Hair corticosterone concentration as a potential stress marker in pet rabbits

Abstract

Background

As a prey species, rabbits tend to hide their illnesses and injuries. Consequently, pet rabbit owners often do not notice that their pet may be suffering.

Methods

Data on the housing and health of 34 dwarf rabbits were collected from a total of 22 rabbit owners. A score from 0 (not stressed) to 3 (highly stressed) was assigned to each rabbit based on the owners’ reports. Afterwards, the hair corticosterone concentration (HCC) of the rabbits was analysed as an indicator of long-term stress.

Results

The mean HCC of all rabbits was 1.54 ± 0.64 pg/mg. Animals with a stress score of 3 had a higher HCC (2.17 pg/mg) than those with a score of 0 (1.09 pg/mg) or 1 (1.44 pg/mg) (p < 0.05). HCC was not significantly affected by sex or age (p > 0.05).

Limitation

Only a small number of rabbits were included in the study. Therefore, the findings need to be confirmed in a larger cohort of animals.

Conclusion

HCC seems to be a promising stress marker in pet rabbits.

INTRODUCTION

Rabbits are popular pets; however, there are hardly any controls on how pet rabbit owners keep their animals, and it is their choice to go to a veterinarian. In many cases, pet rabbit owners do not even notice that their pet is secretly suffering as a consequence of inadequate housing or disease because rabbits, as a prey species, tend to hide their illnesses and injuries.¹ The owner's lack of awareness concerning signs of animal suffering can also result in poor rabbit welfare. Consequently, the animals may be stressed for long periods of time. Determining the stress level in individual animals could therefore be a suitable method to detect deficiencies in the rabbits’ environment or diseases at an early stage.

Stress is associated with increased hypothalamic–pituitary–adrenal axis activity, and the concentration of glucocorticoids (GCs), as the end product of this pathway, is therefore considered an indicator of stress.² GC measurement in hair is considered a promising method to indicate an average GC concentration over a long period.³ The sampling procedure is easy and painless² and a relatively small quantity of hair is required for analysis (≤500 mg).³

So far, there are only a few studies showing a relationship between hair cortisol or corticosterone concentrations and stress in lagomorphs, for example, in pikas,⁴ brown hares,⁵ New Zealand white rabbits^{6, 7} and fattening hybrid rabbits (Hyplus, Hypharm, France).⁸ Therefore, the aim of the present pilot study was to test if the measurement of hair corticosterone in pet rabbits may be an appropriate method to assess their stress level and possibly even to indicate diseases.

MATERIALS AND METHODS

Animals, stress scoring and hair sampling

We asked several German pet rabbit owners, who were mainly students or employees of the University of Veterinary Medicine Hannover, to fill in a questionnaire about their rabbits with the following items: date of birth or approximate age, breed, hair colour and length, sex, neuter status, bodyweight, vaccination status and health-related aspects in the last 6 months. These health-related aspects included chronic pre-existing illnesses, acute pre-existing illnesses, veterinary treatments and specific stressful situations. Additionally, a detailed description of the housing system and the number of other rabbits in the same enclosure were requested. We also asked the owners if their rabbit was brushed frequently in the rear neck/front back area, since excessive brushing can lead to locally increased GC production.⁹ All owners gave written consent for their participation in the study.

We collected questionnaires from 28 owners, relating to a total of 69 rabbits. From these, 34 dwarf rabbits belonging to 22 owners were selected for this pilot study. The main selection criteria were light coat colour (white, light grey, grey, light brown), because earlier studies showed an effect of dark colouration on hair cortisol concentration,² breed (dwarf rabbit) and preliminary reports regarding health status and stressful situations.

Based on the questionnaires, we divided the rabbits into four stress groups: 0 = healthy, not exposed to prolonged stress; 1 = mild pre-existing illness and/or minor prolonged exposure to stress; 2 = moderately severe pre-existing illness and/or exposure to prolonged stress; and 3 = severe pre-existing illness and possibly additionally exposed to prolonged stress. The stress score was assigned to each rabbit by two independent observers using a defined basic scoring system (Supporting Information Appendix A1). This scoring system was newly developed for this study because no suitable scoring system from previous studies was available. The two observers did not know the HCC of the rabbits when assigning the stress score. If there was a disagreement between the observers concerning a score for an individual rabbit, additional information from the owner clarified the situation in each case. The different stress factors to which the individual animals were exposed are listed in Supporting Information Appendix A2.

Finally, one hair sample per rabbit was collected and analysed. The rabbit owners received instructions on how to take a hair sample. In the rear neck/front back area, a 2 cm × 2 cm symmetrical area was first shaved as close as possible to the skin using previously cleaned electric clippers or scissors. Care was taken to ensure that the area sampled was as clean as possible and free of faeces. The hair samples were then brought to us in labelled plastic bags (name of owner, name of rabbit). The samples were stored under light-protected and dry conditions at room temperature until they were sent to Dresden LabService, Dresden, Germany, for analysis. There, the hair corticosterone concentration (HCC) was analysed using liquid chromatography with tandem mass spectrometry.

Statistical analysis

Statistical analyses were performed using the statistical software R.10 The level of significance was set at α = 0.05. A linear mixed regression model¹¹ was constructed with HCC as the dependent variable and stress score, sex (male neutered, female intact or female neutered), hair colour (white or non-white) and age category (0–3, 3–5 or 5–10 years) as independent variables using the R package ‘lmerTest’.¹² An F-test was then conducted to analyse which of these variables had a significant impact on HCC. In case of significance, pairwise comparisons (p-value adjustment with Bonferroni–Holm) were conducted using the R package ‘emmeans’.¹³

RESULTS

The selected rabbits (n = 34) were between 1.25 and 10 years old. Among the females (N = 21), about half were neutered (n = 10), and the males were all neutered (n = 13). The mean bodyweight of all rabbits was 2.03 ± 0.60 kg. Nearly all rabbits were vaccinated against myxomatosis (n = 31) and rabbit haemorrhagic disease virus 1/2 (n = 33). Housing ranged from indoor cage housing (at least 3 m²; n = 5 rabbits) to free indoor housing (up to 60 m²; n = 7 rabbits), secured outdoor housing (at least 4 m²; n = 19 rabbits) and free outdoor housing (up to 300 m²; n = 3 rabbits). The group size ranged from one to six rabbits. The animals were brushed rarely or not at all.

The mean HCC of all rabbits was 1.54 ± 0.64 pg/mg (minimum: 0.62 pg/mg; maximum: 3.69 pg/mg). As shown in Figure 1, rabbits with a stress score of 0 had the lowest mean HCC (1.09 ± 0.27 pg/mg), followed by rabbits with a score of 1 (1.44 ± 0.59 pg/mg) and rabbits with a score of 2 (1.60 ± 0.37 pg/mg). The rabbits with a stress score of 3 had the highest mean HCC (2.17 ± 0.84 pg/mg). Compared to rabbits with a score of 3, the HCC decreased significantly for those animals with a score of 1 (p = 0.033) and a score of 0 (p = 0.025). There were no significant differences between the other score contrasts (p > 0.05). White rabbits had a lower mean HCC (1.30 ± 0.53 pg/mg) than non-white rabbits (1.73 ± 0.69 pg/mg, p = 0.022). HCC was not affected by sex or age category (p > 0.05) (Table 1). However, it was descriptively shown that the mean HCC tended to increase with age category (0–3 years: 1.17  ± 0.56 pg/mg, 3–5 years: 1.50 ± 0.47 pg/mg and 5–10 years: 1.72 ± 0.75 pg/mg) (Figure 1).

FIGURE 1

Hair corticosterone concentration (HCC) of rabbits with different stress score levels (0, 1, 2 and 3) and age categories (0–3 years, 3–5 years and 5–10 years). The HCC values (pg/mg) are shown as box plots with minimum, lower quantile, median, upper quantile, maximum and outliers (n = 34)

TABLE 1. Structured table for the linear mixed regression model including the table of coefficients as well as the results of the F-test and the pairwise comparisons with hair corticosterone concentration of rabbits as dependent variable (n = 34).

	Estimate	Standard error	df1	df2	F-value	t-Value	p-Value
Intercept	1.251	0.255
Score 1 (0)	0.253	0.289
Score 2 (0)	0.685	0.316
Score 3 (0)	1.110	0.323
White (non-white)	–0.462	0.170
Female intact (male neutered)	0.299	0.216
Female neutered (male neutered)	–0.290	0.186
3–5 years (0–3 years)	–0.169	0.274
5–10 years (0–3 years)	0.180	0.316
Model term
Score			3	23.68	5.642		0.005^**
Hair colour			1	16.73	6.390		0.022^*
Sex			2	13.13	2.820		0.096
Age category			2	18.46	0.724		0.498
Score contrast
0–1	–0.253	0.309	24.665			–0.818	0.526
0–2	–0.685	0.330	25.000			–2.077	0.193
0–3	–1.110	0.350	23.681			–3.175	0.025^*
1–2	–0.432	0.259	19.077			–1.667	0.336
1–3	–0.857	0.288	23.488			–2.980	0.033^*
2–3	–0.425	0.371	23.484			–1.147	0.526

• * p < 0.05.
• ** p < 0.01.

DISCUSSION

This pilot study showed that rabbits had different HCCs depending on the stress score that we assigned to them, taking into account their health status and experience of stressful situations. The higher the score, the higher the HCC. Diseases and injuries are often associated with discomfort and pain, leading to an increased stress level. In our study, rabbits with a stress score of 3 had suffered from conditions including intestinal obstruction, coccidiosis, gastric emptying disorder, otitis, eye inflammation and tooth extraction within the previous 6 months. Social stress factors identified for rabbits with a stress score of 3 included the loss of a partner rabbit and rank order fights (Supporting Information Appendix A2). Most of these rabbits had a combination of different stress factors.

In previous studies in cattle, it was found that cows with clinical diseases, especially with more than one disease, had greater cortisol concentrations in hair than healthy cows and that these long-term changes in basal cortisol secretion could affect homoeostasis, leading to an increased susceptibility to further diseases.^{14, 15} In addition, it was mentioned that ectoparasitic infestation is stressful for cows, as are changes in housing and social structure, all of which are reflected in hair cortisol concentration.¹⁶ To our knowledge, a relationship between diseases or social stress and HCC in pet rabbits has not been reported before.

In this pilot study, we also showed that white rabbit hair contained less HCC than non-white rabbit hair. However, a comparison of the influence of hair colour on hair GC in different species showed inconsistent results.² For example, in dogs, black hair contained less cortisol than non-black hair,¹⁷ while the opposite was true in pigs and cattle.¹⁸ The underlying mechanisms of GC incorporation into different-coloured hair have not yet been fully investigated and require further research.²

The older pet rabbits get, the more often age-related diseases occur. Therefore, it was not surprising that older rabbits tended to have a higher HCC. Nevertheless, in agreement with our study, it was also shown that faecal GC baseline concentrations in pygmy rabbits were not significantly affected by age.¹⁹ Furthermore, HCC was not influenced by sex. However, neutered female rabbits had slightly lower HCCs than intact ones (1.31 ± 0.57 pg/mg vs. 1.68 ± 0.55 pg/mg). The same was reported for female cats.^{2, 20}

CONCLUSION

In this pilot study, it was shown that pet rabbits with a higher stress score also had a higher HCC, which was mostly associated with illness. Consequently, HCC analysis in pet rabbits could have the potential to provide a non-invasive way to detect stressful situations and diseases, possibly at an early stage. However, further research with a larger sample size is needed to confirm the findings.

AUTHOR CONTRIBUTIONS

Study design: Michaela Fels. Study execution: Dana Hube and Michaela Fels. Data analysis and interpretation: Carmen van Meegen, Swetlana Herbrandt, Dana Hube and Michaela Fels. Preparation of manuscript: Dana Hube and Michaela Fels. Final approval of the manuscript: Nicole Kemper and Michaela Fels.

ACKNOWLEDGEMENTS

We sincerely thank all rabbit owners who provided us with hair samples from their rabbits for their participation in the study.

Open access funding enabled and organized by Projekt DEAL.

CONFLICT OF INTEREST STATEMENT

The authors declare they have no conflicts of interest.

FUNDING INFORMATION

The authors received no specific funding for this work.

ETHICS STATEMENT

The present study did not involve any invasive procedure or treatment of the animals. The authors confirm that the study was conducted in accordance with current German law and no ethical approval was required.

Open Research

DATA AVAILABILITY STATEMENT

The data supporting the findings of this study are available from the corresponding author upon reasonable request.