Development and validation of a machine-learning model for prediction of shoulder dystocia.

OBJECTIVES: To develop a machine-learning (ML) model for prediction of shoulder dystocia (ShD) and to externally validate the model's predictive accuracy and potential clinical efficacy in optimizing the use of Cesarean delivery in the context of suspected macrosomia. METHODS: We used electronic health records (EHR) from the Sheba Medical Center in Israel to develop the model (derivation cohort) and EHR from the University of California San Francisco Medical Center to validate the model's accuracy and clinical efficacy (validation cohort). Subsequent to application of inclusion and exclusion criteria, the derivation cohort included 686 singleton vaginal deliveries, of which 131 were complicated by ShD, and the validation cohort included 2584 deliveries, of which 31 were complicated by ShD. For each of these deliveries, we collected maternal and neonatal delivery outcomes coupled with maternal demographics, obstetric clinical data and sonographic fetal biometry. Biometric measurements and their derived estimated fetal weight were adjusted (aEFW) according to gestational age at delivery. A ML pipeline was utilized to develop the model. RESULTS: In the derivation cohort, the ML model provided significantly better prediction than did the current clinical paradigm based on fetal weight and maternal diabetes: using nested cross-validation, the area under the receiver-operating-characteristics curve (AUC) of the model was 0.793 ± 0.041, outperforming aEFW combined with diabetes (AUC = 0.745 ± 0.044, P = 1e-16 ). The following risk modifiers had a positive beta that was > 0.02, i.e. they increased the risk of ShD: aEFW (beta = 0.164), pregestational diabetes (beta = 0.047), prior ShD (beta = 0.04), female fetal sex (beta = 0.04) and adjusted abdominal circumference (beta = 0.03). The following risk modifiers had a negative beta that was < -0.02, i.e. they were protective of ShD: adjusted biparietal diameter (beta = -0.08) and maternal height (beta = -0.03). In the validation cohort, the model outperformed aEFW combined with diabetes (AUC = 0.866 vs 0.784, P = 0.00007). Additionally, in the validation cohort, among the subgroup of 273 women carrying a fetus with aEFW ≥ 4000 g, the aEFW had no predictive power (AUC = 0.548), and the model performed significantly better (0.775, P = 0.0002). A risk-score threshold of 0.5 stratified 42.9% of deliveries to the high-risk group, which included 90.9% of ShD cases and all cases accompanied by maternal or newborn complications. A more specific threshold of 0.7 stratified only 27.5% of the deliveries to the high-risk group, which included 63.6% of ShD cases and all those accompanied by newborn complications. CONCLUSION: We developed a ML model for prediction of ShD and, in a different cohort, externally validated its performance. The model predicted ShD better than did estimated fetal weight either alone or combined with maternal diabetes, and was able to stratify the risk of ShD and neonatal injury in the context of suspected macrosomia. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.

Elective repeat cesarean delivery in women eligible for trial of labor in Brazil.

OBJECTIVE: To assess the hospital, maternal, and obstetric characteristics associated with elective repeat cesarean delivery (ERCD) among women eligible for trial of labor after cesarean (TOLAC) delivery in Brazil. METHODS: The present data were retrieved from the Birth in Brazil study, a national hospital-based cohort study conducted during 2011-2012. Data were collected from medical records and by interview. Univariate and hierarchical multiple logistic regression analyses were performed to analyze factors associated with ERCD among women with a previous cesarean delivery who were eligible for TOLAC. RESULTS: Among 2295 women considered eligible for TOLAC, 1516 (66.1%) had an ERCD; the overall cesarean delivery rate was 79.4%. In the private sector, almost all deliveries (95.3%) were performed by ERCD. In the public sector, ERCD was associated with socioeconomic (more years of schooling), obstetric (women's preference, no previous vaginal delivery, macrosomia), and hospital (mixed hospital, location in noncapital city, fewer than 1500 deliveries per year) characteristics. CONCLUSION: The ERCD rate in Brazil was high even in a low-risk group, indicating that nonclinical factors may be driving the decision for cesarean delivery. Efforts aiming to reduce cesarean deliveries in Brazil should target women with a previous cesarean delivery.

Primary cesarean delivery rate: Potential impact of a checklist.

BACKGROUND: Cesarean section is the most common surgical procedure performed in developed countries. Its incidence is increasing to a worrisome extent. The 2003 French National Perinatal Survey showed that the inflation in the overall cesarean rate was mainly due to an increase in the first cesarean delivery rate. OBJECTIVE: To evaluate a new tool: a checklist that intent to decrease the first cesarean delivery rate. STUDY DESIGN: Retrospective, observational, multi-center study. A new tool, a "First cesarean delivery" checklist was built according American and French guidelines. Women with full-term of pregnancy, nulliparous or multiparous with a first caesarean delivery including arrest of labor, breech presentation or suspected fetal macrosomia were included. The checklist was applied. Potentially preventable cesareans were analyzed. RESULTS: Among 571 first cesarean section, 178 were eligible to check list application. 147 charts were analyzed in the study. 11.9% of first cesarean deliveries performed were potentially avoidable after applying the checklist. This represented 6.6% of all cesareans. CONCLUSION: The checklist based on the recall of good practices could be an interesting tool to decrease the first cesarean rate.

Long term survival of a patient with Perlman syndrome due to novel compound heterozygous missense mutations in RNB domain of DIS3L2.

Perlman syndrome is a rare overgrowth syndrome characterized by polyhydramnios, macrosomia, distinctive facial appearance, renal dysplasia, and a predisposition to Wilms' tumor. The syndrome is often associated with a high neonatal mortality rate and there are few reports of long-term survivors. We studied a 6-year-old Japanese female patient, who was diagnosed with Perlman syndrome, with novel compound heterozygous mutations in DIS3L2 (c.[367-2A > G];[1328T > A]), who has survived long term. Most reported DIS3L2 mutations have been the homozygous deletion of exon 6 or exon 9, and these mutations would certainly have caused the loss of both RNA binding and degradation activity. We have identified new compound heterozygous mutations in the DIS3L2 of this long-term survivor of Perlman syndrome. The reason our patient has survived long-term would be a missense mutation (c.1328 T > A, p.Met443Lys) having retained RNA binding in both the cold-shock domains and the S1 domain, and through partial RNA degradation. If partial exonuclease functions remain in at least one allele, long-term survival may be possible. Further studies of Perlman syndrome patients with proven DIS3L2 mutations are needed to clarify genotype-phenotype correlation.

Adolescent Pregnancies in the United States: How Obstetric and Sociodemographic Factors Influence Risk of Cesarean Delivery.

Objective To determine how an adolescent's risk of cesarean varies by maternal age and race/ethnicity, and evaluate the contribution of obstetric and sociodemographic factors to mode of delivery. Study Design This is a retrospective cohort study of 604,287 births to women aged 13 to 23 years. Regression techniques were used to determine maternal ages at lowest risk of primary cesarean in each major racial/ethnic group before and after adjustment for various cesarean risk factors. Results Adolescent age was associated with lower risk of cesarean compared with young adults (17.2% at age 13 years vs 24.8% at age 23 years, p < 0.05). After stratification by race/ethnicity, Non-Hispanic Black women had the highest probability of cesarean, while Asian/Pacific Islanders had the lowest probability across all ages. When compared with young adults of the same race/ethnicity, young adolescents continued to have a lower risk of cesarean, decreased by at least 30% until age 18 years (White) and 17 years (other racial/ethnic groups). These associations persisted after adjustment for obstetric and sociodemographic risk factors. Conclusion Young maternal age is protective against cesarean delivery in all racial/ethnic groups. Adolescents also experience racial/ethnic disparities in mode of delivery similar to those observed in adults, which were unexplained by either obstetric or sociodemographic factors.

Vacuum extraction in fetal macrosomia and risk of neonatal complications: a population-based cohort study.

INTRODUCTION: High birthweight is associated with complicated childbirth. The aim of the present study was to investigate the association between birthweight, mode of delivery, and neonatal complications among infants born at term with a birthweight ≥3000 g. MATERIAL AND METHODS: This population-based cohort study used data from the Swedish Medical Birth Register from 1999 to 2012, including 1 030 775 births at >36 completed weeks. Exposure was mode of delivery, categorized into non-instrumental vaginal delivery (VD), emergency cesarean section (CS), vacuum extraction (VE) or cesarean section following attempted vacuum extraction (VE + CS), and birthweight was divided into five categories (3000-3999 g, 4000-4499 g, 4500-4999 g, and ≥5000 g). The following outcomes were assessed: 5-min Apgar score <7, neonatal convulsions, intracranial hemorrhage, and brachial plexus injury. Infants born after VD with a birthweight of 3000-3999 g were used as reference in the logistic regression analysis. RESULTS: The odds ratios for all complications increased at higher birthweights among infants born after VE/VE + CS and VD. The highest risks were seen after VE/VE + CS with an adjusted odds ratio for neonatal convulsions of 2.6 (95% CI 2.1-3.2) in the reference birthweight group and 6.3 (95% CI 4.3-9.2) among infants with a birthweight of ≥4500 g. The corresponding adjusted odds ratios for intracranial hemorrhage were 2.6 (95% CI 1.7-3.9) and 6.7 (95% CI 3.3-13.6) and for brachial plexus injury 4.0 (95% CI 3.3-4.9) and 88.4 (95% CI 71.9-108.4). CONCLUSION: Vacuum extraction is a risk factor for serious neonatal complications, in particular when used in macrosomic fetuses.

[Birth of a large fetus: epidemiologic aspects and analysis of factors of a case in an southern urban public health facility (Abidjan)]. / L'accouchement du gros foetus: aspects épidémiologiques et analyse des facteurs déterminants à propos d'une étude cas-témoin à la Formation sanitaire urbaine d'Abobo sud (Abidjan).

OBJECTIVE: to highlight the factors which influence the childbirth of the large foetus in order to propose adequate conduits. MATERIALS AND METHOD: it acted of a study case--witness who had been held on May 1, 2003 to April 31, 2004. We had included 108 cases of childbirth large foetus and 229 cases of childbirth of foetus of standard weight. The appreciation of the factors which influence the childbirth in these two groups was made thanks to the statistical analysis of the Chi 2 test; it had been considered to be significant with the threshold 5%. RESULTS: the parturients who were confined of the large foetus were referred than those of the group of the witnesses (p = 0,0421). The high way was observed in the event of large foetus (p = 0,0091). When the childbirth of the macrosomes was done by low way, one had noted more dystocie shoulders (0,0091) and more episiotomy (0,00479). The Caesarean for large foetus was carried out when the parturients were allowed in room of work in phase of latency of work (p = 0,028). Among the cesarized parturients, there were more first calf cows (p = 0,00532), young people and parturients whose size was between 150 and 170 cm (p = 0,00069) in the group of study. CONCLUSION: the childbirth of the large foetus is childbirth at the risk both for the mother the foetus. However it is necessary to be patient in front of an excessive uterine height evoking a large foetus. The indication of Caesarean should be posed only in primiparous, young parturients or teenagers, first calf cows of which size lower than 170 cm.

Delivery of the macrosomic infant: cesarean section versus vaginal delivery.

The macrosomic fetus of a diabetic woman faces increased risk for injury at the time of birth. Cesarean section offers the potential for avoiding trauma to the fetus, but can result in increased morbidity in the mother as compared to vaginal delivery. In this article, the advantages and disadvantages of the 2 routes of delivery for the overgrown fetus of a diabetic mother are discussed. In addition, methods for diagnosing macrosomia by ultrasound are examined, along with the benefits and pitfalls of ultrasonic fetal weight estimation in the setting of diabetes. Finally, management approaches for selecting route of delivery for the macrosomic fetus are described and analyzed.

[German-speaking midwifes prefer vaginal mode of delivery to elective Caesarean]. / Hebammen präferieren vaginale Geburt statt Wunschsektio1.

Increasing ceasarean section rates are a world wide concern in obstetrics. One of the latest contributing factors is the elective caesarean section in uncomplicated singleton pregnancy at term. The preference for this mode of delivery was primarily brought forward by obstetric practitioners (Al Mufty, McCarthy, Fisk 1996). A questionnaire, which mainly aimed to ask germanspeaking midwifes in Austria, Germany and Switzerland about their personal choice of delivery mode, was included in one of the issues of the German-language midwifery journal "Die Hebamme". This questionnaire contained 5 half-closed/half open questions describing specific obstetric occurrences. The midwifes were asked to express their preferred mode of delivery and describe their reason for choosing. 446 questionnaires (12 %) were returned. The majority (100 %) of the german speaking midwifes preferred a normal vaginal delivery in an uncomplicated singleton pregnancy at term with a child in cephalic presentation. The rating was about the same (97 %) in the presence of general risk factors which don't indicate a primary caesarean section. Breech presentation and macrosomia are a matter of concern to the midwifes. Midwifes arguing for a first child in breech presentation or with macrosomia > 4.5 kg vote highly significantly more frequently for elective caesarean section than midwifes arguing for at least the second child. The first-rate reasons for the preference of vaginal delivery concern the natural and physiological way of delivery, the personal experience of delivery, the higher risks of caesarean section and the possibility of a later caesarean section in case of fetal distress during first or second stage of labour. Concerns are expressed about the maintenance of competence amongst practitioners, thus influencing the choice of mode of delivery in obstetrics.

[Factors associated with cesarean section in France. Results from the 1995 National Perinatal Survey]. / Les facteurs associés à la pratique d'une césarienne en France. Résultats de l'enquête nationale périnatale de 1995.

OBJECTIVE: to study the relationship between the cesarean section rates and the characteristics of newborns, mothers and maternity units, studying primary and repeat cesarean sections, and cesarean sections before and during labour separately. Population. Representative sample of births in France. METHOD: Univariate and multivariate analyses. RESULTS: The primary cesarean section rate (among women without previous cesarean section) was 11.0% and the repeat cesarean section rate was 62.9%. The main factors associated with primary cesarean sections were breech presentation, small-for-gestational age before term, and high birthweight, for characteristics of newborns, and parity 0,30 years old or more and high pre- pregnancy weight, for characteristics of mothers. These factors were linked to cesarean sections before labour and during labour. Preterm and post-term deliveries, educational level, and nationality played a different role before and during labour. Very few factors were linked to repeat cesarean sections. CONCLUSION: Various factors are associated with cesarean sections; they express the influence of both unfavorable medical conditions and particular medical care organisations.

Predictors of episiotomy use at first spontaneous vaginal delivery.

OBJECTIVE: To identify factors associated with the use of episiotomy at spontaneous vaginal delivery. METHODS: We studied 1576 consecutive term, singleton, spontaneous vaginal deliveries in nulliparas at Brigham & Women's Hospital between December 1, 1994 and July 31, 1995. The association of demographic variables and obstetric factors with the rate of episiotomy use were examined. Adjusted odds ratios (OR) and confidence intervals (CI) were estimated from multiple logistic regression analysis. RESULTS: The overall rate of episiotomy was 40.6% (640 of 1576). Midwives performed episiotomies at a lower rate (21.4%) than faculty (33.3%) and private providers (55.6%) (P =. 001). After controlling for confounding factors with logistic regression, private practice provider was the strongest predictor of episiotomy use (OR, 4.1; 95% CI, 3.1, 5.4) followed by faculty provider (OR, 1.7; 95% CI, 1.1, 2.5), prolonged second stage of labor (OR, 1.8; 95% CI, 1.2, 2.7), fetal macrosomia (OR, 1.6; 95% CI, 1.1, 2.5), and epidural analgesia (OR 1.4, 95% CI, 1.1, 1.8). CONCLUSION: The strongest factor associated with episiotomy at delivery was the category of obstetric provider. Obstetric and demographic factors evaluated did not readily explain this association.

Limited usefulness of fetal weight in predicting neonatal brachial plexus injury.

OBJECTIVE: The objectives were to determine the neonatal morbidity rate from vaginal birth and examine fetal weight-based injury-prevention strategies. STUDY DESIGN: Selected neonatal morbidities were categorized by birth weight for all vertex vaginal deliveries occurring during a 12-year period. Sensitivity, specificity, and predictive values for brachial palsy were calculated at increasing birth weight cutoff levels. A policy of cesarean delivery for macrosomic infants was evaluated. RESULTS: There were 80 cases of brachial palsy among 63,761 infants (0.13%). In mothers without diabetes, rates in the 4500- to 4999-g and >5000-g groups were 3.0% and 6.7%, respectively. A threshold of 3700 g had a sensitivity of 71% and a specificity of 86%; the positive predictive value was 0.56%. To prevent a single case of permanent injury, 155 to 588 cesarean deliveries are required at the currently recommended cutoff weight of 4500 g. CONCLUSIONS: The rates of lasting morbidity do not justify routine cesarean delivery for infants without diabetic complications weighing <5000 g.

When is fetal macrosomia (> or = 4500 g) an indication for caesarean section?

Caesarean and vaginal deliveries of macrosomic infants weighing > or = 4500 g were studied, and pregnant women analysed by indication for caesarean section, presentation, parity and age. Both maternal and neonatal injuries occurred. Puerperal morbidity was noted in women delivered either by caesarean section or vaginally. The control group consisted of 321 parity- and age-matched pregnant women and their newborn infants weighing 3000-3499 g. The two groups were studied according to the same criteria. In the maternity unit of the General Hospital in Sibenik, Croatia, 10852 newborn infants were delivered (only singleton pregnancies included) between 1 January 1984 and 31 December 1993, of whom 321 (2.96%) weighed > or = 4500 g (290 weighing 4500-4999 g, and 31 weighing > or = 5000 g). Caesarean section was performed in 36 (11.2%) and 14 (4.4%) in the macrosomic and control groups, respectively (X2 = 10.50; P < 0.01). Of the 321 women with a macrosomic infant, 10 (3.1%) had a caesarean section for cephalopelvic disproportion and 7 (2.2%) for breech presentation. Caesarean section for vertex presentation was used more frequently in the macrosomic than in the control group (9.0% vs. 3.3%) (P < 0.01), as well as it was used for breech presentation (77.8% vs. 16.7%) (P < 0.01). As regards transverse and oblique lies, no difference was observed. The rates of macrosomic infants delivered from primiparous and grand multiparous women by caesarean section (i.e., 23.1% vs. 5.9% vs. 18.2%) were highly significant (X2 = 19.07; P < 0.001), as were the rates in adolescent pregnant women, in those of optimal childbearing age and in old pregnant women (60.0% vs. 9.0% vs. 26.9%) (X2 = 18.67; P < 0.001). Injuries were sustained by 28 (9.8%) women with a macrosomic infant delivered vaginally and by 12 (3.9%) controls (X2 = 6.25; P < 0.05). No maternal injuries were reported with caesarean delivery in either group. There was no birth trauma in the macrosomic and control infants delivered by caesarean section. With vaginal delivery birth trauma involved clavicular fracture (5.6%), brachial plexus palsy (2.8%) and central nervous system syndrome (2.1%). A total of 30 (10.5%) macrosomic infants and 4 (1.3%) controls, were identified as having birth trauma (X2 = 20.99; P < 0.001). No difference in puerperal morbidity rates were observed between the two groups with regard to caesarean and vaginal delivery (P > 0.05), showing significantly lower rates for vaginally delivered macrosomic infants (12.3% vs. 30.6%) (X2 = 8.51; P < 0.01). There was no perinatal death among those delivered by caesarean section in either group; however, when delivered vaginally, the rates were 0.70% (2 of 285) and 0.65% (2 of 307) for the macrosomic and control infants, respectively (P > 0.05). No women in either the macrosomic or control group died. In conclusion, decision making on management options when delivering a macrosomic infant depends on fetal presentation and maternal age and parity. Vertex presenting macrosomic infants weighing > or = 4500 g should be delivered vaginally, but liberal judgement is suggested in resorting to caesarean section delivery. Abnormal presentation, as well as malpresentations in primiparous women, are an absolute indication for caesarean section, whereas malpresentations in multiparous women are a relative (underlying) indication for caesarean section.

Resultado materno y perinatal en recién nacidos macrosómicos: estudio caso y control / Maternal and perinatal result in macrosomic newborns: case and control study

La macrosomía fetal se considera un factor de riesgo para parto operatorio, asfixia intraparto y traumatismo obstétrico. Algunos recomiendan la operación cesárea para prevenir o reducir esas complicaciones. Se presenta el resultado materno y perinatal de 275 recién nacidos (RM) macrosómicos comparados con un grupo control.Hubo un 8,2 por ciento de macrosomía durante el período estudiado. La ganancia de peso, polihidroamnios y factores de riesgo para diabetes gestacional tuvieron diferencias significativas. No hubo diferencias significativas entre diabetes gestacional, inercia uterina, endometritis postparto, infección de herida operatoria y días de hospitalización. La resolución vaginal del parto y el número de cesáreas fue similar en ambos grupos. Tampoco hubo diferencias en el Apgar al minuto y cinco minutos de vida. Sólo hubo diferencias en la retención de hombros. En conclusión, los RN macrosómicos no tiene una morbimortalidad materna y perinatal superior a la población general, que desaconseje un intento de parto vaginal mediante una prueba de parto vigilada

Shoulder dystocia: a fetal-physician risk.

Trauma that occurs as a result of shoulder dystocia is an important cause of neonatal morbidity. If the occurrence of severe shoulder dystocia, resulting in fetal asphyxia and trauma, could be accurately predicted from maternal risk factors, then a cesarean section would be indicated to prevent the poor outcome. The information available in the obstetric literature, however, is contradictory regarding whether shoulder dystocia can be predicted. In the present study, the patients at greatest risk of shoulder dystocia (all 394 mothers delivering neonates with birth weights greater than or equal to 4000 gm over a 2-year period) were examined. A three-way discriminant analysis was used to determine if a model could be developed that could effectively predict those patients who would be included in each of the groups of no shoulder dystocia, shoulder dystocia without trauma (29 patients), and shoulder dystocia with trauma (20 patients). Three factors, including birth weight, prolonged deceleration phase, and length of second stage labor, were found individually to contribute significantly to the classification. However, when examined in detail, it was noted that while 94% of cases with no shoulder dystocia would be detected, only 16% of the cases of shoulder dystocia with trauma would be predicted by this model. We conclude that in the group of pregnancies delivering neonates greater than or equal to 4000 gm, the occurrence of shoulder dystocia cannot be predicted from clinical characteristics or labor abnormalities, and that the occurrence of shoulder dystocia is not evidence of medical malpractice.