BACKGROUND: In clinical medicine, fetal heart rate (FHR) monitoring using cardiotocography (CTG) is one of the most commonly used methods for assessing fetal acidosis. However, as the visual interpretation of CTG depends on the subjective judgment of the clinician, this has led to high inter-observer and intra-observer variability, making it necessary to introduce automated diagnostic techniques.
METHODS: In this study, we propose a computer-aided diagnostic algorithm (Hybrid-FHR) for fetal acidosis to assist physicians in making objective decisions and taking timely interventions. Hybrid-FHR uses multi-modal features, including one-dimensional FHR signals and three types of expert features designed based on prior knowledge (morphological time domain, frequency domain, and nonlinear). To extract the spatiotemporal feature representation of one-dimensional FHR signals, we designed a multi-scale squeeze and excitation temporal convolutional network (SE-TCN) backbone model based on dilated causal convolution, which can effectively capture the long-term dependence of FHR signals by expanding the receptive field of each layer\'s convolution kernel while maintaining a relatively small parameter size. In addition, we proposed a cross-modal feature fusion (CMFF) method that uses multi-head attention mechanisms to explore the relationships between different modalities, obtaining more informative feature representations and improving diagnostic accuracy.
RESULTS: Our ablation experiments show that the Hybrid-FHR outperforms traditional previous methods, with average accuracy, specificity, sensitivity, precision, and F1 score of 96.8, 97.5, 96, 97.5, and 96.7%, respectively.
CONCLUSIONS: Our algorithm enables automated CTG analysis, assisting healthcare professionals in the early identification of fetal acidosis and the prompt implementation of interventions.

OBJECTIVE: To investigate the predictive value of pre-induction digital examination, sonographic measurements and parity for the prediction of non-reassuring fetal status and cord arterial pH < 7.2 prior to the induction of labor (IOL).
METHODS: This was a prospective observational study, including 384 term pregnancies undergoing IOL. Before the IOL, the Bishop score (BS) by digital examination, sonographic Doppler parameters and the estimated fetal weight (EFW) was assessed. The fetal cord arterial was sampled to measure the pH at delivery. Multivariate logistic regression analysis was performed to identify independent predictors of non-reassuring fetal status and low cord arterial pH.
RESULTS: Forty four cases (11.5%) had non-reassuring fetal status, and 76 cases (19.8%) had fetal cord arterial pH < 7.2. In the non-reassuring fetal status group, the incidence of cord arterial pH < 7.2 was significantly higher than that in the normal fetal heart rate group (χ2 = 6.401, p = 0.011). Multivariate analysis indicated that significant independent predictors of non-reassuring fetal status were nulliparity (adjusted odds ratio [AOR]: 3.746, p = 0.003), EFW < 10th percentile (AOR: 3.764, p = 0.003) and cerebroplacental ratio (CPR) < 10th centile (AOR:4.755, p < 0.001). In the prediction of non-reassuring fetal status, the performance of the combination of nulliparity and EFW < 10th percentile was improved by the addition of CPR < 10th percentile (AUC: 0.681, (95%CI: 0.636 to 0.742) vs 0.756, (95%CI:0.713 to 0.795)), but the difference was not significant (DeLong test: z = 1.039, p = 0.053).. None of the above variables were predictors of cord arterial pH < 7.2.
CONCLUSIONS: The risk of fetal acidosis has increased in cases of non-reassuring fetal status. Nulliparity, small for gestational age and CPR < 10th centile are independent predictors for non-reassuring fetal status in term fetuses, though the addition of CPR < 10th centile could not significantly improve the screening accuracy.

We aimed to examine the impact of a daily pro-inflammatory diet during pregnancy on intrapartum fetal acidemia using a large birth cohort study in Japan. We used data on singleton pregnancies in the Japan Environment and Children\'s Study (JECS) involving births from 2011 to 2014 through vaginal delivery to calculate the maternal dietary inflammatory index (DII). Participants were categorized according to DII quintiles. A multiple logistic regression model was used to estimate the risk of a pro-inflammatory diet on fetal umbilical artery pH. In total, 56,490 participants were eligible for this study. Multiple regression analysis showed that nulliparous women who had undergone vaginal delivery and were consuming a pro-inflammatory diet had an increased risk of pH < 7.10 (adjusted odds ratio [aOR]: 1.64, 95% confidence interval [CI]: 1.12-2.39). Among these women, the risk of pH < 7.10 was not affected by the duration of labor (aOR: 1.64, 95% CI: 1.11-2.42). In conclusion, following a pro-inflammatory diet during pregnancy is a risk factor for fetal acidosis among nulliparous women undergoing vaginal delivery. A high DII diet during pregnancy may modify the intrapartum fetal heart rate pattern via intrauterine inflammation.

An increasing number of delivering women experience major morbidity and mortality. Limited work has been done on automated predictive models that could be used for prevention. Using only routinely collected obstetrical data, this study aimed to develop a predictive model suitable for real-time use with an electronic medical record. We used a retrospective cohort study design with split validation. The denominator consisted of women admitted to a delivery service. The numerator consisted of women who experienced a composite outcome that included both maternal (eg, uterine rupture, postpartum hemorrhage), fetal (eg, stillbirth), and neonatal (eg, hypoxic ischemic encephalopathy) adverse events. We employed machine learning methods, assessing model performance using the area under the receiver operator characteristic curve and number needed to evaluate. A total of 303,678 deliveries took place at 15 study hospitals between January 1, 2010, and March 31, 2018, and 4130 (1.36%) had ≥1 obstetrical complication. We employed data from 209,611 randomly selected deliveries (January 1, 2010, to March 31, 2017) as a derivation dataset and validated our findings on data from 52,398 randomly selected deliveries during the same time period (validation 1 dataset). We then applied our model to data from 41,669 deliveries from the last year of the study (April 1, 2017, to March 31, 2018 [validation 2 dataset]). Our model included 35 variables (eg, demographics, vital signs, laboratory tests, progress of labor indicators). In the validation 2 dataset, a gradient boosted model (area under the receiver operating characteristic curve or c statistic, 0.786) was slightly superior to a logistic regression model (c statistic, 0.778). Using an alert threshold of 4.1%, our final model would flag 16.7% of women and detect 52% of adverse outcomes, with a number needed to evaluate of 20.9 and 0.455 first alerts per day per 1000 annual deliveries. In conclusion, electronic medical record data can be used to predict obstetrical complications. The clinical utility of these automated models has not yet been demonstrated. To conduct interventions to assess whether using these models results in patient benefit, future work will need to focus on the development of clinical protocols suitable for use in interventions.

OBJECTIVE: To determine which parameter of umbilical arterial cord gas analysis, pH, base deficit, lactate concentration, ionogram values, or glucose level index is the best predictor of neonatal morbidity at term.
METHODS: We conducted a 15-month retrospective cohort study that included all nonanomalous, singleton, term births at a single center. The predictive ability of lactate concentration, base deficit, pH, ionogram values, and glucose level were compared using receiver-operating characteristic curves for global and neurological composite morbidity. Optimal cutoff values for lactate concentration, base deficit, and pH were estimated based on their maximum area under the curve.
RESULTS: We included 5161 newborns: 52 (1.01 %) had global composite morbidity, and 17 had (0.33 %) neurological composite morbidity. Blood levels of potassium, calcium, natremia, glucose level, and HCO3- did not differ significantly between groups. pH, partial pressure of CO2, partial pressure of O2, base deficit, and lactate levels differed significantly between neonates in the groups with and without global composite morbidity. Nearly similar results were found for neurological composite morbidity. The predictive ability did not differ between arterial pH and arterial lactate concentration (P = .25) and base deficit (P = .79). Optimal cutoff values to predict global composite morbidity were arterial pH 7.144, venous pH 7.236, and arterial lactate concentration 6.5 mmol/L.
CONCLUSIONS: Acid-base status analysis remains the best objective indicator for predicting neonatal morbidity and can be estimated using pH, lactate, or base deficit. Ionogram cord blood composition and glucose level do not appear to be useful for this purpose.

Cardiotocography (CTG) is currently the most commonly used method for intrapartum fetal monitoring during labor. However, a high false-positive rate of fetal acidosis indicated by CTG leads to an increase in obstetric interventions. We developed a microdialysis probe that is integrated into a fetal scalp electrode allowing continuous measurement of lactate subcutaneously, thus giving instant information about the oxygenation status of the fetus. Our aim was to establish proof of concept in an animal model using a microdialysis probe to monitor lactate subcutaneously.
We performed an in vivo study in adult male wild-type Wistar rats. We modified electrodes used for CTG monitoring in human fetuses to incorporate a microdialysis membrane. Optimum flow rates for microdialysis were determined in vitro. For the in vivo experiment, a microdialysis probe was inserted into the skin on the back of the animal. De-oxygenation and acidosis were induced by lowering the inspiratory oxygen pressure. Oxygenation and heart rate were monitored. A jugular vein cannula was inserted to draw blood samples for analysis of lactate, pH, pco2 , and saturation. Lactate levels in dialysate were compared with plasma lactate levels.
Baseline blood lactate levels were around 1 mmol/L. Upon de-oxygenation, oxygen saturation fell to below 40% for 1 h and blood lactate levels increased 2.5-fold. Correlation of dialysate lactate levels with plasma lactate levels was 0.89 resulting in an R2 of .78 in the corresponding linear regression.
In this animal model, lactate levels in subcutaneous fluid collected by microdialysis closely reflected blood lactate levels upon transient de-oxygenation, indicating that our device is suitable for subcutaneous measurement of lactate. Microdialysis probe technology allows the measurement of multiple compounds in the dialysate, such as glucose, albumin, or inflammatory mediators, so this technique may offer the unique possibility to shed light on fetal physiology during the intrapartum period.

Automated fetal heart rate (FHR) analysis removes inter- and intra-expert variability, and is a promising solution for reducing the occurrence of fetal acidosis and the implementation of unnecessary medical procedures. The first steps in automated FHR analysis are determination of the baseline, and detection of accelerations and decelerations (A/D). We describe a new method in which a weighted median filter baseline (WMFB) is computed and A/Ds are then detected.
The filter weightings are based on the prior probability that the sampled FHR is in the baseline state or in an A/D state. This probability is computed by estimating the signal\'s stability at low frequencies and by progressively trimming the signal. Using a competition dataset of 90 previously annotated FHR recordings, we evaluated the WMFB method and 11 recently published literature methods against the ground truth of an expert consensus. The level of agreement between the WMFB method and the expert consensus was estimated by calculating several indices (primarily the morphological analysis discordance index, MADI). The agreement indices were then compared with the values for eleven other methods. We also compared the level of method-expert agreement with the level of interrater agreement.
For the WMFB method, the MADI indicated a disagreement of 4.02% vs. the consensus; this value is significantly lower (p<10-13) than that calculated for the best of the 11 literature methods (7.27%, for Lu and Wei\'s empirical mode decomposition method). The level of inter-expert agreement (according to the MADI) and the level of WMFB-expert agreement did not differ significantly (p=0.22).
The WMFB method reproduced the expert consensus analysis better than 11 other methods. No differences in performance between the WMFB method and individual experts were observed. The method Matlab source code is available under General Public Licence at http://utsb.univ-catholille.fr/fhr-wmfb.

暂无摘要。

OBJECTIVE: Nomograms are statistical models that combine variables to obtain the most accurate and reliable prediction for a particular risk. Fetal heart rate (FHR) interpretation alone has been found to be poorly predictive for fetal acidosis while other clinical risk factors exist. The aim of this study was to create and validate a nomogram based on FHR patterns and relevant clinical parameters to provide a non-invasive individualized prediction of umbilical artery pH during labour.
METHODS: A retrospective observational study was conducted on 4071 patients in labour presenting singleton pregnancies at >34 gestational weeks and delivering vaginally. Clinical characteristics, FHR patterns and umbilical cord gas of 1913 patients were used to construct a nomogram predicting an umbilical artery (Ua) pH <7.18 (10th centile of the study population) after an univariate and multivariate stepwise logistic regression analysis. External validation was obtained from an independent cohort of 2158 patients. Area under the receiver operating characteristics (ROC) curve, sensitivity, specificity, positive and negative predictive values of the nomogram were determined.
RESULTS: Upon multivariate analysis, parity (p < 0.01), induction of labour (p = 0.01), a prior uterine scar (p = 0.02), maternal fever (p = 0.02) and the type of FHR (p < 0.01) were significantly associated with an Ua pH <7.18 (p < 0.05). Apgar score at 1, 5 and 10 min were significantly lower in the group with an Ua pH <7.18 (p < 0.01). The nomogram constructed had a Concordance Index of 0.75 (area under the curve) with a sensitivity of 57%, a specificity of 91%, a negative predictive value of 5% and a positive predictive value of 99%. Calibration found no difference between the predicted probabilities and the observed rate of Ua pH <7.18 (p = 0.63). The validation set had a Concordance Index of 0.72 and calibration with a p < 0.77.
CONCLUSIONS: We successfully developed and validated a nomogram to predict Ua pH by combining easily available clinical variables and FHR. Discrimination and calibration of the model were statistically good. This mathematical tool can help clinicians in the management of labour by predicting umbilical artery pH based on FHR tracings.

OBJECTIVE: In case of abnormal fetal heart rate, there is no consensus on the decision threshold pH scalp leading to a rapid birth. The objective of this study was to compare neonatal issues and cesarean rate in two maternity using different decision thresholds of scalp pH.
METHODS: A comparative retrospective study conducted in two level III maternity units between January 2013 and May 2014, one maternity unit used a decision threshold of 7.20 (maternity unit 7,20), and the other one a threshold of 7.25 (maternity unit 7,25). An adverse neonatal outcome was defined by a composite endpoint of neonatal morbidity. The risk of cesarean was assessed using a multivariate analysis.
RESULTS: One hundred and four patients were included in the maternity unit 7,20 and 163 patients in the maternity 7,25. Adverse neonatal outcome was similar in both maternities (25% vs. 30,1%; P=0.4). The average pH at birth was similar in both maternities, as well as the Apgar score at 5minutes and neonatal transfer rates. However, BE<-12 was more frequent in maternity using 7,20 scalp pH threshold (7% vs. 0%; P<0.01). The cesarean rate was higher in maternity 7,25 (adjusted OR=2.23 95% CI [1.17-4.25]).
CONCLUSIONS: It seems that a decisional threshold fixed to 7,20 could be used reasonably. It could allow to reduce cesarean rate. Other studies are, however, needed to confirm that such threshold of 7,20 does not increase the risk of severe acidosis.