Feline heartworm disease: Solving the puzzle

Heartworm (Dirofilaria immitis) infection in cats was first reported more than 85 years ago.1 However, many cat owners and some veterinarians either remain unaware or do not think that heartworms can cause serious and sometimes fatal disease in cats. Most of us are familiar with the potential consequences of heartworm infections in dogs, but we fail to recognize that heartworm infections in cats can result in demonstrably different clinical responses (Table 1).2,3

A. Ray Dillon, DVM, MS, MBA, DACVIM

Byron L. Blagburn, MS, PhD

PREVALENCE IN CATS

Although the prevalence of heartworm infection in cats has been studied, unique features of feline infections make assessing the true prevalence difficult, if not impossible. A variety of techniques—including radiography, angiography, ultrasonography, and necropsy, as well as microfilariae, antibody, and antigen detection—have been used.4 The use of these different diagnostic methods makes comparison of the various studies difficult. Moreover, because many of these tests were developed for diagnosing adult heartworms in dogs, they may not be directly applicable in cats in which immature adults cause significant disease.

Table 1: Comparison of Canine and Feline Heartworm Infections

The nature of available tests and features of feline heartworm infections detailed in Table 1 probably result in grossly underestimated prevalences of heartworm infection in cats. Most heartworm researchers agree that exposure of cats to heartworm-infected mosquitoes is surprisingly high and that the risk of feline heartworm infection remains a concern in many regions of the country.5,6

CLINICAL MANIFESTATIONS IN CATS

Feline heartworm disease can differ markedly from its canine counterpart and may require several diagnostic tests or procedures to confirm. Although most cats infected with heartworms develop pathologic changes associated with infection, they remain asymptomatic. It is impossible to predict when and under what circumstances infected cats will develop clinical heartworm disease, and some cats will be clinically normal even with significant pulmonary pathology.

Cats with clinical heartworm disease usually present with respiratory signs such as coughing, dyspnea, or both, or intermittent vomiting not associated with eating. Signs can be limited to weight loss or diarrhea without accompanying respiratory signs. When present, the respiratory signs are similar to those observed with feline bronchial disease, which is frequently described as asthma by the owners. A small percentage of cats that develop clinical signs may die suddenly. This peracute presentation also mimics signs of acute dyspnea associated with feline asthma, cardiomyopathy, pleural diseases, or infections with other pulmonary parasites.4,6 Many of these cats are clinically normal before the acute event.2-7

Events in the developmental cycle of heartworm in cats are depicted in the boxed text titled "Events in the Heartworm Life Cycle in Cats." In the past, the signs of feline heartworm disease were usually attributed to the death of adult heartworms and the cat's unique pulmonary reaction to fragments of dead and dying adult worms. Dr. Dillon has, for years, forwarded the notion that immature worms may die in the lungs of infected cats before they mature to adults.6 These worms arrive in the heart 70 to 90 days after initial infection, and because of their small size (less than 1 in), they are carried by the pressure and direction of blood flow to the distal pulmonary arteries. The arrival and death of these immature worms in the lungs at 90 to 120 days after infection can result in coughing and dyspnea.

Events in the Heartworm Life Cycle in Cats

These same developmental and pathogenic events are less likely to occur in dogs, since virtually all immature heartworms mature to adult worms. Early death of immature heartworms at 90 to 180 days in cats and the resulting pulmonary disease have been referred to as the three-month disease cycle, to distinguish it from disease caused by adult worms (six-month or longer disease cycle).6

As mentioned previously, the acute respiratory signs thought to be associated with the death of immature heartworms in cats may mimic those of feline asthma. Pulmonary signs may be caused by the arrival of immature worms as early as 80 to 90 days after infection, which is three months before the heartworms begin to release antigen. Thus, during the initial disease process in cats, commonly used canine heartworm antigen tests, which test for an adult female heartworm antigen, cannot be used to confirm a diagnosis of heartworm disease. To further complicate the puzzle, accompanying pulmonary radiographic lesions are not specific for heartworm disease, and worms or fragments cannot be visualized by using readily available imaging techniques, especially in the presence of immature worms.6

These factors and the diagnostic problems that they impose served as our stimulus to develop a laboratory model of the three-month disease cycle and to demonstrate that marked lesions and disease can result from the death of immature heartworms in the lungs of cats.

SOLVING THE PUZZLE: A STUDY

The study protocol was reviewed and approved by the Auburn University Institutional Animal Care and Use Committee. To create the laboratory model, we acquired heartworm-naïve cats and separated them into three groups. The number of cats in each group was different because of the cats' availability. On Day 0, all cats were subcutaneously infected with 100 infective l3 larvae of D. immitis obtained from laboratory-maintained mosquitoes (Aedes aegypti). At necropsy, 10 cats were evaluated in Group 1, nine in Group 2, and 10 in Group 3.

Group 1 cats served as controls. As such, they received no chemo-intervention that would affect the subsequent development of the heartworms.

Group 2 cats were treated with ivermectin (Ivomec—Merial) orally at a dose of 150 µg/kg every two weeks beginning on study Day 84. We hypothesized that treatment with ivermectin beginning on Day 84 would target the early arriving L5 larvae in the lungs. We also hypothesized that continued administration of ivermectin to cats in this study group would mimic the natural death of developing heartworms in naturally infected cats.

Group 3 cats were given selamectin (Revolution—Pfizer Animal Health) topically at the label dose of 6 mg/kg monthly beginning 28 days after experimental infection and continuing monthly until the termination of the study. The purpose of the monthly treatment with selamectin was to prevent development of heartworms beyond the L4 stage and eliminate them before their arrival in the lungs. For study purposes, cats in this group would correlate to uninfected control cats and would corroborate the importance of monthly heartworm prevention.

HARD: A proposed name for lesions and disease associated with the death of immature heartworms in the lungs of cats

For the study, the cats were group-housed, were not stressed by handling, and were sedated for radiographic examination and other procedures required for data collection.

Data collection procedures

To evaluate the efficacies of treatments given to cats in Group 2 and Group 3 and to confirm the consequences of infection in all groups, numerous procedures were conducted on all cats throughout the study period of 240 days after infection. In addition to daily clinical observations, we collected blood samples for complete blood counts, serum chemistry profiles, and heartworm antigen and feline heartworm antibody detection. We also performed thoracic radiographic examinations and bronchoalveolar lavages at intervals to assess pulmonary pathologic changes and the development of vascular and airway disease. All cats were examined at necropsy about 252 days after experimental infection. All hearts and lungs were removed for gross and histologic evaluation.

Lung lesions (arteries, arterioles, capillaries, alveoli, bronchioles, bronchi, and tracheas) were assigned a lesion score of 0 to 3 with increasing severity. These methods were similar to those used in a previous study.8 Lesion scoring was performed under the guidance of an anatomic pathologist who was blinded to treatment group assignments. Multiple group lesion scores were analyzed by using the nonparametric Kruskal-Wallis test of variance, with specific differences detected by using Levene's test of error variances (p ≤0.05).

Study results

Some cats in Group 1 and Group 2 developed intermittent signs of heartworm infection (lethargy, depression, dyspnea) beginning three months after infection.* Intermittent signs in Group 1 cats continued throughout the second half of the study and were present in some cats up to and at the time of necropsy. This suggested that clinical responses were the result of the death of immature as well as adult heartworms and comparable to what is observed during natural infections. One cat in Group 2 developed acute dyspnea and died 120 days after infection. That cat had radiographic lesions similar to those of other cats in Group 2. Cats in Group 3 remained clinically normal.

Lesions typical of what was observed in the lungs of cats from the three treatment groups are presented in Figures 1-4. Lesion scores from lungs recovered from cats in each of the treatment groups were demonstrably different. Lung arterial lesions were present in Group 1 and 2 cats but were most severe in the Group 1 cats. However, lesions in the alveoli, bronchioles, and bronchi of Group 2 cats were equally as severe as in Group 1 cats. Alveoli, bronchi, and bronchioles from cats treated with selamectin (Group 3) had significantly lower lesion scores than in cats in Groups 1 and 2 and were generally normal in their appearance.

Figures 1A & 1B & 2A & 2B

Cats with abbreviated infections (Group 2) had demonstrable pulmonary vascular lesions, although they were not as severe as lesions observed in the control cats. Vascular morphometry findings in the selamectin-treated cats appeared normal. Statistical analysis supported the fact that lesion scores were significantly lower in selamectin-treated cats than in both nontreated cats and cats with abbreviated infections.

Figures 3A & 3B

Lesions typical of feline heartworm disease were not observed in cats in which selamectin was used as a preventive. Although seemingly intuitive, these studies again demonstrate the benefits of prevention, and there appears to be no evidence that immature heartworms reach the pulmonary vasculature.

Figures 4A & 4B

The number of heartworms and fragments recovered from untreated control cats confirms and corroborates the results of prior experimental infections. A mean of 4.3 live adult heartworms (range = 1 to 12) were recovered from eight Group 1 cats at necropsy. Six of these cats also harbored worm fragments (mean = 8.5 fragments; range = 2 to 13). Two cats harbored live adult worms but did not harbor fragments. Two cats did not harbor live adult worms but did harbor pulmonary fragments (one fragment and 12 fragments, respectively).

A cat in Group 2 died with respiratory disease presumably in response to immature worm death (one immature worm was present in the heart and two immature worms were present in the lungs at necropsy). No worm fragments of live adult heartworms were recovered from eight of the remaining nine cats in Group 2 cats at necropsy. One affected cat harbored one dead worm fragment attached to the tricuspid valve and one small degenerate worm in the lungs. These numbers strongly support the success of the strategy to kill immature worms in the lungs and the success of the laboratory model.

None of the cats in Group 3 harbored adult heartworms or any evidence of immature heartworms in the lungs.

Preliminary examination of radiographs suggests that radiographic findings correlate with histologic lesions in the lungs. Preliminary data based on thoracic radiographic findings in cats in which development of immature worms was interrupted (Group 2) reveal similarities between these cats and cats with other bronchial diseases, including feline asthma. Consequently, as noted earlier, early death of heartworms in cats is important both pathogenically and in its potential for misdiagnosis. Recall that in early infections, antigenic confirmation of heartworm would not be possible and that antibody responses inconsistently reflect adult heartworm infection. Radiographic and laboratory data (results of antigen and antibody tests, complete blood counts, serum chemistry profiles, and bronchoalveolar lavages) will be made available in a future report.

The results of our experimentally induced infections appear to correlate to a recent assessment of pulmonary lesions in uninfected and naturally infected cats subjected to necropsy examination.8 In that study, pulmonary arterial lesions (occlusive medial hypertrophy) were most severe in the caudal lung lobes of cats that were positive for the presence of adult worms and antibodies to developing stages of D. immitis (19/24 cats; 79%).8 Cats in that group would correlate to our experimentally infected control cats (Group 1). The inherent problems with the poor correlation of antibody testing, radiographic lesions, and the presence or absence of adult heartworms at necropsy have been demonstrated in other experimental infections in which radiographic lesions were present and no adult heartworms were present at necropsy.9 Pulmonary arterial lesions in the cats with spontaneous disease were also severe in cats without evidence of adult worms but with positive antibody responses (12/24; 50%).8 Cats in that group would correlate to our cats in which infections were abbreviated while in the lungs (Group 2). Pulmonary arterial lesions were least severe in cats that were negative for both adult worms and antibody responses (3/24; 13%).8 Those cats could correlate to our Group 3 cats, which were receiving selamectin. However, pulmonary arterial lesions due to other causes would likely be more prevalent in cats in the referenced study because those cats came from random sources and were more likely to be exposed to other potential pathogens. The comparative results of the two studies add further validity to our established laboratory model.

*These results are part of a comprehensive study assessing both short- and long-term effects of experimentally induced abbreviated and nonabbreviated heartworm infections in cats. The results of the short-term study are presented here. An additional three groups of cats were studied for 16 months after infection.

ACKNOWLEDGMENTS

The authors gratefully acknowledge the efforts and expertise of the following co-investigators and collaborators on this research: Dr. Michael Tillson; Dr. William Brawner, Jr.; Dr. Calvin Johnson; Dr. Jennifer Spencer; Dr. Elizabeth Welles; Dr. Patricia Rynders; Dr. Anthony Guarino; Ms. Jennifer Rainey; Ms. Sharron Barney; Ms. Sarah Majors; Ms. Elizabeth Files; Ms. Jamie Butler; Ms. Tracey Land; and Ms. Heather Stockdale. The authors also would like to thank Drs. Tom Nelson and Julie Levy for their review of the concept of HARD.

Byron L. Blagburn, MS, PhD

Department of Pathobiology

College of Veterinary Medicine

Auburn University

Auburn, AL 36849-5519

A. Ray Dillon, DVM, MS, MBA, DACVIM

Department of Clinical Sciences

College of Veterinary Medicine

Auburn University

Auburn, AL 36849-5519

REFERENCES

1. Riley WA. Dirofilaria immitis in the heart of a cat. J Parasitol 1922;9:48.

2. Blagburn BL. A review of common intestinal parasites of cats. In: The Veterinary CE Advisor: update on feline parasites. Vet Med 2000;95 (suppl):3-11.

3. Mannella C, Donoghue AR. Feline heartworm disease: facts and myths. Vet Forum 1999;16:50-65.

4. Atkins C. Heartworm disease: an update on testing and prevention in dogs and cats. Vet Med 1998;93 (suppl):3-13.

5. Dillon AR, Atkins CA, Barclay S. Feline heartworm disease: the rising concern about feline heartworm; roundtable on clinical strategies. Vet Forum 1997;14:62-69.

6. Dillon R. Feline heartworm disease: assessing the danger for owners. In: DVM Best Practices. DVM 2003;34 (suppl):23-26.

7. Miller MM. Feline dirofilariasis. Clin Techniques Small Anim Pract 1998;13:99-109.

8. Browne LE, Carter TD, Levy JK, et al. Pulmonary arterial disease in cats seropositive of Dirofilaria immitis but lacking adult heartworms in the heart or lungs. Am J Vet Res 2005;66:1544-1549.

9. Selcer AD, Newell SM, Mansour MS, et al. Radiographic and 2-D echocardiographic findings in 18 cats experimentally exposed to D. immitis via mosquito bites. Vet Radiol Ultrasound 1996;37:37-44.