http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, Early Online: 1–5 ! 2015 Informa UK Ltd. DOI: 10.3109/14767058.2015.1045863
ORIGINAL ARTICLE
Lactate versus pH levels in fetal scalp blood during labor – using the Lactate Scout System
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
Christina Rørbye, Anette Perslev, and Carsten Nickelsen Department of Obstetrics and Gynecology, University of Copenhagen, Hvidovre Hospital, Hvidovre, Denmark
Abstract
Keywords
Objective: To assess if lactate measured with the Scout Lactate System is a reliable alternative to pH in intrapartum monitoring of the fetus. Methods: A prospective study analyzing (1) the correlation between scalp lactate measured by the Scout Lactate System and the Automatic Blood Laboratory (ABL), (2) the correlation between lactate and pH measured in scalp blood and (3) the correlation between fetal scalp lactate and umbilical cord SBE. The sensitivity/specificity and positive/negative predictive values of lactate in predicting low pH were analyzed and expressed as Receiver Operating Curves (ROC). Results: Lactate measured by the Scout Lactate System and the ABL correlated well (r2 ¼ 0.85). Both lactate and pH were measured in 1009 scalp blood samples. The sensitivity and specificity of lactate 4.8 mmol/l in predicting a pH57.20 were 0.63 and 0.85, respectively. The correlation between scalp lactate measured within 15 min prior to delivery and the umbilical cord SBE was low. Conclusion: Monitoring non-reassuring deliveries with scalp lactate instead of pH would have resulted in more (155 instead of 56) instrumental deliveries with no decrease in newborns with severe metabolic acidosis.
Neonatal asphyxia, ROC, sensitivity, specificity
Introduction Intrapartum monitoring of the fetus during labor is crucial to obtain a good neonatal outcome. Cardiotocography (CTG) tracing of the fetal heart rate is used to recognize and intervene in cases of suspected fetal hypoxia in order to prevent adverse perinatal outcome. However, the sensitivity and specificity of CTG patterns are low, and the intraobserver variation of the interpretation is high. While a normal CTG is reassuring, only few of the cases with nonreassuring CTG traces reflects fetuses with hypoxia leading to a high risk of unnecessary interventions. When CTG is non-reassuring, fetal scalp blood sampling and measurements of the pH may assist in discriminating the non-hypoxic fetus from the fetus with asphyxia. Analyzing pH requires 30–50 ml of fetal blood while lactate measurements only require 5 ml resulting in much lower risks of sampling failures [1]. While high lactate values reflect metabolic acidosis, the pH measurement does not discriminate between respiratory and metabolic acidosis. In a previous RCT [2], lactate and pH measurements had the same sensitivity in predicting fetal asphyxia, but the failure rate of pH Address for correspondence: Christina Rørbye, Department of Obstetrics and Gynecology, University of Copenhagen, Hvidovre Hospital, Kettegaard Alle´ 30, DK-2650 Hvidovre, Denmark. Tel: +45 3862 5520. Fax: +45 3862 2135. E-mail: [email protected]
History Received 9 January 2015 Revised 14 April 2015 Accepted 24 April 2015 Published online 25 May 2015
measurements was 11%, while the failure rate of lactate measurements was negligible. In previous studies, lactate has been analyzed by the single-strip method Lactate Pro (Arkray, Kyoto, Japan). A new single-strip method (Scout Lactate, EKF Diagnostics, Germany) requires only 2–5 ml of fetal blood and is based on the same enzymatic amperometric detection method. The aim of this study was threefold: First to test the agreement between scalp blood lactate measured on the Scout Lactate System and on the Radiometer ABL 700 blood–gas analyzer in a clinical setting. Next, to compare lactate and pH measured in the same blood samples during delivery and analyze the sensitivity and specificity of lactate measurements in predicting scalp pH. Finally, to evaluate if lactate is a reliable alternative to pH measurements in intrapartum monitoring of the fetus when CTG is non-reassuring.
Methods The study was performed at Copenhagen University Hospital, Hvidovre during the period of 1 March 2012 and 30 August 2013. We included women in labor at term with a singleton in the cephalic presentation with a non-reassuring CTG either during the first or the second stage of labor. Fetal scalp blood samples were obtained, pH was measured and further management of the labor was decided by the obstetrician on call. On the same blood sample trained auxiliary nurses
2
C. Rørbye et al.
J Matern Fetal Neonatal Med, Early Online: 1–5
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
measured lactate using the Scout Lactate system (EKF Diagnostics, Germany), and the result was recorded outside the labor room – blinded to the obstetrician. The capillary blood pH was measured in a Radiometer BPH5 blood gas analyzer (BPH5) while pH, pCO2, SBE and lactate were measured on a Radiometer ABL700 blood gas analyzer (ABL). The ABL and BPH5 use glass-pH electrodes for pH measurement, and the ABL is equipped with an enzymatic amperometric biosensor for lactate measurement. The sensors are calibrated daily. Intervention or proceeding of labor was based on the pH value only, as the obstetrician was not informed about the lactate value. The umbilical artery blood was analyzed on the ABL immediately after delivery. Metabolic acidosis was defined as pH 57.0 and SBE 510. Statistics The relationship between lactate measured by Scout Lactate and Radiometer ABL was calculated by linear regression analysis. Sensitivity and specificity of lactate in predicting scalp pH were calculated for lactate values from 1 to 10 mml/l and presented in a ROC diagram. Cut-off values of 4.2 and 4.8 mmol/l were chosen. Comparison of scalp lactate and umbilical cord SBE was done when a scalp blood sample was performed within 15 min prior to delivery. Furthermore, it was calculated how many interventions would have been made, had lactate been the gold standard. Ethical considerations The Lactate Scout device and strips were sponsored by EKF Diagnostics, Germany. The company had no impact on the protocol, the results of the study, nor the way they are presented. The investigators have no economical connection to the company. The study is an evaluation of two different laboratory equipments, and according to the ethical committee, no further approval was needed (H-2-2014-FSH78). The analysis was performed on blood samples accepted to be evaluated for fetal acid–base changes as a routine clinical procedure. Measuring lactate did not require any extra intervention. The results had no impact on management of labor, and the procedure implied no risks to neither the women nor the fetuses. The study was approved by the Danish Data Registry (J nr 2012-41-1415).
Results In 264 fetal scalp blood samples lactate was measured both on the Radiometer ABL700 and the Scout Lactate system, and the results were closely interrelated with a coefficient of correlation of r2 ¼ 0.85 (Figure 1). Both pH and lactate were measured in 1009 fetal blood samples during 677 deliveries, see Table 1 for description of the study population. Fetal scalp pH was inversely related to lactate with r2 ¼ 0.73 (p50.0001) as shown in Figure 2. Using pH as the gold standard and pH 57.20 as the limit for acidemia, the sensitivity and specificity of a lactate value 4.8 were 0.63 (45/71) and 0.85 (795/938), respectively (Figure 3). The corresponding positive and negative predictive values were 0.24 (45/188) and 0.97 (795/821). For pH 57.15 the sensitivity and specificity were 0.86 (19/22) and
Figure 1. The correlation between lactate analyzed on the Scout Lactate system and the Radiometer ABL (n ¼ 264). Table 1. Study population. Mean (range) or number (%) Maternal age, years Gestational age, days BMI Nulliparous Multiparous Induction of labor Acute cesarean section Ventouse delivery Umbilical artery pH 57.0 and SBE 412 mmol/l Umbilical artery pH 57.0 and SBE 12 mmol/l
31 284 24 539 138 283 120 199 7 1
(17–42) (239–297) (17–47) (80%) (20%) (42%) (18%) (29%) (1.05%) (0.15%)
0.82 (818/987) (Figure 4). The sensitivity and specificity of a normal lactate 54.2 to predict a pH 7.20 were 0.98 (655/ 665) and 0.82 (283/344). Of the 677 monitored deliveries a scalp pH57.20 was seen in 56 and delivery was performed either by caesarean section (n ¼ 12) or vacuum extraction (n ¼ 44). Seventeen of these women had a scalp lactate value 4.8 mmol/l. One hundred and fifty-five of the women had at least one scalp blood measurement with a lactate 4.8 mmol/l. If intervention had been based on the lactate value, an additional 99 instrumental deliveries would have been performed in these 677 women. In cases with measurement of a scalp lactate within 15 min prior to delivery (n ¼ 104) the lactate value was compared to the arterial umbilical cord standard base excess (SBE) obtained immediately after delivery (Figure 5). A lactate 4.8 mmol/l was found in 50 scalp blood samples with SBE 412 in cord blood samples, thus the false positive rate was 48%.
Discussion In a clinical setting, we found that lactate measured on the Scout Lactate System correlates well with lactate measured
DOI: 10.3109/14767058.2015.1045863
Lactate versus pH in scalp blood
3
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
Figure 2. Lactate versus pH measured in the same fetal scalp blood sample (n ¼ 1009).
Figure 3. ROC for pH 57.20.
on the Radiometer ABL. Assuming pH as the gold standard, the sensitivity and specificity of a high lactate value to predict a low pH value are low. Monitoring non-reassuring deliveries with scalp lactate instead of pH would have resulted in more (155 instead of 56) instrumental deliveries without a decrease in the number of newborns with severe metabolic acidosis (pH 57.0 and SBE 512). Both pH and lactate measuring in fetal scalp blood during labor are valuable tools to discriminate the non-hypoxic fetus from the fetus with acidemia. Whether pH or lactate is superior is still unanswered as they reveal different aspects of the acid–base status of the fetus, and different fetuses might have different acid–base levels of clinical relevance. Several scalp blood samples from the same woman were accepted as individual measurements, as they are believed to be independent. The optimal cut-off level of lactate is not well-established and varies from 4.2 to 6.6 mmol/l [3,7]. It has been shown that lactate increase with increasing gestational age [4] and different cut-offs might provide more precise knowledge of when to intervene. In this study, we used a cutoff of lactate 4.8 and 54.2 mmol/l as these are most often
Figure 4. ROC for pH 57.15.
used. According to the ROC diagram calculated in our study these cut-offs have an acceptable accuracy. During the first stage of fetal distress the fetal–maternal exchange is slightly compromised. The oxygen tension decreases and the carbon dioxide tension increases. The pH in fetal blood decreases, and the hemoglobin dissociation curve is shifted to the right (the Bohr Effect). As long as the oxygen tension in the placenta only decreases slightly, the change in hemoglobin dissociation will facilitate the release of oxygen from the fetal blood in the tissue (oxygen unloading capacity), and aerobic metabolism is retained. This respiratory acidosis is well-tolerated by the fetus identified by a low pH and normal SBE. In the second stage of fetal distress the oxygen tension is further reduced and the oxygen-uptake of the hemoglobin in the placenta decreases as a consequence of the shifted hemoglobin dissociation curve. In this situation there is no benefit of the increased oxygen unloading capacity and the oxygen tension in the fetal tissue is insufficient for aerobic metabolism [5]. During oxygen deficiency the anaerobic metabolism of one mole of glucose results in production of two mole of lactic acid. The metabolic acidosis
4
C. Rørbye et al.
J Matern Fetal Neonatal Med, Early Online: 1–5
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
Figure 5. Fetal scalp lactate (515 min prior to delivery) versus umbilical cord artery SBE.
is an indicator of severe hypoxia and can be identified by a low pH, a low SBE and a high lactic acid concentration in the blood. The fetal well-being during labor is monitored using the acid status of fetal scalp capillary blood. Theoretically, the first warning should be a decrease of the pH during the first stage of distress, while the second stage of fetal distress is characterized by decreasing SBE and increasing lactic acid concentrations as a consequence of an anaerobic metabolism. In the clinical situation of a delivery, the demand for a method with a high sensitivity and specificity to detect fetal acidemia is crucial. The ideal monitoring tool would be able to discriminate between a respiratory and an early metabolic acidosis in time to prevent severe metabolic acidosis. A low sensitivity is unacceptable, as an acidotic fetus may never be missed; while a low specificity would result in unnecessary instrumental deliveries. The use of fetal scalp blood pH as the gold standard in monitoring deliveries with a non-reassuring CTG pattern reduces the use of unnecessary instrumental interventions. However, a low pH cannot discriminate a respiratory and a metabolic acidosis, but as the metabolic acidosis is preceded by a respiratory acidosis, and in order to prevent of metabolic acidosis, intervention could preferably be planned and performed at the time of a respiratory acidosis. In our labor ward we monitor high risk deliveries with CTG, and in 25% we found indication for pH measuring in a fetal scalp blood sample. With this set-up we performed 319 instrumental deliveries in the 667 deliveries with nonreassuring CTG, and only one child (0.15%) was born with severe metabolic acidosis (pH 57.0 and SBE 512). If the 677 deliveries had been monitored with fetal scalp lactate using a level of intervention of 4.8 mmol/l, we would have made an additional 99 interventions, while no cases of acidosis would have been missed. A discrepancy between normal, pre-acidotic and acidotic levels of lactate and pH in the same blood sample has been reported in 55% of scalp blood samples in the study by Liljestro¨m et al. [6]. In this study, the frequency of operative delivery due to fetal distress was 82% of cases with both an abnormal scalp pH and lactate measurement compared to 53% when only one of the parameters was abnormal value.
None of the infants in this study were born with metabolic acidosis. This implies that a combined test could have advantages in fetal surveillance, but this needs further testing in clinical trials. A significant although poor correlation between both fetal scalp lactate and fetal scalp pH versus umbilical cord pH (r ¼ 0.21 and r ¼ 0.40, respectively) was shown by Holzmann et al. [7]. However, among 85 fetal scalp lactate measurements 46.6 mmol/l, only 14 had cord pH levels 57.10. Heinis et al. advocate that scalp lactate provides more accurate information on fetal acid–base status at birth than does pH and/or SBE [8]. Nonetheless, 29% of acidotic children at birth (pH 57.05 and SBE 512) were not identified using a lactate limit of 6.6 mmol/l in the fetal scalp blood sample. Measuring lactate in a scalp blood sample is an easier procedure, as it can be done by a hand hold device using only 5 ml of blood, compared to pH measuring that requires 35–100 ml of blood and an ABL unit located close to the delivery room. These challenges result in failure rates of 11–20% in intended pH measurements in some delivery wards [1,2,9,10]. Furthermore, the time from deciding to do a fetal scalp blood sample to the obstetrician receives the result is shorter when analyzing lactate than pH [11]. In a multicenter RCT comparing the outcome of deliveries monitored by fetal scalp pH (47.25 and57.21) and lactate (54.2 and44.8 mmol) no differences were found in the frequency of cord acidemia [2]. The conclusion was based on an intention to treat analysis but included failed analysis in 10% of the pH group and 1% of the lactate group. At Hvidovre Hospital, Copenhagen, the pH measurement in fetal scalp blood has a failure rate of less than 1%. This may be explained by a high routine of the staff since the department has more than 7000 deliveries annually and scalp blood measurements are performed in all cases of a non-reassuring CTG. A high lactate value might be a better predictor of neurologic disability [12], but whether pH or lactate is superior in monitoring high risk deliveries is still not answered, as they might mirror different clinical aspects of asphyxia. This study however shows, that in our clinical setting, we would have performed more instrumental deliveries with no decrease in newborns with metabolic
DOI: 10.3109/14767058.2015.1045863
acidosis, if lactate had replaced pH in monitoring the fetus during labor. In conclusion, lactate measured on the Scout Lactate system and the ABL correlates well. The sensitivity, specificity, PPV and NPV of a high lactate to predict a low pH are low, thus in our clinical setting, more instrumental deliveries would be performed using lactate instead of pH in monitoring fetuses during labor.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
Declaration of interest No declaration of interest. The authors report no conflicts of interest in connection with the article. The authors alone are responsible for the content and writing of the paper. The Lactate Scout device and strips were sponsored by EKF Diagnostics, Germany. No other funding was obtained.
References 1. Westgren M, Kruger K, Ek S, et al. Lactate compared with pH analysis at fetal scalp blood sampling: a prospective randomized study. Br J Obstet Gynecol 1998;105:29–33. 2. Wiberg-Itzel E, Lipponer C, Norman M, et al. Determination of pH or lactate in fetal scalp blood in management of intrapartum fetal distress: randomised controlled multicentre trial. BMJ 2008;336: 1284–7.
Lactate versus pH in scalp blood
5
3. East CE, Leader LR, Sheehan P, et al. Intrapartum fetal scalp lactate for fetal assessment in the presence of a non-reasurring fetal heart rate trace. Cochrane Database Syst Rev 2010;3:CD006174. 4. Wiberg N, Ka¨llen K, Herbst A, et al. Lactate concentration in umbilical cord blood is gestational age-dependant: a populationbased study of 17 867 newborns. BJOG 2008;115:704–9. 5. Wimberley PD. Fetal hemoglobin: 2,3-diphosphoglycerate and oxygen transport in the newborn premature infant (thesis). Scand J Clin Lab Invest Suppl 1982;160:1–149. 6. Liljestro¨m L, Wikstro¨m AK, Hanson U, et al. Evaluation of the discrepancy between pH and lactate in combined fetal scalp blood sampling. Acta Obstet Gynecol Scand 2011;90:1088–93. 7. Holzmann M, Cnattingius S, Nordstro¨m L. Outcome of severe intrapartum acidemia diagnosed with fetal scalp blood sampling. J Perinat Med 2011;39:545–8. 8. Heinis A, Spaanderman M, Gunnewiek J, Lotgering F. Scalp blood lactate for intra-partum assessment of fetal metabolic acidosis. Acta Obstet Gynecol Scand 2011;90:1107–14. 9. Nordstro¨m L. Fetal scalp blood measurements during labour – lactate or pH? Clin Biochem 2011;44:456–7. 10. Ramanah R, Martins A, Clement MC, et al. Fetal scalp lactate microsampling for non-reassuring fetal status during labor: a prospective observational study. Retal Diagn Ther 2010; 27:14–19. 11. Nordstro¨m L. Fetal scalp and cord blood lactate. Best Pract Res Clin Obstet Gynecol 2004;18:467–76. 12. Kruger K, Hallberg B, Blennow M, et al. Predictive value of fetal scalp blood lactate concentration and pH as markers of neurologic disability. Am J Obstet Gynecol 1999;181:1074–8.
ORIGINAL ARTICLE
Lactate versus pH levels in fetal scalp blood during labor – using the Lactate Scout System
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
Christina Rørbye, Anette Perslev, and Carsten Nickelsen Department of Obstetrics and Gynecology, University of Copenhagen, Hvidovre Hospital, Hvidovre, Denmark
Abstract
Keywords
Objective: To assess if lactate measured with the Scout Lactate System is a reliable alternative to pH in intrapartum monitoring of the fetus. Methods: A prospective study analyzing (1) the correlation between scalp lactate measured by the Scout Lactate System and the Automatic Blood Laboratory (ABL), (2) the correlation between lactate and pH measured in scalp blood and (3) the correlation between fetal scalp lactate and umbilical cord SBE. The sensitivity/specificity and positive/negative predictive values of lactate in predicting low pH were analyzed and expressed as Receiver Operating Curves (ROC). Results: Lactate measured by the Scout Lactate System and the ABL correlated well (r2 ¼ 0.85). Both lactate and pH were measured in 1009 scalp blood samples. The sensitivity and specificity of lactate 4.8 mmol/l in predicting a pH57.20 were 0.63 and 0.85, respectively. The correlation between scalp lactate measured within 15 min prior to delivery and the umbilical cord SBE was low. Conclusion: Monitoring non-reassuring deliveries with scalp lactate instead of pH would have resulted in more (155 instead of 56) instrumental deliveries with no decrease in newborns with severe metabolic acidosis.
Neonatal asphyxia, ROC, sensitivity, specificity
Introduction Intrapartum monitoring of the fetus during labor is crucial to obtain a good neonatal outcome. Cardiotocography (CTG) tracing of the fetal heart rate is used to recognize and intervene in cases of suspected fetal hypoxia in order to prevent adverse perinatal outcome. However, the sensitivity and specificity of CTG patterns are low, and the intraobserver variation of the interpretation is high. While a normal CTG is reassuring, only few of the cases with nonreassuring CTG traces reflects fetuses with hypoxia leading to a high risk of unnecessary interventions. When CTG is non-reassuring, fetal scalp blood sampling and measurements of the pH may assist in discriminating the non-hypoxic fetus from the fetus with asphyxia. Analyzing pH requires 30–50 ml of fetal blood while lactate measurements only require 5 ml resulting in much lower risks of sampling failures [1]. While high lactate values reflect metabolic acidosis, the pH measurement does not discriminate between respiratory and metabolic acidosis. In a previous RCT [2], lactate and pH measurements had the same sensitivity in predicting fetal asphyxia, but the failure rate of pH Address for correspondence: Christina Rørbye, Department of Obstetrics and Gynecology, University of Copenhagen, Hvidovre Hospital, Kettegaard Alle´ 30, DK-2650 Hvidovre, Denmark. Tel: +45 3862 5520. Fax: +45 3862 2135. E-mail: [email protected]
History Received 9 January 2015 Revised 14 April 2015 Accepted 24 April 2015 Published online 25 May 2015
measurements was 11%, while the failure rate of lactate measurements was negligible. In previous studies, lactate has been analyzed by the single-strip method Lactate Pro (Arkray, Kyoto, Japan). A new single-strip method (Scout Lactate, EKF Diagnostics, Germany) requires only 2–5 ml of fetal blood and is based on the same enzymatic amperometric detection method. The aim of this study was threefold: First to test the agreement between scalp blood lactate measured on the Scout Lactate System and on the Radiometer ABL 700 blood–gas analyzer in a clinical setting. Next, to compare lactate and pH measured in the same blood samples during delivery and analyze the sensitivity and specificity of lactate measurements in predicting scalp pH. Finally, to evaluate if lactate is a reliable alternative to pH measurements in intrapartum monitoring of the fetus when CTG is non-reassuring.
Methods The study was performed at Copenhagen University Hospital, Hvidovre during the period of 1 March 2012 and 30 August 2013. We included women in labor at term with a singleton in the cephalic presentation with a non-reassuring CTG either during the first or the second stage of labor. Fetal scalp blood samples were obtained, pH was measured and further management of the labor was decided by the obstetrician on call. On the same blood sample trained auxiliary nurses
2
C. Rørbye et al.
J Matern Fetal Neonatal Med, Early Online: 1–5
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
measured lactate using the Scout Lactate system (EKF Diagnostics, Germany), and the result was recorded outside the labor room – blinded to the obstetrician. The capillary blood pH was measured in a Radiometer BPH5 blood gas analyzer (BPH5) while pH, pCO2, SBE and lactate were measured on a Radiometer ABL700 blood gas analyzer (ABL). The ABL and BPH5 use glass-pH electrodes for pH measurement, and the ABL is equipped with an enzymatic amperometric biosensor for lactate measurement. The sensors are calibrated daily. Intervention or proceeding of labor was based on the pH value only, as the obstetrician was not informed about the lactate value. The umbilical artery blood was analyzed on the ABL immediately after delivery. Metabolic acidosis was defined as pH 57.0 and SBE 510. Statistics The relationship between lactate measured by Scout Lactate and Radiometer ABL was calculated by linear regression analysis. Sensitivity and specificity of lactate in predicting scalp pH were calculated for lactate values from 1 to 10 mml/l and presented in a ROC diagram. Cut-off values of 4.2 and 4.8 mmol/l were chosen. Comparison of scalp lactate and umbilical cord SBE was done when a scalp blood sample was performed within 15 min prior to delivery. Furthermore, it was calculated how many interventions would have been made, had lactate been the gold standard. Ethical considerations The Lactate Scout device and strips were sponsored by EKF Diagnostics, Germany. The company had no impact on the protocol, the results of the study, nor the way they are presented. The investigators have no economical connection to the company. The study is an evaluation of two different laboratory equipments, and according to the ethical committee, no further approval was needed (H-2-2014-FSH78). The analysis was performed on blood samples accepted to be evaluated for fetal acid–base changes as a routine clinical procedure. Measuring lactate did not require any extra intervention. The results had no impact on management of labor, and the procedure implied no risks to neither the women nor the fetuses. The study was approved by the Danish Data Registry (J nr 2012-41-1415).
Results In 264 fetal scalp blood samples lactate was measured both on the Radiometer ABL700 and the Scout Lactate system, and the results were closely interrelated with a coefficient of correlation of r2 ¼ 0.85 (Figure 1). Both pH and lactate were measured in 1009 fetal blood samples during 677 deliveries, see Table 1 for description of the study population. Fetal scalp pH was inversely related to lactate with r2 ¼ 0.73 (p50.0001) as shown in Figure 2. Using pH as the gold standard and pH 57.20 as the limit for acidemia, the sensitivity and specificity of a lactate value 4.8 were 0.63 (45/71) and 0.85 (795/938), respectively (Figure 3). The corresponding positive and negative predictive values were 0.24 (45/188) and 0.97 (795/821). For pH 57.15 the sensitivity and specificity were 0.86 (19/22) and
Figure 1. The correlation between lactate analyzed on the Scout Lactate system and the Radiometer ABL (n ¼ 264). Table 1. Study population. Mean (range) or number (%) Maternal age, years Gestational age, days BMI Nulliparous Multiparous Induction of labor Acute cesarean section Ventouse delivery Umbilical artery pH 57.0 and SBE 412 mmol/l Umbilical artery pH 57.0 and SBE 12 mmol/l
31 284 24 539 138 283 120 199 7 1
(17–42) (239–297) (17–47) (80%) (20%) (42%) (18%) (29%) (1.05%) (0.15%)
0.82 (818/987) (Figure 4). The sensitivity and specificity of a normal lactate 54.2 to predict a pH 7.20 were 0.98 (655/ 665) and 0.82 (283/344). Of the 677 monitored deliveries a scalp pH57.20 was seen in 56 and delivery was performed either by caesarean section (n ¼ 12) or vacuum extraction (n ¼ 44). Seventeen of these women had a scalp lactate value 4.8 mmol/l. One hundred and fifty-five of the women had at least one scalp blood measurement with a lactate 4.8 mmol/l. If intervention had been based on the lactate value, an additional 99 instrumental deliveries would have been performed in these 677 women. In cases with measurement of a scalp lactate within 15 min prior to delivery (n ¼ 104) the lactate value was compared to the arterial umbilical cord standard base excess (SBE) obtained immediately after delivery (Figure 5). A lactate 4.8 mmol/l was found in 50 scalp blood samples with SBE 412 in cord blood samples, thus the false positive rate was 48%.
Discussion In a clinical setting, we found that lactate measured on the Scout Lactate System correlates well with lactate measured
DOI: 10.3109/14767058.2015.1045863
Lactate versus pH in scalp blood
3
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
Figure 2. Lactate versus pH measured in the same fetal scalp blood sample (n ¼ 1009).
Figure 3. ROC for pH 57.20.
on the Radiometer ABL. Assuming pH as the gold standard, the sensitivity and specificity of a high lactate value to predict a low pH value are low. Monitoring non-reassuring deliveries with scalp lactate instead of pH would have resulted in more (155 instead of 56) instrumental deliveries without a decrease in the number of newborns with severe metabolic acidosis (pH 57.0 and SBE 512). Both pH and lactate measuring in fetal scalp blood during labor are valuable tools to discriminate the non-hypoxic fetus from the fetus with acidemia. Whether pH or lactate is superior is still unanswered as they reveal different aspects of the acid–base status of the fetus, and different fetuses might have different acid–base levels of clinical relevance. Several scalp blood samples from the same woman were accepted as individual measurements, as they are believed to be independent. The optimal cut-off level of lactate is not well-established and varies from 4.2 to 6.6 mmol/l [3,7]. It has been shown that lactate increase with increasing gestational age [4] and different cut-offs might provide more precise knowledge of when to intervene. In this study, we used a cutoff of lactate 4.8 and 54.2 mmol/l as these are most often
Figure 4. ROC for pH 57.15.
used. According to the ROC diagram calculated in our study these cut-offs have an acceptable accuracy. During the first stage of fetal distress the fetal–maternal exchange is slightly compromised. The oxygen tension decreases and the carbon dioxide tension increases. The pH in fetal blood decreases, and the hemoglobin dissociation curve is shifted to the right (the Bohr Effect). As long as the oxygen tension in the placenta only decreases slightly, the change in hemoglobin dissociation will facilitate the release of oxygen from the fetal blood in the tissue (oxygen unloading capacity), and aerobic metabolism is retained. This respiratory acidosis is well-tolerated by the fetus identified by a low pH and normal SBE. In the second stage of fetal distress the oxygen tension is further reduced and the oxygen-uptake of the hemoglobin in the placenta decreases as a consequence of the shifted hemoglobin dissociation curve. In this situation there is no benefit of the increased oxygen unloading capacity and the oxygen tension in the fetal tissue is insufficient for aerobic metabolism [5]. During oxygen deficiency the anaerobic metabolism of one mole of glucose results in production of two mole of lactic acid. The metabolic acidosis
4
C. Rørbye et al.
J Matern Fetal Neonatal Med, Early Online: 1–5
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
Figure 5. Fetal scalp lactate (515 min prior to delivery) versus umbilical cord artery SBE.
is an indicator of severe hypoxia and can be identified by a low pH, a low SBE and a high lactic acid concentration in the blood. The fetal well-being during labor is monitored using the acid status of fetal scalp capillary blood. Theoretically, the first warning should be a decrease of the pH during the first stage of distress, while the second stage of fetal distress is characterized by decreasing SBE and increasing lactic acid concentrations as a consequence of an anaerobic metabolism. In the clinical situation of a delivery, the demand for a method with a high sensitivity and specificity to detect fetal acidemia is crucial. The ideal monitoring tool would be able to discriminate between a respiratory and an early metabolic acidosis in time to prevent severe metabolic acidosis. A low sensitivity is unacceptable, as an acidotic fetus may never be missed; while a low specificity would result in unnecessary instrumental deliveries. The use of fetal scalp blood pH as the gold standard in monitoring deliveries with a non-reassuring CTG pattern reduces the use of unnecessary instrumental interventions. However, a low pH cannot discriminate a respiratory and a metabolic acidosis, but as the metabolic acidosis is preceded by a respiratory acidosis, and in order to prevent of metabolic acidosis, intervention could preferably be planned and performed at the time of a respiratory acidosis. In our labor ward we monitor high risk deliveries with CTG, and in 25% we found indication for pH measuring in a fetal scalp blood sample. With this set-up we performed 319 instrumental deliveries in the 667 deliveries with nonreassuring CTG, and only one child (0.15%) was born with severe metabolic acidosis (pH 57.0 and SBE 512). If the 677 deliveries had been monitored with fetal scalp lactate using a level of intervention of 4.8 mmol/l, we would have made an additional 99 interventions, while no cases of acidosis would have been missed. A discrepancy between normal, pre-acidotic and acidotic levels of lactate and pH in the same blood sample has been reported in 55% of scalp blood samples in the study by Liljestro¨m et al. [6]. In this study, the frequency of operative delivery due to fetal distress was 82% of cases with both an abnormal scalp pH and lactate measurement compared to 53% when only one of the parameters was abnormal value.
None of the infants in this study were born with metabolic acidosis. This implies that a combined test could have advantages in fetal surveillance, but this needs further testing in clinical trials. A significant although poor correlation between both fetal scalp lactate and fetal scalp pH versus umbilical cord pH (r ¼ 0.21 and r ¼ 0.40, respectively) was shown by Holzmann et al. [7]. However, among 85 fetal scalp lactate measurements 46.6 mmol/l, only 14 had cord pH levels 57.10. Heinis et al. advocate that scalp lactate provides more accurate information on fetal acid–base status at birth than does pH and/or SBE [8]. Nonetheless, 29% of acidotic children at birth (pH 57.05 and SBE 512) were not identified using a lactate limit of 6.6 mmol/l in the fetal scalp blood sample. Measuring lactate in a scalp blood sample is an easier procedure, as it can be done by a hand hold device using only 5 ml of blood, compared to pH measuring that requires 35–100 ml of blood and an ABL unit located close to the delivery room. These challenges result in failure rates of 11–20% in intended pH measurements in some delivery wards [1,2,9,10]. Furthermore, the time from deciding to do a fetal scalp blood sample to the obstetrician receives the result is shorter when analyzing lactate than pH [11]. In a multicenter RCT comparing the outcome of deliveries monitored by fetal scalp pH (47.25 and57.21) and lactate (54.2 and44.8 mmol) no differences were found in the frequency of cord acidemia [2]. The conclusion was based on an intention to treat analysis but included failed analysis in 10% of the pH group and 1% of the lactate group. At Hvidovre Hospital, Copenhagen, the pH measurement in fetal scalp blood has a failure rate of less than 1%. This may be explained by a high routine of the staff since the department has more than 7000 deliveries annually and scalp blood measurements are performed in all cases of a non-reassuring CTG. A high lactate value might be a better predictor of neurologic disability [12], but whether pH or lactate is superior in monitoring high risk deliveries is still not answered, as they might mirror different clinical aspects of asphyxia. This study however shows, that in our clinical setting, we would have performed more instrumental deliveries with no decrease in newborns with metabolic
DOI: 10.3109/14767058.2015.1045863
acidosis, if lactate had replaced pH in monitoring the fetus during labor. In conclusion, lactate measured on the Scout Lactate system and the ABL correlates well. The sensitivity, specificity, PPV and NPV of a high lactate to predict a low pH are low, thus in our clinical setting, more instrumental deliveries would be performed using lactate instead of pH in monitoring fetuses during labor.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Nyu Medical Center on 07/04/15 For personal use only.
Declaration of interest No declaration of interest. The authors report no conflicts of interest in connection with the article. The authors alone are responsible for the content and writing of the paper. The Lactate Scout device and strips were sponsored by EKF Diagnostics, Germany. No other funding was obtained.
References 1. Westgren M, Kruger K, Ek S, et al. Lactate compared with pH analysis at fetal scalp blood sampling: a prospective randomized study. Br J Obstet Gynecol 1998;105:29–33. 2. Wiberg-Itzel E, Lipponer C, Norman M, et al. Determination of pH or lactate in fetal scalp blood in management of intrapartum fetal distress: randomised controlled multicentre trial. BMJ 2008;336: 1284–7.
Lactate versus pH in scalp blood
5
3. East CE, Leader LR, Sheehan P, et al. Intrapartum fetal scalp lactate for fetal assessment in the presence of a non-reasurring fetal heart rate trace. Cochrane Database Syst Rev 2010;3:CD006174. 4. Wiberg N, Ka¨llen K, Herbst A, et al. Lactate concentration in umbilical cord blood is gestational age-dependant: a populationbased study of 17 867 newborns. BJOG 2008;115:704–9. 5. Wimberley PD. Fetal hemoglobin: 2,3-diphosphoglycerate and oxygen transport in the newborn premature infant (thesis). Scand J Clin Lab Invest Suppl 1982;160:1–149. 6. Liljestro¨m L, Wikstro¨m AK, Hanson U, et al. Evaluation of the discrepancy between pH and lactate in combined fetal scalp blood sampling. Acta Obstet Gynecol Scand 2011;90:1088–93. 7. Holzmann M, Cnattingius S, Nordstro¨m L. Outcome of severe intrapartum acidemia diagnosed with fetal scalp blood sampling. J Perinat Med 2011;39:545–8. 8. Heinis A, Spaanderman M, Gunnewiek J, Lotgering F. Scalp blood lactate for intra-partum assessment of fetal metabolic acidosis. Acta Obstet Gynecol Scand 2011;90:1107–14. 9. Nordstro¨m L. Fetal scalp blood measurements during labour – lactate or pH? Clin Biochem 2011;44:456–7. 10. Ramanah R, Martins A, Clement MC, et al. Fetal scalp lactate microsampling for non-reassuring fetal status during labor: a prospective observational study. Retal Diagn Ther 2010; 27:14–19. 11. Nordstro¨m L. Fetal scalp and cord blood lactate. Best Pract Res Clin Obstet Gynecol 2004;18:467–76. 12. Kruger K, Hallberg B, Blennow M, et al. Predictive value of fetal scalp blood lactate concentration and pH as markers of neurologic disability. Am J Obstet Gynecol 1999;181:1074–8.
