Correlations of Fatty Acid Supplementation, Aeroallergens, Shampoo, and Ear Cleanser With Multiple Parameters in Pruritic Dogs
Seventy-two pruritic dogs were fed one of four diets controlled for n-6:n-3 fatty acid ratios and total dietary intake of fatty acids. Multiple parameters were evaluated, including clinical and cytological findings, aeroallergen testing, microbial sampling techniques, and effects of an anti-fungal/antibacterial shampoo and ear cleanser. Significant correlations were observed between many clinical parameters, anatomical sampling sites, and microbial counts when data from the diet groups was combined. There were no statistically significant differences between individual diets for any of the clinical parameters. The importance of total clinical management in the control of pruritus was demonstrated.
Introduction
Pruritus is often the chief client complaint associated with canine dermatitis. There are many possible causes of pruritus, with allergies (e.g., atopy, fleas, food) and external parasites (e.g., Sarcoptes, Cheyletiella) being amongst the most common.1 Chronic or recurring otitis externa is a common finding in the pruritic dog.2 Bacteria and yeast are often contributing factors to pruritus and to gross dermatological lesions.2 Clinical signs in pruritic dogs are variable depending on the primary and secondary etiologies, severity of pruritus, and anatomical sites of the lesions.34 The immunological, infectious, clinical, diagnostic, and therapeutic aspects of canine atopic dermatitis, the role of inflammatory cells in cutaneous allergic reactions, and the mediators of cutaneous inflammation have recently been reviewed.1–12
Successful management of pruritus and otitis externa requires the identification and control of both primary and secondary causes. Systemic medications, including antibiotics, glucocorticoids, antihistamines, and supplements containing n-3 and n-6 fatty acids are frequently used for symptomatic relief of pruritus.1013–16 Topical shampoos and rinses are often recommended to help control pruritus, as well as yeast and bacterial infections. Concurrent use of multiple medications is frequently indicated.13–17 Specific topical antipruritic, antibacterial, antifungal, and antiseborrheic agents for the skin and ears are typically selected on the basis of the history, physical findings, and diagnostic test results.
The goals of this study were to assess the importance of selected factors (e.g., Malassezia spp., bacteria, aeroallergens) that contribute to pruritus, otitis externa, and gross dermatologic lesions in dogs with a history of seasonal or nonseasonal pruritus; to evaluate clinical responses to selected therapies (e.g., n-6 and n-3 fatty acids, antibiotics, antihistamines, a shampoo, and an ear cleansing solution); and to evaluate two sampling techniques (e.g., skin scrapings, acetate tape assays) for recovery of skin yeast and bacteria under conditions approximating clinical practice.
Materials and Methods
Subjects
Dogs selected for the study were evaluated by the Tufts University School of Veterinary Medicine Dermatology Service between June 1999 and June 2000. The criteria for selection included either seasonal or nonseasonal pruritus that was historically and clinically consistent with an allergic etiology. Specific criteria for selection included a minimum age of 1 year; negative skin scrapings for mites; no oral prednisone, prednisolone, or triamcinolone for at least 2 weeks; no injectable triamcinolone acetonidea for 21 days or methylprednisolone acetateb for 42 days; no oral fatty acid supplementation for 4 weeks; no ketoconazole or itraconazole therapy for 60 days; and no hyposensitization for atopy for a minimum of 4 weeks prior to entrance into the study. The dogs were eligible for the study if they were receiving antibiotics, ear medications, antihistamines, and shampoos. Dogs were excluded if they had gross lesions consistent with a deep pyoderma, mites were identified on skin scrapes, or if an endocrine disease was suspected based on history, clinical evaluation, or laboratory tests. Dietary (food) elimination trials were not required for entrance into the study.
Study Protocol
Computer-generated numbers were used to randomly assign dogs to one of four diet groups. Respective n-6:n-3 ratio, eicosapentaenoic acid (EPA) intake, and docosahexaenoic acid (DHA) intake (mg/kg body weight) for the diets were as follows: Diet A—ratio=1:1, EPA=87 mg/kg, DHA=91 mg/kg; Diet B—ratio=3:1, EPA=88 mg/kg, DHA=83 mg/kg; Diet C—ratio=6:1, EPA=25 mg/kg, DHA=24 mg/kg; and Diet D—ratio=27:1, EPA=0 mg/kg, DHA=0 mg/kg [Table 1]. All of the experimental diets were nutritionally balanced, manufactured by Nestlé Purina Pet Care Company, and exceeded the minimum Association of American Feed Control Officials (AFFCO) standards. Owners were given written instructions on the number of cups of food to give per day based on the dog’s weight. Both investigators and owners were blinded as to which diet a dog was receiving.
Owners were instructed to bathe all dogs weekly with a shampoo containing 2% boric acid and 2% acetic acid,c maintaining a minimum of 5 minutes of contact time before rinsing well. Until the second evaluation (on or about day 21 of the study), antibiotics and/or antihistamines were administered if indicated (based on clinical signs). All antibiotics and antihistamines were discontinued on the day of the second evaluation.
When treatment of otitis externa was indicated, a 2% acetic acid and boric acid cleanserd was used once or twice weekly. Owners were instructed to instill a liberal amount of the cleanser in each ear canal, gently massage the ear, and clean away any debris present on the external portion of the ear.
After the first reevaluation, the assigned diets and weekly baths were continued for the remaining 35 days of the study, and the ear cleaning solution was continued if indicated (based on clinical signs and cytology). A final evaluation occurred on or about day 56 of the study. Dogs were removed from the study if any other food or treats were fed, if the clinical signs became severe enough to require additional treatments, or if a scheduled visit was missed.
Evaluations and Scoring System
All dogs were examined on day 0 and approximately days 21 and 56 of the study. At each visit, thorough dermatological and otic histories and examinations were completed by one of the investigators (Nesbitt). A subjective score of 0 to 3 (0=not observed, 1=mild, 2=moderate, 3=marked) was assigned to multiple dermatological and otic clinical parameters. A complete blood count (CBC) and serum biochemical analysis were performed at each visit.
Clinical dermatological parameters that were graded included pruritus, erythema, and miscellaneous lesions such as alopecia and the presence of skin oiliness, papules, pustules, and scale. Seven anatomical sites were evaluated for each of the clinical parameters. The anatomical sites included the face, feet, ventrum of the trunk, axillae, dorsolateral trunk, extensor carpal surface, and flexor tarsal surface. Using the number scale described above, the following clinical dermatological scores where obtained:
Total skin erythema score: The sum of erythema scores from all anatomical sites (range 0 to 21).
Total skin pruritus score: The sum of pruritus scores from all anatomical sites (range 0 to 21).
Total miscellaneous lesion score: The sum of all individual clinical lesion scores for alopecia, oiliness, papules, pustules, and scale (range 0 to 105).
Total skin score: The sum of all scores from all anatomical sites (range 0 to 147).
Total feet score: The sum of all pruritus, erythema, and miscellaneous lesion scores from the feet (range 0 to 15).
Total axillae score: The sum of all pruritus, erythema, and miscellaneous lesion scores from the axillae (range 0 to 15).
Pinnal observations included pruritus, erythema, epithelial proliferation, scale or crust, and debris. External ear canal observations included erythema, epithelial proliferation (i.e., maculopapular inflammation, cobblestone appearance), debris, edema, and erosions. A grading scale of 0 to 3 for the individual observations was used to obtain the following clinical otic scores:
Total ear erythema score: The sum of pinnal and external ear erythema scores for both ears (range 0 to 12).
Total ear pruritus score: The pinnal pruritus score for both ears (range 0 to 6).
Total ear score: The sum of pinnal and external ear canal observations from both ears at one observation time (range 0 to 30).
Total pruritus score: The sum of the total skin pruritus and total ear pruritus scores (range 0 to 27).
Total ear and skin score: The sum of the total ear and total skin scores (range 0 to 177).
Cytological Sampling Techniques
Four anatomical skin sites were sampled for Malassezia spp. and bacteria. These sites were the axillae, ventral feet, dorsal interdigital spaces, and dorsolateral trunk. Two adjacent areas, approximately 1 cm in diameter and with similar gross appearance, were sampled via skin scraping or acetate tape application.
For the skin scrapings, a no. 10 scalpel blade was used to scrape the debris and superficial corneocytes from the surface. The sample was obtained from the deepest folds and crevices in the dorsal and ventral interdigital spaces and axillae. A superficial scraping was also taken from the dorsolateral trunk. The material that adhered to the scalpel blade was transferred to a glass slide and stained with a modified Wright’s stain.e The slides were not heat-fixed prior to staining. For the acetate tape technique, clear, single-sided acetate tape was used. The adhesive side of the tape was firmly touched to the skin surface. A few drops of modified Wright’s staine were placed on a glass slide, and then the tape was applied to the slide over the stain and smoothed out to distribute the stain uniformly under the tape.
Both ears were sampled using a cotton-tipped applicator carefully inserted into the lumen of the vertical canal to the maximum depth possible, usually at the junction of the vertical and horizontal canals. As the applicator was removed, it was rubbed against the side of the vertical canal to remove debris.
The stained slides (skin and ear samples) and tape preparations (skin samples) were reviewed microscopically under oil-immersion (100×). The entire slide was scanned. Six oil-immersion fields with the largest number of yeast and six fields with the largest number of bacteria were counted. Counts from each of the six fields for each organism were summed and divided by six to obtain a mean Malassezia or bacteria colony count per oil-immersion field. The number of organisms were scored using a scale of 0 to 5, with 0=none seen, 1=one to two Malassezia organisms or bacterial colonies per oil-immersion field (OIF), 2=three to six Malassezia organisms or bacterial colonies per OIF, 3=seven to 10 Malassezia organisms or bacterial colonies per OIF, 4=11 to 20 Malassezia organisms or bacterial colonies per OIF, and 5=>20 Malassezia organisms or bacterial colonies per OIF. These scores were used to obtain the following microorganism counts:
Skin Malassezia count: The sum of the mean Malassezia spp. counts per oil-immersion field obtained from each of the anatomical sites using the tape sampling technique (range 0 to 20).
Ear Malassezia count: The sum of the mean Malassezia spp. counts per oil-immersion field obtained from swabs of both ears (range 0 to 20).
Skin cocci count: The sum of the mean cocci counts per oil-immersion field obtained from all anatomical sites for sampling using skin scrapings and tape preparations (range 0 to 20).
Ear cocci count: The sum of the mean cocci counts per oil-immersion field obtained from swabs of both ears (range 0 to 10).
Skin rod count: The sum of the mean rod counts per oil-immersion field obtained from all anatomical sites for sampling using skin scrapings and tape preparations (range 0 to 20).
Ear rod count: The sum of the mean rod counts per oil-immersion field obtained from swabs of both ears (range 0 to 10).
Total feet Malassezia count: The sum of mean interdigital and ventral feet Malassezia spp. counts (range 0 to 10).
Total feet cocci count: The sum of mean interdigital and ventral feet cocci counts (0 to 10).
Total Malassezia count: The sum of mean total feet, skin and ear Malassezia counts obtained from tape preparations (range 0 to 0.50).
Allergen Antibody Test With Seasonal and Nonseasonal Allergens
On day 0, an allergen-specific IgE enzyme-linked immunosorbent assay (ELISA) test,f containing 48 allergens [Table 2], was performed on serum from each dog in the study. A second ELISA testg containing 11 mold allergens [Table 2] was also performed on serum from 36 of the dogs with a history of nonseasonal pruritus. The seasonal allergens included weed (13), grass (10), and tree (15) pollens. The nonseasonal allergens included all molds, six and 11 respectively in the two Elisa tests,f,g as well as cat epithelium, cockroach, and two dust mite allergens. Flea allergen, which may be either a seasonal or nonseasonal allergen, was included in the Northwest screen.f Dogs with a seasonal history of pruritus with periods of no pruritic symptoms were classified as seasonal pruritic dogs. Dogs that manifested some pruritus at all times, regardless of the time of year, were classified as nonseasonal pruritic dogs.
Statistical Analysis
The Spearman’s rank correlation was used to estimate the correlations among the selected variables to detect linear relationships between observations. Analysis of variance (ANOVA) was used for comparison of the different anatomical sites and for testing of differences between the scraping and tape sampling techniques. The Wilcoxon’s signed rank test was used to determine the differences in means. A permutation test, PROC MULTTEST in SAS, was used to determine differences between groups when several different groups were compared.h A P value of ≤0.05 was considered statistically significant.
Results
Seventy-two dogs entered the study, but only 58 dogs completed the study. Of the 58 dogs, 13 were in diet group A and 15 were in diet groups B, C, and D. The average (range) of ages for dogs in the four diet groups were A: 5.8 (2 to 10) years, B: 5.2 (1 to 11) years, C: 5.4 (3 to 11) years, and group D: 4.7 (1 to 11) years. There were a total of 29 different breeds and five mixed-breed dogs represented in the study. The most common breeds were the golden retriever (n=7), Labrador retriever (n=), and Labrador retriever-cross (n=7). The next most common breeds were the Lhasa apso (n=5), German shepherd dog (n=5), and German shepherd-cross (n=5). There were 28 males and 30 females.
Flea control was used in 36 of the dogs, including four in diet group A, 13 in group B, 10 in group C, and nine in group D. Flea control products included imadoclopridi (n=17), fipronilj (n=15) , lufenuronk (n=3), permethrinl (n=2), and flea collars (n=2). Seven dogs received systemic antibiotics, including cephalexin (n=5), trimethoprim-sulfadiazine (n=1), and amoxicillin-clavulanate (n=1). Dogs receiving antibiotics were distributed across all diet groups, including one dog in diet group B and two dogs in each of groups A, C, and D. Antihistamines were used in 14 dogs, including one in diet group A, four each in groups B and C, and five in group D. Seven of these dogs were treated with diphenhydramine, and six dogs received hydroxyzine HCl. Ear medications were used in 44 dogs, including 11 in diet groups A and C, 12 in group B, and 10 in group D. Diet trials were completed in 11 dogs, including one in diet group A, four in group B, and three in groups C and D. Thirty of the dogs had a nonseasonal pruritus, and 28 dogs manifested seasonal pruritic symptoms. Of the 114 baths recorded in the study, 80 (71.8%) were administered at weekly intervals as directed. Thirty-four (29.8%) were given at intervals >7 days. Baths were administered at intervals between 8 and 14 days in 58.8% (20/34) of the dogs and at intervals >14 days in 47.6% (14/34) of the dogs.
Effects of Adjunctive Treatments
The use of oral adjunctive treatments (i.e., antihistamines only, antibiotics only, or concurrent antibiotics and antihistamines) did not significantly correlate with skin or otic clinical scores on days 0, 21, and 56. For the differences between treatment groups for total skin scores and total ear scores, P values ranged from P=0.334 to P=1.000. All of the CBCs and serum biochemical analyses were within normal reference ranges on all three sampling days.
Effect of Diets on Clinical Scores
No significant differences were detected between the four diet groups for any of the dermatological or otic clinical parameters. Table 3 lists the numerical scores and the average decrease in scores (in percentages) for each of the four diets with respect to various clinical scores for skin and ear parameters. Mean clinical scores for erythema, pruritus, miscellaneous lesions, and total skin values for all dogs in the study are shown in Figure 1.
Comparison of Tape Preparation and Scraping Techniques
The means±standard deviations (SD) for skin cocci counts obtained with the scraping technique were 1.250±2.312 on day 0, 0.587±1.433 on day 21, and 0.737±2.117 on day 5. The means±SD for skin cocci counts obtained with the tape technique were 0.792±1.583 on day 0, 0.359±0.88 on day 21, and 0.446±0.913 on day 56. No significant differences were noted between the mean skin cocci count when scraping, and tape sampling techniques were compared for a single day (day 0 P=0.085, day 21 P=0.420, day 56 P=0.318). However, when the sum of the scores from the three observation days were compared, there was a 38.2% (P=0.045) higher recovery of cocci with the scraping technique [Figure 2].
The means±SD for skin Malassezia spp. counts obtained with the scraping technique were 0.722±1.224 on day 0, 0.683±1.457 on day 21, and 0.750±2.384 on day 56. The means±SD for skin Malassezia spp. counts obtained with the tape technique were 1.750±2.670 on day 0, 0.984±2.012 on day 21, and 1.526±2.797 on day 56. A comparison of scraping and tape techniques for Malassezia spp. showed significant differences on days 0 (P<0.001) and 56 (P=0.007). When the sums of all observations for the 3 days were compared for both sampling techniques, there was a 50.3% (P<0.001) higher recovery with the tape technique [Figure 3].
Anatomical Site Scores and Correlations
On days 0, 21, and 56, the total clinical site-specific scores for the feet and ventrum were significantly higher (P<0.05) than scores for the axilla, extensor carpal surface, flexor tarsal surface, and dorsolateral trunk. On day 0, all of the individual site-specific clinical scores were significantly and positively correlated with total skin score, total skin pruritus, and total skin miscellaneous lesions scores (e.g., alopecia, oiliness, papules, pustules, scale). Similar significant positive correlations were observed between total skin erythema scores and site-specific clinical scores for the feet, ventrum, axillae, and flexor tarsi on day 0. All site-specific skin scores were significantly correlated with total skin score, total skin miscellaneous lesions, total skin pruritus, and total skin erythema scores on day 21 (P=0.02), with two exceptions, namely the face with miscellaneous lesions and extensor carpal surface with skin erythema. On day 56, similar significant positive correlations (P<0.025) were observed between most site-specific skin scores and most clinical parameter scores. Four exceptions included the face with miscellaneous lesions, skin pruritus, skin erythema, and the dorsolateral trunk with skin erythema. There were strong positive correlations (P<0.001) between total skin score and total skin pruritus, total skin erythema, and total skin miscellaneous lesion scores for days 21 and 56.
The total skin score, total skin miscellaneous lesion score, total skin pruritus score, and total skin erythema scores were compared to the site-specific skin yeast and cocci counts as well as the total skin yeast and cocci counts for days 0, 21, and 56. The positive correlations are listed in Table 4. Significant positive correlations were also observed between total feet cocci and Malassezia spp. counts on day 21 (P=0.111) and day 56 (P=0.027).
Correlation of Skin Parameters and Cytological Findings
On day 21, significant positive correlations were observed between skin erythema scores and skin cocci (P<0.001) and Malassezia spp. (P<0.001). Correlations between skin pruritus scores and cocci (P=0.002) and yeast counts (P=0.041) were also significant, as were total skin score and skin cocci counts (P=0.029). On day 56, significant positive correlation was observed between skin erythema and both skin Malassezia spp. (P=0.026) and cocci counts (P<0.001), while no significant correlation was found between skin pruritus or total skin scores and yeast or cocci counts. It should be noted that correlation coefficients detect linear relationships; they do not reflect a cause and effect.
Correlation of Otic Parameters and Cytological Findings
Strong correlations on day 0 were seen between mean total ear scores and mean ear Malassezia spp. (P=0.002), cocci (P<0.001), and rod (P=0.002) counts. On day 21, similar positive correlations were observed between total ear scores and ear yeast (P=0.017), cocci (P=0.002), and rod counts (P=0.005). Similar correlations were observed on day 56 between total ear score and ear cocci (P<0.0001) and rod counts (P=0.002). No correlation was found between total ear score and ear yeast counts (P=0.99).
Strong correlations were seen between ear Malassezia spp. counts and ear pruritus scores (P<0.001) for days 0, 21, and 56. Correlations were observed between ear yeast counts and ear erythema scores for days 0 (P=0.001), 21 (P=0.002), and 56 (P=0.013). Correlations were observed between ear cocci counts and ear pruritus scores for days 21 (P=0.008) and 56 (P=0.001). Significant correlations occurred between ear cocci counts and ear erythema scores for days 0 (P<0.003), 21 (P<0.001), and 56 (P<0.001). Significant correlations were observed between ear rod counts and ear pruritus (P=0.025) and erythema scores (P=0.013) on day 0 and for erythema only (P=0.005) on day 56.
Clinical and Cytological Parameters in Treated and Untreated Ears
For dogs (n=14) that were not treated with topical ear medications, significant reductions occurred in mean total ear scores between days 0 and 21 (P=0.041) and between days 0 and 56 (P=0.014). Significant positive correlations were observed between the means of the total ear scores and Malassezia spp. counts for days 0 to 21 (P=0.006) and days 0 to 56 (P=0.004). Rods were not observed in any of the samples for days 0, 21, and 56. No cocci were observed in the samples on day 56. In the dogs (n=44) treated topically with the 2% boric acid and 2% acetic acid cleanser, significant decreases in the means of the total ear score were observed between days 0 and 21 (P<0.001), days 21 and 56 (P=0.012), and days 0 and 56 (P<0.001). Significant decreases in Malassezia spp. counts were observed between days 0 and 21 (P<0.001) and days 0 and 56 (P<0.001). A significant increase (P=0.013) occurred in cocci counts between days 21 and 56. No significant changes were observed for rod counts between days 0, 21, and 56. Significant positive correlations (P=0.010) were seen between mean rod counts and changes in total ear score between days 0 and 56. Significant positive correlations were observed between the changes in total ear scores and both yeast (P=0.056) and cocci counts (P=0.041) for days 0 to 56 (P=0.014). No significant differences for yeast, cocci, or rod counts were seen between the treated and untreated groups on any of the days.
Correlations of Aeroallergen Reactions and Other Parameters
Results of the ELISA assays are summarized in Table 5. Most dogs with positive reactions had both moderate/strong and weak reactions to the allergens in the panel. Of the 72 dogs that entered the study, 22 dogs were defined as seasonally pruritic and 50 dogs had nonseasonal allergies. Analysis of the data revealed that eight of 22 (36%) seasonally pruritic dogs were negative for seasonal allergens using the Northeast Allergy Panel [Table 2]. Eight of 13 (61%) dogs with positive supplemental mold panel reactions [Table 2] were negative for seasonal allergens. The number of positives ranged from two to 11 reactions (mean 7.1). In the nonseasonally pruritic group, 20 of 50 (40%) dogs tested using the Northeast Allergy Panelf were negative for seasonal allergens, and 17 (34%) were negative for all of the nonseasonal allergens. Four of 27 (15%) dogs were negative when the supplemental mold panel was tested. The 17 dogs that were negative for nonseasonal allergens all had positive mold reactions on the supplemental mold panel, ranging from two to 11 reactions (mean 9.1) per panel.
No significant correlations were observed between positive supplemental mold reactionsg or the positive Northeast allergens and the corresponding total skin scores for days 0, 21, and 56. No significant correlations were found when positive reactions to molds, dust mite, and flea allergen were compared with dogs of either the seasonal or nonseasonal pruritus groups. Significant positive correlations were found between positive reactions to the following components of the Northeast allergen panel: nonseasonal allergens with molds (P<0.001, nonseasonal allergens with dust mites (P<0.001), total allergens with seasonal allergens (P<0.001), and molds with flea allergen (P<0.004).
Discussion
In prior fatty acid studies, >50% reduction in pruritus has been reported with fatty acid administration.18–24 When overall improvement was evaluated, clinical responses were variable with reported improvements of 0% to 22% in five studies, 30% to 49% in five studies, and >50% to 82% in three studies.25–32 Comparing one study with another is difficult because of the many variables within each study design. None of these prior studies were controlled for dietary intake of fatty acids. Variations in the source, dosage, and ratio of n-6:n-3 fatty acids, as well as duration of treatment, may have influenced the individual results from these studies. Bathing with antiparasiticidal, antifungal, or antibacterial shampoos was allowed in many of the studies. Different clinical parameters were also assessed. In the present study, attempts were made to minimize the variables by controlling the ratio and total dietary intake of n-3 and n-6 fatty acids, bathing all dogs with the same antibacterial/antifungal moisturizing shampoo to control secondary factors contributing to pruritus, using the same antibacterial/antifungal ear cleansing solution when signs of otitis externa were present, and having one investigator complete all clinical and cytological evaluations.
In the study reported here, the 48.2% decrease in pruritus scores and 36.7% mean clinical improvement for the combined diet groups over the 8-week trial compares favorably with results of previous studies.2527–30 Erythema and miscellaneous lesions (e.g., alopecia, skin oiliness, papules, pustules, scale) did not improve as much as pruritus, thus contributing to a lower overall clinical response [Figure 1]. These results agreed with the findings of a previous study in which erythema, scale, and coat condition changed less than pruritus in dogs treated with both fatty acid supplementation and antihistamines.33 The strong correlation between total clinical scores and erythema and miscellaneous lesions supported the importance of these clinical parameters in the overall assessment of response to supplementation therapy and possibly other therapies. If fatty acid supplementation has less effect on erythema and miscellaneous lesions than on pruritus (as observed in this study), the overall clinical response resulting from fatty acid supplementation may be dependent upon identifying and controlling multiple secondary factors that contribute to the pruritus. Because of the many clinical interactions and variables in this study, it was not possible to determine the specific influence of n-3 and n-6 fatty acid supplementation on the different clinical parameters.
The failure to demonstrate significant differences in total skin scores between diet groups was most likely the result of persistent skin erythema, miscellaneous skin lesions, ear lesions, or the presence of skin and ear yeast and cocci. Based on these results, it may be postulated that fatty acid supplementation has less effect on the modulation of these factors. Another explanation may be that the pruritus was not controlled sufficiently to decrease erythema and miscellaneous or otic lesions.33
Clinical parameter scores between dietary groups were not significantly different (P≤0.05) on any of the study days. However, the percentage changes in scores between days 0 and 56 for various clinical parameters [Table 3] suggested that total dietary intake and/or specific ratios of n-6:n-3 fatty acids may influence the clinical observations. The percentage decreases in all clinical skin and ear scores between days 0 and 56 were the largest for dogs fed Diet C. The least improvement in the total skin score and combined total skin/ear scores occurred with Diet D. Diet B induced the least changes in skin and ear pruritus. The trend for Diet A to have more effect on pruritus than Diet B was not expected, but perhaps the slightly lower ratio of n-6:n-3 had an impact on the modulation of pruritic mediators. The other three diets had a higher ratio of n-3 fatty acids compared to Diet A. The n-6:n-3 ratio of 6:1 for Diet C was similar to the ratio previously reported to have antiinflammatory effects in healthy dogs.34 A larger sample size for each diet group may have revealed more statistically significant changes between groups.
The preponderance of significant positive correlations between total skin and ear clinical parameter scores and cytological observations strongly supported the suspicion that microbial agents contribute to the clinical signs. The results also supported a relationship between erythema and the presence of Malassezia spp. The lack of correlation on day 56 between pruritus and total skin counts for Malassezia spp. or cocci was likely the result of alterations in the surface microenvironment and microbial populations from the regular bathing. The overall improvement of skin parameter scores may also have resulted from regular bathing, since antihistamine and antibiotic treatments were not correlated with clinical improvement. The variable interval between times of bathing and the dermatological evaluations may have influenced individual clinical scores on a given day. The positive effects that the dietary fatty acids may have had on clinical parameters appeared to be diminished by the presence of yeast and bacteria and the persistence of various gross lesions, especially erythema and miscellaneous lesions. These results supported the recommendation for use of long-term adjunctive therapies in the management of the pruritic dog.
Breed of dog, anatomical site, and underlying disease may affect yeast and bacterial counts on the skin.1333–41 The significant correlations of all the anatomical sites (i.e., face, feet, ventrum, axilla, dorsolateral trunk, extensor carpal surface, flexor tarsal surface) and the total skin scores on day 0 were consistent with previously described common locations of atopic dermatitis lesions.24 A reduction in Malassezia spp., cocci, and aeroallergens on the skin surface as a result of weekly bathing may have contributed to the significant reduction of clinical scores over the 8-week trial. The inverse relationship of yeast and cocci counts and clinical improvement on day 56 supported the conclusion that normal resident flora are not altered by topical antifungal/antibacterial shampoos or fatty acid supplementation and, therefore, do not contribute significantly to clinical signs.40 The feet and axillae had the most persistent gross lesions on days 21 and 56, with many significant positive correlations between the site-specific clinical scores, including erythema and pruritus, and yeast and cocci counts. These correlations supported the presumed role abnormal numbers of microorganisms have in contributing to clinical signs.
Several factors in this study may have influenced individual observations. The variable intervals of time between bathing and the cytological evaluations may have influenced individual yeast and bacterial counts. The administration of antihistamines during the first 3 weeks of the trial may have reduced pruritus and normalized the skin surface environment, resulting in decreased microbial counts for part or all of the study period. Antibiotics administered during the first 3 weeks may have reduced the population of cocci. The ratio of surface bacteria to yeast may have been altered sufficiently to result in an increase in yeast counts. Inadequate contact time of the shampoo on the feet and the axillae may have affected clinical parameters in these anatomical sites.
The higher recovery of Malassezia spp. using the tape technique compared to scrapings may have diagnostic significance. In the areas that are more difficult to access, such as interdigital regions, the tape technique often allows for easier sampling of the deeper crevices.13 In areas with heavy sebaceous accumulations, scraping may result in recovery of higher numbers of Malassezia spp. because of increased exposure of the superficial cells of the skin. Because bacteria are frequently found adhered to the superficial corneocytes, the higher recovery of cocci with scrapings was expected. A large variability in the recovery of yeast and bacteria between sampling sites was experienced and verified not only by large standard deviations of the mean counts, but by a lack of significant differences between techniques on any specific day, despite significant differences in the combined counts for the three sampling days. These variations supported the recommendation that if one technique does not yield expected results, resampling using another technique is indicated.13
The significant decrease in total ear scores on days 21 and 56 in the untreated ear group was consistent with the milder form of otitis externa that commonly accompanies systemic allergies. The influence that yeast and bacteria have on otitis externa was strongly supported by their correlations with ear pruritus and erythema. The 2% boric acid and 2% acetic acid ear cleanser used in this study appeared to be effective in controlling Malassezia spp., but it was less effective in decreasing bacterial populations. The significant decrease in total ear scores between days 0 and 21 and days 0 and 56 may be attributed to both local treatment with the ear cleanser and the overall decrease in the skin clinical parameters.
Most of the dogs in this study fulfilled the criteria for a diagnosis of atopy, although no significant correlation existed between positive allergen reactions and the clinical scores.4243 The large differences in specific reactions to either seasonal or nonseasonal allergens and the clinical history supported the proposition that allergy testing is not a definitive diagnostic test of atopy.78 When performing allergen testing, selection of additional mold test panels is often indicated in the nonseasonal pruritic dog, especially if there is poor correlation between reactions to nonseasonal allergens and the clinical history. This recommendation was supported in the study reported here by the high percentage (85%) of positive reactions on the supplemental mold panel compared to the low number of mold reactions (15%) on the Northeast Allergy Panel in dogs exhibiting nonseasonal pruritus. The previously reported correlation between positive mold allergen reactions and skin yeast infections was not observed in this study.4 The limitations of ELISA testing indicate a need to identify and control the primary and secondary problems associated with the clinical signs in order to achieve maximum clinical improvement.
Since all animals in the study were client-owned dogs and had pruritus as the primary complaint, symptomatic treatment with regular bathing was deemed necessary. A commercial shampoo containing 2% acetic acid and 2% boric acid in a soapless shampoo vehicle with moisturizers was selected, because it had antifungal and antibacterial activity but contained no antiinflammatory agents. The variation noted in the yeast and cocci counts from the different anatomical sites may have resulted from differences in adhesion factors, in normal resident counts, and from the bathing itself. By lowering the pH on the surface of the skin, the shampoo may have affected the colonization of both resident and pathogenic microorganisms.44 The initial significant correlation between Malassezia spp. and pruritus scores, and the subsequent lack of correlation in the predisposed anatomical sites (i.e., axillae, dorsal and ventral interdigital) may have been a reflection of lower surface moisture. With the improvement of clinical parameters (i.e., pruritus, erythema, miscellaneous lesions), the surface skin environment may have favored the regrowth of normal numbers of resident Malassezia spp. and cocci. No predictable pattern was observed when comparing time of bathing with the skin pruritus and erythema scores or with skin microbial counts.
In this study, significant correlations between improvement in clinical signs and total dietary n-3 and n-6 fatty acids, n-6:n-3 ratios, antibiotics, antihistamines, and reactions to aeroallergens were not observed when all dietary groups were combined. The potential favorable influence of fatty acid supplementation on clinical signs must still be considered for several reasons. Firstly, all of the diet groups had significant overall improvements in combined skin and ear scores. Secondly, there were differences between the groups in the percentage changes of various clinical parameters over the 8-week study. Finally, the combined potential effects of routine bathing on the surface micro- and macro-environment and the contribution of improved otic signs to overall clinical improvement are difficult to measure. Additional studies are needed to determine the influence of specific ratios and total dietary intake of n-3 and n-6 fatty acids, and to assess the role of fatty acid supplementation on the long-term prognosis of allergic disease.
This study had several significant limitations. A food allergy was not ruled out in all dogs, so a lack of clinical response to the study diets could have been the result of an adverse food reaction.4445 The use of subjective clinical assessments for evaluating the effects of dietary supplements or medications (either topical or systemic) was another limitation. Similarly, differences might have occurred in the performance of sampling techniques and cytological evaluations. Microbial flora varies considerably among different breeds and anatomical sites, as does the inflammation associated with the yeast and bacterial infections.34374041 Since bathing was performed by the owners, techniques probably differed between dogs and, therefore, may have affected the subjective clinical evaluations and cytological observations. The compliance of each owner in the feeding of a measured amount of food or the proper application of the ear cleansers could not be documented by the investigators. The clinical scoring protocol used in this study did not distinguish between a dog with a few severe lesions and one with more disseminated lesions of lesser severity. It is possible that the initial response of the severe lesions to treatment may have been more rapid than the response of less severe lesions, thus resulting in different clinical scores. The significant positive correlations between clinical signs and microbial populations on the skin, as well as the total body and site-specific clinical scores, highlighted the difficulties of conducting clinical trials to evaluate specific responses to one dietary supplement or treatment.
Conclusion
The results of the current study demonstrated strong correlations between clinical manifestations (pruritus, erythema, alopecia, papules, pustules, scale, and oiliness) and skin and otic yeast and bacterial populations. The measurable influence of specific n-6:n-3 dietary fatty acid supplementation on overall clinical improvement and on pruritus was diminished by the multiple clinical manifestations observed in the pruritic dogs. This study highlights the problems associated with conducting a clinical trial to evaluate a response to specific fatty acid supplementation, while controlling for the multiple secondary clinical manifestations in client-owned dogs. Additional studies are needed to determine the influence of specific ratios and total dietary intake of n-3 and n-6 fatty acids, and to assess the role of fatty acid supplementation on the long-term prognosis of allergic disease.
Acknowledgments
The authors acknowledge the generous donation of products and services from DermaPet, Inc., and Heska Corporation. The authors also gratefully acknowledge the assistance in statistical analyses by Wendell Kerr, Senior Research Statistician, Nestlé Purina Product Technology Center and Jay Harrison, Senior Statistician, Ralston-Purina Company.
Vetalog; Fort Dodge Animal Health, Overland Park, KS 66210
Depo-Medrol; Pharmacia & Upjohn, Kalamazoo, MI 49003
Malacetic shampoo; DermaPet, Inc., Potomac, MD 20854
DermaPet Ear/Skin Cleanser; DermaPet, Inc., Potomac, MD 20854
Diff-Quik; American Scientific Products, Gibbstown, NJ 08027
Allercept; Heska Corporation, Ft. Collins, CO 80525
Canine In-Vitro Test; Greer Laboratories, Lenoir, NC 28645
SAS v. 8.2, SAS Institute, Cary, NC 27511
Advantage; Bayer Animal Health, Shawnee Mission, KS 66201
Frontline; Merial Limited, Duluth, GA 30096
Program; Novartis Animal Health, Greensboro, NC 27408
Biospot; Farnam Pet Products, Phoenix, AZ 85076
Citation: Journal of the American Animal Hospital Association 40, 4; 10.5326/0400270
Citation: Journal of the American Animal Hospital Association 40, 4; 10.5326/0400270
Citation: Journal of the American Animal Hospital Association 40, 4; 10.5326/0400270
Mean skin clinical scores for all dogs on days 0, 21, and 56. Total skin score (——), total skin pruritus score (– – – –), total skin erythema score ( • • • • ), and total skin miscellaneous lesion score (–• • – • •). * P≤0.05 for the change between days 0 and 56.
Comparison of mean skin cocci counts obtained by tape (black bar) and scraping (white bar) techniques on three different days and the overall results from combining all the counts obtained during the study. The mean differences between the two sampling techniques are represented by the gray bar plus the standard error of the difference. * P≤0.05.
Comparison of mean skin yeast counts obtained by tape (black bar) and scraping (white bar) techniques on days 0, 21, and 56 and overall counts obtained by combining the counts for all 3 days for the tape (black) and scrape (white) sampling techniques. The mean differences between the two sampling techniques are represented by the gray bar plus the standard error of the difference. * P≤0.05.
