Abstract
Objectives
Topical treatments, along with systemic oral antifungals, are the mainstay of infection and environmental control for cats with dermatophytosis in animal shelters. This study aimed to provide realistic expectations of the performance of three commonly used topical treatments to help shelters minimize the length of stay and optimize feline welfare.
Methods
A prospective, randomized, semi-blinded clinical trial compared treatment success and time to cure for miconazole chlorhexidine (MC) and hydrogen peroxide (HP) shampoos vs lime sulfur (LS) using a non-inferiority model. Cats with Microsporum canis were enrolled at four US animal shelters, treated with oral itraconazole and randomized into three topical treatment groups. Treatment success was defined as mycological cure by 7 weeks, while inferiority was defined as requiring more than 1 additional week to achieve cure.
Results
A total of 76 cats were enrolled. LS significantly outperformed both alternatives in cure by 7 weeks. Time to cure analysis showed significant differences between LS (mean 27 days, range 7–45) vs MC (37, 14–62) (P = 0.04) and LS vs HP (36, 11–65) (P = 0.06). Although alternative products took significantly longer to cure, confidence intervals (CIs) around the difference of means included the inferiority margin: LS vs MC (9.66, 95% CI 0.7–18.5) and LS vs HP (8.54, 95% CI 0.44–16.6). Therefore, inferiority of the alternative products was suggestive but inconclusive. After controlling for confounders, a Cox proportional hazards analysis confirmed significantly poorer performance of MC (P = 0.003) and HP (P = 0.032) vs LS. Younger age also significantly prolonged treatment (P = 0.039), while intake type, co-housing and low body condition score did not. Shelter staff ratings showed no differences between products in terms of treatment unpleasantness, difficulty or cat reactions.
Conclusions and relevance
In this study, LS outperformed HP and MC in treating cats with M canis. Younger cats took a longer time to achieve cure. If shelters or practitioners wish to use an alternative topical treatment to LS, then HP should be considered.
Plain language summary
Dermatophytosis (ringworm) continues to be a challenge to treat in shelter cats and kittens. This multi-institutional, randomized, partially blinded clinical trial investigated the success of two alternative topical treatments (nazole chlorhexidine (MC) and hydrogen peroxide (HP) shampoos) against the standard lime sulfur (LS) solution in shelter cats receiving oral itraconazole. The authors defined ‘success’ as cure by 7 weeks and set an inferiority margin at a 7-day difference in time to cure between products. IDEXX Labs was used for all fungal cultures. Many cats in the alternative treatment groups failed to cure by 7 weeks: nine (37.5%) in the MC group and six (21.7%) in the HP group; all cats in the LS group were cured in 7 weeks. The average time to cure for MC was 37 days, and 36 for HP vs 27 for LS. The mean differences in times to cure were >7 days: 9.66 days longer for MC and 8.54 days longer for HP, but both confidence intervals included 7 days. So, our test for inferiority was suggestive but inconclusive. A proportional hazards model was created to account for factors that might contribute to differences in treatment time. Of the factors included, only cat age and treatment group showed significance. Older cats and those treated with LS were cured faster. Other factors did not seem to matter: shelter site, co-housing, season of intake, body condition and shelter intake type. Many people resist using LS because of its noxious odor and the perception that cats dislike LS treatment. No treatment type was appreciated to be more difficult, more unpleasant or more aversive to cats than any other. Caretakers in our study reported bleached hair (HP), itchiness (HP and LS) and hypothermia (MC and HP), but no significant side effects from any treatment type. Our recommendation is for shelters to use LS when treating dermatophytosis. If an alternative treatment is needed, HP should be considered over MC, owing to faster treatment and cure times.
Introduction
Microsporum canis is the most common cause of dermatophytosis in cats. 1 Despite generally being a mild, superficial infection, it is of particular importance in animal shelters because of its prevalence, zoonotic and outbreak potential, and the cost and duration of treatment.2,3 Dermatophytosis primarily affects kittens,4 –6 resulting in isolation, confinement and aversive treatments during critical stages of social development. Shelter treatment protocols prioritize a high cure rate, rapid cure, minimizing contamination and mitigating transmission risks.2,7 To achieve these goals, both systemic and topical antifungal treatments are required.2,7,8 Itraconazole and terbinafine are the safest and most effective systemic treatments;9 –11 itraconazole is the only US Food and Drug Authority-approved oral treatment for cats with M canis. 11
Topical antifungal therapy destroys infective fungal spores on hair shafts and is an important adjunct to systemic treatment, which kills the fungal organisms within the skin and hair follicle.7,8,12,13 Twice-weekly topical drenches of lime sulfur (LS) or enilconazole have been recommended for shelter cats.8 –10,14,15 Both have residual antimicrobial action against dermatophytes on the haircoat after application, which appears important to facilitate rapid cure, to avoid positive cultures from cats without active infections and to prevent environmental transmission.7,8 Enilconazole is available in Europe but no longer in North America; veterinary LS is no longer available in Canada except in a compounded form or with special dispensation. Alternatives to LS have also been sought because of its unpleasant odor, corrosion of equipment and yellow coat discoloration. In the authors’ experience, both treatments are unpleasant for cats and caretakers.
Two widely available alternative topical agents have shown promise against M canis dermatophytosis in cats. Miconazole chlorhexidine (MC) has sporicidal effects in vitro 16 and is recommended alongside LS and enilconazole in the feline literature.12,13 It provided faster mycological cure when used with systemic griseofulvin compared with systemic therapy alone17,18 and was superior to single-product miconazole or chlorhexidine products. 19 The only published efficacy study found a poorer 6-week cure rate and a slower rate to mycological cure for MC compared with LS in cats receiving oral itraconazole; several cats went to rescue treatment with LS. 9 Hydrogen peroxide (HP) products have shown sporicidal effects as an environmental disinfectant and in vitro,20 –22 however, they have not been tested in vivo. Despite anecdotal reports that many shelters use MC and HP products, there are concerns about the lack of residual antifungal activity and the lack of evidence for therapeutic equivalence.
The aim of this study was to investigate the efficacy of alternative treatments to LS using a non-inferiority model, allowing for mathematical assessment of clinically relevant differences. The primary objective of this randomized, semi-blinded prospective clinical trial was to compare times to cure for MC shampoo and HP shampoo vs LS when treating M canis dermatophytosis in shelter cats receiving oral itraconazole. A secondary objective was to briefly collect subjective impressions of each treatment from shelter staff, and to report any adverse effects.
Materials and methods
Four shelters (in Washington, Oregon, California and South Carolina) with a high intake prevalence of dermatophytosis and well-established dermatophytosis isolation and treatment wards agreed to enroll cats. Each shelter’s dermatophytosis treatment team received pretrial online training, aligned protocols, treatment products, as well as follow-up with the study manager and blinded investigators.
Healthy, social domestic cats aged 6 or more weeks admitted between October 2022 and December 2023 to the four recruited shelters were included in the study when they met the criteria for M canis dermatophytosis. These criteria were: growth of 10 or more colonies (‘P3’) on their first pretreatment fungal culture or six to nine colonies (‘P2’) of M canis on the first fungal culture and subsequent positive cultures. 8 Cats were excluded if their first fungal culture had five or fewer colonies (‘P1’), if they were diagnosed with other dermatophyte species or if they had received any previous antifungal treatment. Cats were removed from the study if treatment seriously impacted their welfare.
Sample sizes were calculated using the published mean time to cure for cats treated concurrently with oral antifungals and LS, approximately 25 ± 7 days.9,14 An 80% power calculation, with an a priori assumption that 7 days longer mean time to cure represented a meaningfully inferior product in a shelter, gave a minimum sample size of 13 cats per group. 23 Given the lack of data available and the wide range in treatment times reported, we aimed for 20–25 cats per treatment group.
Skin examination and Wood’s lamp testing of all cats were conducted at intake by trained staff at each shelter. Wood’s lamp-positive cats and their littermates were flagged for enrollment. Each dermatophytosis suspect was sampled, assigned an individual study number and block-randomized to receive one of three study treatments. Littermates were split into pairs or trios and sorted to receive the same treatment during the randomization process so that co-housing would not cause cross-contamination.
All diagnostic and in-treatment hair samples were submitted via courier to IDEXX Reference Laboratories using the 3685 Ringworm PCR plus fungal culture test code. Samples were processed immediately according to the laboratory’s proprietary specifications. Culture plates were examined on day 10, and the number of colonies from each plate counted on days 14 and 21. Results were reported to the unblinded research project manager (KJJ) who shared results with shelters in real time.
Itraconazole 10 mg/ml oral solution (Itrafungol; Virbac) was prescribed to all study cats at enrollment at 5 mg/kg PO q24h for 7 days followed by 7 days off treatment. Weekly pulsing was repeated until cured. All shelters used 1:16 accelerated hydrogen peroxide disinfectant (Rescue; Virox Animal Health) for daily spot-cleaning and twice-weekly environmental sanitation.
The topical treatments investigated in this study were Lime Sulfur 97.8% (Vet Basics; Revival Animal Health), miconazole nitrate/chlorhexidine gluconate 2%/2% plus triz EDTA shampoo (Dechra US) and OgenaShield Pure Oxygen HD Shampoo (Ogena Solutions). For all treatment groups, the topical product was applied every 3–4 days. Full haircoat sampling using the Mackenzie toothbrush technique occurred on the day of treatment, before application, twice weekly. Shelter staff were not blind to treatment types or outcomes. Topical treatment protocols are provided in the supplementary material.
The investigators predefined treatment as ‘failed’ when a cat remained culture positive on the hair sample taken on day 49 (week 7). Treatment with the allocated product continued until failure was confirmed by the study manager, typically approximately 14 days later. Cats in alternative treatment groups were then prescribed rescue treatment with LS. ‘Cure’ for all treatment groups was defined as occurring on the date of sampling of the second consecutive culture that showed no growth.
A survey was conducted at study completion to investigate staff and cat perceptions of each topical treatment. Shelter staff and managers who had applied LS and at least one other treatment product during the study were asked to compare the ease, unpleasantness and cat acceptance of each protocol on a 5-point Likert scale. They were also asked which product they would recommend their shelter use before receiving study results.
Statistical analyses were conducted by blinded investigators in Jamovi version 2.6 (ClinicoPath and TOSTER Modules) and SPSS Statistics Premium v30 (IBM). Demographic data were compared between treatment groups using Welch’s t-test and ANOVA. Success of treatment and time to cure were tested using χ2 test and Welch’s t-test analyses. Non-inferiority was compared using Welch’s t-test with a hypothesis set to non-equivalence. A Cox proportional hazards model was used to assess confounding of related factors and their impacts on time to cure by product using SPSS. Significance was set at P <0.05.
This study was approved by the Institutional Animal Care and Use Committee at Cornell University (project designation #2022-0016) on 1 April 2022 and renewed in April 2023. Informed consent was provided in writing by the medical director at each shelter location.
Results
A total of 89 cats from the four study shelters were enrolled between November 2022 and September 2023. Nine cats were excluded as fomite carriers, one for previously receiving antifungal treatment, two were removed because of stress of treatment and returned to field, and one did not complete its treatment by the end of the study. Of the 76 successfully enrolled cats, 74 (97.4%) were positive on Wood’s lamp: both Wood’s lamp-negative kittens had Wood’s lamp-positive littermates and consecutively positive culture results. In total, 23 cats were treated with LS, 24 with MC and 29 with HP.
There were no significant differences in the distribution of cats across treatment groups with respect to season, shelter, intake type, housing, body condition score (BCS) or age at intake (Table 1).
Demographic and treatment group allocation using Welch’s t-test for categorical variables and ANOVA for age
Data are n (%) or mean ± SD
BCS = body condition score; HP = hydrogen peroxide; LS = lime sulfur; MC = miconazole chlorhexidine
As predefined by the researchers, treatment success was the ability of the product to provide a cure by 7 weeks. No cats in the LS group failed to achieve cure (0.0%, 95% CI 0–14). In total, 9/24 (37.5%, 95% CI 18–59) in the MC group and 6/29 (26.1%, 95% CI 10–48) in the HP group failed to achieve cure, both significantly different from those treated with LS (Table 2).
Rate of treatment failure in 7 weeks by product, compared pairwise with lime sulfur
Data are n (%). Values in bold are statistically significant
df = degrees of freedom; HP = hydrogen peroxide; LS = lime sulfur; MC = miconazole chlorhexidine
Cats that failed treatment were predominantly from shelters 2 (n = 7) and 4 (n = 7) and aged less than 10 weeks at enrollment.
Time to cure
Median total time to cure, defined as the difference between date of first treatment application and sampling date of the second consecutive negative culture plate, was 25 days for LS (range 7–45), 37 days (range 14–62) for MC (P = 0.034) and 36 days (range 11–65) for HP (P = 0.030). These times were significantly different on a Welch’s t-test investigating equivalence.
Non-inferiority analysis
To investigate non-inferiority, Welch’s t-test was again performed for LS vs each alternative treatment. The inferiority margin was set at 7 days (two sampling events), the null hypothesis at a mean difference of 7 or more days and the alternate hypothesis at a difference below 7 days. Both shampoos showed a mean time to cure over 7 days longer than the mean time to cure for LS: MC-LS at 9.66 days (95% CI 0.7–18.5; P = 0.017) and HP-LS at 8.54 days (95% CI 0.44–16.6; P = 0.02). However, the CIs around the mean differences both included 7 days (Figure 1). Thus, our test for clinical inferiority was suggestive but not conclusive.
Equivalence plots for lime sulfur (LS) vs miconazole chlorhexidine + TrizEDTA (MC) and hydrogen peroxide shampoo (HP). The central dot is the mean difference in time to cure; the horizontal black bar along the x-axis represents the 95% confidence interval. The vertical dashed line at 0 represents the mean time to cure for LS, while the dashed line at –7 represents the predetermined inferiority margin. Color changes represent increasing standard deviations away from the difference in means
Controlling confounders
To determine whether other demographic or treatment factors affected differences in time to cure between treatment groups, a Cox proportional hazards model was created for the 73 non-censored cats. When shelter site, intake type, season of intake, housing type and BCS were controlled, only age and treatment groups remained significant. As age at intake increased, cats had a higher chance of cure at any given point in time, while cats in the MC and HP groups had a 23% and 44% chance of cure, respectively, compared with LS-treated cats at any point in time (Table 3, Figure 2).
Cox proportional hazards model for cofounding variables
The P value given on the control group line indicates the overall significance of that factor in the model. Values in bold are statistically significant
Reference group used in the model
BCS = body condition score; CI = confidence interval; df = degrees of freedom; HR = hazard ratio
Kaplan–Meier curve (time to cure) for cats in each treatment group, adjusted for confounders via the Cox proportional hazards model (P = 0.012). The dashed line represents 49 days, the a priori cutoff for treatment ‘success’. HP = hydrogen peroxide; LS = lime sulfur; MC = miconazole chlorhexidine
Treatment product perception
Major barriers to the use of LS include its noxious smell, unpleasant color and perceived feline aversion to application. After the study concluded – but before results were shared – 24 caretakers from the four shelters rated each treatment’s difficulty, unpleasantness and feline aversiveness on a 5-point Likert scale.
Analysis of survey responses found that no treatment was appreciated to be more difficult, more unpleasant or more averse to cats than any other. Participants commented on the unpleasant odor of both HP and LS, itchiness and dry skin in cats treated with HP and LS, bleaching of cat haircoats for HP, as well as ‘hypothermia’ and handling challenges during proscribed 10-min (MC) and 5-min (HP) soaking periods in the shampoo protocols (see the supplementary material). Caretakers also described the alternative product protocols as taking noticeably longer to complete. When asked which product they would like their shelter to use in a standard protocol, 16/23 (70%) respondents preferred LS, 6/23 (26%) preferred HP and 1/23 (4%) preferred MC.
Cat outcomes
All cats were successfully adopted after participating in the study, including one 18-week-old cat diagnosed as progressively positive for feline leukemia virus (time to cure 15 days.) The mean length of time between cure and adoption was 23.8 days (range 3–58); this reflected the lag of 14–20 days between cure and culture growth/reporting. Shelters reported no recrudescence after cure, nor any serious adverse effects for any product.
Discussion
In shelter cats receiving pulsed weekly oral itraconazole, both MC and HP shampoos had significantly more treatment failures before 7 weeks and significantly longer mean times to cure than those treated with LS.
Non-inferiority analysis showed that the alternative products were not equivalent but could not be confirmed as inferior to LS, a result particular to non-inferiority studies. Although the mean time to cure for both products was greater than the predetermined ‘inferiority margin’ (ie, 7 days), the CIs around that difference included our margin. This suggests a better outcome for the alternative treatments than demonstrating clear inferiority (ie, CIs entirely above the inferiority margin), but it offers less certainty than demonstrating equivalence (ie, mean differences below 7 days or no statistically significant difference). 24 A larger sample size would likely have given smaller CIs and better illustrated the true mean difference.
Means and ranges for overall times to cure for LS and MC shampoo seen in this study were similar to those previously reported, even though the products we used had some formulation differences. One study found a mean cure time of 19.8 days (range 10–49) for LS in highly culture-positive cats, 10 while another found a median cure time of 30 days (range 10–69) for LS and 38 days (range 14–93) for MC (P = 0.031). 9 The similarity in times to cure was expected but also reassuring, given that our samples were grown using the proprietary IDEXX dermatophyte culture system rather than the standard point-of-care dermatophyte test media.6,25,26
Although HP-based shampoo is widely used to treat dermatophytosis in animals, this is the first study in the veterinary literature to investigate its time to cure. To facilitate direct comparison, our treatment protocol was slightly different from that recommended by Ogena Solutions, the shampoo manufacturer. With their approval, we used a slightly stronger dilution twice weekly rather than a less concentrated solution applied every other day. We do not know how the difference in protocols affected our outcomes, but more baths/week would present logistical and staffing challenges for shelters and be more aversive for cats.
Although cats receiving itraconazole and LS are likely to be mycologically cured by the time their first negative culture is finalized at 14 days, 27 we were concerned that this accelerated schedule might not hold for cats being treated with alternative products. To decrease the length of shelter stay for study cats, we allowed shelters to release cats in the LS group from isolation after their first negative result. Bi-weekly sampling ensured that at least three subsequent cultures would already be captured during the lag of approximately 14 days between sampling and result reporting. Using two consecutive cultures with no growth at 14–21 days is consistent with previous recommendations for determining mycological cure.12,26
Our Cox proportional hazards analysis showed that increasing age had a significant positive effect on time to cure. Indeed, the seven ‘adult’ cats (aged >5 months) enrolled in this study were all cured by 38 days, regardless of treatment. Age may reflect immune competence, as a reduced ability to mount an immune response to fungal infection could lead to longer cure times. In addition, pulse treatment with itraconazole relies on the drug’s accumulation and persistence in the stratum corneum, which is composed of corneocytes, ceramides, free fatty acids and cholesterol. Both age and body condition can affect the composition of these elements and may impact drug accumulation. 11 Although body condition approached significance in the individual factor analysis, it was not significant in the final model. We also accounted for other factors that did not seem to affect chances of cure over time in our model, including co-housing littermates, intake type and season of intake. Although many of these factors are important epidemiologically,4,28 only age seemed to affect the length of treatment.
In our hazard model, differences in median times to cure between MC vs LS and HP vs LS treatment groups maintained significance, with MC performing more poorly than HP. These differences remained despite a range of unmeasured factors that could have confounded our results, including differences in climate, the convenience sample of kittens and cats entering each shelter, differences in product mixing or application, individual caretaker effects and individual animal stress. Three cats in the HP and MC groups were ultimately cured with LS, and their treatment courses may have been even longer had the therapy not been changed.
Although LS has been widely acknowledged as hard on facilities and unpleasant for cats and personnel, the surveyed caregivers in this study did not find LS more difficult to apply, more unpleasant or more aversive for cats than the tested alternative treatments. In addition to longer times to cure, the alternative products required staff to keep the animal soapy and wet for 5 mins (HP) or 10 mins (MC) while the product took effect, and then to perform a rinsing step. Cats becoming cold during treatment increased handling difficulty over time and required additional staff time. Where available, LS remains the most economical option in the veterinary marketplace.
Regardless of the product used, all study cats had positive adoption outcomes and, after rescue treatment for three cats, cured significantly faster than reported times for itraconazole alone or self-cure.11 –13
Conclusions
Based on the high number of treatment failures and significantly longer times to cure for MC and HP shampoo-treated cats, our overall recommendation is to continue to use LS as the primary topical treatment for dermatophytosis in shelter-housed cats. Shelters may have well-considered reasons for selecting an alternative topical treatment, such as access to products or staff and foster parent compliance. If an alternative product is needed, HP shampoo appears to be a better choice to minimize both time to cure and required staff time.
Supplemental Material
Supplementary Material
Topical treatment protocols.
Footnotes
Acknowledgements
We would like to acknowledge Ms Suzette Moschetti of Cornell University for her administrative and project management assistance. We would also like to acknowledge the contributions of the site managers, veterinarians and cat caretakers at each of the four shelters, without whose efforts this study would not have been possible. We would finally like to honor Dr Kimberly Coyner DVM, DACVD, who helped with the study design and conceptualization before her untimely passing.
Supplementary material
The following file is available as supplementary material:
Topical treatment protocols.
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This work was financially supported by the Cornell Feline Health Center Research Grants Program, a grant made available to the College of Veterinary Medicine, Cornell University, Ithaca, NY, USA. All topical products and their shipping costs were generously donated by their manufacturers, Revival Animal Health, Ogena Solutions and Dechra US. Shelter discounts were generously extended for the purchase of Itrafungol by Virbac and by IDEXX for sample testing. Funding for the publication of this study was provided by the American Society for the Prevention of Cruelty to Animals (ASPCA) Open-Access Publishing Fund.
Ethical approval
The work described in this manuscript involved the use of non-experimental (owned or unowned) animals and procedures that differed from established internationally recognized high standards (‘best practice’) of veterinary clinical care for the individual patient. The study therefore had prior ethical approval from an established (or ad hoc) committee as stated in the manuscript.
Informed consent
Informed consent (verbal or written) was obtained from the owner or legal custodian of all animal(s) described in this work (experimental or non-experimental animals, including cadavers, tissues and samples) for all procedure(s) undertaken (prospective or retrospective studies). No animals or people are identifiable within this publication, and therefore additional informed consent for publication was not required.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
