Machine Learning Algorithm to Predict Acidemia Using Electronic Fetal Monitoring Recording Parameters.

BACKGROUND: Electronic fetal monitoring (EFM) is the universal method for the surveillance of fetal well-being in intrapartum. Our objective was to predict acidemia from fetal heart signal features using machine learning algorithms. METHODS: A case-control 1:2 study was carried out compromising 378 infants, born in the Miguel Servet University Hospital, Spain. Neonatal acidemia was defined as pH < 7.10. Using EFM recording logistic regression, random forest and neural networks models were built to predict acidemia. Validation of models was performed by means of discrimination, calibration, and clinical utility. RESULTS: Best performance was attained using a random forest model built with 100 trees. The discrimination ability was good, with an area under the Receiver Operating Characteristic curve (AUC) of 0.865. The calibration showed a slight overestimation of acidemia occurrence for probabilities above 0.4. The clinical utility showed that for 33% cutoff point, missing 5% of acidotic cases, 46% of unnecessary cesarean sections could be prevented. Logistic regression and neural networks showed similar discrimination ability but with worse calibration and clinical utility. CONCLUSIONS: The combination of the variables extracted from EFM recording provided a predictive model of acidemia that showed good accuracy and provides a practical tool to prevent unnecessary cesarean sections.

The utility of fetal heart rate deceleration's descending slope in searching for a non-National Institute of Child Health and Human Development parameter for the detection of fetal acidosis.

OBJECTIVE: To identify new parameters predicting fetal acidemia. METHODS: A retrospective case-control study in a cohort of deliveries from a tertiary referral hospital-based cohort deliveries in Zaragoza, Spain between 2018 and 2021 was performed. To predict fetal acidemia, the NICHD categorizations and non-NICHD parameters were analyzed in the electronic fetal monitoring (EFM). Those included total reperfusion time, total deceleration area and the slope of the descending limb of the fetal heart rate of the last deceleration curve. The accuracy of the parameters was evaluated using the specificity for (80%, 85%, 90%, 95%) sensitivity and the area under the receiver operating characteristic curve (AUC). RESULTS: A total of 10 362 deliveries were reviewed, with 224 cases and 278 controls included in the study. The NICHD categorizations showed reasonable discriminatory ability (AUC = 0.727). The non-NICHD parameters measured during the 30-min fetal monitoring, total deceleration area (AUC = 0.807, 95% CI: 0.770, 0.845) and total reperfusion time (AUC = 0.750, 95% CI: 0.707, 0.792), exhibited higher discriminatory ability. The slope of the descending limb of the fetal heart rate of the last deceleration curve had the best AUC value (0.853, 95% CI: 0.816, 0.889). The combination of total deceleration area or total reperfusion time with the slope demonstrated high discriminatory ability (AUC = 0.908, 95% CI: 0.882, 0.933; specificities of 71.6% and 72.7% for a sensitivity of 90%). CONCLUSIONS: The slope of the descending limb of the fetal heart rate of the last deceleration curve is the strongest predictor of fetal acidosis, but its combination with the total reperfusion time shows better clinical utility.

Total intrapartum fetal reperfusion time (fetal resilience) and neonatal acidemia.

OBJECTIVE: The main objective is to study the predictive capacity of intrapartum total fetal reperfusion (fetal resilience) by itself or in combination with other parameters as a predictor of neonatal acidemia. STUDY DESIGN: A retrospective case-control study was carried out at the Miguel Servet University Hospital (Zaragoza, Spain) on a cohort of 5694 pregnant women between June 2017 and October 2018. Maternal, perinatal, and cardiotocographic records were collected. Two reviewers blindly described the monitors with the American College of Obstetricians and Gynecologists (ACOG) categorizations and parameters and the non-ACOG parameters. Neonatal acidemia was defined as pH <7.10. The parameters analyzed to predict acidemia were evaluated using the sensitivity for specificity 90% value, and the area under the receiver operating characteristic curve. RESULTS: We recorded 192 infants with acidemia, corresponding to a global acidemia rate of 3.4%. Of these, 72 were excluded for lack of criteria, leaving 120 patients with arterial acidemia included in the study and 258 in the control group. The sensitivity (specificity 90%) of detection of acidemia was 42% for the ACOG III categorization (AUC, 0.524: 95% CI, 0.470-0.578), 24% for fetal reperfusion (AUC, 0.704: 95% CI, 0.649-0.759), 27% for total area of decelerations (AUC, 0.717: 95% CI, 0.664-0.771) and 50% for the multivariate model built from total reperfusion time (AUC, 0.826: 95% CI, 0.783-0.869). The total reperfusion time corresponding to a false negative rate of 10% is 23.75 min, with 28% of fetuses above this time. The AUC and sensitivity for a false negative rate of 10% are equivalent for deceleration area and time of reperfusion (p = .504). CONCLUSION: The total reperfusion time (fetal resilience) and total deceleration area are non-ACOG parameters with a good predictive ability for neonatal acidemia, higher than the ACOG III classification and without statistical differences between them. The discrimination ability of total reperfusion time can be improved using a multivariate model. As a cutoff for its use we suggest 23.75 min in 30 min corresponding to an acidemic classification rate of 90%. New parameters in combination with other maternal, obstetrics, or fetal variables, are required for the interpretation of fetal well-being.

Diagnostic capacity and interobserver variability in FIGO, ACOG, NICE and Chandraharan cardiotocographic guidelines to predict neonatal acidemia.

OBJECTIVE: Despite its routine use in intrapartum care, the technique of fetal cardiotocography has some limitations. The aim of this study is to analyze the predictive capacity and interobserver agreement in the latest versions of four international cardiotocography guidelines: Federation of Gynecology and Obstetrics (FIGO), American College of Obstetrics and Gynecology (ACOG), the National Institute for Health and Care Excellence (NICE) and Chandraharan, used to predict neonatal acidemia. STUDY DESIGN: The last 30 min of 150 cardiotocographic records were analyzed over all the pH ranges and were blindly evaluated by three independent reviewers. The sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operating characteristic curve (AUC) were calculated to assess the predictive capacity of each fetal cardiotocographic guideline. The degree of interobserver agreement was evaluated with the Fleiss Kappa coefficient. RESULTS: Observers found fetal cardiotocography guidelines to have a variable sensitivity and specificity. The Chandraharan classification reached the highest sensitivity (78.79%), while ACOG had the highest specificity (95.73%). On average for the three observers, Chandraharan had the highest discrimination capacity for neonatal acidemia, although this was only moderate (AUC 0.66; 95%CI, 0.55-0.77) and did not differ significantly from the remaining guidelines. The degree of agreement among the three observers, assessed according to the Fleiss Kappa coefficient, was generally acceptable or moderate for all items and classifications, being highest with the FIGO classification (ĸ = 0.35; 95%CI, 0.28-0.41) and lowest with the ACOG (ĸ = 0.23; 95%CI, 0.16-0.30). CONCLUSION: Although all the guidelines have a moderate capacity to predict neonatal acidemia, the Chandraharan guideline has the highest capacity. This follows a different approach from the others in that it relies on interpretations of cardiotocographic traces based on fetal physiology. The degree of interobserver agreement is, in general, acceptable for the four guidelines, and is the highest for FIGO.

Value of preoperative HE4 as predictor of advanced disease in endometrioid endometrial cancer.

OBJECTIVE: To explore the predictive value of serum human epididymis protein 4 (HE4) marker in the preoperative prediction of the risk of advanced disease in the endometrioid subtype of endometrial cancer, and its association with poor prognostic factors. In addition, a cut-off value of HE4 was defined to classify patients according to these results. METHODS: Prospective and multicenter cohort analytical pilot study of patients operated for endometrial cancer at the Miguel Servet University Hospital of Zaragoza (Spain) and the Complejo Universitario of León (Spain) from January 2017 to May 2019. Preoperative serum levels of HE4 were analyzed by clinical and pathologic characteristics. RESULTS: In all, 126 patients were included. A statistically significant association was found between the preoperative HE4 value and node involvement (P = 0.008), late-stage disease (P = 0.003), high histologic grade (P = 0.007), deep myometrial invasion (P = 0.001), lymphovascular space invasion (P = 0.001), and other pathologic factors. In addition, an HE4 cut-off value (156.4 pmol/L) has been determined to predict, preoperatively, which patients will present with early stage disease. CONCLUSIONS: The preoperative marker HE4 is a useful tool in the preoperative study of patients with endometrial cancer as it relates to late-stage disease as well as other prognostic factors in the endometrioid subtype of endometrial cancer.

Bacteriemia con afectación inflamatoria pélvica y miositis secundaria a histeroscopia diagnóstica. Reporte de un caso / Bacteremia with pelvic inflamatory involvement and secondary myositis due to diagnostic hysteroscopy. A case report

Resumen ANTECEDENTES: Las complicaciones infecciosas secundarias a la histeroscopia diagnóstica ambulatoria son bajas (0.6%). CASO CLÍNICO: Paciente de 46 años que a los 6 días posthisteroscopia diagnóstica ambulatoria por pólipos endometriales ingresó al servicio de Ginecología debido a un cuadro de piomiositis del músculo piriforme izquierdo, secundario a bacteriemia por Streptococcus pyogenes luego de un procedimiento ginecológico. Se trató con una dosis subcutánea de 40 mg de enoxaparina al día, y 2 g de ceftriaxona intravenosa cada 24 h, 240 mg de gentamicina intravenosa una vez al día. En las siguientes 48 horas se retiró la gentamicina y se prescribió 1 g de amoxicilina cada 8 h, por vía oral más 300 mg de clindamicina cada 8 h, también por vía oral que la paciente continuó durante 7 días más. En la resonancia magnética de control tomada a los 10 días quedó de manifiesto la tromboflebitis séptica en la vena iliaca interna izquierda y la persistencia del cuadro de piomiositis, con colección pélvica intra y extramuscular. Se decidió reingresarla para tratarla con el siguiente esquema: 1 g de amoxicilina por vía intravenosa cada 8 horas y 900 mg de clindamicina con igual pauta; se suspendió a las 24 h enseguida de la valoración por el infectólogo. Se indicó nuevo tratamiento antitrombótico con 0.5 mL de fraxiparina forte cada 24 horas, por vía subcutánea. El eco-Doppler descartó la trombosis venosa profunda; los hemocultivos que resultaron negativos. La paciente fue dada de alta con prescripción de fraxiparina Forte a dosis de 0.5 mL cada 24 horas durante 6 semanas; la evolución clínica fue favorable. La resonancia magnética de control a los 30 días reportó la desaparición completa del proceso inflamatorio. Luego de un año, la paciente permanece asintomática y continúa en seguimiento médico anual. CONCLUSIONES: La histeroscopia diagnóstica ambulatoria es una técnica eficaz y segura aunque excepcionalmente pueden sobrevenir complicaciones infecciosas derivadas de la técnica. Esto se previene con profilaxis antibiótica. La valoración previa a la histeroscopia es indispensable para advertir factores de riesgo que puedan controlarse o contraindiquen la realización del procedimiento.

Abstract BACKGROUND: Infectious complications due to diagnostic hysteroscopy are uncommon, being only 0.6% of all complications. CLINICAL CASE: A 46-year-old patient who, 6 days after performing an outpatient diagnostic hysteroscopy due to endometrial polyps, was admitted to the Gynecology Department, due to pyomyositis of the left piriformis muscle secondary to Streptococcus pyogenes bacteremia after gynecological procedure. Antithrombotic treatment was initiated with Enoxaparin 40 mg, 1/24h subcutaneously and antibiotic therapy with Ceftriaxone 2g/24h intravenously (iv) and Gentamicin 240 mg/24h iv, which 48 hours later were changed to Amoxicillin 1g/8h orally (VO) and Clindamycin 300 mg/8 hours orally for another 7 days. In the 10-day control magnetic resonance imaging (MRI), septic thrombophlebitis was seen in the left internal iliac vein as well as persistence of the pyomyositis condition, with intra and extramuscular pelvic collections. Therefore, a new admission was decided to start antibiotic therapy again with Amoxicillin 1g/8h IV and Clindamycin 900 mg/8h IV, which was suspended 24 hours after assessment by the Infectious Diseases Unit; and new antithrombotic treatment with Fraxiparin Forte 0.5 mL/24h subcutaneous, by consensus with the Vascular Surgery Service. ECO-DOPPLER was performed, which ruled out Deep Venous Thrombosis (DVT); and negative blood cultures. The patient was discharged with Fraxiparina Forte 0.5 mL/24h for 6 weeks presenting good clinical evolution. A new MRI control was scheduled for the following month, finding a complete resolution of the inflammatory process. Currently, one year later, the patient is asymptomatic and continues in annual follow-up in our consultations. CONCLUSIONS: Outpatient diagnostic hysteroscopy is an effective and safe technique. Infectious complications derived from the technique are infrequent, which is why it is not necessary to perform an antibiotic prophylaxis in patients who are going to undergo this test