Materials and Methods

Electronic medical records of all cats examined at veterinary hospitals in the United States owned by Banfield, The Pet Hospital, between January 1, 2003, and December 31, 2003, were transferred to the Purdue University School of Veterinary Medicine and converted into analysis-ready data sets. During this period, Banfield operated 359 primary-care hospitals in 40 states. All hospitals used standard protocols for the diagnosis and treatment of nematode infections, and all hospitals used the same proprietary software program^a to create electronic medical records. All patient and medical information was collected and stored in an electronic format that was uploaded weekly into a central data warehouse. Medical information entered during each patient visit was identified by unique patient and encounter (office visit) identification numbers.

Fecal samples used for identification of intestinal parasitism were submitted by the owner or, when necessary, collected at the hospital with a fecal loop. A commercial test system^b with zinc sulfate solution^c was used for fecal flotation testing of all fecal samples, except that examination of direct smears of fecal samples diluted with saline (0.9% NaCl) solution was used when there was an insufficient sample to perform flotation testing. A diagnosis of intestinal parasitism was made on the basis of microscopic identification of parasite eggs. Nematode parasites were classified as roundworms, hookworms, or whipworms; no attempts were made to identify genus or species.

For the present study, the following information was abstracted from each electronic medical record: date of visit, age, sex, breed, reproductive status (ie, sexually intact vs neutered), whether a fecal flotation test was performed, and, if so, results of the fecal flotation test. During the study period, attending veterinarians recommended that a fecal examination be routinely performed at least once a year in all cats and that additional examinations be performed as needed in cats with signs of gastrointestinal tract disease.

The prevalence of roundworm, hookworm, and whipworm infection was defined as the number of positive fecal test results divided by the total number of fecal tests performed. Because some cats may have had > 1 fecal test performed during the study period, prevalence was also defined as the number of cats with positive fecal test results divided by the number of cats tested.

Standard software^d was used to map prevalences of parasitism for each of 10 geographic regions in the United States: north Atlantic, east north central, west north central, mid Atlantic, south Atlantic, east south central, west south central, mountain, north Pacific, and south Pacific.^e Regions were based on those used by the CDC for notifiable diseases,¹⁰ except that regions were modified as necessary to allow for a minimum of 11 participating veterinary hospitals/region.^f Prevalence was also calculated on the basis of season during which fecal testing was performed, as follows: summer (June, July, August), fall (September, October, November), winter (December, January, February), and spring (March, April, May).

Risk factors for roundworm and hookworm infection were identified by means of both fixed- and random-effects multivariate logistic regression. However, because the fixed- and random-effects models gave similar results, only results for fixed-effects models are presented. Potential risk factors for whipworm infection were not analyzed because of the low number of cats found to be infected.

Data were analyzed with standard software,^g and values of P < 0.05 were considered significant. The association between potential risk factors and nematode infection was expressed as the OR and 95% CI, whereby an OR > 1 indicates an increased risk of infection and an OR < 1 indicates a decreased risk.

Results

Prevalence of nematode infection—The review of the electronic medical records identified 697,898 unique patient encounters involving 356,086 cats during the study period (Table 1). Mean ± SD number of encounters per cat was 1.96 ± 1.92. Of the 356,086 cats, 60.2% had a single patient encounter during the study period, 18.4% had 2 patient encounters during the study period, and 21.4% had ≥ 3 patient encounters during the study period.

Mean ± SD age of the cats was 4.3 ± 4.5 years (median, 2.6 years; range, 1 month to 22 years). Of the 695,508 encounters for which information on sex was available, 347,834 (50%) involved male cats and 347,674 (49.9%) involved female cats (Table 1). There were 589,351 (84.4%) patient encounters that involved mixed-breed cats. There were 108,547 (15.6%) encounters representing > 40 recognized pure breeds, including Siamese (32,021 encounters [29.5% of all purebred cat encounters]), Persian (17,367 encounters [16%]), and Himalayan (11,831 encounters [10.9%]). The highest number of patient encounters occurred in the south Atlantic region, which also contained the highest number of participating hospitals, whereas the lowest number of patient encounters occurred in the east south central region, which contained the lowest number of participating hospitals. Frequency of fecal examinations for cats and patient encounters was determined for each region. Numbers of hospitals, cats, and patient encounters per region were not associated with the percentage of fecal examinations performed in that region (Table 1).

Fecal flotation tests were performed on 66,819 of the 356,086 (18.8%) cats and during 80,278 of the 697,898 (11.5%) patient encounters. Of the 66,819 cats that had a fecal test performed, 81.6% had a single fecal flotation test performed, 16.7% had 2 fecal flotation tests performed, 1.6% had 3 fecal flotation tests performed, and 0.1% had 4 fecal flotation tests performed. Kittens (0 to < 6 months old) and juveniles (6 to < 12 months old) had a fecal test performed more frequently than young adult cats (1 to < 4 years old), adult cats (4 to < 10 years old), and geriatric cats (≥ 10 years old). Neutered cats had a fecal test performed more frequently than sexually intact cats, and mixedbreed cats had a fecal test performed more frequently than purebred cats. Cats in the south Pacific region had fecal tests performed less frequently than cats in the other regions.

Overall prevalences of roundworm and hookworm infection, determined on the basis of number of positive fecal flotation test results, were 2.92% and 0.63%, respectively (Table 2). Prevalences of roundworm and hookworm infection, determined on the basis of number of cats with positive test results, were 3.46% and 0.75%, respectively. Prevalence of whipworm infection was 0.031% when calculated on the basis of number of positive test results and 0.037% when calculated on the basis of number of cats with positive test results. Younger cats, sexually intact cats, and mixed-breed cats were more likely to be infected with both roundworms and hookworms than were older cats, neutered cats, and purebred cats. Prevalences of roundworm and hookworm infection in kittens were further evaluated by grouping kittens into 2-week age intervals (Figures 1 and 2).

View Full Size

Prevalence of roundworm infection among kittens (n = 17,780) examined at 359 private veterinary hospitals in the United States during 2003; prevalence was calculated as the percentage of fecal examinations (n = 20,831) with positive results. Error bars represent 95% CIs.

Citation: Journal of the American Veterinary Medical Association 228, 6; 10.2460/javma.228.6.885

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Prevalence of hookworm infection among kittens (n = 17,780) examined at 359 private veterinary hospitals in the United States during 2003; prevalence was calculated as the percentage of fecal examinations (n = 20,831) with positive results. See Figure 1 for key.

Citation: Journal of the American Veterinary Medical Association 228, 6; 10.2460/javma.228.6.885

• Download Figure
• Download figure as PowerPoint slide

Prevalences of roundworm and hookworm infection, determined on the basis of number of positive test results, were calculated for season (Figure 3) and region (Figures 4 and 5). The prevalence of roundworm infection was significantly higher during the summer, fall, and winter seasons, compared with the spring season (Table 3). The prevalence of hookworm infection was significantly higher during the summer season, compared with the spring season.

View Full Size

Prevalences of roundworm and hookworm infection as a function of season for cats (n = 66,819) examined at 359 private veterinary hospitals in the United States during 2003; prevalence was calculated as the percentage of fecal examinations (n = 80,278) with positive results.

Citation: Journal of the American Veterinary Medical Association 228, 6; 10.2460/javma.228.6.885

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Prevalence of roundworm infection as a function of region for cats (n = 66,819) examined at 359 private veterinary hospitals in the United States during 2003; prevalence was calculated as percentage of fecal examinations (n = 80,278) with positive results.

Citation: Journal of the American Veterinary Medical Association 228, 6; 10.2460/javma.228.6.885

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Prevalence of hookworm infection as a function of region for cats (n = 66,819) examined at 359 private veterinary hospitals in the United States during 2003; prevalence was calculated as percentage of fecal examinations (n = 80,278) with positive results.

Citation: Journal of the American Veterinary Medical Association 228, 6; 10.2460/javma.228.6.885

• Download Figure
• Download figure as PowerPoint slide

Risk factors for roundworm and hookworm infection—Age, reproductive status, breed, and season were significant risk factors for roundworm infection (Table 3). Kittens, juveniles, and young adult cats were more likely to be infected with roundworms than were geriatric cats; sexually intact cats were more likely to be infected than spayed female cats; mixed-breed cats were more likely to be infected than purebred cats; and cats examined during the summer, fall, or winter were more likely to be infected than cats examined during the spring.

Similarly, age, reproductive status, and season were significant risk factors for hookworm infection (Table 3). Kittens and juveniles were more likely to be infected than geriatric cats, sexually intact cats were more likely to be infected than spayed female cats, and cats examined during the summer were more likely to be infected than cats examined during the spring.

Regional differences in prevalences of roundworm and hookworm infection were found (Table 3). Cats from all geographic regions other than the south Pacific region were more likely to be infected with roundworms than were cats from the mountain region. Cats from the north Atlantic, mid Atlantic, south Atlantic, east south central, and west south central regions were more likely to be infected with hookworms than were cats from the mountain region, and cats from the south Pacific region were less likely to be infected with hookworms than were cats from the mountain region.

Discussion

In the present study, estimated prevalences of roundworm and hookworm infection among pet cats in the United States were 3.46% and 0.75%, respectively. These values are generally lower than values published previously.^5–9,11 For instance, a recent study⁸ of 184 cats examined at animal control agencies in Georgia in which fecal samples were tested by means of the zinc sulfate flotation method reported that prevalence of roundworm infection was 28.2% and that prevalence of hookworm infection was 10.2%.⁸ In a study⁵ of 2,000 cats examined at the Veterinary Hospital of the University of Pennsylvania between 1984 and 1991 in which fecal samples were tested by means of the zinc sulfate centrifugal method, the prevalence of roundworm infection was 16% and the prevalence of hookworm infection was < 1%. A study⁷ conducted at the University of Missouri Veterinary Teaching Hospital in which fecal samples were tested with the magnesium sulfate flotation method found that 2.6% of cats were infected with whipworms categorized as Trichuris spp and Capillaria spp. The largest previous study⁴ of the prevalence of nematode infection in cats in the United States involved 11,995 cats examined at the Iowa State University Veterinary Teaching Clinic between 1966 and 1974 and reported that prevalence of roundworm infection was 2.8%, prevalence of hookworm infection was 1.2%, and prevalence of combined roundworm and hookworm infection was 0.5%, with overall prevalence of nematode infection being 4.5%. However, the method used for examination of fecal samples was not reported in this study.

The zinc sulfate flotation method is considered sensitive for identification of nematode eggs, but the sensitivity is lower when centrifugation is not performed,^1,12 as was the case for samples examined during the time frame of the present study. Direct examination of smears has a low sensitivity because of the small amount of fecal material used.¹ Thus, reported prevalences of nematode infection in the present study are likely to underestimate the true prevalence of infection. However, sensitivity of the fecal flotation test was unlikely to affect results of the multivariate analysis of potential risk factors for infection.

Differences in estimates of prevalence of nematode infection among studies can be explained by the fact that infection status is strongly influenced by host factors such as age^4–7 and sex,^4,7 environmental factors such as living condition (owned vs free ranging)^2,13 and geographic area,^13,14 health status (routine veterinary care vs no routine care),^2,13 and diagnostic method used to detect parasites (examination of direct smears vs fecal flotation vs fecal flotation with centrifugation).¹ Cats examined at veterinary hospitals that participated in the present study were expected to have a low prevalence of nematode infection because they generally were privately owned and because most fecal flotation tests were performed as a part of a routine wellness examination. No attempts were made in the present study to calculate prevalence of nematode infection in healthy cats versus those with signs of gastrointestinal tract illness.

Our finding that prevalence of hookworm infection varied with season in the present study is consistent with results of a previous study⁴ conducted at the Iowa State University Veterinary Teaching Clinic in which prevalence was found to peak in June and July. Seasonal and geographic variations in prevalence of hookworm infection may reflect the optimal temperature for development of hookworm larvae, which is 20°C.¹³ In contrast, in the present study, cats examined during the summer, fall, or winter had a higher risk of roundworm infection than did cats examined during the spring. Roundworm eggs can survive in temperatures as low as −15°C and as high as 37°C, making them extremely resistant to dessication.^13,15 Roundworm eggs, therefore, may survive for longer periods throughout the year and in a greater range of geographic areas than do hook-worm eggs.

Female cats frequently give birth in the spring,^16,17 and the kittens will often be first examined by a veterinarian at 6 to 8 weeks of age. Because kittens are at greater risk for nematode infection than are adult cats, this may in part explain why the observed prevalence of nematode infection was highest during the summer months. However, even when age was included in the multivariate analysis, season remained a significant risk factor, with prevalences of roundworm and hookworm infection higher during the summer than during the spring.

In the present study, we found that kittens < 4 weeks old shed roundworm and hookworm eggs. The CDC recommends that preventative anthelmintic treatment begin at 3 weeks of age in cats, with follow-up treatments given at 5, 7, and 9 weeks of age.¹⁸ Our results emphasize that young kittens can have patent infections. Therefore, treatment should be considered for kittens of all ages to reduce the potential for egg shedding and human exposure.

One advantage of the present study was that we were able to identify geographic patterns in the prevalences of roundworm and hookworm infection in pet cats. The higher prevalence of hookworm infection in the southern regions was expected because this parasite favors sandy soil and warm temperatures that provide an optimal environment for development and persistence of larvae.¹⁴ However, the observed pattern with respect to roundworms, with highest prevalence in the upper Atlantic and east central regions, is not readily explained. Previous studies^13–15,19 have suggested that roundworm infection occurs worldwide in cats, with no obvious geographic differences. Widespread distribution is favored because roundworm eggs are resistant to extreme temperatures, resulting in survival of the eggs for many years.^1,13,15 However, cats in the mountain region in the present study had a significantly lower risk of roundworm infection, compared with cats in all other regions except the south Pacific region. A study²⁰ of nematode infection in dogs found that the prevalence of infection was lowest in the western region of the United States, which corresponded to the Pacific and mountain regions in the present study, and highest in the southeastern region, which corresponded to the mid Atlantic, south Atlantic, and south central regions in the present study.

In the present study, only 25 cats were found to be positive for whipworm infection, which was not surprising given that whipworm infection is considered rare among cats in North America.^1,21 Similarly, a study⁸ of intestinal parasitism among cats in northwest Georgia did not identify any cats with whipworm infection. In contrast, a study⁷ of cats admitted to the University of Missouri Veterinary Teaching Hospital found that 2.6% had evidence of whipworm infection. However, that study grouped Trichuris spp and Capillaria spp together for reporting purposes, and there is some evidence that a diagnosis of whipworm infection among cats in North America may actually represent a misdiagnosis of Capillaria infection or passage of eggs from an ingested rodent.²¹ Whipworms identified by means of fecal flotation tests in the present study were not confirmed by other methods because these cats were clinically normal and whipworm infection in cats does not usually cause clinical disease.^1,21

There were several potential limitations to the present study, despite the large sample size and the fact that cats included in the study were patients examined at full-service veterinary hospitals. Although participating hospitals were located in 40 states, many areas of the United States were underrepresented. No attempts were made to distinguish between sick and healthy cats, and the medical records did not contain sufficient information to classify each cat with respect to housing status (ie, strictly indoors vs outdoor exposure vs strictly outdoors) or degree of association with other cats. Finally, whereas all participating hospitals had the same protocol for parasite examinations and all examinations were performed by trained personnel, not all individuals necessarily performed the fecal flotation test in the same way.

In summary, findings of the present study suggest that the prevalence of nematode infection among pet cats in the United States is low. This low prevalence may reflect the quality of care provided to cats by their owners, the high efficacy of drugs available to treat roundworm and hookworm infections, the emphasis placed on preventive medicine by veterinarians at participating hospitals, or the low sensitivity of the fecal examination method used. Nematode infection in cats poses a potential zoonotic threat; therefore, prevention and treatment should be routinely discussed with individuals who own cats, especially with individuals who own cats < 12 months old.