Вы находитесь на странице: 1из 6

Meat Science 91 (2012) 435–440

Contents lists available at SciVerse ScienceDirect

Meat Science
journal homepage: www.elsevier.com/locate/meatsci

Different scalding techniques do not affect boar taint

Daniel Mörlein a,⁎, Anne Grave a, Ahmad Reza Sharifi a, Mark Bücking b, Michael Wicke a
Department of Animal Sciences, University of Göttingen, Germany
Fraunhofer Institute for Molecular Biology and Applied Ecology IME, Germany

a r t i c l e i n f o a b s t r a c t

Article history: The prevention of unpleasant boar taint is the main reason for castration of male piglets. For animal welfare
Received 25 October 2011 reasons, castration is announced to be banned in the European Community. This study aimed to investigate
Received in revised form 6 February 2012 whether androstenone, skatole and indole in backfat of boars may be reduced by different scalding technol-
Accepted 24 February 2012
ogies. To discriminate ante and post mortem effects, carcasses were sampled before and after scalding in two
abattoirs using either horizontal (TANK) or vertical (TUNNEL) scalding. Backfat samples were analysed using
Boar taint
gas chromatography (androstenone) and liquid chromatography (skatole, indole). Neither TANK nor TUNNEL
Castration scalding did significantly reduce malodorous compounds. Skatole and androstenone in backfat obtained after
Scalding scalding averaged 112 ± 123 ng/g and 1196 ± 885 ng/g melted fat, respectively; significant differences be-
Consumer acceptance tween abattoirs were observed for skatole. Increased skatole levels were tentatively assigned to longer trans-
Slaughter technology port duration. Concluding from recent consumer research and subsequent application of suggested sensory
Duroc rejection thresholds for androstenone (2000 ng/g) and skatole (150 ng/g), nearly 30% of the carcasses may
be unacceptably tainted.
© 2012 Elsevier Ltd. All rights reserved.

1. Introduction suggestions. Bonneau and Chevillon (2011) also showed that the
abundance of skatole is a limiting factor for the acceptance of boar
For animal welfare reasons, castration of male piglets is discussed meat: In the absence of skatole, androstenone up to 2000 to
controversially. It effectively prevents the occurrence of the so called 3000 ng/g fat did not affect consumer acceptance of loin chops with
“boar taint”, that is suggested to be mainly caused by the testicular ste- 5 mm fat cover. However, considerable inter-laboratory differences
roid androstenone (5a-Androst-3ene-16one), skatole (3-methylindole) for determination of malodorous compounds have been reported
and indole (Lundström, Matthews, & Haugen, 2009). By 2018 castration (Ampuero Kragten et al., 2011; Haugen, Brunius, & Zamaratskaia,
of piglets is, however, announced to be banned in the European Com- 2011) making comparisons of different sensory studies difficult. Har-
munity (European Declaration on alternatives to surgical castration, monization of chemical analysis methods is clearly needed (Haugen
2011). et al., 2011) to derive valid rejection thresholds.
Varying sensory rejection thresholds for androstenone (500 to To reduce the risk of consumer rejection when introducing boar
1000 ng/g fat) and skatole (200 to 250 ng/g fat) have been used to es- meat in the market, the proportion of carcasses exceeding the rejec-
timate the percentage of unacceptably tainted carcasses (Lundström tion levels of the malodorous compounds is searched to be minimized
et al., 2009). For androstenone, until May 2010 there was a German by several strategies, i.e. breeding, husbandry, feeding (Zamaratskaia
legal threshold of 500 ng/g fat below of which carcasses would be ac- & Squires, 2009).
ceptable (Fleischhygiene-Verordnung, 2001). Contrarily, Norwegian To the best of our knowledge, no studies have been performed on
consumers did not rate meat patties containing 20% fat with added the effect of slaughter facilities on boar taint compounds. However,
androstenone below 3000 ng/g fat significantly worse than non- subjectively observed differences in relative proportions of tainted
spiked patties (Lunde, Skuterud, Nilsen, & Egelandsdal, 2009). For carcasses in two different commercial slaughter facilities within the
skatole, the acceptance thresholds are suggested to be considerably same company were reported to us recently. Those observations
lower. Very recent studies showed that the threshold for skatole were putatively attributed to technical differences between abattoirs,
should rather be as low as 150 ng/g (Bee & Ampuero, 2008; Lunde, i.e. scalding technology.
Skuterud, Hersleth, & Egelandsdal, 2010) compared to previous Scalding and singeing is applied to pig carcasses during slaughter
for dehairing and hygienic reasons, i.e. decontamination. Commonly,
carcasses are immersed in scalding water tanks at 60 to 65 °C for a pe-
⁎ Corresponding author at: University of Göttingen, Department of Animal Sciences,
Albrecht-Thaer-Weg 3, D-37075 Göttingen, Germany. Tel.: + 49 551 39 5611; fax: + 49
riod of about 6 min, followed by brushing, and finally singeing at ap-
551 39 5587. proximately 1200 °C for 10 s before evisceration (Maribo, Olsen, &
E-mail address: daniel.moerlein@agr.uni-goettingen.de (D. Mörlein). Barton-Gade, 1998; Monin & Talmant, 1995). More recently,

0309-1740/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
436 D. Mörlein et al. / Meat Science 91 (2012) 435–440

traditional scalding in hot water (e.g. tub or tank), herein referred to as (LMP), backfat thickness (FD) and loin muscle thickness (MD)
horizontal scalding, has been replaced by so called scalding tunnels, at 13th/14th rib as estimated via AutoFOM carcass grading device
herein referred to as vertical scalding. For hygienic reasons scalding (Carometec AS, Herlev, Denmark) were provided by the slaughter
of pigs is increasingly performed in hanging (vertical) conditions, company.
where hot water is sprayed onto the carcass rather they be moved
through tanks. Horizontal rather than vertical scalding is suggested 2.3. Sampling at slaughter, before and after scalding
to possibly result in cross contamination with boar taint compounds
originating from back fat, preputial fluid of boars (Patterson, 1968) Before scalding, approximately 30 s after exsanguination, car-
and faecal impurities that may permeate the fat tissue. Recently, soil- casses were individually marked with ear tags. Then, backfat was
ing of pigs has been shown to increase skatole levels in pigs (Aluwé sampled using a custom made cylindrical knife made from stainless
et al., 2011). Contrarily, temperature treatment was shown to reduce steel (approximately 3 cm inner diameter) attached to a cordless
boar taint compounds during processing of boar meat. Skatole rather electric screw driver. Backfat samples were taken from the neck re-
than androstenone was reduced with open heating while no reduction gion approximately 10 cm from the usual carcass split line. Excised
effect was observed during closed heating (Dehnhard, Claus, Herbert, samples were put individually in high density polyethylene (HDPE)
& Hillenbrand, 1994). It is thus hypothesized that vertical scalding bags and stored on dry ice. Approximately 6 h after slaughter all sam-
using hot water spray may potentially reduce the odorous compounds ples were evacuated and stored individually in HDPE bags at − 20 °C
compared to horizontal scalding in hot water tub where cross contam- until analysis. Ear tags were read before singeing and related to the
ination could possibly rather increase the abundance of malodorous carcass i.d. that was automatically printed on the back of each carcass
compounds. This is also of interest as breeding projects to reduce by the AutoFOM printer (Carometec AS, Denmark). The enormous
boar taint are currently under way, e.g. in Germany 1000 boars have slaughter line speed, i.e. approximately 1300 carcasses per hour, did
been examined in several pig performance testing stations and not allow all, but a subsample of all carcasses to be investigated be-
slaughtered in several abattoirs (Frieden, Mörlein, Meier-Dinkel, fore scalding (n = 81). After careful data inspection 9 data sets had
Boeker, & Tholen, 2010). It is therefore important to elucidate poten- to be discarded due to erroneous assignment of samples before/
tial effects of different slaughter technologies; the present study after scalding. Thus, 72 carcasses were available for investigating
aimed at comparing different scalding technologies that are com- the general effect of scalding, i.e. 27 of farm BDu75 (only for abattoir
monly applied, i.e. horizontal vs. vertical scalding, in two commercial TANK) and 45 of farm ADu50 (21 for TUNNEL, 24 for TANK scalding).
abattoirs. Part of the carcasses was sampled both before and after The ADu50 samples were used to evaluate the effect of different scald-
scalding to assess, if any, reduction effects and to discriminate ante ing technologies (Fig. 1).
and post mortem effects, i.e. handling/transportation and scalding. After quick chilling, about 3 h post mortem backfat samples were
excised from the neck region, close to the previously excised samples
2. Materials and methods as described above. Samples were put in HDPE bags, stored on dry ice
and transported to the lab, where they were evacuated and stored in
2.1. Slaughter facilities HDPE bags at − 20 °C until analysis. In total, backfat samples of 123
pigs of 2 farms were collected in 2 abattoirs after quick chilling.
In abattoir A (“TUNNEL”) vertical scalding was performed using
three consecutive chambers applying sprayed water (about 60 °C). 2.4. Chemical analysis for boar taint compounds
The carcasses were singed by gas burners 3 times, i.e. once after
each scalding treatment for approximately 3 s. Total duration of Skin and the third (inner) fat layer were removed from the backfat
scalding and singeing was approximately 6 min. In abattoir B samples. Altogether, about 4.0 g were taken from 5 different spots on
(“TANK”) scalding was performed horizontally in a scalding tank the sample. This sample was then heated in a microwave oven. 500 μl
with hot water (approximately 62 °C). Scalding time was about of pure melted fat phase were transferred into a 1.5 ml tube. Then
4 min. The carcasses were singed only once for approximately 3 s be- 0.2 ml of internal standard solution (5α-Androstan-3-on) was
fore evisceration. In both abattoirs, pigs were stunned with carbon added. The mixed sample was centrifuged in a cooling centrifuge,
dioxide using a backloader system (Butina A/S, Denmark; 90% v/v, freezed out and the fat phase was separated. For further sample
120 s) followed by vertical exsanguination. clean up, a conditioned C18-SPE-column (500 mg, 3 ml-unit) was
washed with methanol, the methanol was evaporated and the sample
2.2. Animals was filled in a 8 ml test tube. 100 μl of a derivatization solution was
added and heated for 30 min at 65 °C in a drying chamber. The organ-
Two different Duroc sired crossbred types of entire male pigs were ic phase was separated by centrifugation and the upper organic phase
raised at 2 commercial farms, i.e. farm ADu50: Duroc*(Large White was transferred into a 2 ml sample vial and evaporated under a nitro-
*Landrace); farm BDu75: Duroc*(Duroc *(Large White *Landrace)). gen flush to an end volume of approximately 150 μl.
Boars were raised separately from female pigs. Pigs were fed com- To quantify androstenone an in-house gas-chromatography/mass
mercial feedstuff (farm ADu50: 13.4 MJ/ME, 3.6% crude protein; farm spectrometry (GC–MS) method was applied as previously described
BDu75: 13.0 MJ/ME, 4.5% crude protein). Boars were raised in groups
of 17 (farm ADu50) and 20 (farm BDu75) animals per pen. Before trans- n = 30 (24) n = 30 (27)
port to the abattoirs, pigs were withdrawn from feeding for 24 h abattoir TANK
(farm ADu50) and 15 h (farm BDu75). At farms, boars were randomly
selected to be transported to either of the abattoirs. The pigs were farm ADu50 farm BDU75

transported in mixed groups. From both farms, duration of transport abattoir TUNNEL
to abattoir TUNNEL did not exceed 2 h while transportation to abat- n = 31 (21) n = 32 (-)
toir TANK was approximately 6 h. Pigs were slaughtered July/6
(TANK) and July/7 2010 (TUNNEL). Average daily temperature was Fig. 1. Design of experiment: Boars were delivered from farm ADu50 and BDu75 to abattoir
19 and 24 °C, respectively. In both abattoirs, lairage time was approx- TANK (horizontal scalding) and TUNNEL (vertical scalding), respectively. Number of
carcasses sampled after scalding (in brackets: before scalding) are given. Samples of
imately 2 h to allow pigs recover from stress induced by transporta- farm ADu50 obtained before scalding were used to compare scalding technology between
tion. According to visual inspection, all animals were resting quietly abattoirs. Samples obtained after scalding (n= 123) were used for ANOVA (farm and
during lairage time. Hot carcass weight (HCW), lean meat percentage abattoir effect).
D. Mörlein et al. / Meat Science 91 (2012) 435–440 437

(Fischer et al., 2011). GC/MS was carried out using a MSD 5973 inert according to Eq. (1).
with GC 6890 network and fast scan upgrade (Software version E
02.00 SP1, Agilent Technology, Waldbronn, Germany) with a 30 m DIFFbefore−after ¼ ASIbefore −ASIafter ð1Þ
Rxi-5HT, 0.25 mm ID, 0.25 μm film thickness column (Restek, Bad
Homburg, Germany) with retention gap (5 m × 0.25 mm ID deacti- Thus, positive DIFF values are associated with true reduction due
vated fused silica). The temperature program was initiated at 130 °C to scalding while negative DIFF values represent increased levels
for 1 min and then raised to 265 °C at a rate of 30 °C/min. Then, tem- after scalding. DIFF values were related to initial values (ASIbefore)
perature was raised to 300 °C at a rate of 5 °C/min, hold for 1 min, using enhanced Bland Altman analysis (Fernandez & Fernandez,
then raised to 360 °C at a rate of 20 °C/min and maintained for 2009).
10 min. Androstanone (5α-Androstan-3-on) was used as an internal PROC NPAR1WAY was used for non parametric ANOVA of non-
standard. Calibration was done with solutions of androstenone in normally distributed data, e.g. to elucidate the effect of different
methanol in the range of 15 to 600 ng per sample (500 μl fat phase) scalding technologies between abattoirs.
equivalent to 30 to 1200 ng/g melted fat. The limit of detection Frequencies of carcasses exceeding sensory thresholds for andros-
(LOD) was 4 ng per sample (equivalent to 10 ng/g melted fat). LOD tenone and skatole were compared between abattoirs using Chi-
was calculated based of the calibration values by applying a statistical square test and Chi-square for equal distribution, i.e. against 50:50
quality assurance program (SQS 2000, V 1.4, Moos, Germany) accord- probability.
ing to DIN 32645. Effects of slaughter facility (TANK vs. TUNNEL) and farm (ADu50 vs.
For the quantification of skatole and indole an in-house RP-HPLC BDu75) as well as their interaction effects on carcass traits and boar
procedure with fluorescence detection was used: The system was a taint compounds in backfat samples obtained after scalding were
HPLC-system Dionex Ultimate 3000 with a RF 2000 Fluorescence assessed with analysis of variance (ANOVA) using PROC GLM proce-
Detector (Dionex, Idstein, Germany). Chromatographic parameters dure with model (2).
were: Column: Hypersil ODS C18 5 μm, 150 × 4.60 mm (MZ Analysen-
technik, Mainz, Germany); pre-column: Hypersil ODS C18 5 μm, yijk ¼ μþfarmi þ slaughterj þ farmi  slaughterj þ eijk ð2Þ
10 × 4.60 mm (MZ Analysentechnik, Mainz, Germany); mobile
phase: 60% 0.02 M acetic acid, 25% acetonitrile and 15% i-propanol All carcass traits were normally distributed and could thus be an-
isocratic. The column oven temperature was set at 40 °C and the in- alyzed without transformation. Due to the skewed nature of the
jection volume was 10 μl. The detection parameters were: excitation androstenone, skatole and indole values, data were transformed
and emission wavelength: 270 and 350 nm, respectively; emissions with natural logarithm before ANOVA. This was done after careful
gain and response: 4.0 and 0.5 s (emission sensitivity: medium). Re- outlier analysis (see Results and discussion).
tention times for indole, 2-methylindole (internal standard) and ska- Linear correlations of odorous compounds and carcass perfor-
tole were 4.69, 6.27, and 7.05 min, respectively. Calibration was done mance data were analyzed at farm level using PROC CORR (SAS).
with solutions of skatole and indole in methanol in the range of 15 to
170 ng per sample (400 μl fat phase) equivalent to 40 to 450 ng/g liq- 3. Results and discussion
uid fat. The calibration was done without a matrix, because no ska-
tole/indole-free fat matrix was available. The LOD of both indole 3.1. Carcass traits and boar taint compounds
and skatole was 2 ng per sample (equivalent to 5 ng/g melted fat).
Standard deviations of replicate measurements for both methods Carcass performance data and boar taint compounds data of all
were in the normal range, i.e. between 1 and 5%. carcasses are given in Table 1. Average hot carcass weight was
99.8 ± 7.1 kg. Mean backfat thickness was 16.8 ± 2.6 mm. The
mean hot carcass weight nearly 100 kg is slightly higher compared
2.5. Statistical analyses to average carcass weight in Germany (Mörlein, Link, Werner, &
Wicke, 2007). As estimated by ultrasound based, fully automated
Statistical analyses were performed using SAS 9.2 (SAS Institute grading device (AutoFOM), lean meat percentage (LMP) averaged
Inc., Cary, USA). 56%. LMP is supposed to be underestimated as the estimation for-
To elucidate potential reduction of odorous compounds due to dif- mula currently implemented in the AutoFOM system is based on
ferent scalding technologies, androstenone, skatole, and indole values a representative set of only gilts and castrates. Based on previous
obtained before and after scalding were compared. Difference (DIFF) carcass dissection trials, entire male pigs were reported to have
of androstenone, skatole, and indole (ASI) was computed for each higher lean proportion compared to castrates (Lundström et al.,
carcass as the difference of ASI-values before and after scalding 2009).

Table 1
Descriptive statistics of carcass traits and boar taint compounds of all investigated carcasses.

Trait N Mean STD CV Median P25 P75 Min Max

HCW [kg] 123 99.78 7.12 7.14 100.00 95.40 104.80 83.40 115.40
LMP [%] 123 56.29 2.90 5.15 56.60 54.70 58.10 47.20 63.30
FD [mm] 123 16.78 2.60 15.47 16.66 14.76 18.60 10.57 24.43
MD [mm] 123 59.42 5.30 8.91 60.26 56.21 63.57 42.73 70.59

Boar taint compounds after scaldinga

Indole 123 71.23 49.75 69.85 55.91 40.41 89.91 17.52 339.80
Skatole 123 112.35 123.18 109.64 73.50 51.06 126.31 15.55 1129.61
Androstenone 123 1196.36 885.13 73.99 953.74 646.43 1501.74 22.16 5692.60

FD = backfat depth, MD = loin muscle depth, LMP = lean meat percentage (as estimated by AutoFOM), HCW = hot carcass weight.
Androstenone, skatole and indole are given in ng/g, related to melted fat.
STD = standard deviation, CV = coefficient of variation, P25/P75 = 25th/75th percentile.
438 D. Mörlein et al. / Meat Science 91 (2012) 435–440

Average concentrations of skatole and androstenone in backfat

obtained after scalding were 112 ± 123 ng/g and 1196 ± 885 ng/g
melted fat, respectively. Androstenone and skatole greatly varied
from 21 to 5700 ng/g and 15 to 1130 ng/g melted fat respectively.
To evaluate the proportion of potentially tainted carcasses, relative
frequencies with respect to various thresholds for androstenone and
skatole levels in backfat after scalding are given in Table 2. Concluding
from recent consumer research and subsequent application of rejec-
tion levels for androstenone (2000 ng/g melted fat) and skatole
(150 ng/g melted fat), 30% of the carcasses may be unacceptably
tainted (Table 2) as either of the thresholds was exceeded. Consider-
ing the potential inter-laboratory variation, the proportion of poten-
tially tainted carcasses may not safely be estimated. However,
conclusions from relative differences within the study as intended
in the present investigation are feasible.

3.2. Reduction of odorous compounds by scalding technology

Fig. 2. Androstenone before scalding vs. androstenone reduction (calculated as difference
Androstenone and skatole values before scalding and their abso- before minus after scalding) without respect to abattoir. Horizontal lines denote the 95%
lute change due to scalding are shown in Figs. 2 and 3, respectively. upper and lower limits of reduction and the average reduction level.

Clearly, for the majority of the samples there is no reduction due to

scalding. Some carcasses showed even elevated levels after scalding.
As the 95% confidence interval of the regression line of reduction vs. and androstenone. Subsequently, frequencies of tainted carcasses
level before scalding includes the “zero bias” line there is no signifi- were different between abattoirs, too (Table 5). Carcass weight was
cant reduction due to scalding. Comparing horizontal and vertical not affected by farm or abattoir (Table 6). Abattoir significantly affect-
scalding, there is no evidence for a significant difference between ed lean meat percentage, backfat thickness and loin muscle thickness.
the technologies studied, neither for androstenone nor for skatole Backfat thickness and loin muscle thickness were significantly higher
(Table 3). The observed variation of odorous compound levels before for farm ADu50. For all fat samples after scalding, a slight correlation
and after scalding is rather suggested to be within the order of the between androstenone and skatole was found (r = 0.23, p b 0.05),
intra-assay variation of the chemical analysis (about 1 to 5%) or slight while skatole and indole are highly correlated (r = 0.75, p b 0.001).
inter-individual variation between two sampling locations within Correlation analysis at farm level revealed a closer relationship
carcass (Haugen et al., 2011). between androstenone and skatole (r = 0.37, p b 0.01) as well as for
From previous investigations both a contamination due to androstenone and indole (r = 0.58, p b 0.001) for farm BDu75 while
horizontal scalding and a potential reduction due to repeated vertical there was no significant correlation for farm ADu50. Correlations of
scalding followed by short time singeing were anticipated. Long term boar taint compounds with carcass traits were computed at farm
heat application, e.g. when producing cooked ham or Wiener type level, too. For farm BDu75, androstenone was significantly decreased
sausages, was shown to effectively reduce skatole. Only long term with lean meat percentage (r = −0.44) while this was not observed
“open heating” reduced also androstenone (Dehnhard et al., 1994). for farm ADu50. Skatole and indole were not significantly affected by
In the present study neither horizontal (TANK) nor vertical (TUNNEL) carcass traits. Apparently, slaughter weight does not affect androste-
scalding effectively reduced any of the boar taint compounds as none or skatole content in backfat (data not shown). From previous
evaluated by chemical analysis. Most probably, short duration of studies it is known that correlations of androstenone and skatole
heat application during scalding and singeing at slaughter does not are not consistent (Zamaratskaia & Squires, 2009). As this study
affect boar taint compounds. Obviously, cross contamination with focused primarily at the slaughter technology, other management
androstenone or skatole in TANK scalding is not occurring either. factors such as feeding or husbandry were not standardized.

3.3. ANOVA and correlations

Despite no effect of scalding itself was found, ANOVA revealed

significant differences for, skatole and indole in backfat samples
obtained after scalding between abattoirs (Table 4). Farm also signif-
icantly affected abundance of malodorous compounds. Boars from
farm BDu75 showed significantly increased levels of skatole, indole

Table 2
Relative frequencies (%) of potentially tainted carcasses according to suggested sensory
threshold levels for androstenone and skatole in backfat (after scalding), n = 123 (w/o
exclusion of extreme values).

Androstenone Skatole (ng/g melted fat)

(ng/g melted fat)
b=150 >150 b= 250 >250 Total

b=500 11.38⁎ 1.63 0.81 13.82

> 500 b= 1000 34.96⁎ 4.88 0.81 40.65
> 1000 b= 2000 24.39⁎ 5.69 3.25 33.33
> 2000 8.13 0.81 3.25 12.20
Fig. 3. Skatole before scalding vs. skatole reduction (calculated as difference before
Total 78.86 13.01 8.13 100.00
minus after scalding) without respect to abattoir. A single extreme skatole value
⁎ Suggested not to be tainted according to recent consumer research. (> 1000 ng/g melted fat) was discarded for regression analysis and graphical output.
D. Mörlein et al. / Meat Science 91 (2012) 435–440 439

Table 3
Arithmetic means of odorous compounds before (B) and after (A) scalding and their relative change (DIFFB-A%) with respect to scalding technology (TUNNEL vs. TANK).

TUNNEL; n = 21 TANK, n = 24 p#

Before (B) After (A) DIFFB-A% Before (B) After (A) DIFFB-A%
[ng/g] [ng/g] [ng/g] [ng/g]

Indole 42.5 36.1 14.0 55.4 60.9 − 10.6 b0.001

Skatole 59.6 60.7 − 6.3 88.0 86.3 2.6 n.s.
Androstenone 942.8 990.5 − 6.1 1280.8 1293.5 − 1.0 n.s.

(Only farm ADu50) *androstenone, skatole and indole are given in ng/g, related to melted fat.
DIFFB-A% = (ASIbefore − ASIafter scalding)/ASIbefore * 100% (calculated for each carcass and then averaged).
p refers to significance level of nonparametric ANOVA/mean comparison of DIFFB-A% between abattoirs.

Because scalding technology itself did not affect taint compounds, intestinal passage rate; while there is low energy available for gut
varying abundance of malodorous compounds in samples obtained microbes, they increasingly produce indoleic compounds from endoge-
after scalding may be attributed to pre-slaughter treatment, e.g. trans- nous tryptophan (Claus et al., 1994). Longer feed withdrawal may thus
portation and handling. With respect to different transportation time increase both skatole synthesis and resorption rate into plasma. 12 to
to the abattoirs, several factors are known to stress pigs, e.g. mixing 15 h of pre-slaughter fasting is commonly practiced to reduce the risk
unfamiliar animals, feed withdrawal, noise, etc. (Rosenvold, 2003). of microbial cross contamination, mortality during transport and
With stress, increased cortisole levels may advance catabolism, apoptosis carcass yield. No more than 24 h of fasting is practiced, as extended
and gut mucosa cell debris, and thus increased endogenous tryptophan feed withdrawal and lairage time may also compromise animal welfare
may be used for skatole synthesis (Claus, Weiler, & Herzog, 1994). Previ- (Rosenvold, 2003). Such a short term skatole deposition into fat tissue
ously, injection of human chorionic gonadotropin (hCG) was shown to remains questionable.
increase androstenone, but not skatole, in plasma and fat (Chen,
Zamaratskaia, Madej, & Lundstrom, 2006) within a few days. To the
best of our knowledge, androstenone or skatole levels in backfat have 4. Conclusions and implications
not been reported to increase as shortly as to explain the observed differ-
ences in the present study. Boars originating from two commercial farms were each slaugh-
Breed type as well as husbandry and feeding were reported to tered in two abattoirs applying different scalding technologies.
affect boar taint compounds (Zamaratskaia & Squires, 2009). However, Neither vertical (TUNNEL) nor horizontal (TANK) scalding did rele-
the present study did not aim at investigating these effects but mainly vantly reduce or change any of the major boar taint compounds'
slaughter technology. Two commercial farms were randomly selected. level in backfat. Nevertheless, significant differences of skatole and
Thus, the obvious “farm”-effect was likely to occur while breed type, indole levels were observed between abattoirs and are suggested to
husbandry conditions and feeding are confounded. As the average be associated with longer transport duration, feed withdrawal and
skatole levels were higher for farm BDu75, it is assumed that, both cross- ante mortem stress. However, it remains unclear whether this is
bred type (75% Duroc vs. 50% Duroc), prolonged feed withdrawal and physiologically feasible. Differences between farms are suggested to
less dietary energy (13 MJ vs. 13.4 MJ) may have caused increased be caused by genetic effects, feeding and husbandry. A follow up
skatole levels. Higher Duroc proportion is likely to elevate skatole levels study will focus on the effects of feed withdrawal on the abundance
(Zamaratskaia & Squires, 2009). Feed withdrawal results in lower of androstenone and skatole.

Table 4
Effect of farm and abattoir and their interaction on odorous compounds in backfat of
entire male pigs after scalding (LS-means; all values are given in ng/g melted fat). The authors wish to thank Johann Lehnhof, Wilhelm Wemheuer and
Erwin Tönges for their help collecting the samples at slaughter. We
Effect Skatole Indole Androstenone
(n = 120) (n = 121) (n = 122) gratefully acknowledge the financial support of the study by Tönnies
Fleisch and the support of Wilhelm Jaeger and Hansjörg Eynck at
Farm (F)
ADu50 62.3a 45.3 761.0a
BDu75 107.0b 78.7 1083.6b

Abattoir (A) Table 5

TANK 99.5a 73.0a 934.7a Effect of scalding technology on frequency of potentially tainted carcasses with respect
TUNNEL 67.0b 48.9b 882.2a to different sensory threshold levels.

Abattoir × Farm (A × F) Skatole# Androstenone#

TUNNEL × ADu50 56.1a 36.4 798.4
>150 ng/g >250 ng/g >500 ng/g > 1000 ng/g >2000 ng/g
TUNNEL × BDu75 79.9b 65.6 974.8
TANK × ADu50 69.2ab 56.5 725.3 TUNNEL 7 1 54 28 4
TANK × BDu75 143.2c 94.3 1204.5 (n = 63)
TANK (n = 60) 19 9 52 28 11
p-values (ANOVA) Chi-square 7.79 7.40 0.02 0.0 4.12
F b.0001 b.0001 0.0135 p+ b0.0053 0.0065 0.8784 1.0 0.0423
A b.0001 b.0001 0.6824 Chi-square50/50 5.54 6.40 0.03 0.0 3.27
A×F 0.0387 0.5529 0.2773 P50/50++ 0.019 0.011 0.846 1.0 0.071
*Extreme values greater than P75 + 3*(P75-P25) were removed before ANOVA, i.e. Androstenone and skatole values (after scalding) are given in ng/g, related to
indole > 238.41 ng/g (2 extremes deleted); skatole > 352.06 ng/g (3 extremes deleted); melted fat.
androstenone > 4067.67 ng/g (1 extreme deleted), data were transformed with natural P refers to significance level of Chi square test.
logarithm, LS-means were back-transformed. Values with different superscripts within P50/50 refers to significance level of Chi square test for equal distribution (50:50)
columns are significantly different. between abattoirs.
440 D. Mörlein et al. / Meat Science 91 (2012) 435–440

Table 6 Dehnhard, M., Claus, R., Herbert, E., & Hillenbrand, M. (1994). Skatol- und Androstenon-
Effect of farm and slaughter and their interaction on carcass traits of entire male pigs konzentrationen in Fleischerzeugnissen aus Eberschlachtkörpern. Die Ebermast.
with respect to abattoir and farm of origin (LS-means; n = 123). Angewandte Wissenschaft Heft 449. Schriftenreihe des Bundesministeriums für
Ernährung, Landwirtschaft und Forsten. (pp. 55–71).
Effect Carcass traits European Declaration on alternatives to surgical castration. Retrieved September 14, 2011,
from. http://ec.europa.eu/food/animal/welfare/farm/initiatives_en.htm. (2011).
HCW [kg] LMP [%] FD [mm] MD [mm] Fernandez, R., & Fernandez, G. (2009). Validating the Bland–Altman method of
agreement. SAS Conference Proceedings: Western Users of SAS Software 2009. San
Farm (F)
Jose, California Retrieved from. http://www.lexjansen.com/wuss/2009/pos/POS-
ADu50 100.23 56.12 17.56a 61.33a
BDu75 99.32 56.41 16.04b 57.46b Fischer, J., Elsinghorst, P., Bücking, M., Tholen, E., Petersen, B., & Matthias, W. (2011).
Development of a candidate reference method for the simultaneous quantitation
Abattoir (A) of the boar taint compounds androstenone, 3a-androstenol, 3β-androstenol, ska-
TUNNEL 100.16 57.16a 16.30a 61.28a tole, and indole in pig fat by means of stable isotope dilution analysis-headspace
TANK 99.39 55.38b 17.30b 57.50b solid-phase micro. Analytical Chemistry, 83, 6785–6791.
Fleischhygiene-Verordnung (2001). Verordnung über die hygienischen Anforderungen
Abattoir × Farm und amtlichen Untersuchungen beim Verkehr mit Fleisch. Bekanntmachung vom
TUNNEL × ADu50 100.21 57.03 16.76 63.62 29. Juni 2001. Bundesgesetzblatt, 1366.
TUNNEL × BDu75 100.12 57.28 15.83 58.94 Frieden, L., Mörlein, D., Meier-Dinkel, L., Boeker, P., & Tholen, E. (2010). Boar taint levels
TANK × ADu50 100.25 55.22 18.35 59.04 and performance data in Pietrain sired crossbred males in Germany. 61st Annual
TANK × BDu75 98.53 55.53 16.25 55.97 EAAP Meeting, Heraklion (pp. 142).
Haugen, J., Brunius, C., & Zamaratskaia, G. (2011). Review of analytical methods to
measure boar taint compounds in porcine adipose tissue: The need for harmonised
methods. Meat science (10.1016/j.meatsci.2011.07.005). Elsevier Ltd.
F 0.4865 0.5750 0.0008 b.0001
Lunde, K., Skuterud, E., Hersleth, M., & Egelandsdal, B. (2010). Norwegian consumers'
A 0.5501 0.0006 0.0245 b.0001 acceptability of boar tainted meat with different levels of androstenone or ska-
A×F 0.5297 0.9567 0.1867 0.3329 tole as related to their androstenone sensitivity. Meat Science, 86(3), 706–711,
HCW = hot carcass weight, LMP = lean meat percentage (as estimated by AutoFOM), doi:10.1016/j.meatsci.2010.06.009.
Lunde, K., Skuterud, E., Nilsen, A., & Egelandsdal, B. (2009). A new method for differen-
FD = backfat depth, MD = loin muscle depth.
tiating the androstenone sensitivity among consumers. Food Quality and Preference,
Within column, least squares (LS-) means with the same superscript letters are not
20(4). (pp. 304–311). Elsevier Ltd., doi:10.1016/j.foodqual.2009.02.001.
significantly different (p b 0.05). Lundström, K., Matthews, K. R., & Haugen, J. E. (2009). Pig meat quality from entire
males. Animal, 3(11), 1497–1507, doi:10.1017/S1751731109990693.
References Maribo, H., Olsen, E. V., & Barton-gade, P. (1998). Comparison of dehiding versus scald-
ing and singeing: Effect on temperature, pH and meat quality in pigs. Meat Science,
Aluwé, M., Bekaert, K. M., Tuyttens, F. A. M., Vanhaecke, L., De Smet, S., De Brabander, H. F., 50(2), 175–189.
et al. (2011). Influence of soiling on boar taint in boars. Meat Science, 87(3). (pp. Monin, G., & Talmant, A. (1995). Effects on carcass weight and meat quality of pigs
175–179). The American Meat Science Association, doi:10.1016/j.meatsci.2010.10.005. dehaired by scalding or singeing post-mortem. Meat Science, 39, 247–254.
Ampuero Kragten, S., Verkuylen, B., Dahlmans, H., Hortos, M., Garcia-Regueiro, J. A., Dahl, Mörlein, D., Link, G., Werner, C., & Wicke, M. (2007). Suitability of three commercially
E., et al. (2011). Inter-laboratory comparison of methods to measure androstenone in produced pig breeds in Germany for a meat quality program with emphasis
pork fat. Animal, 5(10), 1634–1642, doi:10.1017/S1751731111000553. on drip loss and eating quality. Meat Science, 77(4), 504–511, doi:10.1016/
Bee, G., & Ampuero, S. (2008). Wie hat die elektronische Nase gelernt, den Ebergeruch j.meatsci.2007.04.030.
zu erkennen? KTBL Schrift, 469, 60–66. Patterson, R. L. S. (1968). Acidic components of boar preputial fluid. Journal of the Science
Bonneau, M., & Chevillon, P. (2011). Acceptability of entire male pork with various levels of of Food and Agriculture, 19, 38–40.
androstenone and skatole by consumers according to their sensitivity to androstenone. Rosenvold, K. (2003). Factors of significance for pork quality—A review. Meat Science,
Meat Science. : Elsevier Ltd., doi:10.1016/j.meatsci.2011.07.019. 64(3), 219–237, doi:10.1016/S0309-1740(02)00186-9.
Chen, G., Zamaratskaia, G., Madej, A., & Lundstrom, K. (2006). Effect of hCG administration Zamaratskaia, G., & Squires, E. (2009). Biochemical, nutritional and genetic effects
on the relationship between testicular steroids and indolic compounds in fat on boar taint in entire male pigs. Animal, 3(11), 1508–1521, doi:10.1017/
and plasma in entire male pigs. Meat Science, 72(2), 339–347, doi: S1751731108003674.
Claus, R., Weiler, U., & Herzog, A. (1994). Physiological aspects of androstenone and
skatole formation in the boar—A review with experimental data. Meat Science,
38(2), 289–305, doi:10.1016/0309-1740(94)90118-X.