Harnessing the open access version of ChatGPT for enhanced clinical opinions.

With the advent of Large Language Models (LLMs) like ChatGPT, the integration of Generative Artificial Intelligence (GAI) into clinical medicine is becoming increasingly feasible. This study aimed to evaluate the ability of the freely available ChatGPT-3.5 to generate complex differential diagnoses, comparing its output to case records of the Massachusetts General Hospital published in the New England Journal of Medicine (NEJM). Forty case records were presented to ChatGPT-3.5, prompting it to provide a differential diagnosis and then narrow it down to the most likely diagnosis. The results indicated that the final diagnosis was included in ChatGPT-3.5's original differential list in 42.5% of the cases. After narrowing, ChatGPT correctly determined the final diagnosis in 27.5% of the cases, demonstrating a decrease in accuracy compared to previous studies using common chief complaints. These findings emphasize the necessity for further investigation into the capabilities and limitations of LLMs in clinical scenarios while highlighting the potential role of GAI as an augmented clinical opinion. Anticipating the growth and enhancement of GAI tools like ChatGPT, physicians and other healthcare workers will likely find increasing support in generating differential diagnoses. However, continued exploration and regulation are essential to ensure the safe and effective integration of GAI into healthcare practice. Future studies may seek to compare newer versions of ChatGPT or investigate patient outcomes with physicians integrating this GAI technology. Understanding and expanding GAI's capabilities, particularly in differential diagnosis, may foster innovation and provide additional resources, especially in underserved areas in the medical field.

ChatGPT: a pioneering approach to complex prenatal differential diagnosis.

This commentary examines how ChatGPT can assist healthcare teams in the prenatal diagnosis of rare and complex cases by creating a differential diagnoses based on deidentified clinical findings, while also acknowledging its limitations.

Chat Generative Pre-trained Transformer: why we should embrace this technology.

With the advent of artificial intelligence that not only can learn from us but also can communicate with us in plain language, humans are embarking on a brave new future. The interaction between humans and artificial intelligence has never been so widespread. Chat Generative Pre-trained Transformer is an artificial intelligence resource that has potential uses in the practice of medicine. As clinicians, we have the opportunity to help guide and develop new ways to use this powerful tool. Optimal use of any tool requires a certain level of comfort. This is best achieved by appreciating its power and limitations. Being part of the process is crucial in maximizing its use in our field. This clinical opinion demonstrates the potential uses of Chat Generative Pre-trained Transformer for obstetrician-gynecologists and encourages readers to serve as the driving force behind this resource.

Remote patient monitoring for management of diabetes mellitus in pregnancy is associated with improved maternal and neonatal outcomes.

BACKGROUND: Diabetes mellitus is a common medical complication of pregnancy, and its treatment is complex. Recent years have seen an increase in the application of mobile health tools and advanced technologies, such as remote patient monitoring, with the aim of improving care for diabetes mellitus in pregnancy. Previous studies of these technologies for the treatment of diabetes in pregnancy have been small and have not clearly shown clinical benefit with implementation. OBJECTIVE: Remote patient monitoring allows clinicians to monitor patients' health data (such as glucose values) in near real-time, between office visits, to make timely adjustments to care. Our objective was to determine if using remote patient monitoring for the management of diabetes in pregnancy leads to an improvement in maternal and neonatal outcomes. STUDY DESIGN: This was a retrospective cohort study of pregnant patients with diabetes mellitus managed by the maternal-fetal medicine practice at one academic institution between October 2019 and April 2021. This practice transitioned from paper-based blood glucose logs to remote patient monitoring in February 2020. Remote patient monitoring options included (1) device integration with Bluetooth glucometers that automatically uploaded measured glucose values to the patient's Epic MyChart application or (2) manual entry in which patients manually logged their glucose readings into their MyChart application. Values in the MyChart application directly transferred to the patient's electronic health record for review and management by clinicians. In total, 533 patients were studied. We compared 173 patients managed with paper logs to 360 patients managed with remote patient monitoring (176 device integration and 184 manual entry). Our primary outcomes were composite maternal morbidity (which included third- and fourth-degree lacerations, chorioamnionitis, postpartum hemorrhage requiring transfusion, postpartum hysterectomy, wound infection or separation, venous thromboembolism, and maternal admission to the intensive care unit) and composite neonatal morbidity (which included umbilical cord pH <7.00, 5 minute Apgar score <7, respiratory morbidity, hyperbilirubinemia, meconium aspiration, intraventricular hemorrhage, necrotizing enterocolitis, sepsis, pneumonia, seizures, hypoxic ischemic encephalopathy, shoulder dystocia, trauma, brain or body cooling, and neonatal intensive care unit admission). Secondary outcomes were measures of glycemic control and the individual components of the primary composite outcomes. We also performed a secondary analysis in which the patients who used the two different remote patient monitoring options (device integration vs manual entry) were compared. Chi-square, Fisher's exact, 2-sample t, and Mann-Whitney tests were used to compare the groups. A result was considered statistically significant at P<.05. RESULTS: Maternal baseline characteristics were not significantly different between the remote patient monitoring and paper groups aside from a slightly higher baseline rate of chronic hypertension in the remote patient monitoring group (6.1% vs 1.2%; P=.011). The primary outcomes of composite maternal and composite neonatal morbidity were not significantly different between the groups. However, remote patient monitoring patients submitted more glucose values (177 vs 146; P=.008), were more likely to achieve glycemic control in target range (79.2% vs 52.0%; P<.0001), and achieved the target range sooner (median, 3.3 vs 4.1 weeks; P=.025) than patients managed with paper logs. This was achieved without increasing in-person visits. Remote patient monitoring patients had lower rates of preeclampsia (5.8% vs 15.0%; P=.0006) and their infants had lower rates of neonatal hypoglycemia in the first 24 hours of life (29.8% vs 51.7%; P<.0001). CONCLUSION: Remote patient monitoring for the management of diabetes mellitus in pregnancy is superior to a traditional paper-based approach in achieving glycemic control and is associated with improved maternal and neonatal outcomes.

Vasa previa: avoiding incising the membranes at cesarean delivery.

We present our technique for cesarean delivery of prenatally diagnosed vasa previa in which we avoid incising the membranes and fetal vessels. This technique allows direct visualization of the fetal blood vessels and may prevent blood loss from the baby at the time of birth.

Performance of a Multianalyte 'Rule-Out' Assay in Pregnant Individuals With Suspected Preeclampsia.

BACKGROUND: The ability to diagnose preeclampsia clinically is suboptimal. Our objective was to validate a novel multianalyte assay and characterize its performance, when intended for use as an aid to rule-out preeclampsia. METHODS: Prospective, multicenter cohort study of pregnant individuals presenting between 280/7 and 366/7 weeks' with preeclampsia-associated signs and symptoms. Individuals not diagnosed with preeclampsia after baseline evaluation were enrolled in the study cohort, with those who later developed preeclampsia, classified as cases and compared with a negative control group who did not develop preeclampsia. Individuals with assay values at time of enrollment ≥0.0325, determined using a previously developed algorithm, considered at risk. The primary analysis was the time to develop preeclampsia assessed using a multivariate Cox regression model. RESULTS: One thousand thirty-six pregnant individuals were enrolled in the study cohort with an incidence of preeclampsia of 30.3% (27.6%-33.2%). The time to develop preeclampsia was shorter for those with an at-risk compared with negative assay result (log-rank P<0.0001; adjusted hazard ratio of 4.81 [3.69-6.27, P<0.0001]). The performance metrics for the assay to rule-out preeclampsia within 7 days of enrollment showed a sensitivity 76.4% (67.5%-83.5%), negative predictive value 95.0% (92.8%-96.6%), and negative likelihood ratio 0.46 (0.32-0.65). Assay performance improved if delivery occurred <37 weeks and for individuals enrolled between 28 and 35 weeks. CONCLUSIONS: We confirmed that a novel multianalyte assay was associated with the time to develop preeclampsia and has a moderate sensitivity and negative likelihood ratio but high negative predictive value when assessed as an aid to rule out preeclampsia within 7 days of enrollment. REGISTRATION: The study was registered on Clinicaltrials.gov (Identifier NCT02780414).

Use of cervical elastography at 18 to 22 weeks' gestation in the prediction of spontaneous preterm birth.

BACKGROUND: Accurate identification of the women who will have spontaneous preterm birth continues to be a great challenge. The use of cervical elastography for prediction of preterm birth is promising, but several limitations exist. Newer cervical elastography technology has been developed that may prove useful in evaluation of risk of preterm birth. OBJECTIVE: This study aimed to develop standard cervical elastography nomograms for singleton pregnancies at 18 to 22 weeks' gestation using the E-Cervix ultrasound application, assess intraobserver reliability of the E-Cervix elastography parameters, and determine whether these cervical elastography measurements can be used in the prediction of spontaneous preterm birth. STUDY DESIGN: This was a prospective cohort study of pregnant women undergoing cervical length screening assessment via transvaginal ultrasound examination at 18 to 22 weeks' gestation. A semiautomatic, cervical elastography application (E-Cervix) was used during the transvaginal examination to calculate 5 quantitative parameters (internal os stiffness, external os stiffness, internal -to -external os stiffness ratio, hardness ratio, and elasticity contrast index) and create a standard nomogram for each one of them. The intraobserver reliability was calculated using Shrout-Fleiss reliability. Cervical elastography parameters were compared between those who delivered preterm (<37 weeks) spontaneously and those who delivered full term. A multivariable logistic regression model was performed to determine the ability of the cervical elastography parameters to predict spontaneous preterm birth. RESULTS: A total of 742 women were included, of which 49 (6.6%) had a spontaneous preterm delivery. A standard nomogram was created for each of the cervical elastography parameters from those who had a full-term birth in the index pregnancy (n=693). Intraobserver reliability was good or excellent (intraclass correlation, 0.757-0.887) for each of the cervical elastography parameters except external os stiffness which was poor (intraclass correlation, 0.441). In univariate analysis, none of the cervical elastography parameters were associated with a statistically significant increased risk of spontaneous preterm birth. In a multivariable model adjusting for history of preterm birth, gravidity, ethnicity, cervical cerclage, and vaginal progesterone use, increasing elasticity contrast index was significantly associated with an increased risk of spontaneous preterm birth (odds ratio, 1.15; 95% confidence interval, 1.02-1.30; P=.02). CONCLUSION: Cervical elastography parameters are reliably measured and are stable across 18 to 22 weeks' gestation. Based on our findings, the elasticity contrast index was associated with an increased risk of spontaneous preterm birth and may be a useful parameter for future research.

Coronavirus disease 2019 infection and placental histopathology in women delivering at term.

BACKGROUND: There is a paucity of data describing the effects of coronavirus disease 2019 on placental pathology, especially in asymptomatic patients. Although the pathophysiology of coronavirus disease 2019 is not completely understood, there is emerging evidence that it causes a severe systemic inflammatory response and results in a hypercoagulable state with widespread microthrombi. We hypothesized that it is plausible that a similar disease process may occur in the fetal-maternal unit. OBJECTIVE: This study aimed to determine whether coronavirus disease 2019 in term patients admitted to labor and delivery, including women without coronavirus disease 2019 symptomatology, is associated with increased placental injury compared with a cohort of coronavirus disease 2019-negative controls. STUDY DESIGN: This was a retrospective cohort study performed at NYU Winthrop Hospital between March 31, 2020, and June 17, 2020. During the study period, all women admitted to labor and delivery were routinely tested for severe acute respiratory syndrome coronavirus 2 regardless of symptomatology. The placental histopathologic findings of patients with coronavirus disease 2019 (n=77) who delivered a singleton gestation at term were compared with a control group of term patients without coronavirus disease 2019 (n=56). Controls were excluded if they had obstetrical or medical complications including fetal growth restriction, oligohydramnios, hypertension, diabetes, coagulopathy, or thrombophilia. Multivariable logistic regression models were performed for variables that were significant (P<.05) in univariable analyses. A subgroup analysis was also performed comparing asymptomatic coronavirus disease 2019 cases with negative controls. RESULTS: In univariable analyses, coronavirus disease 2019 cases were more likely to have evidence of fetal vascular malperfusion, that is, presence of avascular villi and mural fibrin deposition (32.5% [25/77] vs 3.6% [2/56], P<.0001) and villitis of unknown etiology (20.8% [16/77] vs 7.1% [4/56], P=.030). These findings persisted in a subgroup analysis of asymptomatic coronavirus disease 2019 cases compared with coronavirus disease 2019-negative controls. In a multivariable model adjusting for maternal age, race and ethnicity, mode of delivery, preeclampsia, fetal growth restriction, and oligohydramnios, the frequency of fetal vascular malperfusion abnormalities remained significantly higher in the coronavirus disease 2019 group (odds ratio, 12.63; 95% confidence interval, 2.40-66.40). Although the frequency of villitis of unknown etiology was more than double in coronavirus disease 2019 cases compared with controls, this did not reach statistical significance in a similar multivariable model (odds ratio, 2.11; 95% confidence interval, 0.50-8.97). All neonates of mothers with coronavirus disease 2019 tested negative for severe acute respiratory syndrome coronavirus 2 by polymerase chain reaction. CONCLUSION: Despite the fact that all neonates born to mothers with coronavirus disease 2019 were negative for severe acute respiratory syndrome coronavirus 2 by polymerase chain reaction, we found that coronavirus disease 2019 in term patients admitted to labor and delivery is associated with increased rates of placental histopathologic abnormalities, particularly fetal vascular malperfusion and villitis of unknown etiology. These findings seem to occur even among asymptomatic term patients.

Second trimester marginal cord insertion is associated with adverse perinatal outcomes.

Objectives: To determine the feasibility in visualizing placental cord insertion (PCI) during second-trimester fetal anatomical survey and the association between marginal cord insertion (MCI) and preterm delivery (PTD) and low birth weight (LBW). Our secondary objectives were to evaluate the association of MCI with adverse composite obstetrical and neonatal outcomes. Methods: A prospective cohort study was performed over a 28-month period. Women with singleton pregnancies presenting for routine anatomical survey between 18 and 22 weeks' gestation were included. PCI site was visualized on 2D grayscale and color Doppler and the shortest distance from the sagittal and transverse planes to the placental edge were recorded. MCI was diagnosed when any of measured distances was ≤2 cm. Correlations were assessed via bivariate chi-squared, independent t-test analyses and Fisher's exact tests. Regression models evaluated associations between MCI and adverse composite outcomes. Results: Three hundred one women were included and PCI was feasible in all cases. The incidence of MCI was 11.3% (n = 34). Baseline characteristics between those with and without MCI were similar, except for story of prior PTD, which was greater among those with MCI (17.65 versus 7.17%, p = .04). MCI was associated with increased likelihood of LBW (RR four; 95%CI, 1.46-10.99) and PTD (RR 3.2; 95%CI, 1.53-6.68); in multivariate analysis, we found associations between MCI and composite adverse obstetrical (RR 2.33; 95%CI, 1.30-4.19) and neonatal (RR 2.46; 95%CI, 1.26-4.81) outcomes. Conclusions: Evaluation of PCI is feasible in all cases. Second-trimester MCI is associated with increased likelihood for LBW, PTD, and composite adverse obstetrical and neonatal outcomes.

Improving the ultrasound detection of isolated fetal limb abnormalities.

OBJECTIVE: The prenatal detection rate of isolated fetal limb abnormalities ranges from 4 to 29.5%. Our aim was to determine the accuracy of a detailed ultrasound protocol in detecting isolated fetal limb abnormalities Methods: This is a retrospective study of infants born at our institution with isolated limb defects from 2009 to 2014. Antepartum and postpartum records were reviewed for genetic testing results. We routinely image both upper and lower extremities, including all long bones, hands, feet, fingers and toes. Posturing, muscular tone and movement are also noted. RESULTS: During the study period, there were 52 neonates born with isolated fetal limb abnormalities who had received a fetal anatomic survey in our ultrasound unit and 15 930 sonograms performed with normal findings; 36 out of the 52 had been prenatally diagnosed (detection rate 69%). The specificity of the protocol was 100% as there were no false positive cases, the positive predictive value was 100% and negative predictive value 99.8%. Forty-three of 52 neonates had normal genetic testing either prenatally or postnatally; 9 neonates did not undergo genetic testing. The average additional time required for this detailed protocol was <5 min for second trimester sonogram. CONCLUSION: A minimal investment in time for detailed evaluation of fetal limbs more than doubles the previously reported prenatal detection rate.

The use of cervical sonography to differentiate true from false labor in term patients presenting for labor check.

BACKGROUND: Cervical length by transvaginal ultrasound to predict preterm labor is widely used in clinical practice. Virtually no data exist on cervical length measurement to differentiate true from false labor in term patients who present for labor check. False-positive diagnosis of true labor at term may lead to unnecessary hospital admissions, obstetrical interventions, resource utilization, and cost. OBJECTIVE: We sought to determine if cervical length by transvaginal ultrasound can differentiate true from false labor in term patients presenting for labor check. STUDY DESIGN: This is a prospective observational study of women presenting to labor and delivery with labor symptoms at 37-42 weeks, singleton cephalic gestation, regular uterine contractions (≥4/20 min), intact membranes, and cervix ≤4 cm dilated and ≤80% effaced. Those patients with placenta previa and indications for immediate delivery were excluded. The shortest best cervical length of 3 collected images was used for analysis. Providers managing labor were blinded to the cervical length. True labor was defined as spontaneous rupture of membranes or spontaneous cervical dilation ≥4 cm and ≥80% effaced within 24 hours of cervical length measurement. In the absence of these outcomes, labor status was determined as false labor. Receiver operating characteristic curves were generated to assess the predictive ability of cervical length to differentiate true from false labor and were analyzed separately for primiparous and multiparous patients. The diagnostic accuracies of various cervical length cutoffs were determined. The relationship of cervical length and time to delivery was also analyzed including both use and nonuse of oxytocin. RESULTS: In all, 77 patients were included in the study; the prevalence of true labor was 58.4% (45/77). Patients who were in true labor had shorter cervical length as compared to those in false labor: median 1.3 cm (range 0.5-4.1) vs 2.4 cm (range 1.0-5.0), respectively (P < .001). The area under the receiver operating characteristic curve for primiparous patients was 0.88 (P < .001) and for multiparous patients was 0.76 (P < .01), both demonstrating good correlation. The area under the receiver operating characteristic curves were not significantly different between primiparous and multiparous (P = .23). The area under the receiver operating characteristic curve for primiparous and multiparous patients combined was 0.8 (P < .0001), indicating a good overall correlation between cervical length and its ability to differentiate true from false labor. Overall, a cervical length cutoff of ≤1.5 cm to predict true labor had the highest specificity (81%), positive predictive value (83%), and positive likelihood ratio (4.2). There were no differences in cervical length prediction between primiparous and multiparous patients. Cervical length was positively correlated with time to delivery, regardless of the use of oxytocin. CONCLUSION: In differentiating true from false labor in term patients who present for labor check, a cervical length of ≤1.5 cm was the most clinically optimal cutoff with the lowest false positive rate-due to its highest specificity-and highest positive predictive value and positive likelihood ratios. Its use to decide admission in patients at term with labor symptoms may prevent unnecessary admissions, obstetrical interventions, resource utilization, and cost.