A novel therapeutic diet can significantly reduce the medication score and pruritus of dogs with atopic dermatitis during a nine-month controlled study.

BACKGROUND: Canine atopic dermatitis (cAD) is a common chronic relapsing pruritic skin disease for which management commonly relies on life-long use of immunomodulatory drugs. A number of the medications used are associated with adverse effects and the potential for complications during long-term use. HYPOTHESIS: The goal of the study was to determine if a complete and balanced diet formulated for therapeutic benefit could contribute towards management of cAD. We hypothesised that the diet would reduce pruritus while also reducing the requirement for medication during the study period. ANIMALS,
MATERIALS AND METHODS: Forty privately owned dogs, having undergone a comprehensive diagnosis for cAD, were randomly allocated to two groups, each group being fed one of two diets (test or control) for up to nine months. We assessed pruritus, Canine Atopic Dermatitis Extent and Severity Index-(4th iteration) and medication score, the latter reflecting the medication required to maintain a satisfactory quality of life for the animal.
RESULTS: Both diets were well-accepted and -tolerated. There was a significant improvement in the pruritus score after three months of feeding the therapeutic diet (P = 0.0001). No such improvement was observed at any time point in the group of dogs given the control diet. There was a reduced drug requirement for dogs receiving the therapeutic diet after three months (P = 0.058), and that decrease was significant at six months (P = 0.021) and nine months (P = 0.018). No improvement was seen at any time point in the control group.
CONCLUSION: The results suggest that a novel therapeutic diet can assist in the management of cAD by helping to control pruritus and reducing reliance on medication.

Introduction

Canine atopic dermatitis (cAD) is a multifactorial disease associated with immunological dysregulation and skin barrier abnormalities. Treatment is based mainly on the use of anti‐inflammatory and/or immunomodulatory drugs, while the use of epidermal barrier treatments is mainly considered supportive. Immunomodulatory drugs unfortunately may be associated with detrimental adverse effects. It is well‐known that dietary components or additives may positively influence the course of human atopic dermatitis (AD). The influence of these phytochemicals on allergic reactions has been studied in some detail and potential efficacy is established. In veterinary medicine, the role of essential fatty acids and some anti‐oxidants has been assessed. , , Dietary supplements often are advocated for mild to moderate atopic disease as part of a multimodal approach, and some studies have shown a benefit. However, administration increases the burden of treatment, requiring dosing at minimum once per day. Therefore, incorporation of helpful ingredients into a complete meal format can add to convenience, with the potential to see concomitant improvements in compliance. What is more, it has been illustrated that food‐integrated nutrients are more effective than supplements. While some diets have been developed for the management of AD in dogs, well‐controlled studies are still lacking.

The goal of this study was to assess the clinical efficacy of a diet specifically developed to target the pathological mechanism of cAD to assist in the long‐term management of this condition.

Methods and materials

Owner consent, Ethics and General Data Protection Regulation (GDPR) statement

The protocol was approved by the local Kanton Zurich ethical committee (ZH114/16) as well as the Royal Canin Ethical Research Committee. An informed consent form was signed by all dog owners who participated. Any personal data relating to participants in the study was the minimum required and was held only for as long as needed to execute the study, according to GDPR.

Dogs

Privately owned atopic dogs with nonseasonal pruritus were included. All dogs were treated in private practice clinics. The diagnosis of AD was made using standard criteria, after exclusion of ectoparasites and other complicating diseases such as bacterial and yeast infections. Dogs with active bacterial and/or yeast infections, as well as individuals responding adequately to an eight weeks elimination diet and re‐challenge with previous diet, namely, dogs classified as food allergic were not included. Moreover, dogs with ongoing allergen‐specific immunotherapy for <12 months or dogs controlled with this intervention alone were not enrolled. Included dogs needed to be receiving ongoing ectoparasite treatment and this treatment was continued throughout the study. Concomitant medications (see Appendix 1 for list), excluding all nutritional supplements, were allowed throughout the study to comply with ethics and ensure an optimal quality of life (QoL), and were carefully recorded throughout.

Inclusion consultation and initial clinical scoring

All enrolment and clinical evaluations were performed by qualified and experienced veterinary dermatologists. Included dogs were initially evaluated for skin lesions and pruritus using Canine Allergic Dermatitis Extent and Severity Index, 4th iteration (CADESI‐04) and Hill’s pruritus Visual Analog Score (pVAS), respectively. Medications used during the month before inclusion were recorded. A medication score was made as published previously. Briefly, as depicted in Table 1, monthly treatments were recorded, with the specific drug, dose and frequency of use being attributed a numerical score. The overall score for each month was determined by summing daily scores. The figures reported for medication score are the monthly totals divided by the number of days in the month, thereby providing a daily average.

Table 1

Medication scoring system (as described previously ) used to grade medication consumption by dogs at the start of the study and at each subsequent check point

Medication and dose range	Score attributed
No concurrent medication	0
Shampoo therapy	5
Ear medication (topical)	5
Other topical therapy	5
Antihistamines	10
Frequent antibiotics (>21 days)	20
Less frequent antibiotics (<21 days)	10
Prednisolone
≥1 mg/kg/day	40
0.5–1 mg/kg/day	30
0.2–0.5 mg/kg/day	20
≤0.2 mg/kg/day	10
Ciclosporin (5 mg/kg)
Once each day	30
Every other day	20
Every three days	10
Every four days	5
Oclacitinib
Twice daily	40
Once each day	30
Every other day	20
Every three days	10

Follow‐up consultations and drop‐out

Follow‐up consultations were made after one month (phone call), three, six and, when possible, nine months. The one month consultation was aimed at recording any problems associated with the study, for example diet palatability, in addition to recording the medication and pruritus scores. Consultations after three, six and nine months consisted of general examination of the dogs, CADESI‐04, pVAS and medication score. At each time point the consumption of the diets was reported by the owners using a binary yes/no question with comments if needed. In addition, throughout the study owners were asked to evaluate the quality of the dogs' stools in order to assess how well the diet was being tolerated. Stool quality was assessed according to a five‐point scale, illustrated by picture (grade 5, hard; 4, well‐formed, easy to pick up; 3, well‐formed but soft; 2, very wet but not liquid; and 1, liquid diarrhoea).

Owners were permitted to withdraw their dogs from the study at any time and for any reason. They were encouraged to feed their dogs exclusively with the diet provided for at least three months. In the event of a dog being withdrawn, owners who were not satisfied with the efficacy of the diet were given the opportunity to swap to the test diet for the remainder of the study period. In this case the diet provided was renamed (code Pebble) to maintain blinding. Drop‐out reasons were recorded. Dogs maintained on any of the diets for a minimum of three months were considered for statistical analyses.

Diets

The study was double‐blinded and placebo‐controlled. Enrolled dogs were randomly allocated into one of the two study groups. One group of dogs received the test diet (code Flame) while the other group received a standard premium diet (placebo: code Ice). Both of the diets contained chicken as the protein source and were produced at a commercial manufacturing facility in France. The diets were dry kibble format (Figure 1), and were complete and balanced for dogs at maintenance energy requirement (95 kcal/kg0.75 NRC 2006). A comparison of the key nutritional components of the two diets can be found in Table 2. The diets were packaged in identical neutral bags, differentiated only by their code names. Both diets were composed of standard pet food materials, the test diet was supplemented with turmeric and licorice extracts. The turmeric extract (Curcuma longa) was provided by Arjuna (Kerala, India) and contained 95% curcuminoids with enhanced bioavailability. The licorice extract (Glycyrrhiza glabra), supplied by Naturex (Avignon, France) was prepared from the root of the plant. The extract was standardised for glycyrrhizin concentration (12% w/w). The concentration of the omega‐6 fatty acid linoleic acid in the Flame diet was increased by replacing a proportion of poultry and pork fat in the Ice diet with soya oil and fish oil. The omega‐3 fatty acids were supplemented using the fish oil. As stated previously, neither the owners nor the investigators were informed as to the nature of the diet. Owners were instructed to feed only the study diets provided for the entirety of the study. When dogs were transitioning from their previous diet to the study diets, owners were advised to change gradually over six days (two days each of 75%:25%, 50%:50%, 25%:75%).

Figure 1

Physical appearance of dry format diets.

As Pebble diet was the same as the Flame diet, it is not shown here.

Table 2

Concentrations of key nutritional components of the two diets fed during the study. Details of the diets are given for the proximates and a group of other ingredients which differed significantly between the Test (Flame) and Control (Ice) diets

	Units	Flame	Ice
Dry matter	%	90.50	90.51
Moisture	%	9.50	9.49
Protein	%	22.54	22.51
Fat	%	13.97	13.99
Ash	%	6.90	6.89
CFIB (crude fibre)	%	2.90	2.90
Linoleic acid	%	3.90	2.20
EPA+DHA	%	0.54	0.01
Vitamin E	mg/kg	900.64	162.14
Taurine	mg/kg	4500	600
Lutein	mg/kg	5	0.74
Curcuma extract	mg/kg	300	0
Licorice root extract	mg/kg	200	0

EPA, eicosapentaenoic acid, DHA, docosahexaenoic acid.

Outcome measures

Drop‐out reasons and unexpected events were recorded. The percentages of dogs with >30% and >50% improvement of the clinical score were computed for both groups. Additionally, the number of dogs returning to normal (CADESI‐04 < 16, pVAS < 2.5 and medication score < 10) were recorded in both groups.

Statistical methods

Statistical analyses were performed using JMP v14 (JMP; Cary, NC, USA). Linear mixed models were used for evaluating the impact of diet, time and time‐by‐diet interactions on CADESI‐04, pVAS and medication score. Animal was modelled as random term.

Log‐transformation was used when necessary to meet the statistical assumptions of a linear mixed model (normally distributed residuals and homoscedasticity). Tukey's honestly significant difference (HSD) was used for the post hoc multiple comparisons. Level of significance was set at 5%. Medians (min–max) were provided for the studied parameters.

Results

Study population detail

Forty dogs were initially included and allocated to one of the two study groups, one to be fed the test diet Flame (n = 21) and the other to be fed the placebo diet Ice (n = 19). The male:female ratio was 21:19 and the mean age was 5.2 years (SD ±1.78). No age or sex difference was found between groups. Some breeds were over‐represented, in particular French bulldogs (n = 11), West Highland white terriers (n = 3), Labrador retrievers (n = 3) and American Staffordshire terriers (n = 3). Multiple dogs of the same breed were present in both groups.

At inclusion, the median CADESI‐04 scores in the Flame and Ice groups were 19 (11–59) and 18 (6–69), respectively. The median pruritus scores were 5.5 (2.8–6.8) and 4.6 (3–8.5) respectively. The medication scores were 20 (10–45) and 30 (5–60), respectively. There was no significant difference between the groups for any of the three parameters at inclusion.

Six dogs were not considered for subsequent analysis because they did not receive the diet for the required minimum of three months. Three were in the Flame group (one refused the diet, one developed persistent diarrhoea five weeks after diet transition, and one developed mammary tumours). Three were in the Ice group (one dog refused the diet, two were lost to follow‐up). Six owners elected to feed their dogs with the Pebble diet after three months of feeding (n = 5; four Flame, one Ice) or six months (n = 1, Ice).

One month data were available for 36 dogs (19 Flame, 17 Ice), while three month data were analysed for 32 dogs (17 Flame, 15 Ice). Six month analyses were made on 24 dogs (12 Flame, 12 Ice) and nine month analyses on 21 dogs (14 Flame, 7 Ice). There was no significant difference between the two diet groups for dropout frequency (P = 0.75).

Score comparisons

Pruritus (pVAS)

There was clear evidence of an improvement in the pruritus score for the Flame diet group after three months compared to the start of the study, the median score declining from a baseline value of 5.5 to 3.0 (ranges 3–7.6 at baseline and 0–6 at three months; P =0.0001). There was no significant difference between pVAS within the Flame group between 0 and six months (median 4.35 to 2.13), but the score reduction was significant by nine months from baseline 5.1 to 2.8 (3–6.3 and 1–3.8, respectively; P = 0.0043). Within the Ice group, no changes in pVAS were observed between any time points. Figure 2a shows the change in pVAS during the nine month study period. Variation in baseline pVAS was a consequence of cases leaving the study before each time point.

Figure 2

Standard boxplots (showing median with 25th to 75th percentiles) for clinical parameters assessed for two diet groups over the nine month feeding period.

Case numbers at each time point, Flame/Ice: time zero 21/19; three months 17/15; six months 12/12; nine months 14/7). Scores as a function of time and diet using: (a) Pruritus Visual Analog Score (pVAS); (b) medication; and (c) Canine Allergic Dermatitis Extent and Severity Index, 4th iteration (CADESI‐04).

Medication score

The medication score was used to determine whether and to what extent the diets were able to reduce the dependency on nondietary interventions in order to maintain an acceptable QoL. After three months, the median medication score had decreased from 20 to 10 (ranges 10–30 at baseline and 0–30 at three months; P = 0.058) for the Flame diet group. By contrast, in the Ice diet group the median medication score remained the same between baseline and three months at 25 (5–50 and 5–60, respectively; NSD). After six months the medication score showed a significant reduction in the Flame group (median 20 versus 10, ranges 10–30 and 0–20; P = 0.021), which was maintained up to nine months (20 versus 5, ranges 10–30 and 0–20; P = 0.018). In addition, the average medication scores at six and nine months were significantly lower than those at one month (P = 0.050 and P = 0.039, respectively). No significant changes in scores were observed for the Ice diet group between any of the time points. Figure 2b shows the medication score during the nine month study period.

CADESI‐04

There was a reduction in median CADESI‐04 score from 16 (10–59) at the beginning of the feeding period to 12 (2–36) after three months in the Flame diet group. The median score in the Flame group was 14.5 (4–73) at six months, and then 8 (4–40) after nine months. In the Ice group, the median CADESI‐04 after three months of feeding decreased from 18 (6–69) to 13.5 (4–48), increasing to 21 (1–84) after six months, before dropping again to 15 (0–44) by the end of the study. None of these figures showed a significant change from baseline CADESI‐04 values. Figure 2c shows CADESI‐04 scores for both diet groups during the nine month study period.

Score improvement

At each time point the number of dogs showing a > 30% and >50% improvement of the clinical scores were computed (Table 3). The analysis shows that for 11 of 11 data points (CADESI‐04, pVAS and medication score combined) there was a larger proportion of cases that improved >30% in the Flame group than in the Ice group. This also was seen for nine of 11 of the data at the >50% threshold. Table 3 also illustrates that, in the Flame diet group, there was typically a progressive increase in the proportion of dogs showing >30% and >50% improvement over time. The pattern was not repeated in the Ice diet group.

Table 3

Improvement of the clinical score for both diet groups: the percentage of cases at each assessment point which showed either a > 30% or a > 50% improvement in pruritus (Visual Analog Score, pVAS), Canine Allergic Dermatitis Extent and Severity Index, 4th iteration (CADESI‐04) or medication score

Flame	One month	Three months	Six months	Nine months	Ice	One month	Three months	Six months	Nine months
pVAS					pVAS
>30%	47	82	46	64	>30%	17	36	31	36
>50%	21	29	46	60	>50%	0	21	8	18
CADESI‐04					CADESI‐04
>30%		44	42	72	>30%		37	31	42
>50%		22	29	55	>50%		25	23	25
Medication					Medication
>30%	10	53	64	82	>30%	6	7	8	18
>50%	5	29	36	73	>50%	6	7	8	9

Return to normal

Finally, it is of note that six dogs eventually achieved a normal score for the three parameters monitored in the Flame group, whereas just one dog attained this level of improvement in the Ice group.

Dogs transferred to Pebble diet

Owners elected to change the diet for six dogs. Data for these dogs were included in the analysis before but not after the change was performed. Dogs which transitioned to Pebble did not show a significant improvement: it should, however, be taken into account that four of these dogs were already on the same Flame diet and that, in the main study, it was shown that an improvement was present after three months. This suggests that stabilisation of the improvement could be expected, and was indeed the case for three of four dogs. One of the two dogs in the Ice diet group improved after switching to the Pebble diet.

Palatability and digestive tolerance

As mentioned in the Study population section above, one dog in each group refused the diet. Owners reported that the diets were well‐accepted by all other dogs. In the Flame group, one dog developed chronic diarrhoea several weeks after the diet transition. This dog was removed from the study; causality was not established. In this group, one dog had very dry faeces (grade 5), 10 had grade 4 (optimal according to the scale used) and six had grade 3 faeces. In the Ice diet group, the stools were considered grade 5 in two dogs, while 14 had grade 4 and one grade 3 faeces.

Discussion

This study demonstrates that a commercially‐prepared diet supplemented with a targeted mixture of antioxidants and with a modified fatty‐acid composition reduces pruritus and has a drug‐sparing effect on dogs with AD. There was modest itch improvement following one month of feeding the diet, although statistical significance was reached only after three months. The lesional scores showed improvement in both groups at certain times but statistical significance was not reached. This is probably because the study was somewhat underpowered for CADESI‐04, accentuated by the fact that only moderate to mild cases were included, thereby reducing the scope for improvement of CADESI‐04 score. The main purpose of the study was to achieve itch improvement with reduced requirement for medication, factors likely to positively impact the QoL of affected patients as well as their owners.

The clearly illustrated drug‐sparing effect also is of value because drugs used for the control of the clinical signs of cAD may be associated with adverse effects, compliance issues and, in some cases, considerable financial cost. The reduction of the medication score was evident after three months and became significant from six months onwards. It should be emphasized that AD is a life‐long disease with early onset of clinical signs. In this regard, any safe, easy‐to‐use intervention capable of decreasing the dose or dose‐duration of potentially problematic drugs is welcomed.

Both diets tested in this study were formulated to deliver high‐quality maintenance nutrition and were comparable in all aspects apart from a small group of ingredients included in the Flame test diet to target mechanistic aspects of canine atopic dermatitis. Amongst these, vitamin E and lutein provide antioxidant benefits to the skin. , In support of this activity, lutein has been found to concentrate in the skin following oral consumption in humans. The essential omega‐6 polyunsaturated fatty acid linoleate is known to play a vital role in maintaining the structural integrity of the epidermal barrier, intrinsic as it is in the formation of certain ceramides. What is more, linoleic acid intake has been shown to influence aspects of integument performance as well as skin lipid composition in dogs. , Taurine has been reported to contribute to keratinocyte hydration, thereby also contributing to barrier function. The theoretical immunomodulatory role of the test diet is provided by three components: omega‐3 fatty acids, turmeric and licorice. Multiple studies have indicated a role for the omega‐3 fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in the management of canine skin inflammation. , Turmeric is a source of phenolic compounds known as curcuminoids. These have significant anti‐inflammatory capacity, particularly in relation to cell‐mediated immunoreactivity. The inclusion of licorice root extract could provide an immunomodulatory benefit. A constituent of licorice, the triterpenoid saponin glycyrrhizin, has demonstrated the ability to suppress interleukin (IL)‐4 levels and restore the immune balance of T helper (TH1/TH2) cells in a mouse allergy model. In addition, it was shown that glycyrrhizin attenuates B‐cell production of allergen‐specific immunoglobulin (Ig)E and IgG1 and can act as a mast cell stabilizer. These all are activities which have potential value to influence the immune dysregulation underlying cAD. In combination, this group of functional ingredients appears to have value in the management of disease symptoms.

Previous studies have indicated the value of a dietary approach to the management of canine allergic skin disease. Success with respect to food‐derived allergies has been especially impressive. , In cases where the allergy is at least partly environmentally derived, results have been more variable. A recent review elegantly summarises much of the work to date. A common issue highlighted is that many of the purported nutritional or nutraceutical‐based therapeutics still lack efficacy data derived from well‐designed, placebo‐controlled studies.

There was excellent acceptance and digestive tolerance of the diets used in the study. The test diet was refused by only one dog and one other developed chronic diarrhoea, although no association between the diet and the gastrointestinal signs was established. There was some improvement in all scores in the group of dogs using the control diet, although statistical significance was not reached. The moderate improvement in the control group could have been a consequence of the benefits of providing high‐quality nutrition to all participants.

It was notable that the main improvements in the test diet group were observed after three months. It is not surprising as other studies have shown that nutritional therapies can often take some time to build to maximal skin effects. , Classical drug administration would consequently still be important during the early phase of dietary treatment in dogs suffering from severe AD. It is possible, however, that the diet alone may be sufficient to adequately control the clinical signs of atopic dogs with mild or moderate clinical signs. Further studies are needed to confirm this potential benefit. The results from this study were encouraging over the nine month duration. The number of cases lost before each time point was not ideal, reducing the number of cases completing the full study. What is more, it may have been valuable to increase the study period to establish whether the diet could maintain the reduced medication requirement for an extended period. A further limitation of the study was inclusion of mild and moderate cases only. It would have been interesting to see if the diet also could contribute to the management of severely atopic dogs.

Conclusion

In summary, to the best of the authors' knowledge, this is the first time that a diet designed specifically to assist in the management of cAD has been shown to be efficacious in the framework of a double‐blinded, placebo‐controlled study. Importantly, it is seen that the QoL of atopic dogs could be improved using the dietary intervention and that the approach might mitigate against complications associated with drug therapy, providing significant benefit in a relatively short time frame.

Author contributions

Adrian Watson: Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Resources, Validation, Writing‐original draft, Writing‐review & editing. Ana Rostaher: Data curation, Investigation, Methodology, Resources, Writing‐review & editing. Nina Fischer: Data curation, Investigation, Methodology, Resources, Writing‐review & editing. Claude Favrot: Conceptualization, Data curation, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing‐original draft, Writing‐review & editing.

Appendix 1: List of medications permitted for management of AD during the study

Click here for additional data file.