an increased number of hours outdoors and were also outdoors at dusk, at dawn, or after dark were compared with all others, that the odds of prophylaxis failure were significantly increased. This suggested that increased exposure to mosquitoes increased the odds of heartworm prophylaxis failure among dogs in our study and is consistent with the findings of a previous study14 of risk factors for heartworm infection. The lack of association between risk factors for exposure to mosquitoes found in the univariate analyses, in contrast to the findings in the multivariate analysis, will assist with planning and analysis of data in future studies. Because 99% of dogs in our study were receiving heartworm preventives, the association of increased exposure to mosquitoes with failure of heartworm prophylaxis in the multivariate analysis is consistent with resistance of D immitis larvae at the L3-L4 stage to macrocyclic lactones or with lack of compliance. Although the present study did not directly address the role that lack of compliance plays in the odds of heart-worm infection, most published reports4–9 support the efficacy of various macrocyclic lactones and therefore the implication that compliance may be the predominant contributor to prophylaxis failure.
The association of prophylaxis failure with advice on preventive products and dose of heartworm preventives provided by a veterinarian may be attributable to a closer relationship between respondents that experienced failure of prophylaxis and their veterinarians, leading to increased compliance and testing. In the present study, respondents reporting prophylaxis failure were 3 times as likely to have a heartworm test performed at least once a year when compared with respondents reporting successful prophylaxis. Lack of testing may be associated with a perception that prophylaxis was successful, resulting in the true rate of failure to be underestimated. In addition, those reporting prophylaxis failure were significantly more likely to test for heartworm infection prior to changing products and to also have a second test performed after the product change.
Results of the multivariate logistic regression model indicated that residence south of the 37th parallel was independently and significantly associated with failure of heartworm prophylaxis. Dog owners reporting prophylaxis failure may be more likely to be familiar with the name of the heartworm test that was performed after discussing the situation with their veterinarian. Dogs kept outdoors for longer periods and those kept out at dusk, at dawn, or after dark were more likely to have successful prophylaxis. However, owners of dogs that were outdoors at dusk, at dawn, or after dark and also spent an increased number of hours outdoors during the mosquito season were more likely to report prophylaxis failure. In addition, respondents who had their dogs tested for heartworm infection less than once a year and also had the test performed prior to April 1 were more likely to report prophylaxis failure. Infrequent testing for heartworm infection and testing at the wrong time of year may lead to missed diagnoses or to the conclusion that heartworm prophylaxis has failed when infection may have been present prior to initiation of prophylaxis.
Because the geographic distribution of respondents resembled that of reported cases of heartworm infection in dogs, we believe the survey results can be generalized to hunting dog owners and handlers nationally. Limitations of the present study include the inability to generalize results to the entire population of dog owners and reduced power because of the relatively small number of respondents that reported prophylaxis failure when compared with the number reporting success. Respondents that experienced prophylaxis failure may have been highly motivated to participate in the study; therefore, the annual incidence of 9% for prophylaxis failure among survey respondents may be much greater than that in the population of club members and the population of dog owners at large. Members of the hunting dog club may have a greater financial investment and a closer relationship to their dogs because of inclusion of dogs in sporting activities and therefore greater motivation to seek information and comply with AHS recommendations for prevention of heartworm infection than the average dog owner. Membership in the hunting dog club also provided access to educational materials on heartworm prevention. There were several instances in which an important difference was detected between respondents reporting prophylaxis failure and those reporting success (eg, the association of year-round administration with successful heartworm prophylaxis); however, because of the relatively small number of respondents reporting failure in the previous 12 months or at any time in the past, these differences did not attain significance and the questions remain unanswered. We believe the results of our study can be extrapolated to dog owners, dog trainers, and kennel operators that are motivated to join a club, have easy access to information on heartworm prophylaxis, and therefore may be more knowledgeable about and compliant with heartworm prevention than the general population of dog owners.
Lack of regular testing and testing at the wrong time of year may lead owners and veterinarians to miss heartworm infection, associate infection with the wrong year or season, or associate the wrong preventive with infection. This is particularly problematic when the dog owner first starts a dog on a certain heartworm preventive or changes products after 1 or more years. Reducing the number of hours dogs spend outdoors during the hours of dusk, of dawn, or after dark, when mosquito activity is present, is recommended to reduce the odds of heartworm prophylaxis failure, whether attributable to lack of compliance in preventive administration or resistance of D immitis larvae to macrocyclic lactones. Additional studies are needed to assess the value of recommending year-round administration of heartworm preventives. The results of the present study do not directly address the possibility of resistance of the L3-L4 stage of D immitis larvae to macrocyclic lactones; however, our data identifies practices associated with suspected failure of heartworm prophylaxis. These practices, when addressed by veterinarians through client education, may reduce the likelihood of heartworm disease in dogs and improve the ability to identify the cause of suspected prophylaxis failure.
ABBREVIATIONS
| AHS | American Heartworm Society |
| CI | Confidence interval |
| IQR | Interquartile range |
| OR | Odds ratio |
MySQL [database online]. Uppsala, Sweden: MySQL AB, 2007. Available at: www.mysql.com. Accessed Dec 20, 2010.
Excel, Microsoft Office 2007, Microsoft Corp, Redmond, Wash.
SAS, version 9.2, SAS Institute Inc, Gary, NC.
References
- 1.↑
Hampshire VA. Evaluation of efficacy of heartworm preventive products at the FDA. Vet Parasitol 2005; 133:191–195.
- 2.↑
Snyder DEWiseman SCruthers LR, et al. Ivermectin and milbemycin oxime in experimental adult heartworm (Dirofilaria immitis) infection of dogs. J Vet Intern Med 2011; 25:61–64.
- 3.↑
Blagburn BLSpencer JANobles J, et al. American Heartworm Society Quarterly E-Bulletin. Heartworm preventive failures: resistance, clinic or client compliance. Available at: www.heartwormsociety.org/enewsletter/march2008/enewsletter_p=13.html. Accessed Dec 20, 2010.
- 4.
Blair LSWilliams EEwanciw DV. Efficacy of ivermectin against third-stage Dirofilaria immitis larvae in ferrets and dogs. Res Vet Sci 1982; 33:386–387.
- 5.
Clemence RGSarasola PGenchi C, et al. Efficacy of selamectin in the prevention of adult heartworm (Dirofilaria immitis) infection in dogs in northern Italy. Vet Parasitol 2000; 91:251–258.
- 6.
Genchi CPoglayen GKramer LH, et al. Efficacy of moxidectin for the prevention of adult heartworm (Dirofilaria immitis) infection in dogs. Parassitologia 2001; 43:139–141.
- 7.
Lok JBKnight DHNolan TJ, et al. Efficacy of an injectable, sustained-release formulation of moxidectin in preventing experimental heartworm infection in mongrel dogs challenged 12 months after administration. Vet Parasitol 2005; 128:129–135.
- 8.
Lok JBWashabau RJHeaney K, et al. Six-month prophylactic efficacy of moxidectin sustained release (SR) injectable for dogs against experimental heartworm infection in growing puppies. Vet Parasitol 2005; 133:233–241.
- 9.
McTier TLShanks DJWatson P, et al. Prevention of experimentally induced heartworm (Dirofilaria immitis) infections in dogs and cats with a single topical application of selamectin. Vet Parasitol 2000; 91:259–268.
- 10.↑
Prichard RK. Is anthelmintic resistance a concern for heartworm control? What can we learn from the human filariasis control programs? Vet Parasitol 2005; 133:243–253.
- 11.↑
American Heartworm Society. Diagnosis, prevention and management of heartworm (Dirofilaria immitis) infectrion in dogs. Available at: www.heartwormsociety.org/veterinary-resources/canine-guidelines.html. Accessed Dec 29, 2010.
- 12.
Glickman LTGrieve RBBreitschwerdt EB, et al. Serologic pattern of canine heartworm (Dirofilaria immitis) infection. Am J Vet Res 1984; 45:1178–1183.
- 13.
Selby LACorwin RMHayes HM Jr. Risk factors associated with canine heartworm infection. J Am Vet Med Assoc 1980; 176:33–35.
- 14.↑
Theis JHFranti CLambert L, et al. Risk factors for heartworm infection in dogs living in Sierra Nevada foothills and Sacramento Valley counties; public health implications. Calif Vet 1984; 38:13–17.
- 15.
Walters LL. Risk factors for heartworm infection in Northern California. In: Soil MDKnight DH, eds. Proceedings of the heartworm symposium ′95. Batavia, Ill: American Heartworm Society 1995;5–26.
- 16.↑
Knight DHLok JB. Seasonality of heartworm infection and implications for chemoprophylaxis. Clin Tech Small Anim Pract 1998; 13:77–82.
- 17.↑
Nelson CTMcCall JWRubin SB, et al. 2005 guidelines for the diagnosis, prevention and management of heartworm (Dirofilaria immitis) infection in dogs. Vet Parasitol 2005; 133:255–266.
