Performance of machine-learning approach for prediction of pre-eclampsia in a middle-income country.

OBJECTIVE: Pre-eclampsia (PE) is a serious complication of pregnancy associated with maternal and fetal morbidity and mortality. As current prediction models have limitations and may not be applicable in resource-limited settings, we aimed to develop a machine-learning (ML) algorithm that offers a potential solution for developing accurate and efficient first-trimester prediction of PE. METHODS: We conducted a prospective cohort study in Mexico City, Mexico to develop a first-trimester prediction model for preterm PE (pPE) using ML. Maternal characteristics and locally derived multiples of the median (MoM) values for mean arterial pressure, uterine artery pulsatility index and serum placental growth factor were used for variable selection. The dataset was split into training, validation and test sets. An elastic-net method was employed for predictor selection, and model performance was evaluated using area under the receiver-operating-characteristics curve (AUC) and detection rates (DR) at 10% false-positive rates (FPR). RESULTS: The final analysis included 3050 pregnant women, of whom 124 (4.07%) developed PE. The ML model showed good performance, with AUCs of 0.897, 0.963 and 0.778 for pPE, early-onset PE (ePE) and any type of PE (all-PE), respectively. The DRs at 10% FPR were 76.5%, 88.2% and 50.1% for pPE, ePE and all-PE, respectively. CONCLUSIONS: Our ML model demonstrated high accuracy in predicting pPE and ePE using first-trimester maternal characteristics and locally derived MoM. The model may provide an efficient and accessible tool for early prediction of PE, facilitating timely intervention and improved maternal and fetal outcome. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Eficiencia de un enfoque de aprendizaje automático para la predicción de la preeclampsia en un país de ingresos medios OBJETIVO: La preeclampsia (PE) es una complicación grave del embarazo asociada a morbilidad y mortalidad materna y del feto. Dado que los modelos de predicción actuales tienen limitaciones y pueden no ser aplicables en situaciones con recursos limitados, se propuso desarrollar un algoritmo de aprendizaje automático (AA) que ofrezca una solución con potencial para desarrollar una predicción precisa y eficiente de la PE en el primer trimestre. MÉTODOS: Se realizó un estudio de cohorte prospectivo en Ciudad de México para desarrollar un modelo de predicción de la PE pretérmino (PEp) en el primer trimestre utilizando AA. Para la selección de variables se utilizaron las características maternas y los múltiplos de la mediana (MdM) obtenidos localmente para la presión arterial media, el índice de pulsatilidad de la arteria uterina y el factor de crecimiento placentario sérico. El conjunto de datos se dividió en subconjuntos de datos de entrenamiento, de validación y de test estadístico. Se empleó un método de red elástica para la selección de predictores, y el rendimiento del modelo se evaluó mediante el área bajo la curva de características operativas del receptor (ABC) y las tasas de detección (TD) con tasas de falsos positivos (TFP) del 10%. RESULTADOS: El análisis final incluyó a 3050 mujeres embarazadas, de las cuales 124 (4,07%) desarrollaron PE. El modelo de AA mostró una buena eficiencia, con un ABC de 0,897, 0,963 y 0,778 para la PEp, la PE de aparición temprana (PEat) y cualquier tipo de PE (todas las PE), respectivamente. Las TD con TFP del 10% fueron del 76,5%, 88,2% y 50,1% para la PEp, PEat y todas las PE, respectivamente. CONCLUSIONES: Nuestro modelo de AA demostró una alta precisión en la predicción de la PEp y la PEat utilizando características maternas del primer trimestre y MdM calculados localmente. El modelo puede proporcionar una herramienta eficiente y accesible para la predicción temprana de la PE, facilitando la intervención oportuna y la mejora de los resultados maternos y del feto.

Comorbidity, poverty and social vulnerability as risk factors for mortality in pregnant women with confirmed SARS-CoV-2 infection: analysis of 13 062 positive pregnancies including 176 maternal deaths in Mexico.

OBJECTIVE: Mortality in pregnancy due to coronavirus disease 2019 (COVID-19) is a current health priority in developing countries. Identification of clinical and sociodemographic risk factors related to mortality in pregnant women with COVID-19 could guide public policy and encourage such women to accept vaccination. We aimed to evaluate the association of comorbidities and socioeconomic determinants with COVID-19-related mortality and severe disease in pregnant women in Mexico. METHODS: This is an ongoing nationwide prospective cohort study that includes all pregnant women with a positive reverse-transcription quantitative polymerase chain reaction result for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) from the Mexican National Registry of Coronavirus. The primary outcome was maternal death due to COVID-19. The association of comorbidities and socioeconomic characteristics with maternal death was explored using a log-binomial regression model adjusted for possible confounders. RESULTS: There were 176 (1.35%) maternal deaths due to COVID-19 among 13 062 consecutive SARS-CoV-2-positive pregnant women. Maternal age, as a continuous (adjusted relative risk (aRR), 1.08 (95% CI, 1.05-1.10)) or categorical variable, was associated with maternal death due to COVID-19; women aged 35-39 years (aRR, 3.16 (95% CI, 2.34-4.26)) or 40 years or older (aRR, 4.07 (95% CI, 2.65-6.25)) had a higher risk for mortality, as compared with those aged < 35 years. Other clinical risk factors associated with maternal mortality were pre-existing diabetes (aRR, 2.66 (95% CI, 1.65-4.27)), chronic hypertension (aRR, 1.75 (95% CI, 1.02-3.00)) and obesity (aRR, 2.15 (95% CI, 1.46-3.17)). Very high social vulnerability (aRR, 1.88 (95% CI, 1.26-2.80)) and high social vulnerability (aRR, 1.49 (95% CI, 1.04-2.13)) were associated with an increased risk of maternal mortality, while very low social vulnerability was associated with a reduced risk (aRR, 0.47 (95% CI, 0.30-0.73)). Being poor or extremely poor were also risk factors for maternal mortality (aRR, 1.53 (95% CI, 1.09-2.15) and aRR, 1.83 (95% CI, 1.32-2.53), respectively). CONCLUSION: This study, which comprises the largest prospective consecutive cohort of pregnant women with COVID-19 to date, has confirmed that advanced maternal age, pre-existing diabetes, chronic hypertension, obesity, high social vulnerability and low socioeconomic status are risk factors for COVID-19-related maternal mortality. © 2021 International Society of Ultrasound in Obstetrics and Gynecology.

Increased levels of soluble fms-like tyrosine kinase-1 are associated with adverse outcome in pregnant women with COVID-19.

OBJECTIVE: In addition to the lungs, the placenta and the endothelium can be affected by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Soluble fms-like tyrosine kinase-1 (sFlt-1) and placental growth factor (PlGF) are markers of endothelial dysfunction and could potentially serve as predictors of severe coronavirus disease 2019 (COVID-19). We aimed to investigate the association of serum concentrations of sFlt-1 and PlGF with the severity of COVID-19 in pregnancy. METHODS: This was a prospective cohort study carried out in a tertiary care hospital in Mexico City, Mexico. Symptomatic pregnant women with a positive reverse-transcription quantitative polymerase chain reaction test for SARS-CoV-2 infection who fulfilled the criteria for hospitalization were included. The primary outcome was severe pneumonia due to COVID-19. Secondary outcomes were intensive care unit (ICU) admission, viral sepsis and maternal death. sFlt-1 levels were expressed as multiples of the median (MoM). The association between sFlt-1 and each adverse outcome was explored by logistic regression analysis, adjusted for gestational age for outcomes occurring in more than five patients, and the predictive performance was assessed by receiver-operating-characteristics-curve analysis. RESULTS: Among 113 pregnant women with COVID-19, higher sFlt-1 MoM was associated with an increased probability of severe pneumonia (adjusted odds ratio (aOR), 1.817 (95% CI, 1.365-2.418)), ICU admission (aOR, 2.195 (95% CI, 1.582-3.047)), viral sepsis (aOR, 2.318 (95% CI, 1.407-3.820)) and maternal death (unadjusted OR, 5.504 (95% CI, 1.079-28.076)). At a 10% false-positive rate, sFlt-1 MoM had detection rates of 45.2%, 66.7%, 83.3% and 100% for severe COVID-19 pneumonia, ICU admission, viral sepsis and maternal death, respectively. PlGF values were similar between women with severe and those with non-severe COVID-19 pneumonia. CONCLUSION: sFlt-1 MoM is higher in pregnant women with severe COVID-19 and has the capability to predict serious adverse pregnancy events, such as severe pneumonia, ICU admission, viral sepsis and maternal death. © 2021 International Society of Ultrasound in Obstetrics and Gynecology.

Performance of fetal middle cerebral artery peak systolic velocity for prediction of anemia in untransfused and transfused fetuses: systematic review and meta-analysis.

OBJECTIVE: To evaluate the performance of fetal middle cerebral artery peak systolic velocity (MCA-PSV) ≥ 1.5 multiples of the median (MoM) for the prediction of moderate-severe anemia, in untransfused and transfused fetuses. METHODS: A systematic search was performed to identify relevant observational studies reported in the period 2008-2018 that evaluated the performance of MCA-PSV, using a threshold of 1.5 MoM for the prediction of fetal anemia. Diagnosis of fetal anemia by blood sampling was the reference standard. A hierarchical summary receiver-operating characteristics (hSROC) curve was constructed using random-effects modeling. Subgroup and meta-regression analyses, according to the number of previous intrauterine transfusions, were performed. RESULTS: Twelve studies and 696 fetuses were included in the meta-analysis. The area under the hSROC curve (AUC) for moderate-severe anemia was 83%. Pooled sensitivity and specificity (95% CI) were 79% (70-86%) and 73% (62-82%), respectively, and positive and negative likelihood ratios were 2.94 (95% CI, 2.13-4.00) and 0.272 (95% CI, 0.188-0.371). When considering only untransfused fetuses, prediction improved, achieving an AUC of 87%, sensitivity of 86% (95% CI, 75-93%) and specificity of 71% (95% CI, 49-87%). A decline in sensitivity for the prediction of moderate-severe anemia by MCA-PSV ≥1.5 MoM was observed (estimate, -5.5% (95% CI, -10.7 to -0.3%), P = 0.039) as the number of previous transfusions increased. CONCLUSIONS: MCA-PSV ≥ 1.5 MoM for the prediction of moderate-severe anemia in untransfused fetuses shows moderate accuracy (86% sensitivity and 71% specificity), which declines with increasing number of intrauterine transfusions. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.

Rendimiento de la velocidad sistólica máxima de la arteria cerebral media fetal para la predicción de la anemia en fetos sometidos a transfusión y no sometidos: revisión sistemática y metaanálisis OBJETIVOS: Estimar las diferencias en la frecuencia del diagnóstico del útero septo mediante tres definiciones diferentes y determinar si estas diferencias son significativas en la práctica clínica, y examinar la relación entre el diagnóstico del útero septo, por medio de cada una de las tres definiciones, y la infertilidad o el aborto espontáneo previo, así como con el costo de la recomendación de cirugía. MÉTODOS: Este estudio fue un análisis secundario de los datos de un estudio prospectivo de 261 mujeres en edad reproductiva que asisten de forma consecutiva a una clínica privada especializada en el diagnóstico y tratamiento de las malformaciones congénitas del útero. El nuevo análisis de los conjuntos de datos se realizó de acuerdo con tres maneras diferentes de definir el útero septo: siguiendo las recomendaciones de la Sociedad Americana de Medicina Reproductiva (ASRM, por sus siglas en inglés), una actualización de 2016 de las de la Sociedad Americana de la Fertilidad de 1988 (ASRM-2016: profundidad de la hendidura interna del fondo uterino ≥1,5 cm, ángulo de la hendidura interna <90o y profundidad de la hendidura externa <1 cm); con base en las recomendaciones de la Sociedad Europea para la Reproducción Humana y Embriología/Sociedad Europea de Endoscopía Ginecológica (ESHRE/ESGE, por sus siglas en inglés), publicadas en 2013 y revisadas en 2016 (ESHRE/ESGE-2016: profundidad de la hendidura interna del fondo uterino >50% del espesor de la pared uterina y profundidad de la hendidura externa <50% del espesor de la pared uterina, cuando se mide el espesor de la pared uterina por encima de la línea interostial/intercornual); y utilizando una definición publicada el año pasado que se basaba en la decisión tomada con mayor frecuencia por un grupo de expertos (Malformación Uterina Congénita según los Expertos; CUME, por sus siglas en inglés) (CUME-2018: profundidad de la hendidura interna del fondo uterino ≥1 cm y profundidad de la hendidura externa del fondo uterino <1cm). Se comparó la tasa de diagnóstico del útero septo utilizando cada una de estas tres definiciones y, para cada una, se estimó la relación entre el diagnóstico y la infertilidad y/o el aborto espontáneo previo, y se anticiparon los costos asociados con su implementación mediante un método de estimación conjetural. RESULTADOS: Aunque el 32,6% (85/261) de las mujeres cumplieron con los criterios de una de las tres definiciones de útero septo, sólo el 2,7% (7/261) de ellas se pudieron definir como con útero septo de acuerdo con las tres definiciones. Significativamente, se diagnosticaron más casos de útero septo usando los criterios de ESHRE/ESGE-2016 que usando los de ASRM-2016 (31% vs 5%, riesgo relativo (RR)=6,7, P<0.0001) o de CUME-2018 (31% vs 12%, RR=2,6, P<0.0001). También se observaron casos frecuentes que no pudieron ser clasificados definitivamente por ASRM-2016 (zona gris: ni normal/arcuado ni septo; 6,5%). No hubo diferencias significativas (P>0,05) en la prevalencia de útero septo en mujeres con infertilidad vs mujeres fértiles, según ASRM-2016 (5% vs 4%), ESHRE/ESGE-2016 (35% vs 28%) o CUME-2018 (11% vs 12%). El diagnóstico del útero septo fue significativamente más frecuente en mujeres con aborto espontáneo previo, según los criterios de ASRM-2016 (11% vs 3%; P=0,04) y de CUME-2018 (22 vs 10%; P=0,04), pero no según los criterios de ESHRE/ESGE-2016 (42% vs 28%; P=0,8). Los cálculos mostraron que los costos globales para el sistema de salud dependerían en gran medida de los criterios utilizados desde el punto de vista clínico para definir el útero septo, siendo los costos asociados con la definición de ESHRE/ESGE-2016 potencialmente de 100-200 mil millones de dólares adicionales durante 5 años, en comparación con los asociados a las definiciones ASRM-2016 y CUME-2018. CONCLUSIONES: La prevalencia del útero septo según las definiciones de ESHRE/ESGE-2016, ASRM-2016 y CUME-2018 difiere considerablemente. Una limitación importante de la clasificación ASRM, que debe ser abordada, es la alta proporción de casos no clasificables originalmente denominados, por nosotros, como en la 'zona gris'. La alta tasa de sobrediagnóstico del útero septo en función de ESHRE/ESGE-2016 puede llevar a un uso innecesario de la cirugía y, por lo tanto, a un riesgo innecesario en estas mujeres y puede imponer una carga financiera considerable a los sistemas sanitarios. Se deben fomentar los esfuerzos para definir criterios clínicamente significativos y aplicables de forma universal para el diagnóstico del útero septo.

Type 2 diabetes detection based on serum sample Raman spectroscopy.

In this work, we propose to the Raman spectroscopy as a new technique for the detection of the type 2 diabetes using blood serum samples. The serum samples were obtained from 15 patients who were clinically diagnosed with type 2 diabetes mellitus and 20 healthy volunteers. The average spectra showed equally intense peaks as, 695 cm-1, the doublet of tyrosine at 828 and 853 cm-1, phenylalanine at 1002 and 1028 cm-1, the phospholipid shoulder at 1300-1345 cm-1, and proteins (amide I) at 1654 cm-1. The major differences were found at 661 and 1404 cm-1 (glutathione), 714 (polysaccharides), 605 (Phe), 545 cm-1 (tryptophan), and the shoulder of amide III at 1230-1282 cm-1, where seem to disappear in the diabetes spectrum. On the contrary, the region that is more highlighted due to that diabetes peaks are clearly more intense was 897-955 cm-1. Principal component analysis and linear discriminate analysis were employed for developing discrimination method. The first three principal components provided a classification of the samples from healthy and diabetes patients with high sensitivity and specificity. In addition, when the first principal component was plotted as a function of the Raman shift, it revealed these shifts accounted for the greatest differences between control and diabetes samples, which coincided with the shifts of spectral differences shown by mean spectra. Our results demonstrated that serum sample Raman spectroscopy promises to become a non-invasive support tool of the currently applied techniques for type 2 diabetes detection, decreasing the false-positive cases.

[Prevalence of HPV high-risk serotypes detected by PCR in patients with normal cervical cytology at the Hospital Regional Adolfo López Mateos, ISSSTE]. / Prevalencia de genotipos de alto riesgo de VPH detectados por PCR en pacientes con citología normal del Hospital Regional Adolfo López Mateos, ISSSTE.

Background: Is fundamental to determine the prevalence of human papiloma virus (HVP) high-risk serotypes in local and regional population in order for health providers to offer patients, vaccines and treatments against specific population-based serotypes. Objetive: To determine the prevalence of HPV High risk serotypes detected by PCR in patients with normal cytology from the ISSSTE Adolfo Lopez Mateos Regional Hospital. Methods: An observational, descriptive, prospective study was conducted from cervical cytologies and high risk HPV test by PCR in patients from the Regional Hospital Adolfo López Mateos, ISSSTE, during the period January 2013-December 2015. Cases of patients with negative cervical cytology were included. Information about age, the result of cervical cytology and high risk HPV test by PCR was obtained. The overall prevalence of HPV infection and the most prevalent serotypes by age groups were calculated. Results: A total of 3258 cervical smears were performed, of which 2557 were negative (78.4%), from this, the global prevalence of HPV infection was 10.2% (n=262). We found that 1.8% (n = 45) of negative reports had HPV16 infection, 0.5% (n=13) had HPV18 and 8.9% (n = 227) were infected by Viral Pool of other high-risk serotypes. The prevalence of infection by viral pool of high risk serotypes was 11.5% in women <20 years, 12.9% in women between 20-29 years and 22.2% in women between 30-39 years. This prevalence was lower in patients older than 40 years (p<0.05). Conclusion: A higher prevalence of viral pool high risk serotypes was found in patients with normal cytology, than the HPV16 and HPV-8 prevalence, which was significantly higher in women younger than 40 years.