Abstract
Background
Dirofilaria immitis (causative agent of heartworm disease) is a mosquito-borne parasite that is endemic in canids internationally and can cause serious disease in dogs.
Objective
The objective of this study was to predict the earliest time for infective D. immitis exposure risk in multiple cities in Ontario. In addition, the potential impact of an increase in the daily mean temperature was evaluated.
Animals and procedure
The earliest date of D. immitis exposure risk for dogs within a calendar year was estimated for multiple cities in Ontario, based on the time to accumulation of 130 heartworm development units (HDUs).
Results
The earliest date of accumulation of 130 HDUs between 1996 and 2023 was May 25 (Windsor in 2012), and the latest was August 7 (Thunder Bay in 2004, Tobermory in 2014). The median date ranged from June 7 (Windsor) to July 18 (Thunder Bay). There were no significant changes in onset date over the study period for any location (all Q > 0.12).
When 1°C was added to the mean daily temperature, the date until 130 HDUs was reached decreased by a median of 4.5 d for Windsor (range: 2 to 12 d), 7 d for Sault Ste. Marie (range: 4 to 17 d), and 6 d for Kitchener-Waterloo (range: 3 to 11 d). This resulted in an earliest date of May 19 for Windsor, June 8 for Kitchener-Waterloo, and June 18 for Sault Ste. Marie.
Conclusion and clinical relevance
Described data suggest that current recommendations for heartworm prophylaxis in Ontario still provide abundant coverage for the estimated onset of risk, even when factors that could result in an earlier onset are considered.
RÉSUMÉ
Estimation du début du risque d’exposition à Dirofilaria immitis (ver du cœur) dans certaines villes de l’Ontario de 1996 à 2023
Contexte
Dirofilaria immitis (agent responsable de la dirofilariose) est un parasite transmis par les moustiques qui est endémique chez les canidés à l’échelle internationale et qui peut provoquer une maladie grave chez les chiens.
Objectif
L’objectif de cette étude était de prédire le moment le plus hâtif du risque d’exposition à D. immitis infectieux dans plusieurs villes de l’Ontario. De plus, l’impact potentiel d’une augmentation de la température moyenne quotidienne a été évalué.
Animaux et procédure
La date la plus hâtive du risque d’exposition à D. immitis pour les chiens au cours d’une année civile a été estimée pour plusieurs villes de l’Ontario, en fonction du temps d’accumulation de 130 unités de développement du ver du cœur (HDU).
Résultats
La date la plus hâtive d’accumulation de 130 HDU entre 1996 et 2023 était le 25 mai (Windsor en 2012) et la plus tardive était le 7 août (Thunder Bay en 2004, Tobermory en 2014). La date médiane s’échelonnait du 7 juin (Windsor) au 18 juillet (Thunder Bay). Il n’y a eu aucun changement significatif de la date de début au cours de la période d’étude pour aucun endroit (tous les Q > 0,12).
Lorsque 1 °C a été ajouté à la température quotidienne moyenne, la date à laquelle 130 HDU ont été atteintes a diminué d’une médiane de 4,5 jours pour Windsor (intervalle : 2 à 12 jours), de 7 jours pour Sault Ste. Marie (intervalle : 4 à 17 jours) et de 6 jours pour Kitchener-Waterloo (intervalle : 3 à 11 jours). Cela a donné lieu à une date la plus rapprochée, soit le 19 mai pour Windsor, le 8 juin pour Kitchener-Waterloo et le 18 juin pour Sault Ste. Marie.
Conclusion et pertinence clinique
Les données décrites suggèrent que les recommandations actuelles pour la prophylaxie contre le ver du cœur en Ontario offrent toujours une couverture abondante pour l’apparition estimée du risque, même lorsque des facteurs qui pourraient entraîner une apparition plus hâtive sont pris en compte.
(Traduit par Dr Serge Messier)
INTRODUCTION
Dirofilaria immitis (causative agent of heartworm disease) is a mosquito-borne parasite that is endemic in canids internationally and can cause serious disease in dogs. The life cycle involves ingestion of microfilariae (pre-L1 stage), typically from an infected wild or domestic canid, by a feeding female mosquito of a competent vector species. The parasite then develops through L1 and L2 stages, with subsequent molting into the infective L3 stage inside the mosquito’s Malpighian tubules. The L3 stage is then transmitted back to canids (and less commonly, other species such as felids) when the mosquito feeds again. Larval development in the mosquito is temperature dependent, taking 10 to 14 d at high temperatures and humidity (e.g., average daily temperature of 27°C and 80% humidity) but much longer at lower temperatures, with little or no development at or below average daily temperatures of 14°C (1–3).
This temperature dependency means heartworm risks vary seasonally in temperate climates, based on both when mosquitoes first appear in the environment and how long it takes before they can transmit parasites. The onset of transmission risk can be estimated by calculating heartworm development units (HDUs), the number of degrees by which the mean daily temperature exceeds 14°C (e.g., a mean temperature of 17°C would provide 3 HDUs for that day) (2). Development of D. immitis to the infective L3 stage is based on accumulation of 130 HDUs (2,4), so determination of the earliest day in a year when 130 HDUs is accumulated can be used as a proxy for heartworm transmission risk.
Heartworm control in dogs typically focuses on monthly administration of preventive medications. Currently available monthly preventives, on label, kill parasites up to 4 wk of age in dogs (the L3 and early L4 stages) (5); therefore, starting administration within 1 mo of the onset of risk of exposure should be effective. In work from the 1980s that used climate data from 1957 to 1986 (2), the earliest risk of heartworm transmission was calculated using HDUs for locations across Ontario and concluded that the earliest transmission occurred in Windsor (June 1). Furthermore, although the start date was later for all other risk areas in Ontario, June 1 was used as the start date for all risk areas, to avoid confusion. Thus, because of the 1-month reach back activity, the first monthly heartworm preventive did not theoretically need to be given until July 1. However, heartworm prevention guidance for Ontario has typically recommended starting heartworm prophylaxis on June 1 (6), which would cover exposure starting at the beginning of May. But as climate changes, alterations in vector-borne diseases have occurred and will continue (7), raising concerns about whether current practices reflect the current risks of heartworm transmission (6).
The objective of this study was to predict the earliest time for infective D. immitis exposure risk in multiple cities in Ontario, based on daily temperature data and HDUs. In addition, the potential impact of an increase in the daily mean temperature was evaluated.
MATERIALS AND METHODS
Daily mean temperature data were obtained online from the Government of Canada (https://climate-change.canada.ca/climate-data/#/daily-climate-data). A convenience sample of locations distributed across Ontario was chosen (Figure 1). Data for years 1996 to 2023 were collected when available. If data were lacking for specific days within a year, they were obtained from an alternate site in the same city, when available. Otherwise, the mean temperature of the preceding and following days was used. However, if data were unavailable for more than 10 d of the year (excluding January, February, November, and December), that year was excluded from the dataset.
FIGURE 1.
Locations of cities in Ontario where date of accumulation of 130 heartworm development units was studied.
Heartworm development units were calculated by subtracting 14 from the mean daily temperature (degrees Celsius). Negative values were reported as 0. The date at which a sum of 130 HDUs was accumulated was determined for each city and year.
The median and range of the time to accumulation of 130 HDUs were described. Linear regression was used to evaluate the effect of year on date to accumulation of 130 HDUs. Because of the number of cities that were evaluated and the potential for Type I error, P-values were adjusted to account for false discovery rate using the Benjamini-Hochberg correction, with resulting Q values < 0.05 indicating statistical significance. Data were analyzed using JMP 17 software (SAS Institute, Cary, North Carolina, USA).
To estimate the effect of climate change, HDUs were recalculated for a subset of cities after adding 1°C to the daily mean temperature. This temperature increase was chosen as an example of the impact from short-term effects of climate change, based on the reported increase in mean spring temperatures in Canada of 1.6°C between 1948 and 2022 (8).
RESULTS
Median and range estimates for the earliest date in a year when 130 HDUs had accumulated are reported in Table 1 and Figures 2 and 3. The earliest date of accumulation of 130 HDUs between 1996 and 2023 was May 25 (Windsor in 2012), and the latest was August 7 (Thunder Bay in 2004, Tobermory in 2014). The median date ranged from June 7 (Windsor) to July 18 (Thunder Bay). Before adjustment for false discovery rate, there was a significant effect of time for Toronto (decreasing date over time, P = 0.011) and Kingston (increasing date over time, P = 0.039). However, after adjustment, there were no associations between year and time until accumulation of 130 HDUs for any location in Ontario over the study period (all Q > 0.13).
TABLE 1.
Median and range of date of accumulation of 130 heartworm development units in selected cities in Ontario, with indication of the earliest and latest dates for data from 1996 to 2023.
| City | Year range | Date of accumulation of 130 heartworm development units | ||
|---|---|---|---|---|
|
| ||||
| Earliest | Median | Latest | ||
| Kingston | 1996 to 2023 (excl 2002, 2008) | June 10 (1999) | June 26 | July 9 (2009) |
| Kitchener-Waterloo | 1998 to 2023 | June 10 (2010) | June 23 | July 8 (1997) |
| London | 1996 to 2022 (excl 2002) | June 4 (2010) | June 18 | July 1 (2019) |
| Orillia | 1996 to 2023 (excl 2008, 2012 to 2014) | June 10 (2021) | June 19 | July 6 (2015) |
| Ottawa | 1996 to 2023 | June 5 (2010) | June 19 | July 2 (2019) |
| Peterborough | 1996 to 2023 | June 12 (1999) | June 25 | July 11 (2009) |
| Sault Ste. Marie | 1996 to 2023 (excl 2012) | June 25 (2007) | July 16 | August 4 (1996) |
| Sudbury | 1996 to 2022 | June 15 (2010) | July 1 | July 24 (2000) |
| Thunder Bay | 1996 to 2022 | July 2 (2021) | July 18 | August 7 (2004) |
| Tobermory | 2007 to 2023 | July 3 (2020) | July 15 | August 7 (2014) |
| Toronto | 1996 to 2023 | May 30 (2010) | June 14 | June 26 (2019) |
| Welland | 1996 to 2023 (excl 2006) | June 12 (2012) | June 25 | July 4 (2009) |
| Windsor | 1996 to 2023 | May 25 (2012) | June 7 | June 24 (1997) |
FIGURE 2.
Dates of accumulation of 130 heartworm development units in selected cities in Ontario, with trend lines, from 1996 to 2023.
FIGURE 3.
Quantile box plot of the dates of accumulation of 130 heartworm development units in selected cities in Ontario from 1996 to 2023.
When 1°C was added to the mean daily temperature, the date until 130 HDUs was reached decreased by a median of 4.5 d for Windsor (range: 2 to 12 d), 7 d for Sault Ste. Marie (range: 4 to 17 d), and 6 d for Kitchener-Waterloo (range: 3 to 11 d). This resulted in an earliest date of May 19 for Windsor, June 8 for Kitchener-Waterloo, and June 18 for Sault Ste. Marie (Figure 4).
FIGURE 4.
Effect of a 1°C increase in the mean daily temperature on the date of accumulation of 130 heartworm development units (HDUs) in Windsor and Sault Ste. Marie, Ontario, using data from 1996 to 2023.
DISCUSSION
These data indicate that the predicted onset of heartworm exposure risk can vary by more than 1 mo, both within locations year-to-year and between locations in Ontario. Although the date to accumulation of 130 HDUs within a calendar year was typically earlier than in a study from 1989 that assessed regional transmission periods in Ontario (2), even the earliest date in the warmest location was still well within the timeframe that is recommended for heartworm prophylaxis in Ontario (i.e., first treatment on June 1) (6).
The earliest date of accumulation of 130 HDUs was May 25, in Windsor (in 2012), which historically was used to provide heartworm prevention recommendations for Ontario as it had the warmest climate (2). Used as per label recommendations, all heartworm monthly preventive treatments kill parasites up to 4 wk of age in dogs (the L3 and early L4 stages). Therefore, administration within 1 mo of the onset of risk of transmission should be effective. Accordingly, the recommended June 1 start date for prophylaxis (6) is well within the adequate time period. Transmission risk would have to start before May 2 for there to be risk, and no locations in Ontario had earliest transmission risk dates close to that date.
Although there is good evidence supporting the temperature and time dependency of larval maturation in mosquitoes (2,4), the 130 HDU threshold is not absolute. An experimental study of D. immitis development in Aedes aegypti (a mosquito species not currently established in Canada) generally supported this threshold, as there was a marked increase in detection of L3 D. immitis in Malpighian tubules of infected mosquitoes after 130 HDUs (4). However, L3 larvae developed before accumulation of 130 HDUs in some mosquitoes, particularly at higher temperatures. At 19°C, the first L3 were detected after 125 HDUs, indicating some potential for earlier infectivity, but decreasing the threshold to 125 HDUs would typically only shorten the time to risk of infective larvae being present by 1 to 2 d in locations from the current study (data not presented).
In earlier work, calculation of the accumulation of HDUs has been done in 2 ways: totalling HDUs until 130 is reached, or doing that over a rolling 30-day period (4,9,10). The latter is based on an assumption that adult mosquitoes only live for a maximum of 30 d, though this cannot be stated with confidence, particularly given the range of different potential mosquito vector species (10). Therefore, we chose not to use a 30-day rolling window to avoid underestimating the onset of risk should some infected mosquitoes live for more than 30 d.
The HDU concept has scientific underpinning; however, it is still unclear how precisely it reflects natural risk under highly variable natural conditions with complex climatic conditions and different mosquito species and D. immitis strains. Also, it has been demonstrated that using average daily temperature may underestimate HDU accumulation when compared to hourly measurements (4). That study, based on data from Ithaca, New York (USA), predicted the onset of risk for 2012 to be 7 d earlier for 2012 and 2 d earlier for 2013 when using hourly versus daily data. Hourly data were not available for the current study. It is reasonable to assume that the risk could start somewhat sooner than the current data indicated. However, this would still be well within the timeframe for effective heartworm prophylaxis using a June 1 starting date for administration of a monthly preventive dose. In addition, temperatures at weather stations do not account for microclimates (e.g., sewers, inside structures) and heat sinks that would result in exposure of mosquitoes to higher temperatures and facilitate faster development of D. immitis larvae, again underestimating the onset of risk. The effect of this in the field is unknown but should not be ignored. However, at present, it does not appear to be a significant, clinically relevant issue in Ontario.
Focusing on HDUs also does not consider the potential for overwintering of mosquitoes harboring infective larvae or larvae with arrested development. The effect of this is unknown as it has not been studied in Canada or in similar areas. Although microfilariae can survive freezing (11), the overall risk to the Ontario dog population is probably limited, as the prevalence of infection in Ontario mosquito populations is presumably low based on the low prevalence of heartworm in Ontario dogs (12). As factors such as overwintering and microclimates would likely contribute to limited risk of exposure outside of the predicted risk period, these unknown aspects preclude any zero-risk determinations. Thus, there are multiple situations in which risk could theoretically exist before the 130 HDU threshold is first attained in a year. However, these are assumed to be of minor clinical significance (13). Thus, monitoring HDUs likely indicates the periods of greatest or most realistic risk and allows for objective comparison between regions and over time, but should not be considered an indicator of absolute risk.
This study focused on the onset of the exposure risk period. The end of the potential transmission period was not calculated in this study because of a lack of adequate evidence to define it. Earlier work with Canadian climate data noted that the transmission season for a year ends on the latest day that 130 HDUs occur within the previous 30 d (9) Not only has this not been adequately investigated, but it is based on the assumption that mosquitoes only live for a maximum of 30 d — an incorrect assumption for all potential vector species in Ontario. For example, Aedes vexans, a common mosquito species in Ontario (14) and a known D. immitis vector (15), can survive for up to 11 wk (16), albeit with poorer survival in cooler conditions. Therefore, although HDU determination might identify the last day when heartworm larvae could have developed into infective larvae, it does not necessarily indicate the end of the transmission period, because that is also dependent on the lifespan of the adult female mosquito harbouring infective larvae.
Climate change is altering various vector ranges and disease transmission risks (7). As temperatures increase, an earlier onset of heartworm risk is expected. Numerical decreases in time to reach 130 HDUs were identified for most regions evaluated across Ontario, but these were not statistically significant over the period from 1996 to 2023. There is abundant year-to-year variation, so the lack of a significant difference is perhaps not surprising over a relatively short time span. Adding 1°C to the daily mean temperature had limited effect on the onset of risk, particularly compared to the broader, year-to-year variation. Therefore, short-term effects on the risk period are presumably limited and should not alter current recommendations. However, the predicted acceleration of climate change will likely have more profound effects over time. Further modelling of the impact of climate change over a broader range of temperature changes is indicated.
In summary, described data suggest that current recommendations for heartworm prophylaxis in Ontario still provide abundant coverage for the estimated onset of risk, even when factors that could result in an earlier onset are considered. Historically, a single starting date for heartworm transmission has been quoted for Ontario (June 1) and is based on examination of climate data from 1957 to 1986 (2) for Windsor, Ontario, where transmission starts earliest in the province. In this study, examination of climate data for 1996 to 2023 indicated that transmission still began earliest in the Windsor area, and that, despite real concerns about the impact of climate change on vector-borne diseases, current heartworm prophylaxis recommendations should still cover the main risk period throughout Ontario. CVJ
Footnotes
Copyright is held by the Canadian Veterinary Medical Association. Individuals interested in obtaining reproductions of this article or permission to use this material elsewhere should contact Permissions.
REFERENCES
- 1.Kartman L. Factors influencing infection of the mosquito with Dirofilaria immitis (Leidy, 1856) Exp Parasitol. 1953;2:27–78. [Google Scholar]
- 2.Slocombe JOD, Surgeoner GA, Srivastava B. Determination of the heartworm transmission period and its use in diagnosis and control. Proceedings of the Heartworm Symposium. 1989:19–26. [Google Scholar]
- 3.Kutz FW, Dobson RC. Effects of temperature on the development of Dirofilaria immitis (Leidy) in Anopheles quadrimaculatus Say1 and on vector mortality resulting from this development. Ann Entomol Soc Am. 1974;67:325–331. [Google Scholar]
- 4.Ledesma N, Harrington L. Fine-scale temperature fluctuation and modulation of Dirofilaria immitis larval development in Aedes aegypti. Vet Parasitol. 2015;209:93–100. doi: 10.1016/j.vetpar.2015.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.AHS. Current Canine Guidelines for the Prevention, Diagnosis, and Management of Heartworm (Dirofilaria immitis) Infection in Dogs. Holly Springs, North Carolina, USA: American Heartworm Society; 2020. [Last accessed October 1, 2024]. Available from: https://d3ft8sckhnqim2.cloudfront.net/images/pdf/AHS_Canine_Guidelines_11_13_20.pdf?1605556516. [Google Scholar]
- 6.CPEP. Canadian Parasitology Expert Panel Guidelines for the Management of Parasites in Dogs and Cats. Saskatoon, Saskatchewan: Canadian Parasitology Expert Panel; 2009. [Last accessed October 1, 2024]. Available from: https://research-groups.usask.ca/cpep/documents/cpep-booklet.pdf. [Google Scholar]
- 7.Ludwig A, Zheng H, Vrbova L, Drebot MA, Iranpour M, Lindsay LR. Increased risk of endemic mosquito-borne diseases in Canada due to climate change. Can Commun Dis Rep. 2019;45:91–97. doi: 10.14745/ccdr.v45i04a03. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Government of Canada [Internet] Temperature change in Canada. [Last accessed October 1, 2024]. [updated July 25, 2024]. Available from: https://www.canada.ca/en/environment-climate-change/services/environmental-indicators/temperature-change.html.
- 9.McGill E, Berke O, Peregrine AS, Weese JS. Epidemiology of canine heartworm (Dirofilaria immitis) infection in domestic dogs in Ontario, Canada: Geographic distribution, risk factors and effects of climate. Geospat Health. 2019;14 doi: 10.4081/gh.2019.741. [DOI] [PubMed] [Google Scholar]
- 10.Slocombe JOD, Srivastava B, Surgeoner GA. The transmission period for heartworm in Canada. Proceedings of the American Heartworm Society Conference. 1995:43–48. [Google Scholar]
- 11.Fortin JF, Slocombe JOD. Temperature requirements for the development of Dirofilaria immitis in Aedes triseriatus and Ae. vexans. Mosquito News. 1981;41:625–633. [Google Scholar]
- 12.McGill E, Berke O, Weese JS, Peregrine A. Heartworm infection in domestic dogs in Canada, 1977–2016: Prevalence, time trend, and efficacy of prophylaxis. Can Vet J. 2019;60:605–612. [PMC free article] [PubMed] [Google Scholar]
- 13.Slocombe JOD. Heartworm in Canada in 2010 with comments on Ontario. Vet Pract Newslett. 2011 March;:31. [Google Scholar]
- 14.Giordano BV, Turner KW, Hunter FF. Geospatial analysis and seasonal distribution of west Nile virus vectors (Diptera: Culicidae) in southern Ontario, Canada. Int J Environ Res Public Health. 2018;15:614. doi: 10.3390/ijerph15040614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hendrix CM, Bemrick WJ, Schlotthauer JC. Natural transmission of Dirofilaria immitis by Aedes vexans. Am J Vet Res. 1980;41:1253–1255. [PubMed] [Google Scholar]
- 16.Costello RA, Brust RA. Longevity of Aedes vexans under different temperatures and relative humidities in the laboratory. J Econ Entomol. 1971;64:324–325. [Google Scholar]