Previous shoulder dystocia with gestational diabetes: caesarean or vaginal birth at 37 weeks
A 34-year-old woman, G3P2, attends the antenatal clinic at 37+4 weeks. She has gestational diabetes treated with metformin 1 g twice daily since 26 weeks; fasting glucose is 5.6 mmol/L and HbA1c 6.1%. Her BMI is 34 kg/m². Her second delivery three years ago was complicated by shoulder dystocia at an estimated fetal weight of 4.3 kg, which resolved after the McRoberts manoeuvre and suprapubic pressure; the baby had no brachial plexus injury. The current ultrasound estimates fetal weight at 4.0 kg with an abdominal circumference above the 95th centile and normal liquor volume. She is adamant she does not want a caesarean and asks why she cannot simply be induced at 38 weeks. The unit has a consultant-led labour ward, 24-hour theatre and a special care baby unit, but no facility for fetal blood sampling.
1. Counsel her on elective caesarean section versus planned vaginal birth, using figures and explaining how you would reach a shared decision.
Read the worked answer
Start with what she is really asking: will the baby's shoulder get stuck again, and can that be prevented?
Recurrence risk. After one shoulder dystocia, recurrence is quoted at roughly 6–20%, commonly about 10%. That is a genuine increase but still means about nine in ten labours will not repeat it. The risk of a persisting brachial plexus injury is much smaller - permanent injury follows well under 1% of shoulder dystocias when manoeuvres are performed promptly, and most neonatal palsies resolve within 6–12 months.
Caesarean section. It reduces but does not abolish shoulder dystocia (it can occur at caesarean, especially in a deeply impacted head) and it carries its own costs: increased bleeding, infection, thromboembolism, a longer stay, a scarred uterus for future pregnancies (about 0.5% rupture risk if she labours next time) and, if performed before labour at 38 weeks, an increase in transient tachypnoea of the newborn.
Current estimated weight. 4.0 kg on scan is not itself an indication for caesarean. Accepted thresholds are an estimated weight of about 4500 g in a diabetic pregnancy, or 5000 g in a non-diabetic one. Ultrasound near term has an error of ±10%, so her baby could be 3.6 kg or 4.4 kg.
Induction at 38 weeks. There is no good evidence that induction prevents shoulder dystocia; it modestly reduces birthweight but increases the chance of a failed induction and caesarean, particularly with an unfavourable cervix - which then exposes her to both a caesarean and its risks without the benefit she wanted.
Shared decision. Present the numbers plainly, note that her previous dystocia was managed successfully without injury and that her unit has theatre and a special care unit, and offer either planned caesarean or a carefully conducted trial of labour with a documented plan. Whatever she chooses, document the discussion in her own words, and arrange a consultant review before 39 weeks.
2. She declines caesarean section. Set out your intrapartum management plan for a planned vaginal birth.
Read the worked answer
Antenatal preparation
- Senior review and a written plan in the notes, copied to the labour ward, labelled 'previous shoulder dystocia'.
- Optimise glycaemic control and agree timing of induction; in metformin-treated gestational diabetes, delivery by 39–40 weeks is usual, but induction should not be brought forward to 38 weeks solely to shrink the baby.
- Document that she understands recurrence risk and has chosen vaginal birth.
On the labour ward
- Consultant-led care, continuous CTG, intravenous access, group and save, and an anaesthetist aware of the case in case a rapid caesarean or difficult fetal extraction is needed.
- Avoid prolonged oxytocin augmentation and a long second stage; if progress is poor at 4 cm or beyond with adequate contractions, move to caesarean rather than persisting.
- Consider an epidural early: it does not prevent shoulder dystocia but makes manoeuvres and any emergency caesarean easier to conduct.
- Position for the second stage to avoid lithotomy with a fixed sacrum; be ready to use the all-fours position.
- Alert the neonatal team at the start of second stage.
If the shoulder does not deliver
- Call for help, start the clock, McRoberts plus suprapubic pressure first, then internal rotational manoeuvres (Rubin II, Wood's screw), then delivery of the posterior arm, then all-fours. Never fundal pressure. Episiotomy only if needed to gain access.
- Document times, sequence of manoeuvres, which shoulder was anterior, cord gases and the baby's arm movement at delivery.
Postpartum - examine the baby's upper limbs formally, inform the paediatrician, and debrief the mother. A perineal/third-degree tear check and thromboprophylaxis assessment still apply.
3. What would change in your advice if the estimated fetal weight were 4.6 kg rather than 4.0 kg?
Read the worked answer
The threshold is crossed. In a woman with diabetes and a previous shoulder dystocia, an estimated fetal weight of about 4500 g or more makes elective caesarean a reasonable and commonly recommended option. At 4.6 kg I would move from offering a choice to actively recommending caesarean section at around 39 weeks, and would say so clearly rather than presenting a neutral menu.
The reasoning is that above this weight both the absolute risk of shoulder dystocia and the risk of a persisting brachial plexus injury rise steeply, and the previous dystocia means her pelvis has already failed at 4.3 kg. Ultrasound error (±10%) is not a reason to discount a 4.6 kg estimate: it could be 4.1 kg or 5.1 kg, and the management decision must be safe at the upper end of that range. I would also confirm the scan was performed by an experienced sonographer and repeat it if there is doubt.
What would not change: her right to decline, the need to document the discussion, the glycaemic management, and the fact that induction for suspected macrosomia has not been shown to prevent shoulder dystocia or brachial plexus injury - a large trial of induction for suspected macrosomia found fewer shoulder dystocias but no difference in brachial plexus injury, at the cost of more caesareans.
If she still declines caesarean, the intrapartum plan from the previous answer applies with a lower threshold for caesarean in labour, a low threshold to stop oxytocin, and a neonatal team present. I would also discuss that if labour does not progress normally, caesarean should not be delayed in an attempt to achieve vaginal birth at any cost.