Weaning age and post-weaning nursery feeding regime are important in improving the performance of lightweight pigs

Abstract

The aim was to investigate the effect of weaning age, weaning weight, and nursery feeding regime on post-weaning performance. The focus was on pigs weaned light, as they may be better off when weaned at a later age and/or offered a specialist nursery feeding regime. Piglets (n = 1,448) from one farrowing batch of 110 sows that farrowed over 2 wk were individually weighed and their morphometric measurements were taken at birth. Pigs were weaned on the same day, but variation in birth date resulted in variable weaning ages (mean age day 34.1, SD = 2.5). The youngest 50% at weaning were classified Y and the oldest 50% as O; within an age class, the lightest 50% were classified L, the heaviest 50% as H, and housed accordingly. Pigs were individually weighed at weaning, 7 and 15 wk post-weaning. At weaning, Y were 6 d younger and 1.4 kg lighter than O pigs, whereas L were 3.2 kg lighter than H pigs. Pigs were randomly allocated to a 3-stage superior (SUP) or control (CON) nursery feeding regime, with SUP pigs having a 65% greater allowance (on a kg/pig basis) of the first and second stage feeds than the CON. Pigs weaned Y had a higher mortality rate from weaning to 7 wk post-weaning than pigs weaned O (9.14% vs. 4.98%; P = 0.046). As expected, age and weight significantly (P < 0.001) affected performance to both 7 and 15 wk post-weaning: at 15-wk pigs weaned Y were 5.5 kg lighter than pigs weaned O; pigs weaned L were 9.0 kg lighter than H pigs. It was estimated that pigs weaned YL needed ~4 d more (P = 0.018) to reach 60 kg BW than pigs weaned OL. Feed intake was not affected by feeding regime, age and weight, or their interactions. Performance was not affected by feeding regime (P > 0.05), but was affected by the weight × feeding regime interaction (P = 0.044) to 7 wk post-weaning: L pigs on SUP were 1.2 kg heavier than L pigs on the CON regime; this was not the case for the H pigs. Performance up to 7 wk post-weaning was positively associated with birth weight to cranial circumference ratio and weaning weight (P < 0.05) for both YL and OL pigs; for the latter, additional performance predictors were weaning age (P = 0.044) and feeding (P = 0.027). Improved growth for L pigs up to 7 wk post-weaning could be obtained by a greater allowance of the nursery diets. However, weaning at a later age benefitted the performance of L pigs to a common BW, suggesting that this might be a strategy with longer term benefits.

Introduction

The management of lightweight piglets, which have become more common with the increase in the litter size of modern sow genotypes, remains a challenge (Ocepek et al., 2017). Post-weaning, these pigs have a higher mortality risk, grow slower, and consequently need more time to reach slaughter weight than their heavier counterparts (Collins et al., 2017). Farmers try to manage this challenge either by offering lightweight pigs a specialist nursery feeding regime, such as using readily digestible/high-quality nutrient sources, synthetic amino acids, and palatable ingredients (Beaulieu et al., 2010; Collins et al., 2017), or by weaning them at a later stage, for example, by “split weaning” (Pluske and Williams, 1996; Vesseur et al., 1997; Abraham et al., 2004). The latter involves weaning the heavier piglets and leaving the lightweight piglets on the sow for a longer period of time.

There are good reasons why either of the above strategies, or their combination, might work. Lightweight pigs have a more immature digestive system (Cranwell et al., 1997; Pluske et al., 2003) and a lower feed intake post-weaning (Magowan et al., 2011) than heavyweight pigs. Prolonging milk intake may enhance gut maturity (Fanaro et al., 2003; Schack-Nielsen and Michaelsen, 2007), thus reducing post-weaning growth check. In addition, lightweight pigs may have different nutrient requirements, as a result of differences in gut maturity and stage of development, compared with heavy pigs of the same age. Previous research has suggested that feeding lightweight pigs for a longer period of time on a specialist nursery feeding regime, may influence their subsequent performance (Magowan et al., 2011; Douglas et al., 2014). Usually, these effects have been investigated only on the basis of weaning weight, without considering weaning age and the interactions between the two.

Because both strategies for managing lightweight pigs are associated with significant costs and disruption of the flow of pigs from a batch, it is imperative to know what their consequences on long-term performance would be. The objectives of this experiment were to investigate 1) the consequences of delayed weaning, increased allowance (on a kg/pig basis) of a nursery feeding regime, and their combination on the performance of lightweight pigs to slaughter, and 2) whether all lightweight pigs are able to benefit equally from the above strategies. We have previously shown that not all lightweight pigs are alike and that the ones having a reduced post-weaning performance are those weaned light and disproportional, that is, piglets born long and thin and/or with a greater head circumferences in relation to birth weight (Douglas et al., 2016; Huting et al., 2018). The latter may be related to intrauterine growth restriction (IUGR), which is often characterized by a dolphin-like forehead (Amdi et al., 2013). Here, we investigated whether these morphometric predictors interact with the management of lightweight piglets, that is, the time of weaning and the allowance of the nursery regime.

Materials and Methods

Experimental Design

The design was a 2 × 2 × 2 factorial with 2 ages (young vs. old), 2 weaning weights (light vs. heavy), and 2 nursery feeding regimes (control vs. superior) as post-weaning treatments. The experiment took place on a Large White × Landrace sow, farrow-to-finish commercial farm in Great Britain from February to July 2017, and the experiment was approved by the Animal Welfare and Ethical Review Body (AWERB project ID 419) of Newcastle University. Piglets (n = 1,448) born in a single farrowing batch were followed from birth to 15 wk post-weaning.

Animals, Housing, and Management

The farrowing batch consisted of 110 sows of different parities: 22 gilts and 88 sows (mean parity 4.38, SD = 2.39; parity range 2 to 13). Sows and gilts were placed in farrowing pens with a crate and a heated creep area (floor heating) and farrowed over a 2-wk period. The majority of cross-fostering took place within the first 4 d of life to standardize litter sizes to the number of functional teats and to create litter uniformity (AHDB Pork, 2017). Fourteen sows (12.7%) were weaned shortly after birth and were replaced by other sows due to space restrictions or poor performance. The latter resulted in that 53.9% of the piglets were eventually cross-fostered. Piglets had ad libitum access to a 2-phase pelleted creep feed regime provided in a hopper from 10 d of age (day 9.78, SD = 1.67) to weaning; diet 1 of the nursery feeding regime (22.0% CP, 10.4 MJ NE/kg diet) was provided during the last week prior to weaning (see Table 1 for diet specifications).

Table 1.

Ingredient composition, on an as-fed basis, and chemical analysis of the post-weaner feeds used. Pigs were randomly allocated to either a high feed allowance or a standard nursery feeding regime^1,2

Diet	1	2
Ingredient g/kg
Micronized barley	150	50.0
Wheat	218	450
Micronized wheat	50.0	50.0
Micronized oats	100	—
Fishmeal	72.5	25.0
Soya Hypro	127	210
Full-fat soybean	30.0	25.0
Pig weaner vitamin/trace element supplement3	5.00	5.00
Dried skim milk powder	40	—
Whey	146	34.7
l-Lysine HCL	3.17	4.93
dl-Methionine	1.98	2.12
l-Threonine	1.91	2.39
l-Tryptophan	0.38	0.24
l-Valine	0.70	1.28
Pan-Tek Robust CB	0.15	0.15
Sucram	0.10	0.10
Benzoic acid	5.00	5.00
Limestone flour	1.79	-
Dicalcium phosphate	6.19	15.1
Salt	-
Binder (LignoBond DD)4		4.17
Sodium bicarbonate		3.57
Soya oil	28.2	11.0
Analyzed composition, % as fed
CP	22.1	20.2
Crude fiber	2.10	2.40
Moisture	9.70	11.2
Ash	5.70	6.20
Calculated composition, % as fed or as specified
NE, MJ/kg	10.4	9.39
Calcium	0.75	0.71
Phosphorus	0.68	0.68
Lactose	12.5	2.50
Lys	1.54	1.40
SID Lys5	1.40	1.28
Met	0.61	0.52

¹All diets were identical between the 2 feed allowances. However, pigs fed the superior feeding regime had a 65% greater allowance of the first and second stage feeds (on a kg/pig basis) before moving to the grower regime than pigs fed the standard feeding regime.

²Diets were supplied by Primary Diets, ABAgri, Ripon, North Yorkshire, UK.

³It provided per kilogram of complete diet 11,500 IU of vitamin A, 2,000 IU of vitamin D₃, 100 IU of vitamin E, 4 mg of Vitamin K, 27.5 µg of vitamin B₁₂, 15 mg of pantothenic acid, 25 mg of nicotinic acid, 150 µg of biotin, 1.0 mg of folic acid, 160 mg of Cu (CuSO₄), 1.0 mg of iodine (Ca (IO₃)₂), 150 mg of Fe (FeSO₄), 40 mg of Mn (MnO), 0.25 mg of Se (bone morphogenetic protein), and 110 mg Zn (ZnSO₄).

⁴Borregaard LignoTech, Sarpsborg, Norway.

⁵SID, standardized ileal digestible.

All pigs were weaned on the same day at a mean age of 34.1 d (SD = 2.5), irrespectively of their date of birth. At weaning, pigs were moved to the fully slatted weaner facilities making groups of 60 pigs per pen with stocking density consistent with UK legislations. Pigs were grouped on the basis of their weaning age, weaning weight, and nursery feeding regime. Pigs were fed using an automatic rationed liquid feeding system with one feeder per pen pair. Pigs remained in the same group and room up to 7 wk post-weaning; thereafter, pigs were mixed to create smaller groups depending on the pen size (range 20 to 35 pigs per pen), ensuring a similar stocking density between pens and moved to 1 of the 3 fully slatted finisher rooms on site.

Experimental Procedures

Pre-weaning performance

Piglets were individually weighed to the nearest 1 g within 12 to 24 h after birth (BiW). At the same time, additional morphometric measurements were taken and piglets were individually ear tagged (Dentag, Toptags, Kelso, UK) for identification purposes. The morphometric measurements taken from each individual piglet were as follows: crown to rump length (CRL, cm), snout to crown length [head length (HL), cm], abdominal circumferences (AC, cm) taken at the anterior side of the umbilicus, and cranial circumferences (CC, cm). Additional variables were created, such as ponderal index [PI; BiW/(CRL)³, kg/m³], body mass index (BMI; BiW/(CRL)², kg/m²), BiW:cranial circumferences (BiW:CC, kg/cm), and snout to crown length: BiW (HL:BiW, cm/kg; Douglas et al., 2016; Huting et al., 2018). In addition, piglet individual BiW was expressed in relation to birth litter average (relative BiW) by dividing piglet BiW by the mean BiW of the birth litter (Paredes et al., 2012; Huting et al., 2018), as it appears to be an important parameter influencing light piglet pre-weaning performance (Huting et al., 2018).

From the piglets that died pre-weaning, additional recordings were taken where possible, including the cause of death (e.g., nonviable, starvation, anemic, crushed, savaging, meningitis, and diarrhea), BW, and foster sow. Piglets were retagged (Suretag flag, Dalton tags, Newark Nottinghamshire, UK) at 20.7 (SD = 2.8) d of age. All pigs were weaned on a Thursday morning (day 34.1, SD = 2.5), but piglets were individually weighed 2 to 3 d prior to weaning (day 31.7, SD = 2.6); this was done randomly, to spread the workload and enable allocation to treatments. Piglets were given an oral suspension of Baycox (Bayer plc, Reading, UK) at days 3 to 5 of life (day 3.89, SD = 1.45). At the same time, litter details (sow ID, piglet ID) were recorded, as by this time, the majority of cross-fostering had taken place.

Nursery feeding regime

The 3-phase nursery feeding regime (Primary Diets, ABAgri, Ripon, North Yorkshire, UK) consisted of diet 1 (22.0% CP, 10.4 MJ NE/kg feed), diet 2 (20.2% CP, 9.39 MJ NE/kg feed), and a “grower” diet fed from weaning to 7 wk post-weaning; a “finisher” diet was fed between 7 and 15 wk post-weaning. All diets (Table 1), including the grower and finisher diet, met or exceeded the nutrient requirements for pigs of this size and age (NRC, 2012) and were identical between the experimental treatments. However, diet 1 and diet 2 were fed in different quantities. Pigs allocated to the higher nursery feeding allowance (SUPERIOR) regime were fed a 65% greater allowance of diet 1 and diet 2 (3.80 and 3.33 kg/pig, respectively) compared with pigs allocated to the control (CON) regime (2.30 and 2.00 kg/pig, respectively). After this, an extra 2 kg of diet 2 was mixed with the grower diet to ease diet transition of both the CON and SUPERIOR pigs. The pigs were fed ad libitum until they consumed the amount of feed allowance they were allocated to. The amounts of feed in the CON regime were based on the normal feeding strategy of the pig unit. The difference between the 2 regimes was set to 65% to ensure the economic viability of the SUPERIOR regime, which was around 50% more expensive per pig compared with the CON regime ($6.84 vs. $4.62 per pig).

The cost of the feeds (£/ton June 2018; exchange rate June 2018 £1 = US$1.33) was provided by the feed company: diet 1 (£688/ton, $ 915/ton), diet 2 (£ 474/ton, $ 630/ton), and grower diet (£240/ton, $ 319/ton). This was used to calculate the following: 1) total feed cost per pig, 2) total cost per kilogram gained (weaning to 7 wk post-weaning), and 3) the margin over feed (MOF; as per Douglas et al., 2014):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{M}\mathrm{O}\mathrm{F}=[(\mathrm{G}\mathrm{a}\mathrm{i}{\mathrm{n}}_{\mathrm{k}\mathrm{g}}\times \mathrm{P}\mathrm{r}\mathrm{o}{\mathrm{p}}_{\mathrm{c}\mathrm{a}\mathrm{r}\mathrm{c}\mathrm{a}\mathrm{s}\mathrm{s}})\times \mathrm{P}\mathrm{r}\mathrm{i}\mathrm{c}{\mathrm{e}}_{\mathrm{k}\mathrm{g}\mathrm{c}\mathrm{a}\mathrm{r}\mathrm{c}\mathrm{a}\mathrm{s}\mathrm{s}}]-\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}{\mathrm{l}}_{\mathrm{f}\mathrm{e}\mathrm{e}\mathrm{d}\mathrm{c}\mathrm{o}\mathrm{s}\mathrm{t}}}\end{array}}$$

where Gain_kg is the kilogram gain between weaning to 7 wk post-weaning, Prop_carcass the proportion of carcass weight from live weight (0.75), Price_kgcarcass the price per kilogram carcass weight (as of June 2018, $2.62; AHDB Pork), and Total_feedcost the total feed cost per pig.

Post-weaning performance

A total of 960 pigs were selected from 1,135 healthy (e.g., free from lameness, viable, >4 kg) individuals available at weaning. On the basis of weaning age, pigs were split into 2 groups: the bottom 50% were classified young (Y, < mean weaning age; day 32.4, SD = 0.3) and the rest classified as old (O, > mean weaning age; day 35.9, SD = 0.3). Within a weaning age class, the lightest 50% (≤ mean weaning weight for Y or O) were classified light (L), the heaviest 50% pigs (> mean weaning weight for Y or O) were classified heavy (H). The latter was done between pigs of the same date of birth. It should be noted, however, that a significant proportion of pigs was born on the same day and weaned at day 34 (n = 329). As the experiment aimed at assessing the effect of weaning age on post-weaning performance and days can make a substantial difference in the subsequent responses to weaning, pigs weaned at day 34 were excluded from subsequent statistical analysis. By doing so, the remaining Y pigs were weaned at 31.8 (SD = 0.2; range 29 to 33) d and the remaining O pigs were weaned at 37.5 (SD = 0.3; range 35 to 41) d of age (see the “Statistical Analysis” for more details).

The day before weaning, pigs were marked using different color markers and different markings indicating their post-weaning treatments (age, weight, and nursery feeding regime) on the following morning (Thursday). Within each age and weight classes, pigs were allocated to one of the 2 different nursery feeding regimes: SUPERIOR or CON, while balancing for weaning age, weaning weight, and sex. Each of the 8 treatment groups comprised 2 parallel pens that shared one feeder.

Feed intake was recorded on a weekly basis per pen pair. Irrespective of weaning age, pen weights were taken at 1, 3, 4, and 5 wk post-weaning using a platform weighing scale. Individual BW were taken at 2, 7, and 15 wk post-weaning, 1 wk before the biggest pigs of the batch were sent to slaughter, after which the trial finished.

Statistical Analysis

The effects of weaning weight, weaning age, nursery feeding regime, and their interactions on post-weaning performance were evaluated using the PROC MIXED procedure of SAS version 9.4 (SAS Inst. Inc. Cary, NC). The residual variance of the data was tested for normality using the UNIVARIATE procedure. Data were expressed as least-square means, with approximate standard errors of the differences of means unless stated otherwise. Statistical significance was assessed at the 5% level and tendencies were set at 10%. Several covariance structures were tested, but variance components resulted in the lowest Akaike information criteria (AIC) and were selected for the final model. Sow parity effect could not be fitted in our statistical models because of the unbalanced or small number of sows in certain parities.

Post-weaning performance

Post-weaning mortality was evaluated with a chi-square test. The statistical unit between weaning and 7 wk post-weaning was pen mean nested within pen pair and was analyzed excluding the pigs that were weaned on day 34. As only one feeder was available per treatment, the statistical unit for feed intake and MOF was pen pair. As a result of mortality and pig removals due to illness, post-weaning ADFI was corrected using the following formula over a given period:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{D}\mathrm{F}\mathrm{I}=\frac{(\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}{\mathrm{l}}_{\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}})}{(\mathrm{T}\mathrm{i}\mathrm{m}{\mathrm{e}}_{\mathrm{p}\mathrm{i}\mathrm{g}}/24\mathrm{h})}}\end{array}}$$

where Total_intake is the total amount consumed per pen pair (g) and Time_pig the total time pigs reside in the pen (h).

Only individual pig data (i.e., BW 2 to 3 d prior at weaning, 2, 7, and 15 wk post-weaning) could be used for the analysis when pigs weaned at day 34 were excluded. The statistical unit was pen mean. Pen means were based on the number of Y or O pigs that were in the pen. A WEIGHT statement was used to account for differences in the number of pigs per treatment on which the average was based. The homogeneity of variance was tested using the Levene’s test (HOVTEST) and graphical diagnostics in PROC GLM. Where data were unbalanced, the denominator degrees-of-freedom Satterthwaite was used for adjusting the degrees of freedom to unequal variance and studentized maximum modulus for multiple comparisons, in all other cases protected difference was used to compare means. Because pigs were mixed after 7 wk post-weaning, the subsequent statistical unit was pen mean for each constituent treatment group, based on the number of YL, YH, OL, and OH within each pen, taking post-weaning nursery feeding regime into consideration. The pen means were blocked by pen nested within finisher room. A WEIGHT statement was used to account for the difference in the number of pigs the pen mean was based on. For both models (i.e., weaning to 7 wk post-weaning and 7 to 15 wk post-weaning), adjusted group size (i.e., adjusting for mortality) was added to the model as covariate:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{G}\mathrm{r}\mathrm{o}\mathrm{u}\mathrm{p}\mathrm{s}\mathrm{i}\mathrm{z}\mathrm{e}=\frac{(\mathrm{T}\mathrm{i}\mathrm{m}{\mathrm{e}}_{\mathrm{p}\mathrm{i}\mathrm{g}}/24\mathrm{h})}{\mathrm{T}\mathrm{i}\mathrm{m}{\mathrm{e}}_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}}}\end{array}}$$

where Time_pig is the total time (h) pigs reside in the pen and Time_total the total period (d). A WEIGHT statement was used to account for differences in the number of pigs per treatment the average was based on.

Performance to 20 and 60 kg

The performance of the pigs from birth and weaning to a common BW (i.e., 20 and 60 kg) was also investigated. The former was chosen because it was the BW of the lightest pig within the cohort at 7 wk post-weaning, whereas the latter was the BW of the lightest pig at 15 wk post-weaning. The data were analyzed using the PROC MIXED models as stated previously.

Predictor variables for post-weaning performance

The statistical unit for both univariate and multivariate models applied was pig blocked by pen nested within pen pair. All potential predictor variables (i.e., BiW, total number of born, morphometric characteristics, relative BiW, sex, pre-weaning litter size, weaning age, weaning weight, pre-weaning ADG, and nursery feeding regime) and their effect on post-weaning performance of pigs of the different age and weight classes (i.e., YL, YH, OL, and OH) were fitted in a univariate mixed model (PROC MIXED); those that were significant (P < 0.05) were taken forward in the multivariate analysis. Different multivariate models were built following a forward and backward stepwise procedure for variables that were highly correlated (r > 0.70; PROC CORR), using the AIC criteria (the smaller the better) to determine the model of best fit. Only factors with a probability below 0.05 were retained in the final model.

Results

Pre-weaning Performance

The average number of piglets born alive was 13.7 (range 2 to 21) with an average BiW of 1.49 kg (SD = 0.30) and birth litter CV of 19.5 (SD = 7.0). Average litter size at 2 to 3 d prior to weaning was 12.4 (SD = 1.9) with an average weaning weight and ADG of 9.01 kg (SD = 1.63) and 237 g/d (SD = 43), respectively. Pre-weaning mortality rate of piglets born alive was 17.2%, of which 25.3% was attributed to crushing.

Post-weaning Performance

The number of pigs per treatment after the “intermediate” pigs were excluded from the analysis, and overall mortality rate can be found in Table 2. Table 3 shows the effect of age, weight, and nursery feeding regime on post-weaning performance of pigs. Supplementary Table S1 shows the same information but with the “intermediate” pigs included. A total of 188 pigs were considered YO, 162 pigs were considered YH, 145 pigs were considered OL, and 136 pigs were considered OH. Post-weaning mortality was not significantly (P > 0.05) affected by weaning weight or nursery feeding regime (Table 2). Pigs weaned Y (day 31.8, SD = 0.2), however, had a significantly (P = 0.046) higher post-weaning mortality rate than pigs weaned O (day 37.5, SD = 0.3).

Table 2.

The number of pigs and mortality rate per treatment (i.e., weaning age, weaning weight and nursery feeding regime) during the different stages (PW; post-weaning) of production. At weaning, pigs were split into different groups based on their weaning age (young vs. old) and weaning weight (light vs. heavy) and pigs were randomly allocated to the dietary treatment (superior vs. standard feed allowance)^1,2

Weaning age (A)	Young				Old				Total
Weaning weight (W)	Light		Heavy		Light		Heavy		Young	Old	Light	Heavy	SUP	CON
Regime (R)	SUP	CON	SUP	CON	SUP	CON	SUP	CON
Number of pigs
Weaning (WW)	92	96	79	83	71	74	68	68	350a	281b	333a	298b	310	321
Week 2 PW	92	96	76	82	68	72	68	68	346a	275b	327A	294B	303	318
Week 7 PW	88	95	75	79	68	69	68	68	337a	273b	320A	290B	299	311
Week 15 PW	84	86	75	73	67	69	65	66	318a	267b	306	279	291	294
Mortality rate, %
Total	8 (8.70%)	10 (10.4%)	4 (5.06%)	10 (12.0%)	4 (5.63%)	5 (6.76%)	3 (4.41%)	2 (2.94%)	32a (9.14%)	14b (4.98%)	27 (8.11%)	19 (6.38%)	19 (6.13%)	27 (8.41%)

¹At weaning, pigs were split into 2 groups based on their age excluding pigs that were weaned at day 34: the bottom 50% were classified young and the upper 50% were classified old. Within weaning age group pigs were classified light (lightest 50%) or heavy (heaviest 50%) on the basis of their weaning weight (WW). Pigs were randomly allocated to the dietary treatment either fed a high feed allowance regime (SUP) in which pigs had a 65% greater allowance of the first and second stage feeds (on a kg/pig basis) or the control regime (CON).

²Data were analyzed with a chi-square test.

^a,bNumbers with different superscripts differ significantly (P < 0.05).

^A,BNumbers with different superscripts tended to differ (P < 0.10).

Table 3.

The effect of weaning age (young vs. old), weaning weight (light vs. heavy), and nursery feeding regime (superior vs. standard) on post-weaning (PW) performance^1,2

Weaning age (A)	Young				Old
Weaning weight (W)	Light		Heavy		Light		Heavy			Significance3
Regime (R)	SUP	CON	SUP	CON	SUP	CON	SUP	CON	SED	A	W	R	A × W	W × R	A × R	A × W × R
Age, d
Weaning	31.7	31.7	31.9	31.9	37.5	37.6	37.4	37.4	0.430	<0.001	0.897	0.824	0.353	0.970	0.849	0.903
BW, kg
Weaning (WW)	6.93	6.89	10.1	10.1	8.27	8.37	11.5	11.5	0.250	<0.001	<0.001	0.754	0.962	0.980	0.501	0.945
Wk 2 PW	12.8	13.0	16.4	18.0	14.5	15.4	18.2	17.9	1.04	0.004	<0.001	0.129	0.156	0.849	0.435	0.113
Wk 7 PW	35.3	34.1	39.5	40.9	40.0	38.7	43.0	43.8	1.30	<0.001	<0.001	0.889	0.171	0.044	0.673	0.815
Wk 15 PW	91.0	89.8	98.2	98.4	96.4	94.8	104	104	1.35	<0.001	<0.001	0.527	0.752	0.380	0.908	0.990
ADG, g/d
WW to Wk 2 PW	353	370	380	485	375	421	411	387	55.9	0.955	0.108	0.108	0.116	0.820	0.241	0.081
WW to Wk 7 PW	535	519	558	587	602	576	604	615	24.0	0.001	0.005	0.937	0.185	0.046	0.435	0.816
Wk 2 to Wk 7 PW	620	587	640	632	701	643	690	718	29.8	0.001	0.016	0.130	0.997	0.034	0.811	0.182
Wk 7 to Wk 15 PW	929	931	986	954	944	952	1034	1006	16.2	0.004	<0.001	0.242	0.154	0.121	0.826	0.949

¹Individual BW were taken at 2 to 3 d (day 31.7, SD = 2.6) prior to weaning (day 34.6, SD = 3), 2, 7, and 15 wk post-weaning (day 144, SD = 2).

²At weaning, pigs were split into 2 groups based on their age excluding pigs that were weaned at day 34: the bottom 50% were classified young and the upper 50% were classified old. Within weaning age group, pigs were classified as light or heavy on the basis of their weaning weight (WW). Pigs were then randomly allocated to the different pens and nursery feeding regime: either fed a high feed allowance regime (SUP) or the control regime (CON). All diets were identical between the 2 feed allowances; however, pigs fed the SUPERIOR regime were fed a 65% greater allowance of the first and second stage feeds.

³The experimental unit for all models was pen mean with main effects of interest: weaning age (A), weaning weight (W), nursery feeding regime (R), and their interactions: A × W, W × R, A × R, A × W × R. A weight statement was used to account for differences in the number of pigs on which the pen means were based. From weaning to 7 wk PW, pigs remained in the same group and pen pair was used as a random factor. From 7 wk PW to finisher, pigs were mixed in smaller groups; therefore, pen mean nested within room was used as a random factor. Adjusted group size was inserted as covariate to all models.

The interaction between age × regime, age × weight, and age × weight × regime did not significantly affect (P > 0.05) performance up to 7 wk post-weaning (Table 3). However, significant interactions were observed between weight × feeding regime for BW at 7 wk post-weaning (P = 0.044), for post-weaning ADG obtained between weaning and 7 wk post-weaning (P = 0.046) and for ADG between 2 and 7 wk post-weaning (P = 0.034). Pigs weaned L and fed the SUPERIOR regime weighed 1.2 kg heavier at 7 wk post-weaning than L pigs fed the CON regime (37.6 kg, SD = 0.6 and 36.4 kg, SD = 0.6, respectively). Similarly, pigs weaned L and fed the SUPERIOR regime showed higher ADG between weaning and 7 wk post-weaning and between 2 and 7 wk post-weaning (569 g/d, SD = 12 and 660 g/d, SD = 14, respectively) than L pigs fed the CON regime (548 g/d, SD = 11 and 615 g/d, SD = 14, respectively). Pigs weaned H were 1.00 kg lighter at 7 wk post-weaning, gained less between weaning and 7 wk post-weaning, and between 2 and 7 wk post-weaning when having access to the SUPERIOR regime (41.3 kg, SD = 0.2; 581 g/d, SD = 12; and 665 g/d, SD = 15, respectively) than those fed the CON regime (42.3 kg, SD = 0.2; 601 g/d, SD = 12; and 675 g/d, SD = 15, respectively).

Age and weight significantly (at most P = 0.016) affected post-weaning performance (Table 3). At weaning Y pigs weighed 1.4 kg lighter than pigs weaned O (8.49 kg, SD = 0.17 and 9.91, SD = 0.19, respectively); this weight difference increased to −5.5 kg at 15 wk post-weaning. Significant weight differences (P < 0.001) were also observed between pigs weaned L and H. At weaning, pigs weaned L (day 34.6, SD = 0.3) were 3.2 kg lighter (7.60 kg, SD = 0.17 vs. 10.8, SD = 0.2) than pigs weaned H (day 34.6, SD = 0.3); this weight difference increased to −8.0 kg at 15 wk post-weaning (93.0 kg, SD = 3.6 vs. 101, SD = 4). The main effect nursery feeding regime did not significantly (P > 0.05) influence post-weaning performance.

Performance to 20 and 60 kg

The interaction between weaning weight and nursery feeding regime significantly affected ADG from birth to 20 kg of weight (P = 0.046) and the number of days it took the pig to reach 20 kg (P = 0.027). Although no differences were seen for H pigs fed the SUPERIOR or CON nursery feeding regimes, L pigs fed the SUPERIOR regime gained weight faster (348 g/d, SD = 5) and needed less time to reach 20 kg (53.9 d, SD = 0.7), when compared with L pigs fed the CON regime (341 g/d, SD = 5 and 55.2 d, SD = 0.7, respectively). Also weaning age and weaning weight significantly affected ADG from birth to 20 kg (P = 0.002 and P < 0.001, respectively) and the number of days it took pigs to reach 20 kg (P = 0.001 and P < 0.001, respectively). Pigs weaned L had a lower ADG (344 g/d, SD = 5) and needed ~6 more days (54.6 d, SD = 0.7) to reach 20 kg compared with pigs weaned H (385 g/d, SD = 5 and 48.2 d, SD = 0.7, respectively). Pigs weaned Y on the other hand, gained weight faster (371 g/d, SD = 5) and needed less time (50.3 d, SD = 0.7) to reach 20 kg BW compared with pigs weaned O (358 g/d, SD = 5 and 52.4 d, SD = 0.7, respectively).

Similar results were seen for ADG between birth and 60 kg BW, with a significant (P < 0.05) effect found for weaning age and weaning weight for ADG (P = 0.020 and P < 0.001, respectively) and the number of days taken (P = 0.015 and P < 0.001, respectively); however, the interaction between weaning weight and nursery feeding regime was no longer significant (P > 0.05). Pigs weaned L gained less (566 g/d, SD = 7) and needed more time to reach 60 kg (104 d, SD = 1) compared with pigs weaned H (604 g/d, SD = 7 and 97.3 d, SD = 1.4, respectively). On the other hand, pigs weaned Y gained more (591 g/d, SD = 7) and needed less time to reach 60 kg (99.5 d, SD = 1.4) than pigs weaned O (579 g/d, SD = 8 and 102 d, SD = 2, respectively).

Also the performance from weaning to 60 kg (i.e., ADG and days to reach 60 kg) was affected by weaning age (P = 0.055 and P = 0.007, respectively) and weaning weight (P = 0.007 and P < 0.001, respectively), but not by their interactions (P > 0.05). Pigs weaned L needed more time (72.3 d, SD = 1.7) and gained weight slower (734 g/d, SD = 16) between weaning and 60 kg compared with pigs weaned H (65.0 d, SD = 1.8 and 766 g/d, SD = 17, respectively). Pigs weaned Y, however, tended to gain weight slower (739 g/d, SD = 17) and needed more time to reach 60 kg (70.4 g/d, SD = 1.7) compared with pigs weaned O (761 g/d, SD = 19 and 66.8 d, SD = 1.9, respectively). Although only numerical (P > 0.05) differences were seen between pigs weaned YL and OL with respect to ADG between weaning and 60 kg of weight (723 g/d, SD = 15 vs. 745 g/d, SD = 18), pigs weaned YL required more time (P = 0.018) to reach 60 kg when compared with OL pigs (74.1 d, SD = 1.6 vs. 70.4 d, SD = 1.8).

Feed Intake and Economic Analysis

Total intake per pig (kg/pig), G:F ratio, total feed cost per pig, total feed cost per kg gain, and MOF costs from weaning to 7 wk post-weaning were not affected significantly by regime, age, and weight; the same (P > 0.05) was the case for the interactions between weight × age, weight × regime, and age × regime. The total feed cost per pig was slightly higher for the pigs fed the SUPERIOR compared with those fed the CON regime ($22.2/pig, SD = 0.9 vs. $20.9/pig, SD = 0.9). Pigs weaned Y had a numerically lower MOF cost and total feed intake ($35.5, SD = 1.1 and 55.9 kg/pig, SD = 2.0, respectively) than pigs weaned O ($37.6, SD = 1.1 and 60.6 kg/pig, SD = 2.0, respectively). Also, pigs weaned L had a numerically lower MOF cost and total feed intake ($35.5, SD = 1.1 and 55.3 kg/pig, SD = 2.0, respectively) compared with pigs weaned H ($38.1, SD = 1.1 and 61.2 kg/pig, SD = 2.0, respectively). Pigs weaned YL had a numerically lower MOF cost compared with pigs weaned YH ($33.6 vs. $38.2); OH and OL had a MOF cost of $37.9 (SD = 1.6) and $37.4 (SD = 1.6), respectively.

Predictor Variables for Post-weaning Performance

Correlations between variables for the different weaning age and weight groups (i.e., YL, YH, OL, and OH) are shown in Supplementary Table S2. The univariate analysis assessing which predictor variables significantly affected (P < 0.05) post-weaning performance (ADG between weaning and 7 wk post-weaning) is shown in Table 4. The final multivariate models are shown in Table 5. The final multivariate model for YL pigs rejected BiW as best predictor and showed that BiW:CC (P < 0.001) and weaning weight (P < 0.001) were positively associated with post-weaning ADG. The best model fit for OL pigs consisted of BiW:CC (P = 0.044), weaning age (P = 0.043), weaning weight (P < 0.001), and nursery feeding regime (P = 0.027); all were positively associated with post-weaning ADG and with OL pigs fed the SUPERIOR regime having a higher growth. The final multivariate model for YH comprised BiW:CC (P < 0.001) and weaning weight (P = 0.042), and for pigs weaned OH comprised BMI (P = 0.001), BiW:CC (P = 0.001), and weaning weight (P = 0.003).

Table 4.

Statistical significance (P-value) of the different predictor variables for post-weaning ADG (weaning and 7 wk post-weaning) fitted in the univariate models for piglets weaned at different ages and weights¹

Weaning age	Young		Old
Weaning weight	Light	Heavy	Light	Heavy
Predictor variable
Birth weight, kg	<0.001	<0.001	0.001	<0.001
Relative birth weight2	0.121	0.068	0.011	0.001
Total piglets born	0.008	0.001	0.099	0.001
Sex	0.814	0.960	0.271	0.521
Crown to rump length, cm	<0.001	<0.001	0.002	0.004
Snout to ear length, cm	0.319	0.001	0.014	0.001
Abdominal circumference, cm	0.002	<0.001	0.001	<0.001
Cranial circumference, cm	0.026	<0.001	0.005	<0.001
Body mass index3, kg/m2	0.100	0.003	0.028	<0.001
Ponderal index4, kg/m3	0.429	0.436	0.826	<0.001
Birth weight:cranial circumference, kg/cm	<0.001	<0.001	0.001	<0.001
Snout to ear length: birth weight, cm/kg	<0.001	<0.001	0.001	<0.001
Weaning weight, kg	<0.001	0.009	<0.001	0.046
Weaning age, d	0.243	0.425	0.009	0.004
Pre-weaning ADG, g/day	0.010	0.191	<0.001	0.816
Litter size pre-weaning5	0.921	0.999	0.270	0.516
Nursery feeding regime6	0.051	0.139	0.041	0.680

¹At weaning the youngest 50% were classified “young” and the oldest 50% “old” excluding intermediate pigs (pigs weaned at day 34); within age class, pigs were split into 2 groups based on their weaning weight (light, bottom 50%; heavy, upper 50%). Morphometric measurements were taken within 12 to 24 h post-partum, pigs were weighed at birth (day 0), at 2 to 3 d before weaning (day 31.7; SD = 2.6), and at 7 wk post-weaning. Significant predictor values are given in bold.

²Relative birth weight = (Birth weight piglet/mean birth weight birth litter).

³Body mass index = birth weight (kg)/[crown rump length (m)]².

⁴Ponderal index = birth weight (kg)/[crown rump length (m)]³.

⁵Pre-weaning litter size = [(total time (h) piglets reside within litter/pen)/24 h]/total period in day.

⁶At weaning pigs were randomly allocated to the different nursery feeding regimes: either fed a high feed allowance regime in which pigs had a 65% greater allowance of the first and second stage feeds or the control regime.

Table 5.

Final multivariate models (coefficient and SE) for different predictor variables for post-weaning ADG (weaning to 7 wk post-weaning) for piglets from different weaning weights and weaning ages^1,2

Weaning age	Young		Old
Weaning weight	Light	Heavy	Light	Heavy
Predictor variable
Body mass index3, kg/m2	—	—	—	9.53 (2.86)
Birth weight: cranial circumference, kg/cm	1,758 (444)	3,136 (535)	1,354 (663)	2,590 (735)
Weaning age, d	—	—	10.5 (5.13)	12.8 (4.2)
Weaning weight, kg	19.4 (4.7)	10.2 (5.0)	36.6 (6.7)	—
Nursery feeding regime4	—	—	39.5 (17.6)	—

¹At weaning, pigs were split into 2 groups based on their weaning age: the bottom 50% were classified “young” and the upper 50% were classified “old,” within weaning age group pigs were classified “light” (bottom 50%) or “heavy” (upper 50%) on the basis of their weaning weight.

²Morphometric measurements were taken within 12 to 24 h post-partum, pigs were weighed at birth (day 0), at 2 to 3 d before weaning (day 31.7; SD = 2.60), and at 7 wk post-weaning. Pigs weaned at an “intermediate” age were excluded from the analysis.

³Body mass index = birth weight (kg)/[crown rump length (m)]².

⁴The CON regime was set at 0. At weaning, pigs were randomly allocated to a different nursery feeding regime: either fed a high feed allowance regime in which pigs had a 65% greater allowance of the first and second stage feeds or the control regime.

Discussion

We investigated 2 management strategies that aim to address the challenge associated with pigs weaned lightweight: delaying weaning and specialist nutrition. Both strategies are based on the assumption that such pigs have different (nutritional) requirements than their “normal” weight counterparts. Our hypothesis was that such pigs would perform better when weaned at a later age and/or when fed more of the nursery feeding regime. Pigs weaned light that would benefit most from these strategies would be those born disproportional, that is, piglets born with a dolphin-like forehead and those born long and thin (Douglas et al., 2016; Huting et al., 2018). These hypotheses were based on the findings that 1) post-weaning growth check decreases with increasing weaning age (Colson et al., 2006; van der Meulen et al., 2010; Leliveld et al., 2013) and 2) lightweight pigs have an immature digestive system compared with their heavier counterparts (Cranwell et al., 1997; Pluske et al., 2003); thus, increasing weaning age might be especially beneficial for them. We hypothesized that although piglets born disproportional have been suggested to have a reduced post-weaning performance (Douglas et al., 2016; Huting et al., 2018) and are known to be less efficient in nutrient utilization (Wu et al., 2006), they may benefit when they are given a high-quality nursery regime over a longer period of time. In addition, specialist feeding regimes (e.g., starter diets using readily digestible/high-quality nutrient sources with a high nutrient density), which have been shown to be effective for lightweight pigs (Magowan et al., 2011; Douglas et al., 2014), may have different effects on the performance of pigs which have been weaned either young or old.

In the European Union, pigs are usually weaned at 4 wk of age (European Commission, 2008). However, there are concerns about the health and welfare implications of this practice, especially in relation to the ban on the use of antibiotic growth promoters (European Commission, 2003a) and the restrictions on the prophylactic use of copper and zinc (European Commission, 2003b, 2016). Weaning is accompanied by a drop in intake and a reduced digestive and absorptive capacity, negatively affecting gut health (Pluske et al., 1997; Lallès et al., 2007). Pigs weaned older have a more mature digestive system (Cranwell et al., 1997; Pluske et al., 2003), with delayed weaning positively affecting the mucosal permeability and the motility of the gut (Moeser et al., 2007), decreasing the proliferation of pathogenic bacteria (Wellock et al., 2007; Leliveld et al., 2013), and reducing mortality rate (Leliveld et al., 2013). All these, together with the higher post-weaning feed intake (van der Meulen et al., 2010), suggest that pigs weaned older recover more quickly from the post-weaning diet transition than pigs weaned young. This suggests that increasing weaning age might be the way forward under current EU regulations for reducing the potential negative effects the ban on antibiotic growth promotors may have on performance and welfare. The fact that improvements in performance can be seen beyond a 30 d weaning age, as seen in this study, is significant as it supports that benefits can be realized with increasing weaning age and thus an “improvement ceiling” had not been reached. However, it is appreciated that piglets are weaned significantly earlier than the ages investigated in this experiment; for example, in North America, weaning ages are more commonly in the 18- to 24-d range. Therefore, it remains to be seen whether in these cases the benefits of later weaning can also be seen in the younger weaning ages (e.g., days 21 to 30), as opposed to the older ages investigated here. In other words, it is possible that an age threshold needs to be achieved for the later weaning benefits to materialize.

To our knowledge, our study is one of the first attempts to disentangle the effects of weaning age vs. weaning weight. Our work provides new evidence that although weaning weight does predict subsequent growth and health, weaning age alone is a critical factor. There was no interaction between weaning age and weaning weight in our experiment, suggesting that both L and O pigs benefited equally from the delay in weaning. Previous large-scale investigations, such as during the “AGEWEAN” project (Edge et al., 2008), have suggested that delaying weaning to 6 wk of age has beneficial consequences on the number of pigs needing veterinary treatment or dying post-weaning. This was also the finding of our study, as post-weaning mortality was almost double for pigs weaned earlier (Y pigs). However, when the overall performance of the pigs in the “AGEWEAN” study was evaluated, weaning pigs at 4 wk of age were still considered to be more efficient (Edge et al., 2008), although the authors emphasized that not all their pigs were followed until slaughter. Colson et al. (2006) and Leliveld et al. (2013) similarly reported that, although the growth depression was larger and lasted longer for early compared with late weaned pigs (3 vs. 4 wk of age), this only influenced performance during the intermediate post-weaning period and had no long-term consequences (to 10 wk of age). Despite that our pigs were weaned relatively old, our study suggests that pigs weaned Y gained less between weaning and 60 kg BW and needed almost 4 extra days from weaning to 60 kg BW than pigs weaned O. Although weaning L pigs later may be an attractive way to improve their post-weaning performance, the additional sow costs (e.g., feeding, housing) need to be taken into consideration to evaluate the cost-effectiveness of such a strategy.

In this study, we were particularly interested in whether increasing weaning age for L pigs would be beneficial, as the aforementioned concerns may be exacerbated in the lightweight pigs. Split weaning, that is, weaning the heaviest piglets of the litter 3 to 7 d before the rest of the litter, is thought to reduce the weaning-to-estrus interval (Soede and Kemp, 2015) and to increase pre-weaning performance for the remaining piglets, compared with piglets that were weaned at the same calendar age, but not split weaned (Pluske and Williams, 1996; Vesseur et al., 1997). Literature is inconsistent with respect to whether pigs that remained on the sow longer after split weaning could retain the BW advantages at weaning in later life. Some suggest that the BW advantage does not persist in the long term (split weaning at day 21 vs. weaning at day 28; Pluske and Williams, 1996; Vesseur et al., 1997), while under tropical conditions, it was reported that L pigs that remained on the sow until day 53 after split weaning (at day 28) were heavier at 15 wk of age compared with L pigs that were either weaned early (day 28) or as a whole litter at the same age (day 53; Abraham et al., 2004). Although split weaning was not practiced in this study and all pigs were weaned at the same day irrespective of the date of birth, the results presented demonstrate that L pigs benefitted from later weaning (O), needing approximately 4 d less to reach 60 kg compared with L pigs that were weaned 6 d earlier (Y). Leliveld et al. (2013) also found that an increase of 1 kg in weaning weight resulted in a 2.3 or 1.8 kg weight advantage at 10 wk of age for pigs weaned at 3 or 5 wk, respectively, thus emphasizing that weaning age is more important for pigs weaned L. Pigs end up L at weaning as a result of 1) being born L and/or disproportional, 2) insufficient colostrum intake, 3) direct and indirect competition for milk intake, or 4) due to illness (Declerck et al., 2016; Douglas et al., 2016; Huting et al., 2017, 2018). Increasing weaning age for these pigs might stimulate gut maturity, with components in the milk altering the intestinal microflora important for the development of the gastrointestinal tract and the immune system, as shown in human infants (Fanaro et al., 2003; Schack-Nielsen and Michaelsen, 2007). In addition, creep feed intake increases with increasing weaning age (Defra, 2007), positively influencing post-weaning feed intake (Muns and Magowan, 2018).

Nutritionists have been suggesting specialist feeding regimes for L pigs as another way to compensate for the immature digestive system and lower feed intake of L pigs, and to maximize post-weaning performance (Douglas et al., 2014). These range from enhanced diet composition, improving the digestibility and palatability of the nursery diets (Beaulieu et al., 2010; Douglas et al., 2014), or feeding a greater allowance (kg/pig basis) of the nursery diets (Magowan et al., 2011; Douglas et al., 2014; Muns and Magowan, 2018). This may increase voluntary feed intake, but the use of readily digestible nutrients may reduce the amount of substrate passing the large intestine, thus improving intestinal health (Wellock et al., 2009). In this experiment, we opted for feeding a greater allowance because of its practicality, which was achieved by feeding lightweight pigs for a longer period of time on the nursery feeding regimes. The experimental outcomes suggest that pigs weaned L were able to improve post-weaning performance when having access to the SUPERIOR regime. Mahan et al. (1998) also found that feeding the first phase regime for 2 wk instead of 1 wk reduced the number of days it took L pigs to reach slaughter weight; they reported, however, that weaning weight still had a greater effect on post-weaning performance than feed allowance. Likewise, in this experiment, the increased amount of kilograms per pig of diet 1 and 2 was not enough to maximize the growth of L pigs. What is interesting is that the performance of the H pigs was not affected by the increased amount of the nursery feeding regime consumed, questioning the need for this strategy for this class of pigs. We expected that the higher feed allowance would also improve the performance of pigs weaned H, but that it would not be economically viable.

It should be emphasized that the advantages of the SUPERIOR regime for the L pigs soon after weaning were not sustained in the long term, that is, the interaction between feeding regime and weaning weight was no longer significant by 15 wk post-weaning, in agreement with Magowan et al. (2011). When their performance was considered to 60 kg, there was neither any longer a nursery feeding regime effect or a weaning weight × regime interaction. Also, others (Magowan et al., 2011) who fed 100% more of the first-stage nursery feeds showed that the BW advantage during the immediate post-weaning period was not sustained in the long term. This questions the successfulness of the higher allowance nursery feeding regime for lightweight pigs.

We have shown previously that some morphometric traits of the lightweight piglets are better predictors of post-weaning performance than birth weight (Douglas et al., 2016; Huting et al., 2018). Also in this study, it became clear that not all lightweight pigs are alike and some are able to perform better than the others. Irrespective of weaning age, pigs weaned L most likely to remain light were characterized by a disproportionate head circumference in relation to size (low BiW:CC) and a lower weaning weight. The larger head in relation to size may characterize piglets that suffered from IUGR and remain stunted throughout life (Amdi et al., 2013). On the other hand, post-weaning performance of pigs weaned OL was positively associated with weaning age. The greater weaning age variation for pigs weaned OL when compared with YL might explain why weaning age affected the performance of OL pigs only. What was interesting was that the nursery feeding regime influenced OL pigs only. Although L pigs are thought to have a less developed digestive system (Cranwell et al., 1997; Pluske et al., 2003), creep feed intake increases with weaning age (Defra, 2007). The latter may have prepared the digestive tract of OL pigs for weaning, resulting a smaller growth check. The feed allowance for YL pigs might not have been sufficient to support their performance.

Conclusions

We have found that both increasing weaning age and feeding higher quantities of the first diets in the post-weaning nursery feeding regime are beneficial for pigs weaned L. We did not find an interaction either between weaning age and weaning weight, or between weaning age and feeding regime. Although feeding higher amounts of the nursery diets could be easily implemented in pig units that separate pigs on the basis of weaning weight, the benefits were not sustained in the longer term. Although more research is needed, it is recommended that delayed weaning may yield longer term improved post-weaned performance for pigs weaned light.