RESUMEN
Given the barriers to early detection of gestational diabetes mellitus (GDM), this study aimed to develop an artificial intelligence (AI)-based prediction model for GDM in pregnant Mexican women. Data were retrieved from 1709 pregnant women who participated in the multicenter prospective cohort study 'Cuido mi embarazo'. A machine-learning-driven method was used to select the best predictive variables for GDM risk: age, family history of type 2 diabetes, previous diagnosis of hypertension, pregestational body mass index, gestational week, parity, birth weight of last child, and random capillary glucose. An artificial neural network approach was then used to build the model, which achieved a high level of accuracy (70.3%) and sensitivity (83.3%) for identifying women at high risk of developing GDM. This AI-based model will be applied throughout Mexico to improve the timing and quality of GDM interventions. Given the ease of obtaining the model variables, this model is expected to be clinically strategic, allowing prioritization of preventative treatment and promising a paradigm shift in prevention and primary healthcare during pregnancy. This AI model uses variables that are easily collected to identify pregnant women at risk of developing GDM with a high level of accuracy and precision.
Asunto(s)
Diabetes Mellitus Tipo 2 , Diabetes Gestacional , Niño , Embarazo , Femenino , Humanos , Recién Nacido , Diabetes Gestacional/diagnóstico , Estudios Prospectivos , Inteligencia Artificial , México/epidemiología , Factores de RiesgoRESUMEN
Resumen OBJETIVO: Determinar, conforme al Índice de Robson, la tasa de cesáreas en pacientes atendidas, en un periodo de nueve meses, en el Centro de Investigación Materno-Infantil del Grupo de Estudios al Nacimiento. MATERIALES Y MÉTODOS: Estudio retrospectivo, transversal, descriptivo y monocéntrico efectuado en pacientes embarazadas atendidas en el Centro de Investigación Materno Infantil del Grupo de Estudios al Nacimiento, (CIMIGen) que finalizaron el embarazo por cesárea y en quienes se utilizó el índice de Robson para su clasificación. Parámetros de estudio: indicaciones de la cesárea, riesgo de pérdida del bienestar fetal, alta de progresión del trabajo de parto, falla en la inducción del trabajo de parto, macrosomía, periodo intergenésico corto (menos de 18 meses). RESULTADOS: Se obtuvieron 569 pacientes de las que 228 finalizaron el embarazo mediante cesárea y 341 por parto, lo que representó una tasa de cesáreas del 39.9%. Al aplicar el método de clasificación de Robson, los grupos con mayor contribución relativa a la tasa global de cesáreas fueron: grupo 1 (17.62%), grupo 2, subdividido en sus dos categorías: 2a con 19.38% y 2b 17.18%; y el grupo 5.1 (22.91%) y grupo 5.2 (3.96%). Las principales indicaciones de cesárea fueron: 1) riesgo de pérdida del bienestar fetal (18.9%), 2) falta de progresión del trabajo de parto (16.7%), 3) falla en la inducción del trabajo de parto (11.1%), 4) macrosomía (7.2%) y 5) periodo intergenésico corto (7.2%). CONCLUSIONES: El índice de Robson señaló a los grupos 1, 2 y 5 como los mayores contribuyentes a la tasa de cesáreas en CIMIGen. Esta tendencia, grupos 1 y 2 con porcentajes elevados, también se observa en otros centros de atención en México, lo que pudiera indicar una práctica de atención obstétrica que debe revisarse. Esto también se ve en otros países, aunque los porcentajes son distintos pero siguen siendo considerables en los grupos 1, 2 y 5.
Abstract OBJECTIVE: To determine, according to the Robson Index, the caesarean section rate in patients attended, over a period of nine months, at the Maternal and Infant Research Centre of the Childbirth Studies Group. MATERIALS AND METHODS: Retrospective, cross-sectional, descriptive, single-centre study carried out in pregnant patients attended at the Maternal and Infant Research Centre of the Childbirth Study Group (CIMIGen) who terminated gestation by caesarean section and in whom the Robson index was used for classification. Study parameters: indications for caesarean section, risk of loss of fetal well-being, high labour progression, failure of labour induction, macrosomia, short inter-gestational period (less than 18 months). RESULTS: We obtained 569 patients of whom 228 terminated pregnancy by caesarean section and 341 by delivery, representing a caesarean section rate of 39.9%. Applying Robson's classification method, the groups with the highest relative contribution to the overall caesarean section rate were: group 1 (17.62%), group 2, subdivided into its two categories: 2a with 19.38% and 2b 17.18%; and group 5.1 (22.91%) and group 5.2 (3.96%). The main indications for caesarean section were 1) risk of loss of fetal well-being (18.9%), 2) failure of labour to progress (16.7%), 3) failure of induction of labour (11.1%), 4) macrosomia (7.2%) and 5) short inter-gestational period (7.2%). CONCLUSIONS: Robson's index pointed to groups 1, 2 and 5 as the largest contributors to the caesarean section rate at CIMIGen. This trend, groups 1 and 2 with high percentages, is also seen in other facilities in Mexico, which may indicate an obstetric care practice that needs to be reviewed. This is also seen in other countries, although the percentages are different but still considerable in groups 1, 2 and 5.
RESUMEN
OBJECTIVES: Normal thyroid activity has an essential role in fetal development, its deficiency may hamper fetal neurodevelopment and neonatal growth. The quantitation of thyroid hormones although useful, still exposes differences on cut off levels to diagnose thyroid deficit accurately that can elicit under or over diagnosis of thyroid dysfuntion. METHODS: A total of 839 pregnant patients were studied for thyroidal clinical assessment through quantitation of thyroid-stimulating hormone (TSH) and free thyroxine (FT4) circulating levels. Patients evaluated for prenatal and neonatal outcomes. Thyroid function deficiencies were determined with the American Thyroid Association (ATA) 2011 and 2017 values. Statistical analysis searched for associations between variables, odds ratios (OR) and correlations were calculated to evaluate the reliability of the cutoff values recommended by the ATA. RESULTS: Mean age of our cases was 27.5 + 5.83 years at diagnosis, mean gestational age at first consultation was 23.8 + 10.5 weeks. Mean TSH levels detected were: 2.5 + 1.89 mIU/L, total T3: 3.55 + 4.1 ng/dL, FT4: 3.14 + 4.4 ng/dL. The ATA 2011 values yielded 332 hypothyroidism cases vs. 507 euthyroid patients, a total incidence of 39.6% vs. the ATA 2017 values, diagnosing 100 hypothyroidism cases and 739 euthyroid patients, total incidence of 11.9%. Association with complications were not significant. CONCLUSIONS: Using ATA 2017 values showed a decreased population with gestational hypothyroidism, hence preventing overdiagnosis and over-treatment. No significant complications were associated, requiring the determination of new regional values. Education and sensibilization of our population is needed to comply with early prenatal consultation and thyroid function testing.