In women with preexisting type 1 diabetes mellitus, delivery usually is recommended at 38–39 weeks of gestation to reduce the risk of late fetal death and complications associated with fetal macrosomia.1,2 Indeed, macrosomia is increased in newborns of women with type 1 diabetes mellitus; 49–63% of newborns are large for gestational age, and 20–25% weigh more than 4,000 g.3–5 Macrosomia is associated with an increased risk of shoulder dystocia and brachial plexus injury.6
Cesarean delivery essentially eliminates the risk of brachial plexus injury. Accordingly, the rate of cesarean delivery in women with type 1 diabetes mellitus is twofold to fourfold higher than in the general population, ranging from 45% to 73%.3,5,7–10 However, cesarean delivery is associated with increased maternal morbidity.11 Furthermore, a uterine scar contraindicates induction of labor for many obstetricians and places women at increased risk for uterine rupture, placenta previa, or accreta, or all of these, and for the need for hysterectomy after a second cesarean delivery.12 Thus, avoidance of unnecessary primary cesarean delivery has important implications for future pregnancies, and hence the importance of identifying factors associated with cesarean delivery in women with type 1 diabetes mellitus. Induction of labor at 38 weeks of gestation in insulin-requiring women with diabetes decreases the rate of macrosomia and does not increase the rate of cesarean delivery.13 In retrospective studies, a prior cesarean delivery and nulliparity have been found to be independent predictors of cesarean delivery in women with diabetes.10,14 However, these studies included primarily women with gestational diabetes mellitus and their conclusions may not apply to women with type 1 diabetes mellitus.
The primary objective of the present study was to identify possible risk factors associated with cesarean delivery, either without labor or during a trial of labor, in nulliparous women with type 1 diabetes mellitus managed with standardized protocols regarding the timing and the route of delivery. The secondary objective was to estimate maternal and neonatal morbidity associated with planned cesarean delivery and with the route of delivery when a trial of labor was attempted.
MATERIALS AND METHODS
All nulliparous women with type 1 diabetes mellitus and a single pregnancy who were consecutively delivered after 22 weeks of gestation of a liveborn neonate in the Department of Obstetrics and Gynecology, Cochin - Saint Vincent-de-Paul Hospital between 1997 and 2008 were eligible for the study. Women were excluded if they were multiparous, had a fetus with major congenital malformation, or had intrauterine fetal death. After receiving approval from the Institutional Review Board Comité de Protection des Personnes se prêtant à la Recherche Biomédicale Ile de France 3, we performed a nested case-control study within this prospective cohort.
Preconception care and management of diabetes during pregnancy were standardized as previously reported.15 Glycemic control was assessed by hemoglobin A1C (Hb A1C levels measured by high-performance liquid chromatography (normal: 4.2–5.7%) during the first and second trimesters and at delivery. Gestational age was determined from the date of the last menstrual period confirmed by first-trimester ultrasonography. Maternal baseline characteristics included age, ethnicity, marital status, smoking habit, and prepregnancy body mass index (BMI). Diabetes characteristics included duration of diabetes, retinopathy, nephropathy, and chronic hypertension. Chronic hypertension, gestational hypertension, and preeclampsia were defined according to the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy.16 Antenatal nonstress test was performed twice weekly from 32 weeks of gestation to delivery.
The timing of delivery was guided by gestational age according to a standardized protocol. When delivery was indicated for maternal or fetal reason, it was decided upon a case-by-case benefit–to–risk ratio estimation. In the absence of complication, delivery was planned at 38 weeks of gestation. The route of delivery was chosen according to the following obstetric conditions: a cesarean delivery without labor was planned for fetal malpresentation (breech or transverse lie) or placenta previa, and encouraged for suspected fetal macrosomia. Suspected fetal macrosomia was based on a symphysis-fundal height measurement more than 40 cm combined with Leopold's maneuvers or a fetal abdominal circumference more than 360 mm, or both, on ultrasonography performed in the week before delivery.17 A trial of labor was attempted in patients with cephalic presentation, normal placental insertion, and no suspicion of fetal macrosomia. Bishop score was evaluated at admission,18 and the following protocol was used for all women: when the Bishop score was lower than 6, the cervix was ripened with a vaginal insert containing 10 mg of dinoprostone (Propess; Ferring Pharmaceuticals, Gentilly, France) for 24 hours then labor was induced with intravenous oxytocin regardless of the Bishop score; when the Bishop score was 6 or higher, the labor was induced with oxytocin. Oxytocin was started at 2.5 milliunits/min and increased by 2.5 milliunits/min at 30-minute intervals until strong contractions occurred at 3-minute intervals or labor progressed. The maximum dose of oxytocin used was 25 milliunits/min. An amniotomy was performed only when the fetal vertex was well applied to the cervix. Augmentation of labor was considered if the frequency of contractions was fewer than three contractions every 10 minutes in the absence of cervical change. Intrauterine pressure catheters were not systematically used. Epidural anesthesia was placed only on maternal request. A prophylactic intravenous administration of 1 g of cefazoline was systematically given at cord clamping during cesarean deliveries. During labor and delivery, a standardized protocol was applied to achieve normoglycemia as previously described.19
The indications for cesarean delivery in labor were classified as failed induction, arrest disorders, or nonreassuring fetal heart rate (FHR). Failed induction was defined as a failure to achieve dilatation of 4 or more centimeters after 12 hours of membrane rupture and oxytocin administration. Arrest disorders were defined as no change in cervical dilatation during the active phase for at least 3 hours of adequate uterine contractions. Second stage arrest was defined as a lack of progress for 3 hours.Operative vaginal deliveries were performed only at full cervical dilatation, with the fetal head at +1 station or lower. Indications for operative vaginal delivery were an absence of progression of the second stage of labor or a nonreassuring FHR. Shoulder dystocia was defined as a delivery that required additional obstetric maneuvers following failure of gentle downward traction on the fetal head to effect delivery of the shoulders.
Maternal complications included third- and fourth-degree perineal lacerations, postpartum hemorrhage, wound infections and hematomas, and postpartum endometritis. Postpartum hemorrhage was defined as an estimated blood loss of more than 500 mL after vaginal delivery and more than 1,000 mL after cesarean delivery. Wound infection was defined as an infected abdominal or episiotomy wound. Wound hematomas requiring drainage were reported. Postpartum endometritis was defined as fever higher than 38°C, fundal tenderness, or foul-smelling lochia with no other source of fever requiring more than 24 hours of antibiotics. Neonatal outcomes included birth weight, sex, 5-minute Apgar score, umbilical artery pH, and neonatal complications. Preterm delivery was defined as delivery before 37 weeks of gestation. Birth weight according to gestational age referred to the French growth standard curves.20 Neonates who were large for gestational age (LGA) were defined as having birth weight in the 90th percentile or higher for gestational age, and macrosomia as birth weight 4,000 g or more. Neonatal complications included admission to neonatal intensive care unit, hypoglycemia, respiratory distress syndrome, and intraventricular hemorrhage. Neonatal hypoglycemia was defined as the occurrence of a serum glucose level of less than 40 mg/dL (2.2 mmol/L) despite systematic prevention by the injection of glucagon (0.3 mg/kg) at birth. Respiratory distress syndrome was defined as the need for oxygen therapy or invasive ventilation for more than 24 hours.
Maternal demographic, medical, and obstetric factors were evaluated for association with cesarean delivery without labor compared with trial of labor. When trial of labor was attempted, the same factors were evaluated for association with cesarean delivery in labor. We calculated the sample size to achieve a power of 80% with the following hypothesis: 1) a prevalence of 0.20 of the studied factor in the control group, 2) an odds ratio (OR) of 3.0 or more between the studied factor and the outcome (α=0.05; case-control ratio 0.8). Under these assumptions, the calculated sample size was 144 women. Independent factors associated with cesarean delivery and odds ratios were identified by logistic regression (STATA 9.2, College Station, TX). If linearity with independent outcome could not be assumed, continuous variables were recoded in nominal variables to be included in the logistic model. For variables with more than 3% missing data, an additional class for missing data was created to avoid a loss of power in multivariable analysis. All possible interactions were tested. In case of collinearity between two or more variables, the most clinically important was chosen. To avoid over-fitting and collinearity, we limited the number of variables included in the final model to 10% of the prevalence of the primary outcome in the sample. Variables were kept in the model if P values were lower than .05.
During a 12-year period (1997–2008), 209 nulliparous women with type 1 diabetes mellitus and a single pregnancy were consecutively delivered of a liveborn neonate in the same center. Their characteristics are shown in Table 1.
Cesarean delivery was performed without labor in 94 women (45%) and a trial of labor was attempted in 115 women (55%). Among those attempting a trial of labor, 54 (47%) were delivered by cesarean section and 61 (53%) had a vaginal delivery. The overall cesarean delivery rate was 71%.
Among the 94 women who had a cesarean delivery without labor, the reasons for cesarean section were suspected fetal macrosomia (n=56), fetal malpresentation (n=7), nonreassuring FHR (n=13), and other maternal reasons (n=18). Other maternal reasons were preeclampsia (n=9), contracted pelvis (n=2), placenta previa (n=2), maternal request (n=2), ketoacidosis (n=1), thrombocytopenia (n=1), and previous laparoscopic myomectomy (n=1).
Among the 115 women who attempted a trial of labor, labor was spontaneous in 18 and induced in 97 patients. Among 54 women who had a cesarean delivery in labor, the reasons for cesarean section were failed induction (n=25), arrest disorders (n=20), and nonreassuring FHR (n=9). Sixty-one women delivered vaginally, spontaneously in 25 cases and operatively in 36 cases. Among 71 women who had a preinduction cervical ripening, 23 (32%) had cesarean delivery for failed induction. Shoulder dystocia occurred in three neonates weighing 3,960, 4,010, and 4,070 g, and one transient brachial plexus injury was reported. The relative risk of shoulder dystocia in an undiagnosed neonate weighing 4,000 g or more was 56 (95% confidence interval [CI], 8–412).
We first compared women who had a cesarean delivery without labor with those who attempted a trial of labor. In the univariable analysis, the following factors were associated with cesarean delivery without labor: first-trimester Hb A1C higher than 6%, Hb A1C at delivery higher than 6%, presence of a nephropathy, gestational weight gain more than 15 kg, and suspected fetal macrosomia. In the multivariable analysis, gestational weight gain more than 15 kg (39% compared with 23%; OR, 2.2; 95% CI, 1.1–4.5) and suspected macrosomia (79% compared with 21%; OR, 13.1; 95% CI, 5.3–32.2) remained independently associated with cesarean delivery (Table 2). Among the 56 planned cesarean deliveries for suspected fetal macrosomia, 36 neonates (64%) weighed less than 4,000 g. Among the 162 women who delivered at term, the sensitivity of identifying a newborn weighing 4,000 g or more was 83% (95% CI, 61–95%), and the specificity was 77% (95% CI, 69–83%). The positive predictive value of a suspicion of macrosomia was 37% (95% CI, 24–52%), and the negative predictive value was 96% (95% CI, 91–99%).
Among women who had a trial of labor, we compared those who had a cesarean delivery with those who had a vaginal delivery. In the univariable analysis, the following factors were associated with an increased risk of cesarean delivery in labor: age older than 35 years, prepregnancy BMI more than 25, preconception Hb A1C more than 6%, and Bishop cervical score of 3 or lower. Gestational weight gain more than 15 kg was not associated with this outcome (25% compared with 18%; OR, 1.5; 95% CI, 0.6–3.6). In the multivariable analysis, prepregnancy BMI more than 25 (84% compared with 39%; OR, 7.5; 95% CI, 1.9–22.4) and Bishop score of 3 or lower (66% compared with 25%; OR, 5.9; 95% CI, 2.2–16.1) remained independently associated with cesarean delivery in labor (Table 2).
Preconception care, presence of a nephropathy, Hb A1C levels during pregnancy, preeclampsia, and preterm delivery were not associated with cesarean delivery, either without labor or during a trial of labor. Maternal complications are reported in Table 3. No third- or fourth-degree perineal lacerations were reported.
The neonatal outcomes are reported in Table 4. Eighty-four neonates (40%) were LGA and 24 (12%) weighed 4,000 g or more. The proportions of LGA neonates and those weighing 4,000 g or more were higher in the planned cesarean delivery group than in the trial of labor group. The other outcomes were not different among the three groups.
In This cohort of 209 nulliparous women with type 1 diabetes mellitus, the rate of cesarean delivery was 71%.
We found that a gestational weight gain more than 15 kg and a suspicion of fetal macrosomia were independently associated with cesarean delivery without labor. However, the accuracy of assessment for fetal macrosomia was poor. When a trial of labor was attempted, the rate of cesarean delivery was 47%. Prepregnancy BMI more than 25 and a Bishop score of 3 or lower were associated with cesarean delivery. The rate of shoulder dystocia was 3%. The rates of wound infection and endometritis were 0.7% and 3%, respectively.
The strengths of this study were that all the data were prospectively collected and the variables were defined and measured according to standard definitions. No eligible woman was excluded from the analysis, all the women were managed according to a standardized protocol, and there was no loss of follow-up. Finally, logistic regression analysis was performed to adjust for potential confounders. A limitation of our study was its single-center setting. Another limitation was its sample size. The smallest OR that could be detected for factors with a prevalence of 0.20 was 2.5 for association with cesarean delivery without labor and 3.4 for association with cesarean delivery in labor. To detect an association with an OR of 2.0 or higher, a sample size of 393 women was required. However, type 1 diabetes mellitus is uncommon (about 3 per 1,000 pregnancies), and morbidity associated with cesarean delivery is a real concern.
The rate of cesarean delivery was very high in this series, but was similar to that in previously reported series.3,5,7–10 In women with diabetes, the risk of shoulder dystocia ranged from 8.4% to 16.7% for neonates weighing 4,000 to 4,500 g.6 Furthermore, it has been reported that 84% of the neonates born to mothers with diabetes who have shoulder dystocia weighed more than 4,000 g.21 Thus, avoidance of vaginal delivery of macrosomic fetuses in women with type 1 diabetes mellitus would eliminate most cases of shoulder dystocia. In one study, an estimated weight threshold of 4,250 g as an indication for cesarean delivery reduced the rate of shoulder dystocia without increasing the rate of cesarean section.22 In that study, most women had gestational diabetes mellitus and no conclusion can be drawn for women with type 1 diabetes mellitus. In our study, 27% of women had a cesarean delivery without labor for suspected macrosomia, but neonatal birth weight was less than 4,000 g in 64% of these cases. Thus, an accurate prediction of macrosomia and the use of an estimated fetal weight threshold for cesarean delivery are crucial for choosing the optimal route of delivery. In our study, the predictive positive value of a suspicion of identifying a newborn weighing 4,000 g or more was only 37% and led to unnecessary cesarean deliveries. In other studies, the predictive positive value of identifying a newborn weighing 4,000 g or more clinically and sonographically was higher (44–81%) among diabetic women.23–25 Prospective studies in women with diabetes have shown that clinical estimates of macrosomia are as predictive as those derived with ultrasonography.26
The greater predictive value in the other reports might be explained by a higher prevalence of macrosomia (19–26%) in those studies as compared with ours.
Gestational weight gain more than 15 kg was associated with cesarean delivery without labor. This is in accordance with the increase of cesarean delivery observed in women with gestational diabetes mellitus who had gestational weight gain above the Institute of Medicine guidelines.27 However, the ideal gestational weight gain remains unknown.
When a trial of labor was attempted, we identified two independent predictors of cesarean delivery. A prepregnancy BMI more than 25 was associated with an increased risk of cesarean delivery. This is in accordance with the 50% increase of cesarean delivery observed in nulliparous overweight women without diabetes compared with women with normal BMI.28 This suggests that weight loss should be encouraged as part of preconception care in overweight women with type 1 diabetes mellitus. In our study, a gestational weight gain more than 15 kg was not associated with an increased risk of cesarean delivery. A Bishop score of 3 or lower was also associated with an increased risk of cesarean delivery. The evidence clearly points to the fact that labor induction in the nondiabetic nulliparous woman with an unfavorable cervix increases the risk of cesarean delivery.29 In our study, failed induction was defined as the failure to achieve dilatation of 4 cm or more after 12 hours of membrane rupture and oxytocin administration. Although there is no universally admitted definition of “failed induction,” in nulliparous women a latent phase lasting up to 18 hours during induction of labor allows the majority to achieve a vaginal delivery without increased maternal or neonatal morbidity.30 Allowing a prolongation of the latent phase in women with type 1 diabetes mellitus could increase the risk of chorioamniotitis and postpartum endometritis and should be evaluated. However, our results suggest that in the absence of complications, it may be worth waiting until 39 weeks of gestation and the achievement of a Bishop score higher than 3 to increase the success of labor induction. The risk of infection after cesarean delivery has been addressed in only a limited number of studies. In one study, 13% wound infections and 6% endometritis were reported in women with type 1 or type 2 diabetes mellitus after cesarean delivery, either elective or during labor, but prophylactic antibiotics were not systematically given.11 In another study, the rate of postoperative infection was similar in women with or without diabetes when antimicrobial prophylaxis was routinely given for cesarean delivery during labor.31 In our study, all women received prophylactic antibiotics during cesarean delivery, and the rates of wound infection and endometritis were low, 0.7% and 3%, respectively.
In conclusion, in women with type 1 diabetes mellitus, the challenge for obstetricians remains to find the optimal balance between maternal and fetal risks. Our results identified several factors that are potentially modifiable to decrease the rate of cesarean deliveries. The proportion of cesarean deliveries was much higher in mothers of macrosomic neonates in contrast to those women with type 1 diabetes mellitus with normal-weight neonates. Thus, prevention of macrosomia by tight glycemic control may decrease the need for cesarean delivery in women with type 1 diabetes mellitus.32 Accurate diagnosis of macrosomia is also important to minimize the risk of shoulder dystocia and to decrease the rate of unnecessary cesarean deliveries. Magnetic resonance imaging that gives access to maternal pelvic measurements and fetal shoulder width,33 although not often if ever used, might help to improve the accuracy of predicting the risk of shoulder dystocia. In the absence of macrosomia and when glycemic control is good, a conservative management with fetal surveillance waiting for a favorable cervical status should be investigated in a randomized controlled study to potentially decrease the rate of failed inductions in this high-risk population.
1. Boulvain M, Stan C, Irion O. Elective delivery in diabetic pregnant women. The Cochrane Database Syst Reviews 2001, Issue 2. Art. No.: CD001997. DOI: 10.1002/14651858.CD001997.
2. Sacks DA, Sacks A. Induction of labor versus conservative management of pregnant diabetic women. J Matern Fetal Neonatal Med 2002;12:438–41.
3. Evers IM, de Valk HW, Visser GH. Risk of complications of pregnancy in women with type 1 diabetes: nationwide prospective study in the Netherlands. BMJ 2004;328:915–8.
4. Macintosh MC, Fleming KM, Bailey JA, Doyle P, Modder J, Acolet D, et al. Perinatal mortality and congenital anomalies in babies of women with type 1 or type 2 diabetes in England, Wales, and Northern Ireland: population based study. BMJ 2006;333:177–82.
5. Jensen DM, Damm P, Moelsted-Pedersen L, Ovesen P, Westergaard JG, Moeller M, et al. Outcomes in type 1 diabetic pregnancies: a nationwide, population-based study. Diabetes Care 2004;2:2819–23.
6. Nesbitt TS, Gilbert WM, Herrchen B. Shoulder dystocia and associated risk factors with macrosomic infants born in California. Am J Obstet Gynecol 1998;179:476–80.
7. Pregnancy outcomes in the Diabetes Control and Complications Trial. Am J Obstet Gynecol 1996;174:1343–53.
8. Wylie BR, Kong J, Kozak SE, Marshall CJ, Tong SO, Thompson DM. Normal perinatal mortality in type 1 diabetes mellitus in a series of 300 consecutive pregnancy outcomes. Am J Perinatol 2002;19:169–76.
9. Diabetes and Pregnancy Group, France. French multicentric survey of outcome of pregnancy in women with pregestational diabetes. Diabetes Care 2003;26:2990–3.
10. Kjos SL, Berkowitz K, Xiang A. Independent predictors of cesarean delivery in women with diabetes. J Matern Fetal Neonatal Med 2004;15:61–7.
11. Takoudes TC, Weitzen S, Slocum J, Malee M. Risk of cesarean wound complications in diabetic gestations. Am J Obstet Gynecol 2004;191:958–63.
12. Silver RM, Landon MB, Rouse DJ, Leveno KJ, Spong CY, Thom EA, et al, for the National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. Maternal morbidity associated with multiple repeat cesarean deliveries. Obstet Gynecol 2006;107:1226–32.
13. Kjos SL, Henry OA, Montoro M, Buchanan TA, Mestman JH. Insulin-requiring diabetes in pregnancy: a randomized trial of active induction of labor and expectant management. Am J Obstet Gynecol 1993;169:611–5.
14. Yogev Y, Ben-Haroush A, Chen R, Glickman H, Kaplan B, Hod M. Active induction management of labor for diabetic pregnancies at term: mode of delivery and fetal outcome—a single center experience. Eur J Obstet Gynecol Reprod Biol 2004;114:166–70.
15. Lepercq J, Coste J, Theau A, Dubois-Laforgue D, Timsit J. Factors associated with preterm delivery in women with type 1 diabetes: a cohort study. Diabetes Care 2004;27:2824–8.
16. Diagnosis and management of preeclampsia and eclampsia. ACOG Practice Bulletin No. 33. American College of Obstetricians and Gynecologists. Obstet Gynecol 2002;99:159–67.
17. Pedersen JF, Molsted-Pedersen L. Sonographic estimation of fetal weight in diabetic pregnancy. Br J Obstet Gynaecol 1992;99:475–8.
18. Bishop EH. Pelvic scoring for elective induction. Obstet Gynecol 1964;24:266–8.
19. Lepercq J, Abbou H, Agostini C, Toubas F, Francoual C, Velho G, et al. A standardized protocol to achieve normoglycaemia during labour and delivery in women with type 1 diabetes. Diabetes Metab 2008;34:33–7.
20. Mamelle N, Munoz F, Martin JL, Laumon B, Grandjean H. Fetal growth from the AUDIPOG study. II: application for the diagnosis of intrauterine growth retardation [in French]. J Gynecol Obstet Biol Reprod (Paris) 1996;25:71–7.
21. Langer O, Berkus MD, Huff RW, Samueloff A. Shoulder dystocia: should the fetus weighing greater than or equal to 4000 grams be delivered by cesarean section? Am J Obstet Gynecol 1991;165:831–7.
22. Conway DL, Langer O. Elective delivery of infants with macrosomia in diabetic women: reduced shoulder dystocia versus increased cesarean deliveries. Am J Obstet Gynecol 1998;178:922–5.
23. Benson CB, Doubilet PM, Saltzman DH. Sonographic determination of fetal weights in diabetic pregnancies. Am J Obstet Gynecol 1987;156:441–4.
24. McLaren RA, Puckett JL, Chauhan SP. Estimators of birth weight in pregnant women requiring insulin: a comparison of seven sonographic models. Obstet Gynecol 1995;85:565–9.
25. Best G, Pressman EK. Ultrasonographic prediction of birth weight in diabetic pregnancies. Obstet Gynecol 2002;99:740–4.
26. Johnstone FD, Prescott RJ, Steel JM, Mao JH, Chambers S, Muir N. Clinical and ultrasound prediction of macrosomia in diabetic pregnancy. Br J Obstet Gynaecol 1996;103:747–54.
27. Cheng YW, Chung JH, Kurbisch-Block I, Inturrisi M, Shafer S, Caughey AB. Gestational weight gain and gestational diabetes mellitus: perinatal outcomes. Obstet Gynecol 2008;112:1015–22.
28. Poobalan AS, Aucott LS, Gurung T, Smith WC, Bhattacharya S. Obesity as an independent risk factor for elective and emergency caesarean delivery in nulliparous women: systematic review and meta-analysis of cohort studies. Obes Rev 2009;10:28–35.
29. Vrouenraets FP, Roumen FJ, Dehing CJ, van den Akker ES, Aarts MJ, Scheve EJ. Bishop score and risk of cesarean delivery after induction of labor in nulliparous women. Obstet Gynecol 2005;105:690–7.
30. Simon CE, Grobman WA. When has an induction failed? Obstet Gynecol 2005;105:705–9.
31. Riley LE, Tuomala RE, Heeren T, Greene MF. Low risk of post-cesarean section infection in insulin-requiring diabetic women. Diabetes Care 1996;19:597–600.
32. Mello G, Parretti E, Mecacci F, La Torre P, Cioni R, Cianciulli D, et al. What degree of maternal metabolic control in women with type 1 diabetes is associated with normal body size and proportions in full-term infants? Diabetes Care 2000;23:1494–8.
33. Kastler B, Gangi A, Mathelin C, Germain P, Arhan JM, Treisser A, et al. Fetal shoulder measurements with MRI. J Comput Assist Tomogr 1993;17:777–80.