when epidural does not work - epidural malfunction


Epidural not working

Failure of epidural to provide adequate pain relief is low. Published data most likely overestimates it. In the vast majority of laboring women epidurals result in excellent pain relief.

Epidural that is not functioning properly is one of the most frustrating situations for obstetric anesthetists. However, every medical procedure, including labor epidural, has its failure rate. Sometimes surgical operations do not produce the desired result, drugs don’t help some patients, so some epidurals don’t work.

The incidence of failure depends on the institution and the level of training of anesthetist performing it. Teaching hospitals tend to have higher failure rate, because many epidurals there are done by trainees and the fact that patient population in these institutions tends to be more complicated.

The latest – and the most comprehensive – review of epidural success has been performed 2004. Out of 19,259 epidurals 12% did not function adequately. However, 46% of those were corrected by simple manipulations which are discussed below. 7.1% of patients had their epidurals re-done. 1.9% of patients had their epidurals replaced more than once. In the end, 98.8% of patients reported having adequate pain relief during their labor.

An earlier review in 1998 reported higher failure rate: 13.1% of epidurals performed on 4240 patients had to be re-done. However 98% of women were satisfied with their pain relief. Apparently older equipment and old fashioned drug dosages were to blame for higher rate of failure.

Many practicing anesthetists will find the rates of failure reported in the above studies surprisingly high, and it seems that the numbers above do not reflect real life situation. In both of the above studies the failure has been defined as the inability to achieve adequate pain relief for labor after one attempt. Changing the position of the catheter, changing the strength of local anesthetic and administering additional boluses of local anesthetic mixture solves the problem in most cases within 20-30 minutes, yet it is recorded as primary failure.

As already mentioned, failures are more common in academic hospitals where majority of epidurals are done by registrars, and it is reasonable to suggest that in private hospitals where epidurals are administered by qualified specialists the rate of problems, including non-working epidural, is significantly lower. On the other hand, data from private hospitals doesn’t get published, and this statistic becomes unfavorably skewed.

There are several possible causes of epidural failure. In order for the epidural to work properly local anesthetic mixture should be distributed within epidural space so that appropriate nerve roots are blocked. If it does not happen it is usually either due to the incorrect placement of the epidural needle, epidural catheter or complicated anatomy of the epidural space. Placement of the epidural needle is based on relatively subjective feeling of loss of resistance to pressure, therefore if the tissues around epidural space are not dense enough it is possible to make the mistake and get the wrong impression of the position of the tip of the needle.

Other factors sometimes associated with the rate of epidural success are the type of the epidural catheter used (soft tip, hard tip, wire reinforced, single or multi-orifice), already mentioned experience of the anesthetist, the distance to which the catheter is threaded into the epidural space and, finally, expectations of the patients. There is also a separate debate regarding which is best for loss of resistance, air or saline.

Incorrect placement of the needle tip most commonly happens in patients with complicated anatomy. The examples include obese patients and those with spinal deformities. Because of higher degree of failure the latter is relative contra-indication to epidural, and some anesthetists are reluctant to perform them in such patients. Most commonly the tip of incorrectly placed epidural needle ends up in the subcutaneous fat. This mistake is usually detected quickly, within 20-30 minutes. Curiously, some patients experience placebo effect even if local anesthetic mixture is injected under the skin. The relief, however, is short living, and epidural needs to be redone.

Another possibility, albeit quite rare, is so called subdural-extra-arachnoid block. The structure encompassing the spinal cord and the surrounding cerebro-spinal fluid, in fact, consists of three layers: Dura Mater, Arachoid Mater and Pia Mater (from outside in), tightly bound together. There is no space between them. However, with “luck” the tip of the needle and consequently the epidural catheter may end up just under the Dura Mater without penetrating the other two layers. When local anesthetic mixture is injected it separates these anatomical structures and spreads unpredictably, producing patchy block that is often insufficient for labor analgesia. Subdural- extra-arachniod block is discussed in more detail in the appropriate chapter.

It happens sometimes that the epidural space is identified correctly and the catheter placed without problems but the epidural does not work. The reason as to why it happens is best explained in the review quoted at the end of this chapter: “for those who have studied the epidural space, it may seem amazing that epidurals ever work. The epidural space is filled with fat, connective tissue, and an extensive venous plexus. It behaves… like a container filled with sand and variously sized pebbles, such that drugs must traverse a maze of obstacles to reach the nerves.

According to published data up to five to eight percent of epidurals may fail despite of the correct placement of the catheter, and it is generally believed that it happens either because of some sort of abnormal anatomical barrier or due to the tip of the catheter being not in the best position that insures the most adequate spread of local anesthetic. Considerable amount of research has been done in order to understand how drugs behave after being injected into the epidural space and why epidurals fail. In some cases bands of connective tissue are present in the epidural space acting as barriers to the spread of anesthetic drugs. These structures, called the dorsal median connective tissue band (DMCTB), are rather rare and cannot explain most cases of unilateral block.

It has also been established is that when the catheter is threaded through the epidural needle its tip seldom stays in the midline and often moves in the wrong direction: downwards instead of up. This probably is the main reason of incomplete blocks and that is why giving additional doses of local anesthetic mixture helps. Extra anesthetic diffuses into the cerebro-spinal fluid and eventually reaches concentration necessary for pain relief.

Another cause of incomplete patient satisfaction with epidural is sacral sparing. During the first stage of labor pain is conducted by the nerves at low thoracic to high lumbar levels, typically T10 – L1, while second stage pain is transmitted by sacral nerves, coming out of the lower part of the spine, sacrum. Epidural catheters are usually inserted at the level of L2-3 or L3-4, and epidural analgesia initially does not cover sacral segments. After a while the drugs are well diluted in the cerebro-spinal fluid and lower segments are blocked as well. However, if labor progresses fast by the time the woman is ready to push the pain that originates in the vagina and the perineum will not be adequately blocked. Similarly to the situation above, injection of larger and/or more concentrated local anesthetic mixture often helps. Some anesthetists prefer performing combined spinal-epidural block if labor is close to the second stage, to ensure analgesia of sacral segments.

Properly placed epidurals catheters may migrate. In this case well functioning epidural stops working after a few hours. The body is a dynamic system, and artificial objects placed in it – such as epidural catheters – may move. Inadequate block due to these causes is often helped by administering larger amount of local anesthetic mixture. Alternatively the epidural needs to be re-inserted.

Epidural catheter may also migrate either into the subdural space – becoming a spinal – or into a blood vessel, in which case the drugs intended for the epidural space will be inadvertently administered into the bloodstream. In the former case pain relief at first improves, following by possible symptoms of high block. In the latter case epidural becomes inefficient and there is danger of local anesthetic toxicity. Fortunately, due to low concentrations of local anesthetic mixtures used in current practice, local anesthetic toxicity is unlikely.

Patients with back problems – either deformed spine or chronic back pain – also experience higher rate of failure. It has been observed that in patients with chronic sciatica pain epidural takes longer time to work. Presence of herniated discs or scar tissue in the spinal structure, such as after back surgery, makes it difficult to perform epidural block and may also affect the spread of local anesthetics, thus increasing the risk of epidural failure.

Finally, different patients have different attitudes to pain and different expectations of pain relief, and the same degree of neuraxial block in labor pain may be perceived differently by different patients. This is usually solved without problems by giving less tolerant patients larger doses of drugs in their epidurals. There is, however, one situation where pleasing the patient in labor is problematic, namely, when epidural is started before induction of labor.

In order to preserve the ability to push during the second stage of labor local anesthetics in epidurals are used in low concentrations. Common side-effect of this is that laboring women are often aware of uterine contractions, often reporting feeling cramping. Strictly speaking, pain relief brought about by labor epidural is not one hundred percent complete. However, if epidural is started when labor was in progress and the woman experienced contraction pain in its full intensity, the level of pain during these cramps is acceptable by most patients. On the other hand, without the experience of painful uterine contractions such cramps feel uncomfortable and are often perceived as pain. That is why many obstetricians and anesthetist are often reluctant to start epidurals before the induction of labor, until the first stage of labor is established.

In spite of some degree of failure, epidurals bring relief to the overwhelming majority of patients. Partial or patchy block can be salvaged by simple maneuvers without inconveniencing the woman, and epidural needs to be re-inserted in very small number of patients.


K Arendt, S Segal. Why Epidurals Do Not Always Work. Rev Obstet Gynecol. 2008;1(2):49-55.



Dr. Eugene Smetannikov is a practicing anesthesiologist with the interest in obstetric anesthesia. He is the author of the most comprehensive book on the subject, The Truth About Labor Epidural