Limitations in the clinical prediction of intrapartum fetal asphyxia

Predictive analytics for early detection of hospital-acquired complications: An artificial intelligence approach

BACKGROUND

Hospital-acquired complications (HACs) have an adverse impact on patient recovery by impeding their path to full recovery and increasing healthcare costs.

OBJECTIVE

The aim of this study was to create a HAC risk prediction machine learning (ML) framework using hospital administrative data collections within North Metropolitan Health Service (NMHS), Western Australia.

METHOD

A retrospective cohort study was performed among 64,315 patients between July 2020 to June 2022 to develop an automated ML framework by inputting HAC and the healthcare site to obtain site-specific predictive algorithms for patients admitted to the hospital in NMHS. Univariate analysis was used for initial feature screening for 270 variables. Of these, 77 variables had significant relationship with any HAC. After excluding non-contemporaneous data, 37 variables were included in developing the ML framework based on logistic regression (LR), decision tree (DT) and random forest (RF) models to predict occurrence of four specific HACs: delirium, aspiration pneumonia, pneumonia and urinary tract infection.

RESULTS

All models exhibited similar performance with area under the curve scores around 0.90 for both training and testing datasets. For sensitivity, DT and RF exceeded LR performance while on average, false positives were lowest for LR-based models. Patient's length of stay, Charlson Index, operation length and intensive care unit stay were common predictors.

CONCLUSION

Integrating ML-based risk detection systems into clinical workflows can potentially enhance patient safety and optimise resource allocation. LR-based models exhibited best performance.

IMPLICATIONS

We have successfully developed a "real-time" risk prediction model, where patient risk scores are calculated and reviewed daily.

The Association of Placental Abruption and Pediatric Neurological Outcome: A Systematic Review and Meta-Analysis

Placental histopathology provides insights, or "snapshots", into relevant antenatal factors that could elevate the risk of perinatal brain injury. We present a systematic review and meta-analysis comparing frequencies of adverse neurological outcomes in infants born to women with placental abruption versus without abruption. Records were sourced from MEDLINE, Embase, and the CENTRAL Trials Registry from 1946 to December 2019. Studies followed the PRISMA guidelines and compared frequencies of neurodevelopmental morbidities in infants born to pregnant women with placental abruption (exposure) versus women without placental abruption (comparator). The primary endpoint was cerebral palsy. Periventricular and intraventricular (both severe and any grades of IVH) and any histopathological neuronal damage were the secondary endpoints. Study methodologic quality was assessed by the Ottawa-Newcastle scale. Estimated odds ratios (OR) and hazards ratio (HR) were derived according to study design. Data were meta-analyzed using a random effects model expressed as pooled effect sizes and 95% confidence intervals. We included eight observational studies in the review, including 1245 infants born to women with placental abruption. Results of the random effects meta-analysis show that the odds of infants born to pregnant women with placental abruption who experience cerebral palsy is higher than in infants born to pregnant women without placental abruption (OR 5.71 95% CI (1.17, 27.91); I2 = 84.0%). There is no statistical difference in the odds of infants born to pregnant women with placental abruption who experience severe IVH (grade 3+) (OR 1.20 95% CI (0.46, 3.11); I2 = 35.8%) and any grade of IVH (OR 1.20 95% CI (0.62, 2.32); I2 = 32.3%) vs. women without placental abruption. There is no statistically significant difference in the odds of infants born to pregnant women with placental abruption who experience PVL vs. pregnant women without placental abruption (OR 6.51 95% CI (0.94, 45.16); I2 = 0.0%). Despite our meta-analysis suggesting increased odds of cerebral palsy in infants born to pregnant women with placental abruption versus without abruption, this finding should be interpreted cautiously, given high heterogeneity and overall poor quality of the included studies.

[Factors related to severe neonatal acidosis in planned cesarean section. A case-control study]

OBJECTIVES

To identify factors related to the occurrence of severe neonatal acidosis in case of planned caesarean section.

METHODS

Case-control study conducted between 1997 and 2016 among women with planned caesarean delivery at term. Cases were women whose neonates had neonatal arterial pH<7,0. For each case, two planned caesarean sections with neonatal pH≥7,0 were selected as controls. Women whose fetus had a congenital malformation and those whose anesthesia was not spinal anesthesia were excluded.

RESULTS

Among the 5014 planned cesarean sections of the study period, 38 severe neonatal acidosis were observed (incidence of 0,76% CI95 [0,54-1,04]). Compared to 72 controls, the 36 caesareans with severe neonatal acidosis were associated with more frequent maternal obesity (BMI≥30kg/m²), higher ephedrine doses, longer time from skin incision to infant delivery, and more extraction difficulties. After logistic regression, only maternal obesity remained associated with a significant increase in the risk of severe neonatal acidosis, ORa=3,73, 95%CI (1,11-12,56).

CONCLUSIONS

In case of planned cesarean section, the main risk factor for severe neonatal acidosis is the existence of maternal obesity.

LyTONEPAL: long term outcome of neonatal hypoxic encephalopathy in the era of neuroprotective treatment with hypothermia: a French population-based cohort

Background

Hypoxic-ischemic encephalopathy (HIE) is a rare neonatal condition affecting about 1‰ births. Despite a significant improvement in the management of this condition in the last ten years, HIE remains associated with high rates of death and severe neurological disability. From September 2015 to March 2017, a French national cohort of HIE cases was conducted to estimate the extent of long-term moderate and severe neurodevelopmental disability at 3 years and its determinants.

Methods

This prospective population-based cohort includes all moderate or severe cases of HIE, occurring in newborns delivered between 34 and 42 completed weeks of gestation and admitted to a neonatal intensive care unit. Detailed data on the pregnancy, delivery, and newborn until hospital discharge was collected from the medical records in maternity and neonatology units. All clinical examinations including biomarkers, EEG, and imaging were recorded. To ensure the completeness of HIE registration, a registry of non-included eligible neonates was organized, and the exhaustiveness of the cohort is currently checked using the national hospital discharge database. Follow-up is organized by the regional perinatal network, and 3 medical visits are planned at 18, 24 and 36 months. One additional project focused on early predictors, in particular early biomarkers, involves a quarter of the cohort.

Discussion

This cohort study aims to improve and update our knowledge about the incidence, the prognosis and the etiology of HIE, and to assess medical care. Its final objective is to improve the definition of this condition and develop prevention and management strategies for high-risk infants.

Trial registration

NCT02676063. Date of registration (Retrospectively Registered): February 8, 2016.

Pathology of perinatal brain damage: background and oxidative stress markers

PURPOSE

To review historical scientific background and new perspective on the pathology of perinatal brain damage. The relationship between birth asphyxia and subsequent cerebral palsy has been extensively investigated. The role of new and promising clinical markers of oxidative stress (OS) is presented.

METHODS

Electronic search of PubMed-Medline/EMBASE database has been performed. Laboratory and clinical data involving case series from the research group are reported.

RESULTS

The neuropathology of birth asphyxia and subsequent perinatal brain damage as well as the role of electronic fetal monitoring are reported following a review of the medical literature.

CONCLUSIONS

This review focuses on OS mechanisms underlying the neonatal brain damage and provides different perspective on the most reliable OS markers during the perinatal period. In particular, prior research work on neurodevelopmental diseases, such as Rett syndrome, suggests the measurement of oxidized fatty acid molecules (i.e., F4-Neuroprostanes and F2-Dihomo-Isoprostanes) closely related to brain white and gray matter oxidative damage.

A risk stratification model to predict adverse neonatal outcome in labor

OBJECTIVE

The development and evaluation of a labor risk model consisting of a combination of antepartum risk factors and intrapartum fetal heart rate (FHR) characteristics that can reliably identify those infants at risk for adverse neonatal outcome in labor.

STUDY DESIGN

A nested case-control study of term singleton deliveries at the nine hospitals between March 2007 and December 2009. Eligibility criteria included: gestational age ≥ 37.0 weeks; singleton pregnancy; documented continuous FHR monitoring for ≥ 2 h before delivery; assessment of FHR tracing at least every 20 min; and, available maternal and neonatal outcomes. Adverse neonatal outcome was defined as nonanomalous infants admitted to the newborn intensive care unit with either a 5 minute Apgar score <7 or an umbilical artery pH<7.1. Initial risk score was determined using data available at 1 h after admission. Patients with an initial risk score between 7 and 15 were considered high risk. Intrapartum risk scores were then created for these patients using FHR tracing data and labor characteristics.

RESULT

A total of 51 244 patients were identified meeting study criteria. Of the antepartum variables evaluated (n=31), 10 were associated with an adverse outcome. The high-risk group made up 28% of the population and accounted for 59.8% of the adverse outcomes. Intrapartum characteristics were then evaluated in this high-risk group. Intrapartum evaluation identified the highest risk group with a C/S rate of 40% and adverse outcome rate of 11.3%.

CONCLUSION

Incorporation of maternal and antepartum risk factors with FHR analysis can improve the ability to identify the fetus at risk in labor.

[Severe neonatal acidosis: comparison and analysis of obstetrical practices in two French perinatal centers]

OBJECTIVES

To identify and compare risk factors for severe neonatal acidosis, defined by an umbilical artery pH inferior to 7.00, and clinical practices in two different perinatal centers.

PATIENTS AND METHODS

In a retrospective study, from 2003 to 2008, in two university perinatal centers (Poitiers and Saint-Antoine in Paris) on all term pregnancies complicated by severe neonatal acidosis (umbilical artery pH<7.00), we elected to compare the following risk factors: maternal characteristics, medical and obstetrical histories, progress of pregnancy, labour and delivery as well as the neonatal status.

RESULTS

Among 23,508 births, 177 term newborns had severe neonatal acidosis. The rate was similar for both perinatal centers of Poitiers and Saint-Antoine (0.92% and 0.77% respectively). Factors associated with severe neonatal acidosis were similar in both centers: maternal age, thick meconium, prior cesarean section. There were differences in obstetrical practices between the two centers: there were more caesarean sections and assisted vaginal deliveries in Paris and more inductions of labour in Poitiers.

CONCLUSION

Severe neonatal acidosis is associated with the geographical origin, the progress of labour and the mode of delivery. It seems that severe neonatal acidosis is unrelated to cesarean delivery.

Risk factors for severe neonatal acidosis

OBJECTIVE

Neonatal asphyxia may have severe consequences in term newborns. Our purpose was to identify possible risk factors of severe acidosis during pregnancy and labor.

METHODS

In a case-control study from January 2003 to December 2008 in three university perinatal centers (two French and one Canadian hospitals), we analyzed 226 women with term pregnancies complicated by severe neonatal acidosis (umbilical artery pH less than 7.00). Cases were individually matched with controls with a normal acid-base status (pH 7.15 or greater) paired by parity. Groups were compared for differences in maternal, obstetric, and fetal characteristics. Univariable and logistic conditional regression were used to identify possible risk factors.

RESULTS

Among 46,722 births after 22 weeks, 6,572 preterm births and 829 stillbirths or terminations of pregnancy were excluded. From the 39,321 live term births, 5.30% of pH values were unavailable. Severe acidosis complicated 0.63% of 37,235 term structurally normal pregnancies. By using multivariate conditional regression, maternal age 35 years or older (35.0% compared with 15.5%; odds ratio [OR] 5.58, 95% confidence interval [CI] 2.51-12.40), prior neonatal death (3.5% compared with 0%), prior cesarean delivery (24.7% compared with 6.6%; OR 4.08, 95% CI 1.71-9.72) even after excluding cases of uterine rupture, general anesthesia (8.4 compared with 0.9%; OR 8.04, 95% CI 1.26-50.60), thick meconium (6.4% compared with 2.8%; OR 5.81, 95% CI 1.72-19.66), uterine rupture (4.4% compared with 0%), and abnormal fetal heart rate (66.1% compared with 19.8%; OR 8.77, 95% CI 3.72-20.78) were independent risk factors of severe neonatal acidosis.

CONCLUSION

Prior cesarean delivery, maternal age 35 years or older, prior neonatal death, general anesthesia, thick meconium, uterine rupture, and abnormal fetal heart rate are independent risk factors of severe neonatal acidosis.

LEVEL OF EVIDENCE

II.

[Fetal heart rate during labour: definitions and interpretation]

Continuous fetal heart rate monitoring is widely used during labor even in low risk pregnancies. Consensus is necessary to define and interpret accurately the different FHR patterns. The normal FHR tracing include baseline rate between 110-160 beats per minute (bpm), moderate variability (6-25 bpm), presence of accelerations and no decelerations. Uterine activity is monitored simultaneously: contractions frequency, duration, amplitude and relaxation time must be also normal. Abnormal baseline heart rate during 10 minutes or more is termed tachycardia above 160 bpm (except for FIGO above 150) and bradycardia below 110 bpm. Variability is minimal below 6 bpm and absent when non visible. Decelerations are classified as early, variable, late, and prolonged. Early and late decelerations have an onset gradual decrease of FHR, in contrast variable decelerations have an abrupt onset. Early deceleration is coincident in timing with uterine contraction. Variable deceleration is variable in onset, duration and timing, and may be described as typical or non reassuring. Late deceleration is associated with uterine contraction; the onset, nadir, and recovery occur after onset, peak and end of the contraction. Prolonged deceleration is lasting more than two but less 10 minutes, with almost onset abrupt and no repetition. Electronic fetal monitoring is a method to detect risk of fetal asphyxia; analysis and interpretation of FHR patterns are difficult with a high false positive rate, increasing operative deliveries. The patterns who are predictive of severe fetal acidosis include recurrent late or variable or prolonged decelerations or bradycardia, with absent FHR variability, and sudden severe bradycardia. The other FHR patterns are not conclusive and defined as non reassuring; obstetrical risk factors must be considered and other method (like scalp sampling for pH) utilised to evaluate fetal state.

Fetal acid-base and neonatal status after general and neuraxial anesthesia for elective cesarean section

OBJECTIVE

To analyze the relation between fetal acid-base and neonatal status in an observational cohort study of 900 consecutive women with singleton pregnancies at term undergoing elective cesarean section.

METHODS

The women were divided into 3 groups according to the type of anesthesia administered. Fetal acid-base status was assessed from umbilical cord blood (both artery and vein) and intermediate neonatal outcome was noted.

RESULTS

Epidural anesthesia was associated with the highest pH. The lowest pH and the highest pC0(2) values were associated with spinal anesthesia. Although maternal general anesthesia was associated with the highest values for partial pressure and saturation of oxygen in umbilical arterial blood, the newborns were more likely to be depressed than those born following spinal (P=0.0016) or epidural (P=0.0002) anesthesia.

CONCLUSIONS

If fetal oxygenation is the goal, general anesthesia provides the highest values for partial pressure and saturation of oxygen in umbilical arterial blood. However, epidural anesthesia was associated with better fetal and neonatal status than either spinal or general anesthesia.

The Current Crisis in Obstetrics

Of the issues leading to legal actions in obstetrics, the most important are events occurring before delivery that are deemed to account for the birth of a physically or mentally challenged child. In determining causation in the clinical setting, the diagnosis of fetal asphyxia can be made using blood gas and acid-base assessment. However, there are many subsidiary questions that in most cases cannot be answered, including when the asphyxia began, the severity and nature of the asphyxia during the exposure, the quality of the cardiovascular compensation, and when the brain damage occurred. When scientific proof is not available, the dilemma for the court is the requirement to reach a conclusion about the timing of brain damage on the balance of probabilities. Although it is of value, clinical risk scoring using fetal heart rate (FHR) monitoring may result in false positive predictions of fetal asphyxia. The problem in FHR monitoring is the lack of a detailed algorithm for the interpretation of FHR patterns with appropriate recommendations for management. Until such an algorithm is developed, health care workers cannot be expected to respond to fetal heart rate patterns consistently. Responsibility for the crisis in obstetrics must rest with the members of the health care disciplines who provide expert testimony. Progress made in research encourages us to assume that more is known about the causes of brain damage in the clinical setting than in fact is known. Similarly, health care professionals, parents, and lawyers often assume current methods of prediction and diagnosis to be more effective than they actually are.

Combination of early perinatal factors to identify near-term and term neonates for neuroprotection

OBJECTIVE

To determine early predictors of abnormal outcome at > or =24 months' age in neonates at risk for hypoxic-ischemic brain injury.

STUDY DESIGN

A prospective cohort study with developmental follow-up of > or =24 months. Infants were selected based on risk factors, and neurologic outcome was determined. Variables affecting the outcome were evaluated with univariate and multivariate methods, and a scoring system was devised to predict adverse outcome.

RESULTS

A total of 41 infants born > or =35 weeks' gestational age with possibility of hypoxic-ischemic insult were enrolled. In all, 39 (95%) had known outcomes, of whom 17 (48%) had an abnormal neurologic outcome, including five deaths. The variables within the first hour of life correlating with the adverse outcome were 1- and 5-minute Apgar scores, intubation in the delivery room and cord/initial base-deficit > or =20 mmol/l. A scoring system was derived based on significant variables, and a score > or =5 had a 90% positive predictive value for abnormal outcome. Seizures, multiorgan failure and abnormal imaging studies were also significantly associated with abnormal outcome.

CONCLUSIONS

The proposed scoring system, being highly predictive of outcome at 24 months' age, may be potentially useful in selecting subjects for preventive or therapeutic interventions to prevent or minimize neurologic morbidity due to hypoxic brain injury.

Calcitonin gene-related peptide contributes to the umbilical haemodynamic defence response to acute hypoxaemia

Despite clinical advances in obstetric practice, undiagnosed fetal hypoxaemia still contributes to a high incidence of perinatal morbidity. The fetal defence to hypoxaemia involves a redistribution of blood flow away from peripheral circulations towards essential vascular beds, such as the umbilical, cerebral, myocardial and adrenal circulations. In marked contrast to other essential vascular beds, the mechanisms mediating maintained perfusion of the umbilical circulation during hypoxaemia remain unknown. This study determined the role of calcitonin gene-related peptide (CGRP) in the maintenance of umbilical blood flow during basal and hypoxaemic conditions. Under anaesthesia, five sheep fetuses were instrumented with catheters and a Transonic probe around an umbilical artery, inside the fetal abdomen, at 0.8 of gestation. Five days later, fetuses were subjected to 0.5 h hypoxaemia during either i.v. saline or a selective CGRP antagonist in randomised order. Treatment started 30 min before hypoxaemia and ran continuously until the end of the challenge. The CGRP antagonist did not alter basal blood gas or cardiovascular status in the fetus. A similar fall in Pa,O2 occurred in fetuses during either saline (21 +/- 0.8 to 9 +/- 0.9 mmHg) or antagonist treatment (20 +/- 0.9 to 9 +/- 1.2 mmHg). Hypoxaemia during saline led to significant increases in arterial blood pressure, umbilical blood flow and umbilical vascular conductance. In marked contrast, hypoxaemia during CGRP antagonist treatment led to pronounced falls in both umbilical blood flow and umbilical vascular conductance without affecting the magnitude of the hypertensive response. In conclusion, CGRP plays an important role in the umbilical haemodynamic defence response to hypoxaemia in the late gestation fetus.

Reflections on the occurrence and significance of antepartum fetal asphyxia

This chapter documents the growing evidence of the occurrence and significance of antepartum fetal asphyxia. Perinatal mortality studies demonstrate that the majority of fetal deaths, including those due to asphyxia, occur in the antepartum period. Epidemiological studies of cerebral palsy conclude that a minority of cases of cerebral palsy are due to intrapartum fetal asphyxia and implicate antenatal events, including asphyxia. Cordocentesis studies have confirmed that antepartum fetal asphyxia occurs in the growth-restricted fetus and might contribute to the increase of stillbirths and cerebral palsy in these children. Blood gas and acid-base studies in the immature fetus have demonstrated the increased prevalence of fetal asphyxia, and particularly of moderate or severe asphyxia, of which the majority might occur in the antepartum period. The ultimate determination of the prevalence and significance of antepartum fetal asphyxia requires the development of a non-invasive specific measure of fetal asphyxia.

Influence of maternal, obstetric and fetal risk factors on the prevalence of birth asphyxia at term in a Swedish urban population

BACKGROUND

The influence of maternal, obstetric and fetal risk factors on the prevalence of birth asphyxia at term in a Swedish urban population.

OBJECTIVE

To investigate risk factors for Apgar score-defined birth asphyxia, birth asphyxia with hypoxic-ischemic encephalopathy and birth asphyxia-related death/disability.

MATERIAL AND METHODS

Retrospective case-control study in term neonates with birth asphyxia defined as Apgar score < 7 at 5 min. Cases originating from nonasphyctic causes (e.g. infection, maternal sedation) were excluded. Hypoxic-ischemic encephalopathy was diagnosed according to criteria by Sarnat. Maternal, obstetric and fetal risk factors were registered in 225 cases of birth asphyxia diagnosed in 42 203 live births occurring in the urban Swedish population studied. A matched control group was used for statistical evaluation.

RESULTS

Asphyxia was associated with single civil status, OR = 7.1 (95%CI 2.0, 27.6); intrauterine meconium release, OR = 4.1 (95%CI 1.8, 9.8); operative delivery, OR = 8.7 (95%CI 3.4, 24.6); breech delivery, OR = 20.3 (95%CI 3.0, 416.5); oxytocin augmentation, OR = 2.9 (95%CI 1.4, 6.3); cord complication, OR = 15.8 (95%CI 2.1, 341.5); external compression to assist delivery OR = 6.2 (95%CI 1.3, 45.7); and cardiotocography score, OR = 0.5 (95%CI 0.4, 0.6). Normal fetal heart rate variability, OR = 0.4 (95%CI 0.2, 0.6), repeated late decelerations irrespective of amplitude or repeated variable decelerations, OR = 29.4 (95%CI 5.7, 540.8) or occasional late or variable decelerations, OR = 2.2 (95%CI 1.3, 3.8), and no accelerations, OR = 5.2 (95%CI 2.0, 16.4), were associated with asphyxia. Operative or instrumental delivery was more common in all three asphyxia groups compared with controls. Leanness was a risk factor for asphyxia and for hypoxic-ischemic encephalopathy. Maternal age, smoking and illnesses, time of delivery (day/night, seasonal) and previous caesarean section were not associated with birth asphyxia.

CONCLUSIONS

An association between neonatal asphyxia and cardiotocography parameters, intrauterine meconium release, operative delivery, breech delivery, single civil status, oxytocin augmentation, cord complication, external compression to assist delivery and neonatal leanness was found. Abnormal fetal heart rate variability, repeated late decelerations irrespective of amplitude or repeated variable decelerations, occasional late or variable decelerations and no accelerations were associated with asphyxia.

Long-term developmental outcome of asphyxiated term neonates

Asphyxia remains one of the main causes of later disability in term infants. Despite many publications identifying possible predictors of outcome in this population of interest, little is known of the long-term developmental outcome of asphyxiated term neonates. Observational studies have largely focused on short-term outcomes, with an emphasis on significant neurologic sequelae and intellectual impairments. This article reviews the literature that has described the developmental outcome of asphyxiated term newborns. As part of this review, we have also highlighted the evolution of the definition of asphyxia and delineated appropriate markers that should be used in future research on this population.

Reproductive risk factors of fetal asphyxia at delivery: a population based analysis

To investigate reproductive maternal risk factors of intrapartum fetal asphyxia, we analyzed 556 women with singleton pregnancies complicated by intrapartum fetal asphyxia who gave birth at Kuopio University Hospital from January 1990 to December 1998. The general obstetric population (N=21746) was selected as the reference group and logistic regression analysis was used to identify independent reproductive risk factors. The incidence of intrapartum fetal asphyxia was 2.5%. Placental abruption, primiparity, alcohol use during pregnancy, low birth weight, preeclampsia, male fetuses, and small-for-gestational age births were independent risk factors of intrapartum asphyxia, with adjusted relative risks of 3.74, 3.10, 1.75, 1.57, 1.49, 1.48 and 1.33, respectively. Most cases of intrapartum fetal asphyxia occur in low-risk pregnancies and, therefore, risk screening in antenatal care cannot accurately predict which women will eventually need emergency care for fetal asphyxia.

The prediction and prevention of intrapartum fetal asphyxia in term pregnancies

OBJECTIVE

This study was undertaken to examine the roles of clinical risk scoring, electronic fetal heart rate monitoring, and fetal blood gas and acid-base assessment in the prediction and prevention of intrapartum fetal asphyxia in term pregnancies.

STUDY DESIGN

The outcomes of 166 term pregnancies with biochemically confirmed fetal asphyxia (umbilical artery base deficit at delivery, >12 mmol/L) were examined. This population included 83 pregnancies delivered abdominally matched with 83 pregnancies delivered vaginally. Antepartum and intrapartum clinical risk factors and neonatal complications were documented. Fetal assessments included fetal heart rate patterns in the fetal heart rate record and fetal capillary blood gas and acid-base assessments. Fetal asphyxia was classified as mild, moderate, or severe on the basis of umbilical artery base deficit (cutoff >12 mmol/L) and neonatal encephalopathy and other organ system complications.

RESULTS

Fetal asphyxial exposures were as follows: mild, 140; moderate, 22; and severe, 4. Intervention and delivery during the first or second stage of labor occurred in 98 of the 166 pregnancies. Predictive fetal heart rate patterns were the primary indication leading to intervention and delivery during the first or second stage of labor. Clinical risk factors when present were secondary indications in the clinical decision to intervene. Fetal blood gas and acid-base assessment was a useful supplementary test in 41 pregnancies. Intervention and delivery may have prevented the progression of mild asphyxia in 78 pregnancies and may have modified the degree of moderate or severe asphyxia in 20 pregnancies.

CONCLUSION

Although fetal heart rate patterns will not discriminate all asphyxial exposures, continuous fetal heart rate monitoring supplemented by fetal blood gas and acid-base assessment can be a useful fetal assessment paradigm for intrapartum fetal asphyxia. Such an assessment paradigm will not prevent all cases of moderate or severe fetal asphyxia. However, prediction and diagnosis with intervention and delivery during the first or second stage of labor could prevent the progression of mild asphyxia to moderate or severe asphyxia in some cases.

Fetal distress

The term 'fetal distress' should be replaced by 'suspected fetal compromise' because the diagnosis of 'fetal distress' is often unproven. Cardiotocography remains the cornerstone of making the diagnosis, but as a test it is renowned for its high sensitivity and low specificity. It has reduced intrapartum fetal mortality but not long-term neonatal morbidity or the incidence of cerebral palsy. There is no doubt that when obvious signs of fetal compromise, such as late decelerations in the presence of intrauterine growth retardation and oligohydramnios, are present, the diagnosis of fetal compromise is relatively simple. Often, however, the subtle signs of fetal compromise are missed; these are a change in the grade of meconium in the amniotic fluid, a rising base-line fetal heart rate, the absence of accelerations, the presence of 'atypical' variable decelerations or a combination of the above. To date, there is no test available to replace the cardiotocograph, although fetal pulse oximetry is the most promising adjunctive test. Above all, no test result obtained in isolation must detract from the whole clinical picture.

Intrapartum Fetal Surveillance. Is It Worthwhile?

Potentially significant intrapartum fetal asphyxia occurs in approximately 20 per 1000 births. Moderate and severe fetal asphyxia exposure with newborn morbidity occurs in 3 to 4 1000 births, with brain damage and subsequent disability in at least 1 per 1000 births. Although the prevalence of moderate and severe asphyxia is modest, prevention is important because of the serious implications of this complication to the child, family, and society. Because of the limited predictive value of clinical risk factors, the interpretation of patterns in a fetal heart rate record has become the primary screening test for intrapartum fetal asphyxia. Despite extensive clinic experience and numerous clinical trials, the benefits of EFM as a screening test have not been established, and harm may occur owing to unnecessary intervention. This observation raises serious ethical issues. When an intervention is initiated by the clinician rather than the patient, the clinician under greater obligation to ensure that the benefits outweigh the harm. Several factors complicate the demonstration of benefits of EFM as a screening test. There is no consensus regarding a protocol of fetal surveillance for low-risk patient who account for approximately 25% of intrapartum fetal asphyxia. Moderate and severe asphyxia cannot be prevented when asphyxial exposure has occurred before labor or before the onset of fetal surveillance. Prediction of intrapartum fetal asphyxia cannot occur when the quality of the record does not permit interpretation. Interpretation of predictive fetal heart rate patterns cannot occur unless the record is consistently and carefully scored. Prediction of most cases of intrapartum fetal asphyxia on the basis of fetal heart rate patterns is possible but difficult. Because the goal of intrapartum fetal surveillance is the prevention of moderate and severe fetal asphyxia, prediction must be achieved before fetal decompensation. Prediction must occur before absent baseline fetal heart rate variability evident in the record, which is uniformly associated with cerebral dysfunction and, in some cases, brain damage. The possibility of fetal asphyxia must be considered when, within a 1-hour window of recording, there are two or more cycles of minimal baseline fetal heart rate variability and two or more cycles of late or prolonged decelerations or both. Because approximately 9 of 10 predictive fetal heart rate patterns are false-positive, supplementary tests to confirm the diagnosis and to identify false-positives to prevent unnecessary intervention are essential. Until such time as additional fetal assessment tests are validated, blood gas and acid-base assessment of fetal blood can provide a definitive diagnosis and identify false-positive predictions.

Historical perspectives: forgotten past, unpredictable future

Intrapartum surveillance has in recent years become a matter of debate. Following its earlier development, first in auscultation and then 40 years ago in electronic monitoring, obstetricians accepted its use with great, perhaps too great, enthusiasm. Years later, attempts to evaluate the actual consequences of this use led to disappointment: although its benefit on perinatal mortality is acknowledged, two observations lead one to reconsider the legitimacy of its use. First the apparent lack of beneficial influence on neonatal long-term morbidity, and second the definite increase in the rate of caesarean section. Furthermore, recent comparative studies, despite some discrepancies, seem to indicate that clinical monitoring by auscultation leads to results as good as those from electronical monitoring, particularly with respect to fetal mortality and infant morbidity. These observations obviously merit careful consideration; some explanations may be put forward to explain these apparently surprising results. From a practical point of view, this discussion leads to two opposite choices for obstetric policy: either to 'go back' to auscultation or to try to identify indicators more specific to fetal asphyxia and increased risk of cerebral palsy, leading to more precise and fewer indications for caesarean section. This chapter on historical perspectives may be useful in pointing out what were the goals of the obstetric pioneers involved in electronic monitoring: definitely not to build theoretical considerations on the pathophysiology of fetal distress, but to gather continuous information about the fetal heart rate in the hope of detecting changes announcing fetal asphyxia before it becomes irremediable, and hence preventing fetal death. These promises have been fulfilled. It follows that continuous clinical monitoring, which provides the same kind of information, is quite likely to lead to similar clinical results. It also follows that this relatively cumbersome method has really nothing to do with the 'classical' clinical surveillance in use before the widespread acceptance of electronical monitoring. It may be beneficial to experiment with this specific type of clinical surveillance; it would be dangerous, however, to 'go back' to the type of monitoring practised 40 years ago.

Monitoring technology and the clinical perspective

Currently available technology requires a new look to reduce intervention as well as to improve the detection of the truly at-risk fetus. Iatrogenic causes of so-called fetal distress, in particular the administration of uterotonics without due attention to avoiding hyperstimulation, predominate as a reason for intervention. There needs to be a better definition of the starting point, i.e assessment of the fetal condition and identification of any risk factors, such as oligohydramnios and growth retardation, that might diminish fetal reserve. This will allow 'customization' of surveillance and management according to the needs of each individual fetus. There also needs to be better training and better agreement about the end-point of monitoring. For prospective surveillance, the aim is to avoid rather than to identify damage, and the definition of the appropriate point for intervention needs to come from better consensus on what is and what is not acceptable management based on current knowledge. New technology holds the promise that it can give trended information during labour, allow early recognition of problems and reduce unnecessary intervention. However, there is a need to ensure reliability and reproducibility of the readings before a new method is released. Co-operation with industry is essential, but the roles need to be well defined and the ultimate responsibility for establishing the role of a new technique has to come from the clinicians involved in intrapartum care.