Vous êtes sur la page 1sur 131

University of Alberta, Swine Research & Technology Centre University of Minnesota, College of Veterinary Medicine

Leman Pre-Conference Reproduction Workshop Saturday, September 17, 2005

Effective Management of Replacement Gilts

Workshop Sponsor: Intervet Inc.

Proceedings index

Introductory comments (George Foxcroft)

Part 1: Establishing the principles. The biological basis for implementing effective replacement gilt management (George Foxcroft) The Non-Negotiables of managing an effective Gilt Development Unit (GDU) (Joaquin Sprke) Health status and biosecurity as essential components of GDU systems in the North American industry (Laura Batista)

Part 2: Putting principles into practice. Designing effective boar stimulation systems as a critical feature of the GDU (Eduardo Beltranena) Use of pharmacological interventions to further improve GDU efficiencies (George Foxcroft) Capturing gilt pool information (Brad Thacker)

Part 3: Proof of principle. Experience with the Pipestone Project (Joaquin Sprke) Experience using MATRIX and P.G. 600 in a commercial swine operation - impact on gilt pool size (Matt Ackerman)

Introductory Comments
The phenotypes of commercial dam-line sows reflect 20 years of selection for lean growth performance in their terminal line offspring, in parallel with less intense selection for females with high fertility and good mothering ability. The balance between these selection pressures results in some characteristic differences between dam-lines that need to be appreciated: However, the failure to achieve top 10 % performance in most breeding herds (30 pigs per sow per year being a popular benchmark) is due to inadequate management and a failure to apply a number of basic production practices. Improving the efficiency of gilt pool management and achieving a consistent flow of eligible and select gilts for entry to the breeding herd is considered to the essential driver of good breeding herd performance. As the first of a series of collaborative Reproduction Workshops, George Foxcroft (Leader of the Swine Reproduction-Development Program University of Alberta) will facilitate a workshop directed at consultant veterinarians, breeding stock suppliers, and breeding herd managers that will set the benchmarks for improved replacement gilt management as the essential driver of breeding herd performance.

Part 1: Establishing the principles.

The biological basis for implementing effective replacement gilt management

George Foxcroft1,2, Eduardo Beltranena2, Jenny Patterson2 and NoelWilliams3. Canada Research Chair in Swine Reproductive Physiology, Swine Reproduction-Development Program, Swine Research & Technology Centre, University of Alberta, Edmonton, Alberta, T6G 2P5, Canada; 3PIC North America, 3033 Nashville Rd, Box 348, Franklin, KY 42134 USA E-mail: george.foxcroft@ualberta.ca
2 1

INTRODUCTION Further improvements in gilt development programs can lead to major increases in breeding herd efficiency. Whether gilts are reared in in-house gilt multiplication systems, or supplied by a breeding company, proper selection and management of replacement gilts in purpose-built Gilt Development Units (GDUs) has a proven impact on lifetime non-productive days (NPD). However, the full economic benefit of improved GDU management goes beyond a simple reduction in NPD. The trend towards larger breeding sow herds seems to be decreasing the efficiency of breeding herd management. PigChamp data for 2002 showed that on larger breeding sow farms in the USA and Canada, annual herd replacement rates were often between 60 and 70 %, with a number of important consequences. A larger pool of replacement gilts is needed to meet increased replacement requirements Breeding herd parity distribution is unstable and biased towards lower parity females Chronic over-crowding of pens in the gilt development area is needed to meet replacement needs Negative impacts on health and welfare result Pressure to meet breeding targets results in less fertile gilts being bred using pharmacological interventions Gilts are bred below target weights General performance and morale of GDU staff declines and staff retention is low To reverse these trends, effective gilt management programs are urgently needed that will meet replacement targets from a smaller pool of gilts, and with gilts with improved lifetime breeding performance. This will ultimately reduce annual replacement rates (target for top 30% of breeding herds should be <50%), improve sow fitness, decrease sow death losses, and increase labor efficiency and space utilization. Cost-benefit analysis strongly favors improved efficiencies within the GDU, but implementation of effective GDU programs is still not widespread in the pig industry. Therefore, we will again review the key elements of good GDU management. By making gilt management more efficient, we improve both the utilization of space and labor, and actually achieve a flow of eligible (service-ready) gilts within the design specifications of the gilt facility. The main purpose of this review is 1) to define the phenotypic characteristics of 5

commercial dam-line gilts and sows, as the basis for developing appropriate recommendations for improved gilt management, and 2) to discuss the type of management strategies that allow effective implementation of effective Gilt Development Units (GDUs) as an essential driver of breeding herd productivity. DEFINING MEASURABLE AND MEANINGFUL INDICATORS OF BREEDING HERD PERFORMANCE When measuring the success of management programs within the GDU and mature sow herd, the key indicators of breeding herd performance need to be carefully defined, and should reflect the most meaningful measures in terms of overall economic performance. Our producers are trying to make money, and should not be encouraged to see a simplistic measure of productivity like maximal numbers of pigs produced, at any cost, as a worthwhile goal. If a production system is not fully integrated, the terms of contracts at each level of the production chain should reflect the value of the pigs produced. Indeed, recognition of the need for a correct balance between the quality and the number of weaned pigs produced is not always apparent in the contracts agreed. Consequently, this is often also not reflected in the priorities given to improving breeding herd performance. In terms of producing a reliable supply of weaned pigs at the critical nursery stage of production, the most important breeding herd key performance indicators (KPI) are probably; 1) uniform numbers of pigs weaned per week, 2) the weight and age of the pigs weaned, and 3) the least variation possible in age and weight at weaning. In turn, if properly rewarded, these KPIs determine the key factors that will be the focus of the breeding herd. As has been repeatedly emphasized in the assessment of key determinants of the number of pigs born and weaned per week, the single biggest factor needing attention is meeting breeding targets, with the second largest risk factor being farrowing rate. As shown in Table 1, these factors far outweigh the impact of achieving overall increases in the number of pigs born/litter, or variations in pre-weaning mortality. Thus, the primary focus of the breeding herd should be identifying the gilts and sows available on a weekly basis to meet projected breeding targets and to improving the breeding management of these gilts and sows. Table 1. The relative importance (%) of different components of breeding herd efficiency for achieving a uniform weaned pig flow to the nursery (after Dial, 2002). % Number of sows served Farrowing rate 60 30 5 5

Number born alive per litter

Mortality of pigs born alive

In the push concept of breeding herd management, a focus on establishing a well managed Gilt Development Unit (GDU) ensures a constant supply of gilts per week, and at the same time improves breeding management within the GDU. A constant input of high quality gilts into the

breeding herd, with increased longevity, in turn stabilizes the parity structure of the breeding herd. This helps in preventing the somewhat erratic contribution that weaned sows are often seen to make to weekly breeding targets. A constant input of select gilts to the breeding herd also prevents the tendency for a reduction in the voluntary culling of sows to achieve weekly breeding targets. All these factors will prevent breeding farms from entering the death spiral that is frequently seen in many of our larger production systems (Williams et al., 2005). The key management practices that will best serve our industry, and the characteristics of the dam-lines that best suit these practices, will be the focus of the rest of this discussion. UNDERSTANDING THE PHENOTYPE OF COMMERCIAL DAM-LINE SOWS 1) Changes in lean tissue growth rates in dam-line females. The major changes in the lean growth potential, and associated changes in the overall tissue metabolism of contemporary damline sows is not adequately recognized. Compared to selection for reproductive merit, the much greater heritability of growth traits has resulted in improvements lean growth performance in terminal line pigs that is the very basis of a competitive pork production industry in world meat markets. Inevitably, existing dam-lines carry these same traits to a greater or lesser degree. In the major dam-lines used in contemporary pork production in North America, inadequate attention is paid to the changes in basic sow metabolism resulting from this increased potential for lean tissue deposition and an associated lack of fatness in our current dam-lines. Traditional management practices that were established even 20 years ago need to be re-evaluated, if we are to capture the full economic potential of the modern breeding sow and her offspring, in terms of greatly improved nutrient utilization. In terms of the threshold growth rates needed to sure that growth rate per se is not limiting the onset of sexual maturation in the gilt, the earlier data of Beltranena et al. (1991) suggested that only when growth rate was below 0.55kg/day from birth to onset of boar stimulation at 160 days of age, was there any delay in onset of pubertal estrus (Figure 1). The more recent data presented in Figure 2, from a study of Genex grandparent females and their F1 progeny at the University of Alberta, support these conclusions. This leads to the generalization that with unrestricted feeding during the grow/finish phase, and recommended space allocations to gilts during development, it is unlikely that growth rate in commercial dam-line gilts will limit age at the onset of first estrus. Furthermore, the data in Figure 2 emphasize that age at first estrus is very largely dependent on the age at which effective stimulation with boar pheromones and direct boar contact is applied. Recent comments that pubertal estrus is occurring at older ages in todays commercial dam-lines seems to us to have little substance, unless of course boar stimulation is delayed. The true distribution of age at first estrus is clearly evident when one re-plots the data in Figure 2 for the F1 gilts that were first exposed to boar contact at 135 days of age, as in Figure 3. Age at pubertal estrus is now seen to almost normally distributed, with some gilts reaching puberty within days of first boar contact, whilst other gilts may only not show pubertal estrus after 50 days of continuous boar contact. However, the data in Figure 3 seem to support the curvilinear best fit to the data shown in Figure 1, suggesting a tendency for the highest growth rates to

Figure 1. Relationship between growth rate and age at pubertal estrus in gilts first stimulated with boars at 140 days of age. (After Beltranena et al., 1991). Feed restriction in gilts achieving a lower growth rate was associated with a delay in the onset of puberty.

Age at puberty, days




100 0.25 0.35 0.45 0.55 0.65 0.75 0.85 Growth rate from birth to puberty, kg/d

Figure 2. Effect of puberty stimulation in the gilt commencing either at 160 d (Closed squares; ) or 135 d (Open diamonds; ) of age. Both sets of data indicate that the highest growth rates achieved by feeding gilts ad libitum with diets aimed to maximize lean growth potential may result in a delay in the onset of first estrus. (Data from Patterson, 2001).
day 135 day 160

215 205 Age at puberty (d) 195 185 175 165 155 145 135 125 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9

Lifetime growth rate (kg/d)

be associated with a marginal delay in pubertal estrus. This may be problematic, in that late maturing and fast growing gilts may become overweight by the time they are bred, and as discussed later, this is one of the major risk factors for poor retention in the breeding herd (Williams et al., 2005). Figure 3. Normal distribution of age at first recorded heat (pubertal estrus) in Genex F1 gilts provided good boar contact from 135 days of age. (Data from Patterson, 2002)

50 45 40 35 30 25 20 15 10 5 0 130 136 142 148 154 160 166 172 178 184 190 196 Age at Puberty (d)

Finally, data from another dam-line genotype shown in Figure 4 reinforces the view that growth performance of most commercial dam-line gilts is unlikely to place any constraint on the age at pubertal estrus, and that pubertal estrus can still be induced at a relatively early age with good boar stimulation. The data in Figure 4 show comparable data from a gilt re-population study conducted in collaboration with the Prairie Swine Centre Inc., involving PIC Camborough 22 gilts, provided good boar contact from a pen average of 140 days. These data also serve to demonstrate the total lack of any relationship between growth rate and the population of gilts that did, or did not, have a recorded pubertal estrus within 40 days of commencing boar stimulation. Finally, because of the considerable variability in growth performance within a group of gilts, it is wrong for producers to assume that some arbitrary age will effectively define the physical development of gilts at stimulation or breeding. Gilt pool managers seem to ignore the enormous variation in growth rate among groups of gilts, and also the rather uncertain relationship between weight and back-fat.


Figure 4. Relationship between growth rate and age at puberty in response to daily boar stimulation from 140 days of age (open diamonds). Gilts not recorded in estrus by 180 days were designated Non-Responders (closed diamonds). (Prairie Swine Centre and University of Alberta, Swine Technology and Research Centre, unpublished data, 2003).

200 190 180 170 160 150 140 130 120 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 Growth Rate (kg/d) (birth to d100 of age)

The extremes that were encountered at the time of first recorded estrus in the gilts that produced the data shown in Figure 3, are shown in Figure 5. The extremes of age at sexual maturity and growth rate, result in gilts with induced first estrus soon after commencing stimulation with mature boars at around 130 days of age but only a threshold growth rate of 0.64 kg/d (Fig. 5a), compared with gilts with first recorded estrus at 189 days of age and the highest growth rates recorded at 0.80 kg/d (Fig. 5b). These extremes result in first estrus gilts differing in body weight by as much as 73 kg. In terms of gilt conditioning for physical fitness and longevity in the breeding herd, early maturing/slower growing gilts would need to be provided with high energy "fattening" diets to achieve 135 kg body weight and at least 18 mm of back-fat at breeding. In contrast, late maturing/fast growing gilts probably need to be subjected to restrict feeding during development to prevent excessive growth being a cause of lameness and eventual culling. The unavoidable conclusion from these data is that age is not a good measure of weight or fatness, and the only way to be certain that gilts are at target weight for breeding is to weigh them! As age at sexual maturity can also vary from 130 to over 200 days, it is impossible to set some arbitrary age and assume that this defines some general level of sexual maturity. Clearly, by starting to assess whether gilts will show a standing heat in response to boar contact at over 200 days of age, there is little opportunity to determine the relative sexual maturity of individual females. The only benefit from introducing boar contact at such a late stage is the very short period over which pubertal estrus will be observed. However, very efficient boar stimulation programs can involve relatively little labor input per gilt bred, and yet increase the lifetime performance of truly select gilts substantially.

Age at puberty or removal


Figure 5. An illustration of the extremes of growth performance within a contemporary group of commercial dam-line gilts provided with direct boar contact from 135 days of age to stimulate onset of first estrus. (Data from Patterson, 2001) (a)
Gilt # 114 160 140 Weight (kg) 120 100 80 60 40 80 90 100 110 120 130 140 150 160 170 180 190 200 Age (d) Gilt # 18 Puberty = 189 d Weight = 130kg Backfat = 17.4 Puberty = 189 d Weight = 155 kg Backfat = 34.8 mm


160 140 Weight (kg) 120 100 80 60 40 80 90 100 110 120 130 140 150 160 170 180 190 200 Age (d)
Gilt # 155 Puberty = 131 d Weight = 82 kg Backfat = 12.6 mm
Gilt # 99 Puberty = 133 d Weight = 86 kg Backfat = 11.3 mm

The lack of any reliable association between age, and onset of sexual maturity or body weight, implies that these essential benchmarks must be assessed independently in a well-managed GDU and used to allocate gilts to appropriate breeding groups. The aim should be to have gilts as sexually mature as possible before target breeding weight is reached, with the minimal requirement that breeding occurs at least at second estrus.

Interestingly, the earlier studies of Beltranena et al. (1993) already indicated that the fatness of the gilt was unrelated to the rate of sexual maturity, and this conclusion has also been supported in subsequent experiments. Moreover, in most gilt pools, there is usually a very weak association between weight and measured back-fat, as shown in Figures 6 and 7. From a management perspective, simply relying on an increase in overall body weight to produce a predictable change in back-fat in all gilts, or to assume that some arbitrary age will be associated with


target levels of back-fat in all gilts, is unrealistic. Equally, back-fat can in no way be interpreted as indicating the likely body weight of a replacement gilt.

Figure 6. Associated changes in sow body weight and back-fat in (a) Camborough 22 and (b) Genex gilts between breeding and farrowing. Dashed lines indicate average weight and back-fat at each time. (Unpublished data, University of Alberta, 2005) (a)
30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0 75 85 95 105 115 125 135 145 155 165 175 185 195 205 215 225 235 245

Backfat (mm)

Weight (kg)

30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0 75 85 95 105 115 125 135 145 155 165 175 185 195 205 215 225 235 245

Backfat (mm)

Weight (kg)


CONSIDERATION OF DIFFERENCES AMONG DAM-LINES (GENOTYPES) Breeding companies producing a wide variety of high index dam-line gilts, have generally failed to provide adequate data about the key performance characteristics of even major commercially used dam-lines. Based on recent collaborative studies with two major commercial dam-lines, we conclude that the phenotype of the gilts and first parity sows clearly reflects the extent to which selection for increased lean tissue gain is reflected in these terminal dam-lines. As can be seen in Figure 6, the level of fatness (back-fat measured at the P2 position in both cases) during gilt development tends to be different. Furthermore a maternal weight gain of 50 kg from breeding to farrowing results in a very different response in back-fat gain. The critical question then becomes, to what extent is this relative leanness of the terminal damline likely to affect lifetime productivity of the sow? From existing data, it is hard to suggest that there are any inherent differences in lifetime reproductive performance that can be ascribed to the relative fatness of the sows per se (Williams et al., 2005). The lack of a consistent relationship between overall sow body weight and back-fat thickness is also seen in data collected over three parities from the gilts shown in Figure 6a. The changes in sow body weight and back-fat over three successive parities, for those sows that were available to record data on each occasion are shown in Figure 7a and 7b, respectively. As can be seen, because gilts were bred by design at third estrus, and there was a lack of any relationship between body weight and rate of sexual maturity once the critical threshold has been passed, this resulted in a wide range of body weights at breeding at immediately after these gilts farrowed their first litter. In general, the pattern of increase in lean body mass over successive parities would meet most conventional targets (Figure 7a), and the changes in measured back-fat were variable and lower that would be suggested as ideal even for the Camborough sow (Fig. 7b). However, as discussed later in these proceedings, the lower than targeted levels of back-fat do not seem to be critical for sow longevity in the breeding herd, or for sow lifetime productivity. Another notable feature of the data shown in Figure 7a is the persistent difference in sow body weight over three parities, despite the fact that management practices in this herd would allow feed intake in gestation to vary with respect to perceived weight and body condition of the sows after breeding. This emphasizes the need to focus on entering gilts into the breeding herd at known and recorded weights, as probably the only reliable way of insuring that lifetime changes in sow weight will be consistent with longevity and good lifetime productivity. In terms of the phenotypic characteristics of contemporary commercial dam-lines, therefore, implementation of effective gilt pool management strategies will allow producers to meet targets for body condition (weight, back fat) and physiological maturity (age at first estrus, and estrus at breeding), to reduce annual replacement rates (target for top 30% of breeding herds should be <50%) by improving sow fitness and reducing sow death losses, and increase labor efficiency and space utilization. A comprehensive understanding of the characteristics of specific damlines is however, essential, when developing the optimized breeding management programs of the future.


Figure 7. Mean body weights of gilts bred at third estrus, regardless of body weight, with the data representing the body weight (a) and back-fat (b), respectively, for the lowest, the middle and the highest 10% of gilt weights when first recorded as heat-no-serve after introducing direct contact with boars at 140 days of age when first bred. (a)
290 270 250 230 Weight (kg) 210 190 170 150 130 110 90 hns BRD F1 Time F2 F3 Light Medium Heavy

18 17 16 Backfat (mm) 15 14 13 12 11 10 SHBF BRDBF F1BF Time F2BF F3BF Heavy Light Medium

The next section of this introductory paper will consider other biological characteristics of sexual maturity and relative fertility in gilts, and how these can also be taken into account to improve overall breeding herd performance All these advantages can ultimately be achieved whilst maintaining the economic efficiencies of smaller, well managed, gilt pools.


UNDERSTANDING THE BASIS OF EFFECTIVE GILT MANAGEMENT PROGRAMS. Taking into consideration what we know about gilt development, and incorporating the above key points, we suggest the in-house gilt management system shown in Figure 8. Identifying select gilts at an early age is a critical part of a successful gilt development program. This selection process will involve three steps. Figure 8. Schematic diagram of an efficient gilt management system.

Pre-Select 1
Pre-Select 1 occurs at the time the gilts leave the nursery. At this time gilts must have good conformation, 12-14 teats and be free of hernias or ruptures. As more data become available, it may also be appropriate to exclude gilts with inadequate growth rate at this stage.

After gilts leave the nursery an opportunity exists to condition gilts to achieve adequate weights and body condition at puberty to sustain lifetime performance. As discussed above, available data consistently show that at commercially acceptable growth rates (0.55 0.80 kg/d) (birth to 100 days of age), there is no relationship between growth rate and age at puberty (Figures 1 through 4). Experience in commercial practice suggests that specific gilt "conditioning" diets can be used to increase general physical fitness of gilts, particularly in some of the leaner dam-lines. In studies in which we attempted to slow growth in gilts with high fiber diets from 50 kg until puberty induction (Patterson et al., 2002a), we had very little impact on bodyweight at first estrus.


Pre-Select 2 Pre-Select 2 will occur at 140 days of age. Ideally, gilts will be assessed for weight, growth rate and backfat depth at this time. Gilts must achieve a lifetime growth rate of at least 0.6 kg/d. Figure 9 illustrates why it is important to remove gilts with low growth rates. Figure 9. Actual weight and growth rate at 140 days of age versus recorded age at puberty in response to daily direct boar contact from 140 days, and predicted estrus at breeding to achieve a minimum target breeding weight of 135 kg Actual predicted breeding weight (kg) is shown within each of the cells. (Prairie Swine Centre/University of Alberta, Swine Research and Technology Centre, unpublished data 2003).

Age at Puberty

165d 135 137 137 135 145 140 149 140 149

170d 138 140 140 138 148 143 136 145 153

175d 140 142 143 141 137 147 140 149 158

180d 143 145 146 144 141 135 144 153 162

185d 145 136 136 148 144 139 148 157 167

190d 148 139 139 137 148 143 152 162 171

0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90


70 77 84 91 98 105 112 119 126

143 146 146 145 141 136 145 136 144


1st estrus 2nd estrus 3rd estrus 4th estrus 5th estrus 6th estrus 7th estrus

In this example, a slow growing (< 0.6kg/d) but early maturing gilt (first estrus at 160 days) would weigh approximately 96 kg at first estrus. If this gilt was bred in the appropriate weight range (135 150 kg body weight), she would need to be bred at 4th or 5th estrus and would accumulate nearly 84 days in the gilt stimulation/pre-breeding area. The lack of inherent growth performance, the accumulated NPD involved, and the real risk of such gilts being bred below


target weight when breeding targets are not being met, all suggest that exclusion of lower growth rate gilts is the preferred management option. Even a slow growing (<0.6 kg/d) and late maturing (190 days) gilt would accumulate 30 days in stimulation and an additional 42 days to reach minimum breeding weight. Therefore, at Pre-Select 2, gilts not achieving a growth rate of 0.6 kg/d at 140 days of age would not be permitted to enter the stimulation phase. In a study conducted at the University of Alberta, 13% of 228 gilts would have been culled because they did not meet the minimal growth criteria. This percentage may be higher if gilts are subjected to vaccination programs for PRRS and other diseases. In this case it may be necessary to adjust our benchmarks for entry-to-service interval to acknowledge the reality of the situation. At Pre-Select 2 gilts will be further examined to ensure that all gilts have good conformation, locomotion, 12-14 teats and are still free of hernias, ruptures and other ailments. Again, conformation data obtained at Pre-Select 2 can be used to set up gilts on fattening diets if needed. The number of gilts required to enter the stimulation phase will depend on the breeding requirements of the herd. In a trial recently completed at Prairie Swine Centre, the results indicated that approximately 125% of breeding gilt requirements should enter the stimulation phase (expecting 22% not to cycle and 3% to be culled) to obtain the required number of gilts that are naturally cyclic within 40d. If the target number of gilts needed to enter the gilt pool cannot be met with gilts that meet minimal growth targets at Pre-Select 2, an appropriate number of Non-Select gilts can enter the puberty induction phase, as a last resort, accepting that these gilts will either tend to be bred below target breeding weight, or will accumulate excessive NPD before breeding.

Final Selection Puberty Induction The age to begin puberty stimulation will depend on a number of factors. Generally, as illustrated in Figure 10, a younger age at stimulation corresponds to a decreased age at puberty, but requires more days in stimulation; conversely, older gilts at stimulation are typically older at puberty, but require fewer days of stimulation.

However, stimulating gilts at an earlier age (as shown in Figure 11) has several benefits. Stimulation at a young age enables the producer to identify gilts that are most sexually mature (Foxcroft and Aherne, 2001) Stimulating gilts early would permit a producer to cull non-cycling gilts as market animals, reducing the number of gilt NPD and the financial cost to the producer Gilts can be better managed to achieve a target weight (135 150 kg) and body condition Early stimulation allows a producer to synchronize estrus in gilts using products like Regumate (Matrix in the USA) and thus meet breeding requirements from a smaller pool of select (service eligible) gilts


Finally, early stimulation of gilts permits producers to take advantage of the increased productivity of gilts bred at second or third estrus
It is important to understand that stimulation of early onset of puberty does not mean that these gilts have to be bred at first estrus, or at a light weight.

Figure 10. Relationship between age at stimulation, days to first estrus, and age at first estrus. (Levis, 2000)

60 Days to First Estrus 50 40 30 20 10 0 110

210 200 Age at First Estrus 190 180 170 160 150 190








Age at Stimulation

Figure 11. Number of gilts per day showing pubertal estrus after stimulation with direct boar contact from approximately 140 days of age. 112 gilts out of 509 (22%) did not exhibit first estrus. (Prairie Swine Centre/U of A SRTC, 2003)
40 35 Number of Gilts 30 25 20 15 10 5 0 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 Days to First Estrus

20 days stimulation = 53% Pubertal 30 days stimulation = 77% Pubertal 40 days stimulation = 96% Pubertal


Historically, age at puberty has been shown to be normally distributed when growth rate is not limiting. The full extent of this variation in age at first estrus is most apparent if gilts are exposed to mature boars at an early age (say 140 days as in the study presented in Figure 11). As previously mentioned, puberty induction at an early age serves to identify the precocious animals. In a recent experiment, out of 508 gilts stimulated with direct daily boar contact from 140d of age, 77% of gilts were pubertal within 30 days of stimulation. By comparison, when stimulation is delayed to at least 160 days (Figure 12), 84% of gilts had a recorded first estrus by the same 30-day cut-off. If a large proportion of gilts are required to reach a synchronous puberty, commencing boar exposure at an older age is desirable (Levis, 2000). This is also probably most efficient in terms of labor and space utilization increasing eligible gilts/pen space/day. Figure 12. Number of gilts per day showing pubertal estrus after stimulation with daily direct boar contact from greater than 160 days of age (University of Alberta Swine Research & Technology Centre, unpublished data)

120 100 Number of Gilts 80 60 40 20 0 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 Days to First Estrus

20 days stimulation = 67% Pubertal 30 days stimulation = 84% Pubertal 40 days stimulation = 93% Pubertal

It is becoming increasingly important to identify the 75 to 80% of gilts that respond earlier to boar stimuli, because there are sound biological reasons, and increasing amounts of production data, to support the suggestion that later- maturing gilts will have reduced lifetime fertility. Data from a gilt development study conducted at the Prairie Swine Centre, Saskatoon, is examining the relationship between age at puberty and lifetime performance in Camborough 22 and L42 gilts. The gilts were housed in groups of twenty and received 20 min direct daily exposure to an epididimectomized boar, starting at 140.0 4.7 d of age. Gilts attaining puberty by 180d of age were deemed to be select gilts and classified as Early (EP), Intermediate (IP) and Late (LP) with respect to age at first estrus. Gilts were deemed to be Non-select (NP) if first estrus was 19

not shown by 180 d of age. Select gilts were bred at third estrus, regardless of age or weight. Non-select gilts were added to the gilt pool by production staff using available techniques (i.e. mixing, additional boar contact and treatment with PG 600) after entry to the sow farm. To determine sow lifetime performance, data on sow body weight, loin and backfat depth at farrowing and weaning, total litter size born alive, dead and mummies, weaning-to-estrus interval and reason for culling were collected over three parities. Some initial results are presented in Figure 13. Figure 13. Breeding, pregnancy, farrowing, weaning, and rebreeding rate over three parities as a percentage of gilts originally on inventory (Prairie Swine Centre and University of Alberta, Swine Technology and Research Centre, unpublished data, 2004)

100 90 80 70 60 50 40 30 20
2 Br ee d 1 1 3 3 2 R eb re ed R eb re ed R eb re ed w w w Fa rr o Fa rr o Fa rr o Fa rr o

Retention Rate (%)





As a percentage of the total number of gilts on inventory at the start of stimulation in each group, fewer Non-Select gilts were bred than any of the classes of Select gilts. Consequently for NP gilts, pregnancy rate, farrowing rate, weaning rate and the percent rebred after weaning after first parity (expressed as a % of gilts originally on inventory) were lower than for EP, IP or LP gilts (Figure 13). Furthermore, considering only those gilts successfully weaned as parity 1 sows, class of gilt affected (P < 0.02) the percentage of animals pregnant as parity 2 sows (EP: 94.2; IP: 87.2; LP: 91.0; and NP: 76.6 %). Similarly, breeding herd efficiencies (Non-Productive Days/pig born) declined as age at puberty increased, when gilts were bred at third estrus irrespective of weight or age (Table 2). Taken together, these data lead to the obvious suggestion that response to a standardized protocol of boar stimulation can be used to identify the 75-80% of gilts that are likely to be most fertile. As illustrated in Figure 8, to meet breeding targets, or in start-up situations, it may be necessary to retain Non-Select gilts as part of the breeding herd. However, retention of Non-Select gilts within the herd would;

Incur costs of unknown numbers of additional NPD Represent less efficient use of pen space within the gilt pool Still not guarantee that gilts would eventually cycle It is also important to emphasize that even if these gilts are bred, their expected fertility would be low. It may be good management practice to already designate these Non-Select gilts at parity 1 culls, if they are included in the herd to meet initial breeding targets.
Taking these factors into account, and considering expected cost-benefits of efficient use of space and time, we recommend that the puberty induction phase begins when gilts reach 160 days of age and continue until they exhibit their first estrus or until 190 days of age, whichever comes first.

FURTHER REFINEMENTS TO STANDARDIZE BREEDING WEIGHT OF GILTS The results of the study at Prairie Swine Centre for which data are shown in Figure 9, indicate that early exposure (135 - 140 days of age) of gilts to boars resulted in a large variation in weights and ages at puberty, ranging from 75.8 to 151.4 kg, and 132 to 190 d, respectively. Because all gilts were bred at third estrus in this study, this variation in weight at puberty resulted in weights at breeding ranging from approximately 100 to 190 kg. These large ranges present several problems to the producer. Gilts that are heavy at breeding incur increased lifetime feed costs for maintenance and may cause welfare problems because of potentially larger physical size as mature sows Conversely, gilts that are lightweight at breeding may lack the necessary body reserves to sustain body condition through several parities Recent studies at the University of Alberta, and elsewhere, suggest that a minimum body weight after farrowing of 175-180 kg may be necessary to protect against excessive loss of protein mass during the first lactation (Quesnel and Prunier, 2003; Clowes et al., 2003a,b). As suggested by Foxcroft (2002), a body weight of 135-140kg at breeding, assuming a 35-40 kg weight gain during the first gestation, would theoretically result in body weight after farrowing being 175 kg or greater. Development and implementation of gilt management strategies that ensure that all gilts achieve adequate body tissue reserves at farrowing are necessary. To overcome the problems associated with large variations in weight, a stricter selection program should be implemented, stipulating that all gilts weigh between 135 150 kg at breeding. As shown in Figure 9, if gilts from this study had been bred according to a desired target weight of 135-150kg, they would have bred at their 1st through 7th estrus. However, 1) if during Pre-Select 1 and Pre-Select 2 the slowest growing gilts were already culled, and 2) an upper limit of 3rd estrus for breeding was stipulated, the number of non-productive days would be dramatically reduced. Using these tighter selection criteria, and as shown in Figure 14, it was predicted from


the data available that 10, 32 and 58% of gilts would be bred at their 1st, 2nd and 3rd estrus, respectively, to meet the target weights at breeding. Figure 14. Distribution of predicted estrus that gilts would be bred to achieve a target body weight of 135-150kg at breeding using stricter criteria for minimal growth performance of > 0.0kg/day at boar stimulation. (Prairie Swine Centre, and University of Alberta, Swine Research and Technology Centre, unpublished data 2004)

35% 30%

25% 20% 15% 10% 5% 0% 1 2 3 4 5 6


COST-BENEFIT ANALYSIS OF GILT MANAGEMENT PROGRAMS The next real challenge is to develop comprehensive economic analyses of different gilt replacement programs to demonstrate the net cost/benefits of particular systems. One component of this analysis will be to provide a cost benefit analysis of the capital costs of retrofitting existing facilities or including additional costs in GDU designs against the increase in labor efficiency and use of space. Although we have not reached this point in our own program, we believe that many concerns about extra labor requirements to implement more efficient GDU protocols are unfounded. This might be illustrated briefly as follows; Basic requirements of good estrus induction programs are well defined in research literature and relatively simple to understand. Practical implementation requires effective interaction between high libido boars with pre-pubertal gilts that have achieved a minimum growth performance at the time of stimulation. Adequate interaction requires attention to the boar:gilt ratio and the amount of contact time allowed. In an ideal system, these programs can still result in around 85% of gilts being recorded in pubertal estrus within three or four weeks of initial boar contact. This system also triggers a management decision about the fate of the 15 to 20% of non-responders. Based on several recent studies and assuming around 20 minutes per day of boar stimulation on a pen basis, it


will be possible to identify thre manpower needed to run this intensive-type GDU system. The critical next step is to compare the economic inputs v. outputs of such systems against equally well audited but less labor-intensive GDU systems, taking account of the same key performance indicators in each case. In situations where inadequate boar contact occurs, often with boars showing low libido and being managed by frustrated staff in overcrowded facilities, the success of the estrusinduction process can rapidly fall to 40% or less over the same three to four week period. In terms of the number of eligible gilts/pen space/per day, this constitutes a very inefficient use of the gilt facility. In labor terms this doubles the labor budget to around per eligible gilt. Presented in these terms, we start to appreciate the inherent futility of poor gilt management programs, with over 50% of the time spent in gilt stimulation being non-productive. The economic cost of such inefficiency is then further compounded by the need for extra pen space to house the non-responsive gilts beyond the 28-day stimulation period until they are eventually bred. Considered in terms of a 50% v. 85% farrowing rate with the same amount of time spent breeding sows over a four-week period, our inability to internalize the relative efficiency of gilt management programs is perhaps more obvious. Other more complex cost/benefit analyses are needed before the true value of specific gilt management programs will be clearly established. For example, we know from the literature that ovulation rate, and therefore, litter size increase with breeding estrus. If we supposed a 0.7 pig increase in litter size in gilts bred at 2nd rather than at 1st estrus, the extra cost of this 0.7 pig benefit will depend on the housing (capital depreciation) and feed costs and could vary from approximately $13 to $20 . Similarly, if we consider that the increase in litter size between the 2nd and 3rd estrus would be only 0.2 pigs, the extra cost of achieving this 0.2 pig benefit might be unacceptable in some housing situations. However, consideration should also be given to the potential economic benefit that increased weight and fatness resulting from delaying breeding estrus may have on sow lifetime productivity, offsetting the cost of improving first litter size by breeding at 2nd or 3rd estrus. Similarly, if early induction of 1st estrus can be an outcome of more efficient boar stimulation programs applied at a younger age, then the benefits of increased first litter size can be achieved without increased housing costs within the GDU. These and other variables will eventually be integrated into a full economic model to determine the relative cost benefits of GDU options. Ultimately, these economic models must include a full risk/benefit analysis that recognizes the impact of GDU management on overall production efficiency. Weekly variability and lack of uniformity (weight, age and health status) in weaned pigs moved to nurseries, is a key benchmark of breeding herd efficiency and is largely related to efficiency in meeting GDU breeding targets. Therefore, not only the relative costs of producing a bred gilt, but the economic value in meeting breeding targets must be assessed in determining the GDU as a critical cost centre within the overall production system.


CONCLUSIONS Our former colleague and valued mentor, Frank Aherne, provided the Swine Research & Technology Centre in Edmonton with its unofficial raison detreIn God we trust everything else requires data. We have some excellent data on factors that determine the efficiency of gilt development programs that reflect a longstanding commitment to improve our knowledge base through industry-relevant R & D programs. This R & D must now be extended to the creation of realistic economic models on which to base key management decisions about GDU systems and to demonstrate the value of improved GDU programs in large production systems. We have no doubt that such analyses will confirm the role of improved GDU management in efficient pork production systems. ACKNOWLEDGEMENTS We are grateful for the financial support of the funding partners supporting the research of the Swine Reproduction-Development Program at the Swine Research & Technology Centre, University of Alberta (Alberta Pork, AARI, NSERC, and Hypor Inc.), and in addition, the financial support of Sask Pork and the Agricultural Development Fund of Saskatchewan for the collaborative research based at the Prairie Swine Centre Inc. The support of Pipestone Veterinary Clinic, Intervet Inc. and PIC Technical Services (N. America) in conducting ongoing collaborative studies in commercial sow farms is also greatly appreciated. We also acknowledge those private and corporate sponsors that have supported the establishment of the Swine Research & Technology Centre as one of the premier swine research facilities worldwide. George Foxcroft presently holds a Canada Research Chair in Swine Reproductive Physiology. LITERATURE CITED AND SUGGESTED FURTHER READING Beltranena, E., Aherne, F.X., Foxcroft, G.R. and Kirkwood, R.N. 1991. Effects of pre- and postpubertal feeding on production traits at first and second estrus. J. Anim. Sci., 69, 886893. Beltranena, E., Foxcroft, G.R., Aherne, F.X. and Kirkwood, R.N. 1991. Endocrinology of nutritional flushing in gilts. Can. J. Anim. Sci., 71, 1063-1071. Beltranena, E., Aherne, F.X. and Foxcroft, G.R. 1993. Innate variability in sexual development irrespective of body fatness in gilts. J. Anim. Sci., 71, 471-480. Clowes, E.J., Aherne, F.X. and Foxcroft, G.R. 1994. Effect of delayed breeding on the endocrinology and fecundity of sows. J. Anim. Sci., 72, 283-291. Clowes, E.J., Aherne, F.X., Foxcroft, G.R. and Baracos, V.E. 2003a. Selective protein loss in lactating sows is associated with reduced litter growth and ovarian function. J. Anim. Sci., 81, 753-764. Clowes, E.J., Aherne, F.X., Schaefer, A.L., Foxcroft, G.R. and Baracos, V.E. 2003b. Parturition body size and body protein loss during lactation influence performance during lactation and ovarian function at weaning in first parity sows. J. Anim. Sci., 81, 1517-1528.


Foxcroft, G.R. 2002. Fine Tuning the Breeding Program. Saskatchewan Pork Industry Symposium 2002. Saskatoon, Saskatchewan. Foxcroft, G.R. 2004. Pharmacological considerations for optimizing reproductive efficiency. Proceedings of the 35th Annual Conference of the Amercian Association of Swine Veterinarians; proceedings of pre-conference symposium on Optimizing reproductive efficiency, pp 17-29. Foxcroft, G.R. and Aherne, F.X. 2001. Rethinking management of the replacement gilt. In: Advances in Pork Production, Ed. R.O. Ball, Dept. Agricultural, Food and Nutritional Science, University of Alberta, Volume 12, 197-210 Foxcroft, G., Ruiz-Sanchez, A., Town, S., Barry, J., Clowes, E., Willis, H., Beltranena, E., Pettitt, M. and Patterson, J. 2002. Fine tuning the breeding program. Proceedings of the Saskatchewan Pork Industry Symposium, 2002, pp49-61. Foxcroft, G., Patterson, J., Beltranena, E. and Pettitt, M. 2004. Identifying the true value of effective replacement gilt management. In: Proceedings of the Manitoba Swine Seminar, Volume 18, 35-51. Kirkwood, R.N., Aherne, F.X. and Foxcroft, G.R. 1998. Effect of PG600 at weaning on reproductive performance of primiparous sows. Swine Health Prod., 6,51-55. Levis, D. G. 2000. Housing and management aspects influencing gilt development and longevity: A Review. 2000 Allen D. Leman Swine Conference. Patterson, J.L. 2001. Factors influencing onset of puberty in gilts. M.Sc. Thesis, University of Alberta, Edmonton, Canada Patterson, J.L., Willis, H.J., Kirkwood, R.N. and Foxcroft, G.R. 2002. Impact of boar exposure on puberty attainment in gilts. Theriogenology, 57, 2015-2025. Patterson, J.L., Ball, R.O., Willis, H.J., Aherne, F.X. and Foxcroft, G.R. 2002. The effect of lean growth on puberty attainment in gilt. J. Anim. Sci., 80, 1299-1310. PigChamp. 2002 Breeding Herd Summary for Canada. http://www.pigchamp.com/2002Datashare.htm Quesnel, H. and A. Prunier. 2003. Endocrine mechanisms mediating nutritional effects on fertility in the gilt and sow. Proceedings of I Congreso Lactinoamericano de Nutricin Animal. Willis, H.J., Zak, L. and Foxcroft, G.R. 2003. Lactation length effects on sow endocrinology, folliculogenesis and in vitro and in vivo embryo development. J. Anim. Sci., 81, 20882102. Zak, L.J., Cosgrove, J.R., Aherne, F.X. and Foxcroft, G.R. 1997a. Pattern of feed intake, and associated metabolic and endocrine changes, differentially affect post-weaning fertility in the primiparous lactating sow. J. Anim. Sci., 75, 208-216. Zak, L.J., Xu, X., Hardin, R.T. and Foxcroft, G.R. 1997b. Impact of different patterns of feed intake during lactation in the primiparous sow on follicular development, oocyte maturation and embryo survival. J. Reprod. Fertil., 110, 99-106.


The Non-Negotiables of Effective Gilt Development Unit (GDU) Management

Joaqun Sprke R.1 and Eduardo Beltranena2

Asesoras Portec Ltda., Santiago, Chile; 2Swine Reproduction-Development Program, SRTC, University of Alberta, Edmonton, Alberta, T6G 2P5, Canada. E-mail: joaquin@asportec.cl

INTRODUCTION The North-American industry is facing further changes to update the industry and to become more competitive in a global market. For many years, one of the most challenging areas for the industry has been the performance and efficiency of the sow farms, resulting in sub-optimal number of weaned piglets /sow/year. In recent years, many producers realized that to be more competitive in this particular area, changes were necessary in many different aspects of sow farm management. It is well established that there is a relationship between litter size in parity 1 sows and lifetime productivity. Thats the reason why GDU management and gilt reproductive performance became a major objective for the industry. Different strategies have been developed, and most of them are based on the way gilts are flown into and managed in sow farms. Gilt Development Unit (GDU) management and implementation of basic principles are key factors in a good start of a female. Many different procedures can be implemented to achieve the objective, but here I want to focus on the most important concepts and management procedures, the ones that can be considered as Non-negotiable. THE FIRST STEP IS THE FIRST PRIORITY The most important factor is that the producers recognize the real importance of a good gilt development program and the huge impact it has on the entire sow farm productivity. Commitment from the different management levels in a production system hierarchy is a basic requirement to succeed in the implementation of a good GDU management program. GILTS AS THE ROW MATERIAL FOR THE BREEDING HERD The simple concept of establishing realistic targets for gilt flow is a very common reason for failure. Unrealistic replacement rate (too low, about 50%) and too high expectations from the multiplier herd are the most common factors that lead to less gilts in the pipeline. The consequences for the whole breeding herd are reduced breeding targets, breeding of marginal


sows, etc. Good planning is critical. Realistic replacement rate of a highly productive herd is between 50% and 55%. Budgeting less than 1.5 weaned gilts per bred gilt is too optimistic, leaving no room for rejection on the basis of physical and reproductive traits. Figure 1 shows the proportion of bred gilts for every 100 weaned gilts available for initial selection.

Targets for gilt flow

Weaned Phase Selects Into sow herd Bred 0 10 20 30 40 50 60 % retaine d 70 80 90 100

% of Gilts Retained

Figure 1: Gilt retention from weaning to breeding (J. Sprke, 2005)

FACILITIES -A BASIC NEED It is too obvious that a farm must have designated area(s) for the GDU. Unfortunately, too many times this is not the case. Sow farms do not have the designated space and/or facilities to implement an adequate GDU management program. This issue has become much more critical in the last years with the health challenges faced and the need to implement effective isolation and acclimatization programs. Ideally, the modern gilt flow should provide adequate space and facilities to accomplish gilt health protocols before they are introduced into the destination herd. However, depending on the frequency of gilt introductions and ages, the practical flow must be designed to allow simultaneous health acclimatization (conditioning) and also reproductive management associated with puberty induction. An ideal scenario is to have gilts already acclimatized and in the sow herd at 150 days of age. Many times this is not the case, and additional effort is needed to accomplish the objectives of health and reproduction conditioning of gilts before breeding in offsite facilities.


OBJECTIVES OF A GILT DEVELOPMENT PROGRAM The main objective is to have the right quantity of well-acclimatized gilts available for breeding, at the right weight and with at least one recorded estrus. GDU management has to be efficient in the use of space and labor, and also in keeping gilt costs under control. The general goals for the GDU are based on: Effective heat induction process - over 90% of gilts cycling within 4-6 weeks of puberty stimulation Allow gilts to grow well, to be at 300 lb or more at breeding Obtain an early detected estrus in order to get control over the gilt pool. Observing a heat-no-serve estrus allows us to manage the groups more efficiently based on expected week of breeding Reject gilts with low reproductive potential at the proper market weight THE SEVEN NON-NEGOTIABLE ASPECTS OF GDU MANAGEMENT 1. Acclimatization: Health conditioning of the gilt pool is part of the foundation of successful GDU management programs. Without stable and well acclimatized gilts, all the effort for reproductive conditioning will likely be worthless. 2. Trained staff: One of the commitments of the management is to provide one of the most important resources needed in GDU management, qualified staff. The staff designated to run the GDU has to be well trained in managing the whole gilt pool during puberty stimulation, with special focus on working with boars to induce and detect early pubertal estrus. 3. Age, bodyweight and back fat of gilts: Which is the right age, bodyweight and fatness of gilts at puberty and breeding has being forever discussed. Different approaches are taken around the world.
Age is more important as a factor determining the onset of puberty. Before 140d of age, the investment in labor related to boar exposure is probably not worth the reduced pubertal response observed. Between 140 and 180d, more gilts respond sooner the longer boar stimulation is delayed. Nonetheless, the proportion of gilts responding to boar stimulation within 30d is not different. Thereafter, there is no good evidence supporting age as an important factor determining the breeding time of gilts. Bodyweight at first breeding is likely a more important factor determining first parity gilt performance and influencing lifetime performance. Increasing breeding weight improves litter size at first litter, but there is a cost related to adding non productive days (NPD) as gilts most likely are bred at second or third estrus.


Recent data demonstrates the impact of weight, backfat and estrus at breeding on gilt and sow productivity for up to three parities. Results (based on experimental data and cost/benefit analysis) illustrate that gilts should be bred at a target weight of 135 150 kg. Gilts weighing less than 135 kg (300 lb) at breeding had less total pigs born over 3 parities than gilts weighing over 135 kg (Figure 2). Achieving the desired weight allows for proper body mass to be achieved at first farrowing (> 180kg) assuming a sow weight gain of 35 40 kg gain during first gestation.

34.0 Total Born 32.8 33.0 32.0 31.0 30.0 < 135 135-148 148-159 Weight, kg 31.1 32.8

33.1 32.3



Figure 2. Impact of weight at first service on total born through 3 parities (N. Williams, 2005)
Backfat; Results illustrate that backfat at first service appears to have minimal impact on productivity through 3 parities (Figure 3). Therefore, gilts should be targeted to be bred at 135-150 kg, regardless of age and /or backfat level.

34.0 Total Born 33.5 33.0 32.5 32.0 <12 32.54

33.68 33.33 32.79







> 20

Backfat, mm

Figure 3. Impact of backfat at first service on total born through 3 parities (N. Williams, 2005)


Breed on second or third estrus: Gilts bred at their first estrus have significant lower number born in their first parity, and consequently, lifetime born. At farm specific breeding age, gilts need to have at least one detected estrus without breeding or Heat-No-Service (HNS). Gilts with recorded pubertal estrus have significant advantages to the management of the gilt pool: a) their breeding week becomes predictable, b) it reduces labor related to boar exposure of cycling gilts, c) the majority of these gilts are likely bred at their second estrus if they have reached the targeted breeding weight, d) a high percentage of gilts will be bred on their 2nd or 3rd estrus; a few, ideally none, on their first estrus
Thus, breed gilts at their 2nd estrus (if body weight is at least 135 kg). Data in Figure 4 demonstrates that the productivity of sows through three parities is similar irrespective of breeding at second or third estrus.

34.0 Total Born

33.9 33.4 33.1




31.0 2nd 3rd 4th Estrous Cycle 5th

Figure 4. Impact of estrus cycle at first service on total born through three parities (N. Williams, 2005)

4. Boar Power: Boar power is a critical factor. The boar is the fundamental tool for stimulating gilts to reach onset of puberty and conduct estrus detection. Too often this basic tool is under estimated. GDUs are then managed with lack of boar numbers or low quality boars. It is mandatory to establish a program for regular introduction of heatdetection teaser boars to keep always the right boar power at the sow farm and at the GDU. Heat-detection boars need to be mature (over 10 months of age) and sexually active to be effective. To keep them active, boars must breed or get collected 1-2 times/week. We consider adequate boar power to be 1 boar for every 60 gilts being exposed. 5. Estrus Induction: Not only inducing but also recording estrus is a very important activity in the GDU management. The boar plays a critical role to induce pubertal estrus.


Personnel and boar power are the key elements to achieve a high percentage of cycling gilts within 4-6 weeks of first boar exposure. Effective boar exposure should involve direct physical contact with boar, minimum once a day, and for approximately 15 minutes per gilt pen. Boar exposure should start 4-6 weeks before the desired time of breeding. Over 80-90% of gilts are likely to show estrus within 4-6 weeks. This response should be tracked for every gilt group in order to correct or adjust if targets are not being achieved. Based on data from the University of Alberta summarized in Figure 2, nonresponding (>40d) or late-responding gilts (>30d) are gilts with inherited low reproductive potential. The best option is to remove them and not breed. Removing such non-cycling gilts is an additional selection process that is highly recommended. Additional gilts need to be initially considered in the flow to be able to select for reproductive potential.

100 90 Retention Rate (%) 80 70 60 50 40 30 20 Breed Farrow Rebreed Farrow Rebreed Farrow Rebreed Farrow 1 1 2 2 3 3 4 EP IP LP NR

Figure 5: Breeding, pregnancy, farrowing, weaning and rebreeding rate of Parity 1 sows as a percentage of gilts originally on inventory (Patterson et al, 2004)

6. Flush feeding before breeding: The level of feeding before breeding is critical for gilt performance. If gilts are restrict-fed, the level of feeding must be increased to at least 6.5 lb or 3 kg per gilt for 7-10 days before breeding. This objective is easily accomplished if gilts are in pens and with finisher self - feeders. Like weaned sows, crated gilts should be fed at least twice a day to reduce feed wastage and entice high intakes, in the order of 8-9 lbs./day.


HOW TO IMPLEMENT AN EFFECTIVE GDU MANAGEMENT PROGRAM A practical approach to implementing an adequate gilt puberty management is to use the concept of the Magic 42. There should be heat induction period of three weeks when intensive work with boars is needed to trigger the first heat, and record the HNS dates. The second period of 3 weeks (from week -3 to day 0) is a period where the remaining late-responding gilts finally achieve their first estrus. The ones that have already cycled show their second estrus. Following this protocol, at breeding age it is possible to have full control over the gilt pool. This leaves the farm in a privileged position of having the control over the gilt pool, and finally over the breedings. Figure 6 shows the different periods of the Magic 42 program.

The "Magic 42" concept


Day 0 2nd Estrus

Day +21

1st Estrus

BREEDING 2nd-3rd
>90% with HNS Selection of non-cycling gilts Control over the gilt pool

Intensive daily boar exposure

Continue boar exposure >70% with HNS

Figure 6: The Magic 42 gilt puberty management process (J. Sprke, 2003)

SUMMARY OF KEY ASPECTS OF GDU MANAGEMENT: Physical selection of gilts must be completed before puberty induction Boar stimulation for onset of puberty should start at 150-160 days of age. Depending on flow and breeding weight it can start at maximum 180 days of age Gilts are exposed for a minimum of 30 days, and ideally 42 days, (4 to 6 weeks) Intensive boar exposure should be focused during first 3 weeks to induce the first estrus Estrus has to be recorded for each gilt Non-cycling gilts have to be identified and culled prior to the maximum acceptable market weight for the pork chain. Flush feeding should be conducted 7 10d before breeding in gilts are being restrict-fed. Have the majority of gilts at 300 or more lb at breeding Boar power should be provided at rate of 1 teaser boar:60 gilts being exposed Trained personnel is needed to run an effective and efficient GDU management program Pharmacological aids are not recommended for estrus induction, unless the program is not working


CONCLUSIONS Adequate gilt management and preparation of the gilts for first breeding has a significant effect on the performance of the gilts in the sow farm. Due to the relationship between gilt performance and lifetime productivity, it is highly recommended to manage gilts properly during puberty to maximize their output in the first parity. Weight and estrus number at breeding are the major requirements for gilts to be bred. Gilts bred at 300-320 lb (130-145 kg), and in second or third estrus, achieve the highest productivity. Early puberty induction in gilts is essential to record the first estrus, and to be able to select the gilts that didnt show estrus after 2 days of boar exposure. Late puberty gilts should not be kept as replacement gilts for the breeding herd unless the plan is to cull them after weaning the first litter. A good gilt puberty management program can bring an additional value of up to $150 per gilt introduced into the herd.

LITERATURE CITED AND RECOMMENDED REFERENCES: G. Dial et al. 2001. The application of improved gilt pool management: An industry perspective. Advances in Pork Production, Vol 12, 185-195. F. Aherne, 2001. Gilt Production and Management. Pigletter Feb-Apr 01 Donald G. Levis, 2000. Housing and management aspects influencing gilt development and longevity: A review. A.D. Leman Conference 2000. Sam K. Baidoo et.al. 2000. Increasing Sow longevity: Body conditioning and feeding. A.D. Leman Conference 2000. P.E. Hughes et al. 1997. The effects of contact and transport on the efficacy of the boar effect. Animal Repr. Sc.. 46:159-165 D.W. Rozeboom et al., 1996. Influence of Gilt Age and Body Composition at First Breeding on Sow Reproductive Performance and Longevity. J. Anim. Sc. 74:138-150 G. Foxcroft and F. Aherne, 2000. Management of Gilts and First Parity Sows: Parts II, III, IV, V and VI. VII Simposio Internacional de Reproduccin e Inseminacin Artificial, Brasil. J. Patterson and F. Aherne, 2001. Effect of Age of Gilt at first Boar exposure on age at puberty. Alberta Revealing Research. G. Foxcroft and F. Aherne, 2001. Rethinking Management of the Replacement Gilt. Advances in Pork Production, Vol. 12. Pg 197 J. Sprke, 2004. PIC internal communication E. Beltranena, et. al. 2005. 12 Non-negotiable aspects of gilt development. Western Hog Journal. Summer 2004. 45 50. M. Young, 2005. Gilt development: A review of the literature. AASV Gilt Development and Acclimatization Seminar, Toronto. J. Sprke, 2005. Gilt development programs in North and South America. AASV Gilt Development and Acclimatization Seminar, Toronto. N. Williams, 2005. Non-Negotiables in Gilt Development. Banff seminar. J. Patterson, 2004. Gilt selection for improved lifetime productivity. 2004 Joint Meeting St. Louis, MO.


Health Status and Biosecurity in Gilt Development Units

Laura Batista Universit de Montreal , Facult de mdecine vtrinaire, CP 5000 , St. Hyacinthe, J2S 7C6, Qubec, Canada E-mail: laura.batista@umontreal.ca INTRODUCTION There are several ways to introduce new genetic material into a breeding herd. Live animals may be purchased or produced within a system as selects, breeder weaners or weaned pigs. New genes may also enter a herd via semen, embryos or cesarean-derived piglets. Regardless, the goal of any introduction is to improve genetics while maintaining herd health. Therefore the health status of the donor herd should be scrutinized and compared closely with that of the recipient herd. Herds should match as closely as possible in health status. Developing gilt development units (GDU) systems that meet the needs of isolation, acclimation and overall biosecurity within production systems is a key component of breeding herd management. The concept of offsite, or segregated on-site, GDUs may actually facilitate herd depopulation/re-population programs when dealing with PRRS and other prevalent diseases. In general we could say that other than feed cost and market price, disease is the greatest economical risk element of swine production units. Biosecurity of GDUs encompasses all efforts towards disease exclusion and disease challenge reduction. Biosecurity has two elements, external biosecurity and internal biosecurity. External biosecurity focuses efforts toward keeping out new disease agents and internal biosecurity focuses efforts toward minimizing the challenge level of diseases presently in the herd. OFF-SITE OR ON-SITE GDUS? There is no obvious answer to this question. Both systems offer advantages and disadvantages. It is certain that off-site units offer an extra security to avoid both disease transmission from incoming animals or to incoming animals. However there are added costs, management, and labour constraints related to an off-site unit. For this reason, the swine industry is tending to shift towards on-site GDUs. BASIC BIOSECURITY RULES FOR GDUS Location and People


Location is one of the most important considerations in determining long term herd health of a unit. Considerations include distance between swine production units pig density in local area, probability of other pigs coming to the area, water supply, and environmental concerns. People can be important carriers of disease. Therefore it is important to restrict access to buildings with a perimeter fence and by padlocking all outside doors. An unforgiving attitude in relation to biosecurity is required because one mistake can lead to serious problems. Every worker and visitor on the farm must have this unforgiving attitude. There can be no exceptions. Require clean time between units for visitors. Provide shower in / shower out facilities with clean and dirty areas. A major fault with most shower facilities is their lack of privacy and hygiene. Always provide clothing and boots. Another very practical and effective system is the Danish bath:
1. 2. 3. 4. 5. 6. 7.

in door remove top clothes and shoes go over a grate (denotes leaving dirty area) wash hands, scrub nails with soap and water put on farm clothes and footwear through a footbath (emphasizes clean area) into the unit

Equipment and Transport In the last five years, new research related to porcine reproductive and respiratory syndrome (PRRS) has proved the importance of fomites and transport in the dissemination of disease in the swine industry. All incoming equipment should be disinfected. Today many units are equipped with UV and fogging chambers and are used to disinfect ALL incoming equipment, no exceptions should be permitted. A protocol for all transport should be established and respected. All trucks (animal transportation, feed, etc.) should arrive washed and disinfected to the GDU. The famous 5 Ds (disinfectant, dilution, delivery, duration and drying) should be enforced to avoid contamination through transport. Units should have a loadout area to prevent entry by truck drivers. Loadout areas should have a roof and fencing and should use material that is easy to wash and disinfect. The chute gate should have a guillotine mechanism which prevents animals from re-entering the loadout holding area. Boots should be provided for the drivers and should be worn will working the chute area. These can be kept in a garbage can next to the chute. Clean, disinfect, and dry. The loadout should be washed, disinfected and dried immediately after the truck leaves the unit. Insects and domestic and non-porcine feral animals Other animal species and insects can be vectors for disease. For example, birds can transmit TGE (36 hour survival and 25 mile range) and can carry Erysipelas and Salmonella. Rats can range distances of 1 to miles whereas mice range very little. Mosquitoes and flies have been shown to


be vectors of porcine reproductive and respiratory syndrome virus (PRRSV). Even thought control of rodents, flies and mosquitoes can be very difficult, it is very necessary. HEALTH GOALS OF GDUS The first goal is to create stability. Properly managing replacement gilts creates a key link in providing homogeneity of health between incoming stock and the existing sow herd. This homogeneity and stabilization of immunity is very critical in maintaining a balanced, sustainable production system. PRRS has made us re-engineer the concept of gilt acclimatization in order to avoid the introduction of nave, viremic or persistently infected animals into positive herds. To reach this goal, gilts should go through a period of isolation, acclimation and recovery prior to being introduced into the new herd. Similar systems should be established for both internally and externally sourced animals. Isolation Period Isolation is defined as segregation from the main herd. It is a period of time to observe pigs for any health changes and conduct. In this period health supervision is very important and in general it has lasts around 7-14 days. Observation of clinical signs, pertinent serology (PRRSV, Swine Influenza, Mycoplasma, etc.), and necropsies are performed in order to assure the health status of the incoming animals. This period also allows the provider (genetic company o source farm) to contact you if there has been any health change in the source herd. Health information should be available through monthly veterinary visits, production data and routine diagnostics. Isolation protects the receiving herd from exposure to known clinical diseases and allows identification of incubating diseases or contamination during transport. It also prevents an overwhelming exposure of diseases present in the recipient farm to incoming animals before they are prepared. The herd health risk is a function of agent, separation distance, animal population, age of the pig, biosecurity and physical location of the isolation unit with respect to the sow population. As the size of the group in isolation increases, and separation distance decreases, the risk of aerosol disease transfer increases; this statement is particularly important for on-site GDUs. Flow assumes that animals in isolation are of high risk to the main herd. The building should have a separate shower and entry. Ideally, one person should take care of these animals. However, employees within the main unit can manage the isolation unit as long as it is visited as the last task of the day and the personnel maintain separate boots and clothes and shower in or out.


The isolation facility must be all-in, all-out in order to maintain a period of true isolation, it should not be operated continuous flow. Each time a group of animals enters this unit it should be perfectly washed and disinfected. Acclimatization Period Acclimatization allows incoming animals to become accustomed to new facilities, recipient herd pathogen levels and endemic disease agents in the receiving herd. It involves vaccination of animals to develop acquired immunity, and exposure to receiving herd pathogens (controlled exposure) to develop natural immunity. The purpose of acclimatizaton is to stimulate individual animal and population immunity. The acclimatization program should be herd and system specific, based on the health status of the incoming animals and the receiving herd; therefore its duration is variable. It is important to design a system of disease profiling to identify if natural or vaccine exposure has been successful in exposing gilts to agents such as Porcine Reproductive and Respiratory Syndrome (PRRS), Mycoplasma, Parvovirus amongst others. Remember that serology will only indicate if the individual has been exposed to the virus/bacteria, not whether it is truly infected. Many tests are available such as ELISA, SVN, IFA, VI, PCR, IHC, etc., some detect antigen, some antibodies; they all have advantages and disadvantages, non are perfect. Antigen detecting tests can provide more information about whether the animal is actually harbouring virus or bacteria. However, a negative result from an antigen detecting test should not be considered to be 100% accurate. An accurate test result is highly dependent of the stage of the disease, sample handling, and sample collection technique. Recovery Period The last phase of health management for incoming animals is the recovery period. This period begins after isolation and acclimation, and plays a very important role when animals have been exposed to PRRSV. The recovery period allows stabilization of individual and group immunity, reducing risk of disease spread to the receiving sow herd. After the emergence of PRRSV the recovery period has extended. It is commonly accepted that a recovery of 60 days or longer is necessary to reduce the risk of disease transmission, before adding them to the existing herd. Adult animals are remain viremic to PRRSV for a period of 2130 days post-exposure; however persistence and shedding can lasts for more than 80 days. Therefore it is important to allow a long recuperation period. This situation has complicated the GDUs management. In order to provide a 60-day recovery period, design your GDU according to the expected breeding age. If your target breeding age is 210 days, and you need at least an 60-day period for all the health management of replacement animals, meaning they need to be received at 140-150 days of age. If the farm has just one gilt developer facility, animals should arrive at nine-week intervals to avoid continuous flow. In this is the case, you will need to receive staggered-age gilts in a batch program.


There are two very practical options. The first one is to have two or more gilt developer units. This allows alternating the flow between the two units. Each gilt developer unit needs to be designed based on the inventory of animals, length of occupation and exit age and weight of the gilts. Maintaining a separate period of isolation followed by acclimation and recovery allows gilt delivery every 90 days. The second option is to receive gilts at 5kg follow on-site nursery and finisher flow and 50-60 days before transfer into the herd, the gilts are moved into a separate area where they will finish their recovery phase. This has proved to be a very effective gilt acclimatization strategy. However, in the case of PRRSV, if the farm is stable and there is no recirculation in the nursery, incoming gilts will not be exposed, and will be nave at the end of the acclimatization period. In this case gilts can be vaccinated against PRRSV when they enter the recovery phase in an isolated barn. The same procedures must be followed for animals in an internal, closed-herd multiplier as from an outside source. Don't assume herd health of those sources is similar to that of your herd. Profiling has proven that internally produced replacement animals aren't consistently healthy. They need to go through the same acclimation and recovery period as externally produced replacements.

CONCLUSIONS The benefits of properly conducted GDU have been identified by numerous researchers. The health assurance program should address both external and internal biosecurity. It should minimise the opportunity for diseases to enter the unit (external biosecurity), as well as minimising the effects of diseases within the herd (internal biosecurity). New breeding animals should be obtained from one source whose existing health status is compatible with the receiving herd. The source-herd health status should be evaluated critically for the presence of diseases that are of particular concern such as porcine reproductive and respiratory syndrome (PRRS), Mycoplasmosis, and Swine Influenza among others. The gilt developer unit system needs to incorporate all aspects of health and proper management. The design and sizing of these facilities depends on the capacity for the gilt pool within the breeding herd facility, the age and weight of service, the age and weight of the incoming animals and the length of each period. Finally, success depends on following detailed procedures and designating a person accountable for implementation.

REFERENCES 1. Agence canadienne dinspection des aliments. La bioscurit la ferme. Guide pratique.


http://inspection.gc.ca/francais/anima/heasan/fad/biosecurf.shtml 2. Amass S. F. 2000. Investigation of people as mechanical vectors for porcine reproductive and respiratory syndrome virus. Swine Health and Production, 8 (4) : 161-166. 3. Amass S. F., Clark L.K. 1999. Biosecurity considerations for pork production units. Swine Health and Production, 7 (5): 217,-228. 4. Amass S. F., Vyerberg B.D., Ragland D., et al. 2000. Evaluating the efficacy of boot baths in biosecurity protocols. Swine Health and Production, 8(4):169-173. 5. Amass, S. F. 2001. Biosecurity to prevent mechanical transmission of swine pathogens: threat or fiction ? In Proc. Allen D. Leman Swine Conference, p. 43-45. 6. Amass, S. F. 2002. Biosecurity: what does it all mean. Proc. 33rd AASV Annual Meeting. p. 279-281. 7. Batista L, Pijoan C, Dee S, Olin M, Molitor T.A., Joo, H.S Xiao Z, Murtaugh M (2004) Virological and immunological responses to porcine reproductive and respiratory syndrome virus (PRRSV) in a large population of breeding age female swine. Can J Vet Res 8. Batista, L, Dee, SA, Rossow, K, Deen, J and Pijoan, C (2002) Assessing the duration of porcine reproductive and respiratory syndrome virus persistence and shedding in a large population of breeding age female swine. Can Jour Vet Res 66:196-200. 9. Batista L, Torremorrel M and Pijoan C (2002) Experimental exposure to porcine reproductive and respiratory syndrome virus (PRRSV) in gilts during acclimatization. J Swine Health Prod; 10, 4:147-150. 10. Boutin, R., Broes A. 2001. La bioscurit la ferme: un must pour tous les levages. Colloque sur la production porcine, p. 58-81. 11. Broes, A. 2002. Les mesures de bioscurit dans les levages porcins qubcois. Journe : "De la dmarche hygine la bioscurit", ISPAIA - SOGEVAL, Ploufragan, France. 12. Connor, J.. F. 2002. Biosecurity becomes necessity for 21st Century pig production. Swine health and epidemiology report. Special biosecurity issue. p. 1. http://animalagriculture.org/swine/29981%20NIAA%20Swine%20Healthbio.pdf 13. Corrigan, R. M. 2002. Overview of rodent control for commercial pork operation. Swine health and epidemiology report. Special biosecurity issue. p. 8. http://animalagriculture.org/swine/29981%20NIAA%20Swine%20Healthbio.pdf 14. Dee, S. A. 1999 An overview of methods for measuring the impact of sanitation procedures for swine transport vehicles. Swine Health fact Sheet. Vol. 1, No.2. 15. Dee, S.A. 2005 The 5Ds of transport sanitation. International Pigletter. Vol. 25. No. 3b http://www.porkscience.org/documents/Other/transportsanitation.pdf 16. Dewey C., Friendship R. Health status as a risk factor: A systems and Unit Perspective. http://gov.mb.ca/agriculture/livestock/pork/swine/bab11s15.html 17. Fangman Th. J. 1998. Securing your investment by securing the health of your herd. http://www.agebb.missouri.edu/commag/inst/fangman.htm 18. Moore C. 1992. Biosecurity and minimal disease herds. Veterinary Clinics of North America: Food animal practice. 8 (3), 461-474. 19. Morrison, B. 2002. Biosecurity is like insurance I want as little as I need. Western Hog Journal. p. 30-37. 20. National Pork Board and American Association of Swine Veterinarians. Biosecurity Guide For Pork Producers. http://www.porkscience.org/documents/Other/final%20biosecurity%20book.pdf


21. Stark K.D.C. 1999. The role of infectious aerosols in disease transmission in pigs. Vet. J.158, 164-181. 22. Thompson Robert. 2002. Transportation cleaning and disinfecting. Swine health and epidemiology report. Special biosecurity issue. p. 10. http://animalagriculture.org/swine/29981%20NIAA%20Swine%20Healthbio.pdf 23. Thompson, R. W. 2001. Transmission of pathogens via transportation vehicles. Proc. Allen D. Leman Swine Conference, p. 37-42. 24. Tubbs, R.. 2002. Isolation of incoming breeding stock to prevent disease. Swine health and epidemiology report. Special biosecurity issue. p. 4. http://animalagriculture.org/swine/29981%20NIAA%20Swine%20Healthbio.pdf


Part 2: Putting principles into practice


Designing effective boar stimulation systems as a critical feature of the Gilt Development Unit
Eduardo Beltranena1, Jennifer L. Patterson2 and George R. Foxcroft2 Swine Reproduction-Development Program, Swine Research & Technology Centre, 1 AAFRD, and 2University of Alberta, Edmonton, Alberta, Canada. E-mail: eduardo.beltranena@gov.ab.ca INTRODUCTION The Gilt Development Unit (GDU) is a physically distinct area of modern breeding barns. This area houses replacement gilts while they are stimulated to reach puberty (1st estrus) and are subsequently bred for the first time. Boar exposure is the single most important factor in stimulating early puberty in gilts. Therefore, a properly designed area where gilts can be exposed to boars is key to both successful breeding and labor management. We have conceived our Boar Exposure Area (BEAR) to be the hub of effective boar power delivery in the GDU. FEATURES AND LAYOUT Figure 1 shows our BEAR concept layout. It encompasses retrofitting three typical (8 x 16; 2.5m x 5.0m), fully slatted, hog finishing pens. The two side pens are equipped with entry/exit swing gates (8; 2.4m) that pivot on either end. The pens are not altered on the far sides. The only optional additions are two posts per corner to round up the corners. These posts prevent the gilts from being cornered and injured by the teaser boars. Corner posts also grant get away protection to the herdsperson in case a boar should turn on him. The center portion of our BEAR design includes: a) Four to six teaser boar crates (24 - 30w x 6 - 8l x 46h; 60 75cm w x 1.8 2.4m l x 1.15m h). These boar crates have spindle gates with boar-secure latches or drop pins at both ends. Nipple drinkers are made available on either end of the crates so boars can drink when facing either way. Boars are fed using pans that slide under the crates gates. A water drip system is installed on top to cool the boars in summer time. b) One horny gilt pen (4w x 6 - 8l x 40h; 1.2m w x 1.8 2.4m l x 1m h) equipped with swing gates on three sides that pivot on either end. c) One platform scale for finishing hogs (2w x 6 8l x 40h; 61cm w x 1.8 2.4 m l, x 1m h). Attachable crowding gates on either side of the scale are used only when weighing gilts, if desired. d) One narrow herdspersons pass-thru space (8 - 12 w x 6 - 8 l x 40h; 20 30cm w x 1.8 2.4m l x 1m h; without bars on top) between the scale and boar crates.


Figure 1. SRTCs recommended Boar Exposure Area (BEAR) for the Gilt Development Unit. Installing a BEAR merely involves retrofitting two or three conventional growout pens. Drawing by Crystal Korth, AAFRD.

TYPICAL FLOW AND MANAGEMENT 1. Upon reaching 150 180d of age, a group of no more than 15 gilts housed in a similar size pen (8 x 16; 2.5m x 5.0m) is herded into one of the two boar exposure pens located on either side of the teaser boars crates through Gate 1 or 2 (Figure 1). 2. The herdsperson should observe the gilts carefully. Any gilts showing signs of standing heat (rigid, ears up) as they enter the BEAR and face the caged boars should be identified (record ear tag and mark). The herdsperson should then walk to the standing gilt and score and record the redness and swelling of the vulva. He should then determine the extent of standing reflex by sitting on or applying pressure on the gilts back. Any confirmed standing gilt(s) should then be pushed into the horny gilt pen.



Horny gilt pen

Figure 2. Top view of the Boar Exposure Area (BEAR) showing two groups of gilts being simultaneously exposed to a teaser boar on each side. Teaser boars are housed in crates (center) and may enter, exit or face on either direction. A scale to weigh gilts (top center) and a horny gilt pen (bottom center) to hold heat-standing gilts are also depicted. The herdsperson holding a chase board, is shown standing at the pass-thru between pens ready to intervene should it be necessary. 3. The herdsperson should then release a high-libido teaser boar (older than 10 months) to stimulate the remaining non-standing gilts. The herdsperson then moves to the passthrough space or corner behind the posts. He should note the time and then observe the gilts reacting to the physical teaser boar contact for a few moments. 4. If other gilt(s) show(s) signs of standing upon direct physical contact with the teaser boar, the herdsperson should intervene. He must approach the engaged pair from the gilt side with a hand-held board. Try to separate the two by shoving the board between them. Either the boar is then escorted into its crate or the newly identified heat-standing gilt is pushed into the horny gilt pen. Whatever is easier to do, but the herdsperson must not expose his back to the boar at any moment. 5. According to the time elapsed, the agitation of the boar and the ambient temperature, the herdsperson may decide to release the same or a different teaser boar into the pen with


the remaining gilts again for another round of interaction. Observe for additional gilts that may be close to standing in heat from the pass-thru stall or behind the posts in a pen corner. 6. If no other gilts are showing signs of standing in heat, the herdsperson may leave the teaser boar unattended in the pen with the gilts for a moment. He would then herd the next group of gilts to be exposed on the opposite side pen (Figure 2). 7. Once 15 20 minutes of direct physical boar contact have elapsed, the teaser boar should be escorted to its stall using a board. Non-standing gilts can then return to their home pen. Preferably, the herdsperson should first release the standing gilt(s) out of the horny gilt pen, through Gate 3 into the alley. Her penmates can then chase her as they are subsequently released into the alley through Gates 4 or 5. 8. Gilts that have shown their heat-no-serve (HNS) pubertal estrus are marked with a different colour at 5 7 day intervals. Gilts may return to their home pen, but days later they should preferably be relocated to individual stalls (crate snake). Boar stimulation can then be focused on gilts that have not stood yet. Mixing gilts at Day 14 may provide an additional stimulus. Gilts that do not stand within 23 days may be treated with PG600 (Intervet) and exposed for an additional week of direct physical teaser boar contact. Nonresponders should then be promptly sent to slaughter. 9. We weigh gilts within days of showing their heat-no-serve pubertal estrus. This weight is used to predict weight at fertile breeding. Gilts are bred for the first time between 135 150kg at either their second or third estrus according to their growth rate while in the GDU. 10. Teaser boars (sterile or intact) in the GDU must be allowed to breed a heat-standing gilt or be hand-collected while jumping a dummy weekly. Doing so will maintain a high libido, interest in the gilts and prevent build-up of sexual frustration.

BENEFITS Our BEAR layout is simple, cost effective to adopt and to implement in new or refurbished barns. The layout provides two gilt-checking areas and a pass-thru space between them. One person can simultaneously check two groups of gilt by circulating from one side to the other. The BEAR provides housing for up to six boars. The crates allow the teaser boars to face and enter/exit both gilt checking pens. Initial fenceline exposure to aroused, pheromone-loaded, salivating and frothing boars is a very powerful stimulus to induce early puberty in gilts. It allows for earlier identification of heat-standing gilts and these are seen standing for a longer period of time (Table 1).


Table 1. Response of sows to back pressure, boar exposure, both or both conducted in the Boar Exposure Area (BEAR) after weaning (modified from data of Langendik et al. (J.Anim. Sci. 2000. 78:871).

1. Back pressure Sows in estrus, % Onset of heat, h Duration of heat, h 46 107 22

2. Presence of boar 56 106 29

Both 1 and 2 Both 1 and 2 90 99 42 in BEAR 97 93 55

Gilts identified standing can be parked in the horny gilt pen. A teaser boar can then impart additional physical contact to the remaining non-standing gilts. To establish the breeding at the desired weight range, a scale and crowding gate(s) are positioned to weigh gilts after the pubertal heat-no-serve estrus. Protection posts round corners to prevent gilts from being injured by the boar. Both the pass-thru space and corner posts offer protection for the safety of the herdsperson.


Use of pharmacological interventions to further improve GDU efficiencies


George R. Foxcroft and 2Eduardo Beltranena

Canada Research Chair in Swine Reproductive Physiology, Swine Reproduction-Development Program, Swine Research & Technology Centre, University of Alberta, and 2 AAFRD, Edmonton, Alberta, Canada. E-mail: george.foxcroft@ualberta.ca

INTRODUCTION The efficiency with which gilts are introduced to the breeding herd is a major determinant of breeding herd efficiency, even in established operations. Whether gilts are reared in house or supplied by a breeding company, the entry-to-service interval is a major factor contributing to the total non-productive days (NPD) in the herd. Even in well-managed units, gilts may contribute up to 30% of NPD. In units in which gilt pool management is inadequate, the contribution of gilt entry-to-service interval to NPD can be much greater. Secondly, a failure to meet breeding targets has a major impact on total pigs weaned per inventoried female per year and results in under-utilization of farrowing room capacity. In the situation of large start up operations, the efficiency of gilt pool management is crucial to ensuring that expected production targets are met. There are two components to an efficient gilt pool management program: 1. Effective induction of pubertal estrus (cyclicity) and selection 2. Efficient allocation of cyclic gilts to meet weekly breeding targets When gilts are obtained from a multiplier, it is generally assumed that they are already cyclic, or will be induced to cycle in response to transport and mixing effects. If gilts are managed "inhouse" from an earlier stage of their development, then effective stimulation of pubertal estrus becomes an important management factor. In both situations, estrus synchronization techniques allow breeding targets to be met on a weekly basis from a smaller sized gilt pool than is needed when synchronization of estrus is not used. However, the financial costs of estrus synchronization must be less than the cost of maintaining a larger gilt pool, if estrus synchronization is to be widely employed in the industry.

SYNCHRONIZATION OF PUBERTAL ESTRUS AS AN AID TO BREEDING MANAGEMENT Responses to stimulation with boars. The factors that may contribute to puberty onset and possible use of different hormonal strategies to induce first estrus were comprehensively reviewed by Hughes (1982) and Paterson (1982). Little has changed in the development of such techniques since these reviews were published. Use of natural boar pheromones can be very effective in inducing a fairly synchronous onset of first heat, as long as gilt can be managed to


optimize the response to exposure to boars (see Hughes, 1982). More recent data on the response to stimulating gilts with mature boars has been extensively reviewed earlier in these proceedings and will be considered in commercial practice later. In theory, if the responses to boar stimulation was very consistent on a week to week basis, then as long as the size of the gilt pool could be increased to provide the required number of cyclic gilts per week, there would be little need to use estrus synchronization to meet breeding targets. However, if gilts are older and heavier at the time of transportation or first exposure to boars, there will be a fairly synchronous expression of first estrus. The assumption here is that many gilts have already exceeded the weight and age thresholds to attain puberty but have not been exposed to the stimulus to trigger final sexual maturity. When these stimuli are applied, a large group of gilts reach estrus at the same time. In this situation we can see that around 30% of the gilts already showed estrus within the first 7 days of boar contact and a higher percentage overall had cycled within 40 days of exposure to boars. Although the number of gilts per week available for breeding is greater in this situation, there are two disadvantages. First, it may be less effective in identifying gilts that are the most sexually mature. Second, in order to meet regular breeding targets it is necessary to develop some system of heat synchronization, because the number of gilts in pubertal estrus in any one week is very variable. In order to meet regular weekly breeding targets for gilts, there will either have to be much larger gilt pool to be certain that in any one week enough gilts showing a natural heat are available to match breeding targets. The large availability of gilts in the first week will likely exceed breeding requirements. Some of these gilts must therefore either be kept for a complete 21 day cycle and become part of a later breeding group, or a proportion of the week-one gilts must be synchronized to meet breeding requirements at a later specified time. The basis for using pharmacological techniques for delaying estrus until gilts are required to be bred, whereby a more even distribution of gilts can be created, with obvious efficiencies related to the smaller size of the gilt pool needed. An excellent summary of the basis for using the synthetic progestin, allyl-trenbolone (also known as altrenogest, Regumate and Matrix) is available through the Intervet website, and this information was largely written and edited by Dr. Billy Flowers.

Use of exogenous hormones to induce cyclicity. The use of low dose combinations of exogenous gonadotropins was also reviewed by Paterson (1982). The most extensive data relate to the use of the product PG 600 (Intervet) which contains 400 iu eCG (PMSG) combined with 200 iu hCG. In gilts in the late pre-pubertal stage, induction of a fertile estrus can be achieved in a high proportion of gilts, although resulting litter size can be more variable than is seen in gilts bred to a natural estrus. A number of studies report problems with the predictability with which gilts continue to cycle after a PG 600-induced first estrus and ongoing daily exposure to boars was found to partly resolve this problem (Paterson and Lindsay, 1981). Others reported problems with a lack of behavioral estrus, even when gilts ovulated to treatment (Paterson et al., 1984). This was found to be related to the immediate ovulation of the more mature follicles on the ovary at the time of treatment, presumably in response to the hCG component of the treatment


(Steadman and Foxcroft, unpublished data). The resulting increase in plasma progesterone was thought to block the expression of behavioral estrus, even in the presence of elevated estrogen levels in response to the development of a second wave of ovulatory follicles in response to the PMSG component of the vehicle. There appeared to be a greater risk of these early induced ovulations, the heavier (more mature) the gilts were at the time of treatment.
Collectively, these data suggest that one way of avoiding some of these problems may be to breed gilts immediately after PG 600 treatment and not risk the lack of continuing cyclicity. However, in a recent GDU trial carried out in collaboration with Pipestone Vet. Clinic, the details for which will be reported later in this Workshop, we encouraged the barn staff to use available information on the reproductive status of the gilt to improve the criteria for deciding on PG6oo intervention, and management techniques to optimize the response4s to treatment.

The relative sexual maturity of the gilt pool in question was known, as more than half their littermates usually had a recorded boar-induced natural first estrus (HNS) at the time opportunity gilts were treated with PG600. We avoided treatment of gilts that were in the luteal or late follicular phase of an existing estrous cycle, because early and intensive management of gilts through the BEAR system described earlier effectively identified truly non-cyclic gilts (non-responders) as the target group of opportunity gilts available for PG600 treatment. Treatment of gilts in the late proceptive period approaching their first natural estrus was avoided using a simple vulval scoring records ; gilts could be in the proceptive period of their first pubrtal estrus would have vulval scores of 2 to 5, and were not treated with PG600. Very intense contact with mature boars continued over the 5-day period after PG600 treatment; gilts were required to show a normal HNS in order to be retained as opportunity females. After a PG600 induced HNS, second estrus was synchronized by feeding Matrix, in the hope that this progestagen priming would improve the efficiency of heat detection and AI compared to experience with breeding gilts at the induced PG600 heat, or at a natural second heat without progestagen synchronization. Using this rigorous approach to PG600 treatment, the results of a recent commercial trail seem to demonstrate a useful role for PG600 treatment for recruiting eligible opportunity gilts into a specific breeding week (Figure 1). Over 90% of the gilts treated with PG600 were recorded as having a normal HNS event and were usually then bred at their second natural heat after synchronization with the oral progestagen, Matrix. Over 80% of treated gilts were recorded in estrus within the 5-day period of intensive boar contact originally required after treatment. In a normal gilt flow, the expectation would be that up between 60 and 70% of gilts given effective stimulation in a BEAR system would be recorded as having a natural HNS record within a 28-day period. The best of the remaining gilts that are contemporary to these females would then be eligible for PG600 treatment if there is an anticipated shortfall of known HNS gilts to meet future weekly breeding targets. In gilt cohorts where the percentage of gilts showing a boar-induced HNS event is unexpectedly low, a greater proportion of opportunity


gilts will require PG600 treatment. As will be seen later, the combined HNS responses to boar stimuli, plus PG600 treatment of opportunity gilts, are effective in: Identifying a total of 80 to 90% of the 4-week cohort as eligible gilts, with a recorded HNS event, within the 28-day period allowed. The breeding results from these opportunity gilts appear to be very acceptable. The fate of the low percent of non-eligible gilts, with no recorded HNS event, must in our opinion be to cull them as heavy market pigs. The costs and number of pen space days needed to produce a farrowing record from the residual non-responders is difficult to justify (see further comments below). Retention of non-responders beyond the set period for becoming eligible for breeding, simply leads to overcrowding in the GDU, or frequently the re-location of non-cyclic gilts to the gestation area where intensive management is very unlikely to occur. Figure 1. Response of a carefully controlled population of non-cyclic gilts, identified as opportunity females and treated with PG600 to induce HNS (Hawkeye 2 collaborative project; unpublished information 2005)

HNS < 10d after PG600 treatment

60.0% 50.0% Percent (%) 40.0% 30.0% 20.0% 10.0%

90.1 % HNS (n = 799)

0.0% 1 3 5 7 9 Days from PG600 treatment to HNS

Another possible approach is to breed gilts at the PG 600-induced estrus and then terminate this first pregnancy before day 30 to 35 with prostaglandin-F2 (PGF2 ). Although this treatment protocol may raise ethical questions, it provides a ready supply of gilts that can be returned to estrus in a controlled way and avoids the erratic litter sizes associated with PG 600 use. A more recent approach has been to develop the technology to induce pseudo-pregnancy in gilts by 50

treating cyclic gilts with estrogen depot preparations 7-11 days after ovulation (see Cushman et al., 1998). This mimics the anti-luteolytic effect of embryonic estrogen by extending the life of the newly formed corpora lutea. PGF2 can then again be used to induce estrus in gilts at any time after pseudo-pregnancy is established.


Our experience with the management of the 10 to 15% of gilts that fail to show a natural boarinduced heat , and also fail to respond within 5mor 6 days to initial PG600 treatment, reflects many anecdotal reports from the industry. Half of these gilts eventually can be bred and will farrow. However, the predictability of cyclicity is low and their lifetime fertility is probably very questionable. The other half of these gilts never produce a farrowing record and yet can accumulate 50 or more NPD in some systems before a culling decision is made. We have attempted to confirm these ideas by more carefully determining the reproductive status of these gilts without a recorded HNS event. Plasma progesterone assays indicate that half of these gilts had already cycled but had never been detected in estrus, whilst the other half were truly infantile and had still not reached puberty. Treatment of the population of presumed non-cyclic gilts with PG600 produced predictable results. Gilts that were already cyclic and in the luteal phase of the cycle at treatment, invariably developed cystic ovarian follicles, and although they were bred at the induced standing heat, they were infertile. Truly acyclic gilts responded to PG600 and ovulated and could potentially be included in appropriate breeding groups. However, the potential risks for future herd fertility when using exogenous hormones to induce pubertal estrus in gilts that are still non-cyclic well beyond the age range at which most of their contemporaries have already shown a natural, spontaneous, heat needs careful consideration.

SYNCHRONIZATION OF ESTRUS IN CYCLIC GILTS If a large enough gilt pool is maintained, then gilts will probably be available to meet approximately 80% of breeding targets. An alternative to increasing the size of the gilt pool to meet the remaining 20% of breeding requirements is to use estrus synchronization techniques. Use of oral progestagens. Effective synchronization of estrus in the gilt or sow is possible, and the most commonly used technique is the feeding of the synthetic progesterone analogue, allyl trenbolone (Regu-Mate). Feeding this orally active progestagen for 14 to 18 days in randomly cyclic gilts will result in effective estrus synchronization over a 4 to 6 day period after the last day of feeding (see earlier reviews of Webel, 1980; Martinat-Bott et al., 1985). If the stage of the estrous cycle is known, the number of days that Regumate needs to be fed can be considerably reduced, without loss of efficacy. The fertility of gilts is generally nor reported to be affected by Regumate treatment and indeed may be improved. We find this particularly true for AI use, because fresh semen can be ordered in a very predictable way and we can concentrate on gilt breeding over a concentrated period of time.


Our ongoing experimental use of Regumate for synchronizing estrus in known cyclic gilts resulted in 97% of gilts treated showing estrus within the designated breeding week (Figure 2).

Number of Days After Regu-mate Withdrawal vs. Percent of Gilts Showing Standing Heat
40.00 percent of gilts showing standing heat 35.00 30.00 25.00 20.00 14.75 15.00 10.00 5.00 0.00 0.00 1 2 3 4 5 6 7 8 N=217 9 10 num ber of days after Regu-m ate w ithdraw al 0.00 4.15 1.38 1.38 0.00 14.75 32.72 30.88

Figure 2. Results of recent use of Regumate to synchronize third estrus in known cyclic gilts with periods of treatment ranging from 5 to18+ days (Swine Research & Technology Centre, University of Alberta, in-house data, 1997) By finely adjusting the dose of Regumate fed (between the recommended range of 15 and 20mg/day), and the day preceding the breeding week on which Regumate treatment is terminated, it is possible to insure that all gilts are in estrus within a 5-day period. This is clearly illustrated in Figure 3, which presents the same data as in Figure 3, but as the cumulative percentage of gilts in estrus. The advantages of creating a predictable breeding week may be as much related to improvements in breeding technique and semen quality, as it is to the pretreatment with a progestagen per se. However, recent studies suggest that the tendency for litter size to marginally increase as a direct physiological response to at least Regumate use at the 20mg.day dose level, seems to merit further investigation.


Number of Days After Regu-mate Withdrawal vs. Cummulative percent of Gilts Showing Standing Heat
120.00 cummulative percent of gilts showing standing heat 100.00 79.72 80.00 60.00 40.00 20.00 0.00 0.00 1 2 3 4 5 6 7 8 9 N=217 10 0.00 1.38 16.13 94.47 98.62 100.00 100.00


number of days after Regu-mate withdrawal

Figure 3. Cumulative percentage of gilts in estrus after Regumate withdrawal. Withdrawal of treatment three days before the start of the breeding week, induces estrus in close to 100% of gilts during the 5-day breeding period. Because complete records of the date of pubertal and second estrus are kept as part of our routine gilt pool management system, we only introduce Regumate treatment from day 12 of the previous cycle in those gilts which are not due to show spontaneous estrus in the targeted breeding week. We then continue feeding Regumate until four days before the start of the breeding week. Depending on the date of the previous estrus, this results in treatment periods ranging from 5 to 18 days, representing a considerable saving in treatment cost. To insure that problems with low-dose treatment are not encountered, we generally require a minimum treatment period of 5 days in any gilt. To-date we have no evidence that the duration of Regumate treatment has any effect on gilt fertility (Figure 4). Irrespective of the duration of Regumate treatment, data accumulated from our routine use of Regumate also indicate, no effect of Regumate treatment on either litter size, or the distribution of litter size, when gilts are bred by AI at either second or third estrus (Figure 5). Therefore, recent data from our use of Regumate confirm the efficacy of the product. Our results also indicate the possibility of reducing the total amount of product used by maintaining appropriate records in the gilt pool, and only treating gilts that will not naturally cycle within the breeding week. Taking this approach, and combining Regumate use with effective puberty stimulation with boars, it is possible to meet all weekly breeding targets by treating only 25% of gilts within the gilt pool, and limiting the period of treatment to an average of 12 days. Overall, therefore, the use (costs) of Regumate treatment can be limited to an average of less than 3 days per gilt bred, and this may prove to be a very cost-effective technique for meeting breeding targets. However, economic models of effective gilt management are needed before the full economic impact of this, or any other, gilt pool management system, can be established.


Days on Regumate vs Total Born

18 16 14 Total Born 12 10 8 6 4 2 0 0 5 10 15 Days on Regum ate 20 25 30 y = 0.0326x + 10.413 R2 = 0.0029

Figure 4. Lack of any relationship between duration of Regumate feeding and subsequent litter size born (Swine Research & Technology Centre, University of Alberta, in-house data, 1998)

Regumate Non-Regumate 25.00%

Regumate (n= 73) mean = 10.7 Non - Regumate (n = 114) mean = 10.4

% of Total Born Per Litter





0.00% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

# Total Born per Litter

Figure 5. Litter size born in Regumate treated gilts and contemporary, non-treated females, showing the distribution of different litter sizes. No difference in either mean litter size, or the distribution of different sized litters is apparent (Swine Research & Technology Centre, University of Alberta, 1997-1998)


As will be apparent later, the translation of these early experiences with allyl-trenbolone marketed as Regumate, to validation of Matrix as the branded name for allyl-trenbolone on the N.American market has produced equally consistent results. In recent years there have renewed attempts to develop an intra-vaginal progesterone releasing device (PRID) as used in sheep and cattle. Information on the performance of a progesterone releasing CIDR for use in gilts has recently been presented by Day ( 2000). This alternative approach seems promising in terms of the synchrony of estrus observed; However, the reported conception rates of gilts bred after withdrawal of the CIDR wre below acceptable norms for the industry. Use of luteolytic agents. In other species, the use of natural or synthetic prostaglandins to cause luteolysis in known cyclic females can be an effective synchronization technique. Some degree of synchrony can be achieved with the use of PGF2 in cyclic gilts, however, the corpora lutea of the pig are only sensitive to PGF2 from day 12 of the estrous cycle. At most, estrus can therefore be advanced by some 5 days. Nevertheless, this may still be useful in bringing a number of gilts into a tighter breeding group. However, it is essential to have accurate records of gilt cycles if this technique is to be applied effectively. Recent studies on the use of multiple PGF2 injections early in the cycle have not been encouraging (Gadsby et al., 1997).

REFERENCES Beltranena, E., Aherne, F.X. and Foxcroft, G.R. (1993) Innate variability in sexual development irrespective of body fatness in gilts. J. Anim. Sci., 71, 471-480. Cushman, R.A., Davis, P.E., Boonyaprakob, U., Hadgpeth, V.E., Burns, P.D. and Britt, J.H. (1998) Use of slow-release estradilo-17 (SRE) and PGF2 to induce pseudopregnancy and control estrus in gilts. J. Anim. Sci., (Suppl. 1): 216. Day, B.N. (2000) Controlled synchronization of estrus and ovulation in swine. In Proceedings of the VII simposio Internacional de Reproducao e Inseminacao Artificial em Suinos, EMBRAPA, Brazil, pp 19-38. Elsaesser, F. and Foxcroft, G.R. (1978) Maturational changes in the characteristics of oestrogen-induced surges of luteinizing hormone in immature domestic gilts. Journal of Endocrinology, 78, 455-456. Foxcroft, G.R., Elsaesser, F., Stickney, K., Haynes, N.B. and Back, H.L. (1984) Ovarian oestrogen-dependent maturation of the luteinizing hormone/follicle stimulating hormone surge mechanism during prepubertal development in the gilt. Journal of Endocrinology, 101, 371-380. Gadsby, J., Graffinger, M. and Almond, G.W. (1997) Induction of short cycles in pigs with prostaglandin F analogs. Proc. 14th. Int. Pig Vet. Congr. pp564 (Abstr). Hughes, P.E. (1982) Factors affecting the natural attainment of puberty in the gilt. In : 'Control of Pig Reproduction', Eds. Cole, D.J.A. and Foxcroft, G.R.: Butterworths, Lond, Chp. 6, pp. 117-138. Martinat-Bott, F., Bariteau, F., Badouard, B. and Terqui, M. (1985) Programmes for controlled reproduction. J. Reprod. Fertil. Suppl. 33:211-228.


Paterson, A.M. (1982) The controlled induction of puberty. In : 'Control of Pig Reproduction', Eds. Cole, D.J.A. and Foxcroft, G.R.: Butterworths, Lond, Chp. 7, pp. 139-160. Paterson, A.M. and Lindsay, D.R. (1981) Induction of puberty in gilts: 2. The effect of boars on maintenance of cyclic activity in gilts induced to ovulate with PMSG and HCG. Anim. Prod. 32:51-54. Paterson, A.M., Pearce, G.P., Reed, H.C.B. and Foxcroft, G.R. (1984) The reproductive performance of gilts induced into puberty with oestradiol benzoate or a combination of PMSG and HCG. Animal Production, 38, 121-128. Webel, S.K. (1982) The control of ovulation. In: 'Control of Pig Reproduction', Eds. Cole, D.J.A. and Foxcroft, G.R.: Butterworths, Lond, Chp. 10, pp. 197-210.

ACKNOWLEDGEMENTS We acknowledge the work of all our present and former colleagues in conducting many of the studies mentioned in this review and the financial and in-kind support of our research from the Natural Sciences and Engineering Research Council, Alberta Pork Producers Development Corporation, Alberta Agriculture Research Institute, Pig Improvement (Canada) Ltd. and Alberta Swine Genetics Corporation.


Capturing Gilt Pool Information

Brad Thacker, DVM, PhD, MBA Intervet Inc.

Capturing Information involves the following general steps: 1. Recording the appropriate events 2. Entering data into a record system. PigChamp will be used for this presentation. 3. Retrieving data from PigChamp. 4. Comparing data versus expected/planned values (benchmarking)

For this presentation, the gilt pool process will be limited to the time from entry into the breeding herd through first service.


Slide 2

Gilt Pool Management






Generic Management Model: 1. Success is dependent on all 5 basic activities. 2. Accurate monitoring and critical evaluation validates initial plan and promotes continuous improvement. 3. Evaluation process transforms data in to information which then leads to action.


Slide 3

Appropriate Events
Begin with the end in mind
Flow parameters
Littles Law: Inventory = Time in Process X Entry/Exit Rate

Retention rate Reproductive performance

Fertility Litter size

Flow Parameters: Littles Law is a principal in operations management that is used to understand the relationship between: 1. Inventory- In this case the number of gilts in the gilt pool. 2. Time in Process- The amount of time that the gilts spend in the gilts pool. 3. Entry/Exit Rate- The number of gilts that enter and exit the gilt pool over time. This rate is dependent on the replacement (cull plus death) rate if herd size is stable.

Retention Rate: This parameter can have several starting and ending points. The starting point depends on when the gilts enter the process. Accordingly, the process needs to be defined and is preferably consistent over time such as entering gilts at approximately the same age. The end could be at one of several time points. In this presentation, it will be at the gilts first service.

Reproductive performance: The ultimate output variables of the process are fertility (farrowing rate) and litter size. In most operations, this data is readily available.


Slide 4

Appropriate Events
Facilitate economic evaluation
Gilt purchases need to maintain sow herd inventory
Gilt retention and sow survival rate

Gilt pool inventory

Entry to service or removal intervals

Gilt farrowing rate Gilt litter size

Pre-process economics of the gilt pool: negotiated price of each gilt (purchase price or internal transfer cost)

In-process economics depends on: 1. Entry-to-service interval for gilts that are served. 2. Retention rate and entry-to-removal interval for gilts that are culled or die.

After process economics depends on the reproductive performance (farrowing rate and litter size), survival (removal rate) and flow (service targets).


Slide 5

Data Entry in the Gilt Pool

Individual animal identification is preferred Coding non-routine data
Use available variables in Records Details

Critical reproductive events and dates

Entry Start boar exposure
Not commonly done Often the same as entry

Heat-no-service First service

Individual animal identification enables calculation of key intervals such as entry-to-first service and entry-to-removal, monitoring treatment compliance and evaluating reproductive performance with regard to treatments and animal attributes (such as weight or age).

Record Details can be used to enter special codes or designations such as study treatments.

Ideally, all heats are recorded whether or not the gilt is served. Recording when boar exposure starts enables evaluation of heat stimulation and heat detection activities.


Slide 6

Data Entry in the Gilt Pool

Intervention (treatment) events
PG 600: single date Matrix
Single date (begin or end) Both begin and end dates

Animal fitness events

Age at entry or birth date Weight
Weight target Growth rate target

Back fat

Recording intervention events enables compliance monitoring. Activities such as heat stimulation and detection can be evaluated if failure to exhibit heat results in a treatment such as PG 600.

Animal fitness emphasizes the point that attributes other than reproductive events need to be considered when servicing gilts. In this workshop, weight at service and growth rate are the primary attributes that are used to evaluate gilt fitness at their first service.


Slide 7

Retrieving Data: Routine Reports

Performance Monitor
Entry to service interval Inventories

Farrowing Rate Report

Overall service target

The PigChamp Performance Monitor provides gilt only information regarding entry-to-service intervals and inventory information including average inventory, beginning and ending inventory, gilts entered and gilts removed.

The Farrowing Rate Report is the main report used to determine if overall service targets are achieved.


Slide 8

Retrieving Data: Routine Reports

Parity Distribution Report
Reproductive Performance Retention Rate

Removal Report
Retention rate Removal type: cull, death Reasons for removal

The Parity Distribution Report presents reproductive performance information on gilts and inventory information that can be used to calculate retention rates.

Removal Report information provides retention rate information and the reasons why animals are removed.


Slide 9

Retrieving Data: Database Application

Useful canned routines
Tally (frequency distribution)
Averages can be misleading Compliance monitoring capability Entry to PG 600 intervals Entry to service interval is a good example

The Database Application menu in PigChamp enables generating reports that present data as distributions rather than totals or averages. For example, the Tally routine can be used to evaluate entry-to-service intervals. In the Performance Monitor, entry-to-service intervals are presented as averages for those gilts that were served during the time period presented in the report. The Tally provides a frequency distribution of when the gilts are served, which can be used to determine the effectiveness of heat stimulation programs (boar stimulation, gilt responsiveness) and identify compliance failures. A compliance failure example could be finding gilts with entry-to-service intervals that exceed 60 days even though the protocol states that gilts should be served or removed by 60 days after entry.


Slide 10

Retrieving Data: Database Application

Extract raw data using List Data Routine
Often for special analyses
Comparing treatments During implementation phase

Use Excel, Access and/or statistical packages for additional data organization and analysis

In some situations, extracting raw data from PigChamp is useful for more detailed analyses. One line of raw data includes the animals ID, the event, the event date and information related to the event. I prefer to save or export the data as a text file. The next steps are to open the data in Excel and use Excel to organize and clean up the data. Further statistical analyses can be done within Excel or the data can be transferred to statistical programs with more advanced features such as SAS, JMPIN or Statistix.


Slide 11

Compare data against defined process
Compare with plan (compliance)
Administer PG 600 on day 25 after the start of boar exposure Feed Matrix for 14 days

Within process parameters

Percentage of gilts in heat after 25 days of boar exposure Percentage of gilts treated with PG 600 Gilt retention rate

In this example, heat stimulation and detection is done for 25 days. Gilts that exhibit heat are then served on their second or third heat depending on weight. Gilts that fail to cycle by 25 days are considered opportunity gilts and are treated with PG 600. If not in heat within 7 days, they are culled. Those exhibiting heat are placed on Matrix to provide a minimum heat-no-service-toservice interval of at least 25 days and to synchronize the gilts to meet service targets.

The expected percentage of gilts in heat after 25 days is >80% and the expected percentage of gilts that cycle after PG 600 treatment is >90%. Gilt retention rate is based on the percentage that fail to cycle (<2%) and those culled for other, non-reproductive reasons (<3%).


Slide 12

Compare data against defined process
Process output
Farrowing rate Litter size Survival

Example process/operation (see next page)

Output variables
Synch = synchronization or ability to time services Fertility refers to both farrowing rate and litter size

Once the gilt is served, the gilt pool process ends so reproductive performance and survivability is measured after process. Other presentations within this workshop present farrowing rate and litter size information that demonstrate what is expected with the various options that are presented on the next page.


Slide 13

Gilt Breeding Flow Chart

Heat Stimulation and Detection for 25 days

Natural heat (>80%)

No heat, not needed, culled

Opportunity gilt, given PG 600 (20%)

No heat, Culled (<10%)

Served on 1st heat (0%)

Skip cycle (all)

Induced Heat (>90%) Matrix prior to service on 2nd heat

Matrix prior to service

No Matrix prior to service

Served on 2nd heat, No Matrix

Served at induced heat (0%)

- synch, + fertility

+++ synch, +++ fertility

+/- synch, +++ fertility

+/- synch, ++ fertility

+++ synch, +++ fertility

+++ synch, +/- fertility

This diagram describes the flow and expected outputs of an example gilt breeding program. All potential options are presented but in this example no gilts should be served at their first estrus. In some of the boxes, the expected rates are presented. The outputs, degree of synchronization and fertility (in this case farrowing rate and litter size), are qualitatively graded.


Slide 14

Service Targets
Statistical Process Control Charting
Voice of the process (data) versus voice of the customer (target)


Statistical Process Control (SPC) involves a number of concepts that can be used to evaluate the performance of any process. Charting is a key activity in SPC programs. The chart above presents services per week (Sunday to Saturday) over a 3 year period of time. The y axis presents the number of services and the x axis is the week. A fairly rigid heat no service program was implemented around week 100. Soon after, their was a PRRS outbreak that resulted in three weeks of low numbers followed by two weeks of higher numbers. Thereafter, the services per week has been much more consistent than before the program was implemented. Overall, the six sigma (normal) range of services per week was 110 to 236. Since week 110, the range is 135 to 194 and since week 130, the range is 142 to 188.


Part 3: Proof of principle


Experience with the Pipestone Project

Joaquin Sprke, Jennifer Patterson, Eduardo Beltranena & George Foxcroft



Pre-Selection 1 will occur at the time the gilts leave the hot room. Gilts will be examined and any gilts with conformation and health problems will be culled. Any gilts with clearly stunted growth will also be culled. Gilts will be re-grouped and re-penned by weight class, estimated visually.





Pre-Selection 2 will occur at the time of tattooing (approximately 160 days of age). Gilts will be examined and any gilts with conformation and health problems will be culled. Any gilts with obvious stunted growth will also be culled. A minimum of 155 gilts per four week group are needed to enter stimulation, representing at least 110% of gilts finally expected to be selected as recorded heat-noserve females (HNS).


At approximately 160 days of age (on a Monday), gilts will be moved daily to the designated heat-checking pen/stimulation area. When a group of gilts first enters the stimulation/heat check area, they are immediately permitted FENCELINE contact with the two or three boars to facilitate heat detection; the back-pressure-test (BPT) is performed by the trained technicians to determine if any gilts are exhibiting the standing reflex. Any gilts with a detected HNS are marked and recorded as 1st Heat-No-Serve (1HNS) on preprinted puberty stimulation record cards. Gilts in estrus are removed from the stimulation area and placed in an adjacent holding pen. This will avoid the necessity of preventing boars from attempting to mount estrous females during the stimulation period.


The pen group of 20 gilts will be equally divided into two groups of 10 gilts and placed on either side of the boar stalls, providing a boar to gilt ratio of no greater than 13:1 during periods of direct contact.


Only mature, vasectomized boars with high libido and acceptable temperament are used for both fenceline and direct contact. Under direct control of a stockperson, and with dry flooring, boars are permitted to mount, but not penetrate, a gilt in heat. Records are required that all boars are hand-collected at least once a week to maintain high libido.


A single vasectomized boar is then permitted to have full contact with the group for approximately 10 minutes daily. Physical signs of pending estrus, such as redness, degree of swelling, mucosal discharge from the vulva, and proceptive (soliciting) behavior towards the boar, are recorded daily during the stimulation period using an agreed scoring system (1 to 5 scale). Only gilts that are exhibiting full standing estrus will be scored at 5 (a recorded HNS). Heat-no-service will be recorded using individual gilt ID on a pre-printed puberty stimulation card.


Any gilt recorded in standing heat will be held back at the end of that stimulation period and weight at HNS recorded


Based on this weight, gilts will be designated to be bred at second or third estrus. Weight at breeding is targeted to be at least 135 kg (300 lb). Any gilt weighing at least 260 lb or greater at first HNS will be relocated to a stall in the pre-breeding area. Stalled gilts should already be organized by designated breeding week Gilts not at target weight at first HNS are designated to be bred at third estrus, remain in group pens for another cycle. After a second recorded HNS, gilts will then be moved from the pen group and placed in gilt stalls.


Physical signs of pending estrus, such as redness, degree of swelling, mucosal discharge from the vulva, and proceptive (soliciting) behavior towards the boar, are recorded daily during the stimulation period using an agreed scoring system (1 to 5 scale). Only gilts that are exhibiting full standing estrus will be scored at 5 (a recorded HNS).


At d14 all gilts that do not have a recorded HNS are mixed and relocated to new pens. Gilts can be further sorted visually by weight at this time.


Any gilt not recorded as HNS after 23 days can be used as opportunity gilts if anticipated breeding targets will not be met. As needed, the heaviest opportunity gilts available are treated with PG600 and continued to be checked daily for standing heat. The required number of PG600 treated opportunity gilts, with PG600-induced heats, will be added to heat-no-serve gilts already in the gilt stalls, to create the required breeding weeks. Before gilts are administered PG600, heat check sheets should be carefully examined. Any gilt showing signs of pending estrus (vulva scores 3-4), but was not recorded as HNS, may likely be close to ovulating and should not be given PG600. If treated with PG600 at this time gilts are at risk of developing cystic, anovulatory follicles. Another possibility is that gilts may be showing silent heats (showing vulval and behavioral signs of entering pro-estrus, but not displaying a full standing heat during regular heat checking). These gilt should not enter the breeding herd.


After PG600 injection, gilts remain in pen groups and must receive daily boar contact in the stimulation pen for an additional 5 days (d23-d28). All signs of pending estrus are recorded. If gilts exhibit a HNS they are considered to be select opportunity gilts and placed in stalls within an appropriate breeding week.



All gilts that are placed in stalls should have daily fence-line contact with active, mature boars until mated and should be fed as close to appetite as possible, using two feed drops per day. This will also facilitate administration of Matrix in the feed drops the previous evening and rapid consumption of a relatively small volume of Matrix impregnated feed the next morning. Floor troughs should be free of water at the time that any Matrix dosed feed is delivered.


Gilts will receive MATRIXTM starting Friday, approximately 20 days before their expected breeding date, regardless of the day of the estrous cycle that treatment starts.




Use of Matrix allows very tight and precise breeding weeks to be established. Even four to five weeks in advance, the staff know that gilt breeding targets will be met AI coincides with optimal semen delivery and fresher extended semen is used for breeding Standing heats in the gilts stalled within a breeding week is virtually 100% guaranteed Breeding staff are therefore spending their time breeding and not heat checking


Appears that approximately 3-4%more gilts induced to show HNS with PG600 are not bred at the predicted heat after Matrix withdrawal, compared to boar induced , naturally cyclic gilts. However, less than 10% of gilts designated ay HNS to be part of successive breeding groups are not bred. This compares well with around 15% of gilts in less intensive GDU programs being declared eligible to be bred but never have a farrowing record.


Again some significant difference seems to be emerging in the pattern of days on which natural, boar induced HNS gilts show estrus after Matrix withdrawal, compared to PG600 induced HNS. Pregnancy rates and farrowing rates are above targets set


Litter size Born Alive is running more than 1.5 pigs above target Good uniformity of litter size born is being seen. Experience suggests that greater variability in litter size born would occur if gilts were bred at the first PG600-induced heat.





Using MATRIX and P.G. 600 in a commercial swine operation - impact on gilt pool size
Matt Ackerman, DVM Swine Veterinary Services, PC


When Matrix was brought to the market place we at Swine Veterinary Services had many discussions about how to best recommend the product in different farm situations. We know that most of the farms we work with do not have a heat no service program. This is a decision by farm management to trade larger giltpools for less up front labor. Of course not having a Heat No Service program always leaves us wondering what went wrong if breeding targets are not met.


As we discussed different programs, we felt the most cost effective approach was to tighten up the giltpool so that clients arent accumulating larger and larger pools of non-cycling animals. The goal was to reduce labor by focusing them on smaller groups of eligible animals.


The way this would look on a calendar. 1. 2. 3. 4. 5. Put the gilts into crates one week prior to going on Matrix so they get use to eating in a crate. Put the gilts on Matrix for 14 days (from a Wednesday to a Tuesday) Heat Check the gilts on the following Monday to Friday. On Friday, treat all non-cycling gilts with PG600 Heat Check the gilts for one more week.

Cull the remaining gilts. It is our assumption that if a gilt doesnt cycle off of Matrix, but does cycle off of PG600, it is because she was not cycling when she was put onto Matrix.


It really doesnt matter how the barn is laid out. What matters is that you have a system for getting the gilts into the crates and making sure that the right gilts get a full dose of Matrix every day for 14 days.


The Matrix literature itself set expectation very high. When our clients read that 85% of gilts will come into heat in a 6 day period, they expect 85% consistently. What was interesting is that most of our owner operators and 4-Hers got along quite well with Matrix. Unfortunately, a number of our larger operations struggled with making it work.


A number of our larger operations reported results more like this. Unfortunately, those results were often hard to substantiate. It seemed that while our clients are very good at PigCHAMP record keeping, they had trouble tracking their Matrix usage and results. The perception was that the results were variable and they were spending money on product. They were quick to discontinue Matrix use without getting to the bottom of the issue why werent gilts cycling?


As we started down a troubleshooting path, there were a number of issues that came to light. We quickly figured out that if we were going to get compliance at the farm level, we were going to have to have a better system for tracking the results.


We worked with our clients to design a record keeping system to track each group of gilts coming into the farm. We used PigCHAMP as well as a spreadsheet kept at the farm. Here is the whole spreadsheet. It is actually broken down into 3 parts. I will walk you through each part on the next 3 slides. The important thing to see on this slide is that the start breeding dates run down the left hand side of the screen. This format is repeated in each of the 3 sections. Each row is an individual set of gilts all the way through the Matrix, Matrix & PG-600 and culling program. Section 1 tracks the number of animals that cycle off of Matrix Section 2 tracks the number of animals that cycle off of Matrix followed by PG-600 10 days post Matrix Section 3 attempts to track the percent still pregnant after RTU of both treatments combined


As you can see, the breeding dates are on the left. The number of animals treated in column D Row 9 Columns E through L calculate the percent of cycling animals that stand in heat any given day. On the initial program that we were using we had 79% of the gilts cycle back in a 10 day period and 93% of those were Pregnant at RTU (Real Time Ultrasound) for what we calculate as a 72.9% (79% x 93%) gilt utilization off of Matrix (percent of gilts treated with Matrix that were pregnant at RTU). Remember conception rate is over 90%. Employees were so focused on the Matrix claim of 85% of gilts coming into heat, that we found it important to track the pregnancy rate and keep them focused on end result not just how many cycled.


In the next section, we track how many animals didnt cycle per group and end up being treated with PG600. We then track the percent of gilts cycling on an given day. On the initial program that we were using we had 66% of the gilts cycle back in a 10 day period and 90% of those were RTU positive for what we calculate as a 59.2% gilt utilization off of Matrix (percent of gilts treated with Matrix that were pregnant at RTU). Once again, please remember conception rate is over 90%. It is our assumption that if a gilt doesnt cycle off of Matrix, but does cycle off of PG600, it is because she was not cycling when she was put onto Matrix.


On the 3rd and final part of the spreadsheet we tracked the percent of gilts that were culled due to the fact that they never cycled in the 14 day window. On the initial program that we were using we had 7% of the gilts culled because they were noncycling in the 14 day window. Historically gilts culled for not cycling [Percent of Gilts entered and removed (no service) from the Productivity Analysis] had been a double digit number after 42 days in the giltpool. We also tracked the total utilization rate which is the percentage of gilts that are pregnant at RTU off their first service regardless if they received Matrix or both Matrix and PG600. At this time it was 82%. This number doesnt include gilts that breed back on their second cycle. As I stated, we still wanted to get 85% of the animals to cycle off of Matrix.


We had several weeks when a higher percentage of gilts were reported not to cycle. The employees wanted to blame the weekend person for not giving the Matrix or not getting the right dose into them which could cause them to go cystic. When we did postmortems on these gilts, we found that they were cycling normally. This set of ovaries had both CLs (Corpora lutea) and follicles.


This set of ovaries had CLs (Corpora lutea) and CHs (Corpora hemorrhagica).


This set of ovaries had follicles.


All three sets had normal cycling ovaries. Conclusion Focus on Heat detection


We used PigCHAMP to verify the on farm data. We coded No Matrix (only for animals that came into heat the first couple of days of Matrix and were taken off the product), Matrix and Matrix & PG 600 into the origin field on the individual sow records. Any animals that were bred previous to the start of the program were not coded as any of the three, but are still included in the Total column. The big concerns before starting the program were: 1. Will Matrix cause less fertile animals to come into heat and thereby hurt our born live and farrowing rate? No, they were both better with Matrix. 2. Will Matrix & PG 600 cause less fertile animals to come into heat and thereby hurt our born live and farrowing rate? No, it is better to have those animals bred than culled.


Before Matrix our farms struggled a great deal to consistently achieve Mating Target. Therefore, they over bred to compensate for short weeks.


Being on a Matrix program has allowed us to consistently achieve service target. It has also allowed us to identify bad weeks in RTU and ramp up services accordingly.


We therefore readjusted our targets, bringing the top end down so that we can increase our average wean age.


Our entire gilt process has become more efficient. Instead of checking 120 per week to find 25 in heat, we know just check 32. Maintaining a smaller gilt pool leads to a lower replacement rate.


The cost calculations are pretty straightforward. Because they are so obvious, a lot of times that is the only part people want to focus on.


Calculating the savings is more difficult because it is hard to get a solid handle on a number of the daily costs to which a system has become accustom. I choose to spell out my assumptions and let the client tell me what number they disagree with. In this case, I chose to use a savings based approach versus a increase in income based approach, due to my clients situation. Watch out for client assumptions. Are home made gilts really free? On this system they do internal multiplication and not much value is put on gilt premium. They are not as interested in more pigs weaned above a certain goal They are more interested in having 1050 to 1150 pigs produced per week that have an 18 day wean age. How much less space will you need? Can you do anything with that extra space? They have been able to reduce the amount of rented off-site gilt developer, since they are now developing gilts for 3 weeks within the sow farm with the space they have gained from reducing the gilt pool.


The assumptions from the previous screens are then calculated into an ROI. What value do you put on the positive side effects of this process?


Our Management team continues to challenge us to increase our level of success with this product. We continue to refine our process to try to maximize our gilt utilization. We continue to focus on boar exposure and heat detection. One thing that became clear as we continued to work with the crew is that they had not been doing heat detection and boar exposure while animals were on Matrix It is important that animals on Matrix receive daily boar exposure so that they are receiving that pheromone stimulus. It is also important that gilts have been trained to the boar exposure and heat detection process when they come off of Matrix.


Also as we reviewed our process we found that using Heat No Service the week that gilts were first put into crates helped reduce the number of animals on Matrix. Our utilization number are calculated by taking the number of animals pregnant at RTU divided by the number of animal put on that treatment. The big improvement was in Utilization off of PG-600. The reality is that less animals needed this treatment since they were found on the front end of the process.


On the initial program that we were using we had 79% of the gilts cycled back in a 10 day period and 93% of those were RTU positive for what we calculate as a 72.9% (79% x 93%) gilt utilization off of Matrix (percent of gilts treated with Matrix that were pregnant at RTU). On this more aggressive approach we found that 88% of the animals are getting serviced but only 82% of them are staying bred. This gives us a 71.4% utilization off of Matrix. Slightly less utilization despite a higher percent of services.


On the initial program that we were using we had 66% of the gilts cycled back in a 10 day period and 90% of those were RTU positive for what we calculate as a 59.2% gilt utilization off of Matrix (percent of gilts treated with Matrix that were pregnant at RTU). Once again, please remember conception rate is over 90%. On our more intense Boar Exposure & Heat Detection program we had 82% of the gilts cycled back in a 10 day period and 92% of those were RTU positive for what we calculate as a 75.4% gilt utilization off of Matrix. Much more cost effective results.


On the initial program that we were using we had 7% of the gilts be culled because they were non-cycling in the 14 day window. Now it is down to 2%. None of these numbers include the animals that are now being Heat No Serviced for the one week period prior to going onto Matrix.



Our basic recommendation for Matrix use.