- Email: [email protected]
Canine dirofilariasis in the region of Conceição Lagoon, Florianópolis, and in the Military Police kennel, São José, State of Santa Catarina, Brazil
Veterinary Parasitology 113 (2003) 239–242
Canine dirofilariasis in the region of Conceição Lagoon, Florianópolis, and in the Military Police kennel, São José, State of Santa Catarina, Brazil R.T. Araujo a , C.B. Marcondes a,∗ , L.C. Bastos b , D.C. Sartor c a
Departamento Microbiologia Parasitologia, CCB, UFSC, Florianópolis (SC) 88040-900, Brazil b Coord. Pós-graduação Farmacologia, CCB, UFSC, Florianópolis (SC), Brazil c Cl´ınica Lovely Dog, Florianópolis (SC), Brazil Received 5 May 2002; received in revised form 25 January 2003; accepted 10 February 2003
Abstract Eighty dogs from a locality west of the Conceição Lagoon, in the Santa Catarina Island, Florianópolis Municipality, State of Santa Catarina, Brazil, were examined for dirofilariasis by thick smear and by modified Knott’s method. Twelve of them (15%) were positive, five only by Knott’s, two only by thick smear and five by both methods. Positive dogs were observed in several parts of the locality, indicating that the parasite is widespread. Prevalence was significantly higher in older animals. Size, sex, hair length, place of sleeping and hour of blood collection did not significantly influence the results. Forty dogs of the Military Police kennel, at São José Municipality, Santa Catarina State, were all negative. Studies on the transmission and control of dirofilariasis in the region should be developed. © 2003 Elsevier Science B.V. All rights reserved. Keywords: Dirofilariasis; Dirofilaria immitis; Dog; Brazil; Lung; Santa Catarina
1. Introduction Canine dirofilariasis is potentially a very serious infection of dogs and other canids, occurring also in cats and in humans. It is widely distributed in Brazil; canine infection has been studied mostly in coastal areas, but it can also be found in other regions (Labarthe et al., 1999). General prevalence in Brazil is 9.2% (Ahid et al., 1999), but prevalence may vary in the Brazilian States from 1.1% in Rio Grande do Sul to 12.5% in Alagoas, and 12% in Santa Catarina (Labarthe et al., 1997a,b). Prevalence is 43 and 46% in two localities in São Lu´ıs, Maranhão State (Ahid et al., 1999). ∗ Corresponding author. Fax: +55-48-331-9258. E-mail address: [email protected] (C.B. Marcondes).
0304-4017/03/$ – see front matter © 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0304-4017(03)00077-3
240
R.T. Araujo et al. / Veterinary Parasitology 113 (2003) 239–242
Prevalence of canine dirofilariasis was studied in the neighborhood of Conceição Lagoon, a tourist region of Santa Catarina Island, in southern Brazil. Venous blood was obtained from 80 dogs with defined residence, west of the Conceição Lagoon. The Lagoon is situated in the center of the Santa Catarina Island, Florianópolis Municipality. The co-ordinates of extremes of the area, obtained by GPS 45 (GarminTM ) varied from 27◦ 34 56.1 S 48◦ 27 30.3 W to 27◦ 36 11.1 S 48◦ 27 46.7 W, and 27◦ 35 52.4 S 48◦ 28 02.9 W to 27◦ 35 54.8 S 48◦ 27 48.5 W. Houses usually have several fruit trees in their backyards, and the higher parts of the area are forested. Blood was also obtained from 40 dogs of the Military Police kennel, in the nearby São José Municipality (27◦ 36 55 S 48◦ 37’39 W, 8 m a.s.l.), on the mainland. The area near the kennel is plain and, besides some eucalyptus trees, has no vegetation, and is exposed to strong winds. In both studies, only dogs older than 6 months and not treated for dirofilariasis in the last 12 months were studied. Blood of each dog was used for preparing one-thick smear by spreading it in about 1 cm2 using another slide. Preparations were exposed to distilled water and stained by the Giemsa method. Another sample (1 ml) was processed by modified Knott method (Georgi et al., 1990). Briefly, the sample was mixed with 15 ml of EDTA. In the laboratory, this material was gently mixed with 10 ml of formaline at 2%(2 ml of formaldehyde at 37% in 98 ml of distilled water), to break the erythrocytes and fix the leukocytes and the microfilariae. After 2–3 min, the material was centrifuged for 5 min, at 1500 rpm, discarding the supernatant. Thirty-five microliter of the sediment was transferred to each of three slides, preparing three smears of similar sizes, staining them by Giemsa method, and calculating the number of microfilariae per milliliter. Microfilariae from all positive dogs were examined, to differentiate them from those of Dipetalonema sp. Ten microfilariae, randomly chosen, were measured, to check the specific identification. In the Conceição Lagoon region, 12 of the 80 dogs were positive (15%); five of them were positive in both the tests (6.25%), five only by modified Knott’s (6.25%) and two only in thick smear (2.5%). None of the 40 dogs of the kennel of the Military Police was positive. The absence of dirofilariasis in the dogs of the kennel is probably due to the conditions of the terrain. They are not suitable for mosquito breeding. Neither the curved tail nor the cephalic hook, which characterize microfilariae of Dipetalonema sp. and differentiate them from those of Dirofilaria immitis (described by Leidy in 1856, Hoffmann, 1987), was observed in any of the 976 microfilariae, from all positive dogs. The morphometric study of 10 microfilariae showed that the length varied from 212.8 to 289 m (mean length = 242.9 ± 28.9 m) and the width varied from 2.5 to 6 m (mean width = 3.85 ± 1.2 m). Age of the animals, analyzed by χ2 , significantly influenced the infection index; it was higher in older dogs (Table 1). No significant influence of sex, time of blood collection (morning versus afternoon), place where dogs slept (out versus in the houses), hair length (short versus median versus long) or breed (defined breed versus mongrel) on infection index was observed (Table 2). The distribution of the positive dogs over the locality indicates that no condition of the terrain seems to influence the infection. Infections were distributed over all the area. Steindel (1999, personal communication), studying Chagas disease in the most southern
R.T. Araujo et al. / Veterinary Parasitology 113 (2003) 239–242
241
Table 1 Influence of the age of the dogs on the prevalence of dirofilariasis in animals from Florian´opolis, Santa Catarina, Brazil Age classes (years) 1–4 5–8 >8 Total
Positive 4 (7.8%) 5 (23.8%) 3 (37.5%) 12 (15%)
Negative
Total
47 (92.2%) 16 (76.2%) 5 (62.5%)
51 (63.8%) 21 (26.3%) 8 (10%)
68 (85%)
80 (100%)
χ2 = 6.36; χ2 = 5.99 (2 d.f.; 5%).
locality in the western edge of the Lagoon (“Canto da Lagoa”), in 1980s, observed a D. immitis prevalence of 23% in the dogs. Modified Knott’s method seems to be better than thick smear. However, since two dogs were positive only by the last method, it should not be discarded. The examination of more material (2–3 slides) in the Knott’s method may improve its efficiency and should be evaluated. The 57.1% of dogs from Recife (Pernambuco State) with adult worms of D. immitis in the heart could not be detected by the Knott method (Alves et al., 1999); this indicates the use of more than one method for the dogs is advised. However, the low efficiency of the Knott’s method in the dogs from Recife was attributed to the use of microfilaricide drugs (Alves et al., 1999), which were not used in the dogs of our study, as far as informed by their owners. Further studies on the efficiency of these rather simple methods should be developed. Even though serology is probably more sensitive than the parasitological techniques, the latter is simpler, can be used in any laboratory, and permits an examination of the morphology of the microfilariae. Additionally, there is no risk of false positive. In the future, there may be problems with false negative results with test that detect microfilariae and veterinarians in this area will need to use antigen tests. Comparing the dimensions of the microfilariae with those of previous studies (Flynn, 1973; Georgi et al., 1990; Hoffmann, 1987) indicates that the larvae from Florianópolis are in the lower part of the range. N. Labarthe (2000, personal communication) does not usually measure larvae, observing only their morphology. An increase in prevalence from 1 to 5 years of age was observed in dogs in Padua, Italy (Capelli et al., 1996); this result is similar to those of the present study. Analyzing the data of Alves et al. (1999), a significantly greater proportion of male dogs with microfilariae could Table 2 Values of χ2 for the influence of some features of the dogs on the prevalence of dirofilariasis in animals from Florian´opolis, Santa Catarina, Brazil Features of the dogs
Calculated χ2
From the table χ2 (d.f.) (5%)
Sex Time of blood collection (morning vs. afternoon) Sleep (out vs. in the house) Hair length (short vs. median vs. long) Breed (defined breed vs. mongrel)
1.76 0.02 0.02 0.88 2.87
3.84 (1) 3.84 (1) 3.84 (1) 5.99 (2) 3.84 (1)
242
R.T. Araujo et al. / Veterinary Parasitology 113 (2003) 239–242
be observed (χ2 = 4.42, cf. χ12 d.f.; 5% = 3.84). However, this number included dogs with microfilariae of Dipetalonema reconditum (Grassi, 1890) as well as of Dirofilaria immitis. The analysis of numbers of dogs infected with each worm did not indicate significant results (χ2 = 3.44 and 0). The periodicity of the microfilariae in the peripheral blood, although not significant in the dogs of the present study, should be more thoroughly studied, due to previous observation of periodicity (e.g. Nogami et al., 2000). The periodicity, which seems to vary in different regions, could furnish further information on the efficiency of potential vectors and in the choice of the time for examination of the dogs. The high prevalence of dirofilariasis in the locality dictates that the population is informed and that dogs are inspected and treated. The area, besides being densely populated, receives many tourists in the summer, mostly from nearby states and from Argentina. The prevalence of dirofilariasis in dogs from Argentina can be as high as 60% in localities near Buenos Aires (Grubissich, 1999). Since most Argentinean tourists at Florianópolis come from the Buenos Aires region and many of them transport their dogs, there is a great probability of circulation of the parasite between both the regions in summer. The sanitary inspection of dogs in the frontier should be enforced. Also, dirofilariasis should be considered in cases of unexplained human pulmonary nodules in the region of Florianópolis. Acknowledgements To Dr. Lawrence Hribar (Keys Mosquito Control District, Florida), for checking the English language and commenting on the manuscript. To Dr. Norma Labarthe (UERJ, Rio de Janeiro) for comments on the work. References Ahid, S.M.M., Lourenço-De-Oliveira, R., Saraiva, L.Q., Dirofilariose Canina na Ilha de São Lu´ıs, Nordeste do Brasil: Uma Zoonose Potencial. Cad. Saúde Públ., Rio de Janeiro, 15, 405–412. Alves, L.C., Silva, L.V.A., Faustino, M.A., McCall, J.W., Supakonderj, P., Labarthe, N.W., Sanchez, M., Caires, O., 1999. Survey of canine heartworm in the city of Recife, Pernambuco, Brazil. Mem. Inst. Oswaldo Cruz 94, 587–590. Capelli, G., Poglayen, G., Bertotti, F., Giupponi, S., Martini, M., 1996. The host–parasite relationship in canine heartworm infection in a hyperendemic area of Italy. Vet. Res. Commun. 20, 320–330. Flynn, R.J., 1973. Parasites of Laboratory Animals, vol. 1. The Iowa State University Press, Ames. 442 pp. Georgi, J.R., Georgi, M.E., Theodorides, V.J., 1990. Parasitology for Veterinarians, 5th ed. W.B. Saunders Company, Philadelphia. 412 pp. Grubissich, J.A., 1999. Dirofilariasis canina. Holliday News 2, 8–12. Hoffmann, R.P., 1987. Diagnóstico de Parasitismo Veterinário. Porto Alegre, Sulina. 156 pp. Labarthe, N.V., Araújo, A.M., Bordin, E.L., Larsson, M.E., Guerreiro, J., 1997a. Update on the distribution of Dirofilaria immitis in dogs in Brazil. In: Proceedings of the XVIISA-VA World Congress, Roma, Italy. pp. 287–289. Labarthe, N., Almosny, N., Guerrero, J., Duque-Araújo, A.M., 1997b. Description of the occurrence of canine dirofilariasis in the State of Rio de Janeiro Brazil. Mem. Inst. Oswaldo Cruz 92, 47–51. Labarthe, N., Lourenço-De-Oliveira, R., Serrão, M.L., 1999. A transmissão da Dirofilaria immitis no Rio de Janeiro. Vetores Pragas 2, 20–25. Nogami, S., Murasugi, E., Shimazaki, K., Maeda, R., Harasawa, R., Nakagaki, K., 2000. Quantitative analysis of microfilarial periodicity of Dirofilaria immitis in cats. Vet. Parasitol. 92, 227–232.
Canine dirofilariasis in the region of Conceição Lagoon, Florianópolis, and in the Military Police kennel, São José, State of Santa Catarina, Brazil R.T. Araujo a , C.B. Marcondes a,∗ , L.C. Bastos b , D.C. Sartor c a
Departamento Microbiologia Parasitologia, CCB, UFSC, Florianópolis (SC) 88040-900, Brazil b Coord. Pós-graduação Farmacologia, CCB, UFSC, Florianópolis (SC), Brazil c Cl´ınica Lovely Dog, Florianópolis (SC), Brazil Received 5 May 2002; received in revised form 25 January 2003; accepted 10 February 2003
Abstract Eighty dogs from a locality west of the Conceição Lagoon, in the Santa Catarina Island, Florianópolis Municipality, State of Santa Catarina, Brazil, were examined for dirofilariasis by thick smear and by modified Knott’s method. Twelve of them (15%) were positive, five only by Knott’s, two only by thick smear and five by both methods. Positive dogs were observed in several parts of the locality, indicating that the parasite is widespread. Prevalence was significantly higher in older animals. Size, sex, hair length, place of sleeping and hour of blood collection did not significantly influence the results. Forty dogs of the Military Police kennel, at São José Municipality, Santa Catarina State, were all negative. Studies on the transmission and control of dirofilariasis in the region should be developed. © 2003 Elsevier Science B.V. All rights reserved. Keywords: Dirofilariasis; Dirofilaria immitis; Dog; Brazil; Lung; Santa Catarina
1. Introduction Canine dirofilariasis is potentially a very serious infection of dogs and other canids, occurring also in cats and in humans. It is widely distributed in Brazil; canine infection has been studied mostly in coastal areas, but it can also be found in other regions (Labarthe et al., 1999). General prevalence in Brazil is 9.2% (Ahid et al., 1999), but prevalence may vary in the Brazilian States from 1.1% in Rio Grande do Sul to 12.5% in Alagoas, and 12% in Santa Catarina (Labarthe et al., 1997a,b). Prevalence is 43 and 46% in two localities in São Lu´ıs, Maranhão State (Ahid et al., 1999). ∗ Corresponding author. Fax: +55-48-331-9258. E-mail address: [email protected] (C.B. Marcondes).
0304-4017/03/$ – see front matter © 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0304-4017(03)00077-3
240
R.T. Araujo et al. / Veterinary Parasitology 113 (2003) 239–242
Prevalence of canine dirofilariasis was studied in the neighborhood of Conceição Lagoon, a tourist region of Santa Catarina Island, in southern Brazil. Venous blood was obtained from 80 dogs with defined residence, west of the Conceição Lagoon. The Lagoon is situated in the center of the Santa Catarina Island, Florianópolis Municipality. The co-ordinates of extremes of the area, obtained by GPS 45 (GarminTM ) varied from 27◦ 34 56.1 S 48◦ 27 30.3 W to 27◦ 36 11.1 S 48◦ 27 46.7 W, and 27◦ 35 52.4 S 48◦ 28 02.9 W to 27◦ 35 54.8 S 48◦ 27 48.5 W. Houses usually have several fruit trees in their backyards, and the higher parts of the area are forested. Blood was also obtained from 40 dogs of the Military Police kennel, in the nearby São José Municipality (27◦ 36 55 S 48◦ 37’39 W, 8 m a.s.l.), on the mainland. The area near the kennel is plain and, besides some eucalyptus trees, has no vegetation, and is exposed to strong winds. In both studies, only dogs older than 6 months and not treated for dirofilariasis in the last 12 months were studied. Blood of each dog was used for preparing one-thick smear by spreading it in about 1 cm2 using another slide. Preparations were exposed to distilled water and stained by the Giemsa method. Another sample (1 ml) was processed by modified Knott method (Georgi et al., 1990). Briefly, the sample was mixed with 15 ml of EDTA. In the laboratory, this material was gently mixed with 10 ml of formaline at 2%(2 ml of formaldehyde at 37% in 98 ml of distilled water), to break the erythrocytes and fix the leukocytes and the microfilariae. After 2–3 min, the material was centrifuged for 5 min, at 1500 rpm, discarding the supernatant. Thirty-five microliter of the sediment was transferred to each of three slides, preparing three smears of similar sizes, staining them by Giemsa method, and calculating the number of microfilariae per milliliter. Microfilariae from all positive dogs were examined, to differentiate them from those of Dipetalonema sp. Ten microfilariae, randomly chosen, were measured, to check the specific identification. In the Conceição Lagoon region, 12 of the 80 dogs were positive (15%); five of them were positive in both the tests (6.25%), five only by modified Knott’s (6.25%) and two only in thick smear (2.5%). None of the 40 dogs of the kennel of the Military Police was positive. The absence of dirofilariasis in the dogs of the kennel is probably due to the conditions of the terrain. They are not suitable for mosquito breeding. Neither the curved tail nor the cephalic hook, which characterize microfilariae of Dipetalonema sp. and differentiate them from those of Dirofilaria immitis (described by Leidy in 1856, Hoffmann, 1987), was observed in any of the 976 microfilariae, from all positive dogs. The morphometric study of 10 microfilariae showed that the length varied from 212.8 to 289 m (mean length = 242.9 ± 28.9 m) and the width varied from 2.5 to 6 m (mean width = 3.85 ± 1.2 m). Age of the animals, analyzed by χ2 , significantly influenced the infection index; it was higher in older dogs (Table 1). No significant influence of sex, time of blood collection (morning versus afternoon), place where dogs slept (out versus in the houses), hair length (short versus median versus long) or breed (defined breed versus mongrel) on infection index was observed (Table 2). The distribution of the positive dogs over the locality indicates that no condition of the terrain seems to influence the infection. Infections were distributed over all the area. Steindel (1999, personal communication), studying Chagas disease in the most southern
R.T. Araujo et al. / Veterinary Parasitology 113 (2003) 239–242
241
Table 1 Influence of the age of the dogs on the prevalence of dirofilariasis in animals from Florian´opolis, Santa Catarina, Brazil Age classes (years) 1–4 5–8 >8 Total
Positive 4 (7.8%) 5 (23.8%) 3 (37.5%) 12 (15%)
Negative
Total
47 (92.2%) 16 (76.2%) 5 (62.5%)
51 (63.8%) 21 (26.3%) 8 (10%)
68 (85%)
80 (100%)
χ2 = 6.36; χ2 = 5.99 (2 d.f.; 5%).
locality in the western edge of the Lagoon (“Canto da Lagoa”), in 1980s, observed a D. immitis prevalence of 23% in the dogs. Modified Knott’s method seems to be better than thick smear. However, since two dogs were positive only by the last method, it should not be discarded. The examination of more material (2–3 slides) in the Knott’s method may improve its efficiency and should be evaluated. The 57.1% of dogs from Recife (Pernambuco State) with adult worms of D. immitis in the heart could not be detected by the Knott method (Alves et al., 1999); this indicates the use of more than one method for the dogs is advised. However, the low efficiency of the Knott’s method in the dogs from Recife was attributed to the use of microfilaricide drugs (Alves et al., 1999), which were not used in the dogs of our study, as far as informed by their owners. Further studies on the efficiency of these rather simple methods should be developed. Even though serology is probably more sensitive than the parasitological techniques, the latter is simpler, can be used in any laboratory, and permits an examination of the morphology of the microfilariae. Additionally, there is no risk of false positive. In the future, there may be problems with false negative results with test that detect microfilariae and veterinarians in this area will need to use antigen tests. Comparing the dimensions of the microfilariae with those of previous studies (Flynn, 1973; Georgi et al., 1990; Hoffmann, 1987) indicates that the larvae from Florianópolis are in the lower part of the range. N. Labarthe (2000, personal communication) does not usually measure larvae, observing only their morphology. An increase in prevalence from 1 to 5 years of age was observed in dogs in Padua, Italy (Capelli et al., 1996); this result is similar to those of the present study. Analyzing the data of Alves et al. (1999), a significantly greater proportion of male dogs with microfilariae could Table 2 Values of χ2 for the influence of some features of the dogs on the prevalence of dirofilariasis in animals from Florian´opolis, Santa Catarina, Brazil Features of the dogs
Calculated χ2
From the table χ2 (d.f.) (5%)
Sex Time of blood collection (morning vs. afternoon) Sleep (out vs. in the house) Hair length (short vs. median vs. long) Breed (defined breed vs. mongrel)
1.76 0.02 0.02 0.88 2.87
3.84 (1) 3.84 (1) 3.84 (1) 5.99 (2) 3.84 (1)
242
R.T. Araujo et al. / Veterinary Parasitology 113 (2003) 239–242
be observed (χ2 = 4.42, cf. χ12 d.f.; 5% = 3.84). However, this number included dogs with microfilariae of Dipetalonema reconditum (Grassi, 1890) as well as of Dirofilaria immitis. The analysis of numbers of dogs infected with each worm did not indicate significant results (χ2 = 3.44 and 0). The periodicity of the microfilariae in the peripheral blood, although not significant in the dogs of the present study, should be more thoroughly studied, due to previous observation of periodicity (e.g. Nogami et al., 2000). The periodicity, which seems to vary in different regions, could furnish further information on the efficiency of potential vectors and in the choice of the time for examination of the dogs. The high prevalence of dirofilariasis in the locality dictates that the population is informed and that dogs are inspected and treated. The area, besides being densely populated, receives many tourists in the summer, mostly from nearby states and from Argentina. The prevalence of dirofilariasis in dogs from Argentina can be as high as 60% in localities near Buenos Aires (Grubissich, 1999). Since most Argentinean tourists at Florianópolis come from the Buenos Aires region and many of them transport their dogs, there is a great probability of circulation of the parasite between both the regions in summer. The sanitary inspection of dogs in the frontier should be enforced. Also, dirofilariasis should be considered in cases of unexplained human pulmonary nodules in the region of Florianópolis. Acknowledgements To Dr. Lawrence Hribar (Keys Mosquito Control District, Florida), for checking the English language and commenting on the manuscript. To Dr. Norma Labarthe (UERJ, Rio de Janeiro) for comments on the work. References Ahid, S.M.M., Lourenço-De-Oliveira, R., Saraiva, L.Q., Dirofilariose Canina na Ilha de São Lu´ıs, Nordeste do Brasil: Uma Zoonose Potencial. Cad. Saúde Públ., Rio de Janeiro, 15, 405–412. Alves, L.C., Silva, L.V.A., Faustino, M.A., McCall, J.W., Supakonderj, P., Labarthe, N.W., Sanchez, M., Caires, O., 1999. Survey of canine heartworm in the city of Recife, Pernambuco, Brazil. Mem. Inst. Oswaldo Cruz 94, 587–590. Capelli, G., Poglayen, G., Bertotti, F., Giupponi, S., Martini, M., 1996. The host–parasite relationship in canine heartworm infection in a hyperendemic area of Italy. Vet. Res. Commun. 20, 320–330. Flynn, R.J., 1973. Parasites of Laboratory Animals, vol. 1. The Iowa State University Press, Ames. 442 pp. Georgi, J.R., Georgi, M.E., Theodorides, V.J., 1990. Parasitology for Veterinarians, 5th ed. W.B. Saunders Company, Philadelphia. 412 pp. Grubissich, J.A., 1999. Dirofilariasis canina. Holliday News 2, 8–12. Hoffmann, R.P., 1987. Diagnóstico de Parasitismo Veterinário. Porto Alegre, Sulina. 156 pp. Labarthe, N.V., Araújo, A.M., Bordin, E.L., Larsson, M.E., Guerreiro, J., 1997a. Update on the distribution of Dirofilaria immitis in dogs in Brazil. In: Proceedings of the XVIISA-VA World Congress, Roma, Italy. pp. 287–289. Labarthe, N., Almosny, N., Guerrero, J., Duque-Araújo, A.M., 1997b. Description of the occurrence of canine dirofilariasis in the State of Rio de Janeiro Brazil. Mem. Inst. Oswaldo Cruz 92, 47–51. Labarthe, N., Lourenço-De-Oliveira, R., Serrão, M.L., 1999. A transmissão da Dirofilaria immitis no Rio de Janeiro. Vetores Pragas 2, 20–25. Nogami, S., Murasugi, E., Shimazaki, K., Maeda, R., Harasawa, R., Nakagaki, K., 2000. Quantitative analysis of microfilarial periodicity of Dirofilaria immitis in cats. Vet. Parasitol. 92, 227–232.