Although the number of rabbits (Oryctolagus cuniculus) produced in Europe is important, they are under-represented in welfare research. Studies on the effects of housing conditions have focused on performance and mortality. We conducted a meta-analysis to further understand the impact of the housing environment on growing rabbits. Whilst providing a robust quantifiable review, meta-analyses are restricted to existing literature. We included information on behaviour, mortality and performance. Twenty-seven peer-reviewed and conference publications, with a total of 99 experimental treatments, were used. We collected information about rabbit age, bodyweight, sex and breed; allometric space allowance, pen height, group size, environmental temperature, floor type, substrate and enrichment use, lighting, diet and medicated feed. Predictive equations for each response variable were calculated using multiple regression models. Higher space allowance was found to increase locomotor and social activity, and to reduce resting and comfort behaviour. Restricted pen height increased ingestive behaviour; comfort behaviours decreased as space allowance increased, although these might have been confounded with self-directed behaviours in literature. Mortality remained stable at varying space allowances, but increased with larger group sizes and enrichment objects. Growth rate, feed intake and feed conversion were reduced with higher space allowances and larger group sizes, and by provision of substrate and enrichment objects. Findings suggest that higher space allowance and unrestricted pen height are beneficial for rabbit behaviour, but might have undesired consequences if considered independent from other aspects. The challenge of promoting welfare in commercial conditions was highlighted, as a number of parameters which improved behavioural expression reduced performance. In certain cases welfare inputs complemented performance, including providing non-medicated feed and higher space allowance in cooler climates. Although our results should be interpreted with caution given the limitations of the included variables, they are expected to contribute to the improvement of current and new rabbit housing systems to optimise welfare.

Effect of graded levels of high-glucosinolate mustard (Brassica juncea) meal as substitute of soya-bean meal (SBM) in broiler rabbit diets was studied. Forty weaning rabbits of Soviet Chinchilla and White Giant breed were randomly allocated to one of four experimental diets containing mustard meal (MM) 0, 80, 160 and 245 g/kg. The experiment lasted for 8 weeks. MM had 54.8 mg total glucosinolates (TGLSs) per g dry matter (DM). Diets had TGLS 3.8, 8.4 and 11.98 mg/g DM in 80, 160 and 245 g MM diets, respectively. MM-incorporated diets had higher digestible and linearly (P < 0.01) higher metabolisable energy (ME) content. However, the effect on total tract apparent digestibility of DM, and crude protein was quadratic. Average daily gain (ADG) reduced (P < 0.05) linearly with increasing MM levels in diet, still 80 and 160 g MM diets had similar ADG compared to that of SBM diet. Caecum weight reduced linearly (P < 0.05) with increasing MM levels in diet. The pH of caecal content ranged between 5.85 and 6.19, total N between 1.19 and 1.48 (g per 100 g) and total volatile fatty acids between 4.7 and 5.8 mmol per 100 g, and they were not statistically different. NH3-N ranged between 31.2 and 39.0 mg per 100 ml, and reduced linearly (P < 0.05) while trichloroacetic acid-precipitable nitrogen increased linearly (P < 0.01, ranged between 114 and 247 mg per 100 ml) with increasing MM levels in diet. Blood haemoglobin, packed cell volume and lymphocytes were higher (quadratic effects, P < 0.05) on 245 MM diet, whereas white blood cell count reduced linearly (P < 0.01). Serum aspartate aminotransferase increased linearly (P < 0.01) while alanine aminotransferase and alkaline phosphatase activity, protein, erythrocytes sedimentation rate and red blood cell counts were not affected by MM. Serum Cu, Na and K content increased linearly (P < 0.05) with increasing MM levels. Liver Cu concentration showed quadratic (P < 0.05) increase. Rabbits tolerated 8.4 mg TGLS per g diet (160 g MM per kg) during active growth without any apparent effect on health and growth. It is concluded that MM can replace up to 66% SBM protein in rabbit feeding, whereas complete replacement of SBM with MM reduced feed intake and ADG by 23% and 13%, respectively. Further studies are required to confirm these inclusion levels and glucosinolate tolerance of rabbits.