Authors' reply: A re-evaluation of the role of rotational forceps: retrospective comparison of maternal and perinatal outcomes following different methods of birth for malposition in the second stage of labour



This article is corrected by:

  1. Errata: Erratum Volume 121, Issue 10, 1322, Article first published online: 25 August 2014

Authors’ reply


We thank the readers for their interest and for raising several important questions on our paper.[1]

We thank the readers[2-4] for highlighting the problem of many obstetricians abandoning the practice of vaginal operative deliveries nationally and internationally in favour of primary caesarean section. Safe use of rotational forceps deliveries needs high-quality training. We believe that the UK obstetrics specialist training programme should aspire to provide the necessary level of advanced training to deliver safe acute obstetrics services to women.

We agree that the data from the Kielland forceps group shows a higher (but comparable to the rates previously reported for nonrotational vaginal deliveries) rate of obstetric anal sphincter injury (OASIS).[2] Perineal injury is an unavoidable risk with vaginal birth including rotational deliveries when compared with caesarean section. However, we do not think our data are suitable to answer the readers’ question whether differential OASIS rates occur in deliveries completed by trainees or consultants. OASIS rates were relatively rare (total of 24) in our retrospective data set and it only contained data on the operator who completed the delivery not the doctor who initiated the delivery. For example there were three in deliveries completed by a consultant (rate of 2.2%) and 13 in deliveries completed by a senior registrar (rate of 2.6%). It is possible that many of the deliveries completed by consultants/senior trainees would have been initiated by a junior trainee. Although we agree with the readers that OASIS is an important outcome measure for rotational forceps, this research question requires a larger, prospective data collection to be answered conclusively.

In our hospital, trainees first get to observe and perform several Kielland forceps births under direct supervision by a trainer, then they must have a minimum of three competent OSATS performance assessments by a consultant highlighting that the trainee is competent in accordance with the Royal College of Obstetricians and Gynaecologists training. The number of directly supervised Kielland forceps deliveries that each trainee requires before achieving competency varies in a similar way to when trainees are supervised for other surgical procedures such as caesarean sections. For example, not all senior registrars who train in the Liverpool Women's Hospital achieve competence in Kielland forceps deliveries during their usual 12-month placement. The technical ability required to perform a Kielland forceps delivery is relatively easy to teach and assess; however, the decision-making, patient selection and rationality of the trainee to recognise the limitations of the instrument that make the procedure safe are all more difficult to teach and to assess. The Liverpool Women's Hospital obstetrics unit with a large number of deliveries (largest in the UK) and high consultant presence in the labour ward, has the necessary infrastructure to safely teach this advanced skill without usually needing an extended time in training. In smaller units or units that already have low instrumental delivery rates or where Kielland forceps use has been abandoned, reinstating Kielland forceps practice/training is likely to be difficult.

Simulated training may have a place in introducing the trainees to Kielland forceps and other rotational delivery methods[2] but we do not believe that it has the ability to replace the clinical training/assessment of operators of rotational deliveries. A competent performance on a simulator may give false reassurance on a practitioner's ability because it is mainly assessing technical ability in a non-clinical setting. Technical ability is only part of the overall skill needed for the safe use of Kielland forceps. Therefore we would like to stress the importance of high-quality clinical training in advanced obstetric practice in the labour ward, delivered by increased consultant presence but we cannot envisage a reliable alternative that is either easier or quicker to deliver.

We therefore propose a national prospective data collection on all deliveries in second stage with malposition to collect comparative and accurate safety data to guide future practice.

We disagree with Gale et al.[3] regarding their comment on a discrepancy between our results and the abstract. The data presented in our manuscript supports the conclusion that the Kielland forceps deliveries are likely to be the most effective method without any statistical evidence of an increased risk in the maternal or neonatal outcomes when compared with rotational ventouse or primary caesarean section. Our abstract and the full text paper clearly state that the data suggest that the neonatal and perinatal outcomes are comparable, without any statistical evidence of differences between all three methods. Therefore we conclude that all methods appear to be equally safe and we did not comment that Kielland forceps delivery was the safest method.

The second point raised by Gale et al.[3] on lack of data on station, presence of caput/moulding and contractions has already been mentioned in our discussion. Although we agree that this is a limitation in our retrospectively collected data set, we would also like to point out that all these factors are assessed subjectively without any standardised, reproducible or universally accepted method to measure them. We can therefore never be sure of the similarities and differences between the women undergoing operative deliveries in the second stage. Furthermore, we argue that despite the lack of these data it is likely that the women who underwent primary caesarean section are likely to have high station of the head and there may be a selection bias in this group. And this selection bias may select caesarean section deliveries that are going to be technically easier with fewer associated complications than those primary second-stage caesarean section deliveries with low station of the head. Therefore we do not believe that the outcomes considered are favouring Kielland forceps due to lack of this information.

As previously mentioned, operative deliveries in the second stage for malposition are complex acute clinical situations needing experienced clinicians to provide advanced obstetric care.

Although many obstetricians may perceive and promote recourse to primary second-stage caesarean section without trial of rotational delivery for malposition in the second stage, there is no clear evidence to show superior safety outcomes with this as the standard obstetric practice.

If the reduction in training time does not allow a standard trainee in the UK to achieve fluency with the rotational delivery method of choice (and difficult second-stage caesarean section), this needs to be addressed appropriately by the Royal College of Obstetricians and Gynaecologists and the national training programme. However, we do not accept the answer to this lack of suitable trainers or training opportunities is the abandonment of an advanced obstetrics skill that is clearly useful in reducing caesarean section without increasing the risk of maternal and neonatal adverse outcomes. Almost all large obstetric units in the country seem to have obstetricians with the required skills to perform and teach rotational forceps deliveries. Therefore, the units lacking this particular skill via collaboration can potentially acquire proficiency to offer training.