Although epidural analgesia has an excellent track record for relieving the pain of labour, a proportion of epidurals fail to deliver adequate pain relief. Approximately 10% of women will have initially unsatisfactory blocks, and around 2% of these will be persistently inadequate. Poor blocks may be divided into those involving inadequate spread, and those where spread is apparently adequate but the degree of block is insufficient.
Problems and special considerations
Inadequate spread
This may be one of several types:
• A block that, while bilateral, is less effective on one side of the body than the other, usually due to insertion of an excessive length of catheter into the epidural space.
The unequal block is thought to result from placement of the catheter tip to one side of the epidural space, and even its escape from the epidural space via a lateral foramen. This may even result in a very limited block involving only one or two dermatomes. Lengths greater than ~5-6 cm are generally regarded as at increased risk of transforaminal escape of the catheter tip, while lengths less than 2-3 cm are more likely to result in the catheter’s falling out, especially during a long labour.
• Satisfactory cranial and caudal spread, but limited to one side of the body only. Contrast studies of this type of block have shown that it is associated with distribution of fluid to one side of the epidural space. The most likely explanation is the existence of a dorsal midline septum arising from the posterior aspect of the dura mater, which acts as a barrier to the free spread of local anaesthetic. Unilateral block has also been shown to be more common in cases of scoliosis, and this is also presumed to be due to anatomical barriers to spread of local anaesthetic in the epidural space.
• Missed segment, whereby one or more segments remain unblocked despite normal analgesia above and below. This was more commonly seen when low volumes of concentrated local anaesthetic solution were routinely used during labour, but is relatively uncommon with the higher-volume, low concentrations used in modern techniques. Missed segments were thought to arise from isolation of some nerve roots from the local anaesthetic by longitudinal septa or air used in loss-of-resistance techniques, or from relative resistance of larger nerve roots (typically lumbar) to penetration by local anaesthetic.
• Limited cranial or caudal spread despite attempts to extend it with further doses, for example because of a horizontal septum preventing flow or presumed scarring in patients who have undergone spinal surgery.
• Catheters with single terminal eyes are commonly used in the USA, whereas most UK practitioners prefer multi-holed, blind-ending catheters. Studies have demonstrated a higher incidence of unsatisfactory blocks with the former, mostly due to unilateral blocks and missed segments. This is probably due to a ‘streaming’ effect, whereby all the solution is directed along a single track, encouraging longitudinal spread at the expense of lateral flow.
Inadequate density of block
A block that is insufficient despite apparently adequate spread may occur when excessively dilute mixtures of local anaesthetic are used, when the intensity of pain is increased (e.g. occipito-posterior position, or placental abruption or impending uterine rupture), or in some women who just seem to have lower pain thresholds. A differential block may also be seen; this may be due to different nerve fibres being affected variably by the local anaesthetic, or to variable central somatosensory integration.
Other maternofetal factors may be associated with an increased incidence of breakthrough pain during labour epidural analgesia. These include nulliparity, a raised maternal body mass index (BMI), a longer duration of labour, a larger fetus and abnormal fetal presentations. Rarely, severe breakthrough pain that does not improve with additional epidural dosing may be an indication of more serious pathology such as uterine rupture.
Management options
The key to managing poor blocks is early detection. All mothers who have had an epidural block in labour should be checked by the anaesthetist within 20-30 minutes of the first dose and the level of analgesia tested with a suitable stimulus. Continued review of the patient and the efficacy of epidural analgesia is mandatory throughout subsequent labour. Any complaint of persistent pain at any time during the labour should prompt further testing and review. Dislodgement or disconnection of the epidural catheter should first be excluded by visual inspection of the insertion site and infusion set. Confirmation of block height and assessment of the distribution of pain may enable the anaesthetist to determine the appropriate management strategy. If an epidural cannot be made to function adequately within an hour of troubleshooting, the anaesthetist and woman should consider the benefit of resiting it.
Generalised abdominal pain with uterine contractions may indicate an insufficient height of neuraxial block, which should extend to a dermatomal level of T8-10 during the first stage of labour. This may be managed by the administration of an additional dose of the epidural local anaesthetic-opioid mixture (a bolus of up to 20 ml has been suggested). This should preferably be given manually, as this is thought to aid wider spread in the epidural space. Occasionally, a solution containing either stronger local anaesthetic or fentanyl, or both, might be required for intense pain during augmented labour.
Perineal pain that results from failure of epidural analgesia to provide effective blockade of the larger sacral nerve roots that are less penetrable by neuraxial drugs (sacral sparing) may respond to topping up in the sitting position, with either the standard low-dose epidural solution or a higher-concentration local anaesthetic (e.g. 0.25% bupivacaine), or with fentanyl (50-100 µg). Supplementation with a pudendal nerve block may be used if delivery is imminent. A combined spinal-epidural (CSE) should be considered in situations when the pain is resistant to treatment.
A limited unilateral block is usually due to insertion of an excessive amount of epidural catheter, and may be rectified by pulling back the catheter to leave 3-5 cm in the epidural space. Unfortunately, once a ‘track’ has been established for the local anaesthetic solution, it may persist despite this manoeuvre, and the only solution may be to remove the catheter and re-site it in a different space. Even if the catheter was originally inserted to the optimum distance, the possibility of it being drawn further into the epidural space should not be discounted; this has been shown to happen as a result of traction imposed by movements of the vertebrae and activity of the spinal muscles.
Unilateral block can occasionally be overcome by laying the patient on the affected side and administering large volumes of local anaesthetic, with or without opioid. It is presumed that this encourages spread of the solution up and down the epidural space and thus beyond the boundaries of any midline septum, or allows any breaches in the septum to be exploited. These manoeuvres may be effective but there will often be a marked tendency for the block to affect the ‘good’ side more than the other. In the more recalcitrant unilateral block, the catheter may need to be re-sited in another interspace. Sometimes, when the above manipulations have been unsuccessful, a paramedian approach may help, since the catheter has been shown to travel a straighter course in the epidural space when inserted via this route. A CSE technique may be used to overcome poor spread or missed segments, and for some anaesthetists this is the technique of choice following a failed epidural, as it allows rapid onset of analgesia and it has been suggested that the epidural catheter is less likely to fail if a needle-through-needle CSE technique has been used compared with a plain epidural technique. Other causes for poor block should not be overlooked.
With a missed segment, the above techniques may also be tried. Sometimes, a stronger solution of local anaesthetic may be effective, and this may also be useful in cases of apparently adequate spread but breakthrough pain within the ‘blocked’ dermatomes, such as back pain or perineal pain. Addition of opioid is usually employed too. The woman should be warned that re-siting will not necessarily lead to a resolution of the problem if it has been caused by anatomical variation.
Continuous infusion techniques may fail if the syringe pump is not functioning properly, has been incorrectly set up or has become occluded. The commonest reason for a previously satisfactory block failing is the epidural catheter falling out - another good reason for checking the catheter first.
Finally, it should be remembered that an inadequate epidural for labour is likely to be inadequate for caesarean section should one be required in an emergency; in such situations it might be appropriate to consider spinal anaesthesia instead. Furthermore, in women with a high risk of caesarean section, the anaesthetist should have a low threshold for re-siting an epidural if the latter is less than perfect during labour.
Key points
• Blocks should be checked regularly to allow early detection of failure.
• In the event of a poor block, especially when previously satisfactory, the catheter should be checked.
• Poor spread is often caused by an excessive length of catheter in the epidural space.
A length of 3-5 cm is usually considered optimum.
• Re-siting the catheter early is often a better option than repeated ‘fiddling’.
Further reading
Agaram R, Douglas MJ, McTaggart RA, Gunka V. Inadequate pain relief with labor epidurals: a multivariate analysis of associated factors. Int J Obstet Anesth 2009; 18: 10-14.
Hermanides J, Hollmann MW, Stevens MF, Lirk P. Failed epidural: causes and management. Br J Anaesth 2012; 109: 144-54.
Pan PH, Bogard TD, Owen MD. Incidence and characteristics of failures in obstetric neuraxial analgesia and anesthesia: a retrospective analysis of 19,259 deliveries. Int J Obstet Anesth 2004; 13: 227-33.