- Email: [email protected]
Determination of reference intervals for umbilical cord arterial and venous blood gas analysis of healthy Thoroughbred foals
Theriogenology 118 (2018) 1e6
Contents lists available at ScienceDirect
Theriogenology journal homepage: www.theriojournal.com
Determination of reference intervals for umbilical cord arterial and venous blood gas analysis of healthy Thoroughbred foals Sunita S. Jeawon a, *, Lisa M. Katz a, Noreen P. Galvin b, Ursula M. Fogarty c, Vivienne E. Duggan a a b c
UCD School of Veterinary Medicine, Veterinary Sciences Centre, University College Dublin, Belfield, Dublin 4, Ireland Phoenix Equine Group, Kildare, Co. Kildare, Ireland Irish Equine Centre, Johnstown, Co. Kildare, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 March 2018 Received in revised form 18 May 2018 Accepted 21 May 2018 Available online 25 May 2018
Although umbilical cord blood gas analysis is considered the best way to assess in utero oxygenation in human neonates, there is limited evaluation of this method in equine neonatology. Our objectives were to assess the practicality of obtaining umbilical cord blood gas samples in the field and to determine umbilical cord arterial and venous blood gas reference intervals (RI) for healthy, newborn foals. Thoroughbred foals >320 days gestation from healthy mares with uneventful pregnancies at one stud farm were evaluated. All parturitions were observed, with paired umbilical arterial and venous whole-blood samples obtained immediately following parturition for blood gas and lactate concentrations measured in duplicate. Apgar scores were assigned immediately and 10 min after birth, with all foals subsequently examined on days 1e28 to monitor for development of perinatal asphyxia syndrome. Foals were excluded from analysis based on abnormalities of stage 2 labour, Apgar scores and gross and histological placental assessment. Data was analysed using a Student's t-test, Pearson's correlation and the Robust method with P 0.05 significant. Umbilical cord samples were simple to obtain with minimal disruption to the foaling environment. Of the n ¼ 34 foals assessed, n ¼ 7 were excluded based on premature placental separation deliveries. The mean time for stage 2 labour and blood gas analysis after parturition was 17.3 ± 5.1 min and 5.0 ± 2.3 min, respectively. RI were identified for umbilical arterial and venous pH (7.19e7.42 vs. 7.34e7.44), PO2 (15.5e48.39 mmHg vs. 16.6e52.7 mmHg), PCO2 (49.5e82.29 mmHg vs. 45.4e63.1 mmHg), SO2 (9.19e76.89% vs. 39.9e84.88%), bicarbonate (27.3e38.7 mmol/l vs. 27.7 e37.8 mmol/l), base excess (0.36e12.9 mmol/l vs. 1.97e13.1 mmol/l), TCO2 (28.99e40.3 mmHg vs. 29.0 e39.5 mmHg) and lactate (1.4e7.3 mmol/l vs. 1.3e4.9 mmol/l). Umbilical arterial samples had lower pH (P < 0.0001), PO2 (P ¼ 0.002) and SO2 (P < 0.0001) and higher PCO2 (P < 0.0001) and lactate (P < 0.0001) than venous samples. The initial Apgar score was positively correlated to umbilical arterial SO2 (r ¼ 0.4, P ¼ 0.05) and negatively with umbilical arterial TCO2 (r ¼ 0.6, P ¼ 0.004). Overall, umbilical cord sampling was simple and minimally disruptive, with RI obtained for blood gas measurements. RI for umbilical blood gas measurements from a larger population of healthy and unhealthy foals is required to evaluate the accuracy of this method for assessing in utero oxygenation. © 2018 Published by Elsevier Inc.
Keywords: Blood gas analysis Umbilical cord Equine neonate Hypoxia Perinatal asphyxia syndrome
1. Introduction Arterial and venous blood gas analysis can be used as a diagnostic tool in veterinary medicine to help evaluate an animal's
* Corresponding author. E-mail addresses: [email protected] (S.S. Jeawon), [email protected] (L.M. Katz), [email protected] (N.P. Galvin), ufogarty@ irishequinecentre.ie (U.M. Fogarty), [email protected] (V.E. Duggan). https://doi.org/10.1016/j.theriogenology.2018.05.024 0093-691X/© 2018 Published by Elsevier Inc.
oxygenation, ventilation and acid-base status [1]. In human neonatology, umbilical cord blood gas analysis immediately at the time of birth is thought to provide the most sensitive representation of neonatal acid-base status prior to birth [2]. Hypoxic syndromes have been described in human neonates [3,4] with the combination of Apgar scoring [5,6] and umbilical cord blood gas analysis used by neonatologists to assess the likelihood that a hypoxic event occurred in utero, either during parturition or more chronically during the pregnancy [2,7,8]. These assessments help
2
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
identify at-risk neonates, allowing for early medical intervention [9e11], reduction in morbidity and mortality rates as well as adverse clinical sequelae, in particular related to neurodevelopmental disease [12e14]. Similar to the hypoxic syndromes seen in human neonates, perinatal asphyxia syndrome (PAS) in neonatal foals is likely caused by hypoxic-ischemic damage that occurred during pregnancy or parturition [15]. Whilst originally thought to primarily affect just the central nervous system, it is now known to affect multiple other body systems [16]. The overall occurrence of PAS is cited to be between 1 and 2% of all foals born [17]. Although numerous risk factors have been identified for PAS, currently there are no confirmed pathognomonic biological or biochemical parameters that can be used to indicate the presence of this disease [18]. Researchers have described associations between decreased thyroid hormones [19], increased systemic creatinine and decreased systemic glucose concentrations [17,20] with risk of neonatal illness such as PAS. Ringger et al. (2011) also reported on increased concentrations of the biomarker UCHL1 in foals affected by PAS [21]. However, affected foals are more commonly not identified until after clinical signs develop, at which point significant systemic damage by inflammatory mediators may have already occurred [15]. Although the use of arterial and venous blood gas analysis in equine neonatology has been described in the literature [22e24], to-date, the use of umbilical cord blood gas analysis in equine neonates has only been reported by one group in which umbilical arterial blood gas parameters were evaluated in prematureinduced and term-induced foals [25]. Therefore, the purpose of this study was a) to investigate the practicality of obtaining umbilical cord blood samples from newborn foals on a stud farm and b) determine umbilical cord arterial and venous blood gas reference intervals (RI) in a group of healthy, newborn Thoroughbred (Tb) foals. 2. Materials and methods University College Dublin's Animal Research Ethics Committee approved this study. Owner consent was also granted. 2.1. Sample population During the 2017 breeding season, Thoroughbred broodmares with no history of systemic illness or placentitis during pregnancy and a gestational length >320 days were included in the study; exact gestational length was recorded for each mare once they foaled. All mares were from the same stud farm and managed under similar circumstances. The mares were closely monitored by on-site staff 24-hours a day, with at least 2 trained staff members attending each foaling. After foaling, data from foals of mares in the study group were retrospectively excluded from analysis if there was any abnormality of the second stage of parturition (e.g., premature placental separation [PPS]), if the Apgar score of the foal was <6 at each assessment, if the clinical appearance of the foal after parturition and/or the gross and histopathological appearance of the placenta were abnormal. Data from excluded foals are only reported for comparison. 2.2. Experimental protocol All mares were observed from outside the stable as they went through stage 1 labour. The time of chorioallantoic membrane rupture was noted, with the mares then allowed to progress through the second stage of labour with minimal interference, although all mares had one staff member guiding the foal's head
and one applying gentle traction on its legs until the foal was safely delivered. The duration of stage 2 of parturition was recorded for every animal, calculated from the time of chorioallantoic membrane rupture to the birth of the foal. One author (SSJ) attended all parturitions and obtained all immediate post-partum clinical and umbilical cord blood samples. As soon as the foal was expelled, the umbilical arteries and vein were identified via visual inspection and manual palpation. For each blood gas sample measurement, 1 ml of blood was collected anaerobically into a heparinised disposable blood gas syringe (RAPIDLyte, Cruinn Medical Ltd, Dublin, Ireland). The umbilical cord was not clamped for sampling purposes, with the arterial sample always obtained first from the largest umbilical artery (as this was the easiest one from which to obtain a blood sample) using a 21 g needle, followed by venous blood collection using a 23 g needle. The timing of each sample collection in relation to foetal expulsion was recorded. Rectal temperature was taken from each foal to temperature-correct blood gas analysis. Immediately following umbilical cord blood collection, a 4parameter Apgar score was assigned to the foal using a modified system previously described by Vaala (1994) [26] (Table 1), with a second Apgar score assigned 10 min after the initial one; scores of 6 were considered to be within normal limits. Immediately after completion of the initial Apgar score, umbilical blood samples were evaluated in duplicate using a portable blood gas analyser (Vetscan i-Stat® 1, Abaxis UK Ltd, York, United Kingdom). Parameters measured included pH, PCO2, PO2, HCO3, TCO2, SO2, base excess/ deficit and lactate. The arterial sample was always evaluated first followed by the venous sample. The time from blood sampling to analysis was recorded. For all mares, the time to passage of the foetal membranes was recorded, with the membranes immediately collected for analysis. The membranes were initially laid out on a designated table to visually inspect that the entire placenta had been passed, with the presence of any tears noted before being weighed on-site with commercial weighing scales. The foetal membranes were then submitted for full gross and histopathological evaluation. If parturition occurred during the weekend, the foetal membranes were stored in a cold room until the subsequent Monday. All gross and histopathological evaluations were performed by one author (UMF) blinded to any information about the parturition other than the date of foaling and weight of the foetal membranes. Approximately 2 cm2 of pregnant horn/non-pregnant horn body pouch, cervical area, allantoamnion and amniotic cord were taken for histopathological examination. Any pathological findings were recorded as absent, mild, moderate or severe (Supplementary Tables 1e3). The foetal membranes were then classified into group 1 (absent or mild pathological findings) or group 2 (at least one section characterised by moderate-to-severe pathological findings). Data from foals with foetal membranes placed into group 2 were excluded from analysis. Complete physical examinations were carried out on all foals at days 1, 2, 3, 7, 14, 21 and 28 by 1 of 2 authors (SSJ or NPG) to monitor for development of PAS. A blood sample was taken from all foals between 10 and 14 h after birth for measurement of serum IgG concentrations. 2.3. Data analysis Calculations and analyses were performed using MedCalc statistical software (www.medcalc.org, MedCalc Software, v. 13.1.00, Mariakerke, Belgium) and SPSS (IBM SPSS Statistics for Windows, v. 24.0, Armonk, NY: IBM Corp). Measurements were evaluated for normality using a D'Agostino-Pearson test for normal distribution. Boxplots and histograms were assessed and any extreme outliers identified. The measurements were further assessed using Reed
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
3
Table 1 Apgar Score Parameters adapted from Vaala (1994) [26]. Score
2
1
0
Heart rate Respiratory rate Muscle tone, postural responses Response to stimuli
More than 60 BPM and regular Regular Maintains sternal recumbency and attempts to stand Head shake, sneeze or cough
Less than 60 BPM or irregular Irregular Some flexion of limbs and muscle tone
Absent Absent Laterally recumbent, no muscle tone No response to stimulation
Weak ear movement when stimulated, facial grimace
et al. (1971) and Tukey (1977) outlier tests with extreme outliers removed. For the umbilical blood samples, coefficients of variation (CV) from duplicate measurements for each analyte were determined. Reference intervals were then calculated by a Robust Method (Clinical and Laboratory Standards Institute guidelines C28-A3) [27] with 90% confidence intervals (CI) calculated. Reference interval values are presented as median ± CI. All other values are reported as the mean ± sem. Pearson's correlation coefficient was used to evaluate significance and strength of associations between the blood gas analysis variables and foaling and physical parameters, with r 0.4 signifying correlation. Pairwise scatter plots were constructed between all variables to confirm linear distribution. Parameters were considered to be moderately associated with each other if r 0.5 and strongly associated if r 0.7 [28]. Univariate analysis was carried out using a Student's t-test to compare between the umbilical cord arterial and venous blood gas analysis parameters. Significance was set at P 0.05. 3. Results A total of n ¼ 34 mares and associated foals (n ¼ 24 females, n ¼ 10 males) were evaluated. The mean age of the mares was 9 ± 3.5 (range 4e21) years, with an average of 5 parities (range 1e14). All foals were full-term (>320 days old) with a mean gestational age of 344 ± 8 (range 329e363) days. The mean birth weight was 56 ± 7.6 (range 37e67) kg and 58 ± 8 (range 44e69) kg for the filly and colt foals, respectively. The mean length of stage 2 labour was 17 ± 5 (range 9e26) minutes. Of the n ¼ 34 foals assessed, n ¼ 7 (n ¼ 4 females and n ¼ 3 males) were excluded from analysis based on PPS deliveries. Initial Apgar scores were carried out on average 2.9 (range 1e6) minutes after the foal was born. Initial Apgar scores were 4 (n ¼ 1), 6 (n ¼ 3), 7 (n ¼ 3) or 8 (n ¼ 27). All foals with PPS had initial Apgar scores of 8 except for 1 foal, which had an initial Apgar score of 4; this latter foal was retrospectively identified to come from a mare
with a history of having compromised foals at birth, and therefore deemed to be high-risk. Second Apgar scores were completed on average 10.9 ± 1.2 (range 8e13) minutes after the initial one. All foals had a 10-minute Apgar score of 8 except for the foal from the mare deemed to be high-risk, which had a score of 6. Time to expulsion of the foetal membranes after parturition was on average 31 ± 14.6 (range 4e69) minutes and the mean placental weight was 7 ± 1.5 (range 4e11) kg. The average total cord length was 52 ± 8.9 (range 40e70) cm and 50.1 ± 9.9 (range 35e66) cm for the normal foals and PPS foals, respectively. Of the normal foals, 15/27 (56%) were from right horn pregnancies and 12/27 (44%) were from left horn pregnancies. Of the PPS foals, 6/7 (86%) were from right horn pregnancies and 1/7 (14%) from left horn pregnancies. Of the n ¼ 34 foals in the study, only 1 foal (from the high-risk mare) showed clinical signs of PAS in the first 72 h of life. All other foals were clinically normal for the first 28 days of life. The average time from birth to the first umbilical cord sample acquisition was 1.2 ± 0.8 (range 1e5) minutes. The average time from sampling to analysis was 5.0 ± 2.3 (range 2e12) minutes. Each assay took 120 s to complete after the patient details had been entered. Paired arterial and venous samples were obtained from n ¼ 29 foals, arterial samples alone from n ¼ 3 and venous samples alone from n ¼ 2. Based on the constructed boxplots, 2 measurements were excluded from analysis as extreme outliers. Summary statistics are shown in Tables 2 and 3 for the normal foals and in Tables 4 and 5 for the foals excluded from analysis due to PPS. Tables 6 and 7 summarise RI identified for arterial and venous samples. In comparison to venous samples, arterial samples had a significantly lower pH (7.39 ± 0.03 vs. 7.32 ± 0.05, P < 0.0001), PO2 (34.4 ± 8.19 mmHg vs. 29.2 ± 11.5 mmHg, P ¼ 0.002) and SO2 (62.0 ± 15.2 %vs. 43.9 ± 18.1%, P < 0.0001) and significantly higher PCO2 (54.1 ± 4.66 mmHg vs. 64.5 ± 6.84 mmHg, P < 0.0001) and lactate concentration (3.19 ± 0.83 mmol/l vs. 4.42 ± 1.36 mmol/l, P < 0.0001). There were no significant differences between venous and arterial bicarbonate (32.8 ± 2.38 mmol/l vs. 33.2 ± 2.75 mmol/ l), base excess (7.56 ± 2.63 mmol/l vs. 7.1 ± 3.18 mmol/l) and TCO2
Table 2 Summary statistics for paired umbilical cord arterial and venous samples from n ¼ 27 healthy Thoroughbred foals. pH Arterial
Venous
PCO2 (mmHg)
PO2 (mmHg)
Arterial
Base Excess (mmol/l)
HCO 3 (mmol/l)
SO2 (%)
Venous
Arterial
Venous
Arterial
Venous
Arterial
Venous
Arterial
Minimum 25% quartile Median ± SD
7.22 7.33 49.95 7.28 7.37 60.0 7.30 ± 0.05 7.39 ± 0.03 66.0 ± 7.7
45.8 51.8 53.7 ± 4.1
14.0 29.4 36.0 ± 8.6
1.5 4.8 7.0 ± 2.97
3.0 5.4 7.5 ± 2.6
27.9 31.4 33.4 2.7±
28.0 31.1 32.9 ± 2.4
18.0 45.0 31.0 54.3 44.0 ± 16.0 66.3 ± 12.1
75% quartile Maximum Mean ± SE
7.35 7.40 70.7 7.43 7.43 76.95 7.32 ± 0.01 7.39 ± 0.01 65.6 ± 1.6
58.1 63.1 54.3 ± 0.8
38.5 55.0 36.0 ± 1.7
8.5 14.5 7.0 ± 0.6
9.3 13.0 7.5 ± 0.5
34.9 39.9 33.3 ± 0.5
34.99 37.4 32.7 ± 0.5
55.8 82.0 42.4 ± 3.2
70.3 86.5 63.7 ± 2.5
95% CI of duplicates CV of duplicates (%)
0.0687 e0.1253 0.097
1.4695 e2.6972 2.1
16.0 23.0 28.0 (23.1 e30.9) 33.5 55.5 28.0 (24.7 e30.9) 3.6100 e6.6839 5.1
4.0132 e7.4423 5.7
9.8020 e18.6011 14.1
8.7081 e16.4543 12.5
1.4313 e2.6266 2.0
1.4470 e2.6556 2.1
5.2906 e9.8618 7.6
2.8931 e5.3409 4.1
0.0492 e0.0897 0.07
1.5833 e2.9073 2.2
SD: standard deviation; SE: standard error; CI: confidence interval; CV: coefficient of variation.
Venous
4
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
Table 3 Summary statistics for paired umbilical cord arterial and venous samples from n ¼ 27 healthy Thoroughbred foals.
Minimum 25% quartile Median ± SD 75% quartile Maximum Mean ± SE 95% CI of duplicates CV of duplicates (%)
Arterial TCO2 (mmol/l)
Venous TCO2 (mmol/l)
Arterial lactate (mmol/l)
Venous lactate (mmol/l)
29.5 33.3 34.5 ± 2.7 36.3 41.5 35.4 ± 0.5 5.2135e9.7151 7.8
29.0 32.8 34.7 ± 2.4 36.5 39.0 34.2 ± 0.5 1.5727e2.8878 2.2
1.95 3.4 4.2 ± 1.4 5.5 7.2 4.36 ± 0.3 1.4585e2.6768 2.1
1.98 2.6 2.99 ± 0.8 3.7 4.95 3.2 ± 0.8 1.0649e1.9514 1.5
SD: standard deviation; SE: standard error; CI: confidence interval; CV: coefficient of variation.
Table 4 Summary statistics for paired umbilical cord arterial and venous samples from n ¼ 7 Thoroughbred foals from mares with premature placental separation. pH
Minimum 25% quartile Median ± SD 75% quartile Maximum Mean ± SE 95% CI of duplicates CV of duplicates (%)
PCO2 (mmHg)
PO2 (mmHg)
Base Excess (mmol/l)
HCO 3 (mmol/l)
SO2 (%)
Arterial
Venous
Arterial
Venous
Arterial
Venous
Arterial
Venous
Arterial
Venous
Arterial
Venous
7.26 7.298 7.31 ± 0.02 7.32 7.32 7.30 ± 0.008 0.0177 e0.0845 0.05
7.35 7.37 7.38 ± 0.02 7.39 7.399 7.38 ± 0.009 0.0874 e1.3544 0.7
61.8 62.6 64.6 ± 2.5 67.3 67.8 64.9 ± 0.9 0.6378 e3.0853 1.9
49.8 51.3 52.6 ± 1.3 52.8 52.9 51.97 ± 0.6 0.1685 e2.6249 1.4
18.5 24.9 31.5 ± 15.3 33.8 66.0 33.8 ± 5.8 1.5688 e7.7240 4.6
31.5 32.6 39.0 ± 5.4 40.5 45.0 37.5 ± 2.4 0.6003 e9.6493 5.0
2.0 5.1 6.0 ± 1.9 6.5 8.0 5.6 ± 0.7 2.6631 e13.386 7.9
6.0 6.0 6.5 ± 1.0 7.4 8.5 6.8 ± 0.5 0.7513 e12.211 6.3
29.3 30.99 32.2 ± 1.6 32.8 34.3 31.9 ± 0.6 0.5147 e2.4841 1.5
30.8 31.1 31.3 ± 0.7 32.1 32.6 31.6 ± 0.3 0.1634 e2.5455 1.4
20.5 37.0 49.5 ± 20.7 55.0 87.0 49.6 ± 7.8 2.6822 e13.487 7.95
56.0 58.3 67.5 ± 9.3 73.4 79.0 66.6 ± 4.2 0.5213 e8.3315 4.4
SD: standard deviation; SE: standard error; CI: confidence interval; CV: coefficient of variation.
Table 5 Summary statistics for paired umbilical cord arterial and venous samples from n ¼ 7 Thoroughbred foals from mares with premature placental separation.
Minimum 25% quartile Median ± SD 75% quartile Maximum Mean ± SE 95% CI of duplicates CV of duplicates (%)
Arterial TCO2 (mmol/l)
Venous TCO2 (mmol/l)
Arterial lactate (mmol/l)
Venous lactate (mmol/l)
31.0 33.3 34.5 ± 1.8 35.0 36.5 34.1 ± 0.7 0.6467e3.1290 1.9
32.0 32.4 33.0 ± 0.9 33.4 34.5 33.0 ± 0.4 0.1632e2.5424 1.4
2.4 2.7 3.1 ± 1.4 4.2 6.2 3.5 ± 0.5 0.3389e1.6302 0.98
2.2 2.4 2.7 ± 0.5 3.1 3.5 2.8 ± 0.2 0.1873e2.9225 1.6
SD: standard deviation; SE: standard error; CI: confidence interval; CV: coefficient of variation.
Table 6 Reference interval for n ¼ 25 paired umbilical cord arterial and venous samples from healthy Thoroughbred foals calculated using the Robust method. The confidence intervals for the reference limits were estimated using bootstrapping (percentile interval method Efron & Tibshirani 1993). pH Arterial Lower limit 7.19 90% CI 7.17 e7.22 Upper limit 7.42 90% CI 7.38 e7.46
PCO2 (mmHg)
PO2 (mmHg)
Venous
Arterial
Venous
Arterial
7.34 7.32 e7.35 7.44 7.42 e7.45
49.5 44.76 e54.93 82.29 78.36 e85.99
45.4 42.94 e47.45 63.1 59.63 e65.75
15.5 13.08 e18.71 48.39 40.88 e57.01
Base Excess (mmol/l)
HCO 3 (mmol/l)
SO2 (%)
Venous
Arterial
Arterial
Venous
Arterial
Venous
16.6 11.83 e23.48 52.7 46.89 e58.54
0.36 1.97 1.23e2.30 0.68e3.47
27.3 25.63 e28.97 38.7 36.83 e40.51
27.7 26.48 e29.16 37.8 36.38 e39.07
9.19 0.125 e18.27 76.89 65.73 e86.78
39.9 31.35 e47.49 92.1 84.88 e96.66
12.9 11.04 e14.96
Venous
13.1 11.50 e14.54
CI: confidence interval.
(34.2 ± 2.44 mmHg vs. 35.4 ± 3.54 mmHg). Initial Apgar score was positively correlated to arterial SO2 (r ¼ 0.4, P ¼ 0.05) and negatively correlated to TCO2 (r ¼ 0.6, P ¼ 0.004). 4. Discussion To the authors' knowledge, this is the first paper that describes obtaining paired umbilical cord venous and arterial blood gas
measurements from naturally born foals in a field setting. The technique of taking the umbilical cord samples was relatively simple, with minimal disruption to the foaling environment. Failure to obtain a paired arterial and/or venous sample in 4 of the foals was either due to the mare standing prematurely and rupturing the umbilical cord before the venous sample could be taken or due to the mare lying awkwardly against the wall as she foaled, preventing adequate access to the umbilical artery. The ease of identifying the
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
5
Table 7 Reference interval for n ¼ 25 paired umbilical cord arterial and venous samples from healthy Thoroughbred foals calculated using the Robust method. The confidence intervals for the reference limits were estimated using bootstrapping (percentile interval method Efron & Tibshirani 1993).
Lower limit 90% CI Upper limit 90% CI
Arterial TCO2 (mmol/l)
Venous TCO2 (mmol/l)
Arterial lactate (mmol/l)
Venous lactate (mmol/l)
28.99 27.43e30.83 40.3 38.37e42.08
29.0 NC 39.5 NC
1.4 0.77e2.09 7.3 6.32e8.05
1.3 0.797e1.76 4.9 4.27e5.43
CI: confidence interval. NC: not calculated, too many equal values in small sample size.
different vessels was aided by the fact that the vein was consistently the largest vessel and the largest artery usually had a palpable pulse. The average time of 5 min between the samples being taken and analysed is unlikely to have skewed the results, as research into the effect of timing on umbilical cord blood sampling in human neonates has shown no significant difference in the results obtained once the sampling has occurred within a 30 min window from birth [2]. It is of great importance to first identify normal reference intervals for umbilical blood gas measurements from healthy foals so that the use of this method for assessing foals for risk of PAS can be investigated. Umbilical arterial blood gas parameters have been previously reported for 8 premature-induced and full-terminduced foals [22], with similar umbilical arterial pH values (7.3 ± 0.01) to those from the present study. However, the umbilical arterial PO2 was higher (43.2 ± 3.9 mmHg) and the PCO2 (53.0 ± 1.8 mmHg) and base excess (0.2 ± 1.2 mmol/l) lower than what was identified in the present study [22]. A primary reason for differences in reported umbilical arterial blood gas values between the studies would be due to differences in experimental design. The study reported by Rose et al. (1982) involved a much smaller sample population of foals in which parturition was induced, with differences in the timing of sampling as well as differences in blood gas analysers used. Although CSLI guidelines [27] recommend using nonparametric methods to determine population-based RIs from at et al. least 120 healthy reference individuals as described by Geffre (2009) [29], limitation to such large numbers in clinical veterinary practice and in particular neonates often result in this being impossible. Because the reference population of interest was welldefined in the present study with inclusion/exclusion criteria used to confirm health, reference intervals could be estimated using a much smaller sample size through the Robust method using bootstrapping, as highlighted in CSLI guideline (C28-A3) [27]. Furthermore, the samples were run in duplicate and were taken in an environment mimicking that in which the majority of practitioners would employ the technique. As all samples were taken by the same person on a single stud farm; external factors such as foaling protocols, management of the mares and sample timing were also all tightly controlled for consistency. The umbilical artery is considered the most important vessel to sample as it represents the foetal acid-base status independent of the maternal circulation, which the venous sample reflects as well [30]. This explains the significantly lower umbilical arterial pH and significantly higher PCO2 and lactate concentration as compared to the umbilical vein measurements in the present study. These differences are similar to those reported for measurements between the two vessels in human neonatology [8]. The benefits of having both arterial and venous samples for blood gas analysis are emphasised in the human literature [2,3,8,30], where it is reported that comparing the venous and arterial blood gas parameters can help identify whether an acute or chronic hypoxic insult has occurred. With acute cord compression or foetal bradycardia, abnormalities will predominantly be seen in the arterial sample, with
a large arterio-venous difference existing between the parameters of both vessels. However, in more chronic hypoxic insults relating to reduced placental transfer of oxygen, both vessels will yield abnormal blood gas measurements with minimal arterio-venous differences [2,3]. Such information could be similarly useful in the assessment of neonatal foals, with equine neonates prone to both acute asphyxial injury at the time of parturition as well as more chronic hypoxia linked with placental insufficiency. Identifying whether there is an acute or more chronic problem may aid in generating a more realistic prognostic outcome. Classification of an acidosis as either primarily respiratory, metabolic or mixed in origin is also important, as a metabolic acidosis is thought to be more harmful than a respiratory one, as it signifies a later stage of foetal hypoxia when anaerobic pathways are taking over to supply energy to the foetus [2,31,32]. An increase in lactate is associated with this process and mirrored by a concurrent drop in the base excess measurement [31]. Umbilical cord lactate concentration has also been shown to correlate significantly with pH, base deficit and PCO2, and to be predictive of low Apgar Scores and certain other short-term morbidities [33]. Although human neonatologists are now largely in agreement in defining damaging foetal acidemia as an umbilical arterial pH of <7.0 and a base deficit of 12 mmol/l, the reported lactate measurements vary in the literature from 3.2 to 7.0 mmol/l [33e36]. Recent work from Tuuli et al. (2014, 2016) proposes an optimal cut-off for both umbilical cord arterial (3.9 mmol/l) and venous (3.4 mmol/l) lactate concentrations, beyond which values are predictive of neonatal morbidity at term [37,38]. Measurement of umbilical cord blood lactate concentrations in equine neonates could similarly give further insight into the presence or absence of a metabolic acidosis and complement the information gained from umbilical cord blood gas analysis findings further. In this study, the Apgar scoring system modified for use in equine medicine [26] appeared to be a useful scoring system for assessing equine neonates. The one truly compromised foal in the present study had a notably different initial Apgar score in comparison to the rest of the study foals. The modified Apgar scoring system was simple to use with clear definitions regarding how the parameters included were to be quantified. Apgar scoring remains a widely used and relevant scale in modern human medicine to help assess the newborn's condition and aid prediction of neonatal survival [6], and should likely be used to a greater degree in equine neonatology. The positive correlation between the initial Apgar score and umbilical arterial SO2 in this study was interesting, as foals that had high in utero SO2 values were likely breathing efficiently immediately following birth and therefore less likely to be marked as a 1 on the respiratory section of the Apgar scale. Conversely, the negative correlation between Apgar score and umbilical arterial TCO2, a measure of CO2 in multiple states (physical solution, proteinbound, HCO3, CO3 and H2CO3), is most likely reflective of the fact that foals with a higher in utero TCO2 were less likely to be breathing as efficiently at birth, and therefore were more likely to be marked as a 1 on the respiratory section of the Apgar scale. These
6
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
findings could indicate that efficiency of ventilation in the immediate period after birth is influenced by the in utero SO2 and TCO2 levels; however, further research into this area is required. 5. Conclusion In summary, this is the first paper that describes paired umbilical cord venous and arterial blood gas measurements from naturally born foals in a field setting. Overall, umbilical cord sampling was simple and minimally disruptive with RI obtained for blood gas measurements. The use of the modified Apgar scoring in tandem with umbilical cord blood gas analysis could be useful as an indicator of in utero oxygenation of the newborn foal. However, reference intervals for umbilical blood gas measurements from a larger population of healthy and unhealthy foals is required to fully evaluate the accuracy of this method for assessing in utero oxygenation and acid-base status. Conflicts of interest The authors declare no conflicts of interest. Acknowledgements The authors wish to thank the management and staff of the stud farm for collaborating on this project, in particular Gretta Clarke, Meaghan Hegarty and Gillian Murray who greatly facilitated ease of sample collection. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.theriogenology.2018.05.024. References [1] Clutton RE. Blood gas analysis. In: McGorum BC, Dixon PM, Robinson NE, Schumacher J, editors. Equine respiratory medicine and surgery. London: Elsevier Health Sciences; 2006. p. 201e9. [2] Blickstein I, Green T. Umbilical cord blood gases. Clin Perinatol 2007;34: 451e9. [3] Bobrow CS, Soothill PW. Causes and consequences of fetal acidosis. Arch Dis Child Fetal Neonatal Ed 1999;80:F246e9. [4] Jonsson M, Agren J, Norden-Lindeberg S, Ohlin A, Hanson U. Neonatal encephalopathy and the association to asphyxia in labor. Am J Obstet Gynecol 2014;211(667):E1e8. [5] Apgar V. A proposal for a new method of evaluation of the newborn infant. Curr Res Anesth Analg 1953;32(4):260e7. [6] Casey BM, McIntire DD, Leveno KJ. The continuing value of the Apgar score for the assessment of newborn infants. N Engl J Med 2001;344:467e71. [7] Dudenhausen JW, Luhr C, Dimer JS. Umbilical cord blood gases in healthy term newborn infants. Int J Gynecol Obstet 1997;57:251e8. [8] Armstrong L, Stenson BJ. Use of umbilical cord blood gas analysis in the assessment of the newborn. Arch Dis Child Fetal Neonatal Ed 2007;92(6): F430e4. [9] Azzopardi D, Brocklehurst P, Halliday H, Levene M, Thoresen M, Whitelaw A. The TOBY study. Whole body hypothermia for the treatment of perinatal asphyxial encephalopathy: a randomised controlled trial. BMC Pediatr 2008;8(1):17. [10] Hobson A, Baines J, Weiss MD. Beyond hypothermia: alternative therapies for hypoxic-ischemic encephalopathy. Open Pharmacol J 2013;7:26e40. [11] Dingle J, Tooley J, Liu X, Schull-Brown E, Elstad M, Chakkarapani E, Sabir H, Thoresen M. Xenon ventilation during therapeutic hypothermia in neonatal
encephalopathy: a feasibility study. Pediatrics 2014;133(5):809e18. [12] van Schie PE, Schijns J, Becher JG, Barkhof F, van Weissenbruch MM, Vermeulen RJ. Long-term motor and behavioral outcome after perinatal hypoxic-ischemic encephalopathy. Eur J Paediatr Neurol 2015;19(3):354e9. [13] Pappas A, Korzeniewski SJ. Long-term cognitive outcome of birth asphyxia and the contribution of identified perinatal asphyxia to cerebral palsy. Clin Perinatol 2016;43(3):559e72. [14] Adhikari S, Rao KS. Neurodevelopmental outcome of term infants with perinatal asphyxia with hypoxic ischemic encephalopathy stage II. Brain Dev 2017;39(2):107e11. [15] Gold JR. Perinatal asphyxia syndrome. Equine Vet Educ 2015;29(3):158e64. [16] Vaala WE. Perinatal asphyxia syndrome in foals. Compend Contin Educ Vet 2009;4:134e40. [17] Bernard WV, Reimer JM, Cudd T. Historical factors, clinicopathologic findings, clinical features and outcome of equine neonates presenting with or developing signs of central nervous system disease. Proc. Am. Assoc. Equine Practnrs 1995;41:221e4. [18] Galvin N, Collins D. Perinatal asphyxia syndrome in the foal: review and case report. Ir Vet J 2004;57(12):707e14. [19] Pirrone A, Panzani S, Govoni N, Castagnetti C. Thyroid hormone concentrations in foals affected by perinatal asphyxia. Theriogenology 2013;80:624e9. [20] Pirrone A, Antonelli C, Mariella J, Castagnetti C. Gross placental morphology and foal serum biochemistry as predictors of foal health. Theriogenology 2014;81:1293e9. re S, Morresey PR, Yang C, Shaw G. Biomarkers of brain [21] Ringger NC, Gigue injury in foals with hypoxic-ischemic encephalopathy. J Vet Intern Med 2011;25:132e7. [22] Stewart JH, Rose RJ, Barko AM. Respiratory studies in foals from birth to seven days old. Equine Vet J 1984;16(4):323e8. [23] Hodgson DR. Blood gas and acid-base changes in the neonatal foal. Vet Clin N Am Equine Pract 1987;3(3):617e29. [24] Cruz RKS, Alfonso A, Lourenço MLC, Ulian CMV, Sudano MJ, Oba E, Padovani CR, Ramos PRR, Chiachchio SB. Electrolyte, blood gas and electrocardiographic profile of neonatal foals in the first 48 hours of life. Acta Sci Vet 2015;43:1321. [25] Rose RJ, Rossdale PD, Leadon DP. Blood gas and acid-base status in spontaneously delivered, term-induced and induced premature foals. J Reprod Fertil Suppl 1982;32:521e8. [26] Vaala WE. Peripartum asphyxia. Vet Clin N Am Equine Pract 1994;10: 187e218. [27] CLSI. Defining, establishing and verifying reference intervals in the clinical laboratory. Approved guideline. third ed. Wayne, PA: Clinical and Laboratory Standards Institute; 2008. CLSI document EP28-A3c. [28] Mukaka MM. Statistics corner: a guide to appropriate use of correlation coefficient in medical research. Malawi Med J 2012;24(3):69e71. A, Friedrichs K, Harr K, Concordet D, Trumel C, Braun J. Reference [29] Geffre values: a review. Vet Clin Pathol 2009;38(3):228e98. [30] Belai Y, Goodwin TM, Durand M, Greenspoon JS, Paul RH, Walther FJ. Umbilical arteriovenous PO2 and PCO2 differences and neonatal morbidity in term infants with severe acidosis. Am J Obstet Gynecol 1998;17(Pt 1):13e9. [31] Gjerris AC, Staer-Jenson J, Jorgensen JS, Bergholt T, Nickelsen C. Umbilical cord blood lactate: a valuable tool in the assessment of fetal metabolic acidosis. Eur J Obstet Gynecol Reprod Biol 2008;139:16e20. [32] Victory R, Penava D, daSilva O, Natale R, Richardson B. Umbilical cord pH and base excess values in relation to adverse outcome events for infants delivering at term. Am J Obstet Gynecol 2004;191:2012e28. [33] Westgren M, Divon M, Horal M, Ingermarsson I, Kublickas M, Shimojo N, Nordstrom L. Routine measurements of umbilical artery lactate levels in the prediction of perinatal outcome. Am J Obstet Gynecol 1995;173(5):1416e22. [34] Linet T, Laporte J, Gueye H, Boog G. Microvolume dosage of lactate in cord blood for the evaluation of the neonatal well-being. J Gynecol Obstet Biol Reprod 2002;31(4):352e7. [35] Nordstrom L, Ingemarsson I, Persson B, Shimojo N, Westgren M. Lactate in fetal scalp blood and umbilical artery blood measured during normal labor with a test strip method. Acta Obstet Gynecol Scand 1994;73(3):250e4. [36] Shirey T, St Pierre J, Winkelman J. Cord lactate, pH, and blood gases from healthy neonates. Gynecol Obstet Invest 1996;41(1):15e9. [37] Tuuli M, Stout MJ, Shanks A, Odibo AO, Macones GA, Cahill AG. Umbilical cord arterial lactate compared with pH for predicting neonatal morbidity at term. Obstet Gynecol 2014;124(4):756e61. [38] Tuuli M, Stout MJ, Macones GA, Cahill AG. Umbilical cord venous lactate for predicting arterial lactic acidemia and neonatal morbidity at term. Obstet Gynecol 2016;127(4):674e80.
Contents lists available at ScienceDirect
Theriogenology journal homepage: www.theriojournal.com
Determination of reference intervals for umbilical cord arterial and venous blood gas analysis of healthy Thoroughbred foals Sunita S. Jeawon a, *, Lisa M. Katz a, Noreen P. Galvin b, Ursula M. Fogarty c, Vivienne E. Duggan a a b c
UCD School of Veterinary Medicine, Veterinary Sciences Centre, University College Dublin, Belfield, Dublin 4, Ireland Phoenix Equine Group, Kildare, Co. Kildare, Ireland Irish Equine Centre, Johnstown, Co. Kildare, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 March 2018 Received in revised form 18 May 2018 Accepted 21 May 2018 Available online 25 May 2018
Although umbilical cord blood gas analysis is considered the best way to assess in utero oxygenation in human neonates, there is limited evaluation of this method in equine neonatology. Our objectives were to assess the practicality of obtaining umbilical cord blood gas samples in the field and to determine umbilical cord arterial and venous blood gas reference intervals (RI) for healthy, newborn foals. Thoroughbred foals >320 days gestation from healthy mares with uneventful pregnancies at one stud farm were evaluated. All parturitions were observed, with paired umbilical arterial and venous whole-blood samples obtained immediately following parturition for blood gas and lactate concentrations measured in duplicate. Apgar scores were assigned immediately and 10 min after birth, with all foals subsequently examined on days 1e28 to monitor for development of perinatal asphyxia syndrome. Foals were excluded from analysis based on abnormalities of stage 2 labour, Apgar scores and gross and histological placental assessment. Data was analysed using a Student's t-test, Pearson's correlation and the Robust method with P 0.05 significant. Umbilical cord samples were simple to obtain with minimal disruption to the foaling environment. Of the n ¼ 34 foals assessed, n ¼ 7 were excluded based on premature placental separation deliveries. The mean time for stage 2 labour and blood gas analysis after parturition was 17.3 ± 5.1 min and 5.0 ± 2.3 min, respectively. RI were identified for umbilical arterial and venous pH (7.19e7.42 vs. 7.34e7.44), PO2 (15.5e48.39 mmHg vs. 16.6e52.7 mmHg), PCO2 (49.5e82.29 mmHg vs. 45.4e63.1 mmHg), SO2 (9.19e76.89% vs. 39.9e84.88%), bicarbonate (27.3e38.7 mmol/l vs. 27.7 e37.8 mmol/l), base excess (0.36e12.9 mmol/l vs. 1.97e13.1 mmol/l), TCO2 (28.99e40.3 mmHg vs. 29.0 e39.5 mmHg) and lactate (1.4e7.3 mmol/l vs. 1.3e4.9 mmol/l). Umbilical arterial samples had lower pH (P < 0.0001), PO2 (P ¼ 0.002) and SO2 (P < 0.0001) and higher PCO2 (P < 0.0001) and lactate (P < 0.0001) than venous samples. The initial Apgar score was positively correlated to umbilical arterial SO2 (r ¼ 0.4, P ¼ 0.05) and negatively with umbilical arterial TCO2 (r ¼ 0.6, P ¼ 0.004). Overall, umbilical cord sampling was simple and minimally disruptive, with RI obtained for blood gas measurements. RI for umbilical blood gas measurements from a larger population of healthy and unhealthy foals is required to evaluate the accuracy of this method for assessing in utero oxygenation. © 2018 Published by Elsevier Inc.
Keywords: Blood gas analysis Umbilical cord Equine neonate Hypoxia Perinatal asphyxia syndrome
1. Introduction Arterial and venous blood gas analysis can be used as a diagnostic tool in veterinary medicine to help evaluate an animal's
* Corresponding author. E-mail addresses: [email protected] (S.S. Jeawon), [email protected] (L.M. Katz), [email protected] (N.P. Galvin), ufogarty@ irishequinecentre.ie (U.M. Fogarty), [email protected] (V.E. Duggan). https://doi.org/10.1016/j.theriogenology.2018.05.024 0093-691X/© 2018 Published by Elsevier Inc.
oxygenation, ventilation and acid-base status [1]. In human neonatology, umbilical cord blood gas analysis immediately at the time of birth is thought to provide the most sensitive representation of neonatal acid-base status prior to birth [2]. Hypoxic syndromes have been described in human neonates [3,4] with the combination of Apgar scoring [5,6] and umbilical cord blood gas analysis used by neonatologists to assess the likelihood that a hypoxic event occurred in utero, either during parturition or more chronically during the pregnancy [2,7,8]. These assessments help
2
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
identify at-risk neonates, allowing for early medical intervention [9e11], reduction in morbidity and mortality rates as well as adverse clinical sequelae, in particular related to neurodevelopmental disease [12e14]. Similar to the hypoxic syndromes seen in human neonates, perinatal asphyxia syndrome (PAS) in neonatal foals is likely caused by hypoxic-ischemic damage that occurred during pregnancy or parturition [15]. Whilst originally thought to primarily affect just the central nervous system, it is now known to affect multiple other body systems [16]. The overall occurrence of PAS is cited to be between 1 and 2% of all foals born [17]. Although numerous risk factors have been identified for PAS, currently there are no confirmed pathognomonic biological or biochemical parameters that can be used to indicate the presence of this disease [18]. Researchers have described associations between decreased thyroid hormones [19], increased systemic creatinine and decreased systemic glucose concentrations [17,20] with risk of neonatal illness such as PAS. Ringger et al. (2011) also reported on increased concentrations of the biomarker UCHL1 in foals affected by PAS [21]. However, affected foals are more commonly not identified until after clinical signs develop, at which point significant systemic damage by inflammatory mediators may have already occurred [15]. Although the use of arterial and venous blood gas analysis in equine neonatology has been described in the literature [22e24], to-date, the use of umbilical cord blood gas analysis in equine neonates has only been reported by one group in which umbilical arterial blood gas parameters were evaluated in prematureinduced and term-induced foals [25]. Therefore, the purpose of this study was a) to investigate the practicality of obtaining umbilical cord blood samples from newborn foals on a stud farm and b) determine umbilical cord arterial and venous blood gas reference intervals (RI) in a group of healthy, newborn Thoroughbred (Tb) foals. 2. Materials and methods University College Dublin's Animal Research Ethics Committee approved this study. Owner consent was also granted. 2.1. Sample population During the 2017 breeding season, Thoroughbred broodmares with no history of systemic illness or placentitis during pregnancy and a gestational length >320 days were included in the study; exact gestational length was recorded for each mare once they foaled. All mares were from the same stud farm and managed under similar circumstances. The mares were closely monitored by on-site staff 24-hours a day, with at least 2 trained staff members attending each foaling. After foaling, data from foals of mares in the study group were retrospectively excluded from analysis if there was any abnormality of the second stage of parturition (e.g., premature placental separation [PPS]), if the Apgar score of the foal was <6 at each assessment, if the clinical appearance of the foal after parturition and/or the gross and histopathological appearance of the placenta were abnormal. Data from excluded foals are only reported for comparison. 2.2. Experimental protocol All mares were observed from outside the stable as they went through stage 1 labour. The time of chorioallantoic membrane rupture was noted, with the mares then allowed to progress through the second stage of labour with minimal interference, although all mares had one staff member guiding the foal's head
and one applying gentle traction on its legs until the foal was safely delivered. The duration of stage 2 of parturition was recorded for every animal, calculated from the time of chorioallantoic membrane rupture to the birth of the foal. One author (SSJ) attended all parturitions and obtained all immediate post-partum clinical and umbilical cord blood samples. As soon as the foal was expelled, the umbilical arteries and vein were identified via visual inspection and manual palpation. For each blood gas sample measurement, 1 ml of blood was collected anaerobically into a heparinised disposable blood gas syringe (RAPIDLyte, Cruinn Medical Ltd, Dublin, Ireland). The umbilical cord was not clamped for sampling purposes, with the arterial sample always obtained first from the largest umbilical artery (as this was the easiest one from which to obtain a blood sample) using a 21 g needle, followed by venous blood collection using a 23 g needle. The timing of each sample collection in relation to foetal expulsion was recorded. Rectal temperature was taken from each foal to temperature-correct blood gas analysis. Immediately following umbilical cord blood collection, a 4parameter Apgar score was assigned to the foal using a modified system previously described by Vaala (1994) [26] (Table 1), with a second Apgar score assigned 10 min after the initial one; scores of 6 were considered to be within normal limits. Immediately after completion of the initial Apgar score, umbilical blood samples were evaluated in duplicate using a portable blood gas analyser (Vetscan i-Stat® 1, Abaxis UK Ltd, York, United Kingdom). Parameters measured included pH, PCO2, PO2, HCO3, TCO2, SO2, base excess/ deficit and lactate. The arterial sample was always evaluated first followed by the venous sample. The time from blood sampling to analysis was recorded. For all mares, the time to passage of the foetal membranes was recorded, with the membranes immediately collected for analysis. The membranes were initially laid out on a designated table to visually inspect that the entire placenta had been passed, with the presence of any tears noted before being weighed on-site with commercial weighing scales. The foetal membranes were then submitted for full gross and histopathological evaluation. If parturition occurred during the weekend, the foetal membranes were stored in a cold room until the subsequent Monday. All gross and histopathological evaluations were performed by one author (UMF) blinded to any information about the parturition other than the date of foaling and weight of the foetal membranes. Approximately 2 cm2 of pregnant horn/non-pregnant horn body pouch, cervical area, allantoamnion and amniotic cord were taken for histopathological examination. Any pathological findings were recorded as absent, mild, moderate or severe (Supplementary Tables 1e3). The foetal membranes were then classified into group 1 (absent or mild pathological findings) or group 2 (at least one section characterised by moderate-to-severe pathological findings). Data from foals with foetal membranes placed into group 2 were excluded from analysis. Complete physical examinations were carried out on all foals at days 1, 2, 3, 7, 14, 21 and 28 by 1 of 2 authors (SSJ or NPG) to monitor for development of PAS. A blood sample was taken from all foals between 10 and 14 h after birth for measurement of serum IgG concentrations. 2.3. Data analysis Calculations and analyses were performed using MedCalc statistical software (www.medcalc.org, MedCalc Software, v. 13.1.00, Mariakerke, Belgium) and SPSS (IBM SPSS Statistics for Windows, v. 24.0, Armonk, NY: IBM Corp). Measurements were evaluated for normality using a D'Agostino-Pearson test for normal distribution. Boxplots and histograms were assessed and any extreme outliers identified. The measurements were further assessed using Reed
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
3
Table 1 Apgar Score Parameters adapted from Vaala (1994) [26]. Score
2
1
0
Heart rate Respiratory rate Muscle tone, postural responses Response to stimuli
More than 60 BPM and regular Regular Maintains sternal recumbency and attempts to stand Head shake, sneeze or cough
Less than 60 BPM or irregular Irregular Some flexion of limbs and muscle tone
Absent Absent Laterally recumbent, no muscle tone No response to stimulation
Weak ear movement when stimulated, facial grimace
et al. (1971) and Tukey (1977) outlier tests with extreme outliers removed. For the umbilical blood samples, coefficients of variation (CV) from duplicate measurements for each analyte were determined. Reference intervals were then calculated by a Robust Method (Clinical and Laboratory Standards Institute guidelines C28-A3) [27] with 90% confidence intervals (CI) calculated. Reference interval values are presented as median ± CI. All other values are reported as the mean ± sem. Pearson's correlation coefficient was used to evaluate significance and strength of associations between the blood gas analysis variables and foaling and physical parameters, with r 0.4 signifying correlation. Pairwise scatter plots were constructed between all variables to confirm linear distribution. Parameters were considered to be moderately associated with each other if r 0.5 and strongly associated if r 0.7 [28]. Univariate analysis was carried out using a Student's t-test to compare between the umbilical cord arterial and venous blood gas analysis parameters. Significance was set at P 0.05. 3. Results A total of n ¼ 34 mares and associated foals (n ¼ 24 females, n ¼ 10 males) were evaluated. The mean age of the mares was 9 ± 3.5 (range 4e21) years, with an average of 5 parities (range 1e14). All foals were full-term (>320 days old) with a mean gestational age of 344 ± 8 (range 329e363) days. The mean birth weight was 56 ± 7.6 (range 37e67) kg and 58 ± 8 (range 44e69) kg for the filly and colt foals, respectively. The mean length of stage 2 labour was 17 ± 5 (range 9e26) minutes. Of the n ¼ 34 foals assessed, n ¼ 7 (n ¼ 4 females and n ¼ 3 males) were excluded from analysis based on PPS deliveries. Initial Apgar scores were carried out on average 2.9 (range 1e6) minutes after the foal was born. Initial Apgar scores were 4 (n ¼ 1), 6 (n ¼ 3), 7 (n ¼ 3) or 8 (n ¼ 27). All foals with PPS had initial Apgar scores of 8 except for 1 foal, which had an initial Apgar score of 4; this latter foal was retrospectively identified to come from a mare
with a history of having compromised foals at birth, and therefore deemed to be high-risk. Second Apgar scores were completed on average 10.9 ± 1.2 (range 8e13) minutes after the initial one. All foals had a 10-minute Apgar score of 8 except for the foal from the mare deemed to be high-risk, which had a score of 6. Time to expulsion of the foetal membranes after parturition was on average 31 ± 14.6 (range 4e69) minutes and the mean placental weight was 7 ± 1.5 (range 4e11) kg. The average total cord length was 52 ± 8.9 (range 40e70) cm and 50.1 ± 9.9 (range 35e66) cm for the normal foals and PPS foals, respectively. Of the normal foals, 15/27 (56%) were from right horn pregnancies and 12/27 (44%) were from left horn pregnancies. Of the PPS foals, 6/7 (86%) were from right horn pregnancies and 1/7 (14%) from left horn pregnancies. Of the n ¼ 34 foals in the study, only 1 foal (from the high-risk mare) showed clinical signs of PAS in the first 72 h of life. All other foals were clinically normal for the first 28 days of life. The average time from birth to the first umbilical cord sample acquisition was 1.2 ± 0.8 (range 1e5) minutes. The average time from sampling to analysis was 5.0 ± 2.3 (range 2e12) minutes. Each assay took 120 s to complete after the patient details had been entered. Paired arterial and venous samples were obtained from n ¼ 29 foals, arterial samples alone from n ¼ 3 and venous samples alone from n ¼ 2. Based on the constructed boxplots, 2 measurements were excluded from analysis as extreme outliers. Summary statistics are shown in Tables 2 and 3 for the normal foals and in Tables 4 and 5 for the foals excluded from analysis due to PPS. Tables 6 and 7 summarise RI identified for arterial and venous samples. In comparison to venous samples, arterial samples had a significantly lower pH (7.39 ± 0.03 vs. 7.32 ± 0.05, P < 0.0001), PO2 (34.4 ± 8.19 mmHg vs. 29.2 ± 11.5 mmHg, P ¼ 0.002) and SO2 (62.0 ± 15.2 %vs. 43.9 ± 18.1%, P < 0.0001) and significantly higher PCO2 (54.1 ± 4.66 mmHg vs. 64.5 ± 6.84 mmHg, P < 0.0001) and lactate concentration (3.19 ± 0.83 mmol/l vs. 4.42 ± 1.36 mmol/l, P < 0.0001). There were no significant differences between venous and arterial bicarbonate (32.8 ± 2.38 mmol/l vs. 33.2 ± 2.75 mmol/ l), base excess (7.56 ± 2.63 mmol/l vs. 7.1 ± 3.18 mmol/l) and TCO2
Table 2 Summary statistics for paired umbilical cord arterial and venous samples from n ¼ 27 healthy Thoroughbred foals. pH Arterial
Venous
PCO2 (mmHg)
PO2 (mmHg)
Arterial
Base Excess (mmol/l)
HCO 3 (mmol/l)
SO2 (%)
Venous
Arterial
Venous
Arterial
Venous
Arterial
Venous
Arterial
Minimum 25% quartile Median ± SD
7.22 7.33 49.95 7.28 7.37 60.0 7.30 ± 0.05 7.39 ± 0.03 66.0 ± 7.7
45.8 51.8 53.7 ± 4.1
14.0 29.4 36.0 ± 8.6
1.5 4.8 7.0 ± 2.97
3.0 5.4 7.5 ± 2.6
27.9 31.4 33.4 2.7±
28.0 31.1 32.9 ± 2.4
18.0 45.0 31.0 54.3 44.0 ± 16.0 66.3 ± 12.1
75% quartile Maximum Mean ± SE
7.35 7.40 70.7 7.43 7.43 76.95 7.32 ± 0.01 7.39 ± 0.01 65.6 ± 1.6
58.1 63.1 54.3 ± 0.8
38.5 55.0 36.0 ± 1.7
8.5 14.5 7.0 ± 0.6
9.3 13.0 7.5 ± 0.5
34.9 39.9 33.3 ± 0.5
34.99 37.4 32.7 ± 0.5
55.8 82.0 42.4 ± 3.2
70.3 86.5 63.7 ± 2.5
95% CI of duplicates CV of duplicates (%)
0.0687 e0.1253 0.097
1.4695 e2.6972 2.1
16.0 23.0 28.0 (23.1 e30.9) 33.5 55.5 28.0 (24.7 e30.9) 3.6100 e6.6839 5.1
4.0132 e7.4423 5.7
9.8020 e18.6011 14.1
8.7081 e16.4543 12.5
1.4313 e2.6266 2.0
1.4470 e2.6556 2.1
5.2906 e9.8618 7.6
2.8931 e5.3409 4.1
0.0492 e0.0897 0.07
1.5833 e2.9073 2.2
SD: standard deviation; SE: standard error; CI: confidence interval; CV: coefficient of variation.
Venous
4
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
Table 3 Summary statistics for paired umbilical cord arterial and venous samples from n ¼ 27 healthy Thoroughbred foals.
Minimum 25% quartile Median ± SD 75% quartile Maximum Mean ± SE 95% CI of duplicates CV of duplicates (%)
Arterial TCO2 (mmol/l)
Venous TCO2 (mmol/l)
Arterial lactate (mmol/l)
Venous lactate (mmol/l)
29.5 33.3 34.5 ± 2.7 36.3 41.5 35.4 ± 0.5 5.2135e9.7151 7.8
29.0 32.8 34.7 ± 2.4 36.5 39.0 34.2 ± 0.5 1.5727e2.8878 2.2
1.95 3.4 4.2 ± 1.4 5.5 7.2 4.36 ± 0.3 1.4585e2.6768 2.1
1.98 2.6 2.99 ± 0.8 3.7 4.95 3.2 ± 0.8 1.0649e1.9514 1.5
SD: standard deviation; SE: standard error; CI: confidence interval; CV: coefficient of variation.
Table 4 Summary statistics for paired umbilical cord arterial and venous samples from n ¼ 7 Thoroughbred foals from mares with premature placental separation. pH
Minimum 25% quartile Median ± SD 75% quartile Maximum Mean ± SE 95% CI of duplicates CV of duplicates (%)
PCO2 (mmHg)
PO2 (mmHg)
Base Excess (mmol/l)
HCO 3 (mmol/l)
SO2 (%)
Arterial
Venous
Arterial
Venous
Arterial
Venous
Arterial
Venous
Arterial
Venous
Arterial
Venous
7.26 7.298 7.31 ± 0.02 7.32 7.32 7.30 ± 0.008 0.0177 e0.0845 0.05
7.35 7.37 7.38 ± 0.02 7.39 7.399 7.38 ± 0.009 0.0874 e1.3544 0.7
61.8 62.6 64.6 ± 2.5 67.3 67.8 64.9 ± 0.9 0.6378 e3.0853 1.9
49.8 51.3 52.6 ± 1.3 52.8 52.9 51.97 ± 0.6 0.1685 e2.6249 1.4
18.5 24.9 31.5 ± 15.3 33.8 66.0 33.8 ± 5.8 1.5688 e7.7240 4.6
31.5 32.6 39.0 ± 5.4 40.5 45.0 37.5 ± 2.4 0.6003 e9.6493 5.0
2.0 5.1 6.0 ± 1.9 6.5 8.0 5.6 ± 0.7 2.6631 e13.386 7.9
6.0 6.0 6.5 ± 1.0 7.4 8.5 6.8 ± 0.5 0.7513 e12.211 6.3
29.3 30.99 32.2 ± 1.6 32.8 34.3 31.9 ± 0.6 0.5147 e2.4841 1.5
30.8 31.1 31.3 ± 0.7 32.1 32.6 31.6 ± 0.3 0.1634 e2.5455 1.4
20.5 37.0 49.5 ± 20.7 55.0 87.0 49.6 ± 7.8 2.6822 e13.487 7.95
56.0 58.3 67.5 ± 9.3 73.4 79.0 66.6 ± 4.2 0.5213 e8.3315 4.4
SD: standard deviation; SE: standard error; CI: confidence interval; CV: coefficient of variation.
Table 5 Summary statistics for paired umbilical cord arterial and venous samples from n ¼ 7 Thoroughbred foals from mares with premature placental separation.
Minimum 25% quartile Median ± SD 75% quartile Maximum Mean ± SE 95% CI of duplicates CV of duplicates (%)
Arterial TCO2 (mmol/l)
Venous TCO2 (mmol/l)
Arterial lactate (mmol/l)
Venous lactate (mmol/l)
31.0 33.3 34.5 ± 1.8 35.0 36.5 34.1 ± 0.7 0.6467e3.1290 1.9
32.0 32.4 33.0 ± 0.9 33.4 34.5 33.0 ± 0.4 0.1632e2.5424 1.4
2.4 2.7 3.1 ± 1.4 4.2 6.2 3.5 ± 0.5 0.3389e1.6302 0.98
2.2 2.4 2.7 ± 0.5 3.1 3.5 2.8 ± 0.2 0.1873e2.9225 1.6
SD: standard deviation; SE: standard error; CI: confidence interval; CV: coefficient of variation.
Table 6 Reference interval for n ¼ 25 paired umbilical cord arterial and venous samples from healthy Thoroughbred foals calculated using the Robust method. The confidence intervals for the reference limits were estimated using bootstrapping (percentile interval method Efron & Tibshirani 1993). pH Arterial Lower limit 7.19 90% CI 7.17 e7.22 Upper limit 7.42 90% CI 7.38 e7.46
PCO2 (mmHg)
PO2 (mmHg)
Venous
Arterial
Venous
Arterial
7.34 7.32 e7.35 7.44 7.42 e7.45
49.5 44.76 e54.93 82.29 78.36 e85.99
45.4 42.94 e47.45 63.1 59.63 e65.75
15.5 13.08 e18.71 48.39 40.88 e57.01
Base Excess (mmol/l)
HCO 3 (mmol/l)
SO2 (%)
Venous
Arterial
Arterial
Venous
Arterial
Venous
16.6 11.83 e23.48 52.7 46.89 e58.54
0.36 1.97 1.23e2.30 0.68e3.47
27.3 25.63 e28.97 38.7 36.83 e40.51
27.7 26.48 e29.16 37.8 36.38 e39.07
9.19 0.125 e18.27 76.89 65.73 e86.78
39.9 31.35 e47.49 92.1 84.88 e96.66
12.9 11.04 e14.96
Venous
13.1 11.50 e14.54
CI: confidence interval.
(34.2 ± 2.44 mmHg vs. 35.4 ± 3.54 mmHg). Initial Apgar score was positively correlated to arterial SO2 (r ¼ 0.4, P ¼ 0.05) and negatively correlated to TCO2 (r ¼ 0.6, P ¼ 0.004). 4. Discussion To the authors' knowledge, this is the first paper that describes obtaining paired umbilical cord venous and arterial blood gas
measurements from naturally born foals in a field setting. The technique of taking the umbilical cord samples was relatively simple, with minimal disruption to the foaling environment. Failure to obtain a paired arterial and/or venous sample in 4 of the foals was either due to the mare standing prematurely and rupturing the umbilical cord before the venous sample could be taken or due to the mare lying awkwardly against the wall as she foaled, preventing adequate access to the umbilical artery. The ease of identifying the
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
5
Table 7 Reference interval for n ¼ 25 paired umbilical cord arterial and venous samples from healthy Thoroughbred foals calculated using the Robust method. The confidence intervals for the reference limits were estimated using bootstrapping (percentile interval method Efron & Tibshirani 1993).
Lower limit 90% CI Upper limit 90% CI
Arterial TCO2 (mmol/l)
Venous TCO2 (mmol/l)
Arterial lactate (mmol/l)
Venous lactate (mmol/l)
28.99 27.43e30.83 40.3 38.37e42.08
29.0 NC 39.5 NC
1.4 0.77e2.09 7.3 6.32e8.05
1.3 0.797e1.76 4.9 4.27e5.43
CI: confidence interval. NC: not calculated, too many equal values in small sample size.
different vessels was aided by the fact that the vein was consistently the largest vessel and the largest artery usually had a palpable pulse. The average time of 5 min between the samples being taken and analysed is unlikely to have skewed the results, as research into the effect of timing on umbilical cord blood sampling in human neonates has shown no significant difference in the results obtained once the sampling has occurred within a 30 min window from birth [2]. It is of great importance to first identify normal reference intervals for umbilical blood gas measurements from healthy foals so that the use of this method for assessing foals for risk of PAS can be investigated. Umbilical arterial blood gas parameters have been previously reported for 8 premature-induced and full-terminduced foals [22], with similar umbilical arterial pH values (7.3 ± 0.01) to those from the present study. However, the umbilical arterial PO2 was higher (43.2 ± 3.9 mmHg) and the PCO2 (53.0 ± 1.8 mmHg) and base excess (0.2 ± 1.2 mmol/l) lower than what was identified in the present study [22]. A primary reason for differences in reported umbilical arterial blood gas values between the studies would be due to differences in experimental design. The study reported by Rose et al. (1982) involved a much smaller sample population of foals in which parturition was induced, with differences in the timing of sampling as well as differences in blood gas analysers used. Although CSLI guidelines [27] recommend using nonparametric methods to determine population-based RIs from at et al. least 120 healthy reference individuals as described by Geffre (2009) [29], limitation to such large numbers in clinical veterinary practice and in particular neonates often result in this being impossible. Because the reference population of interest was welldefined in the present study with inclusion/exclusion criteria used to confirm health, reference intervals could be estimated using a much smaller sample size through the Robust method using bootstrapping, as highlighted in CSLI guideline (C28-A3) [27]. Furthermore, the samples were run in duplicate and were taken in an environment mimicking that in which the majority of practitioners would employ the technique. As all samples were taken by the same person on a single stud farm; external factors such as foaling protocols, management of the mares and sample timing were also all tightly controlled for consistency. The umbilical artery is considered the most important vessel to sample as it represents the foetal acid-base status independent of the maternal circulation, which the venous sample reflects as well [30]. This explains the significantly lower umbilical arterial pH and significantly higher PCO2 and lactate concentration as compared to the umbilical vein measurements in the present study. These differences are similar to those reported for measurements between the two vessels in human neonatology [8]. The benefits of having both arterial and venous samples for blood gas analysis are emphasised in the human literature [2,3,8,30], where it is reported that comparing the venous and arterial blood gas parameters can help identify whether an acute or chronic hypoxic insult has occurred. With acute cord compression or foetal bradycardia, abnormalities will predominantly be seen in the arterial sample, with
a large arterio-venous difference existing between the parameters of both vessels. However, in more chronic hypoxic insults relating to reduced placental transfer of oxygen, both vessels will yield abnormal blood gas measurements with minimal arterio-venous differences [2,3]. Such information could be similarly useful in the assessment of neonatal foals, with equine neonates prone to both acute asphyxial injury at the time of parturition as well as more chronic hypoxia linked with placental insufficiency. Identifying whether there is an acute or more chronic problem may aid in generating a more realistic prognostic outcome. Classification of an acidosis as either primarily respiratory, metabolic or mixed in origin is also important, as a metabolic acidosis is thought to be more harmful than a respiratory one, as it signifies a later stage of foetal hypoxia when anaerobic pathways are taking over to supply energy to the foetus [2,31,32]. An increase in lactate is associated with this process and mirrored by a concurrent drop in the base excess measurement [31]. Umbilical cord lactate concentration has also been shown to correlate significantly with pH, base deficit and PCO2, and to be predictive of low Apgar Scores and certain other short-term morbidities [33]. Although human neonatologists are now largely in agreement in defining damaging foetal acidemia as an umbilical arterial pH of <7.0 and a base deficit of 12 mmol/l, the reported lactate measurements vary in the literature from 3.2 to 7.0 mmol/l [33e36]. Recent work from Tuuli et al. (2014, 2016) proposes an optimal cut-off for both umbilical cord arterial (3.9 mmol/l) and venous (3.4 mmol/l) lactate concentrations, beyond which values are predictive of neonatal morbidity at term [37,38]. Measurement of umbilical cord blood lactate concentrations in equine neonates could similarly give further insight into the presence or absence of a metabolic acidosis and complement the information gained from umbilical cord blood gas analysis findings further. In this study, the Apgar scoring system modified for use in equine medicine [26] appeared to be a useful scoring system for assessing equine neonates. The one truly compromised foal in the present study had a notably different initial Apgar score in comparison to the rest of the study foals. The modified Apgar scoring system was simple to use with clear definitions regarding how the parameters included were to be quantified. Apgar scoring remains a widely used and relevant scale in modern human medicine to help assess the newborn's condition and aid prediction of neonatal survival [6], and should likely be used to a greater degree in equine neonatology. The positive correlation between the initial Apgar score and umbilical arterial SO2 in this study was interesting, as foals that had high in utero SO2 values were likely breathing efficiently immediately following birth and therefore less likely to be marked as a 1 on the respiratory section of the Apgar scale. Conversely, the negative correlation between Apgar score and umbilical arterial TCO2, a measure of CO2 in multiple states (physical solution, proteinbound, HCO3, CO3 and H2CO3), is most likely reflective of the fact that foals with a higher in utero TCO2 were less likely to be breathing as efficiently at birth, and therefore were more likely to be marked as a 1 on the respiratory section of the Apgar scale. These
6
S.S. Jeawon et al. / Theriogenology 118 (2018) 1e6
findings could indicate that efficiency of ventilation in the immediate period after birth is influenced by the in utero SO2 and TCO2 levels; however, further research into this area is required. 5. Conclusion In summary, this is the first paper that describes paired umbilical cord venous and arterial blood gas measurements from naturally born foals in a field setting. Overall, umbilical cord sampling was simple and minimally disruptive with RI obtained for blood gas measurements. The use of the modified Apgar scoring in tandem with umbilical cord blood gas analysis could be useful as an indicator of in utero oxygenation of the newborn foal. However, reference intervals for umbilical blood gas measurements from a larger population of healthy and unhealthy foals is required to fully evaluate the accuracy of this method for assessing in utero oxygenation and acid-base status. Conflicts of interest The authors declare no conflicts of interest. Acknowledgements The authors wish to thank the management and staff of the stud farm for collaborating on this project, in particular Gretta Clarke, Meaghan Hegarty and Gillian Murray who greatly facilitated ease of sample collection. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.theriogenology.2018.05.024. References [1] Clutton RE. Blood gas analysis. In: McGorum BC, Dixon PM, Robinson NE, Schumacher J, editors. Equine respiratory medicine and surgery. London: Elsevier Health Sciences; 2006. p. 201e9. [2] Blickstein I, Green T. Umbilical cord blood gases. Clin Perinatol 2007;34: 451e9. [3] Bobrow CS, Soothill PW. Causes and consequences of fetal acidosis. Arch Dis Child Fetal Neonatal Ed 1999;80:F246e9. [4] Jonsson M, Agren J, Norden-Lindeberg S, Ohlin A, Hanson U. Neonatal encephalopathy and the association to asphyxia in labor. Am J Obstet Gynecol 2014;211(667):E1e8. [5] Apgar V. A proposal for a new method of evaluation of the newborn infant. Curr Res Anesth Analg 1953;32(4):260e7. [6] Casey BM, McIntire DD, Leveno KJ. The continuing value of the Apgar score for the assessment of newborn infants. N Engl J Med 2001;344:467e71. [7] Dudenhausen JW, Luhr C, Dimer JS. Umbilical cord blood gases in healthy term newborn infants. Int J Gynecol Obstet 1997;57:251e8. [8] Armstrong L, Stenson BJ. Use of umbilical cord blood gas analysis in the assessment of the newborn. Arch Dis Child Fetal Neonatal Ed 2007;92(6): F430e4. [9] Azzopardi D, Brocklehurst P, Halliday H, Levene M, Thoresen M, Whitelaw A. The TOBY study. Whole body hypothermia for the treatment of perinatal asphyxial encephalopathy: a randomised controlled trial. BMC Pediatr 2008;8(1):17. [10] Hobson A, Baines J, Weiss MD. Beyond hypothermia: alternative therapies for hypoxic-ischemic encephalopathy. Open Pharmacol J 2013;7:26e40. [11] Dingle J, Tooley J, Liu X, Schull-Brown E, Elstad M, Chakkarapani E, Sabir H, Thoresen M. Xenon ventilation during therapeutic hypothermia in neonatal
encephalopathy: a feasibility study. Pediatrics 2014;133(5):809e18. [12] van Schie PE, Schijns J, Becher JG, Barkhof F, van Weissenbruch MM, Vermeulen RJ. Long-term motor and behavioral outcome after perinatal hypoxic-ischemic encephalopathy. Eur J Paediatr Neurol 2015;19(3):354e9. [13] Pappas A, Korzeniewski SJ. Long-term cognitive outcome of birth asphyxia and the contribution of identified perinatal asphyxia to cerebral palsy. Clin Perinatol 2016;43(3):559e72. [14] Adhikari S, Rao KS. Neurodevelopmental outcome of term infants with perinatal asphyxia with hypoxic ischemic encephalopathy stage II. Brain Dev 2017;39(2):107e11. [15] Gold JR. Perinatal asphyxia syndrome. Equine Vet Educ 2015;29(3):158e64. [16] Vaala WE. Perinatal asphyxia syndrome in foals. Compend Contin Educ Vet 2009;4:134e40. [17] Bernard WV, Reimer JM, Cudd T. Historical factors, clinicopathologic findings, clinical features and outcome of equine neonates presenting with or developing signs of central nervous system disease. Proc. Am. Assoc. Equine Practnrs 1995;41:221e4. [18] Galvin N, Collins D. Perinatal asphyxia syndrome in the foal: review and case report. Ir Vet J 2004;57(12):707e14. [19] Pirrone A, Panzani S, Govoni N, Castagnetti C. Thyroid hormone concentrations in foals affected by perinatal asphyxia. Theriogenology 2013;80:624e9. [20] Pirrone A, Antonelli C, Mariella J, Castagnetti C. Gross placental morphology and foal serum biochemistry as predictors of foal health. Theriogenology 2014;81:1293e9. re S, Morresey PR, Yang C, Shaw G. Biomarkers of brain [21] Ringger NC, Gigue injury in foals with hypoxic-ischemic encephalopathy. J Vet Intern Med 2011;25:132e7. [22] Stewart JH, Rose RJ, Barko AM. Respiratory studies in foals from birth to seven days old. Equine Vet J 1984;16(4):323e8. [23] Hodgson DR. Blood gas and acid-base changes in the neonatal foal. Vet Clin N Am Equine Pract 1987;3(3):617e29. [24] Cruz RKS, Alfonso A, Lourenço MLC, Ulian CMV, Sudano MJ, Oba E, Padovani CR, Ramos PRR, Chiachchio SB. Electrolyte, blood gas and electrocardiographic profile of neonatal foals in the first 48 hours of life. Acta Sci Vet 2015;43:1321. [25] Rose RJ, Rossdale PD, Leadon DP. Blood gas and acid-base status in spontaneously delivered, term-induced and induced premature foals. J Reprod Fertil Suppl 1982;32:521e8. [26] Vaala WE. Peripartum asphyxia. Vet Clin N Am Equine Pract 1994;10: 187e218. [27] CLSI. Defining, establishing and verifying reference intervals in the clinical laboratory. Approved guideline. third ed. Wayne, PA: Clinical and Laboratory Standards Institute; 2008. CLSI document EP28-A3c. [28] Mukaka MM. Statistics corner: a guide to appropriate use of correlation coefficient in medical research. Malawi Med J 2012;24(3):69e71. A, Friedrichs K, Harr K, Concordet D, Trumel C, Braun J. Reference [29] Geffre values: a review. Vet Clin Pathol 2009;38(3):228e98. [30] Belai Y, Goodwin TM, Durand M, Greenspoon JS, Paul RH, Walther FJ. Umbilical arteriovenous PO2 and PCO2 differences and neonatal morbidity in term infants with severe acidosis. Am J Obstet Gynecol 1998;17(Pt 1):13e9. [31] Gjerris AC, Staer-Jenson J, Jorgensen JS, Bergholt T, Nickelsen C. Umbilical cord blood lactate: a valuable tool in the assessment of fetal metabolic acidosis. Eur J Obstet Gynecol Reprod Biol 2008;139:16e20. [32] Victory R, Penava D, daSilva O, Natale R, Richardson B. Umbilical cord pH and base excess values in relation to adverse outcome events for infants delivering at term. Am J Obstet Gynecol 2004;191:2012e28. [33] Westgren M, Divon M, Horal M, Ingermarsson I, Kublickas M, Shimojo N, Nordstrom L. Routine measurements of umbilical artery lactate levels in the prediction of perinatal outcome. Am J Obstet Gynecol 1995;173(5):1416e22. [34] Linet T, Laporte J, Gueye H, Boog G. Microvolume dosage of lactate in cord blood for the evaluation of the neonatal well-being. J Gynecol Obstet Biol Reprod 2002;31(4):352e7. [35] Nordstrom L, Ingemarsson I, Persson B, Shimojo N, Westgren M. Lactate in fetal scalp blood and umbilical artery blood measured during normal labor with a test strip method. Acta Obstet Gynecol Scand 1994;73(3):250e4. [36] Shirey T, St Pierre J, Winkelman J. Cord lactate, pH, and blood gases from healthy neonates. Gynecol Obstet Invest 1996;41(1):15e9. [37] Tuuli M, Stout MJ, Shanks A, Odibo AO, Macones GA, Cahill AG. Umbilical cord arterial lactate compared with pH for predicting neonatal morbidity at term. Obstet Gynecol 2014;124(4):756e61. [38] Tuuli M, Stout MJ, Macones GA, Cahill AG. Umbilical cord venous lactate for predicting arterial lactic acidemia and neonatal morbidity at term. Obstet Gynecol 2016;127(4):674e80.