In the United Kingdom, hospital statistics for assisted vaginal delivery rates range from 4% to 26%1. Surveys suggest that the ventouse is becoming more popular than forceps, especially with the development of the newer soft silastic cups2. It has been shown that maternal morbidity is much less with ventouse compared with forceps. The ventouse is also reputed to be easier than forceps for less experienced operators.
The risk of serious neonatal injury associated with ventouse is uncertain. Many studies have demonstrated the safety of soft silastic cups compared with either metal cups or forceps3–8. Despite the alleged safety of the ventouse, it can result in lacerations to the scalp, cephal haematoma, subgaleal haemorrhage and intracranial haemorrhage due to tentorial tears. These may be life threatening. We present two cases of neonatal intracranial haemorrhage following ventouse delivery, which occurred in a district general hospital in the West Midlands within four months.
Case report 1
A 39 year old woman in her second pregnancy was admitted in spontaneous labour at term. At full dilatation cardiotocography showed a fetal bradycardia lasting for 20 minutes, and therefore an instrumental delivery was performed. A silastic vacuum cup was applied to the fetal head, which was at +1 station and in a direct occipito-anterior position. After two pulls, a baby boy weighing 4 kg was born with Apgar scores of 4 at 1 minute and 6 at 5 minutes. The cord pH was 7.26. He responded rapidly to intermittent positive pressure ventilation. However, at 10 hours of age he had an episode of apnoea and again required intermittent positive pressure ventilation with 100% oxygen. He remained hypertonic and irritable with clenched fists. He continued to have recurrent apnoeic attack and bradycardia for which no obvious cause was found. Computed tomography showed a very extensive haematoma over the right side of the brain and cerebellum, with a midline shift to the left. These findings were consistent with supratentorial and infratentorial haematomas (Figs 1 and 2). The infant was referred to the Department of Neurosurgery on his second postnatal day. He required craniotomy and evacuation of the supratentorial haematoma and a subsequent posterior fossa burr hole for drainage of the infratentorial clot. A few weeks later, he had a cranioplasty and re-insertion of the previously removed bone flap. Eventually, he made a good recovery.
Case report 2
A 34 year old woman in her second pregnancy had an uneventful antenatal period. Spontaneous onset of labour occurred at 41 weeks of gestation. A decision for instrumental delivery was made because of delay in the second stage of labour and maternal exhaustion. A silicone vacuum cup was applied to the fetal head, which was at +1 station and in a right occipito-transverse position. Traction effected no descent and the cup slipped. Therefore, a lower segment caesarean section was performed. A baby girl weighing 4.08 kg was born with Apgar scores of 8 at 1 minute and 9 at 5 minutes. The cord pH was 7.3. On the third day of her life, the infant became lethargic and was not feeding well. She also developed jaundice. Computed tomography showed a 13 × 4 mm area of increased attenuation within the brain substance in the left parietal lobe, consistent with a fresh intracerebral haematoma. There was no midline shift or hydrocephalus. There were no fractures. She was treated conservatively and was well on discharge on the 10th postnatal day. Follow up at seven weeks of age did not show any neurological abnormality.
Our two infants developed significant intracranial haemorrhage following ventouse delivery. Neither delivery was particularly difficult, and traction was not prolonged or excessive in either case. With slippage of the cup in the second case, it is possible that the ‘implosive’ force (sudden release of the vacuum) led to the haemorrhage. It is therefore possible to create a life-threatening cerebral haemorrhage in a baby in an uncomplicated ventouse delivery.
Haemorrhagic complications associated with ventouse delivery have an incidence of 0.72% and mortality of 0.2%8. The haemorrhage can be extracranial or intracranial. Extracranial bleeding occurs in three major forms: in a caput succedaneum, in cephal haematoma and in subgaleal haemorrhage. These cerebral insults result from bleeding in different tissue planes between the skin and the cranial bones5–8.
Intracranial bleeding usually presents as a subdural haemorrhage as a result of a tentorial tear. The vacuum extractor can elongate the cranium in the occipito-frontal direction and thus stretch the tentorium, leading to rupture of the vein of Galen, the straight sinus or the transverse sinus. The haemorrhage may extend both supratentorially or infratentorially. Surgical decompression is indicated in the presence of acute hydrocephalus or signs of brain stem compression.
Huang and Shen7 reported nine infants with haemorrhage as a result of a tentorial tear, of whom five were delivered by vacuum extraction. Two of them needed surgical decompression by craniotomy. The rest were treated conservatively and had a normal neurological and developmental outcome. Huang and Lui4 reported an infant born by vacuum extraction who developed haemorrhage as a result of a tentorial tear, extending inferiorly over the cerebellum and superiorly behind the occipital cortex. Suboccipital craniotomy was required. The infant was developmentally normal at four years of age.
A Cochrane systematic review9 of nine randomised trials, involving 2849 primiparous and multiparous women, showed that compared with forceps the vacuum extractor was more likely to fail to achieve a vaginal delivery, and was more likely to be associated with cephal haematoma, retinal haemorrhage and maternal worries about the baby. However, the ventouse was less likely to be associated with regional or general anaesthesia, significant maternal perineal and vaginal trauma and severe perineal pain at 24 hours and was no more likely to be associated with delivery by caesarean section, a low 5 minute Apgar score or the need for phototherapy. In each of these studies, the outcome was judged by the operator, and not by an independent blinded observer; ascertainment bias may, therefore, have occurred. Objective assessments in larger observational studies as well as a small randomised trial have shown an increased number of anal sphincter injuries with forceps10–12. However, a five year follow up of women enrolled in one of the randomised trials did not show any significant difference in long term outcome between the two instruments, for either the mother or the child8. In view of the reduction in maternal injuries, the ventouse has been considered to be the first instrument of choice for operative vaginal delivery13.
The combined evidence obtained from all available controlled trials (1175 babies in the vacuum extractor groups and 1155 babies in the forceps groups) allows conclusions to be drawn only about relatively common neonatal outcomes. Concerns about the risks of intracranial and subgaleal haemorrhage remain3,5,14–16. However, in a recent review of 583,340 live born singleton infants born to nulliparous women, the rate of subdural or cerebral haemorrhage in vacuum deliveries did not differ significantly from that associated with forceps or caesarean section during labour2. Overall, the risks of perinatal trauma using the vacuum extractor are associated with the duration of application, the station of the fetal head at the start of the delivery, the difficulty of the delivery and the condition of the baby at the start of the procedure17,18. Risks increase significantly in babies who are exposed to attempts at both vacuum and forceps delivery2. Failed vacuum extraction has been considered by some obstetricians as an absolute contraindication to forceps. However, where the infant is in good condition, an experienced obstetrician may feel that a trial of forceps is justified in certain circumstances. Certainly, if the first attempt is by the silicone ventouse cup failure is more likely, and a trial of forceps or vacuum extraction with a metal cup may be justified.
There can be no doubt that the ventouse should be considered the first instrument of choice for the majority of operative vaginal deliveries on the basis of the published data. However, although the ventouse is associated with less maternal trauma, it does not necessarily ensure a safe passage for the infant. An otherwise uncomplicated ventouse delivery can be associated with major intracranial haemorrhage in the infant. Part of the mechanism of the insult may be the ‘implosive’ force if slippage of the cup occurs.