Machine Learning for the Prediction of Surgical Morbidity in Placenta Accreta Spectrum.

OBJECTIVE: We sought to create a machine learning (ML) model to identify variables that would aid in the prediction of surgical morbidity in cases of placenta accreta spectrum (PAS). STUDY DESIGN: A multicenter analysis including all cases of PAS identified by pathology specimen confirmation, across five tertiary care perinatal centers in New York City from 2013 to 2022. We developed models to predict operative morbidity using 213 variables including demographics, obstetrical information, and limited prenatal imaging findings detailing placental location. Our primary outcome was prediction of a surgical morbidity composite defined as including any of the following: blood loss (>1,500 mL), transfusion, intensive care unit admission, vasopressor use, mechanical ventilation/intubation, and organ injury. A nested, stratified, cross-validation approach was used to tune model hyperparameters and estimate generalizability. Gradient boosted tree classifier models incorporated preprocessing steps of standard scaling for numerical variables and one-hot encoding for categorical variables. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), positive and negative predictive values (PPV, NPV), and F1 score. Variable importance ranking was also determined. RESULTS: Among 401 PAS cases, 326 (81%) underwent hysterectomy. Of the 401 cases of PAS, 309 (77%) had at least one event defined as surgical morbidity. Our predictive model had an AUC of 0.79 (95% confidence interval: 0.69, 0.89), PPV 0.79, NPV 0.76, and F1 score of 0.88. The variables most predictive of surgical morbidity were completion of a hysterectomy, prepregnancy body mass index (BMI), absence of a second trimester ultrasound, socioeconomic status zip code, BMI at delivery, number of prenatal visits, and delivery time of day. CONCLUSION: By identifying social and obstetrical characteristics that increase patients' risk, ML models are useful in predicting PAS-related surgical morbidity. Utilizing ML could serve as a foundation for risk and complexity stratification in cases of PAS to optimize surgical planning. KEY POINTS: · ML models are useful models are useful in predicting PAS-related surgical morbidity.. · Optimal management for PAS remains unclear.. · Utilizing ML can serve as a foundation for risk and complexity stratification in cases of PAS..

One-step vs 2-step gestational diabetes mellitus screening and pregnancy outcomes: an updated systematic review and meta-analysis.

OBJECTIVE: This was a systematic review and meta-analysis comparing maternal and neonatal outcomes of patients screened with the 1-step or 2-step screening method for gestational diabetes mellitus. DATA SOURCES: PubMed, Scopus, Cochrane, ClinicalTrials.gov, and LILACS were searched from inception up to September 2022. STUDY ELIGIBILITY CRITERIA: Only randomized controlled trials were included. Studies that had overlapping populations were excluded (International Prospective Register of Systematic Review registration number: CRD42022358903). METHODS: Risk ratios were computed with 95% confidence intervals by 2 authors. Unpublished data were requested. Large for gestational age was the primary outcome. RESULTS: The search yielded 394 citations. Moreover, 7 randomized controlled trials met the inclusion criteria. A total of 54,650 participants were screened for gestational diabetes mellitus by either the 1-step screening method (n=27,163) or the 2-step screening method (n=27,487). For large for gestational age, there was no significant difference found between the groups (risk ratio, 0.99; 95% confidence interval, 0.93-1.05; I2=0%). Newborns of patients who underwent 1-step screening had higher rates of neonatal hypoglycemia (risk ratio, 1.24; 95% confidence interval, 1.14-1.34; I2=0%) and neonatal intensive care unit admissions (risk ratio, 1.13; 95% confidence interval, 1.04-1.21; I2=0%) than newborns of patients who underwent 2-step screening. Patients in the 1-step screening method group were more likely to be diagnosed with gestational diabetes mellitus (risk ratio, 1.73; 95% confidence interval, 1.44-2.09; I2=80%) than patients in the 2-step screening method group. In addition, among trials that tested all patients before randomization and excluded patients with pregestational diabetes mellitus, newborns were more likely to have macrosomia (risk ratio, 1.27; 95% confidence interval, 1.21-1.34; I2=0%). Overall risk of bias assessment was of low concern. CONCLUSION: Large for gestational age did not differ between patients screened using the 1-step screening method and those screened using the 2-step screening method. However, patients randomized to the 1-step screening method had higher rates of neonatal hypoglycemia and neonatal intensive care unit admission and maternal gestational diabetes mellitus diagnosis than the patients randomized to the 2-step screening method.

Medical evidence in asylum applications: Medical versus legal approaches.

The purpose of this narrative review is to elucidate the ways the clinicians working on forensic medical evaluations can engage with asylum proceedings. We compare the legal and medical perspectives on different aspects of forensic medical evidence, asylum evaluations, and asylum applications. As asylum seekers must demonstrate a well-founded fear of persecution in order to receive asylee status, legal and medical professionals often need to collaborate in asylum cases. Although significant evidence has demonstrated that an objective expert medical opinion can support asylum claims, few studies have analyzed how the medical professional's role complements or is at odds with the goals of the legal system. This review summarizes and compares key aspects of the medical and legal perspectives on trauma, credibility, autobiographical memory, and medical evidence to better comprehend the role that medical professionals can play in writing medical affidavits for asylum applications. We dissect legal misconceptions surrounding trauma and the consequences of such misunderstandings and make recommendations for medical evaluators who are working in a forensic capacity.