Rodolfo Carvalho Pacagnella

Objective To evaluate whether early postpartum bleeding velocity predicts standard (≥500 mL) and severe (≥1000 mL) postpartum hemorrhage (PPH) and to characterize temporal blood loss trajectories. Design Retrospective cohort study. Setting Tertiary maternity center in Brazil. Population or Sample A total of 266 women undergoing vaginal delivery. Methods Retrospective study of 266 vaginal deliveries with blood loss measured every 5 minutes for 120 minutes using calibrated drapes. Participants were grouped by 24-hour blood loss. Mixed-effects and GEE models assessed bleeding trajectories. Early bleeding velocity was evaluated using ROC analysis, with exploratory time-to-event analyses for progression to clinically significant thresholds. Main Outcome Measures Standard PPH (≥500 mL) and severe PPH (≥1000 mL). Results Early bleeding trajectories differed significantly by hemorrhage severity. Women with severe PPH (≥1000 mL) showed steeper, rapidly accelerating blood loss and higher early bleeding velocity (~24 mL/min vs 6.5 mL/min). Early velocity predicted standard PPH well (AUC 0.844) but less accurately severe PPH (AUC 0.836). Optimal thresholds were 9.8 mL/min (standard) and 12.5 mL/min (severe). Higher early velocity was associated with faster progression to critical blood loss. Conclusions Early postpartum bleeding velocity and trajectories identify women at risk of hemorrhage before conventional thresholds. Integrating dynamic monitoring into care may enable earlier intervention and reduce severe maternal outcomes.

Objective: This study assessed the efficacy of the cervical pessary combined with progesterone to prevent preterm birth in pregnant women with short cervix and previous preterm birth. Methods: This post hoc analysis of the randomized, multicenter P5 trial examined the efficacy of the cervical pessary associated with vaginal progesterone versus progesterone alone for preventing recurrent preterm birth in 155 pregnant women with cervical length ≤30 mm and prior spontaneous preterm birth (sPPTB) (main subgroup), and in 85 women with cervical length ≤25 mm and sPPTB (higher-risk population). The primary outcome was spontaneous preterm birth (sPTB) before 34 weeks; secondary outcomes included sPTB rates before 37, 32, and 28 weeks, analyzed using Odds Ratio (OR) and Kaplan–Meier curves. A secondary objective was to identify predictive factors for sPTB recurrence in the cohort with prior preterm birth (n = 479), irrespective of treatment allocation. Results: Demographic profiles were balanced between groups. The addition of a cervical pessary to progesterone did not result in a significant reduction in sPTB before 34 weeks: to cervix ≤30 mm, OR 1.169 (95% CI 0.524–2.609; p = 0.703) and 1.167 (95% CI 0.466–2.921; p = 0.742) for ≤25 mm; similar null findings were observed across all gestational age thresholds. Kaplan–Meier survival curves demonstrated no significant differences between groups (p > 0.05). Secondary analysis (n = 479) identified principal predictors of sPTB recurrence, regardless of the cervical length: higher education (OR 2.37; 95% CI 0.99–5.63; p = 0.024), previous cervical conization (OR 4.78; 95% CI 1.08–21.19; p = 0.039) previous low birth weight < 2.5 kg (OR 2.43; 95% CI 1.22–4.85; p = 0.051), prior miscarriages (OR 1.36; 95% CI 1.10–1.69; p = 0.005), current twin pregnancy (OR 14.86; 95% CI 4.35–50.68; p < 0.001) and cervical funneling (OR 3.60; 95% CI 1.79–7.24; p < 0.001). Predictive models achieved an AUC of 0.719, with 87.0% sensitivity and 58.8% specificity. Conclusions: These findings do not support the routine use of cervical pessary combined with progesterone in women with dual risk factors. In this Brazilian population, specific clinical and obstetric characteristics—including higher education, cervical funneling, prior low birth weight delivery, previous conization, current twin gestation, and prior miscarriage—could identify women at increased risk for recurrent preterm birth.

Objective: To analyze the knowledge, attitudes, and practices of obstetricians regarding childbirth care. Methods: This was a nationwide cross‐sectional study involving 1267 physicians working in 393 public and private hospitals in Brazil. A self‐administered and anonymous questionnaire, developed specifically for this study, was used. Differences in proportions among obstetricians working in public, mixed, and private services were analyzed using the chi‐square test, with a significance level of 0.05. Results: High levels of knowledge, favorable attitudes, and adoption of best practices in childbirth care were observed for diet, movement, the presence of a companion, and non‐pharmacological methods for pain relief. Limitations were identified regarding the participation of nurse‐midwives in childbirth care, the use of instrumental delivery, promotion of upright birth positions, and the use of episiotomy. Fear of legal action and unnecessary indications for cesarean sections were frequently reported by professionals as reasons for the country's high cesarean section rates, and should be addressed in medical schools and specialization courses. A more positive perception of cesarean delivery as safe, and the defense of cesarean section as a woman's right, were more common in private services, where there was also less acceptance of births attended by nurse‐midwives. Conclusion: Differences observed between public and private services may inform the development of new strategies and strengthen ongoing initiatives in Brazil, with the aim of reducing cesarean section rates and promoting best practices in childbirth care.

Objective To develop and externally validate predictive models for early detection of postpartum hemorrhage by combining clinical, hemodynamic, and temporal blood loss indicators. Design Multicenter retrospective model development and external validation study. Setting A Brazilian tertiary referral hospital and pooled clinical trial and observational datasets from low- and middle-income countries. Population or Sample Model training used data from a Brazilian tertiary hospital cohort (n = 270) and external validation employed pooled datasets from randomized and observational studies (n = 2,379). Methods Predictive models were developed using maternal demographics, delivery characteristics, pre-delivery hemoglobin, blood loss at 30 minutes postpartum, and bleeding velocity. Statistical and machine learning approaches including logistic regression, random forest, and XGBoost were evaluated. Model discrimination and performance were assessed using area under the receiver operating characteristic curve (AUC), F1-score, sensitivity, and additional validation metrics. Main Outcome Measures Binary; occurrence of postpartum hemorrhage defined as blood loss ≥500 mL (yes/no), along with model discrimination, calibration, and clinical predictive performance. Results Early bleeding velocity and blood loss at 30 minutes were the strongest predictors of PPH. Generalized linear models achieved high discrimination (internal AUC ~ 0.98) and strong agreement. Machine learning models performed well but did not consistently outperform logistic regression. In external validation, simpler regression-based models demonstrated equal or superior discrimination (AUCs up to 0.996) and more stable calibration. Reduced predictor models retained good performance (external AUC ~0.89–0.94), while expanded models maintained robust performance (AUC ~0.92–0.98). Conclusions Integration of early hemodynamic and temporal bleeding measurements could substantially improve early prediction of postpartum hemorrhage. These findings support scalable, data-driven tools for timely intervention and reduction of maternal morbidity.

Preterm birth (PTB) is the main cause of perinatal and neonatal morbidity and mortality worldwide. Widespread implementation of guidelines for early identification and management of patients at risk for adverse pregnancy outcomes is still feeble. This work aims to implement a simple and low-cost bundle to access and manage major modifiable risk factors for PTB. We included first trimester pregnant women seen at Health Units from Jaú - SP, Brazil, where PTB prevalence ranges from 13,4% to 18%. The protocol is based on three aspects: a questionnaire to access smoking status and clinical history; Gram staining and of vaginal microbiota evaluation using Nugent's criteria and evaluation of cervical infections; and transvaginal ultrasound. Pregnant women who smoke and are willing to quit will be treated with auricular acupuncture and referred to a support group if necessary. All patients will be advised on intimate hygiene habits, and those with dysbiosis will be treated. Cervical length will be accessed using transvaginal ultrasound, and those diagnosed with cervical shortening will be treated with vaginal progesterone. This study highlights the importance of implementing active measures to reduce PTB risk factors in a high prevalence setting.

A integração de bases de dados é um desafio comum em diversas aplicações de ciência de dados. Frequentemente, a vinculação é realizada por meio de campos de texto livre, como nomes de pessoas, que costumam apresentar inconsistências devido a erros de grafia, abreviações e outras variações. Essas inconsistências podem resultar na representação de um mesmo indivíduo por diferentes registros, dificultando a identificação e análise precisa das informações. Dada a relevância desse problema, várias técnicas têm sido propostas para a vinculação de bases de dados. No entanto, esses estudos têm alto custo, em virtude da necessidade de validação manual qualificada. Este estudo tem como objetivo principal investigar a viabilidade de métodos automáticos de vinculação de dados aplicados no idioma português brasileiro. Os resultados da vinculação automática foram validados manualmente por uma equipe de especialistas, os quais atestaram a viabilidade técnica. Todas as alternativas testadas obtiveram um índice F1-Score superior a 0,90.

Objective This study aims to create a new screening for preterm birth < 34 weeks after gestation with a cervical length (CL) ≤ 30 mm, based on clinical, demographic, and sonographic characteristics. Methods This is a post hoc analysis of a randomized clinical trial (RCT), which included pregnancies, in middle-gestation, screened with transvaginal ultrasound. […] Abstract Objective This study aims to create a new screening for preterm birth < 34 weeks after gestation with a cervical length (CL) ≤ 30 mm, based on clinical, demographic, and sonographic characteristics. Methods This is a post hoc analysis of a randomized clinical trial (RCT), which included pregnancies, in middle-gestation, screened with transvaginal ultrasound. […]

Introduction: To develop and validate a support tool for healthcare providers, enabling them to make precise and critical decisions regarding intensive care unit (ICU) admissions for high‐risk pregnant women, thus enhancing maternal outcomes.Methods: This retrospective study involves secondary data analysis of information gathered from 9550 pregnant women, who had severe maternal morbidity (any unexpected complication during labor and delivery that leads to substantial short‐term or long‐term health issues for the mother), collected between 2009 and 2010 from the Brazilian Network for Surveillance of Severe Maternal Morbidity, encompassing 27 obstetric reference centers in Brazil. Machine‐learning models, including decision trees, Random Forest, Gradient Boosting Machine (GBM), and Extreme Gradient Boosting (XGBoost), were employed to create a risk prediction tool for ICU admission. Subsequently, sensitivity analysis was conducted to compare the accuracy, predictive power, sensitivity, and specificity of these models, with differences analyzed using the Wilcoxon test.Results: The XGBoost algorithm demonstrated superior efficiency, achieving an accuracy rate of 85%, sensitivity of 42%, specificity of 97%, and an area under the receiver operating characteristic curve of 86.7%. Notably, the estimated prevalence of ICU utilization by the model (11.6%) differed from the prevalence of ICU use from the study (21.52%).Conclusion: The developed risk engine yielded positive results, emphasizing the need to optimize intensive care bed utilization and objectively identify high‐risk pregnant women requiring these services. This approach promises to enhance the effective and efficient management of pregnant women, particularly in resource‐constrained regions worldwide. By streamlining ICU admissions for high‐risk cases, healthcare providers can better allocate critical resources, ultimately contributing to improved maternal health outcomes.

Highlights Pregnant women with increased weight gain must be surveilled for hypertension. Non-white and older women are more likely to develop hypertension. Maternal conditions determine most cases of preterm delivery among women with hypertension. ABSTRACT Objective: This study aimed to evaluate the prevalence of hypertensive disorders during pregnancy among Brazilian women with preterm births and to compare the epidemiological characteristics and perinatal outcomes among preterm births of women with and without hypertension. Methods: This was a secondary cross-sectional analysis of

Brazil presents high maternal and perinatal morbidity and mortality. Cases of severe maternal morbidity, maternal near miss, and perinatal deaths are important health indicators and share the same determinants, being closely related to living conditions and quality of perinatal care. This article aims to present the study protocol to estimate the perinatal mortality rate and the incidence of severe maternal morbidity and maternal near miss in the country, identifying its determinants. Cross-sectional study integrated into the research Birth in Brazil II, conducted from 2021 to 2023. This study will include 155 public, mixed and private maternities, accounting for more than 2,750 births per year, participating in the Birth in Brazil II survey. We will collect retrospective data from maternal and neonatal records of all hospitalizations within a 30-day period in these maternities, applying a screening form to identify cases of maternal morbidity and perinatal deaths. Medical record data of all identified cases will be collected after hospital discharge, using a standardized instrument. Cases of severe maternal morbidity and maternal near miss will be classified based on the definition adopted by the World Health Organization. The perinatal deaths rate and the incidence of severe maternal morbidity and maternal near miss will be estimated. Cases will be compared to controls obtained in the Birth in Brazil II survey, matched by hospital and duration of pregnancy, in order to identify factors associated with negative outcomes. Results are expected to contribute to the knowledge on maternal morbidity and perinatal deaths in Brazil, as well as the development of strategies to improve care.