The term epidural is often short for epidural anesthesia, a form of regional anesthesia involving injection of drugs through a catheter placed into the epidural space. The injection can cause both a loss of sensation (anaesthesia) and a loss of pain (analgesia), by blocking the transmission of signals through nerves in or near the spinal cord.
The epidural space (or extradural space or peridural space) is a part of the human spine. It is the space inside the bony spinal canal but outside the membrane called the dura mater (sometimes called the "dura"). In contact with the inner surface of the dura is another membrane called the arachnoid mater ("arachnoid"). The arachnoid encompasses the cerebrospinal fluid that surrounds the spinal cord.
Spinal anaesthesia is a technique whereby a local anaesthetic drug is injected into the cerebrospinal fluid. This technique has some similarity to epidural anaesthesia, and the two techniques may be easily confused with each other.
Epidural anesthesia and analgesia
Injecting medication into the epidural space
is primarily performed for analgesia
. This may be performed using a number of different techniques and for a variety of reasons. Additionally, some of the side-effects of epidural analgesia may be beneficial in some circumstances (e.g. vasodilation
may be beneficial if the patient has peripheral vascular disease
). When a catheter
is placed into the epidural space (see below) a continuous infusion can be maintained for several days, if needed. Epidural analgesia may be used:
- For analgesia alone, where surgery is not contemplated. An epidural for pain relief (e.g. in childbirth) is unlikely to cause loss of muscle power, but is not usually sufficient for surgery.
- As an adjunct to general anaesthesia. The anaesthetist may use epidural analgesia in addition to general anaesthesia. This may reduce the patient's requirement for opioid analgesics. This is suitable for a wide variety of surgery, for example gynaecological surgery (e.g. hysterectomy), orthopaedic surgery (e.g. hip replacement), general surgery (e.g. laparotomy) and vascular surgery (e.g. open aortic aneurysm repair). See also caudal epidural, below.
- As a sole technique for surgical anaesthesia. Some operations, most frequently Caesarean section, may be performed using an epidural anaesthetic as the sole technique. Typically the patient would remain awake during the operation. The dose required for anaesthesia is much higher than that required for analgesia.
- For post-operative analgesia, in either of the two situations above. Analgesics are given into the epidural space for a few days after surgery, provided a catheter has been inserted. Through the use of a patient-controlled analgesia (PCA) infusion pump, a patient may be given the ability to control post-surgical pain medications administered through the epidural.
- For the treatment of back pain. Injection of analgesics and steroids into the epidural space may improve some forms of back pain. See below.
- For the treatment of chronic pain or palliation of symptoms in terminal care, usually in the short or medium term.
The epidural space is more difficult and risky to access as one ascends the spine, so epidural techniques are most suitable for analgesia for the chest, abdomen, pelvis or legs. They are much less suitable for analgesia for the neck, or arms and are not possible for the head (since sensory innervation for the head arises directly from the brain via cranial nerves rather than from the spinal cord via the epidural space.)
There are circumstances where the risks of an epidural are higher than normal. These circumstances include:
- Anatomical abnormalities, such as spina bifida, meningomyelocele or scoliosis
- Previous spinal surgery (where scar tissue may hamper the spread of medication, or may cause an acquired tethered spinal cord)
- Certain problems of the central nervous system, including multiple sclerosis or syringomyelia
- Certain heart-valve problems (such as aortic stenosis, where the vasodilation induced by the anesthetic may impair blood supply to the thickened heart muscle.)
Circumstances in which epidurals should not be used:
- Lack of consent
- Bleeding disorder (coagulopathy) or anticoagulant medication (e.g. warfarin) - risk of spinal cord-compressing hematoma
- Infection near the point of insertion
- Infection in the bloodstream which may "seed" via the catheter into the (otherwise relatively impervious) central nervous system
- Uncorrected hypovolemia (low circulating blood volume)
The diagram at right depicts the various structures of the spinal column. The spinal cord (yellow core) is in intimate contact with the pia mater (blue). The arachnoid (red) exists superficial to the pia mater, and is attached to it by many trabeculae, giving it a spider-like appearance. This space (light blue) is filled with cerebrospinal fluid (CSF) and is called the subarachnoid space. Superficial to the arachnoid is the dura mater (pink) and although they are unattached, they are kept firmly pressed against one another because of pressure exerted by the CSF. Superficial to the dura mater is a space (pale green), known as the epidural space, that exists between it and the internal surfaces of the vertebral bones and their supporting ligamentous structures. This space is likewise pressed closed by surrounding tissue pressure, so it is called a 'potential' space. The vertebral bones (taupe) are attached to one another by the interspinous ligaments (teal). Insertion of an epidural involves threading a needle between the bones, through the ligaments and into the epidural potential space taking great care to avoid puncturing the layer immediately below containing CSF under pressure.
Technique of insertion
Epidural anaesthesia requires a high level of technical proficiency to avoid serious complications, and should always be performed by a trained anaesthetist, using a strict aseptic technique
to reduce the risk of infection.
Position of the patient
The patient may be in the sitting position or lateral position (lying on the side). The sitting patient is asked to slouch and bend forward slightly from the waist to increase the curvature of the spine. The lying patient is asked to draw the knees up to the chin for the same reason.
The anaesthetist palpates the patient's back and identifies a suitable anatomical gap between the bony spinous processes prior to the procedure. The level of the spine at which the catheter is best placed depends mainly on the site and type of an intended operation or the anatomical origin of pain. Since innervation of the chest and abdomen travels under the ribs, the anaesthetist can palpate along the corresponding rib to determine placement of the catheter tip.
Most commonly, the anaesthetist conducting an epidural places the catheter in the mid-lumbar, or lower back region of the spine, although occasionally a catheter is placed in the thoracic (chest) or cervical (neck) region. In adults, the spinal cord terminates at the first lumbar vertebra, below which lies a bundle of nerves known as the cauda equina ("horse's tail"). Hence, lumbar epidurals carry a very low risk of injuring the spinal cord.
Locating the epidural space
The skin is infiltrated with local anaesthetic such as lidocaine
over the identified space. The insertion point is usually in the midline, although other approaches, such as the paramedian approach
, may occasionally be employed. In the paramedian approach, the needle tip passes along a shelf of vertebral bone called the lamina until just before reaching the ligamentun flavum and the epidural space. 'Walking' the needle tip off this lamina allows the clinician to be confident that they are close to the epidural space. This is particularly important in the thoracic spine, where the spinal cord is larger (than in the lumbar spine) and nearly fills the spinal canal increasing the risk of dural puncture and cord damage.
A particular type of needle known as a Tuohy needle is almost invariably used. This needle was specially designed for locating the epidural space safely, and has several specific features for this purpose.
The Tuohy needle is inserted to the interspinous ligament and a loss of resistance to injection technique is used to identify the epidural space. This technique works because the interspinous ligament is extremely dense, and injection into it is almost impossible. The anaesthetist attaches a syringe to the Tuohy needle and advances it slowly.
The syringe may contain air or saline. The principles are the same, but the specifics of the technique are different due to the greater compressibility of air with respect to saline.
When the tip of the needle enters a space of negative or neutral pressure (such as the epidural space), there will be a "loss of resistance" and it will be possible to inject through the syringe. There is now a high likelihood that the tip of the needle has entered the epidural space. A sensation of "pop" or "click" may be felt as the needle breaches the ligamentum flavum just before entering the epidural space. An innovative technique for teaching this sensation of 'loss of resistance' using a banana was described by Leighton in Anesthesiology 70:368-9; 1989 - "A greengrocer's model of the epidural space."
Traditionally anaesthetists have used either air or saline for identifying the epidural space, depending on their personal preference. However, evidence is accumulating that saline may result in more rapid and satisfactory quality of analgesia.
In addition to the loss of resistance technique, realtime observation of the advancing needle is becoming more common. This may be done using a portable ultrasound scanner, or with fluoroscopy (moving X-ray pictures).
Feeding the catheter
After placement of the tip of the Tuohy needle into the epidural space the catheter is threaded through the needle. The needle is then withdrawn over the catheter. Generally the catheter is then withdrawn slightly so that 4-6 cm remains in the epidural space.
The catheter has depth markings on it (see photo) so that the length of catheter in the epidural space can be estimated.
The catheter is a fine plastic tube, down which anaesthetics may be given into the epidural space. Early catheters had a hole at the end ("end-hole catheters"), but were prone to blockage. More modern catheters ("side-hole catheters") have a blind end but three or more side-holes along the shaft near the tip. This not only disperses the anaesthetic more widely around the catheter, but lessens the likelihood of blockage.
The catheter is typically secured to the skin with adhesive tape or dressings to prevent it becoming dislodged.
In some unusual instances, it may not be required to insert a catheter into the epidural space, e.g. for steroid injections; see below. The anesthesiologist may inject medication into the epidural space through the needle, then remove the needle.
A patient receiving an epidural for pain relief typically receives a combination of local anesthetics and opioids
. This combination works better than either type of drug used alone. Common local anesthetics
, and chloroprocaine
. Common opioids include morphine
, and pethidine
(known as meperidine
in the U.S.). These are injected in relatively small doses.
Occasionally other agents may be used, such as clonidine or ketamine.
Bolus or infusion?
For a short procedure, the anaesthetist may introduce a single dose of medication (the "bolus" technique
). This will eventually wear off. Thereafter, the anaesthetist may repeat the bolus provided the catheter remains undisturbed.
For a prolonged effect, a continuous infusion of drugs may be employed. A common solution for epidural infusion in childbirth or for post-operative analgesia is 0.2% ropivacaine or 0.125% bupivacaine, with 2 μg/mL of fentanyl added. This solution is infused at a rate between 4 and 14 mL/hour, following a loading dose to initiate the nerve block.
There is some evidence that an automated intermittent bolus technique provides better analgesia than a continuous infusion technique, though the total doses are identical. , ,
Block height and intensity
Typically, the effects of the epidural are noted below a specific level on the body (dermatome
). This level (the "block height") is chosen by the anaesthetist. The level is usually 3-4 dermatomes higher than the point of insertion. A very high insertion level may result in sparing of very low dermatomes. For example, a thoracic epidural may be performed for upper abdominal surgery, but may not have any effect on the perineum
(area around the genitals) or bladder.
Nonetheless, giving very large volumes into the epidural space may spread the block both higher and lower.
The intensity of the block is determined by the concentration of local anaesthetic drugs used. For example, 15ml 0.1% bupivacaine may provide good analgesia for a woman in labour, but would likely be insufficient for surgery. Conversely, 15ml of 0.5% bupivacaine would provide a more intense block, likely sufficient for surgery. Since the volume used in each case is the same, the spread of drug, and hence the block height, is likely to be similar.
Removing the catheter
The catheter is usually removed when the patient is able to take oral pain medications. Catheters can safely remain in place for several days with little risk of bacterial infection,
particularly if the skin is prepared with a chlorhexidine solution.
Subcutaneously tunneled epidural catheters may be left in place for longer periods, with a low risk of infection or other complications.
Other types of epidural
For some procedures, the anaesthetist may choose to combine the rapid onset and reliable, dense block of a spinal anaesthetic
with the post-operative analgesic effects of an epidural. This is called combined spinal and epidural anaesthesia
The anaesthetist may insert the spinal anaesthetic at one level, and the epidural at an adjacent level. Alternatively, after locating the epidural space with the Tuohy needle, a spinal needle may be inserted through the Tuohy needle into the subarachnoid space. The spinal dose is then given, the spinal needle withdrawn, and the epidural catheter inserted as normal. This method, known as the "needle-through-needle" technique, may be associated with a slightly higher risk of placing the catheter into the subarachnoid space.
The epidural space may be entered through the sacrococcygeal membrane
, using a 22g catheter-over-needle or regular 21G needle. Injecting a volume of 1 cc/kg of local anaesthetic here provides good analgesia of the perineum
and groin areas. This is typically a single-injection technique and a catheter is not normally placed. This is known as a caudal epidural or "caudal".
The caudal epidural is an effective and safe analgesic technique in children undergoing groin, pelvic or lower extremity surgery. It is usually combined with general anaesthesia since children cannot tolerate the injection awake.
Epidural steroid injections
An epidural injection, or epidural steroid injection, may be used to help reduce the pain caused by a herniated disc
, degenerative disc disease
, or spinal stenosis
. These spinal disorders often affect the cervical (neck) and lumbar (lower back) areas of the spine.
The medicine used in the injection is usually a combination of a local anesthetic (e.g. bupivacaine
) and a steroid (e.g. triamcinolone
). The technique and risks of the procedure are similar to those for standard epidural analgesia. The effects of an epidural steroid injection vary, but permanent benefit is unlikely
The technique is believed to work by reducing the inflammation or swelling, or both, of the nerves in the epidural space.
Some patients who have some residual pain after the first injection may receive a second or third epidural steroid injection. Patients who do not receive any relief from the first injection are unlikely to benefit from a second injection.
Benefits of epidural analgesia after surgery
Epidural analgesia has been demonstrated to have several benefits after surgery. These include:
- Effective analgesia without the need for systemic opioids .
- The incidence of postoperative respiratory problems and chest infections is reduced .
- The incidence of postoperative myocardial infarction ("heart attack") is reduced .
- The stress response to surgery is reduced .
- Motility of the intestines is improved by blockade of the sympathetic nervous system .
- Use of epidural analgesia during surgery reduces blood transfusion requirements .
Despite these benefits, no survival benefit has been proven for high-risk patients .
In addition to blocking the nerves which carry pain, local anaesthetic drugs in the epidural space will block other types of nerves as well, in a dose-dependent manner. Depending on the drug and dose used, the effects may last only a few minutes or up to several hours. This results in three main effects:
Pain nerves are most sensitive to the effects of the epidural. This means that a good epidural can provide analgesia without affecting muscle power or other types of sensation. The larger the dose used, the more likely it is that the side-effects will be problematic.
For example, a laboring woman may have a continuous epidural during labor that provides good analgesia without impairing her ability to move around in bed. She requires a Caesarean section, and is given a large dose of epidural bupivacaine. After a few minutes, she can no longer move her legs, or feel her abdomen. Her blood pressure is noted to be lower and she is given an intravenous bolus of ephedrine or phenylephrine infusion to compensate. During the operation, she feels no pain.
Very large doses of epidural anaesthetic can cause paralysis of the intercostal muscles and diaphragm (which are responsible for breathing), and loss of sympathetic function to the heart itself, causing a profound drop in heart rate and blood pressure. This requires emergency treatment, and in severe cases may require airway support. This happens because the epidural is blocking the heart's sympathetic nerves, as well as the phrenic nerves, which supply the diaphragm.
It is considered safe practice for all patients with epidurals to be confined to bed to prevent the risk of falls.
The sensation of needing to urinate is diminished, which often requires the placement of a urinary catheter for the duration of the epidural.
Opioid drugs in the epidural space are very safe (as well as effective). However, very large doses may cause troublesome itch, and rarely, delayed respiratory depression.
Complications of epidural use
- Block failure (about 1 in 20). Partial failure may still give satisfactory pain relief. However, if pain relief is inadequate, another epidural may have to be performed.
- Bloody tap (about 1 in 30-50). It is easy to injure an epidural vein with the needle. In patients who have normal blood clotting, it is extremely rare (e.g. 1 in 100,000) for problems to develop. However, in a patient who has a coagulopathy, the patient may be at risk of epidural hematoma. If blood comes back down the needle, the anesthesiologist will normally place the epidural at another level.
- Accidental dural puncture with headache (common, about 1-3 in 100 insertions) The epidural space in the adult lumbar spine is only 3-5mm deep, which means it is comparatively easy to cross it and accidentally puncture the dura (and arachnoid) with the needle. This may cause cerebrospinal fluid (CSF) to leak out into the epidural space, which may in turn cause a post dural puncture headache (PDPH). This can be severe and last several days, and in some rare cases weeks or months. It is caused by a reduction in CSF pressure and is characterised by postural exacerbation when the patient raises their head above the lying position. If severe it may be successfully treated with an epidural blood patch (a small amount of the patient's own blood given into the epidural space via another epidural needle which clots and seals the leak). Most cases resolve spontaneously with time.
- Catheter misplaced into a vein (uncommon, less than 1 in 300). Occasionally the catheter may be misplaced into an epidural vein, which results in all the anaesthetic being injected intravenously, where it can cause seizures or cardiac arrest in large doses (about 1 in 10,000 insertions). This also results in block failure.
- High block, as described above (uncommon, less than 1 in 500).
- Catheter misplaced into the subarachnoid space (rare, less than 1 in 1000). If the catheter is accidentally misplaced into the subarachnoid space (e.g. after an unrecognised accidental dural puncture), normally cerebrospinal fluid can be freely aspirated from the catheter (which would usually prompt the anaesthetist to withdraw the catheter and resite it elsewhere). If, however, this is not recognised, large doses of anaesthetic may be delivered directly into the cerebrospinal fluid. This may result in a high block, or, more rarely, a total spinal, where anaesthetic is delivered directly to the brainstem, causing unconsciousness and sometimes seizures.
- Neurological injury lasting less than 1 year (rare, about 1 in 6,700).
- Epidural abscess formation (very rare, about 1 in 145,000). The risk increases greatly with catheters which are left in place longer than 72 hours.
- Epidural haematoma formation (very rare, about 1 in 168,000).
- Neurological injury lasting longer than 1 year (extremely rare, about 1 in 240,000).
- Paraplegia (extremely rare, less than 1 in 100,000).
- Arachnoiditis (extremely rare, fewer than 1000 cases in the past 50 years)
- Death (extremely rare, less than 1 in 100,000)) .
The figures above relate to epidurals in healthy individuals.
There is no evidence to support the concern that epidural analgesia increases the risk of anastomotic breakdown following bowel surgery.
Epidural analgesia in childbirth
Safety and efficacy
Epidural analgesia is a relatively safe method of relieving pain in labor. It provides rapid pain relief in most cases. It is more effective than nitrous oxide
, and other common modalities of analgesia in childbirth.
Prolonged labour and risk of instrumental delivery
Epidural analgesia is associated with longer labor
Some researchers claim that it is correlated with an increased chance of operational intervention. The clinical research data on this topic is conflicting. For example, a study in Australia (Roberts, Tracy, Peat, 2000) concluded that having an epidural reduced the woman's chances of having a vaginal birth, without further interventions (such as episiotomy
or caesarean section
) from 71.4% to 37.8%. Conversely, a 2001 study by researchers at the National Institute of Child Health and Human Development
and a 2002 study by researchers at Cornell University
and the University of Ontario
demonstrated that epidurals do not increase the likelihood of a caesarean section. In 2005, a meta-analysis of 21 studies also showed that epidurals do not increase the likelihood of caesarean section, but they do increase the chance of a forceps or ventouse
delivery by 40% (Anim-Somuah, Cochrane Review, 2005). The COMET Study, published in The Lancet in 2001 (vol358, No9275 p19-23) showed that a combined spinal epidural in labor may speed up the labor process by a few minutes, although those women receiving an epidural had a caesarean rate of 28% and only 35% had a normal birth without instrument assisted delivery.
These differing outcomes may be explained by data that demonstrates that the likelihood of increased intervention is directly related to the quality of the institution or practitioner providing the care: epidurals administered at top-rated institutions do not generally result in a clinically significant increase in caesarean rates, whereas the risk of caesarean delivery at poorly ranked facilities seems to increase with the use of epidural
An alternative explanation is that women are more likely to request an epidural during a prolonged or difficult labor, which in turn is more likely lead to an assisted vaginal birth.
Effects on the baby
Some mothers worry that epidural analgesia may harm their newborn. However, although epidural labor analgesia may be associated with slower progress of labor, it has no adverse effect on perinatal outcome and perinatal complications.
One study concluded that women whose epidurals contain the drug fentanyl were less likely to fully breastfeed their infant in the few days after birth and more likely to stop breastfeeding in the first 24 weeks. However, this study has been criticised for several reasons, one of which is that the original patient records were not examined in this study, and so many of the epidurals were assumed to contain fentanyl when almost certainly they would not have. In addition, all patients who used epidurals in labor had also used systemic pethidine, which would be much more likely to be the cause of any effect on breastfeeding due to the higher amounts of medication used via that route. If that were the case, then early epidurals which avoided the need for pethidine may actually improve breastfeeding outcomes, not worsen them.
Prior to 1943, there were few methods of relieving pain in childbirth without risk of harm to the baby. Caesarean sections
under general anesthesia
was used only as an emergency measure. Dr. Robert A. Hingson, Dr. Waldo B. Edwards, and Dr. James L. Southworth working at the United States Marine Hospital
at Stapleton, on Staten Island, New York, developed the technique of continuous caudal anesthesia
In 1912, German physicians had found that the injection of an anesthetic, at the base of the spinal cord, would prevent pain impulses from reaching the brain. Doctors in the United States developed the technique further. For expectant mothers, the injection "only reduced the pangs of childbirth; it did not eliminate them," wrote Dr. Morris Fishbein in the March 1943 issue of Hygeia, and a single nerve blocking injection was used only toward the end of labor.
Drs. Hingson and Southworth combined the concepts of caudal analgesia and the spinal injection in an operation to strip the varicose veins of a Scottish merchant seaman. The surgeons experimented with a continuous infusion of the local anesthetic, rather than removing the needle after the injection, to originate "continuous caudal analgesia". Dr. Hingson then collaborated with Dr. Edwards, the chief obstetrician at the Marine Hospital, to study the use of this technique in childbirth. The two studied the caudal region to determine where a needle could be safely placed to deliver anesthesia to the spinal nerves without placing the drugs into the spinal fluid.
Testing on a human being did not occur until January 6, 1942, when the wife of a Coast Guardsman was brought into the Marine Hospital for a delivery. Because the woman suffered from rheumatic heart disease, general anesthesia could not be safely used for an emergency Caesarean section, and it was believed that she would not survive the stress of labor. With the use of continuous local anesthesia, the woman and her baby survived. According to Dr. Fishbein's article in Hygeia, a total of 589 women in more than twenty participating hospitals gave birth relatively painlessly in 1942.
The results were published in the January 23, 1943, issue of the Journal of the American Medical Association.
- Roberts C, Tracy S, Peat B,Rates for obstetric intervention among private and public patients in Australia: population based descriptive study, British Medical Journal (BMJ), v321:p137, 15 July 2000
- Jun Zhang, Michael K. Yancey, Mark A. Klebanoff, Jenifer Schwarz and Dina Schweitzer, Does epidural analgesia prolong labor and increase risk of cesarean delivery? A natural experiment, American Journal of Obstetrics and Gynecology, Volume 185, Issue 1, July 2001, Pages 128-134.
- Barbara L. Leighton and Stephen H. Halpern, The effects of epidural analgesia on labor, maternal, and neonatal outcomes: A systematic review, American Journal of Obstetrics and Gynecology, Volume 186, Issue 5, Part 2, May 2002, Pages S69-S77. Also available online
- Boqing Chen and Patrick M. Foye, UMDNJ: New Jersey Medical School, Epidural Steroid Injections: Non-surgical Treatment of Spine Pain, eMedicine: Physical Medicine and Rehabilitation (PM&R), August 2005. Also available online