What babies want

I write to congratulate Professor Reynolds and Dr Seed on their meta-analysis of neonatal acid-base balance and anaesthesia for Caesarean section [1] and Dr Clyburn on his thoughtful accompanying editorial [2]. However, I need to temper this congratulation with some adverse comments about the closing sentence of an otherwise interesting and balanced paper. I will reproduce it in full just in case your readers do not have immediate access to a copy of July's issue: ‘If I were an at-risk baby about to be delivered by Caesarean section, I would say: “Please give Mummy epidural anaesthesia using an S-enantiomer local anaesthetic plus opioid, supplementary oxygen and left uterine displacement. If spinal anaesthesia is a must, please use low-dose bupivacaine with opioid and phenylephrine for hypotension” Having dwelled for the previous 14 pages of this paper very much in the area of Level 1a of the Oxford Centre for Evidence-based Medicine classification [3] it is a disappointment to see the plummet so suddenly to Level 5 at the conclusion of an otherwise largely factual paper. This childish approach is unworthy of the paper, the journal and the subspecialty of obstetric anaesthesia. Although the data presented suggest, but by no means prove, that epidural anaesthesia may provide an advantage to the ‘at-risk’ fetus, they support none of the other suggestions made in the closing sentence. Indeed, all that has been shown is a small but statistically significant difference in funic pH. Does this necessarily translate into a clinical difference in neonatal condition at birth or eventual outcome? The results analysed are mostly from non-randomised studies of women undergoing elective Caesarean sections; great caution should have been exercised in extrapolating these results to emergency procedures and at-risk babies. It is surprising that the authors, neither of whom is a practising clinical anaesthetist, feel able to provide clinical advice. I am fearful of the potential consequences of this advice. Let us imagine a trainee anaesthetist faced with an emergency Caesarean section for fetal distress. Unfamiliar with administering de novo epidural anaesthetics but having read this paper and the advice given at its conclusion, the trainee inserts an epidural and injects local anaesthetic down the catheter. Pressurised by the obstetrician, who is anxious about the baby's well-being, the trainee is persuaded to permit the operation to start before the anaesthetic is of sufficient depth or extent. We are all unfortunately familiar with the scenario that would follow – would this be in the interests of the baby or the mother, let alone the anaesthetist or surgeon? I strongly believe that more studies need to be performed before definitive advice like this is offered, and that the journal should not give non-clinicians the opportunity to give unproven and, in my opinion, poorly thought-through clinical advice. I wholeheartedly support the wise counsel provided by Dr Clyburn in his editorial (a practising clinician expressing opinion in an appropriate location in the journal): he would use spinal anaesthesia while taking steps to minimise any adverse impact of the anaesthetic on the at-risk baby. I am impressed that Dr Hartle had the diligence to read our paper to the bitter end (a well-known yacht club in the British Virgin Islands), but only sorry that he found the end so bitter. Please, Dr Hartle, do not include my long-suffering statistician in the blame. Blame the intellectual baby from whom I quoted. Childish approach, yes indeed, that was the aim. When determining management I agree absolutely that maternal welfare must come first, but many discount or mistake what is best for the baby. I am, however, at a loss to know what element of the baby's request is not supported by good evidence. I feared that had I simply said the best interests of the baby would be served by epidural or general anaesthesia (as supported by our findings), people would be at pains to point out the dangers of each. Surely the greater safety of S-enantiomers than of racemic bupivacaine in the event of intravenous misplacement is evidence-based? Meanwhile, although our own analysis only concerned types of anaesthesia, a review is more useful if it is comprehensive. Thus the various elements of care that might also influence fetal welfare and have been addressed extensively elsewhere needed to be included, hence the labourious length of the discussion (and of the baby's request). Dr Hartle states ‘all that has been shown is a small but statistically significant difference in funic pH.’ Not so, the base excess was also significantly different, which is the more remarkable outcome, since pH can merely reflect a difference in maternal ventilation, while a base deficit reflects a genuine metabolic acidosis. We acknowledged in the discussion that the differences are small, but I wished to emphasise the direction of the differences. Funic acid-base balance is one of the few indices of neonatal welfare that actually correlates with outcome and is the best measure of the effect of intrapartum intervention [1]. This is why it interests me. There is often a problem that results may not be applicable to all elements of a population. The lack of data from randomised trials on children is a case in point. Randomised trials commonly limit recruitment to uncomplicated cases, hence it is difficult to find evidence that relates only to compromised infants. In our review, however, two randomised trials addressed only severe pre-eclampsia [2, 3]; in one, comparing spinal with general anaesthesia, all babies had abnormal cardiotocograph traces [3]. The results were both significant and consistent with the overall results. Non-randomised studies were in the main limited to elective cases, in which babies are likely to start from a similar baseline. Such comparability would be unlikely in non-randomised emergency cases. Incidentally (and more power to my critics), since our analysis, a randomised trial comparing spinal with epidural anaesthesia in severe preeclampsia has been published [4], showing no significant difference in neonatal outcome. More such trials are clearly needed. I trust that, even with the difficulty in training that exists today (yes, even I have heard about that), a young anaesthetist would know better than to attempt de novo epidural anaesthesia when there is a need for haste, but I hope that they might be happy to top up a functioning epidural catheter. I also hope to dissuade anaesthetists from telling the scared young mother who wants general anaesthesia that it would be bad for her baby. F. Reynolds St Thomas' Hospital London, UK E-mail: [email protected]