TITLE: Indicators for Fetal Hypoxia: A Review of the Clinical Evidence and Guidelines

DATE: 13 October 2010

CONTEXT AND POLICY ISSUES:

Asphyxia (hypoxia and metabolic acidosis) is a major cause of perinatal morbidity and can be responsible for neonatal mortality and cerebral palsy.1,2 The general condition of newborns is graded by the Apgar score. The score, devised by Dr. Virginia Apgar, is based on appearance, pulse, grimace (reflex irritability), activity, and respiration. Each indicator is given a score from zero (absence), one, or two and added together. Scores totaling less than four are considered critically low, while those above seven are considered normal.3 A low Apgar score has become a proxy for asphyxia.4

While Apgar score can evaluate the health of a newborn baby, cardiotocography can be used for intrapartum fetal monitoring.5 In this situation, an abnormal fetal heart rate pattern may signify asphyxia.6 However, while cardiotocography provides a sensitive measure, it has low specificity so other indicators of fetal hypoxia and acidosis need to be used to improve predictive power.2,6 Candidate indicators include fetal scalp lactate or blood pH, umbilical cord blood pH or base excess, umbilical cord blood gas, and nucleated red blood cell counts.7 These measures have varying rates of success, can be invasive, or require particular expertise and there are differing ideas on which might be the most appropriate choice.8-10

This report examines the evidence-based indicators of fetal distress and their correlation with Apgar scores, and reviews the evidence-based guidelines regarding the use of these indicators in peri- and intra-partum periods of care.

RESEARCH QUESTIONS:

1. What are the evidence-based indicators signaling fetal distress during the peri-partum and intra-partum periods of care?

2. What is the evidence that fetal hypoxia indicators are correlated with Apgar scores?

Disclaimer: The Health Technology Inquiry Service (HTIS) is an information service for those involved in planning and providing health care in Canada. HTIS responses are based on a limited literature search and are not comprehensive, systematic reviews. The intent is to provide a list of sources and a summary of the best evidence on the topic that CADTH could identify using all reasonable efforts within the time allowed. HTIS responses should be considered along with other types of information and health care considerations. The information included in this response is not intended to replace professional medical advice, nor should it be construed as a recommendation for or against the use of a particular health technology. Readers are also cautioned that a lack of good quality evidence does not necessarily mean a lack of effectiveness particularly in the case of new and emerging health technologies, for which little information can be found, but which may in future prove to be effective. While CADTH has taken care in the preparation of the report to ensure that its contents are accurate, complete and up to date, CADTH does not make any guarantee to that effect. CADTH is not liable for any loss or damages resulting from use of the information in the report.

Copyright: This report contains CADTH copyright material. It may be copied and used for non-commercial purposes, provided that attribution is given to CADTH.

Links: This report may contain links to other information available on the websites of third parties on the Internet. CADTH does not have control over the content of such sites. Use of third party sites is governed by the owners’ own terms and conditions.

3. What are the evidence-based guidelines regarding the use of indicators signaling fetal distress during the peri-partum and intra-partum periods of care?

METHODS:

A limited literature search was conducted on key health technology assessment resources, including PubMed, EBSCOHost CINAHL, the Cochrane Library (Issue 9, 2010), University of York Centre for Reviews and Dissemination (CRD) databases, ECRI (Health Devices Gold), EuroScan, international health technology agencies, and a focused Internet search. The search was limited to English language articles published between January 1, 2005 and September 14, 2010. Filters were applied to limit the retrieval to health technology assessments, systematic reviews, meta-analyses, randomized controlled trials and guidelines. A non-randomized studies filter was applied to a focused search (limited to fetal hypoxia) for targeted non-randomized studies.

SUMMARY OF FINDINGS:

Two systematic reviews,5,9 one randomized controlled trial (RCT),8 eight non-randomized trials,1,2,4,6,11-14 and four evidence-based guidelines15-18 were identified for inclusion. One additional relevant RCT was found,19 but was included in the systematic reviews and is not summarized in this report. No relevant health technology assessments were found.

Systematic reviews and meta-analyses

In 2010, East et al.9 published a systematic review to evaluate the effectiveness of fetal scalp lactate sampling, compared with alternative or no testing, in assessing fetal well-being during labour. Studies were eligible for inclusion if they were randomized or quasi-randomized clinical trials that compared fetal scalp lactate testing with no testing or alternative tests such as pH or fetal pulse oximetry to evaluate fetal status in the presence of a non-reassuring fetal heart rate trace during labour. Participants were limited to women in labour with non-reassuring fetal cardiotocography who would qualify for fetal scalp blood testing by standard delivery procedures. Primary outcomes examined were neonatal seizures, encephalopathy, death, or long-term disability. Secondary outcomes included Apgar scores, umbilical arterial pH, and other indicators of fetal well-being. There were no language restrictions. Two studies met the criteria for inclusion, covering 3,348 mother-baby pairs. Studies were assessed for quality based on allocation concealment, blinding, description of withdrawals and dropouts, and other possible sources of bias.

The authors found no statistically significant difference in neonatal encephalopathy [risk ratio (RR) 1.00, 95% confidence interval (CI) 0.32 to 3.09] or death (RR 0.14, 95% CI 0.01 to 2.76) between treatment groups. No neonatal seizures were reported in the included studies. Neither of the reviewed studies included data on other morbidities. There were no statistically significant differences in secondary outcome measures between treatment groups. This includes low Apgar scores at five minutes (RR 1.13, 95% CI 0.76 to 1.68), admission to neonatal intensive care units (ICU) (RR 1.02, 95% CI 0.83 to 1.25), and low umbilical pH (pH < 7.0, RR 0.84, 95% CI 0.47 to 1.50). Similar proportions of fetuses had to undergo additional tests to further evaluate well-being during labour (RR 0.22, 95% CI 0.04 to 1.30). Success rates for sampling,

Fetal Hypoxia Indicators 2

reported in one of the included studies, were 98.7% for lactate testing (309 successes in 313 attempts) and 79.4% (255 successes in 321 attempts) for pH testing.

The authors of this review concluded that while fetal scalp lactate sampling was more likely to be successful and results more rapidly available compared to pH testing, the increased speed and success did not translate into differences in clinical management or neonatal outcomes. The review was constructed to determine the safety and effectiveness of fetal scalp lactate sampling compared to other testing, but not the diagnostic accuracy of fetal lactate assays. Despite comparison of accuracy not being the focus of the review, the authors argue that the lack of difference in clinical outcomes (i.e. rates of other hypoxia markers) between lactate sampling and pH measurement offers limited support that these measures are similar in ability to identify the at-risk fetus. This review is also limited by the small number of included studies (two), which failed to cover all the desired outcomes and were underpowered for assessing differences in low prevalence outcomes.

In a 2008 systematic review, Alfirevic et al.5 evaluated the effectiveness of continuous cardiotocography during labour. Studies were eligible for inclusion if they were randomized or quasi-randomized clinical trials that compared continuous cardiotocography during labour, with and without fetal blood sampling, to no monitoring, intermittent auscultation (with a Pinard stethoscope or handheld ultrasound device), or intermittent cardiotocography. Participants were pregnant women in labour and their babies. Main outcomes of interest included death, neonatal seizures, Apgar scores, and cord blood acidosis. There were no language restrictions for study inclusion. Twelve studies met inclusion criteria, covering 37,615 women. Studies were assessed for quality based on allocation concealment and handling of study attrition.

The authors found that compared to intermittent auscultation, women subject to continuous cardiotocography were more likely to undergo caesarian section for abnormal fetal heart rate or acidosis (RR 2.37, 95% CI 1.88 to 3.00, n = 33,379, 11 trials). Two included trials reported on the use of fetal blood monitoring and found a statistically significant increase in the number of sampling tests performed with continuous cardiotocography (RR 1.24, 95% CI 1.03 to 1.49, n = 13,929) compared with fetal blood monitoring. No significant difference in the risk of cord blood acidosis was found between continuous cardiotocography and auscultation (RR 0.92, 95% CI 0.27 to 3.11) or intermittent cardiotocography (RR 1.43, 95% CI 0.95 to 2.14). There was no statistically significant difference in perinatal mortality between continuous cardiotocography and intermittent auscultation (RR 0.85, 95% CI 0.59 to 1.23, n = 33,513, 11 trials) but continuous cardiotocography was associated with a reduced risk of neonatal seizure (RR 0.50, 95% CI 0.31 to 0.80, n = 32,386, nine trials). There was no significant difference between continuous cardiotocography and intermittent auscultation in cases of Apgar score less than four (RR 1.43, 95% CI 0.61 to 3.34) or less than seven (RR 0.97, 95% CI 0.72 to 1.31) at five minutes. Similarly, the number of cases of Apgar score less than seven was not statistically different between continuous and intermittent cardiotocography (RR 2.65, 95% CI 0.70 to 9.97) at five minutes (no data was given for scores less than four). One included trial compared continuous to intermittent cardiotocography and found no statistically significant differences in any of the included outcome measures including fetal heart rate, umbilical cord arterial pH values, and Apgar scores less than seven at one and five minutes.

The authors concluded that continuous cardiotocography during labour was associated with a reduction in neonatal seizures, but showed no significant differences in other measures of

Fetal Hypoxia Indicators 3

neonatal well-being. Continuous cardiotocography was associated with an increase in caesarean sections. The review found no evidence that continuous cardiotocography has an impact on Apgar score, however the authors noted that this conclusion is limited by the small number of babies with clinically significant low Apgar scores in the included studies. This review is limited by the fact that few clinically relevant neonatal outcomes were reported consistently across the included trials. Only two of the included trials were considered high quality in terms of selection and attrition bias. Finally, the age of the included studies is a limiting factor. Publication dates ranged from 1976 to 1994 and any improvements in equipment, training, or interpretation will not be accounted for.

Randomized controlled trials

In 2010, Westerhuis et al.8 compared fetal monitoring by cardiotocography with ST waveform analysis to cardiotocography alone in an RCT. Adult (18 years or older) women in labour with a high-risk pregnancy, single fetus with cephalic presentation, gestational age greater than 36 weeks, and an indication for internal electronic fetal monitoring were eligible for this study. Of 5,681 women included in the study, 2,832 were randomly assigned to monitoring by cardiotocography with ST analysis and 2,849 were assigned to be monitored by cardiotocography alone. The primary outcome measure was the incidence of metabolic acidosis, defined as umbilical cord arterial blood pH lower than 7.05 and a base deficit calculated in the extracellular fluid compartment above 12 mmol/L. Secondary outcomes included number of Apgar scores below four at one minute, number of Apgar scores below seven at five minutes, neonatal ICU admissions, and number of cases with fetal blood sampling. Two neonatologists blinded to patient allocation reviewed admissions letters and charts to determine whether moderate or severe neonatal hypoxic-ischemic encephalopathy had developed.

Fetal blood sampling occurred less frequently in the ST analysis group (10.6% versus 20.4%, RR 0.52, 95% CI 0.46 to 0.59). The incidence of metabolic acidosis was also lower in this group (0.7% versus 1.1%, RR 0.70, 95% CI 0.38 to 1.28, number needed to treat 252), but this result was not statistically significant. The incidence of metabolic acidosis based on blood pH below 7.05 and base deficit calculated in the umbilical cord arterial blood above 12 mmol/L, instead of extracellular fluid compartment, was also determined because this information is readily available from most blood analyzers and is often used in the clinic. Using this definition, the incidence of metabolic acidosis was lower in the ST analysis group (1.6% versus 2.6%, RR 0.63, 96% CI 0.42 to 0.94, number needed to treat 100) compared with the group who underwent cardiotocography alone. The rates of low Apgar scores at one minute (RR 1.25, 95% CI 0.82 to 1.90) and five minutes (RR 1.24, 95% CI 0.79 to 1.95) were not statistically different between the two treatment groups, nor was the rate of neonatal ICU admission (RR 0.89, 95% CI 0.58 to 1.35). There were three cases of moderate or severe hypoxic-ischemic encephalopathy and three cases of perinatal death in the ST analysis group, and one case of moderate hypoxic-ischemic encephalopathy and two perinatal deaths in the control group. Relative risk was not calculated for these outcomes due to the low numbers.

The authors concluded that cardiotocography with ST analysis decreases the incidence of metabolic acidosis depending on the definition but does not affect Apgar scores, neonatal ICU admissions, or other fetal-safety outcomes compared to cardiotocography alone. They also noted a substantial decrease in the rate of fetal blood sampling associated with the ST group,

Fetal Hypoxia Indicators 4

which is seen a positive outcome given the invasiveness of the procedure. The trial was powered based on the assumption that 3.5% of neonates would experience metabolic acidosis, however the actual observed rates were lower (0.7% and 1.1% for ST analysis and cardiotocography alone respectively). This trial was performed in the Netherlands, where management of obstetric care may differ from other countries and prevent generalizability to a Canadian context.

Non-randomized studies

In a 2010 observational study, Kovalak et al.12 evaluated the association between umbilical cord nucleated red blood cell (NRBC) counts and non-reassuring fetal heart rate patterns during labor. The study group included 41 consecutive patients presenting with non-reassuring fetal heart rate patterns. Controls consisted of 45 pregnant women with no signs of non-reassuring fetal heart rate. Inclusion criteria were single fetus pregnancies between 37 and 42 weeks gestation, birth weight greater than 2,500 grams, and no evidence of congenital or chromosomal abnormalities. All participants were subject to external intrapartum fetal heart rate monitoring. Apgar scores at one and five minutes, and neonatal ICU admissions were recorded. At delivery, umbilical cord blood samples were taken for pH and blood gas analysis, and a venous blood smear was done to determine NRBC counts, expressed as NRBCs per 100 white blood cells.

Apgar scores were lower in the study group compared to controls at one minute (7.93 ± 1.08 versus 8.80 ± 0.59, P < 0.0001) but the difference at five minutes (9.83 ± 0.44 versus 9.98 ± 0.15, P = 0.05) was not statistically significant. There was no difference in the rate of neonatal ICU admissions between the two groups (4.9% versus 0%, P = 0.16). Umbilical arterial blood pH was lower in the study group (7.30 ± 0.09 versus 7.35 ± 0.07, P = 0.015) as was partial pressure of oxygen (pO2) (18.44 ± 5.59 versus 21.84 ± 7.10, P = 0.016). Neonates in the study group had higher NRBC numbers than the control group (13.54 ± 10.27 versus 8.04 ± 7.48, P < 2 0.002). Regression analysis indicated umbilical artery pO2 as a predictor of NRBC count (R = 0.1, P = 0.002).

Overall, this study found that in patients with non-reassuring fetal heart rate patterns during labour, the number of NRBCs in the cord blood of newborns was found to be elevated. However, the authors stated that the clinical utility of NRCB count as a marker for fetal hypoxia is limited by the wide range and low sensitivity compared to umbilical arterial blood pH. The small number of subjects may limit this study’s ability to identify correlations between NRBC counts and events that occurred infrequently in the study population (e.g. neonatal ICU admission).

In 2010, White et al.1 published the results of an observational study evaluating the impact on perinatal outcome of introducing universal umbilical cord blood gas analysis at delivery. All neonates of at least 20 weeks gestation delivered between January 1 2003 and December 31 2006 at King Edward Memorial Hospital in Perth, Australia were included in the study. Therapeutic abortions for fetal anomaly and fetal death in utero diagnosed prior to delivery were excluded. Paired umbilical arterial and venous blood samples were taken at delivery for lactate and blood gas analysis. Neonatal outcomes included Apgar scores, hypoxic-ischemic encephalopathy, and neonatal death. Overall, there were 12,345 validated paired cord gas samples.

Fetal Hypoxia Indicators 5

Over the course of the study, there were statistically significant increases in mean umbilical arterial and venous blood pH, pO2, and bicarbonate, and decreases in lactate and pCO2 (P < 0.0001). There were significant reductions in the number of newborns with umbilical arterial pH less than 7.1 [Adjusted Odds ratio (OR) 0.712, 95% CI 0.532 to 0.953] and with lactate levels greater than 6.1 mmol/L (OR 0.373, 95% CI 0.301 to 0.461). There was no change in the distribution of five-minute Apgar scores in the validated cohort, nor were there statistically significant changes in the rates of hypoxic-ischemic encephalopathy or neonatal death over the course of the study. Over the course of the study, rates of nursery admission (RR 0.746, 95% CI 0.638 to 0.871, P < 0.001) and special care nursery admission (RR 0.751, 95% CI 0.650 to 0.869, P < 0.001) were significantly different, but not neonatal ICU admissions (RR 1.174, 95% CI 0.969 to 1.422, P = 0.231).

This study suggests that the introduction of universal umbilical cord blood gas analysis has clinical benefits, as it is associated with significant improvements in biochemical markers of metabolic acidosis. One possible reason is that the availability of blood gas analysis led to timelier interventions when there was concern for fetal well-being. The authors argue that this extra biomedical information aids clinical assessment and leads to improved perinatal care. The study had limited power to detect rarer but more serious outcomes such as cerebral palsy, despite the large sample size. It is also limited by the inability to incorporate certain unmeasured factors, such as the degree of involvement of senior staff, into the statistical model that may be associated with the improved clinical outcomes.

In 2009, Ghosh et al.11 conducted a double-blind observational study using Doppler examination to identify umbilical vein pulsations and relate them to perinatal outcome. Women in active labour (three to ten centimeters dilation) with pregnancies over 37 weeks and a single fetus with cephalic presentation were eligible for inclusion. Doppler examinations of the umbilical vein were performed in 26 fetuses with normal cardiotocography and 26 fetuses with pathological cardiotocography. Pathological cardiotocography was defined as the presence of repeated late decelerations, repeated variable decelerations with a duration greater than 60 seconds, or a fetal heart rate over 180 bpm with pathological variability. Umbilical vein Doppler results were compared to perinatal outcomes which included operative deliveries for fetal distress (e.g. caesarian or vacuum extraction), pH in umbilical arterial and venous blood, Apgar scores, and admissions to the neonatal ICU.

No umbilical vein pulsations were observed in patients with normal cardiotocography. Pulsations were observed in eight patients with pathological cardiotocography. Of these, six had operative deliveries due to fetal distress. Among patients with pathological cardiotocography, there was an increased risk of operative delivery due to distress in fetuses with umbilical vein pulsations (six versus zero, P < 0.0001). There were no statistically significant differences in Apgar scores, blood pH, or neonatal ICU admissions among neonates with umbilical vein pulsations.

The authors concluded that pulsations in the umbilical vein can be observed in suspected cases of intrapartum hypoxia, and these pulsations are associated with an increased rate of operative delivery due to fetal distress. Doppler examination of the umbilical vein may contribute useful information on fetal condition during delivery. However, this study is limited by the small number of included cases.

Fetal Hypoxia Indicators 6

A 2009 observational study by Ramanah et al.2 examined the predictive value of fetal scalp lactate sampling in the management of non-reassuring fetal status during labour. Single fetus pregnancies with cephalic presentation and at least 37 weeks gestation were eligible for inclusion. During the study period, from January 2003 to December 2007, 7,617 blood gas measurements were performed immediately following delivery. Of these, 450 fetal scalp blood samples were taken for abnormal fetal heart rate at the discretion of the attending obstetrician. Scalp lactate measurements were compared to scalp pH. After delivery, umbilical arterial blood sampling was performed and fetal scalp lactate was compared to umbilical arterial blood pH, umbilical artery lactate, umbilical artery base deficit, and Apgar scores.

Due to sample failure, 387 of 450 fetal scalp blood samples were paired for lactate and pH. Scalp lactate was significantly associated with scalp pH (r = -0.56, P = 0.001), umbilical artery pH (r = -0.39, P = 0.03), umbilical artery lactate (r = 0.48, P = 0.01) and umbilical artery base deficit (r = 0.51, P = 0.01). There was a stronger correlation between these indicators and scalp lactate than with scalp pH. There was no correlation between scalp lactate and Apgar score at one, five, or ten minutes. Receiver Operating Characteristic (ROC) curves were drawn based on umbilical cord pH less than 7.10 and umbilical cord lactate greater than 8 mmol/L. These curves indicate a scalp lactate cut-off of over 5 mmol/L for maximum sensitivity and specificity for predicting pathological pH and 5.3 mmol/L for predicting pathological umbilical cord lactate.

This study indicates that fetal scalp lactate is significantly correlated to scalp pH and is more closely associated to umbilical cord parameters than scalp pH. Scalp lactate measurement may be a useful tool for monitoring fetal asphyxia during labour. No correlation between fetal scalp lactate and Apgar scores was observed. The low number of abnormal Apgar scores in the study population prevented comparison of scalp lactate distribution between normal and pathological scores. One limitation of the study is that all blood analysis was done with the same machine (Rapid Lab 860 analyzer) and not compared to other methods, so calculated cut-off values for predicting pathological state may vary from tests done by other methods.

In 2008, Borruto et al.13 performed an observational study to describe the detection methods and criteria for assessing asphyxia during labour to prevent cerebral palsy, which included comparison of hypoxia indicators. A cohort of 188 deliveries from 2003 to 2006 was included in this study. Inclusion and exclusion criteria were not described. Fetuses were evaluated by means of Apgar score, cardiotocography, and clinical features of distress at birth. Fetal scalp and umbilical cord lactate were measured.

Scalp lactate correlated significantly with umbilical artery lactate (R = 0.49, P = 0.01), but was not associated with Apgar score at one or five minutes. There was no perfect correlation observed between lactate levels and neonatal outcome. Neonates with abnormal baseline fetal heart rate had a higher incidence of low Apgar scores (≤ 7) than those with variable decelerations or transient tachycardia (P < 0.05). A similar association was observed for umbilical cord lactate, with higher levels recorded in neonates with abnormal baseline fetal heart rate (4.66 ± 0.12 mmol/L) and severe variable decelerations (3.95 ± 0.29 mmol/L) compared to those with mild variable decelerations or transient tachycardia (P < 0.01). At one year of age, four patients were lost to follow-up and four had an adverse outcome unrelated to asphyxia. The remaining 180 children showed normal development, but the possible sequelae of asphyxia were the deaths of two infants, slight abnormalities in ten infants, and clear abnormalities in four infants. The nature of these abnormalities was not described. For high lactate concentration, the

Fetal Hypoxia Indicators 7

sensitivity and positive predictive value for adverse outcomes were 14% and 7%. For low five- minute Apgar score, these values were 14% and 21%.

The authors concluded that fetal blood lactate is a useful measure for diagnosis of fetal distress, but is not associated with Apgar scores. Both lactate and Apgar scores were poor predictors of neurological outcomes. The small number of subjects that experienced intrapartum asphyxia is a limitation of this study. The majority of subjects (85.1%) experienced no complications, with the remaining 28 patients (14.9%) experiencing intrapartum asphyxia. This may limit the statistical power for evaluation of asphyxia and hypoxia indicators. No inclusion or exclusion criteria were described, though all subjects had cephalic presentation and fetal heart rate anomalies.

A 2008 study by Csitári et al.6 attempted to determine the reliability of fetal pulse oximetry in cases of abnormal fetal heart rate pattern and fetal oxygen saturation below 30%. A total of 301 single fetus term deliveries were monitored with cardiotocography and fetal pulse oximetry between January 2001 and December 2003. Pregnant women dilated at least two centimeters and gestational age of at least 37 weeks were eligible for fetal pulse oximetry. Participants were divided into two groups; the study group included patients with fetal pO2 values below 30% (52 patients), and the control group included patients with fetal pO2 values above 30% (249 patients). Base excess and pH were measured in umbilical cord arterial and newborn fingertip blood, and five-minute Apgar scores were recorded.

Compared to the controls, participants with fetal pO2 below 30% had significantly lower umbilical cord pH (P < 0.01). Five-minute Apgar scores, umbilical base excess, fingertip blood pH, and fingertip blood base excess were not significantly different between the two groups. Predictive values, sensitivity, and specificity of the correlation between fetal pO2 values and Apgar scores was calculated. The probability of a fetal oxygen saturation above 30% predicting a five-minute Apgar score greater than seven was 98.9%. Conversely, given a high five-minute Apgar score (> 7), the probability of a pO2 greater than 30% was 82.8%. A similar relationship was seen between fetal pO2 and umbilical cord artery pH with a negative predictive value of 93.9% and a sensitivity of 85.0%. During the study there were no cases of neonatal ICU admission or perinatal death reported.

This study suggests that fetal pulse oximetry is a reliable method of estimating fetal condition, and a cut-off fetal oxygen saturation of 30% is a safe limit that can be used to avoid hypoxia and acidosis in cases with abnormal fetal heart rate.

In 2007, Gea et al.14 conducted a study to evaluate lactate levels and NRBC counts in the placental segment of the umbilical vein for diagnosis of hypoxia-ischemia. Inclusion criteria were gestational age less than 37 weeks and weight less than 2,000 g. Congenital malformations, maternal diabetes, and blood group incompatibility were reasons for exclusion. Eligibility criteria were met by 25 premature newborns delivered between April 2004 and January 2005 in a Brazilian hospital. After delivery, blood was sampled from the umbilical cord vein for lactate measurement, blood gas analysis, and NRBC counts. NRBC counts were calculated as the number of nucleated red blood cells per 100 white blood cells.

There was a correlation between umbilical vein lactate and base excess (R2 = 0.72, P < 0.0001). No correlation was found between lactate and Apgar scores. Acidosis was defined as a

Fetal Hypoxia Indicators 8

base excess of -10 mmol/L or lower. For the identification of acidosis, the area under the ROC curve, which indicates test accuracy, was 0.842, which suggests that lactate is a good indicator of acidosis. The sensitivity of a lactate cutoff level of 4.04 mmol/L in relation to acidosis was 62.5%, the specificity was 94.1%, the positive predictive value was 83.3%, and the negative predictive value was 84.2%. NBRC counts were associated with pH (R2 = 0.38, P = 0.009) and base excess (R2 = 0.26, P = 0.009). NRBC count, with a cut-off value of 10 NRBC per 100 white blood cells, was inadequate for identification of acidosis, with an area under the ROC curve of 0.577 (P = 0.26), sensitivity of 37.5%, specificity of 82.4%, positive predictive value of 50%, and negative predictive value of 73.6%. NRBC counts were correlated with lactate levels (R2 = 0.4, P = 0.0008) and Apgar scores (P = 0.03 at one minute and P = 0.02 at five minutes).

Overall, this study found that lactate sampled from placental segment veins was an adequate marker for fetal hypoxia, but NRBC counts were weakly correlated with pH and base excess, and showed low sensitivity and positive predictive values for acidosis. The authors did not recommend the use of NRBC counts for routine clinical diagnosis of perinatal acidosis. A small number of participants may limit this study’s statistical power and generalizability to a broader clinical context.

In 2007, Hogan et al.4 performed a cohort- and case-control study to evaluate how often low five-minute Apgar scores are associated with asphyxia. Cases were 183 term newborns (37 or more completed gestational weeks), with five-minute Apgar scores below seven, born at Lund University Hospital from 1993 to 2002. Intrauterine fetal deaths before admission were excluded. Controls were 183 randomly selected term newborns with five-minute Apgar scores of nine to ten. Obstetric and pediatric files, and cardiotocography traces were reviewed. Main outcomes were interventions for fetal distress, assessment of monitoring traces, and acid-base status at birth.

After excluding infants with severe malformations, 30 infants from the case group had a five- minute Apgar score below four and 143 had an Apgar score of four to six. Abnormal cardiotocograhy, rates of intervention for fetal distress, umbilical cord artery pH less than 7.15, and rates of hypoxic-ischemic encephalopathy or hypoxic death were higher in the groups with Apgar scores of less than four and between four and six compared to controls (summarized in Table 1). All differences between each case group and the controls were statistically significant (P < 0.001), with the exception of arterial pH assay success.

Table 1: Summary of data from Hogan et al.4 Apgar < 4 (n = 30) Apgar 4-6 (n = 143) Control (n = 182) Abnormal cardiotocography 88% 69% 18% rate Intervention rate 83% 48% 9% Artery pH assay success 87% 91% 86% Cord artery pH < 7.15 69% 54% 7% Encephalopathy or death 70% 14% 0%

When infants with severe malformations were excluded, the authors concluded that the majority of Apgar scores less than four and roughly half of those between four and six can be attributed to birth asphyxia. They concluded that a five-minute Apgar score below four is a good proxy for asphyxia. One limitation to this study is the use of pH 7.15 as the cut-off for acidemia, even

Fetal Hypoxia Indicators 9

though, as the authors point out, many newborns with umbilical cord artery pH between 7.0 and 7.15 experience no complications, and lower pH values may be better indicators of asphyxia.20

Guidelines and recommendations

In 2009, the Institute for Clinical Systems Development15 published a health care guideline for the management of labour. The guideline stated that it was developed using a defined process for literature search and review, document development, and revision however, the process for developing the document was not explained. Recommendations were identified by evidence type (e.g. meta-analysis, randomized controlled trial, cohort study, etc.), but strength of recommendation based on the evidence was not indicated.

The guidelines state that tests to assess fetal acid-base status, including fetal scalp lactate sampling, may be useful if available. This conclusion is based on evidence from one meta- analysis and one case series or case report. Based on an RCT, the authors concluded that fetal oxygen saturation is not associated with a reduction in caesarian delivery rate or improvement of the condition of the newborn. The authors recommend cord pH or gases if the one-minute Apgar score is less than three or the five-minute Apgar score is less than six, based on cord pH being a better indicator than Apgar for fetal compromise. This recommendation is based on two case series or case reports.

There is no indication of the strength of each recommendation, though the evidence source for each is given. Each recommendation is based on one or two published studies, but there is no description of how these studies were chosen. Studies used as evidence to support the above recommendations were published between 1986 and 2006, with only one published since 2002 and three predating 1994. More current sources may be available to develop guidelines.

In 2007, the National Collaborating Centre for Women’s and Children’s Health16 published a clinical guideline for intrapartum care. Recommendations were based on evidence gathered from a systematic literature search and reference lists from other published guidelines. Grey literature and hand searches were not undertaken. Evidence for each recommendation was provided and graded from 1++ (High-quality meta-analyses or systematic reviews of randomized controlled trials) to 4 (expert consensus) for intervention studies and Ia (systematic review) to IV (expert consensus) for accuracy of diagnostic tests. Recommendations were not assigned number or letter grades.

Though there was no difference in fetal acid-base, there was high-level evidence (Grade 1+) that ST waveform analysis reduces instrumental vaginal birth and neonatal encephalopathy. In terms of monitoring markers of fetal hypoxia, there was no evidence of a correlation between fetal scalp lactate and better long term outcomes (Grade II). The use of fetal blood sampling with continuous electronic fetal monitoring may reduce the rate of instrumental delivery, but there is no evidence of differences in other outcomes such as neonatal seizures (Grade 2+). The guidelines indicate that lag events in ST analysis are associated with scalp pH and have diagnostic value (Grade II). When compared to fetal blood lactate analysis, pH analysis is more frequently unsuccessful, though there is no evidence of differences in Apgar scores at one or five minutes (Grade 1+). Some of these guidelines are based on lower levels of evidence (e.g. case control studies) and while the quality of the evidence is indicated, the amount of evidence and strength of recommendation was not provided.

Fetal Hypoxia Indicators 10

In 2007, the Society of Obstetricians and Gynecologists of Canada17 published a clinical practice guideline for fetal health surveillance. The objective of this publication was to provide recommendations for the application of fetal surveillance to decrease the incidence of birth asphyxia. Short- and long-term outcomes that potentially indicate birth asphyxia were considered. Recommendations were made based on evidence from a comprehensive review of RCTs published between January 1996 and March 2007. Level of evidence was assigned a grade from I (Evidence obtained from at least one properly randomized trial) to III (opinions of respected authorities, based on clinical experience, descriptive studies, or reports of expert committees). Each recommendation was assigned a letter grade based on strength of recommendation. A grade of A or B indicates good or fair evidence to recommend the action, C indicates conflicting evidence, and D or E indicates fair or good evidence to recommend against the action. Cases where there is insufficient quantity or quality of evidence to make a recommendation are rated I.

The guideline states that admission fetal heart tracing is not recommended for healthy women at term in labour in the absence of risk factors for adverse perinatal outcome (Evidence I, A) but are recommended if risk factors are present (Evidence III, B). In pregnancies at risk for adverse perinatal outcome, electronic fetal monitoring is recommended (Evidence II, A). For assessment of fetal acid–base status, fetal scalp blood sampling is recommended in women with atypical fetal heart tracings at gestations over 34 weeks (Evidence III, C). Cord blood sampling from both umbilical arterial and venous blood is recommended for all births. If only one sample is possible, it should be arterial (Evidence III, B). There is insufficient evidence for the sampling of arterial and venous cord gases when risk factors for adverse perinatal outcomes exist. Fetal pulse oximetry and intrapartum scalp lactate sampling are not recommended for routine use (Evidence III, C), nor is ST waveform analysis (Evidence I, A).

Many of these recommendations are based on expert opinion, which is considered a low grade of evidence, resulting in conflicting evidence which makes it difficult to recommend for or against a specific action. The exception is electronic fetal monitoring which is recommended based on good evidence from non-randomized trials.

In 2006, the Royal Australian and New Zealand College of Obstetricians and Gynecologists18 published guidelines for intrapartum fetal surveillance. This is an update of guidelines published in 2002. Guidelines were developed based on a literature search. Level of evidence was assigned a grade from I (evidence obtained from a systematic review of all relevant RCTs) to IV (evidence obtained from case series). Recommendations were graded based on the level of evidence, either A (adequate RCT evidence as part of a body of literature of good quality and consistency), B (well-conducted clinical studies on the topic of the recommendation), or C (evidence from expert committee reports or clinical experience from respected authorities).

The guidelines state that fetal surveillance in labour, whether intermittent auscultation or electronic fetal monitoring, should be offered to all women (Recommendation: C). Continuous electronic fetal monitoring is recommended when risk factors for fetal compromise are detected (Recommendation: B) or when the fetal heart rate is abnormal (Recommendation: A). No evidence-based recommendations were made regarding hypoxia indicators such as fetal scalp lactate or blood pH.

Fetal Hypoxia Indicators 11

Limitations

Two systematic reviews were identified, but both were limited either by the number or quality of the included studies.

Small study populations, or small fractions of included populations experiencing fetal distress, limited many of the included studies. Four of the eight non-randomized controlled trials had study populations of 52 or fewer patients and in two others the number of patients experiencing severe adverse events (asphyxia or Apgar score below four) was fewer than 30. This may prevent detection of rarer outcomes and limit their statistical power. The included studies covered a wide variety of potential markers of fetal distress, and as such, there is less evidence for any one particular indicator. With the exception of one publication that examined premature newborns, all studies considered only pregnancies that had gone to term. Non-randomized studies, which may be subject to bias, made up the majority of the included research. Furthermore, all studies only included single fetus pregnancies. This limits the ability to generalize results to twin pregnancies or premature deliveries.

Guideline recommendations were often based on limited evidence such as expert consensus, case-control studies, or a small number of higher grade publications.

CONCLUSIONS AND IMPLICATIONS FOR DECISION OR POLICY MAKING:

Several studies were identified that examined various indicators for fetal distress. Two studies examined blood gas analysis or pulse oximetry and found that pulse oximetry is a reliable indicator of fetal distress, while universal use of blood gas analysis was associated with improvements in markers for blood acidosis. Three studies looked at fetal blood lactate, either from the scalp or umbilical cord. These measures were found to correlate with one another and are good indicators for fetal hypoxia. A systematic review determined that scalp lactate sampling is as safe as other tests. One study examining umbilical vein pulsations found they can be detected in cases of intrapartum hypoxia and are associated with an increased rate of operative delivery due to fetal distress. Two studies examined nucleated red blood cells as a marker for fetal hypoxia and found that while NRBCs are associated with other markers for fetal hypoxia, low sensitivity makes them a poor marker for diagnosis of fetal hypoxia or acidosis.

Low Apgar scores were shown to be correlated with asphyxia in one study, but not all markers of fetal distress were found to be correlated with Apgar scores. Three studies demonstrated that fetal lactate does not correlate well with Apgar scores, while one showed an association between Apgar scores and fetal blood oxygen saturation.

Available evidence-based guidelines recommend electronic fetal monitoring for high risk pregnancies. To assess fetal acid-base status, umbilical cord pH or blood gases, or fetal scalp lactate are recommended in the available guidelines. Guidelines suggest that for umbilical cord blood sampling, both arterial and venous blood are recommended, but in the case of only one available sample, arterial is preferred. It was stated in the guidelines that tests for lactate have a higher success rate than pH sampling. Pulse oximetry, ST waveform analysis, and scalp lactate sampling are not recommended for routine use in the identified guidelines. The majority of the available evidence comes from non-randomized studies, many with small study populations,

Fetal Hypoxia Indicators 12

and there is very limited information on pre-term births or non-cephalic presentation. These are factors that should be considered for decision-making.

PREPARED BY: Health Technology Inquiry Service Email: [email protected] Tel: 1-866-898-8439

Fetal Hypoxia Indicators 13

REFERENCES:

1. White CR, Doherty DA, Henderson JJ, Kohan R, Newnham JP, Pennell CE. Benefits of introducing universal umbilical cord blood gas and lactate analysis into an obstetric unit. Aust N Z J Obstet Gynaecol. 2010 Aug;50(4):318-28. PubMed: PM20716258

2 . Ramanah R, Martin A, Clement MC, Maillet R, Riethmuller D. Fetal scalp lactate microsampling for non-reassuring fetal status during labor: a prospective observational study. Fetal Diagn Ther. 2010;27(1):14-9. PubMed: PM19940450

3 . Finster M. The Apgar score has survived the test of time. Anesthesiology. 2005 Feb 4;102(4):855-7. PubMed PMID: 15791116

4. Hogan L, Ingemarsson I, Thorngren-Jerneck K, Herbst A. How often is a low 5-min Apgar score in term newborns due to asphyxia? Eur J Obstet Gynecol Reprod Biol. 2007 Feb;130(2):169-75. PubMed: PM16621222

5. Alfirevic Z, Devane D, Gyte GM. Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour. Cochrane Database Syst Rev. 2006;3:CD006066. PubMed: PM16856111

6. Csitari IK, Pasztuhov A, Laszlo A. The reliability of fetal pulse oximetry: the effect of fetal oxygen saturation below 30% on perinatal outcome. Eur J Obstet Gynecol Reprod Biol. 2008 Feb;136(2):160-4. PubMed: PM17467877

7. Murray DM, Ryan CA, Boylan BG et al. Prediction of seizures in asphyxiated neonates: correlation with continuous video-electroencephalographic monitoring. Pediatrics. 2006 Jul;118(1):41-6. Available from: http://pediatrics.aappublications.org/cgi/content/full/118/1/41 PubMed: PM16818547.

8. Westerhuis ME, Visser GH, Moons KG, van Beek E, Benders MJ, Bijvoet SM, et al. Cardiotocography plus ST analysis of fetal electrocardiogram compared with cardiotocography only for intrapartum monitoring: a randomized controlled trial. Obstet Gynecol. 2010 Jun;115(6):1173-80. PubMed: PM20502287

9. East CE, Leader LR, Sheehan P, Henshall NE, Colditz PB. Intrapartum fetal scalp lactate sampling for fetal assessment in the presence of a non-reassuring fetal heart rate trace. Cochrane Database Syst Rev. 2010;3:CD006174. PubMed: PM20238343

10. Dear P. Establishing probable cause in cerebral palsy. How much certainty is enough? BMJ. 2000 Apr 15;320(7241):1075-6. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1117953/?tool=pubmed PubMed: PM10764378

11. Ghosh GS, Fu J, Olofsson P, Gudmundsson S. Pulsations in the umbilical vein during labor are associated with increased risk of operative delivery for fetal distress. Ultrasound Obstet Gynecol. 2009 Aug;34(2):177-81. PubMed: PM19588466

Fetal Hypoxia Indicators 14

12 . Kovalak EE, Dede FS, Gelisen O, Dede H, Haberal A. Nonreassuring fetal heart rate patterns and nucleated red blood cells in term neonates. Arch Gynecol Obstet. Epub 2010 May 25. PubMed: PM20499074

13. Borruto F, Comparetto C, Treisser A. Prevention of cerebral palsy during labour: role of foetal lactate. Arch Gynecol Obstet. 2008 Jul;278(1):17-22. PubMed: PM18071726

14. Gea Y, Araujo O, Silva LV. Clinical value of lactate measurement and nucleated red blood cell counts in the placental segment of the umbilical vein of premature newborns for diagnosis of hypoxia-ischemia. J Pediatr (Rio J) [Internet]. 2007 Mar [cited 2010 Sep 20];83(2):186-90. Available from: http://www.jped.com.br/conteudo/07-83-02-186/ing.pdf PubMed: PM17380228

15. Institute for Clinical Systems Improvement. Healthcare guideline: management of labor. 3rd ed. Bloomington (MN): Institute for Clinical Systems Improvement; [Internet]. 2009 May. [cited 2010 Sep 15]. Available from: http://www.icsi.org/labor/labor__management_of__full_version__2.html

16. National Collaborating Centre for Women's and Children's Health. Intrapartum care: care of healthy women and their babies during childbirth. Royal college of obstetricians and gynaecologists; [Internet] 2007. [cited 2010 Sep 15]. Available from: http://www.nice.org.uk/nicemedia/live/11837/36275/36275.pdf

17. Liston R, Sawchuck D, Young D, Society of obstetrics and gynaecologists of Canada, British Columbia perinatal health program. Fetal health surveillance: antepartum and intrapartum consensus guideline. J Obstet Gynaecol Can. 2007 Sep;29(9 Suppl 4):S3- S56. PubMed: PM17845745

18. The Royal Australian and New Zealand College of Obstetricians and Gynaecologists. Intrapartum fetal surveillance: clinical guidelines. 2nd ed.The Royal Australian and New Zealand College of Obstetricians and Gynaecologists [Internet]. 2006 May [cited 2010 Sep 15]. Available from: http://www.ranzcog.edu.au/publications/pdfs/ClinicalGuidelines- IFSSecEd.pdf

19. Wiberg-Itzel E, Lipponer C, Norman M, Herbst A, Prebensen D, Hansson A, et al. Determination of pH or lactate in fetal scalp blood in management of intrapartum fetal distress: randomised controlled multicentre trial. BMJ [Internet]. 2008 Jun 7 [cited 2010 Sep 20];336(7656):1284-7. Available from: http://www.bmj.com/content/336/7656/1284.full.pdf PubMed: PM18503103

20. Gilstrap LC, III. Diagnosis of birth asphyxia on the basis of fetal pH, Apgar score, and newborn cerebral dysfunction. Am J Obstet Gynecol. 1989 Sep;161(3):825-30. PubMed: PM2782367.

Fetal Hypoxia Indicators 15