Impact of parity on carcase and metabolic markers associated with oxidative stress during uterine involution in periparturient goat

Abstract This study aimed to verify the impact of parity on tissue mobilisation, metabolic response, oxidative stress and reproductive tract traits during the peripartum period. Nineteen goats with singleton pregnancies were grouped into nine nulliparous (NU) and 10 multiparous (MU). The animals were followed from the fifth week before delivery to the fourth week after delivery. Does, kids and milk production were weighed; the loin muscle depth area, subcutaneous loin, kidney fat thicknesses, size of the uterus and haemodynamics of the uterine artery were measured by ultrasound. Placenta was weighed and the cotyledons were counted and measured. Plasma was assayed for total protein, glucose, cholesterol, triglycerides, creatinine, gamma-glutamyl transferase (GGT), glutamic-oxaloacetic acid transaminase (GOT) and glutamic-pyruvic acid transaminase (GPT), glutathione peroxidase and β-hydroxybutyrate. MU showed a greater live weight loss (+8%) and longer period of body mass loss (+12 days). Milk production was higher in MU and their kids were heavier at birth with greater weight gain. No differences in the effect of parity were found for uterine diameter and placental weight, while the NU goats had a larger cotyledonary surface. Additionally, MU goats had higher levels of glucose, GPT, GGT, cholesterol, total protein and glutathione peroxidase. In MU females, there was a greater muscle mass mobilisation throughout the peripartum period and a superior replacement of adipose tissue after delivery (+1.3 mm). Therefore, MU females appear to have better tissue mobilisation and productive performance dynamics, although parity does not affect uterine involution in goats. Highlights Parity influences metabolic response and oxidative stress during peripartum in goats. Multiparous goats show greater efficiency in the mobilisation of muscle and adipose tissue and better productive performance. Parity does not influence the process of uterine involution.


Introduction
Uterine involution is one of the most important physiological processes in the reproductive tract of ruminants during puerperium and is one of the main obstacles to the resumption of cyclic ovarian activity during the postpartum period.This process promotes a physical reduction in lumen and cervix size after delivery and is driven by local muscle contraction, extracellular matrix renewal, necrosis, shedding of uterine caruncles and endometrial regeneration (Sheldon and Dobson 2004).These significant changes occur during the metabolic state of parturient females, caused by a rapid increase in milk secretion and a decrease in dry matter intake, resulting in a highly unfavourable metabolic condition, negative energy balance (NEB), characterised by marked weight loss due to the depletion of fat reserves and loss of muscle mass, hypoglycaemia and hypoinsulinaemia, with a consequent increase in the levels of circulating ketone bodies (Grummer 1993).b-Hydroxybutyrate (BHB) is the main ketone body present and has been used as a marker for the diagnosis of pregnancy toxaemia in sheep (Zou et al. 2021).BHB is metabolised in the liver as a source of energy and to produce fatty acids (Zou et al. 2021).
In this context, females that are on NEB have reduced concentrations of nutrients and changes in metabolic hormones, including a reduced abundance of glucose, glutamine, and insulin-like growth factor 1 (IGF1), which can modify the inflammatory response and promote oxidative stress due to the high mobilisation of body reserves, and consequently increased the cellular metabolism involved (Celi et al. 2010).These inflammatory processes can interrupt not only uterine function but also affect the hypothalamic-pituitary-ovarian connection, potentially cause subfertility (Sheldon and Dobson 2004) or lead to an extension of postpartum anoestrus, delay the reproductive period, and increase the interval between parturition.Nutritional factors during the postpartum period can also cause oxidative stress by increasing reactive oxygen species at the cellular level (Celi et al. 2010), as observed in cows fed a diet high in starch and non-fibrous carbohydrates during early pregnancy.
In goats, the rate of reduction of the uterus is not uniform, being directly influenced by the puerperium period, with a higher rate of involution in the first two weeks postpartum, being faster in the first sevendays and ending on the 19th postpartum day (Ababneh and Degefa 2005).The recovery of the endometrium, takes around three to four weeks to fully recover the normal tissue architecture (Sheldon and Dobson 2004), and hormonal, metabolic status.After parturition, circulating steroid hormone concentrations decrease to baseline values with a consequent reduction in the luteinizing hormone (LH) pulse frequency, affecting female cyclic activity (Cheong et al. 2016).
The liver has a limited capacity to metabolise lipids, and excessive accumulation of ketone bodies promoted by NEB can lead to pregnancy toxaemia (Zou et al. 2021).Identifying metabolic alterations in goats during this period and preventing metabolic disorders, such as pregnancy toxaemia and hepatic steatosis, are economically beneficial, as they accelerate uterine involution and return to the reproductive cycle (Cheong et al. 2016).Therefore, metabolic disorders in goats, as in other ruminant species, are strongly associated with reproductive performance.Excessive prepartum body condition or accentuated fat and muscle mobilisation at parturition predispose cattle to metabolic risks at the expense of the reproductive response (Goff and Horst 1997).In beef cattle, parity order influences milk production, metabolic hormones and metabolites after calving (Ferreira et al. 2021).However, little is known about goats, especially in tropical regions, where the main populations of this species are found.Therefore, we hypothesise that in goats, the parity order exerts an effective influence on the dynamics of mobilisation of adipose and muscular tissues before and after delivery, affecting the metabolic response, oxidative stress and uterine involution.
This study aimed to investigate nulliparous and multiparous goats during the peripartum period, verify the performance of carcass and metabolic markers and obtain information on the association of these indicators with glutathione peroxidase and uterine involution.

Location and animal ethics
This study was carried out in the facilities of the laboratory of nutrition and ruminant production of the 'Esa u Accioly Vasconcelos' experimental farm School of Veterinary Medicine, Cear a State University, located in Guaiuba, Cear a, in the equatorial zone (4 2 0 23ʺ S and 38 38 0 14ʺ W), Brazil.

Animals and experimental design
Thirty-three Anglo-Nubian crossbred adult goats from the school farm herd had their oestruses synchronised and were naturally mated with Anglo-Nubian bucks.Pregnancies were detected by ultrasonography 45 days after mating and 19 goats with a singleton pregnancy were grouped according to their parity order with nine nulliparous (age, 30.7 ± 3.8 months; mean ± SD) and 10 multiparous (age, 64.5 ± 6.2 months; 2.5 ± 0.5 parity) with a homogeneous body condition score of 2.8 ± 0.2 (score from 1 to 5).All animals were clinically followed up by a trained veterinarian and received the same diet based on chopped elephant grass and concentrate composed of the following ingredients: ground corn grain, wheat bran, soybean meal and a mineral and vitamin mixture.The proportion of concentrate ingredients was adjusted according to the nutritional parameters of the diet during the gestation phase (initial and late gestation) and early lactation and the mineral and vitamin mixture was maintained at 2% on a dry matter (DM) basis of the total diet.The diets used had a roughage: concentrate ratio of 60:40 and were furnished to satisfy the nutritional requirements of singleton pregnant adult non-dairy goats (NRC 2007) for each phase of gestation (initial and late gestation) and early lactation.The animals were kept in collective stalls with free access to salt supplements and water.The experimental period was 63 days, beginning the fifth week before delivery (Àfive weeks) and continuing until the fourth week after delivery.Delivery was induced on day 145 of pregnancy by intramuscular injection of 1 mL of PGF2a (Prolise; ARSA S.R.L., Buenos Aires, Argentina), and the kids were kept throughout the postpartum period.
In vivo performance and milk yield Does and kids were weighed weekly.Milk production was measured weekly during the postpartum period, as described by Celi et al. (2008).In summary, on the day before the measurement, all kids were separated from their mothers at 18:00.The next day at 6:00 am, each kid was weighed before and after nursing.The feeding period did not exceed 30 min.The difference between the pre-and post-feeding weights was recorded as the estimated milk yield of the goats.

Carcass marker measurements
Fat and muscle mass were measured weekly throughout the experimental period using B-mode ultrasound equipment with a 5 MHz linear probe (model Z5 Vet; Mindray Bio-Medical Electronics Co., Shenzhen, China).Longissimus dorsi muscle depth, loin area and subcutaneous loin fat thickness were measured between the third and fourth lumbar vertebrae, following the methodology of Teixeira et al. (2008), and to estimate visceral fat was measured the thickness of kidney fat behind the 13th rib, following the methodology of H€ arter et al. (2014).A convex transducer with a frequency of 3.5 MHz (model Z5 Vet; Mindray Bio-Medical Electronics Co., Shenzhen, China) was used for kidney imaging.Images were captured in triplicate and measured using the previously calibrated ImageJ program (ImageJ, National Institutes of Health, Millersville, USA).During the evaluation, the animal was kept stationary, the areas on the right side of the body were shaved, and the gel was used as a coupling agent to improve the quality of the images.

Placenta and cotyledons measurement
At parturition, the placenta was weighed, and the cotyledons were counted.Ten cotyledons for each placenta were sampled to measure their length and width, according to Camacho et al. (2018).Placental efficiency was determined as the ratio between kid weight, placental weight and cotyledonary efficiency as the ratio between kid weight and cotyledonary area.

Uterine involution and uterine artery haemodynamics
After delivery, ultrasound images of the uterus were taken weekly to measure the diameter of the horn.Briefly, females were kept in a stationary position, and a linear transrectal transducer assisted the B-mode ultrasound with a frequency of 5.0 MHz (DP-2200Vet, from Mindray Bio-Medical Electronics Co., Ltd., Shenzhen, China), made rigid by an extension rod, according to the methodology of Ababneh and Degefa (2005).The probe was gently inserted with reference to the cranial edge of the bladder as a landmark for the positioning of the uterus.The uterine images were captured in triplicate and analysed using ImageJ software (ImageJ, National Institutes of Health, Millersville, USA).The diameter of the uterine horns was measured from the bifurcation and that of the gravid horn was determined according to the horn with the largest diameter.
To assess uterine blood flow, pulsed colour Doppler ultrasound was used at a frequency of 5.0 MHz and colour gain ¼ 60% to visualise the uterine artery and measure the diameter of the uterine artery and capture pulsatile waves.Doppler velocimetric parameters (resistance index, peak systolic) were obtained from an average of three waves (Camacho et al. 2018).

Data analysis
Statistical analyses were performed using Statistica Software, version v. 13.4.0.14 (2018; TIBCO Software, Inc., Palo Alto, CA, USA).Data were initially verified for normality assumptions by Kolmogorov À Smirnov and Bartlett tests, and when these conditions were not respected, the transformation to log10Â was applied.
Data on in vivo performance and reproductive tract traits at delivery were subjected to analysis of variance (ANOVA) of GLM procedures in a factorial arrangement, where the main effect tested was the parity group (nulliparous and multiparous).Regarding fat and muscle mass mobilisation and metabolite concentration, the fixed effects parity group, interval, or period from delivery (BDP, before delivery period [Àfive weeks to À one week]; DP, delivery period day 2; and ADP, after delivery period [first to fourth weeks]) and interaction parity group vs. period.
Data on metabolites, BHB and GPx levels, as well as descriptive ultrasonography data (loin area, kidney fat thickness, subcutaneous loin fat thickness, uterine diameter and uterine artery hemodynamics) and GLM procedures, were used for repeated measures of ANOVA.The effects tested were the parity group, interval of assessment and interaction parity group vs. time.The recorded anatomical images (1, 2, 3), or samples metabolites assays, were repeated measures.All pairwise comparisons were performed using the Newman À Keuls post hoc test, applied when ANOVA indicated a significant difference (p < .05).

Does and kid's in vivo performance and milk yield
The effect of the group was significant (p < .05)for the performance parameters used in the postpartum period (Table 1).Multiparous animals recorded a greater body weight loss (14.9%) than nulliparous (6.9%) (p ¼ .0117)and showed a longer period (38.0 ± 4.3 days vs. 25.7 ± 1.6 days; p ¼ .0106) of body mass loss (Table 1).
The kids in the multiparous group recorded a higher weight at birth at the fourth week postpartum (p < .05)and obtained a daily live weight gain twice as high (p ¼ .0190)as those born to nulliparous does.The mean milk yield during the postpartum period was higher in multiparous animals respect to nulliparous (0.9 ± 0.1 kg vs. 0.7 ± 0.04 kg; p ¼ .0113).The nulliparous group reached peak lactation in the third postpartum week, while multiparous group showed increased production during the four weeks postpartum (Figure 3(A)).

Reproductive tract traits
At parturition, the value of the placental weight (Table 1) were similar (p > .05) between the two groups, while in nulliparous animals, a greater cotyledonary surface (p ¼ .0063)and a lower cotyledon efficiency (p < .001)were recorded.In both parity classes, the diameter of the uterine horn (Figure 1(A)) shrank until the third week postpartum (time effect, p < .001),registering between the second and third week the highest rate of reduction in the uterine lumen (Figure 1(B)).
Concerning the uterine artery, between the first and fourth week postpartum, a reduction in peak systolic blood pressure (Figure 1(C)) and an increase in the resistance index (Figure 1(D)) were observed.The systolic peak was higher in multiparous animals (37.9 ± 2.6 cm/s vs. 30.3± 1.6 cm/s; p ¼ .0082)and the resistance index was higher in nulliparous animals (2.2 ± 0.3 vs. 1.2 ± 0.3; p ¼ .0478).
The parity order did not affect the concentration of BHB (Figure 2(A)).In both groups, the values increased until the second week postpartum and decreased markedly (time effect, p < .001).The multiparous group recorded higher values of GPx from the first to fourth week postpartum (Figure 2(B)).In this group, the highest concentration was in the second week, after which there was a reduction in circulating levels (time effect, p < .001).

Carcase markers and tissue mass mobilization
Figure 3(B-D) illustrates the dynamics of renal fat thickness, lumbar subcutaneous fat and loin area during the experimental interval.No differences were observed between the two groups in the two carcases AT markers.In both parameters the time has an effect on the measurements (p < .001),showing an abrupt drop at delivery.In multiparous animals, an earlier mobilisation of the kidney fat was observed before kidding (Figure 3(B)), and the recovery of fat deposits was continuous during the fourth week postpartum for both groups.In subcutaneous loin fat (Figure 3(C)), there was also a significant interaction (p < .001) between the group and time, where the nulliparous group began fat recovery from the second week after delivery.
Regarding the area of the loin (Figure 3(D)), although the overall mean was higher in the multiparous group than in the nulliparous group (414.4 ± 10.9 mm 2 vs. 355.4± 3.6 mm 2 ; p < .001), it presented a different dynamic (interaction effect, p < .001),characterised by an abrupt drop at delivery and lower area values (p < .01)during the postpartum period.
Table 3 illustrates the mobilisation dynamics before, during and after the delivery of muscle tissue measured as loin depth and fat tissue as the sum of subcutaneous and kidney fat thicknesses.The multiparous animals recorded higher mean values (p < .05) of muscle thicknesses in the three periods considered and a higher fat mass after delivery when compared to nulliparous group parity (4.7 ± 0.4 mm vs. 3.4 ± 0.2 mm; p < .05).All groups showed a reduction in fat and muscle mass at delivery and recovery after parturition, except the muscle in nulliparous animals.

Discussion
The peripartum period is an essential period for maintaining the health and productive performance of goats (Huang et al. 2021).Meeting the energy demand in this period, especially to ensure the final pregnancy, physiological return of the reproductive tract and milk production, requires metabolic adaptations that compensate for deficiencies during NEB, such as the mobilisation of body reserves, which contributes to the supply of lipids and amino acids.(Sch€ aff et al. 2013).The mobilisation of body reserves comes primarily from AT, with a consequent increase in plasma concentrations of nonesterified FA (Sch€ aff et al. 2013) and skeletal muscle tissue, reflected in increased levels of circulating ketone bodies (Zou et al. 2021).
In our study, carcase markers proved to be efficient in describing tissue mobilisation during the NEB period and identifying differences in response between parity classes.Both groups confirmed that for singleton pregnancies in non-dairy animals, the main tissue mobilisation prior to the metabolic effort occurs shortly after delivery and continues until the early postpartum weeks.We observed greater muscle wasting after delivery and an earlier mobilisation of visceral AT, which started before parturition.These results appear to have been the differential of multiparous and are the main justification for higher milk production and consequently better performance of their kids during the experimental interval.Skeletal muscle is an important protein store (van der Drift et al. 2012), and its catabolism reflects plasma concentrations of branched-chain amino acids or 3-methyl histidine (3-MH; Nicastro et al. 2012), used not only for the synthesis of milk proteins but also for hepatic gluconeogenesis.
Marked AT lipid metabolism promotes a decrease in FA synthesis rate and changes in the activity of the LPL enzyme that participates in FA uptake and synthesis (Chilliard et al. 2003), as well as a decrease in the expression of genes that encode important enzymes for AT metabolism in ruminants, such as CPT-1, CPT-2 and ACSL-1, enzymatic proteins related to FA oxidation, regulated by PPAR, and lipogenic genes, such as ACC, FAS and SCD-1, which are regulated by SREBP-1c (Bordoloi et al. 2019).Therefore, the different efficiency in the muscular mobilisation of multiparous goats could have been a consequence of a higher protein stock due to age, as well as a higher efficiency of the metabolic routes used in this phase.
A large mobilisation of body FA promotes a reduction in nutrient concentrations, including a reduction Table 2. Means and standard errors of metabolite dynamics patterns in goats with different parity (multiparous vs. nulliparous) performed before delivery (BDP: one to five weeks before), at delivery (DP: day 2), and after delivery (ADP: first to fourth weeks)..1844GPT, U/L 21.9 ± 0.70 24.0 ± 3.00 20.30 ± 1.10 18.8 ± 1.10 20.7 ± 2.60 17.9 ± 1.50 .0137.2191.9704Note: GGT: gamma-glutamyl transferase; GOT: glutamic-oxaloacetic acid transaminase; GPT: glutamic-pyruvic acid transaminase; BDP: before delivery period; DP: delivery period; ADP: after delivery period.PG: ANOVA effect for the parity group; Period: ANOVA effect for an interval of time assessment used (BDP, DP, ADP).a, b: Columns with different letters between periods for each parity group, differ (p < .05);A, B: Columns with different letters between parity groups for each period.period.Table 3. Means and standard errors of muscle and fat mass mobilisation in goats with different parity (multiparous vs. nulliparous) performed before delivery (BDP: one to five weeks before), at delivery (DP: day 2), and after delivery (ADP: first to fourth weeks).in glucose abundance.In skeletal muscle, FA oxidation is promoted by leptin, produced predominantly by AT (Fuentes et al. 2010), which justifies the efficient synchronisation between AT and skeletal muscle metabolism.
During the peripartum period, if the energy supply is insufficient, the production rate of hepatic ketone bodies increases disproportionately, which can cause excessive accumulation of BHB in the liver and blood, eventually resulting in pregnancy toxaemia (Zou et al. 2021).The metabolic picture presented in this study is a consequence of the metabolic events described.However, the production of BHB was similar between the two groups, following the pattern of NEB, with a peak in the second week of lactation.Based on these values, there were no cases of hyperketonaemia in the study, according to the reference values for this metabolic condition already established for the species (Huang et al. 2021).Greater tissue mobilisation and milk production in multiparous resulted in more sustained glycaemia, higher concentrations of cholesterol and protein in the blood, and consequently, greater hepatic effort reflected in higher values of GGT and GPT, which may suggest an increase in the formation of reactive oxygen species (ROS) that react with organic molecules to produce reactive oxygen metabolites and consequently promote oxidative stress and cytotoxicity by harming cellular macromolecules such as proteins and nucleic acids (Huang et al. 2021).
During late pregnancy and early lactation, dairy cow face difficulties maintaining homeostasis imposed by metabolic adaptations corresponding to energy deficits and AT and muscle catabolic pathways, which can trigger various metabolic disorders such as ketosis and hypocalcaemia (Goff and Horst 1997) throughout the transition period.Increased ROS may reflect the production of antioxidant defense agents (Celi et al. 2010), such as the enzyme GPx, which is considered an indicator of oxidative stress.In this study, changes in GPx activity were similar to the results of Radin et al. (2015), in which higher-parity goats presented higher levels of metabolic intensity and, consequently, varied levels of oxidative stress during the peripartum period.Changes in the oxidative state may be associated with parity in dairy goats during the transition period (Radin et al. 2015).In the present study, the plasma level of GPx in multiparous goats was higher one week after calving than in nulliparous goats, decreasing after the third week.This profile found in multiparous goats was similar to the findings of Celi et al. (2010), who evaluated goats aged between four and five years and observed an increase in plasma ROS concentrations reaching their highest levels two weeks after parturition; consequently, GPx levels remained high in this period when they decreased significantly between weeks 2 and 4 postpartum.These values may explain the greater metabolic effort exerted by multiparous females with higher milk production to maintain redox homeostasis during lactation.Furthermore, since circulating ketone bodies are taken up by the mammary gland and incorporated into milk fat, the postpartum decrease in circulating BHB from the second week onwards can be explained as it occurs during peak milk production.(Radin et al. 2015).
Despite increased oxidative stress and metabolic effort, no differences were observed with respect to the reproductive tracts of the placenta and cotyledons or in the process of uterine involution in the postpartum period.Following the pattern of the species, the latter showed recovery between days 14 and 21 postpartum (Ababneh and Degefa 2005) and was concluded from this in both animals.Zongo et al. (2015) reported complete uterine involution in goats between 18 and 22 days postpartum.In small ruminants, when lochia is no longer present, the width of the endometrium and myometrium decreases markedly from day 14 postpartum, and uterine involution is complete until day 28 (Gray et al. 2003).The diameter of the uterine artery also decreased with time following uterine involution, as described by Elmetwally and Bollwein (2017).
Regarding the hemodynamic results, our findings reflected the physiological process that occurs in this species during postpartum and did not show an association with the metabolic condition.During pregnancy, in small ruminants, there is a marked increase in the volume of uterine blood flow and a decrease in the resistance index (RI) to meet increasing nutritional demands of the developing foetus and placenta (Elmetwally and Bollwein 2017).Important changes occur in blood flow to the uterus.Dramatic hemodynamic changes during uterine involution in the nonpregnant state are poorly understood.However, it has already been shown that during the postpartum period, not only in sheep and goats (Elmetwally and Bollwein 2017) but also in cattle (Krueger et al. 2009;Heppelmann et al. 2013) and women, there is a reduction in blood flow while RI increases, as observed in our study.Elmetwally and Bollwein (2017) observed a significantly lower uterine blood flow as early as the first postpartum week in goats and sheep, with reductions of 80% and 70%, respectively.A similar reduction in blood flow occurs in women during the first postpartum week (Van Schoubroeck et al. 2004) and in cows during the first five days (Heppelmann et al. 2013).These changes in small ruminants have been attributed to a change in the size and weight of the uterus, which after parturition, has a 50% reduction and expulsion of most lochia (Heppelmann et al. 2013) and an increase in vascular resistance (Van Schoubroeck et al. 2004).Krueger et al. (2009) observed a moderate reduction in blood flow in cattle until day 28 postpartum.Therefore, this decrease in blood flow in our study can be explained by the uterine involution that occurs successively in our study period, from the fourth postpartum week until the complete recovery of the uterus.

Conclusions
Multiparous goats showed a greater mobilisation of muscle and AT in the peripartum period under the experimental conditions of this study and better performance in terms of milk yield and weight gain in the kids.However, the dynamics of uterine involution during puerperium are not affected by parity in goats.

Figure 1 .
Figure 1.Uterine diameter (A), uterine diameter shrinkage rate (B), systolic peak (C) and resistance index (D) of uterine artery measured after delivery (four weeks) in nulliparous and multiparous goats.The black arrow (A) illustrates the postpartum week in which no significant differences (p > .05)were observed in relation to the reduction in uterine diameter.The data are plotted as mean ± SEM.The p value for the parity group effect is shown in each figure.* Bars with asterisks indicate where difference between parity (p < .05)occurred.

Figure 2 .
Figure 2. b-hydroxybutyrate (A), glutathione peroxidase (B), cholesterol (C) and total protein (D), serum levels measured from five weeks before delivery to four weeks after delivery in nulliparous and multiparous goats.Data are plotted as mean ± SEM.The p value for the parity group effect is shown in each figure.* Bars with asterisks indicate where difference between parity (p < .05)occurred.

Figure 3 .
Figure 3. Milk yield (A) performed after delivery.Kidney fat thickness (B), subcutaneous loin fat thickness (C) and loin area (D) from five weeks before delivery to four weeks after delivery in nulliparous and multiparous goats.Data are plotted as mean ± SEM.The p value for the parity group effect is shown in each figure.* Bars with asterisks indicate where difference between parity (p < .05)occurred.

Table 1 .
Means and standard errors of in vivo performance of does and kids from delivery to weaning (four weeks postpartum) and reproductive tract traits recorded at delivery classified according to the parity group (nulliparous and multiparous).Days from delivery.PG, ANOVA effect for the parity group.
Loin depth.ÃÃ Sum of kidney and subcutaneous loin fats thickness.BDP: before delivery period; DP: delivery period; ADP: after delivery period; PG: ANOVA effect for the parity group; Period: ANOVA effect for an interval of time assessment used (BDP, DP, ADP).a, b: Columns with different letters between periods for each parity group, differ (p < .05);A, B: Columns with different letters between parity groups for each period.period.