Prevalence and Risk Factors for Heartworm Infection in Cats From Northern Florida

Introduction

Although cats are naturally resistant to Dirofilaria immitis (D. immitis), feline infection is likely to occur anywhere the parasite is found in dogs. When both cats and dogs are necropsied at animal shelters, worms can usually be found in cats at about 5% to 20% of the rate they are found in dogs in the same location.¹ The lack of highly reliable serological testing for cats has made it difficult to know the true prevalence of feline dirofilariasis or what proportion of infected cats are symptomatic. Significant clinical signs were common in 78 cats diagnosed with natural heartworm infection, including dyspnea (40%), vomiting (34%), coughing (29%), sudden death (19%), and neurological signs (14%).²³

Because of the lack of an adequate “gold standard” for diagnosing heartworm infection in cats, accurate determination of its prevalence is a more difficult task than in dogs. In dogs, gross examination of the heart and pulmonary arteries is considered the best method of identifying heartworm infection and is the method against which most other tests are compared.⁴ Because cats are more likely to develop aberrant migration of D. immitis, dissection of the heart and pulmonary arteries may fail to detect worms that have developed in other areas, such as the brain and body cavities.⁵⁶ In addition, immature worms in precardiac developmental stages are unlikely to be detected during routine necropsies. Thus, prevalence rates reported from necropsy studies in cats represent the minimum rates of infection.

Serological testing for heartworms is also more complicated in cats than in dogs. Heartworm antigen tests have decreased sensitivity when low heartworm burdens or only male heartworms are present, both of which occur commonly in feline infections.^5–8 Heartworm-specific antibodies may be found in up to a quarter of asymptomatic cats, suggesting that exposure to heartworms is relatively common in endemic areas.^679–13 The discrepancy between the presence of antibodies and negative necropsy findings may be explained by persistence of antibodies following natural recovery, aborted heartworm infection, or greater sensitivity of the antibody test to precardiac and extracardiac forms of infection compared to necropsy.⁶⁷

Very little has been reported on the host risk factors for naturally occurring heartworm infections in cats. The purposes of this study were to determine the prevalence of feline heartworm infection in an endemic region and to evaluate host risk factors for heartworm, such as gender, hair coat length, ownership status, and infection with the immunosuppressive retroviruses (i.e., feline leukemia virus [FeLV] and feline immunodeficiency virus [FIV]). The relationships between FeLV and FIV and other host factors were also examined.

Materials and Methods

Results

Of 630 cats examined, 254 (40.3%) were males and 376 (59.7%) were females. Only 46 of the males (18.1%) and 32 of the females (8.5%) were sterilized. Shorthaired cats (84.9%) were more common than medium- or longhaired cats (15.1%). Most of the cats were impounded strays (81.1%), contrasted with pets relinquished by their owners (18.9%). Five cats did not have data about their origins recorded, and they were excluded from the logistic regression analyses.

Thirty (4.7%) cats were seropositive for FeLV antigen, and 38 (6.1%) cats were seropositive for FIV antibody [Figure 1]. Male cats overall were at increased risk for both FeLV (P=0.0002) and FIV (P=0.0001) compared to females [Figure 2]. When gender was divided into four categories to include neuter status, there was still a significant association with FeLV (P=0.0002) and FIV (P=0.0009) due to an increased frequency for sexually intact males and a decreased frequency for intact females. Being shorthaired compared to medium- or longhaired was not significantly associated with FeLV (P=0.8) or FIV (P=0.5) status. Stray cats were not at higher risk than pets for FeLV infection (P=0.08) or for FIV infection (P=0.9).

Heartworms were identified at necropsy in 31 (4.9%) cats [Figure 1]. Single worms were found in 17 (54.8%) of the infected cats, two worms were found in eight (25.8%) cats, three worms were found in five (16.1%) cats, and four worms were found in one (3.2%) cat. Of the 52 recovered worms, 23 (44.2%) were males and 28 (53.8%) were females. Only nine (32.1%) of the female worms were gravid. One cat had a dead worm in the pulmonary artery that was too deteriorated to have its gender determined.

Positive antigen tests occurred in 45 (7.1%) cats, and 93 (14.8%) had at least two positive antibody tests. If all necropsy and serological results were combined, 107 (17.0%) cats had at least one criterion for infection with or exposure to heartworms [Figure 1]. Not all cats with heartworms confirmed at necropsy had positive antigen or antibody tests. Nine cats with heartworms found at necropsy (29%) had at least one false-negative antigen test, and four cats (13%) were negative on all antigen tests performed. Twenty-one cats with heartworms found at necropsy (68%) had at least one false-negative antibody test, and two cats (6%) were negative on all antibody tests.

In the logistic regression model examining the relationship between heartworm-positive necropsy results and FIV, gender, source, hair-coat length, and year of study, the only association was a trend toward females having a lower risk (P=0.05) of heartworms being found at necropsy than males (odds ratio, 0.5; 95% confidence intervals, 0.2, 1.01). Feline leukemia virus status could not be included in the logistic regression model, because no cats were both necropsy positive and FeLV positive. However, FeLV status was not significantly associated with infection when tested alone. For the model with a positive antigen test as the outcome, females were at significantly lower risk than males (P=0.03) to have heartworm antigen (odds ratio, 0.5; 95% confidence intervals, 0.3, 0.9). For the model with positive antibody tests as the outcome, there were no significant risk factors. The year of the study was significant, with cats sampled in year 2 having a lower frequency of heartworm antigen (P=0.03) than cats sampled in year 1. The year of the study had no effect on heartworms diagnosed at necropsy or on prevalence of heartworm antibodies.

Discussion

Currently, there is no ideal method for confirming heartworm infections in cats. Necropsy has long been considered a gold standard for postmortem diagnosis, but dissection of the heart and lungs is insensitive for immature infections and ectopic infections. Antigen testing is more sensitive for ectopic infections, but it has decreased sensitivity for low-worm burdens and male worms. Antibody tests are highly sensitive for immature infections and may detect ectopic infections, but they indicate only exposure to heartworms and not necessarily current infections.⁶⁷ Both heartworm antigen and antibodies remain in the circulation following clearance of heartworms; thus, infected cats may test positive for unpredictable lengths of time after recovering from infection. For these reasons, heartworm infection was evaluated in this study by several methods, including necropsy and testing for antigen and antibodies.

Adult heartworms were found in 31 (4.9%) cats, or one out of every 20 cats examined. To the authors’ knowledge, this is the largest reported study of necropsy-confirmed natural heartworm infection in cats. This is a similar infection rate compared to a series of necropsy surveys from animal shelters in the southeastern United States, which collectively identified 72 infected cats out of a total of 1,554 cats during the past 15 years, for an overall prevalence rate of 4.6% (range, 2.8% to 14.0%).^15–19 In the authors’ study, when heartworm infection was defined as the presence of antigen, the infection rate increased to 45 cats, or one in 14. Antibody test results indicated that 93 cats, or one in seven, were exposed to heartworms. Overall, one in six cats had at least one test consistent with heartworm infection or exposure. There was a significantly lower rate of heartworm antigen-positive cats in 2001 than in 1998, but there was no difference in necropsy findings or antibody-positive results, so the significance of this finding is unknown.

Of all the host characteristics examined in this study, only male gender was a significant risk factor for heartworm infection. Although results of antibody testing indicated that males and females had equal exposure to heartworms, necropsy and antigen testing indicated that males were slightly more likely to actually harbor heartworms. This suggests that male cats may have greater susceptibility to infection than female cats. This has been observed in other natural-infection studies²⁵¹² and in experimental-infection studies,²⁰²¹ in which male cats were more readily infected following deliberate exposure than female cats. However, another study of naturally infected client-owned cats found no difference in risk when the gender of infected cats was compared to that of the hospital population in general.³

Because the cat is naturally resistant to D. immitis, it would be logical to assume that immunosuppressed cats would be at increased risk of infection compared to healthy cats. Feline leukemia virus and FIV are immunosuppressive retroviruses with well-characterized immunopathological effects that increase susceptibility to a variety of opportunistic infections. However, this study failed to demonstrate any association between retroviral infection and co-infection with heartworms.

This study had limitations that should be considered before extrapolating the findings to owned pets. All of the cats examined had been admitted to a county shelter. Most of the cats were of unknown age or environmental history. Even the owned cats relinquished to the shelter may not be representative of most pet cats. The very act of relinquishment to a shelter may indicate a low level of attachment of owners to their cats. The low rate of sterilization (12.4%) in the relinquished cats compared to much higher rates reported for pet cats in general (84% to 86%)^22–24 also suggests that these cats may have received minimal investment and veterinary care. Thus, it is possible that the cats in this study were more likely to be minimally supervised, free-roaming, and exposed to mosquitoes than well-cared for indoor house cats. On the other hand, cats relinquished to shelters tend to be younger than the average population, so they may have had a lower cumulative period of risk compared to well-cared for pet cats. Regardless of the lifestyle of cats in this study, several studies have documented significant heartworm infection rates in heartworm endemic areas, despite whether the cats had access to the outdoors or were housed exclusively indoors.^2111214

Conclusion

The results of this study indicate that heartworm infection in adult cats in northern Florida is at least as common as infection with two other important feline pathogens, FeLV and FIV. Although diagnosis of the retroviruses is not usually as challenging as that of feline heartworm infection, all three infectious agents have in common limited treatment options, once infection is established, and the potential for deadly clinical outcomes. Of the three infections, dirofilariasis can be most consistently prevented by a variety of currently available products. The authors conclude that cats in this region should receive heartworm prophylaxis to prevent heartworm infection.

Acknowledgments

The authors acknowledge the staff of Alachua County Animal Services for technical assistance and IDEXX Laboratories, Synbiotics Corporation, SA Scientific Corporation, Antech Diagnostics, and Heska Corporation for contributing testing materials and services.