Failed Vacuum-Assisted Vaginal Delivery: Tool, Training Or Technique?

“There is a skill about vacuum extraction which tests the mind and hands of the operator in a finer way than do forceps deliveries – for this is a gentleman’s way of delivering a baby…”
Geoffrey Chamberlain, Turnbull’s Obstetrics, 1989


Successfully assisting a vaginal delivery reduces the number of parturients who would otherwise have had Caesarean sections in the second stage of labour1,2. Whilst the rate of delivery by Caesarean section continues to rise, it is estimated that the UK national rate for assisted vaginal delivery remains steady in the region of 10-15%2. The use of the obstetric ventouse has superseded the obstetric forceps as the instrument of first choice, the rise in the former mirroring a fall of use of the latter. This probably reflects the belief that there is a lower risk of causing fetal or maternal trauma with the ventouse; influenced by the assumption that a ventouse delivery requires less technical skill than the forceps and is therefore inherently safer4. Many clinicians believe that vacuum-assisted delivery has a ‘favourable opinion of both patient and midwife’6. It is undoubtedly the case that the newer designs of ventouse are less intimidating to the parturient, and equally importantly, have, as yet, not been the subject of too much ‘adverse publicity’. However, assisted vaginal delivery with the vacuum device is more likely to fail than with forceps, with important consequences for the baby, the mother and the accoucheur.

A vacuum-assisted vaginal delivery may be considered to have failed, if alternative or additional procedures are required to achieve delivery. The reported failure rates vary widely, ranging from 9.2% in an Australian study7 to 18.3% in London8 and 74.5% at the Liverpool Women’s Hospital9. When conventional vacuum-assistance devices were compared with the more recent Kiwi device in two studies, the Kiwi ‘failed’ in 12.8% versus 9.4% of instances10; and 34.4% of instances versus 21.4%11 respectively.

These additional procedures are not always complication-free. In a review of singleton fetal and neonatal deaths associated with cranial trauma and cephalic delivery during a national intrapartum-related confidential enquiry12, more than one instrument was used for assisted vaginal delivery in 69% of cases.

Even when an abdominal delivery is resorted to because the vaginal vacuum-assisted approach fails, the fetal risks remain. In addition, the significant maternal risks of extension of uterine incision, traumatic post-partum haemorrhage and lower genital tract trauma have been regularly documented13,14. Indeed the recent RCOG guideline on Operative vaginal delivery3 points out that an unsuccessful vacuum-assisted procedure amongst other adverse events should trigger an incident report and review if necessary.

Recent publications and presentations indicate that failed vacuum-assisted deliveries remain a persistent reality of obstetric practice. In this paper, we attempt to explore questions about the tool, training and technique, and suggest ways of improving upon clinical practice.



Whilst the perfect tool makes up for all the ‘short-comings’ of the user, there are a few such tools in existence.

Like most obstetric tools in common use, the vacuum assistance devices were initially crude devices, whose basic principle was based on a perceived need. That James Simpson in Edinburgh initiated a concept, which was eventually made functional by Tage Malstrom in Sweden15, lends credence to the practice of continued refinement and improvement of such devices till this day. Over the decades, product development has evolved in terms of relative safety, ergonomics, user-friendliness, disposability and cost.

By applying safe limits of suction to the compliant fetal scalp soft-tissue, vacuum-assistance devices are designed to exert a tractional force on the fetal head, increasing the flexion of the presenting part and reducing the diameters ‘presenting’ to the pelvis. They occupy comparatively little additional space (comparative to the fetal skull) in the maternal vagina and are therefore less associated with vaginal trauma than the obstetric forceps.

The applicator cup has evolved from a rigid metallic device to a softer silicone one. In a similar vein, the dimensions of the cup have been modified not only to improve placement, but also to ensure ease of use without it becoming unwieldy7. Over time the traction facility has changed from a metallic rod to chains to rubber tubing’s of diminishing calibre but increasing tensile strength. The suction generators were previously piston-principle based, utilising either hand or foot powered units, to external electric-powered suction generators to the current ‘palm-held’ devices.

With ongoing fears of cross-contamination and infection despite rigorous sterilisation techniques, reusable instruments are being replaced more and more commonly by single use devices or part devices. Whilst this may raise fears of increasing cost per item used, careful product development and marketing have, to date, ensured that this is not the case. The currently available disposable ventouse devices are not without their critics and it remains to be seen as to whether further developments to the intrinsic design of the ventouse improve its efficacy and safety profiles.

Table 1: Evolution and development of the vacuum-assisted delivery device (UK selection)

Name Hallmarks Advantages
Malstrom cup (1954) Metal Cup Novel alternative to the forceps.
Bird’s modification (1969) Separate vacuum channel, centrally-fixed traction chain on cup. Better torque and traction capability.
Bird’s posterior cup (1976) Laterally-positioned vacuum channel on cup. Tangentially enabled traction. Improved application to malpositions of
the fetal occiput.
Silk cup (Silc) Silicone-rubber cup Chignon is not as severe
Kiwi Omnicup Hand-held, all-in-one
‘Easy to use.’ No need for a power source.
‘Whole’-disposable. Light-weight.

It has to be pointed out that, excellent and safe use can be delivered by non-Kiwi vacuum-assistance devices as well.



Consistent failure of an accoucheur to successfully achieve a vaginal delivery when using a vacuum device to is but one reflection of a possible deficiency in training or technique. Other indicators may be: resorting ‘early’ to a Caesarean section rather than employing a second instrument (with its increased risks of inflicting repeated fetal or maternal trauma); a higher than average rate of vaginal lacerations or fetal scalp abrasions; a need to consider all deliveries as a ‘trial’ in theatre’ – with increased risks of psychological trauma to the birthing woman and her partner.

Failed vacuum assistance was the indication for 61% of second-stage Caesarean sections in two District General Hospitals in Northern Ireland13. It is of note that in this study the majority of the failed procedures were undertaken outside office hours, with a Consultant being present during decision-making and delivery in only one third of the instances.

In a small but remarkable study, a Consultant Obstetrician personally re-examined women within 15 minutes of a Registrar’s request for approval to proceed with caesarean delivery on the basis of a clinical assessment of the feto-pelvic relationship16. It was noted that the ‘registrar-perceived position’ of the fetal occiput was wrong in 44% of the patients and the station / degree of descent of the head did not tally closely in 81%. More worryingly the station was deemed to be higher in the majority of women re-examined. It turned out that 63% of the women eventually had successful assisted vaginal deliveries, a testament to the experience of the attending clinician and a reinforcement to the belief that Consultants should be present on the labour ward ’24/7′.

Skill deficiencies may also be suggested from an audit of inter-operator failure rates of vacuum-assisted deliveries. A very small retrospective audit from Middlesex17 showed failure rates ranging between 0% to 22% amongst 5 middle-grade staff. Again, in the majority of the failed instrumental procedures, there was a malposition of the fetal head. The authors stress the need for ‘expert tuition, targeted clinical training and senior supervision’.

Two poster presentations at the Specialist Registrars in Obstetrics and Gynaecology (SPROGs) Conference in Cambridge in 2005, reinforced this further. In an anonymous telephone questionnaire survey of 72 registrars from North London, only 15% reported that they had received any training in the use of a vacuum delivery technique from a consultant18.Yet about 75% of them had performed less than 20 ‘supervised’ assisted procedures before being allowed to undertake them on their own! In a case review series from Ninewells Hospital in Dundee, the ‘provision of supervision’ was poorly documented in 63% of instances where instrumental vaginal deliveries were undertaken by junior clinical staff19. Whether this reflected a true correlation with the actual degree of supervision was not made clear, but is none-the-less, a cause for concern.

Skills training courses in obstetrics in the UK – Advanced Life-Saving Skills in Obstetrics (ALSO) and Management of Obstetric Emergencies and Trauma (MOET) have incorporated some mannequin / model skills stations in the use of the vacuum-assisted devices. However, it is not expected that the knowledge gained at these courses will replace acquired psychomotor skills honed by formal hands-on supervision.

A training video on the correct use of the Kiwi Complete delivery system20 may help, and in our opinion, is most instructive to watch. For all practical purposes, it appears that determining the application distance/flexion point requires a little ritual of measurement of distances to the maternal fourchette. A casual observer of most vacuum-assisted deliveries would be unlikely to see this regularly and religiously undertaken. However, the markings on the ‘traction tubing’ are there for a purpose!

Furthermore, many observers on the delivery suite would be able to recall the blatant act of applying Hibitane cream to all aspects of VAD devices. The lubricant is meant to be applied to the outer surface of the cup as application to the rim and pad will only increase the failure rate!

The consolation to the few perfectionists was also obvious in the (thankfully un-edited) demonstration by the Dr Aldo Vacca: the chignon of one of the babies delivered appeared to be behind the flexion point. In the other, it appeared a little anterior. That, even in the hand of an expert, who developed the Kiwi Omnicup!



Any technique should be expected to improve with experience. However, if a skill is badly taught (and learnt), it is stubbornly perpetuated and through a cascade process disseminated to others!

Earlier vacuum-assisted devices probably had general instructions for use, based on the manufacturer’s recommendations – advice that has remained largely unchanged (and unread) over decades. More recently developed devices, e.g. the Kiwi-Omnicup comes with specific guidance. It states, among other things that correct technique…should not depend on traction alone to effect delivery of the baby22. It is a common misconception that the ventouse works mainly through the strength, rather than the skill, of the accoucheur.

In the Operational Guidelines which accompany the MityVac vacuum assist delivery system (Prism Healthcare, Cooper Surgical Inc, Trumbull USA), there is also a basic reminder about an important pre-requisite: …’do not use device if flexion point is not accessible…’.

Another cautionary note was sounded by a Public Health Advisory document by the US Food and Drug Administration (FDA) some 9 years ago21. It recommended that persons who use vacuum-assistance devices: ‘be versed in their use …’ and…’read and understand the devices instructions.’ For emphasis, the FDA document specifically warned against rocking movements or applying torque to the device. In our own experience, the former manoeuvre appears to be associated with an increased incidence of lateral lower vaginal wall lacerations, the latter with an increase in the degree of trauma ‘experienced’ by the fetal scalp.

The next point of note is the positioning of the accoucheur. McQuivey,23 an author affiliated to Clinical Innovations, recommends that the accoucheur either sits or kneel as he believes that this would encourage pelvis-axis traction in the correct direction. This advice may also apply to non-Kiwi devices. A few experienced clinicians would however, point out that kneeling may allow the accoucheur to innocuously exert an extra force on the presenting part proportional to his or her own body weight! It is our belief that adopting this technique is only likely to increase the rate of detachment of the ventouse cup and will not help to increase the success rate of delivery. It is of interest that whilst an episiotomy is best performed at the time of crowning of the fetal head in cephalic presentations (if needed), when performed early in the course of vacuum assisted delivery it may enable traction to be more effectively directed24. Accoucheurs who have had to resort to a Caesarean section for failed vacuum assistance after an “ill-timed” episiotomy, will be very reluctant to face this embarrassment again! We would therefore suggest that an episiotomy only be performed when a vaginal birth is strongly anticipated and it is felt that the perineal / vaginal tissues are contributing to an outlet ‘obstruction’.

Table II: Suggested benchmarks for training in vacuum-assisted delivery.

Process Component
1 Attending a training workshop with a trainer model, with major emphasis on fail-safe methods on ascertaining position of the fetal occiput/flexion point.
2 Watching a demonstration video*
3 Watching a vacuum-assisted delivery by an Experienced supervisor
4 Hands-on supervision by Experienced Supervisor
5 Performance of vacuum-assisted delivery under direct supervision
6 Documentation of correctness of ‘chignon’ site after each procedure
7 Personal percentages of failed procedures per year (audit target)
8 Performing not less than 25 procedures a year (i.e.: 2/ month).

One of the oft-mentioned shortcomings of the Kiwi-device, and of vacuum-assisted devices in general, is their likelihood to fail with the occiput-posterior positions of the fetal head3,11,24,25,26. In advancing the logic of manual rotation as an option of management, some Clinicians actually use the suction-activated cups to attempt such rotations! Unsurprisingly, the manufacturer’s instructions do not mention this use at all22 and accoucheurs should be aware of the increased risk of fetal scalp trauma.

Abandonment of a vacuum-assisted delivery commonly follows cup detachment. One way of reducing such ‘pop-offs’ is to keep an eye on the traction-pressure guage, and to pre-emptively top-up the suction once the pressure begins to fall. This works safely, provided that the flexion point was correctly identified, traction is in the right direction and delivery continues to be deemed to be feasible.

Tasks and thoughts


There are many ways of doing things well. Good training ensures that correct skills are acquired and built-upon. Most reputable corporate marketing bodies have a strong user-training and after-sales/user-update programme. We are often impressed by study days for users of blood glucose monitors, volume infusion devices etc.

Yet, the safety and danger of vacuum-assisted devices are taken for granted.

We were unable to find a document that mentioned a ‘conversion’ or ‘half-day adaptation course’ for very experienced users of the ventouse/metal cup, who were switching to use of other devices. One still hears the occasional moan of ‘…If only there was a metal cup, I would have delivered the baby…’

There are non-metallic cups which are known to offer satisfactory and scalp-safe potentials. The problems reported by many users and centres may be due to this assumption, and a failure of proper comparison by appropriately trained personnel.

Accoucheurs using the vacuum-assistance devices may want to be reminded there is a high likelihood of failure27 in the following selected circumstances (the comments in parenthesis are ours):

  • Epidural analgesia (dense motor blocks are notorious)
  • Fetal scalp or caput not visible without parting maternal labia (station may even be higher than estimated)
  • Weak infrequent contractions (syntocinon augmentation may be enough to propel towards un-assisted delivery)
  • Malpositions of the fetal occiput
  • Prolonged second stage of labour (it is important here, to rule out a cephalo-pelvic disproportion)
  • Suspected fetal macrosomia (be wary of heroics, especially if Mum is a diabetic)

To observe that a particular tool is likely to fail, compared with another, suggests an inherent shortcoming of such a tool. It is assumed that the tool-users are not only trained, but probably expert. There is also a presumption that all use/attempts have been undertaken under ideal/feasible circumstances. Of course, this is impracticable.

However it has also been pointed out that training in the use of the vacuum delivery devices is probably cursory, unacceptably brief and probably taken for granted.

There is no doubt that any technique may be perfected with structured learning in association with exposure to practical training. With the reduction in clinical exposure consequent upon changes in clinical training, alternative methods of skills acquisition and development must be put in place, if the clinical abilities of each accoucheurs are to be enriched28. Unfortunately there is no published evidence that there are currently suitable tools in the system to facilitate this.

It is noteworthy that,in each hospital delivery suite, Midwives can readily identify a particular accoucheur who is ‘almost always more likely to succeed’ in vaginally delivering a baby by vacuum-assistance.

Why then, do some appear to consistently succeed?

The foregoing perusal of published work suggests that the answer lies between tool, training and technique – and probably, all three, with careful patient selection!

Ten Tips On Technique

1 Select patients carefully: not all babies are best delivered by VAD, even if all criteria are met. This is especially true in the presence of a dense epidural or spinal blockade. Forceps may be more suitable. Individualise the management.
2 Ascertain that uterine contractions are truly strong and of adequate duration (syntocinon augmentation may be more appropriate than VAD).
3 In lithotomy position, ensure that maternal buttocks only minimally overhang the delivery table (to avoid trauma to the lower maternal spine)
4 Lock the wheels of the table.
5 Educate the parturient about what you are about to do – informed consent. Encourage her to give her best ‘push’ yet, about how many times you may pull, and the risks of the use of a second instrument.
6 If doubtful about the likelihood of success, undertake procedure in theatre. (NB the psychological trauma of moving a patient from the relative seclusion of a delivery room to the full glare of an operative suite with its multitudinous attendees should NOT be underestimated!)
7 Start ‘unsupervised’ practice with ‘lift-out’ procedures before attempting VAD of high station, direct-occiput posterior ositions.
8 Except in ‘lift-out’ procedures, the first pull seldom dramatically shifts the head.
9 If cup pops-off, re-examine the landmarks of the fetal head before re-application
10 If it was a ‘difficult’ VAD, it may be that you over-estimated the station of head descent, the application distance or fetal size!
  1. Thomas J, Paranjothy S. The National Sentinel Caesarean Section Audit Report. Royal College Of Obstetricians and Gynaecologists, London. RCOG Press, 2001.
  2. Ferguson E, Owen P. The second stage of labour (Review) Hospital Medicine (2003) 64(4) 210-213.
  3. Royal College Of Obstetricians and Gynaecologists. Operative vaginal delivery. Guideline No 26. RCOG, London (2005).
  4. Hayman R. Instrumental vaginal delivery. Current Obstetrics and Gynaecology (2005) 15: 87-96.
  5. Sachs BP, Castro MA. The risks of lowering the caesarean- delivery rate(Sounding Board). The New England Journal Of Medicine (1999) 340(1) 54-57.
  6. Johanson R. Obstetric procedures. In: Dewhurst’s Textbook Of Obstetrics and Gynaecology for Postgraduates. 6th Edition. (ed: DK Edmonds) Blackwell Science(1999). Pp 308-329.
  7. Vacca A. Vacuum-assisted delivery : an analysis of traction force and maternal and neonatal outcomes. Australia and New Zealand Journal Of Obstetrics and Gynaecology (2006) 46(2) 124-127.
  8. Cotzias CS, Paterson-Brown S. Ventouse: use, cup placement and success. Journal of Obstetrics and Gynaecology (1998) 18(4) 318-321.
  9. DeSilva A, Karagolopolous S, Patel C et al. An audit on rotational operative vaginal delivery. Conference Documentation,SPROGs Annual Conference, SECC Glasgow, June 2006.
  10. Hayman R, Gilby J and Arulkumaran S. Clinical evaluation of a ‘HandPump’ vacuum delivery device. Obstetrics and Gynecology (2002) 100: 1190-1195.
  11. Attilakos G, Sibanda T, Winter C et al. A randomised control trial of a new handheld vacuum device. British Journal Of Obstetrics and Gynaecology (2005) 112: 1510-1515.
  12. O’Mahony F, Settatree R, Platt C et al. Review of singleton fetal and neonatal deaths associated with cranial trauma and cephalic delivery during a national intra- partum related confidential enquiry. British Journal Of Obstetrics and Gynaecology (2005) 112(5) 619-626.
  13. McKelvey AC. The high maternal morbidity of second stage caesarean section. Poster presentation. SPROGs Conference(2005). Abstract Documentation. Robinson College, Cambridge.
  14. Murphy DJ, Liebling RE and Verity L. Early maternal and neonatal morbidity associated with operative vaginal delivery in the second stage of labour: a cohort study. Lancet(2001) 358: 1207-1212.
  15. Miksovsky P, Watson WJ. Obstetric vacuum extraction: state of the art in the new millennium. CME Review article. Obstetrical and Gynecological Survey (2001) 56(11) 736-751.
  16. Olah KS. Reversal of the decision for caesarean section in the second stage of labour on the basis of Consultant vaginal assessment. Journal Of Obstetrics and Gynaecology (2005) 25(2) 115-116.
  17. Iffland CA, Gordon H. Failed forceps and ventouse: an audit of registrars in training. Journal Of Obstetrics and Gynaecology(1996) 16:83-85.
  18. Gupta S, Milestone N, Loo C et al. Instrumental vaginal deliveries: are we training the trainees? A telephone questionnaire of experience and views of Registrars in North London. SPROGs Conference (200%. Abstract Documentation, Robinson College, Cambridge.
  19. Ramalingham U, Youssef R and Murphy DJ. An audit of instrumental vaginal delivery, 1998-2002. SPROGs Conference(2005). Abstract Documentation, Robinson College, Cambridge.
  20. Vacca A. Vaccum extraction-reducing the risk. Clinical Innovations. Videotape© 09/2000. Running Time 38 mins.
  21. FDA Public Health Advisory: Need for CAUTION when using vacuum assisted delivery devices. US Food and Drug Administration. May 21, 1998. (
  22. Kiwi® Complete Vacuum Delivery System with Palmpump. Instructions for use. US Patents 6,059,795; 6,074,399; 6,355.047 and other patents pending. Clinical Innovations.9.P/N 050-0142 Rev.C
  23. McQuivey RW. Vacuum-assisted delivery: a review. Journal Of Maternal Fetal and Neonatal Medicine (2004) 16: 171-179.
  24. Chamberlain G, Steer P. ABC of labour care. Operative delivery. Clinical Review. British Medical Journal (1999) 318: 1260-1264.
  25. Johanson R, Menon V. Soft versus rigid vacuum extractor cups for assisted vaginal delivery. The Cochrane Database Of Systematic Reviews. In: The Cochrane Library, Issue 2. Update Software, Oxford (2000).
  26. Groom KM, Jones BA, Miller N et al. A prospective randomised controlled of the Kiwi Omnicup versus conventional ventouse cups for vacuum assisted vaginal delivery (2006) 113(2) 183-189.
  27. Vacca A. Vacuum-assisted delivery: Improving patient outcomes and protecting yourself against litigation. In: A Supplement to OBG Management(2004) S1-S4.
  28. Sau A, Sau M, Ahmed H et al. Vacuum extraction: is there any need to improve current training in the UK? Acta Obstetricia et Gynecologica Scandinavica (2004) 83: 466-470.
Categories: ARTICLES

About Author