The most popular drug for labour analgesia in Australia1, nitrous oxide, is more than 150 years old. Its administration under patient control heralded the introduction of a similar approach for epidural analgesia in the 1980s.

In the 21st century, regional analgesia (epidural, combined spinal-epidural and, rarely, spinal or continuous spinal techniques) has reached its full gestation in Australasia. As the availability and popularity of epidural analgesia gradually increased through the 1970s and 1980s, anaesthetists lauded the quality of pain relief, but midwifery and obstetric practice was burdened with, and the parturient suffered from, the undesirable consequences of the methods used to achieve the analgesic goals. Lower leg and abdominal weakness with loss of mobility, leg numbness and urinary retention were unwanted or detrimental, and controversy raged about the effect of the regional block on vaginal delivery, the fetus and the mother’s back.

The introduction of so-called ‘low-dose epidurals’, containing local anaesthetic and opioid in the early 1980s, of patient-controlled delivery techniques a short time later, and of the (initially) motor-block-free ‘combined spinal epidural’ (CSE) analgesic technique in the mid-1990s, was followed by more rigorous scientific investigation and led to improved outcomes and resolved a number of disputes.2 We now know regional analgesia does not increase the risk of a caesarean delivery or postpartum low back pain, that ‘mobile’ epidural approaches reduce the instrumental delivery rate and improve maternal satisfaction, and that epidurals are safe for the fetus (indeed preferable to one of the major alternatives, systemic opioid as intramuscular pethidine).

Given this information, it seems surprising that 25 per cent of Australian women in labour are still administered systemic opioids, predominantly intramuscular pethidine and morphine.1 This method has low efficacy against contraction pain, but induces some useful euphoria and sedation in some situations, for example, painful spurious labour or the latent phase of labour. The opioid method of patient-controlled intravenous remifentanil emerged in the 2000s, and interest has been strong.3 Remifentanil is a highly potent mu-opioid agonist, which has an exceptionally short elimination half-time, making it suitable for patient-controlled delivery. It is, at worst, likely to cause only a very transient effect on the neonate. The ‘off-label’ use of this potentially dangerous drug has found a place because the quality of pain relief achieved is the next best to regional analgesia.

So while remifentanil-based opioid analgesia is the most recent innovation, many experimental drugs are in the pipeline and regional analgesia has improved: safe ‘epidural’ services are widely available in Australia (except in small, and often rural, maternity units); and excellent pain relief can be achieved with minimal impact on the progress of labour, the delivery mode (poorer analgesia from an epidural is a marker for higher risk of operative delivery), the fetus or the neonate. Some issues remain unresolved. Good bladder care and attentive fetal monitoring are still basic requirements and the potential effect of regional analgesia on the initiation or maintenance of breastfeeding threatens to be a story akin to that of the association between epidurals and caesarean section of 15–20 years ago.

So, what is new with regional analgesia?

Drugs and techniques

Anaesthetists continue to experiment with poly-pharmacy in epidural solutions, such that drugs like clonidine have found niche roles in combination with opioid, local anaesthetic or both.4 Epidural neostigmine, an anticholinesterase drug that is much more familiar to most doctors as ‘reversal’ for non-depolarising neuromuscular blocking drugs, is the latest adjunct to show promise. For maintenance of analgesia during labour, patient-controlled epidural drug delivery (PCEA) has gradually replaced epidural infusions in many units, based on established advantages including less motor block of the lower limbs, fewer staff interventions to supplement maintenance solutions and its psychological benefits.2 The optimal combination of variables continues to be investigated and, in at least one country, has reached the sophisticated stage of computer-regulated control of drug delivery, based on patterns of demand.

The CSE technique has a number of advantages when initiating ‘epidural’ analgesia, in particular, the faster onset of pain relief, the lack of motor block, reliability in achieving pain-free labour once labour is progressing well and a reduction in subsequent epidural drug consumption.2 It is particularly useful for multiparous women in established labour, parturients in late labour or parturients for whom assisted vaginal delivery is planned. It is surprising, therefore, that only 3.5 per cent of Australian women received CSE analgesia in 2007, with epidural approaches used by 28 per cent.1 This may represent misreporting, but probably also reflects the conservatism of Australian obstetric anaesthetists – who almost 15 years after the introduction of CSE (also inappropriately called a ‘walking epidural’) are loath to change practice – rather than their concerns about safety, which has been clearly established.

Anaesthetists are currently dragging themselves into the world of ultrasonography, principally when performing peripheral nerve blocks. Advocates of pre-insertion scanning of the lumbar spine claim its benefits include better identification of the interspace to be used, fewer needle insertions in obese parturients and great promise for difficult patients with scoliosis or other deformities.5 These claims are starting to be systematically evaluated.

Altered fetal heart rate patterns

Plummeting maternal blood pressure after an initial epidural bolus and, concurrently, severe fetal bradycardia, was not an uncommon event 20 years ago, but is now much less frequent, largely because of the use of lower doses of local anaesthetic (plus opioid), attention to maternal positioning to avoid aortocaval compression and early treatment with potent vasopressor drugs. Similar patterns of fetal bradycardia, but without maternal blood pressure change, remain not infrequent following CSE analgesia. Both experimental animal and human evidence suggest the aetiology is secondary to a sudden reduction in myometrial tocolysis as serum adrenaline concentrations fall rapidly with pain relief.6 The intrathecal opioid dose also appears relevant, and trends toward use of smaller doses of intrathecal fentanyl and earlier administration of a tocolytic drug (intravenous or sublingual glyceryl trinitrate or subcutaneous terbutaline) should be encouraged, because these strategies reduce the risk of, or correct, worrisome fetal bradycardia in most cases.

Maternal urinary retention

With traditional high-concentration epidural solutions, the loss of the ability to void during labour is very high. When more modern ‘low-dose’ combination solutions are used, more women can void spontaneously, but intermittent catherisation rates are still high.7 Fortunately, most women do not consider ability to void as important as retention of strength and mobility8, but good midwifery and nursing care of the bladder is an essential component of safe practice whenever regional analgesia is used.


The benefits of breastfeeding for both the mother and baby are substantial and life long. Allegations that regional analgesia for labour either impairs the initiation of breastfeeding or is causative in the earlier abandonment of breastfeeding have arisen in recent years. This is an emotive topic and a question anaesthetists have an obligation to address. To date, the evidence suggesting a causal link is extremely weak, but well publicised!9 Indeed, it is not plausible that a method that has no major impact on the neonate at birth should lead to cessation of infant breastfeeding activities weeks or months later. The best current evidence, although not watertight, indicates that, unlike systemic pethidine, there is no impact on the initiation of breastfeeding or reduction in its duration after regional analgesia for labour and delivery.10

Developments in alternative methods of analgesia Non-pharmacological approaches to the management of labour pain continue to play a substantial role – 25 per cent of Australian women do not use analgesic drugs, although what proportion use non-pharmacological strategies is unknown. Meta-analysis confirms our suspicions – acupuncture does not have a useful effect11, nor does transcutaneous electrical nerve stimulation12, use of which anecdotally appears in the decline. Meanwhile, simple measures such as water baths are in vogue and lumbar water blocks for back pain13 have been embraced by a few. Time-honoured and safe methods, such as patient-controlled nitrous oxide analgesia, remain very popular and mu-opioid analgesics are used by a quarter of our labouring population.1 Remifentanil patient-controlled analgesia (PCA) appears the next most effective method after regional analgesia3 and low doses of intravenous ketamine show good efficacy, but have been inadequately investigated.14

In many units, remifentanil PCA is now used for ‘fall-back’ when regional analgesia is contraindicated, although a few units make this method more freely available. The major drawback is the danger of profound maternal respiratory depression, even from a small overdose – this is why remifentanil was only licensed for administration in highly monitored environments, such as the operating room and an intensive care unit, where airway management and resuscitation facilities are optimal. In the delivery unit strict regimen protocols, monitoring policies and midwifery education are mandated. To date experience is good3, but the importance of vigilance and systems that minimise drug errors and other mishaps cannot be overemphasised.

And so to the future

Stargazing is a risky pastime! Research directions include the search for new effective, safe, long-acting analgesic drugs suitable for neuraxial administration, whether local anaesthetics or opioids in lipid-based emulsions or other vehicles that provide sustained drug release; or new therapeutic intrathecal analgesics, such as calcium-channel blockers, non-steroidal anti-inflammatory drugs, adrenaline-reuptake inhibitors; or new systemic analgesics such as kappa-opioid receptor agonists and dexmedetomidine. To date, none of these are approved for clinical use. Other paths of investigation are the prediction of those women most likely or least likely to need or benefit from regional analgesia, based on measurement of pain thresholds, endogenous opioid effects, genetic make up or psychometric testing.8

Is it possible that the days of catheterisation of the vertebral canal are numbered? Perhaps, but not in the near future. For those who want them, we should strive for the time being to improve the availability, safety and clinical utility of our most effective methods, namely epidural, spinal-epidural and remifentanil analgesia.