Seventy two Merina breed male lambs (28.03 ± ± 0.05 kg, 90 days old) were used in this trial. Lambs were weaned at 45 days after birth and then fed a commercial concentrate (17% crude protein), containing cereals, soya, calcium carbonate, sodium chloride, mineral-vitamin mix and cereal straw ad libitum. All handling practices were carried out according to statements of the Directive 2010/63/EU (EC, 2010) with regard to the protection of animals used in research and for scientific purposes.

The study was carried out in spring, in two journeys that were performed in April and May, respectively (both in sunny conditions and without rain). Values of the temperature and relative humidity were registered using a WatchDog 15–Temperature/Humidity Monitor. Animals (n=36 per journey) were loaded (starting at 8 am, 7ºC average temperature during loading) randomly in the middle floor and in both sides of the lorry, according to the three spaces allowances: 0.16 m²/animal (SAL; n=24; 12/journey); 0.20 m²/animal (SAM; n=24; 12/journey) and 0.30 m²/animal (SAH; n=24; 12/journey). The same vehicle was used in both journeys.

Lambs were transported by paved roads (334 km, ~ 5½ h) from the farm (Cabeza del Buey, Badajoz, Spain; 38° 43′ N) to the abattoir (Tarancón, Cuenca, Spain; 40° 0′ N) along with other lambs not included in this study. The mean temperature and relative humidity recorded during transport were similar for both journeys, 11ºC and 56% respectively (min. temp. 10.2ºC and 13ºC at the beginning and max. temp. 14ºC and 16ºC at the end of the journey for the first and second journey, respectively). The lorry had two axles and three floors, natural ventilation along the full length of the truck body sides and on each floor, a hydraulic elevator for loading and unloading. When animals arrived to the slaughterhouse were unloaded and rested in lairage for 18 h previous slaughter.

Animals in the slaughterhouse and in each group of density during transport were split randomly in two groups: one group of animals remained in fasting (FAST; n=36) during lairage, while the other group (FEED; n=36) were fed ad libitum the same commercial concentrate as on the farm). During lairage all animals remained in a waiting area, covered with a roof, at a density of 0.36 m² per animal and receiving water ad libitum. The average temperature at lairage was 14.7ºC with a relative humidity of 62% (first journey: max. temp.=18.3ºC, min. temp.=7.3ºC; second journey: max. temp=20.9ºC, min. temp.=12.7ºC). After lairage, lambs were slaughtered using standard commercial procedures.

Blood samples were taken by trained personnel by external jugular venipuncture according to Linares et al. (2008) at three different times. The first sample used to determine the basal blood parameter concentrations was taken on the farm in resting conditions; a second sample was taken immediately after unloading the animals to evaluate the effect of space allowance during transport. The final sample was collected after 18 h lairage, to determine the effect of the lairage treatment (feeding or fasting) before slaughter. Approximately 20 seconds were necessary for each extraction, and in all samples a 0.8×25 mm bevelled needle (Terumo Neolus, Belgium), 5 mL syringe (BD Discardit^TM, Spain) were used; 5 mL of blood was extracted to fill 1 mL tubes containing EDTA (ethylene diaminotetraacetic acid) for haematology and catecholamine determination and 4 mL tubes without additive for cortisol and biochemical parameter concentrations measurements. Blood samples were maintained at 2ºC in a portable refrigerator until they arrived at the clinical analysis laboratory. The tubes were centrifuged at 4000 rpm for 5 min and frozen and kept at −18ºC until processed.

The following physiological indicators were analysed:

Data were analysed with the Statistical Package SPSS 19.0 (IBM, 2010). First a Shapiro-Wilk test was carried out to check the normality and homogeneity of variance of all parameters. Then the repeated measured ANOVA was carried out to determine the effects of space allowance treatment during transport on haematological, hormonal and biochemical blood parameters and the differences with the basal values on farm. A Tukey’s test at a significance level of p<0.05 was carried out to check the differences between pairs of groups.

A General Linear Model (GLM) was used to examine the effects of space allowance treatment during transport (low, medium, high) and treatment during lairage [Fasting (TL-FAST) or Non Fasting (TL-FEED)] and their interactions on haematological, hormonal and biochemical blood parameters after lairage. In addition a Tukey’s test at a significance level of p<0.05 was carried out to check the differences between pairs of groups [space allowance treatment (SA) – treatment during lairage (TL)].

A stepwise discriminant function analysis was carried out to select a linear combination of the independent parameters that best allowed differentiating among the SA groups during transport. A second discriminant analysis was carried out to differentiate among treatment groups during lairage. This last analysis was defined by two canonical discriminant functions which were illustrated by means of a dispersion diagram.

Physiological parameters values for on farm and after transport per SA are showed in Table 1. Journey did not affect any blood parameter values on farm or after transport in each SA. In the present work haematological parameters after transport did not differ significantly from values detected at the farm. SA during transport did not affect these parameters.

After transport, the SAM group showed the highest cortisol value (142.10 nmol/L) while the lowest was found at the SAH (58.48 nmol/L), while SAL group showed intermediate values (101.53 nmol/L) for this parameter. However, there were not significant differences among SA treatments and values were similar to the values found on farm.

In general, adrenaline concentration was lower on farm than after transport but only significant differences (p<0.05) were found between SAL groups (469.51 nmol/L on farm vs 662.59 nmol/L after transport). SA treatment during transport did not affect adrenaline concentration. Noradrenaline values on farm were similar to the values after transport p> 0.05).

ANOVA test showed significant differences on glucose (p<0.001), total proteins (p<0.001) and LDH activity (p<0.01). In general the lowest values for these serum indicators were found on farm and the highest in SAM group after transport. Tukey test showed significant differences (p<0.05) among SA groups in glucose concentration and in the LDH activity: SAH one had the lowest values (4.99 mmol/L; 296.52 U/L respectively) with significant differences with SAM (5.60 mmol/L, 398.96 U/L respectively). The SAL group (0.16 m²/lamb) showed intermediate values for these biochemical parameters (5.34 mmol/L, 350.91 U/L respectively). However, total proteins, creatinine, CK and lactate values were not affected by the space allowance during transport.

The discriminant analysis showed that only the LDH activity marked the difference among SA groups. The centroid values were -0.124, 0.430 and -0.378 for SAL, SAM and SAH respectively; eigen value: 0.125; canonical correlation: 0.334.

Table 2 shows the values of blood parameters in each group of animals after lairage. Respect to haematological parameters, a significant effect (p<0.05) of treatment during lairage was found on RBC, with significant differences (p<0.05) between SAM groups. Cortisol after 18 h ranged between >65 nmol/L (TL-FAST-SAH) and <111 nmol/L (TL-FEED-SAM). The highest value of adrenaline (763.45 nmol/L) was found in TL-FAST-SAM group. There was a strong effect (p<0.001) of lairage treatment on noradrenaline, with highest values in groups TL-FAST-SAL (2816.46 nmol/L) and TL-FAST-SAM (3171.40 nmol/L).

A significant effect of treatment during lairage was found on glucose (p<0.001), total proteins (p<0.01), creatinine (p<0.01) and LDH activity (p<0.001). In general, fasting during lairage decreased glucose values, while increased total proteins (in SAM group), creatinine and LDH activity (in SAH groups).

The discriminant analysis (Table 3) obtained two discriminant functions. The first explained the maximum of the existing differences among fasting groups (1: FAST-SAL, 3: FAST-SAM and 5: FAST-SAH) and no fasting groups (2: FEED-SAL, 4: FEED-SAM and 6: FEED-SAH), while the second function (only 10.3% of variance) discriminated between groups 2 and 3 from all other groups (1, 5, 4, 6). This analysis found that the parameters that truly marked the difference among groups were noradrenaline and glucose (the standardized coefficient of the first canonical function were 0.670 for glucose and 0.680 for noradrenaline while on the second one these values were 0.748 and 0.739, respectively). A dispersion diagram (Fig. 1) explains, in graphical form, the results obtained in the statistical procedure: Groups 1, 3 and 5 (FASTING) are located in the area that corresponds to the positive value of the first function (with noradrenaline and glucose as main parameters). In regard to the second function (with the same main parameters), G2 (FEED-SAL) and G3 (FAST-SAM) are located in the area that corresponds to the positive values of this function while the rest of the groups (G1, G4, G5 and G6) are located in the negative part.

Similar values were found on haematological parameters among the different SA groups. Miranda-de la Lama et al. (2011) found a similar number of RBC, leucocytes and hematocrit in Aragonesa lambs at similar age (100 days) and weight (25 kg) transported for 3 h. Our results could indicate that SA during transport is not a factor that affects these blood parameters. According to Mellor et al. (2002) cortisol and adrenaline values increase with emotional anxiety, while noradrenaline seems to be related to physical stress. In addition, the cortisol response to handling and transport depends upon the species and breed studied (Hall et al., 1998). In this study the space allowance had no significant effect on cortisol values, results that are in agreement with Cockram et al. (1996) and De la Fuente et al. (2012). Our results showed a significant increase in adrenaline concentration after transport in SAL group but not corresponding increase in noradrenaline, in agreement with Åkerstedt et al. (1983). These findings imply that a space allowance between 0.16 and 0.30 m²/lamb did not cause a significant hormonal variation in lambs transported under the conditions provided in the present study.

Our findings showed that plasma glucose activity was influenced by space allowance. According to the results of Bórnez et al. (2009) in Manchega lambs and Tadich et al. (2009) in Corriedale breed, the highest values found in glucose at medium-SA could be associated with a high stress level. However, De la Fuente et al. (2012) did not find any effect of space allowed in this indicator in Assaf breed lambs. A different reaction to SA could be due to both lamb ages and sheep breeds.

SA had not effect on total protein values, results that are consistent with the findings by Cockram et al. (1996). Increased creatinine levels have been associated to dehydration (Montané, 2002). Pollard et al. (2002) in red deer subdued to different treatments previous to slaughter, indicated that the creatinine concentration was not a useful parameter to measure animal welfare, which is in agreement with the results found in the present study.

On the other hand, no differences for lactate values were detected among SA treatments, although other authors such as Pollard et al. (2002) or Miranda-de la Lama et al. (2011) found an increase for this welfare indicator related to stress. According to Grandin (1997) genetic factors, such as temperament, interact in complex ways with an animal’s previous handling experiences and learning determining future reactions during a particular handling procedure.

High levels of serum LDH or/and CK have been correlated with muscle tissue damage as well as with vigorous exercise (Knowless, 1998; Miranda-de la Lama et al., 2010a). Broom & Fraser (2007) indicated that CK activity could be considered as an indicator that permits determining appropriate minimum acceptable space allowances for transported animals. Although, in our study, no significant effects of space allowance during transport were detected on CK activity, an increase in LDH activity after transport, with highest values for the SAM group, were detected. Ibáñez et al. (2002) found a higher LDH activity in lambs transported at low (4 lambs/m²) than at high stocking density (8 lambs/m²) which is in agreement with our results. On the other hand, Grandin (2000) indicated that sheep may not need to lie down during short journeys. This could explain the absence of differences among SAH and SAL groups on LDH activity.

SA is an economical important factor, because reducing space can reduce costs of transport. In general, smaller space allowances lead to lower unit costs of transport since more animals can be carried in the vehicle (SCAHAW, 2002). A range of 0.16 to 0.30 m²/lamb could be acceptable during transport but investigation on behaviour will be required.

In agreement with the findings by De Boer et al. (1989), fasting induced a rise in catecholamine levels, particularly of noradrenaline. In addition, catecholamine produced a contraction of the spleen with the subsequent increase in blood red cells values (Marco & Lavín, 1999). For this reason, the highest RBC values in the FAST-SAM group after lairage could be associated to the higher catecholamine value found in this group. However, our results contrast with the lack of variations in haematological values reported by Liste et al. (2011). Perhaps the different values in noradrenaline sampled after lairage could be related to the significant interaction SA × TL that was observed and could explain the results on RBC.

Increase in cortisol concentration has been associated with feed-deprivation stress in Merino lamb (Zimerman et al., 2013), which contrasts with the results of cortisol found in the present paper. However, cortisol levels can vary greatly as it is a time-dependent response with high individual variability and therefore, comparisons among studies must be done with caution.

Glucose metabolism decreases due to the decline of propionate production, the major precursor for gluconeogenesis, in the rumen caused by low feed intake (Bergman, 1975). The lowest plasma glucose values found after lairage at the FASTING groups could be useful as an indicator of the intensity of stress, which is in agreement with the results by Kannan et al. (2000) for goats. The higher levels of glucose found in the present study when animals had access to feed during lairage (similar values to the basal conditions, on farm) could indicate recuperation overnight.

Liste et al. (2011) indicated that CK increase could be associated with a feed deprivation during lairage. However, in our study CK levels were similar across all TL groups, as found by Zimerman et al. (2011). Gupta et al. (1999) associated the LDH decrease with feed deprivation in equids, which contrast with our results. Obviously this could respond to the different physiology of these two species. In addition, lairage without feed represented an additional stress, especially in lambs of SAM group. For this reason it is speculated that some physiologic parameters such as LDH increased in this treatment (SAM-TLFAST). This could indicate cumulative effects of the transport and lairage conditions.

With regard to the results of the discriminant analysis, the most notable data showed that this analysis only included the glucose and noradrenaline as the main parameters that allow differentiating among lairage treatments In addition, the highest noradrenaline concentration and lowest plasma glucose values found after lairage in the FASTING groups could be useful as indicators of the intensity of stress, as found for goats (Kannan et al., 2000).

Lairage allows the animals to recover from the stress endured during transport and consequently to improve meat quality (Rabaste et al., 2007). Feed deprivation during lairage is a common practice to reduce gut contents, therefore minimizing the risk of carcass contamination. However, the results of this study indicate that fasting cause some physiological changes which would be indicative of compromised welfare. In addition, after 18 h lairage with access to feed, some physiologic parameters associated with stress (such as glucose, noradrenaline) or physical fatigue (such as LDH) returned to basal (farm) levels. Therefore, these results suggest that feeding during lairage could be more appropriate than fasting, especially when animals are stressed after transport, since it seems to allow overnight recovery. Nonetheless, it is not clear whether these changes are indicative of adaptation to feed deprivation, a sign of metabolic depletion or a compromised welfare. Though recommended from these results, the decision of feeding or fasting would also depend on other factors such as the economic cost, carcass hygiene or meat quality.

Some recommendations to meat industry could be given on the basis of the information obtained in this paper: From welfare standpoint a range of space allowance during transport of between 0.16 and 0.30 m²/lamb could be recommended without showing major changes on welfare physiological indicators. Feeding during lairage could be more appropriate than fasting, especially when animals showed more stress after transport, since it allows recuperation overnight.