A "Pedi" Cures All: Toenail Trimming and the Treatment of Ulcerative Dermatitis in Mice

Abstract

Ulcerative Dermatitis (UD) is the most common cause of unplanned euthanasia in mice used in research, with prevalence rates reported between 4 and 21%. UD is characterized by a deep, ulcerative lesion that appears most commonly over the dorsal neck and is attendant with an intense pruritus. The underlying cause of UD is currently unknown, and as a consequence, there are no directed therapies that resolve lesions reliably. However, there is a growing body of evidence that suggests a behavioral component to the onset, maintenance, and progression of UD lesions. Scratching behavior in response to the intense pruritus associated with UD lesions may be an effective target for interventional therapies. We hypothesized that interfering with scratching behavior by trimming the toenails of mice with UD, would resolve UD lesions. To test this hypothesis, we first evaluated the efficacy of toenail trims with a single application of Vetericyn at the time of treatment versus our previous standard of care, topical Tresaderm applied daily. We found that toenail trims were significantly more effective at resolving lesions (n = 39 toenail trims, n = 100 Tresaderm, p<0.0001) with 93.3% of animals healing by 14 days (median time to lesion resolution). Furthermore, dorsal neck lesions did not recur by 42 days after a single toenail trim (n = 54); however, flank lesions did not resolve and the outcome of the two lesion distributions following treatment were significantly different (p<0.0001). Finally, we implemented toenail trims at an institutional level and found similar efficacies (approximately 90%) for toenail trims regardless of one-time topical supplement used (triple antibiotic ointment, Tresaderm, and Vetericyn, n = 55, 58, 18, p = 0.63). This is the first report of a highly effective treatment for one of the most serious welfare issues in laboratory mice.

Fig 1. Diagram of restraint device used for toenail trim.

A 50mL conical tube was modified by the following changes. Cutouts were made along the rim of the tube to allow seating of the mouse knee against the edge and were lightly melted to remove sharp edges. The tip of the tube was removed and lightly melted to ensure air passage as well as facilitate decontamination between uses. Approximate dimensions are shown in millimeters.

Fig 2. Proportion of mice showing complete lesion resolution by treatment type.

Logistic regression analysis showing the proportion of all mice entering treatment that were cured rather than euthanized in the records review. Estimated means and standard errors are shown from the logistic regression (which controls for date of first treatment, and duration of treatment), P < 0.0001.

Fig 3. Survivorship and cumulative mortality as a function of time by treatment group.

A) Kaplan-Meier survivorship plot. Survivorship is shown as the proportion of non-censored (at risk) animals surviving at each time point (rather than the proportion of all animals surviving). Cured animals are censored (i.e. removed from the total count at each time point), thus the three steps seen in the curve for toenail trim treated animals represent the individual euthanasia events for the three (out of 39) mice which were not cured. The rapid drop in survivorship at day 43 for toenail trim animals reflects the fact that only two animals remained at this timepoint (and the final animal was cured on day 49). B) Cumulative mortality, showing the raw mortality events, without the artifacts injected by Kaplan-Meier survivorship calculations.

Fig 4. Resolution of lesion by type and as a function of time.

Repeated-measures logistic regression showing the differential resolution of Flank versus Head & Neck lesions following Toenail Trim treatment. Each mouse was scored for each type of lesion every two weeks following first diagnosis. Although head-and-neck lesions were far more common at first diagnosis, they resolved rapidly, and did not recur (despite toenail regrowth). Flank lesions did not resolve (P = 0.0004). Observed Means and Standard Errors are plotted over the logistic regression curves.

Fig 5. Lesion resolution as a function of topical supplement selected.

Logistic regression analysis showing the proportion of all mice entering treatment that were cured rather than euthanized in the technician trial. All mice were toenail trimmed. However, depending on the veterinarian, a different topical was applied at the same time. Topical treatment had no effect on final cure rate (P = 0.6293). Estimated means and standard errors are shown from the logistic regression (which controls for date of first treatment, and duration of treatment).

Fig 6. Lesion resolution as a function of time on treatment.

A) Kaplan-Meier survivorship plot. Cure Rate is shown as the proportion of non-censored (i.e. non-euthanized) animals surviving at each time point. Euthanized animals are censored (i.e. removed from the total count at each time point). B) Cumulative Cure Rate, showing the raw cure events, without the artifacts injected by Kaplan-Meier survivorship calculations.