Consumer studies on sensory acceptability of boar taint: A review

Boar taint can be found in meat from boars and affects consumer acceptability of pork. The aim of this review is: (1) to describe different aspects of the existing methodologies used in consumer studies when evaluating boar taint from a sensory point of view, (2) to draw conclusions on different studies regarding the acceptability of meat from entire males, and (3) to discuss a possible harmonization of the different aspects to be considered when performing consumer studies on boar taint. This paper focuses on different aspects of studies previously carried out such as the country of assessment, the location of the test, the cooking procedure, the type of meat samples evaluated, the attributes and scales used, consumer profile, the results obtained, and the effect of androstenone sensitivity of the consumers on boar meat acceptability. A discussion on the possibility of a harmonization of the different aspects is also performed and final remarks and considerations have been drawn.     

Effect of dietary chicory on boar taint

Following preliminary screening and feeding trials on farms supplying a commercial abattoir, 360 entire male pigs were used to evaluate the effects of different percentages of chicory (Cichorium intybus L.) on levels of boar taint compounds and sensory aspects in backfat. Pigs were fed 0, 3, 6 or 9% chicory in the diet, 30 pigs being sampled at 3 different times: initially to measure basal levels of skatole and androstenone and after 1 and 2 weeks on the test diets. Cooked samples of backfat were presented to a trained sensory panel for "sniff" tests. Chicory fed at 9% for 2 weeks reduced skatole levels significantly (P<0.001), with 0.55 of pigs below 0.05 μg/g, typical of levels in castrated males. Abnormal odour scores were significantly lower for pigs in this group compared with 0% pigs (P<0.001), however, androstenone concentration was significantly higher in this group after the 2 week feeding period (P<0.005). Thus, feeding 9% chicory for 2 weeks was effective in reducing backfat skatole concentrations and abnormal odour scores of cooked fat but not androstenone concentration.     

Influence of soiling on boar taint in boars

It has been suggested that skatole, one of the main compounds responsible for boar taint, can be lowered by keeping pigs clean, as skatole can be absorbed through skin and/or lungs (Hansen, Larsen, Jensen, HansenMoller & Bartongade, 1994). With this experiment, we further investigated this hypothesis by comparing extremely clean with extremely dirty animals with regard to the occurrence of boar taint. One group of boars was washed daily and pens were mucked on and littered down daily (CLEAN), a second group of boars was rubbed with faeces daily (DIRTY) and a third group of boars was kept in control conditions (CONTROL). The treatment was performed during the last four weeks before slaughter. According to the standardised consumer panel evaluations, boars subjected to extra soiling had a higher concentration of boar taint than boars that were kept extra clean. In contrast, expert panels judged general meat flavour to be inferior in CLEAN than CONTROL pigs. The home consumer panel, the hot iron method, and laboratory analyses, i.e., the presence of indole, skatole and androstenone in fat and serum, all showed no significant differences. So no clear indications towards skatole reduction by improving cleanliness of pigs were found.     

Perceptual masking of boar taint in Swedish fermented sausages

Surgical castration of male piglets has traditionally been practiced to avoid development of boar taint in pork meat which can occur if entire male pigs are raised. Boar taint is commonly characterised as exhibiting the odour and flavour of urine and manure. This study involves sensory characterisation of the possibilities to mask boar taint in meat from entire male pigs by fermentation and smoking to maintain high sensory quality in meat products if castration is prohibited. Model and commercial type Swedish fermented sausage products based on low or high levels of boar tainted fat, three different starter cultures and two different levels of smoking were studied. In the model sausages, liquid smoke masked the perception of boar taint. In contrast, the smoking procedure of the commercial sausages was insufficient to totally mask the perception of boar taint. In both the model and commercial sausages, the aroma development from the starter cultures lowered the perception of boar taint but was insufficient for total perceptual masking. Due to the total masking effect of smoking in the model sausages, it was clear that smoke may present a potential solution to remove the perception of boar taint in fermented sausages if the smoking procedure is optimised.     

A human nose scoring system for boar taint and its relationship with androstenone and skatole

A system for sensory evaluation of boar taint was used to evaluate boar taint in fat samples from 6574 entire males. The term "human nose scoring" has been used to describe this system. The samples from each boar were heated with a hot iron and three panelists assigned scores of 0 to 4. The reproducibility of HNS ranged from 0.19 to 0.32 reflecting natural variation in the ability of human beings to detect different odors. The correlations of HNS with androstenone ranged from 0.22 to 0.52, while those with skatole ranged from 0.31 to 0.89, suggesting that skatole is a better predictor of boar taint. Considering (1) the relationship of HNS with the boar taint compounds, (2) the ability of HNS to capture variation not accounted for by the boar taint compounds, (3) low estimation costs and (4) low time requirements, HNS can be used in large scale evaluations of boar taint.     

Norwegian consumers' acceptability of boar tainted meat with different levels of androstenone or skatole as related to their androstenone sensitivity

The aim of work was to study Norwegian consumers' acceptance of pork meat with different levels of skatole and androstenone. One group of androstenone sensitive consumers (N = 46) and one group of non sensitive consumers (N = 55) participated in a home test and evaluated 11 samples with different skatole (range 0–0.35 ppm) and androstenone (range 0–9.0 ppm) levels. Liking of odour during frying and odour and flavour of the fried meat were evaluated. Results showed that the non sensitive consumers accepted all levels of androstenone in the samples. Sensitive consumers gave a significantly lower liking score for androstenone samples containing 3 ppm (and more) than the reference sample when evaluating these samples above the frying pan, but no significant difference were found between 3 ppm samples and reference samples when liking of fried meat was evaluated. This indicated that the sensitive consumers accepted 3 ppm in fried meat, but not if 3 ppm was present in the sample during the frying process. The same consumer's differentiated skatole samples with regard to flavour at 0.15 ppm. The Norwegian established practise with a threshold value of 0.21 ppm skatole is higher than the value accepted by the consumers.     

Consumer perception of boar meat as affected by labelling information,malodorous compounds and sensitivity to androstenone

This study aimed to assess the influence of two label conditions on the acceptance of boar meat. A central location test was conducted with 145 consumers each assessing 4 pieces of pork loin.     

A comparison of the carcase characteristics of pigs immunized with a ‘gonadotrophin-releasing factor (GnRF)’ vaccine against boar taint with physically castrated pigs

Evaluating the effect of using a GnRF vaccine against boar taint on the carcase characteristics of boars, vaccinated pigs were compared with physically castrated. In total, 554 male pigs were randomly assigned to treatment groups. T01 comprised physically castrated pigs in the first week of life, T02 comprised pigs vaccinated twice before slaughtering. There was neither significant difference between the groups in terms of average liveweight nor in the hot carcase weight. The mean dressing percentage was 1.5% higher for T01 than for T02 (P < 0.0001). The lean meat percentage was significantly higher in T02 (P < 0.0001). Backfat and backmuscle thickness were significantly higher in T01 (P < 0.0001 and P = 0.0099, respectively). Within the EUROP grading vaccinated pigs were in favour (P = 0.0034). There were no significant differences using the AutoFOM system: weights of the boned ham, boned shoulder and loin (P = 0.5102, P = 0.8881 and P = 0.1919, respectively). The weight of the belly was significantly higher (P = 0.0042) in T01 while the lean meat percentage of belly was significantly higher (P < 0.0001) in T02.     

Evaluation of different heating methods for the detection of boar taint by means of the human nose

No automated detection system for boar taint detection is currently available, thus boar taint at the slaughterline can currently only be assessed using the singeing method (olfactory scoring). This study compares several heating methods (microwave, soldering iron and pyropen) and evaluates the effect of habituation, cleaning the soldering iron, singeing the fat twice in the same place, and variations in the technical procedures. All methods seem to be suitable for detecting boar taint but the choice of heating method for sensory scoring of boar taint depends on habituation of the trained assessor and specific conditions applied. The pyropen seems to be most suitable because it does not contact the fat and is easy to handle (wireless). Finally, the intensity score may also be influenced by: contamination from not cleaning the soldering iron, singeing the fat twice in the same place, and the effect of habituation.     

Predicting the likelihood of developing boar taint: Early physical indicators in entire male pigs

Three potential early-age predictors of which boars are likely to develop boar taint (testes volume, skin lesions and dirtiness) were measured on 102 boars every fortnight from 10weeks of age until slaughter. These predictors were correlated with the level of boar taint according to the hot iron method and the concentrations of skatole and androstenone as determined by chemical analysis. The chance of no/low boar taint according to the hot iron method decreased with higher testes volume (weeks 22 and 24) and increased with skin lesion score (weeks 12, 16 and 18). For the concentrations of androstenone and skatole, the strongest correlation was found with testes volume in week 12. Skin lesions in week 16 were negatively correlated with skatole levels. Dirtiness was negatively correlated with skatole concentrations (week 18) but positively correlated with androstenone concentrations (weeks 20 and 22). Testes volume has the greatest potential for predicting the likelihood of developing boar taint.     

Comparison of heating methods and the use of different tissues for sensory assessment of abnormal odours (boar taint) in pig meat

Five heating methods (microwave, hotwire, boiling at 25 °C and 75 °C and melting) were used to generate cooking odours from backfat of entire male pigs and a 'composite' sample consisting of fat and muscle from the head along with salivary glands. The methods elicited significantly different scores for odours from 4 groups of 10 samples differing in their concentrations and ratios of skatole and androstenone. The odours (pork odour, abnormal odour, skatole odour and androstenone odour) were assessed by 3 experienced assessors. Correlations between skatole and androstenone concentrations and abnormal odour score in backfat were higher for skatole, suggesting it is the more important boar taint compound. In the composite sample, androstenone concentration was much higher than in backfat and androstenone was a more important contributor to boar taint. The microwave, hotwire and boiling (75 °C) methods produced the clearest separation between samples and the microwave method was considered the most suitable for on-line use.     

Influence of volatile compounds on the development of off-flavours in pig back fat samples classified with boar taint by a test panel

Dynamic Head Space methodology was applied to evaluate the possible contribution of some volatile compounds to the development of boar taint in pig backfat samples with low concentrations of skatole and androstenone, but which had previously been classified as tainted by a trained test panel. Volatile compounds were collected in a trap of graphited charcoal and analysed by GC–MS in Scan mode. Aldehydes and short chain fatty acids, compounds that play a significant role in the development of undesirable aromas in food products, were the main classes of compounds identified in this study, although the possible contribution of other compounds that were detected in a minor proportion – such as alcohols and ketones – was evaluated. Styrene and 1,4-dichlorobenzene, compounds that may have come from an external contamination, showed a high concentration in the samples classified with boar taint, so these compounds could have been responsible for the development of some off-flavours in the fat samples studied in this work. In the same study, skatole and androstenone were also determined by normal phase HPLC and GCMS, respectively.     

Absence of an effect of dietary fibre or clinoptilolite on boar taint in entire male pigs fed practical diets

This study aimed to evaluate the possibility of reducing boar taint in boars (Piétrain × Hybrid) by addition of different feed ingredients (raw potato starch (RPS) 10%, raw potato starch 10% + wheat bran 5% (RPS + WB), lupins 10%, inulin 5%, clinoptilolite 1%) to a standard diet over a period of 4–6 weeks before slaughter. Control boars (CBOAR) as well as barrows were fed the standard diet. Efficacy of the different feed ingredients was evaluated by different boar taint detection methods: hot iron method, consumer panel, expert panel and laboratory analysis. According to all detection methods, clear differences were noticeable between boars and barrows. No differences in boar taint incidence were found between the boars on the different dietary treatments as assessed by consumers, experts, hot iron method or the concentration of skatole in fat. A significant effect on indole level was found, but no further differentiation could be made. The concentration of backfat androstenone was significantly higher for the inulin and control boar group compared to the lupin group.     

Effect of feeding fermentable fibre-rich feedstuffs on meat quality with emphasis on chemical and sensory boar taint in entire male and female pigs

Skatole, androstenone and other compounds such as indole cause boar taint in entire male pork. However, female pigs also produce skatole and indole. The purpose of this experiment was to minimise boar taint and increase overall impression of sensory quality by feeding entire male and female pigs with fibre-rich feedstuffs. The pigs have been fed three organic diets for either 1 or 2 weeks prior to slaughter of which two diets contained different fermentable fibre-rich feedstuffs – 10–13.3% dried chicory roots or 25% blue lupines. These two treatments were compared with pigs fed with an organic control diet for either 1 or 2 weeks prior to slaughter. Lupines significantly reduced skatole in blood and backfat for both genders after 1 week. Moreover, lupines showed negative impact on growth rate and feed conversion whilst chicory showed no significant differences in this respect. However, the indole concentration was significantly lower in chicory than lupine fed pigs. From a sensory perspective, chicory and lupine feeding reduced boar taint since odour and flavour of manure related to skatole and urine associated to androstenone were minimised. The level of boar taint in the entire male pigs was most effectively reduced after 14 days by both fibre-rich feeds while lupine had the largest influence on “boar” taint reduction in female pigs.     

Determination of the boar taint compound skatole in meat juice by means of stable isotope dilution analysis-direct immersion-solid phase microextraction-gas chromatography/mass spectrometry

A novel SIDA-DI-SPME-GC/MS procedure for the quantitation of skatole in pork meat juice was developed and validated as a substitute for back fat sample analysis. System suitability was evaluated by determining the correlation between skatole concentrations in a subset of 38 paired meat juice and back fat samples selected from 90 fattened boars. High correlation was observed between both matrices and conclusions about the partitioning of skatole as well as of androstenone between fat and lean compartments in vivo were drawn.     

A single nucleotide polymorphism in the CYP2E1 gene promoter affects skatole content in backfat of boars of two commercial Duroc-sired crossbred populations

The prevention of unpleasant boar taint is the main reason for castration of male piglets. This study aimed to investigate how the malodorous compound skatole is affected by a single nucleotide polymorphism (g.2412 C>T at -586 ATG) in the porcine cytochrome p450 II E1 (CYP2E1) gene. 119 boars of two commercial Duroc-sired crossbred populations raised at different farms were investigated. Skatole and androstenone in backfat averaged 114±125ng/g and 1206±895ng/g melted fat, respectively. The frequency of the genotypes CC, CT, and TT was 25, 52, and 23%, respectively. CC boars had the highest average skatole levels (175ng/g) compared to CT (92ng/g) and TT (93ng/g). Applying suggested sensory threshold levels for skatole (>150ng/g) and androstenone (>2000ng/g), 30% of the carcasses may be unacceptably tainted while the proportion of tainted carcasses is significantly higher within genotype CC (56.7%) compared to genotypes CT (24.3%) and TT (14.8%). Effective reduction of tainted carcasses appears feasible applying marker assisted selection.