Salivary pH and gastric ulcers in gestating and lactating sows

<b>General information</b>This study was carried out over 19 weeks between July and December 2021 at SRUC’s Pig Research Centre (SRUC; Scotland’s Rural College) and took place during the last three weeks of gestation and the entire lactation of 38 sows. This study was made possible due to a planned sow culling for management reasons in the herd. The project was reviewed and approved by the Animal Experiments Committee of Scotland’s Rural College (SRUC).All tables mentioned in the text below are provided in the excel file 'Salivary pH and gastric ulcer in sows - Dataset' in the relevant sheet ('Table 1 Diet', 'Table 2 Overall stomach score', and 'Table 3 Lesion score' sheet).<b><i>Study design</i></b>A case-control study was performed to investigate the association between salivary pH and gastric ulcers (identified post-mortem after weaning) in late-gestating and -lactating sows.Animals were selected on the basis of being designated for culling by the farm administration and included in the study as they became available. Animals were included in the trial from day 91 ± 0.6 of gestation (mean ± S.D.) until the end of lactation when their piglets were weaned at 26 ± 1.46 days after farrowing (mean ± S.D.). Sows were moved to the farrowing facility at day 110 ± 0.7 of gestation (mean ± S.D.), and farrowing occurred at day 116 ± 1.6 of gestation (mean ± S.D.). Animals were followed in four batches between July and December 2021 [batch 1 (n = 12) in July-August; batch 2 (n = 10) and 3 (n = 5), September-October; batch 4 (n = 11), November-December]. Saliva samples were taken by the end of gestation (107 ± 0.72 days of gestation) and lactation (25 ± 1.6 days after farrowing). Stomachs were assessed post-mortem approximately 30 min after euthanasia.<b>Animals and housing</b>Thirty-eight primi- and multi-parous sows (hereafter sows; Large White × Landrace) were used in this study (mean parity ± SD, 2.7 ± 1.6; range 1 to 6). Sows were housed in their home commercial pens during gestation and lactation. During gestation sows were housed in pens of 23.2 m² (mean ± S.D.; room temperature: 15.71 ± 4.22 °C; room humidity: 84.78 ± 14.49 %) and provided with ample amount of straw i.e. bedding was always present in abundance (renewed every Monday and Friday). Sows had access to a restricted low-energy commercial diet for gestation (3 kg feed/day; Table 1) in an individual trough once a day (at 0720 h) and one drinker in the pen with <i>ad libitum</i> access to drinking water (bite drinker). Each pen held between 1 to 5 sows (mean 3.8 sows/pen). Lighting was on between 0700 and 2200h.For lactation stage there are two types of housing available at the farm: lactation crate and PigSAFE, and wererestricted by the needs of the farm. Hence, 17 sows were housed in farrowing crates (1.10m² crate within a 3.75 m² pen) and 21 sows were housed in a PigSAFE farrowing pens (8.9 m²). The room temperature and humidity, including both housing types, were 19.49 ± 3.90 °C and 66.75 ± 11.14 % (mean ± S.D), respectively. For the six sows housed in PigSAFE pens, the litters were socialized during lactation (as part of a separate study), by allowing two adjacent litters to mix freely through a gap in the side of the pen from 10 days of age. In both crates and pens sows were provided with ~ 200 to 300 g/sow/day of straw, <i>ad libitum</i> water and two feed rations according to litter size and sow body condition (at 0800 and 1530 h) of standard commercial diet for lactation (Table 1). Artificial lighting was on between 0800 and 1545h. Farm management was maintained as usual during both stages.<b>Saliva sampling</b>Saliva samples were taken at the end of gestation and at weaning before sows were euthanised. Batch 1 was not sampled at all, and Batch 2 was only sampled at weaning due to logistics.Saliva samples were taken at the end of gestation (batch 3 and 4; n = 16), and at weaning when sows were euthanised; sampling occurred on the day before (batch 2 and 4; n = 21) or the same day of euthanasia (batch 3; n = 5) due to logistics. During gestation, samples were taken between 0815 and 0830 h after sows had eaten all their daily ration (fed at ~ 0720 h; n = 16). During lactation, saliva samples were taken between 0730 and 0800 h before their morning ration (fed at 0800 h; n = 21). Batch 3 sampling (lactation) occurred after feeding (n = 5) as feeding occurred at 0715 h instead of the usual time at 0800 h.Saliva was collected by allowing the sows to chew on a cotton swab (Millpledge Veterinary, DX09396) until saturated (typically 20 seconds). Collected swabs were placed in Salivette® tubes (Sarstedt Ag &amp; Co, Germany) and placed in a styrofoam box with ice blocks at ~ 8°C. To extract saliva, swabs were centrifuged ~ 30 min after collection (~ 4°C for 5 min at 3000 rpm). Samples were refrigerated for transport to the lab and subsequently frozen at −80°C (~ 4 to 5 h after collection).<b>pH measurement</b>The pH of the saliva samples was measured using an electronic pH meter (PerpHecT® ROSS® Micro Combination pH Electrode and Fisher Scientific accumet® AE150) after a three-point calibration at pH 4, 7 and 10. Samples were transported in a Styrofoam box with dry ice from the storage freezer (- 80°C) to the laboratory and defrosted on ice and then at room temperature. The samples were assessed in a random order. Before and after measuring pH, the probe was cleaned with double distilled water. For those sows that had an extra Eppendorf tube of sample (i.e. the amount of saliva collected from the cotton swap was enough to fill in two Eppendorf tubes) the pH of both tubes was measured separately and then the values were averaged.<b>Stomach sampling</b>Euthanasia occurred immediately after weaning (mean ± S.D.; days after farrowing 26 ± 1.46) when each sow was removed from her farrowing pen, with her piglets remaining behind as usual.For euthanasia, sows were walked individually into an adjacent building which was not occupied by any other pigs. Before euthanasia, sows were sedated with a mixture of 6.25 ml azaperone (Stresnil; 40 mg/ml), 2.5 ml medetomidine hydrochloride (Domitor; 1 mg/ml), 25 ml midazolam (Hypnovel; 5 mg/ml) and 25 ml ketamine (Ketamidor; 100 mg/ml) by intramuscular injection in both sides of the neck by trained personnel. Total ml reported here is based on a weight of 250 kg and dosage varied depending on sows’ response to sedation. Once fully sedated, a trained personnel euthanised the sow by injecting a lethal dose of pentobarbital sodium (Euthatal; 200 mg/ml) into the heart (mean ± SD; 78.03 ± 33.96 ml). Death was confirmed via heart rate monitoring with a stethoscope and lack of a corneal reflex. Once death was confirmed, the abdominal cavity was opened, and the stomach was cut out by trained personnel. The carcass was removed from the pen and the pen was cleaned by clearing up any faeces, urine and/or blood and disinfected with disinfectant powder (Staldran) before the next sow was brought into the facility.<b>Assessment of the pars oesophagea</b>The stomachs were assessed ~ 30 min after euthanasia. Stomachs were opened through the greater curvature and washed gently with running cold water. The integrity of the pars oesophagea was evaluated by using an overall stomach score (Table 2) and a lesion score (Table 3). The overall stomach score assesses the overall condition of the pars oesophagea by only taking into consideration the most severe lesion present (least to most severe lesion: keratinization, erosion, ulcer, healing tissue, contraction of the oesophagus). The lesion score describes each of these lesions separately in terms of extension (keratinization, erosion, ulcer and contraction of the oesophagus) or type (healing tissue). When referring to both overall stomach score and lesion score they are named as stomach integrity.<b>Statistical analysis</b>The experimental unit was the individual sow, confidence interval was set at 95% and significance level at 0.05. For stomach integrity data, the scores assessing each lesion separately were coded as presence/absence of lesions because of a low number of cases. Also, the score for oesophageal contraction was not analysed as only one sow had a contraction of the oesophagus. Distribution of the data was assessed using histograms.The effect of presence or absence of a lesion on the salivary pH was investigated by using a two-sample t-test (Minitab 17.1.0, 2013 Minitab, LLC). This was tested for gestating and lactating sows separately. The association between the overall stomach score and salivary pH in both gestating and lactating sows was tested using Spearman’s correlations (Minitab 17). Differences in salivary pH between productive stages (gestation vs lactation salivary pH) was investigated using a paired t-test (Minitab 17) for the same sow.The effect of farrowing environment (farrowing crate vs PigSAFE) and parity number (parity 1 + 2 vs 3 or more parities) on stomach integrity was investigated using SPSS 28 regression models. Parity number was grouped into sows with 1 and 2 parities, and sows with 3 or more parities. This was to satisfy the assumptions of the models of minimum number of cases per level of categorical variable. The effect of farrowing environment and parity on the overall stomach score was investigated using an ordinal regression model, and their effect on the presence or absence of a type of lesion (lesion score) was investigated using binomial regression model.To study the association between stomach integrity (overall stomach score and ulcer extension score) and parity number of the sow a Spearman’s correlation test (two-tailed) was used with SPSS 28. As required in the settings of this test in SPSS 28, both scores were set as ordinal variables (Measure), and additionally as numeric (Type). For this analysis parity number was not re-grouped. The association between both scoring systems (lesion score and overall stomach score) was tested the same way.Health-related measures were not included in any of the models as signs of ill health were sporadic and short duration.