Umbilical arterial lactate levels after normal vaginal and elective cesarean delivery: The role of a longer active second stage most significant in high levels after vaginal delivery

To evaluate umbilical arterial lactate concentrations after spontaneous vaginal delivery and after elective Cesarean delivery, and to study the simultaneous effects of maternal and obstetric variables in high lactate levels in vaginally delivered healthy term singletons.


INTRODUCTION
Birth is a risk for the pregnant woman as well as for the fetus.Fortunately, one of the most feared neonatal complications-severe birth asphyxia leading to hypoxic-ischemic cerebral injury-is quite rare, occurring in 0.1%-0.8% of all live births in developed countries. 1,2These numbers are consistently several times higher in many lower income countries. 2When this intrapartum tragedy occurs in newborns, it can lead to devastating consequences, including the need for lifelong special care.To avoid birth asphyxia, fetal well-being during labor is ensured in most western labor wards by different means-mostly cardiotocography, the STANmethod, and/or analyzing fetal scalp blood samples in cases with reassurance.Immediately after birth, lactate values are measured from umbilical blood samples, particularly in those cases with a labor that was possibly compromised.Umbilical values are useful parameters in predicting neonatal morbidity.][9][10][11] However, no clear consensus exists on the normal reference range for umbilical lactate values or for fetal values when converting from vaginal delivery to an emergency Cesarean or other assisted vaginal delivery in the case of threatened intrapartum asphyxia.3][14][15][16] However, no earlier studies exist on the significance of the duration of rupture of membranes (ROM), maternal intrapartum infection, or infertility in umbilical blood lactate levels, and studies on the simultaneous relationships of the maternal and obstetric factors to the umbilical blood lactate levels are sparse.
In this study, we evaluated the range of the umbilical artery (UA) blood lactate level in normal spontaneous vaginal deliveries (SVDs) and elective cesarean deliveries (ECDs), as well as obstetric factors that were related to the increased lactate levels.Our hypothesis was that specific maternal and labor-related factors (e.g., duration of labor and ROM) do disrupt the evaluation of umbilical lactate values after delivery.The aim of this study was to gain more information about the levels before the onset of labor, as well as of the normal birth process and the possible factors that were related to the high lactate values in UA samples and possible subclinical birth asphyxia.

METHODS
This was a prospectively collected birth register-based clinical study conducted in Kuopio University Hospital (KUH) during the 5-year period between August 2013 and August 2018.KUH is a tertiary hospital with annual delivery rates of 2100-2500.All perinatal data were collected prospectively at the time of birth.During the 5-year study period, there were 11 626 deliveries.The cesarean delivery rates varied annually between 12.0% and 13.8%, and vacuum-assisted delivery rates were between 7.7% and 9.7%.
Information on the maternal, fetal, intrapartum, and early neonatal characteristics were routinely collected by the women themselves, by certified midwives, or by laboratory staff during the time of delivery, and entered into a Haikara electronic database.This database has been described in detail in our earlier study reports. 17Briefly, clinical information concerning pregnancy, delivery, early postpartum, and the early neonatal period was obtained from the electronic medical database.Prenatal data included maternal age and parity (0, 1, ≥2), numbers of earlier gestations (0, 1, ≥2), maternal height (cm), weight (kg), and body mass index (BMI) in the first trimester (kg/m 2 ), gestational weight gain (kg), maternal smoking (any smoking during pregnancy, none, missing value), marital status (married, in relationship but not married, single), infertility treatment before index pregnancy (no/yes), the duration of delivery (I and II stage We recorded lactate values in umbilical blood samples in a prospective way in everyday clinical labor practice in KUH for more than 10 years.Umbilical samples were routinely analyzed in nearly all births, even without evidence or any suspicion of fetal compromise and, during the study period, the concentrations of UA lactate were routinely collected in 90.4%-94.6% of all newborns.After clamping of the umbilical cord, the attending midwife collected the UA samples which were carried immediately to the analyzers, which were situated in a nearby labor ward.The UA lactate was determined using a point-of-care testing device, StatStrip ® (Steripolar).
Data were analyzed by SPSS 21.0 Windows software.We reported the normal variation and percentile limits of UA lactate values and selected into analysis only singleton pregnant women who had undergone ECD without the onset of labor or ROM, or had SVD.We excluded all women who had vacuum-assisted, vaginal breech, or emergency cesarean deliveries.We defined the following from our large hospital database: (1) the variation of UA lactate values in newborns who were delivered either by ECD or by SVD; (2) the evaluation of whether the duration of different stages of labor or ROM were associated with the UA lactate values; and (3) the evaluation of whether there were other clinical characteristics that were associated with the higher (equal to or more than 90th percentiles) UA lactate values in cases with a normal early neonatal outcome.Continuous variables were presented as means and standard deviations (SDs) and grouped either by tertiles (maternal and fetal characteristics) or quintiles (duration of labor and ROM) for analysis.Categorical data were presented as numbers and percentages.The analysis of variance was used to compare continuous variables and Pearson's chi-squared test was used for categorical data.A Pearson's correlation test was used to evaluate the linear correlation between Apgar scores and UA lactate values.Logistic regression analyses were conducted to identify contributing maternal, intrapartum, and neonatal factors to high UA lactate values (90th percentile; ≥5.8 mmol/L) in vaginally delivered term neonates with normal early neonatal outcomes (N = 6541).Preterm newborns (<37 gestational week), those with low Apgar scores at 5 min (≤7), need for NICU and/or with diagnosis of intrauterine hypoxia or perinatal asphyxia (International Classification of Diseases and Related Health Problems 10th P20 or P21) were excluded, since we evaluated the variation of high UA lactate values in normal, uncomplicated term deliveries and the possible explicators of high values other than asphyxia.
Variables were included in the final logistic model (forward stepwise method) if they were significant at a p value of less than 0.10 in univariate analysis.Maternal parity correlated strongly with the number of earlier gestations, as well as with maternal age, and it was the more significant variable in univariate analysis compared to the number of earlier gestations and maternal age.Thus, parity was only included in the logistic model to avoid multicollinearity.Statistical significance was set at a p value below 0.05.

ETHICAL APPROVAL
The research was approved by the Ethics Committees of the Kuopio University Hospital (5302500, 84/2014, and 1474/2019) and the Hospital District of Central Finland, Jyväskylä (18 U/2011, 6.10.2016).

RESULTS
Figure 1 describes the flowchart of the 7723 participants with UA lactate samples.Nearly all UA lactate samples (N = 7405 95.9%) were collected from women delivering at term; only 318 (4.1%) women had a preterm delivery.Table 1 describes the mean values and median, 85th, 90th, and 95th percentiles of UA values by the mode of delivery (either by SVD or ECD).Newborns who were born by ECD had significantly lower mean lactate values compared to newborns of women who had undergone SVD (2.42 [SD 0.94] vs. 3.56 [SD 1.62] mmol/L; p < 0.0001) (Table 1).In total, 18.1% (1322/7301) of newborns who were delivered by SVD had UA lactate values higher than 4.80 mmol/L.The analogous number of those who were born by ECD was 2.8% (12/422).
Tables 2 and 3 show the characteristics of the participants and their newborns in relation to the mean lactate levels by the mode of delivery.Table 2 shows that all maternal clinical characteristics except BMI were significantly associated with the UA lactate values in newborns born by SVD.Younger, unmarried, less parous, shorter women, and those who had undergone infertility treatments before pregnancy, who gained more weight during gestation and who had an infection during delivery had higher UA lactate values compared to others.Neonates of smokers had similar UA lactate levels compared to those of non-smokers; levels were significantly lower only in neonates of the missing value group.No significant associations were noted between the same clinical variables and values in newborns delivered by ECD.  by ECD had values associated with the U-shaped mode; preterm newborns and those who were born ≥41 gestational weeks had the highest UA lactate values versus those born at term.However, only nine newborns were born by ECD after ≥41 gestational weeks.There were 10 (0.13%) newborns with birth asphyxia.They had significantly higher UA lactate levels compared to the others (8.46 vs. 3.55 mmol/L; p < 0.0001) and all were delivered by SVD.The need for NICU (16.4% vs. 7.5%; p < 0.0001), as well as mechanical ventilation (4.3% vs. 0.7%; p < 0.0001), was significantly more common in newborns born by ECD compared to those born by SVD.Above all, newborns that needed NICU had significantly higher UA lactate levels in both types of deliveries, but the need for mechanical ventilation was associated with higher UA lactate levels only in newborns who were delivered by SVD.
The duration of labor (first and active second stages and combined) and ROM correlated positively with the UA lactate values in term newborns born vaginally (Table 4).The strongest significant univariate association was recorded between the duration of active stage II and UA levels.
Table 5 shows the multivariable analysis of different significant clinical predictors to the 90th percentile (≥5.8 mmol/L) lactate values of healthy term newborns with normal early outcomes after SVD (N = 6541).The independent risk factors for high UA lactate value levels were the following: longer active second stage of labor, longer duration of ROM, maternal infection, null parity, maternal shortness, and higher gestational age at the time of birth.The longer duration of active second stage of labor was the strongest predictive factor in the model, and these six significant factors explained 18% of the UA lactate level variation.Note: There were significant ( p < 0.0001) associations between every examined time variable and umbilical arterial lactate levels estimated by one-way ANOVA-test.

DISCUSSION
As expected, the data from this large prospectively collected birth register similar to earlier studies, showed that UA lactate concentrations were significantly higher in those newborns who were born by SVD compared to those who were born by ECD. 12,15We were the first to show that high lactate levels in UA samples correlated independently in multivariable analysis with a longer duration of the active second stage of labor and of ROM, but also with maternal shortness, null parity, maternal infection and higher gestational age at birth.Our results provide new and valuable information about the normal birth process, even though we cannot totally exclude the presence of subclinical or transient birth asphyxia in our dataset.In line with the results of The associations between high level of umbilical lactate and prenatal and obstetric variables in 6541 healthy term neonates after spontaneous vaginal delivery.previous studies, the duration of the active second stage of labor was the strong effector to the high lactate levels, 10,13 but we also confirmed the previously reported significant association between higher gestational age and increased umbilical lactate levels. 14,15,18In addition, there were many other clinically significant factors in the analysis, which were related to high UA lactate levels.Some of these, such as first delivery and short stature of the mother, have been considered previously as important risk factors for high fetal or umbilical lactate levels, 16,19 but the duration of ROM, maternal intrapartum infection, and infertility are less-studied clinical variables in this context.Surprisingly, after 2 h of duration of ROM (third quintile), the risk for higher UA lactate levels was nearly two-fold (compared to first quintile, i.e., when it was equal to or less than 38 min); it increased even more thereafter.Unfortunately, we did not have information about the mode of fetal membrane rupture to indicate whether it was spontaneous or caused by amniotomy.Maternal infection was significantly associated with higher UA lactate levels, both in uni-and multivariable analysis, as suggestive for poorer neonatal outcome. 20,21Its significance was independent even the duration of ROM and null parity were controlled in logistic model.The strengths of this study include its large database, which was collected prospectively in one university hospital.We were interested in normal deliveries and thus able to select only SVDs and ECDs in this study, excluding other operative deliveries, which are often due to (or associated with) more intrapartum asphyxia complications.In general, our recent report offers important new indices for delivery units for the evaluation of the UA lactate values in neonates who are born by ECDs and SVDs.It is, however, possible that some normal vaginally born newborns with normal Apgar scores had imminent intrapartum asphyxia and their condition after birth would have been worse if their labor had lasted longer.We hope that our results will help clinicians to interpret the UA lactate values in relation to the duration of labor and ROM, for example, in cases when parturients with expeditious labor, or after elective cesarean without onset of labor, have abnormally high UA lactate values.This suggests that the asphyxiant hazard would have already occurred before the onset of labor or is related to some chronic intrauterine condition.
Our data were indiscriminate in terms of participants.UA samples were routinely collected for analysis after nearly every delivery and sampling succeeded in more than 90% of all labors.Most previous studies that have evaluated UA lactate values in the prediction of neonatal outcome have not included the type of delivery in their study.Unfortunately, we were unable to recall the quality and validity of intrapartum cardiotocography tracings or the possible need for fetal scalp sampling from birth records to understand more profoundly increased UA lactate levels.However, in the total study population, Apgar scores at 5 min were equal to or more than eight in more than 98% of newborns.In addition, our birth register is valid, and its data have been used and analyzed in many earlier studies. 22,23ne weakness of this study is that the present results mirrored the situation after SVD or ECD, although in clinics we are more interested in labors and neonates with a suspicion of birth asphyxia.Second, we were not able to report the use of epidurals or oxytocin in this study.Third, it is always possible that during the labor, maternal levels of lactate are considerably higher because of maternal labor pain and distress.Some studies have shown a correlation between concomitantly measured maternal and umbilical lactate levels. 12,13It is possible to further analyze umbilical venous values and compare them to the arterial values, and the placenta may also have its own lactate-producing role during delivery. 24onetheless, there is no global consensus on definitions of normal lactate levels in umbilical samples 25 but, according to our results, 17.3% of all newborns had UA lactate values greater than 4.80 mmol/L-the internationally accepted value defined as fetal acidemia in intrapartum fetal scalp blood samples. 19,26In newborns delivered by SVD, the 85th percentile value of lactate was 5.10 mmol/L and the 90th percentile value for lactate was 5.80 mmol/L, even in term newborns with normal Apgar scores.However, we need more follow-up studies on the significance of higher lactate values in long-term health outcomes, especially among offspring who are regarded as normal term newborns at the time of birth.As far as we know, there have been only two follow-up studies evaluating this item with minor neurodevelopmental differences between cases and controls. 27,28n conclusion, our study provides more evidence that UA lactate values should always be interpreted taking into consideration the mode of delivery and onset of labor, particularly whether the newborn is delivered by elective cesarean or by the vaginal route.Furthermore, for neonates who are delivered by the vaginal route, the duration of labor and ROM also need to be considered.Characterization of the potential clinical risk factors of high umbilical lactate levels in "normal" newborns is also crucial for understanding the development and recognition of the intrapartum asphyxia.Future research should evaluate the safety of common obstetric interventions during labor and their possible relationship with the development of birth asphyxia, as well as whether high UA lactate values in normal newborns after birth play any role in their long-term health.
[active] and combined I + II), duration of ROM, mode of delivery (either SVD or ECD), maternal intrapartum infection (recorded by International Classification of Diseases and Related Health Problems 10th revision codes [ICD-10], such as O41.1, O75.2, or O75.3), and duration of gestation at the time of delivery (weeks).Sex, birth weight, and 5-minute Apgar score, need for neonatal intensive care unit (NICU) treatment, mechanical ventilation, and length of hospital stay after delivery were recorded from newborns and pediatricians evaluated their birth diagnosis at the time of leaving hospital.Every newborn was characterized according to the ICD-10 codes, and those with possible birth asphyxia were recorded as P20 intrauterine hypoxia or P21 birth asphyxia.The definition was based on at least two of the following clinical signs or findings: (1) Apgar score ≤4 at 5 min; (2) metabolic acidosis pH ≤7.0; (3) failure to establish breathing at birth and need for assisted ventilation; (4) encephalopathy; (5) presence of convulsions in the first 24 h of life; and (6) multiple organ dysfunction (encephalopathy and the involvement of at least one organ).Further, 445 (4.8%) newborns with congenital anomalies were excluded from the study.Congenital anomalies were detected and recorded after birth by the attending midwife or doctor, mostly with Q-diagnosis (ICD-10).

Figure
Figure Umbilical lactate values after birth and maternal and prenatal variables of 7723 parturients with either spontaneous vaginal or elective cesarean delivery.Apgar scores ≤7 at 5 min.UA lactate values correlated inversely with Apgar scores in both newborns born by SVD (r = À0.192;p<0.0001) and by ECD (r = À0.197;p<0.0001).The duration of gestation at the time of birth was significantly associated with the UA lactate values in both groups (Table3).In newborns born by SVD, values correlated positively in a dose-dependent manner with gestational age.Newborns who were born ≥41 gestational weeks had the highest values.Newborns who were born T A B L E 3 Umbilical lactate values after birth and early neonatal characteristics of 7723 singleton newborns in relation to the type of delivery.
T A B L E 1 The mean and percentile values of umbilical arterial lactate in 7723 singleton neonates born by spontaneous vaginal or elective cesarean delivery.Note: p-Values estimated between different modes of delivery by one-way ANOVA-test.Abbreviation: SD standard deviation.TA B L E 2 Note: p-Values estimated between variables by one-way ANOVA-test.Maternal infection was recorded by ICD-10 codes O41.1, O75.2 or O75.3.Abbreviations: BMI body mass index; NA not available.In total, 144 out of 7723 (1.9%) neonates had Umbilical arterial lactate and duration of different stages of labor and ruptured fetal membranes in 7027 term neonates delivered by spontaneous vaginal delivery.
T A B L E 4 Adjusted p-values were estimated by logistic analysis by stepwise forward method.Only significant variables in multivariate analysis are shown in table.Besides shown variables, infertility treatment before ongoing pregnancy, gestational weight gain and marital status were included in the analysis, but they were insignificant in analysis.Abbreviations: CI confidence interval; OR odds ratio. Note: