Birthsong Childbirth Education & Support Services

Epidural Anesthesia

Epidural anesthesia is the most commonly used and requested method of pain relief for women giving birth in hospitals today. Because it so effectively numbs the pain of contractions, women believe it is the best option and it has become an expectation of many women before they've even experienced labor. Women who have received epidurals see them as a quick fix and are seldom given fully informed consent prior to the procedure. There are many possible side effects and complications, not to mention the entire cascade of medical interventions which often ensue following such a procedure. Women who have had less than favorable experiences with their epidurals often wonder why they were not told about the potential adverse outcomes. The best practice is to fully educate yourself prior to any medical procedure and become an informed consumer. The following information is intended to provide you with a comprehensive guide to everything you need to know about epidural anesthesia, including the benefits and the often untold risks.

Epidural Anesthesia
Janelle Durham, 2002

What is an epidural

  • A catheter is placed into the lumbar region of the back, between vertebrae into the epidural space. Medication is injected, typically by continuous pump infusion. May be a “caine” anesthetic or a narcotic analgesic (Fentanyl or morphine), or a combination of these.
    • Some side effects may vary depending on exactly which medication is used. See Simkin, Whalley, and Keppler for an excellent discussion.
  • Provides pain relief in abdomen, back, buttocks, perineum, and legs.
  • The most common type of pain relief. Statistics vary widely by hospital, but rates may be as high as 75-90% of births.

Benefits of Epidurals: why would you ask for one in the first place?

  • Effective pain relief (for 90-95% of women).
  • Reduced pain related to contractions and to interventions such as pitocin, forceps, episiotomy, cesarean.
  • Mother can remain clear-headed, think and converse normally, and can rest, often even sleep for a few hours while her cervix continues to dilate.
  • May relax pelvic floor muscles, allowing cervix to dilate fully. (This is especially true if mother is experiencing a lot of anxiety because of the pain.)
  • Side effects of narcotics are reduced when they’re given by epidural rather than I.V. because less drug is needed.
  • Can lower blood pressure, which may be helpful for hypertension.

How is an epidural administered, what equipment is involved, and what are the side effects?

  • After you request an epidural, the nurse will call the anesthesiologist. May take 15-60 minutes: Women have described this period as “the longest 30 minutes of my life”. This is because they had already decided the contractions were overwhelming, and asked for relief, and it’s hard to wait for that to come. This is the time to use all the comfort measures at your disposal. Meanwhile:
  • You get into bed; you won’t be able to walk around after the epidural. Most hospitals also say no more food and water from this point on.
  • An IV is started, and extra fluids are given to reduce the chance of a drop in blood pressure. The I.V. also allows for easier administration of medications such as pitocin. Pitocin augmentation is about 3 times more likely in epidural labors
    • One of the common side effects of epidural is dystocia (slowed labor progress).  Contractions are decreased in strength and/or frequency. Body may stop producing oxytocin. Early labor may slow or stop
      • Of 8 studies reviewed by Thorp, 7 suggest epidurals are associated with longer first stage labor. This is typically treated with pitocin.
      • 9 of 10 studies reviewed report a significant association between epidural and longer 2nd stage labor. (This may lead to instrumental delivery or c-section.)
  • Usually a blood pressure cuff is placed to regularly monitor blood pressure. If blood pressure drops (as it does for 12% of women): more fluids, oxygen or medications.
  • Temperature is checked regularly: there is a 14.5% chance of fever over 100.4 w/ epidural;  chance increases after 4 hrs or more with epidural. After 7 hours with epidural, fever increases by up to 1ºC (1.8º F) per hour.
    • If you develop a fever, medical staff works on the assumption that you have an infection: treatment for infection may begin. (May involve antibiotics for mom; 48 hours of observation, blood culture, and spinal tap for baby.)
  • External Fetal Monitor is placed to monitor baby’s condition.
    • If mother develops a fever: Baby’s heart rate may become rapid. Baby may develop a fever (30% chance with epidurals)
    • If the mother is given pitocin: contractions can get too long and too strong, reducing baby’s heart rate.
    • If mom’s b.p. drops: decreased fetal heart rate and decreased oxygen supply.
    • Mild to severe fetal distress in 10-15% of babies after epidural. Generally these changes don’t affect the baby’s health at birth (as measured by APGAR scores), but signs of fetal distress can lead to c-section…
    • One thing you can do to help: after an epidural, lie on your side rather than your back; may decrease fetal distress.
  • Prepping and placing the catheter can take 15-25 minutes. Mother is sitting, or lying on her side, with her back arched. Her back is cleaned with betadine and draped with sterile towels. Local anesthetic is injected, then a large needle is pushed through the skin, then the epidural needle is placed, then catheter is placed, and the needle removed. The mother must remain completely still for insertion, even through severe contractions. The catheter is securely taped to the mother’s back, and medication is begun. May take 15-45 minutes to reach full effect.
  • Urinary catheter is placed. (Can’t sense need to urinate, and full bladder may slow labor.) Bladder function may not return to normal for a day or two after birth.
  • Mother may need oxygen mask, due to decreased respiratory rate.
  • With narcotic epidurals, side effects may include moderate to intense itching, difficulty in urination, nausea and vomiting.
  • Shivering is present in 10% of laboring women. With epidurals, the incidence raises to 33%. Although harmless, it may cause concern for mothers.
  • Headache. During labor, or chronic headaches in the months after birth.
  • Support people tend to move further away from birthing moms after epidural: One study showed that instead of being inches away on average before the epidural, they’re feet away afterwards. When the mom is no longer experiencing physical pain, they may assume she no longer needs support; some women report feeling emotionally abandoned by people moving away at this point.

Time for the Birth

- Can you feel the urge to push? Generally not. Medical staff will check cervix to tell you when to start pushing; and will watch monitor to tell you when a contraction comes so that you can push with the contraction.

- Can you push as effectively with an epidural? No. Abdominal muscles are weakened: normal pressures exerted during second stage are 120-135 mm Hg, with an epidural, they barely reach 100 mm Hg. Also, you can’t use your voluntary muscles as well to aid in the pushing, and you can’t move into the most effective positions (i.e. squatting)

- Does the baby rotate as well into the correct position if mom has had an epidural? No. Amongst non-epidural mothers, only 4% had a posterior baby persisting into second stage; after epidural, 19% failed to rotate.

As a result of these factors, there’s an increased risk of: forceps, vacuum extractor, and c-section. Of 24 studies reviewed, 22 studies showed a significant association between epidural and instrumental delivery; other 2 suggested an association, but it was not statistically significant. With narcotics, instrumental delivery: 3-7% With epidural: 15-53%

Thorp referred to some studies where the epidural medication was stopped once the mother reached 8 cm dilation, so that sensation would return for second stage labor. You may wish to consult with your caregiver about whether this practice is currently recommended.

Increased Risk of Cesarean

In a review of 15 available studies, 12 suggested a significant association between epidural and c-section. Risk of c-section generally found to be 2-3 times more likely with epidural.

An influence on this is what point in labor the epidural was administered. One study found that cesarean rates were 11% if epidural was given at 5 or more cm dilation, 16% at 4 cm, and 28% at 3 cm. Another study was even more striking, finding that cesarean rates increased to 26% when epidural was given at 4 cm, dilation, 33% at 3 cm, and 50% when the epidural was given at 2 cm dilation. So, the longer you can wait to have an epidural, the better.

Possible Effects After the Birth.

Backache. 10% of new moms develop a backache for the first time that lasts at least 6 weeks. Among women who’d had epidurals, the number jumps to 18%. This may be due to poor positioning during birth: women with epidurals may not be able to sense discomfort when they are in a position which is straining muscles, so support people need to pay attention to keeping mom in a comfortable, healthy position.

Medications cross the placenta and may have subtle side effects on the baby, including more difficulty in self-soothing, subtle changes in reflexes.

Decrease in maternal oxytocin during labor may interfere with oxytocin release after birth. Bonding and milk letdown reflexes may be decreased.

Some uncontrolled studies have been done, which aren’t conclusive due to lack of controls, but interesting nonetheless: unmedicated mothers reported that their babies were more sociable, more rewarding, and easier to care for. Unmedicated moms were more responsive to their babies’ cries. Women who had epidurals smiled less at their infants


Statistics cited are from “Epidural Anesthesia in Labor: An Evaluation of the Risks and Benefits” by Thorp and Breedlove, Birth, June 1996. This was a literature review article, which summarized the results of numerous studies involving thousands of births. Other information from: Pregnancy, Childbirth, and the Newborn by Simkin, Whalley, and Keppler (1991 edition). Maternity & Women’s Health Care by Lowdermilk, Perry, and Bobak (6th Edition, 1997). Family-Centered Maternity and Newborn Care by Celeste R. Phillips (Fourth edition, 1996). “Epidural Epidemic” by Dozer and Baruth, Mothering, July-August 1999. “What no one tells you about Epidurals” by Penny Simkin; “The Cascade of Interventions” by Pam England; and Epidural’s Effects on Babies” by Beverley Lawrence Beech, in Mothering, March-April 2000.

“So you Want An Epidural” by Kim James.


Medical Risks of Epidural Anesthesia During Childbirth
By Lewis Mehl-Madrona, M.D., Ph.D. and Morgan Mehl-Madrona

Epidural anesthesia has become increasingly popular for childbirth. The popular book, What to Expect when You’re Expecting, for example, portrays epidurals as perfectly safe. The risks, however, may be greatly underplayed.

Epidurals and pain relief

For the most part, epidural analgesia does effectively relieve labor pain.1 Obstetrical anesthesiologists continue to state that epidural analgesia has other, potentially catastrophic, adverse effects but, with safe clinical practice, these problems are extremely rare. We will suggest in the material that follows that these complications are not extremely rare, and that women are not receiving adequate informed consent about what these complications are and their accompanying frequency. Nor are they being offered any serious alternatives to epidural anesthesia. Despite this, anesthesiologists such as Eberle and Norris argue that specific anaesthetic techniques ... or obstetrical management can limit or eliminate these risks of epidural labour analgesia. What must be remembered for any technical procedure, is that it is studied in major academic centers where highly skilled professors supervise residents and all outcomes are monitored closely. The actual practice, however, takes place in smaller institutions by less qualified individuals so that the actual complication rates of any procedure (obstetric, cardiac, pulmonary) are always higher than what are found in studies.

Overall complications rates for epidural anesthesia

A general estimate of the overall complication rate of epidural anesthesia is 23%.2

1. Effects of epidurals on cesarean rate:

When the dose is too large or when it sinks down into the sacral ("tailbone") region of the body, the perineum and the vagina are anesthetized. Anesthetic is intentionally injected into this area late in labor to deaden all sensation. When it "accidentally" happens earlier in labor, the muscles of the pelvic floor are prematurely relaxed, thereby interfering with the normal flexion and rotation of the baby's head as it passes through the birth canal. This interference can lead to abnormal presentations which are more dangerous for the baby or to what is called "failure to descend," an indication for Cesarean birth.

Thorp, et al3 studied 711 consecutive nulliparous women at term, with cephalic fetal presentations and spontaneous onset of labor. They compared 447 patients who received epidural analgesia in labor with 264 patients who received either narcotics or no analgesia.

The incidence of cesarean section for dystocia was significantly greater (p < 0.005) in the epidural group (10.3%) than in the nonepidural group (3.8%). There remained a significantly increased incidence (p < 0.005) of cesarean section for dystocia in the epidural group after selection bias was corrected and the following confounding variables were controlled by multivariate analysis: maternal age, race, gestational age, cervical dilatation on admission, use of oxytocin, duration of oxytocin use, maximum infusion rate of oxytocin, duration of labor, presence of meconium, and birth weight.

The incidence of cesarean section for fetal distress was similar (p > 0.20) in both groups. There were no clinically significant differences in frequency of low Apgar scores at 5 minutes or cord arterial and venous blood gas parameters between the two groups. They concluded that epidural analgesia in labor increases the incidence of cesarean section for dystocia in nulliparous women.

Frequently the epidural is so effective that it eliminates uterine contractions. The nerves which tell the uterus to contract are all anesthetized. The uterus becomes quiet and must be driven artifically with the hormone oxytocin (Pitocin or Syntocinon).

As the cervix becomes fully dilated and the head descends, the woman (in a normal birth) feels pain and pressure in the lower pelvis and rectum. The last injection of anesthetic during the process of epidural anesthesia occurs after the head has rotated and come down onto the perineum. Higher concentrations of anesthetic are used to assure perineal relaxation. Sometimes the mother is sat upright or at least at a 45 degree angle to be certain that the anesthetic will descend to the sacral nerve roots. When the sacral nerve roots are blocked, the woman looses the urge to push.

After controlling for potentially confounding variables with multiple logistic regression analysis, Adashek, et al4 found that epidural anesthesia was an independent risk factor for cesarean birth among women over age 35 (R = 0.195, p < 0.001).

At the 1997 meeting of the American Society for Anesthesiology, a press release was issued about four studies involving a combined total of more than 22,000 women claiming that labor epidural analgesia does not increase a woman's risk of having a cesarean delivery.5 Three of the studies were presented at the annual meeting of the American Society of Anesthesiologists. The fourth appeared in the September 1997 issue of the medical journal of Anesthesiology.

“The findings have significant implications for physicians, patients and insureres,” said one of the reserchers, Steven T. Fogel, M.D., an anesthesiologist at Washington University School of Medicine in St. Louis. "Some physicians and insurance companies limit patient access to epidurals because they srongly believe thta epidurals can prolong labor, "Dr. Fogel said. "Delaying or withholding an epidural forces the pregnant woman to suffer needlessly and does not lower the cesarean rate."
Each year, about one-million women choose epidural blocks for safe and effective pain relief durig labor.

A study by Dr. Fogel and colleagues at Washington University analyzed labor and delivery data on 7,000 patients. The researchers compared the cesarean rates among first-time mothers during the 12 monts before and the 16 months after epidural analgesia became available on request at the hospital. They found no significant differences in cesarean section rates before and after the services introduction. "Epidural analgesia did not stop normal labor or cause cesarean deliveriew, because the rate of cesareans did not change," Dr. Fogel said.

Women may still require cesarean delivery following epidural blocks, but no direct cause-and-effect relationship has ever been established. "labor epidural anlagesia can be associated with , but does not cause, cesarean section delivery," Dr Fogel said. "An abnormal labor can produce extreme pain for mothers, and this pain leads women to ask for epidurals," he said.

"Our study has the advantage of following cesarean rates before and after the introduction of a popular epidural service," Dr Fogel said. "Since patient demographics and cesarean section rates did not change, we can safely and accurately conclude that the total number of cesarean delieveries performed was not affectred by the availability of epidural analgesia."

Similar findings emerged from an analysis of data on more than 13,000 first-time mothers who gave birth between 1989 and 1995 at St. Luke's Roosevelt Hospital Center, a teaching hospital of Columvia University in New York City.

The analysis showed that themother's age, the baby's birth weight and use of the labor-inducing drug oxytocin increased the women's risk for cesarean sections. "Not only was epidural analgesia not a significant risk factor but it was associated with slightly decreased cesarean section delivery risk." anesthesiologist David J. Birnbach, M.D., said.

By relieving pain, epidural analgesia may help some patients relax and this relazation could facilitate labor, Dr. Birnbach said. "We're not suggesting the technique will prevent cesarean sections," he stressed, "but in our practice, epidurals are clearly not a cause." In addition, the use of epidural analgesia almost doulbed at the hospital during the five years studied (from 37% to 68%) but the percentage of cesarean sections deliveries did not increase, Dr. Birnbach said.

A third study at Boston's Beth Isreal Deaconess Medical Center reviewed the labor and delivery records of more than 2300 women who recieved ultra-low dose solutions of epidural medications. The ultra-low dose epidurals allow many women's albor to proceed comfortable, Stephen D. Pratt, M.D., said. For women requireing more apin relief, additional medication can be given by the same epidural route.

The Boston study showed that women who needed addtional pain relief beyond the ultra-low dose wre more likely to have cesareans. Women who needed more than two additional doses were twice as likely to require a cesarean section as those who do not. "The slow, abnormal labor that leads to cesarean section is more painful than normal labor and therefore requires stronger epidural medication," Dr Pratt said. "Epidural do not cause cesarean sections. Rather women who have abnormal labor may be in more pain and therfore are more likely to need an epidural and stronger medication."

The above three studies were retrospective studies done in choice situations/hospitals that had actively taken a new approach to reducing cesarean sections as well as hospitals that maintain such high cesarean section rates that the comparisons would be insignificant due to cesarean rates of over 30%.

The only prospective study was the fourth one, conducted at the University of Texas Southwestern Medical Center in Dallas.6 Seven hundred, fifteen women of mixed parity in spontaneous labor at full term were randomly assigned to receive either epidural anesthesia (EA) or patient-controlled intravenous meperidine analgesia (PCMA). Epidural analgesia was maintained with a continous infusion of 0.125% bupivacaine with 2 pg/ml fentanyl. Patient controlled analgesia was maintained as 10-15 mg meperidine given every 10 minutes as desired by the patient using a patient-controlled pump.

A total of 358 women were randomized to receive epidural anesthesia and 243 (68%) complied. Similarly, 357 women were randomized to receive patient-controlled analgesia, and 259 (73%) complied with that protocol. Five women randomized to PCMA requested epidurals. Based upon an intent to treat analysis, there was no difference in the rate of cesarean deliveries between the two groups (EA, 4%, 95% CI: 1.9-6.2%; PCMA, 5%, 95% CI: 2.6-7.2%. Women in the epidural group reported lower pain scores during labor and delivery.

What is amazing is that this study was used to argue that epidurals did not increase risk for cesarean birth. What is lacking, of course, is a reasonable control group, such as a doula. When one looks at the amazingly high amounts of meperidine that women could self-administer, it is no wonder that so few requested an epidural, since they could hardly be expected to be very alert. A woman could receive 90 mg of meperdine per hour. In actuality, the average amount of meperidine a woman self-administered was 200 mg with the highest amount being 500 mg. To give a sense of the amounts being given here, routinely in the emergency department, for treating ureterolithiasis (kidney stones), which most women agree is a more severe pain than labor, I would rarely administer more than 150 to 200 mg of meperidine over 12 hours. The average length of labor during which these women received their meperidine was under 12 hours. No one has asked the question of how such a massive dose of meperidine would affect the cesarean rate.

The authors stated, "Patient-Controlled intravenous analgesia is widely used in the United States to manage postoperative pain, although use during childbirth has been limited. One concern is newborn respiratory depression from increased narcotic administration to the mother. Many women in our study used more than 200 mg of meperidine during the course of their labor, but only 3.4% of infants were given naloxone to reverse respiratory depression. The mothers were visibly sedated but were invariably arousable and none experienced respiratory depression."

Since neither of us have never had to give an infant naloxone in over 1000 deliveries, we wonder what the indications would actually be. We suspect these infants were quite depressed to receive naloxone. Our experience with patient controlled analgesia is that the reason the patient stops pushing the button is that she gets so high a dose that she falls asleep. I suspect the mothers were more compromised than the authors think, but none had to be intubated or artificially ventilated, therefore, “none experience respiratory depression.”

The authors also under-emphasized the fact that all patients were enrolled at the same time in an aggressive program to reduce cesarean birth, consisting of:

  • No use of electronic fetal monitoring, even for epidurals (except for one test strip on admission), thereby necessitating nurse auscultation of the fetal heart rate and one-on-one patient care, both well known to reduce the cesarean rate.
  • No patient admitted to the hospital before 4-5 cm of cervical dilation, also known to lower the cesarean rate.
  • No drugs or epidurals until the woman was 5 cm dilated, also known to reduce the cesarean rate.
  • All births attended by CNMs, which is known to reduce the cesarean rate by one-third of what obstetricians would do. For example, a study from Los Angeles County-USC Hospital showed a 4% cesarean rate in a 95% Hispanic population when they were attended by nurse-midwives.7
  • Use of a black and Hispanic population. In an editorial discussion, the authors, themselves, comment on black and Hispanic populations having historically lower cesarean rates

The authors' real conclusions are this: If you do everything possible to reduce the cesarean rate for all patients, and then, if you compare epidural anesthesia with high dose, self-administered narcotic analgesia, there is no difference in the cesarean rate. This is not really a very outstanding conclusion, though the popular press read this result as saying epidural anesthesia is now proven safe and unlikely to increase the cesarean rate (without addressing the question, “over what?”

Some infants (0.8%) in the epidural group also required naloxone. Two transfers to NICU occurred in the epidural group and 3 in PCMA group.

Looking just at those who had epidurals vs. those who didn't, epidurals prolonged the first stage of labor and increased the incidence of oxytocin administration. Fever developed in more women during epidural anesthesia. There was no difference in the number of cesareans in the epidural vs. the PCIA group (the range from 3% to 7% cesarean rate).

Regarding other complications, 24% of women having epidurals had fever compared to 6% in PCMA. Nine percent had forceps with epidural compared to 3% with PCMA. There was a15-19% range of meconium during labor (not different between groups), which we find rather high. There was a rate of 1.2-1.8% of infants suffering meconium aspiration (not different between groups), but also rather high in our experience.

2. Significant low blood pressure (hypotension):

Significant low blood pressure is a complication of epidural anesthesia.1 The ways that epiduralized patients must lie accentuate this. Their position is limited since they are essentially paralyzed people for the duration of the epidural. Hypotension occurs among almost one-third of patients with serious hypotension occurring about 12% of the time.8

Maternal hypotension is a major risk for the baby. The epidural blocks the nerves which regulate blood pressure. It causes the blood in the body to pool, keeping it from being pumped around the body in the proper manner. The arteries dilate and relax their usual, necessary level of tension, making it difficult for the heart to pump blood to the baby. These changes lead to a decrease in the output of the mothers heart. Less blood per unit time can reach the placenta and therefore the baby.

The baby is completely dependent on the mothers heart to pump blood to the placenta to satisfy its needs. All of its oxygen comes across from the placenta. All of the food for its brain and other organs comes across the placenta. Brains cannot live without a relatively constant supply of oxygen and glucose. Without this they become damaged.

A good blood flow is needed to the uterus between contractions so it can get fresh oxygen. During the contraction, blood flow to the uterus is cut off by the muscles contracting. When the contraction stops, the uterus must quickly refill with fresh blood containing oxygen for the baby. If the amount of blood flowing to the uterus is reduced, the baby may not be able to get the oxygen it needs. Then what is called fetal distress may occur. The babys system does not get enough oxygen and goes into distress. Its blood retains excess acid, the oxygen levels go low and tissues and vital organs begin to fail from excess acid, lack of oxygen and lack of fuel

Animal studies have shown that lack of oxygen to the baby (called fetal hypoxia) can cause significant damage to the baby’s brain even without the pH of the baby’s umbilical cord blood being affected.9 (The pH of the babys umbilical cord blood at birth is usually used as an indicator of whether or not fetal hypoxia has occurred.)

Severe low blood pressure can also result from compression of the mothers blood vessels (aorta and vena cava) since all mothers must lie essentially flat on their back after epidural anesthesia (they cannot feel or move their back, pelvis and legs).10

3. Fetal distress:

Fetal heart rate decelerations can occur following the use of epidurals.1 Babies can develop fetal distress after epidural anesthesia.11 This may be caused by the mothers blood pressure getting so low that blood cannot be adequately pumped into the uterus to deliver oxygen to the baby. As we mentioned above, epidurals make it difficult for the muscles in the arteries of the lower body to respond and to keep blood adequately flowing through the body. The ability of the heart to respond to changing needs of the body is impaired.12 Eberle and Norris1 suggest that [i]nduction of maternal analgesia may transiently alter the balance between factors encouraging and discouraging uterine contraction. A temporary increase in the uterotonic effects of endogenous or exogenous oxytocin may then produce a tetanic contraction with subsequent decrease fetal oxygen delivery and resultant fetal bradycardia.

Most babies of mothers receiving epidural anesthesia develop episodes of slow heart rate (bradycardia).13 While this does not usually affect the healthy baby, it can be disastrous for the baby that is alreay compromised from some other problem (often unknown to the doctors).

Adverse effects on the baby indicative of insufficient oxygen reaching the baby (late decelerations) can occur. These changes may also result from a toxic effect to the baby of the local anesthetic given in the epidural.

The transient low blood pressure which always occurs after epidural anesthesia has been found to lead to significantly lower the babys blood pH.14 This indicates excess blood acid, usually meaning that the baby is not getting enough oxygen. Anesthesiologists don’t think that this makes any difference in the babys outcome, but we suspect, if we studied babies already at risk for other reasons, we would find that epidural significantly worsens compromised babies and may lead to a Cesarean birth when the baby might have otherwise tolerated a vaginal birth. We suspect that more detailed research would identify a group of babies who would have tolerated unmedicated, normal birth, but who are unable to handle the added stress of the epidural, leading in the worst causes to death or permanent disability.

4. IV Cannulation:

Accidental injection of the anesthetic solution into the blood stream can occur and can cause the mother to twitch, have convulsions, or lose of consciousness. Seizures can occur from the toxic effects of the anesthetic agent entering the blood stream.15 Local anesthetic toxicity occurred among 12 women in 1000 epidurals.16

5. Trauma to blood vessels:

Trauma to blood vessels can occur as a result of epidural anesthesia.17 In one study, bleeding in the spinal column and unintentional placement of the catheter into an artery or vein occurred 0.67% of the time (67 women of every 1000 epidurals).18 The catheter actually escapes outside of where it is supposed to go 1 to 6% of the time.19

Hemorrhages can occur around the spinal cord and even within the skull following epidural anesthesia.20 These were associated with persistent backaches or headaches. Failure to treat these problems usually results in permanent paralysis. Surgery must be performed within 8 hours of the onset of paralysis or the prognosis is poor. Chronic subdural hematoma has resulted from epidural anesthesia and has even presented as post-partum psychosis.21

6. Punctured dura:

The actual dura may be punctured as a result of epidural anesthesia. Because of the large size of the needle used, severe headache may also result. Dural punctures have been found to occur about 1.8% of the time.22 Unintentional dural puncture occurred in 61 of 1000 epidurals in a University hospital (resulting in spinal anesthesia).23

7. Infection:

An infection can develop at the site of injection. Bacterial meningitis can occur from contamination during placement of the epidural.24 An abscess can also form at the site where the epidural catheter is placed.25 26

8. Backache:

Backache is a common complication. Back pain commonly occurs after epidural anesthesia (18.9% of the time27 ). Upper back pain can happen at some distance from the site where the epidural is injected.28 The back pain can last very long-term.29 Nineteen percent of women had long-term backache after epidural anesthesia.30 It probably results from a combination of its effects on the nerves and from extreme postures and stretching that occurs after the epidural during labor.

Low back pain after epidural anesthesia for childbirth is also frequently mentioned.31

9. Broken catheters:

Occasionally the catheter has broken and a small piece is left in place. It usually causes no ill effects.

10. Abnormal uterine contractions:

Uterine contractions can become weaker and less frequent. An oxytocin infusion is then necessary to improve labor and produce good strength contractions Mothers having epidurals have longer labors and have a higher incidence of the use of oxytocin than mothers having non-medicated deliveries.32

There are important risks of giving oxytocin also. Administration of this hormone to the mother during labor can cause:

  • Dangerously high blood pressure.
  • Abnormal heart rhythms.
  • Nausea and vomiting.
  • Sustained uterine contractions which last too long and result in the baby going into distress from lack of oxygen. When this is too severe, the uterus can rupture. The epidural can mask the strength of the uterine contractions so that no one knows that how strong they are, making uterine rupture more possible.
  • Hemorrhage around the brain.
  • Retention of water leading to convulsions and coma.
  • Bleeding in the pelvis and increased incidence of postpartum hemorrhage.
  • Death of the baby.
  • Jaundice of the baby.

11. Second stage labor effects:

With large doses the patient loses the desire and the ability to bear down and push. This results in an increased use of forceps and vacuum extractions over women having unmedicated deliveries.33

12. Inadequate pain relief:

The epidural is generally inadequate 7.1% of the time leading to supplementation with intravenous pain medication 4.0% of the time and a general anesthetic 3.1% of the time (in one study).34

13. Accidental spinal anesthesia:

When an epidural accidentally turns into a spinal anesthetic, many complications can occur:

  • Postspinal headaches.
  • Dysfunction of the bladder is frequent
  • Occasionally numbness and tingling (paresthesias) of the lower limbs and abdomen develop, and sometimes there is a temporary loss or diminution of sensation in these areas.
  • Unilateral footdrop (paralysis of the muscle that lifts the foot) has occurred.
  • Permanent nerve damage (conditions called chronic, progressive adhesive arachnoiditis or transverse myelitis) can occur. These lead to paralysis of the lower parts of the body.
  • Deaths have been reported.
  • Difficult breathing
  • Increased incidence of forceps deliveries.

The reliability of spinal anesthesia with 5% hyperbaric lignocaine was studied among 30 patients undergoing elective Cesarean. Twelve patients had hypotension and four developed severe postspinal headaches. The block progressed to the C2 dermatome in four patients and was associated with dysphagia. This was totally unpredicted and was thought due to altered cerebrospinal fluid dynamics in late pregnancy.35

14. Maternal heart attack or spinal cord ischemia:

The lack of ability of the heart to pump blood around the body (from low blood pressure or pooling of blood) can become so severe that a heart attack occurs or the spinal cord will suffer damage from not enough blood reaching it.36

15. Asthmatics:

Asthmatics can get suddenly worse during epidural anesthesia37 with more wheezing and inability to breath.

16. Medication interactions:

A hidden danger of epidural anesthesia is its interaction with medications (prostaglandins) commonly used to soften the cervix and start labor. The use of prostaglandins is common at hospitals and creates a potentially dangerous situation in which the usual medications used to treat low blood pressure during labor will no longer work.38

17. Interactions occur with other illnesses:

As an example, women who have migraines can have more visual disturbances after epidurals.39

18. Maternal fever:

Maternal fever and even the severe condition called malignant hyperthermia (dangerously high fever) can result.40 41

19. Respiratory arrests:

Mothers can stop breathing (respiratory arrest)42 43 and can experience other breathing difficulties.44 Greenhalgh* reported a 19 year old obstetric patient who had a respiratory arrest shortly after receiving intrathecal sufentanil and bupivacaine as part of a combined epidural/spinal technique for pain relief.

20. Other neurological disabilties:

Other neurological disabilties (including a condition called Horners syndrome) can develop along with hoarseness (from even just one dose of epidural anesthetic).45 Clayton46 reported an incidence of Horner's syndrome during epidural anesthesia for elective Caesarean section of 4%. The incidence of Horner's syndrome with epidural anesthesia for vaginal delivery was 1.33%. They found it impossible to predict which patients would develop a Horner's syndrome. Even the nerves to the face can be blocked, sometimes temporarily, sometimes permanently.47 Tremors and shakes can occur.48 49

Paresthesias (persistent tingling from sensory nerves) occurred in 0.16% of patients in one study (1.6 per 1000) with an incidence of persistent neuropathy of 0.04% (4 per 10,000).50 Four of these patients had a neuropathy which eventually resolved. In another study 3.0% of patients had tingling of the hands or fingers, while 26 of almost 5000 women had persistent tingling or numbness in the lower back, buttocks or legs.51

Dizziness and fainting can become a problem after epidurals. One study found these symptoms persisting in 2.1% of women.52

21. Nausea and vomiting:

Twenty to 30 percent of women experience nausea after epidural anesthesia, while 3 to 7% have vomiting.53

22. Allergic conditions:

A dangerous allergic condition with shock (called anaphylaxis) can occur.54 The woman develops a red rash (erythema), itches, and her lungs fill up with fluid (pulmonary edema). Excessive lung fluid is also found in the babies in these cases.55

23. Heart problems:

Mothers can experience excessively slow heart rates (bradycardia), heart block in which the electrical activity of the chambers of the heart become dissociated and sometimes even stoppage of the heart (cardiac arrest).56

24. Headache:

Headache after epidural is a persistent problem that is more pronounced in younger patients.57 One study found its incidence after epidural anesthesia to be 4.6%, significantly more often than women not having epidurals.58 It usually occurs from the effects of puncturing the dura. Headache can also occur from air getting into the spinal fluid (called an iatrogenic pneumocephalus). The air is introduced into the spinal fluid and column when the test dose is given that is assumed to be in the extra-dural space. When the person giving the epidural feels a loss of resistance to the injection of air, this is when a pneumocephalus can occur. The patient that was described complained immediately of severe headache on both sides of her forehead followed by vomiting. The baby had to be delivered by Cesarean with general anesthesia. The patients headache resolved in 24 hours after Cesarean.

Of 34 women with spinal headache (from 4766 epidurals), nine had long-term disability from headache. Five of these were from accidental dural puncture and four occurred after accidental spinal block.59

25. Motor blockade:

Epidural anesthesia can produce motor blockade, resulting in temporary paralysis, even of respiratory muscles.60

26. Use in VBAC:

Epidurals are sometimes used with women desiring VBAC. Leung, et al.61 studied the maternal and fetal consequences of uterine rupture during VBAC. They concluded that significant neonatal morbidity occurred when > or = 18 minutes elapsed between the onset of prolonged deceleration and delivery. In 106 cases of uterine rupture at their institution between Jan 1, 1983 and June 30, 1992, seven charts were incomplete and excluded; of the remainder, 28 patients had complete, 13 patients had partial, and 58 patients had no fetal extrusion into the maternal abdomen. Maternal characteristics or intrapartum events were not predictive of the catastrophic extent of uterine rupture. There was one maternal death. Complete fetal extrusion was associated with a higher incidence of perinatal mortality and morbidity.

27. Technical considerations:

Epidural anesthesia is a technical procedure that requires significant skill to place correctly. Many papers document the technical aspects of this procedure which are not insignificant. For example, 23% of epidural catheters inserted more than 2 cm into the epidural space required manipulation. Epidural catheters inserted 8 cm within the epidural space were more likely to result in iv cannulation. Epidural catheters inserted 2 cm within the epidural space were more likely to become dislodged. Epidural catheters inserted 2 to 4 cm within the epidural space required replacement more often than catheters inserted deeper. 62

28. Herpes simplex assocation:

Epidural analgesia is associated with recurrence of herpes simplex blepharitis after cesarean section when epidural morphine is given.63

Case Examples (mild problems)

Here is an example of a 30 year old woman having her first baby who was admitted to the hospital at 2 cm dilation at 11:05 pm.64 By 4:30 am, she was 4 cm dilated with her waters broken and requested epidural anesthesia. Her mother and her husband were in attendence coaching her. The epidural was started at 5:01 am and within 30 seconds, the patients heart rate began to climb steadily and precipitously from 88 to 174, levelling off at that rate.

The certified registered nurse anesthetist (CRNA) discovered that the patient had previously experienced heart palpitations: when stressed or with heavy exercise. She had begun to feel mild nausea and dizziness. Oxygen was administered and an iv drug (adenosine) was given. A second dose followed five minutes later. The abnormal heart rate (a supraventricular tachycardia) returned to normal and the woman had a normal vaginal delivery within 6 hours of this episode. This occurred presumable from a sensitivity to medication placed through the epidural catheter. When complications such as this occur, fetal scalp electrodes are usually placed to monitor the babys EKG. The therapy sometimes causes severe hypotension.

A case has been reported in which a 24 year old woman with a past history of mild backache had an epidural anesthesia. She had an acute vagal reaction with loss of consciousness and her head falling forward. Thirty-six hours later, she complained of severe pains all over her spine, together with sciatica and spasm of the muscles on either side of the spinal column (paravertebral muscles). None of these symptoms responded to drug treatment. The pain eventually disappeared after she wore a cervical collar, taking anti-inflammatory drugs and having spinal manipulations (what chiropractors and osteopaths do) It was thought that her backache was due to what is called a posterior articular joint syndrome in which the forward fall of her head strained the posterior joints of the spinal column.65

Permanent disability from epidural anesthesia

a. A disabling condition called spinal arachnoiditis can develop after epidural anesthesia. Of 6 such women, 3 were permanently confined to a wheelchair three years after their initial evaluation.66 None of these patients had any prior spinal surgery or trauma or problems with the spinal cord including previous hemorrhage, infections or other known causes of arachnoiditis. They had no neurological symptoms prior to epidural anesthesia. The diagnosis was confirmed by a medical test called myelography in all cases. The epidurals were uneventful and performed according to standard methods.

Arachnoiditis is probably caused from the epidural injection of foreign substances (the anesthetic itself or contaminants in the solution) into the spinal canal.

Subarachnoid cysts can occur in the spine from arachnoiditis produced from the epidural anesthetic.67

b. Paralysis can occur.68 The injection of the local anesthetic into the epidural space can result in the veins becoming engorged, the spinal cord suffering from a lack of oxygen (hypoxia) and the woman developing acute neurological problems. Some of these deficits can become permanent.69 Paralysis can also occur from bleeding into the area during the epidural injection with the formation of a pocket of blood pressing on the spinal cord (hematoma). It can also occur from infection or trauma.

Cranial nerve paralysis can occur at quite a distance from the site of the epidural. This is thought to occur from traction on the spinal cord.70

Paralysis can occur from a condition called anterior spinal artery syndrome after epidural anesthesia during labor.71 Paralysis can occur when the blood flow to the spinal cord becomes so limited that tissue dies. This is called an infarction. A case report of an infarction after epidural anesthesia has been published in which leg paralysis occurred and did not recover. A loss of sensation to pain and temperature also occurred to the level of the mid-chest which partially resolved.72

Of 108 non-fatal complications in one study, five were associated with permanent disability.73 These included damage (neuropathy) to a single spinal nerve, acute toxicity from the local anesthetic, and problems associated with accidental puncture of the dura to become a spinal anesthetic.

Another patient developed paralysis after epidural anesthesia probably due to the anterior spinal artery or central arteries being blocked during the epidural and leading to death of part of the spinal cord.74

Neuropathy is a condition in which sensory changes occur (loss of sensation or hypersensitivity to sensation) with or without chronic pain. Neuropathy occurs after epidural anesthesia. It can occur from thrombosis of an artery from trauma from the epidural injection or from the catheter. A lack of adequate blood flow (and therefore oxygen) is called ischemia. This can cause neuropathy also.75

Deaths from epidural anesthesia

Here are some examples of woman who have died from epidural anesthesia to illustrate the dangers.

A healthy, 31 year old woman having her third child requested epidural anesthesia and developed an acute condition of fluid in the lungs (pulmonary edema). She could not be successfully resuscitated. The baby also died.76

Sudden stoppage of the heart (cardiac arrest) can occur during epidural anesthesia.77 78 along with other heart rhythm changes.79 Sudden cardiac arrest may be caused by air getting into veins during placement of the epidural.80

Respiratory insufficiency can occur and cause death. In one case report, the patient initially developed pain in the shoulder-neck region after epidural anesthesia, followed by fever and an elevated white blood count. This led to a high-level (arms and legs) paralysis with an inability to breath. Many problems then developed with the heart and arterial system. An abscess was found and the patient was eventually stabilized with antibiotics. The patient required chronic mechanical ventilation and died of recurrent pneumonia after 5 months of intensive care. The incidence of breathing difficulties in one study was 0.54%, although only 5% of those patients required prolonged artificial ventilation. Total spinal anesthesia occurred in 0.013% of the epidurals but more than half of these auses required intubation and prolonged mechanical ventilation. Partial spinal anesthesia (sub-arachnoid block) occurred in 0.04% of the cases.81

Women are almost never given informed consent for epidurals. Even if they were just read two paragraphs from the package insert that comes with the medication used for epidurals (manufactured by Abbott Laboratories), they might think twice. The package insert states:

Local anesthetics rapidly cross the placenta, and when used for epidural, caudal or pudendal anesthesia, can cause varying degrees of maternal, fetal and neonatal toxicity....Adverse reactions in the parturient, fetus and neonate involve alternations of the central nervous system, peripheral vascular tone and cardiac function....
Neurologic effects following epidural or caudal anesthesia may include spinal block of varying magnitude (including high or total spinal block); hypotension secondary to spinal block; urinary retention; fecal and urinary incontinence; loss of perineal sensation and sexual function; persistent anesthesia, paresthesia, weakness, paralysis of the lower extremities and loss of sphincter control all of which may have slow, incomplete or no recovery; headache; backache; septic meningitis; meningismus; slowing of labor; increased incidence of forceps delivery; cranial nerve palsies due to traction on nerves from loss of cerebrospinal fluid.

Who would sign a consent if it included the above language? The degree to which the facts about the risks of epidural anesthesia are hidden from women in labor is astonishing.

Mothers who have a fever are signficantly more likely to have had epidural anesthesia.82 Therefor increasing the incidence of septic workups on newborns and the subsequent complications of this procedure.

We would hope that pregnant women would have a more informed explanation of the risks of epidural anesthesia than what is currently given, and would suspect that many would make other choices if true informed consent was given (before labor, when the woman still has time to prepare to cope with childbirth pain in other ways).

Critique of 3 other studies

As we said the study we just discussed was the only prospective study. The other three studies are papers not yet published so we can only comment on the abstracts of these studies, they were "presented" at the conference of the American Society of Anesthesiologists in November of 1997. Presented means that the paper was talked about at the conference and has not necessarilly been accepted by a major medical journal for publication as of yet and may/or may not be.

1st additional study:

TITLE: Does epidural analgesia protect against cesarean section in nulliparous patients?

AUTHORS: DJ Burnbach, MD, A Grunebaum, MD, DJ Stem, MD, B Katgaem MD, MM Kuroda, MPH, DM Thys, MD

AFFILIATION: Department of Anesthesiology and Ob/Gyn, St. Luke's-Roosevelt Hospital Center, College of Physicians and Surgeons of Columbia University, New York, NY

INTRODUCTION: Recent articles have suggested that epidural analgesia may increase the risk of cesarean section. (1,2) In an effort to evaluate whether epidural analgesia during labor is associated with an increased risk of cesarean section at our institution, we evaluated the labor data of unlliparous patients who delivered at the hospital between 1989-1995.

METHODS: From January 1, 1989 -December 31, 1995, there were a total of 31,670 deliveries at the St. Luke's-Roosevelt Hospital Center. The 13,203 nulliparous patients who delivered under the care of an obstetrician during that period of time constituted the study population. The following were the independent variables: type of labor analgesia administered, patient age, obstetric care provider (private vs. non-private), birth weight, and use of oxytocin for stimulation or induction. The dependent variable was the delivery mode. A non-hierarchical logistic regression was performed to determine which of these independent variables contributed to the risk of cesarean section.

RESULTS: Logistic analysis demonstrated that the highest risks for cesarean section were associated with oxytocin induction and maternal age. As shown in Table 1, other significant predictors of cesarean delivery were oxytocin for stimulation, patient status (private patients were at an increased risk of cesarean section) and birth weight. Epidural analgesia, on the other hand, was associated with a slightly decreased risk for cesarean section. For example, patients under 30 years of age who were induced with oxytocin who received epidural analgesia had a 32% cesarean section rate (230/719) versus a 39.1% cesarean section rate (70/179) for the same group of patients who did not receive an epidural anesthetic. Patients greater than 29 years of age who were induced with oxytocin and received an epidural had a 45.4% cesarean section rate (114/251) versus a 58% cesarean section rate (51/88) for the same group of patients who did not receive an epidural.

Table 1:

Independent Variables Odds Ratio Confidence Interval
Oxytocin Induction 1.93 1.70 - 2.19
Maternal Age (per 10 years) 1.60 1.53 - 1.68
Birth Weight (per 1000g) 1.20 1.13 - 1.27
Oxytocin Stimulation 1.17 1.06 - 1.30
Patient Status (pvt. vs clinic) 1.10 1.01 - 1.16
Epidural Analgesia 0.89 0.81 - 0.97

DISCUSSION: Our data support other studies that have shown an increased risk of cesarean section with use of oxytocin, increased maternal age, and private patient status.(3) When controlled for other variables, the administration of epidural analgesia was associated with a decreased risk of cesarean section. Based on our data, we suggest that epidural analgesia as practiced at our institution, is associated with a decrease in cesarean section rate in nulliparous patients.

1. Obstet Gynecol 1996;88:993-1000
2. Am J Obstet Gynecol 1993;169:851-8
3. Am J Obstet Gynecol 1993;168:1881-5


A major statistical problem with the study, which prevents the authors from making the conclusions they have made, is the failure to consider the interaction between epidural anesthesia and oxytocin stimulation. Many studies have shown that epidural anesthesia increases the need for oxytocin stimulations [refs]. A proper statistical procedure would have been to use an interactive term for epidural and oxytocin stimulation.

I would have started with a multivariate analysis of variance which would have considered the interaction between epidural anesthesia and the other variables. The authors believe (erroneously) that they are dealing with independent variables. Clearly we all know that there is an association between oxytocin induction and epidurals (more difficult labors; more likely to have an epidural), maternal age (younger and older mothers are more likely to have epidurals), oxytocin stimuation (women who have epidurals are more likely to need oxytocin stimuation because of the desultory effect of the epidural on uterine contractions, private patients are more likely to have epidurals.

The more sophisticated approach to their paper would have been to use a technique such as structural equations modeling (LISREL, or related procedures) to test paths of effects. For example, one can test the hypothesis that epidurals affect the cesarean rate through their effect on the need for uterine stimulation. That could still be done with the authors' data, and we plan to challenge them to an independent data analysis (by our colleagues at the University of Pittsburgh) using path analysis to test these hypotheses.

Simply using logistic regression in this context was inappropriately simplistic, but done probably because it supported their bias (epidurals are good!). I suspect the authors are well-informed and have excellent biostatistical consultation, and chose not to report these other analyses that they probably did, because these other analyses did not support their position.

I would draw your attention to another part of the abstract. Is it reasonable practice to do cesareans on 58% of women who are older than 29 years and are being induced? Are these numbers generalizable? I was shocked to read this figure. In all my experience in obstetrics, I can't imagine how you can perform a cesarean on 58% of the women over age 29 whom you are inducing.

I would very much like to see their criteria for cesarean. With such high rates (45% in the epidural group and 58% in the non-epidural group), I question how generalizable these results are to good practice elsewhere. Perhaps in an environment (consider how stressful this environment must be) in which such high cesarean rates occur, epidurals do decrease the risk for cesarean if you are being induced, but I can't believe that such high cesarean rates constitute safe and prudent practice.

I would like to see their overall complication rates, including post-partum infection and the backache/headache complications of epidurals. I suspect their morbidity is quite high.

2nd & 3rd additional studies:

Two other studies shed some light on this debate. The first is Newton ER, Schroeder BC, Knape KG, Bennett BL. Epidural analgesia and uterine function. Obstet Gynecol 1995; 85:749-55.

"Continuous epidural analgesia with bupivacaine and fentanyl did not result in a change in myometrial contractility in the first hour after initiation of analgesia. However, despite more oxytocin therapy, the rate of cervical dilation was significantly slower in the epidural group than in the nonepidural group (1.9 versus 5.6 cm/hr, p < 0.001). Operative deliveries were more common in patients with epidural analgesia than in those without it (12 of 62 versus 2 of 124, p < 0.0001). After epidural analgesia, myometrial contractility is maintained with oxytocin, but the ability of the uterus to dilate the uterus is reduced significantly."

Also important, Albers LL, Anderson D, Cragin L, Daniels SM, Hunter C, Sedler KD, Teaf D. The relationship of ambulation in labor to operative delivery. JNM 1997; 42(1):4-8.

"Women who ambulated for a significant amount of time during labor (compared with those who did not ambulate) had half the rate of operative delivery (2.7% versus 5.5%)." It's very hard to ambulate with either an epidural or PCIA.

Two others studies have been widely quoted in the popular press. They included a study by Dr. Steven Fogel, an anesthesiologist at Washington University School of Medicine, in St. Louis. Dr. Fogel looked at data on 7,000 patients delivering over a 28 month period at his hospital. He compared cesarean rates for first time mothers before the introduction of an epidural on demand anesthesia service to cesarean rates after the introduction of the service, finding no change. What isn't reported is how obstetrical practices changed during this time interval, a very significant factor. Such historical studies are notoriously unreliable because policies change so rapidly in obstetrical management. Without such information, the study is meaningless.

The final study was done at Boston's Beth Isreal-Deaconess Medical Center by anesthesiologist Steven Pratt. This study reviewed the labor and delivery records of more than 2300 women who received ultra-low dose solutions of epidural medications. The study showed that women who needed additional pain relief beyond the low-doses were more likely to have cesareans than women who did not. Dr. Pratt argued that it was harder labors that caused cesareans and not epidurals. This study was so anecdotal as to be hardly worth commenting upon. What Dr. Pratt failed to note was that several others studies have shown that patient satisfaction with low-dose epidurals is very low and that more than half of women have additional medication. Pratt's study actually supports the point of view that epidurals lead to increased cesareans, when this information is taken into account.

What is amazing is the propagandizing that went on in the press. The gold standard in medicine consists of randomized, controlled trials. These have been done and do generally show increased cesarean rates. To try and argue against these much better quality studies with retrospective studies using historical controls is poor science, but clearly good propaganda. To conclude we list some of these studies and their conclusions:

Prospective Clinical Trials Investigating the Association between Epidural Analgesia and Cesarean Birth Rates by Randomizing Women to a Narcotic versus an Epidural Group:

Reference and Parity Sample Size Relative Risk of Cesarean
Birth with Epidural Analgesia (95% CI)

1. Combined 111 2.8 (0.8-10.0)
2. Nulliparous 93 11.4 (5.8-16.9)*
3. Nulliparous** 693 2.6 (1.5-4.3)*

Parous** 637 3.8 (1.3-11.0)*
Combined*** 869 2.3 (1.3-4.0)*
All three prospective trials combined 1073 2.5 (1.6-4.0)****,*

* Relative risk is statistically significant at least at p < 0.05
** This represents the odds ratio for all women in the study adjusted by multivariate logistic regression analysis.
*** The cesarean birth rate was significantly greater (p = 0.002) in the epidural group (9%, 39/432) compared with the narcotic group (4%, 17/437).
**** Comparison of the proportions by Mantel-Haenszel yields a chi-squared of 18.1 and a p-value of 0.00003.

1. Phillipsen T, Jensen NH. Epidural block or parenteral pethidine as analgesic on nulliparous labor; a randomized study concerning progress in labour and instrumental deliveries. Eur J Obstet Gynecol Reprod Biol 1989; 30:27-33.
2. Thorp JA, Hu DH, Albin RM. The effect of intrapartum epidural analgesia on nulliparous labor: A randomized prospective trial. Am J Obstet Gynecol 1993; 169:851-858.
3. Ramin SM, Gambling DR, Lucas MJ. Randomized trial of epidural versus intravenous analgesia during labor. Obstet Gynecol 1995; 86:783-789.


Eberle RL, Norris MC. Labour analgesia: A risk-benefit analysis. Drug-Saf 1996; 14(4):239-251.
Kantor G. Obstetrical epidural anesthesia in a rural Canadian hospital. Can J Anaesth 1992; 39:390-3.
Thorp JA, Parisi VM, Boylan PC, Johnston DA. The effect of continuous epidural analgesia on cesarean section for dystocia in nulliparous women [see comments]. American Journal of Obstetrics & Gynecology 1989 Sep;161(3):670-5.
Adashek JA, Peaceman AM, Lopez-Zeno JA, Minogue JP, Socol ML. Factors contributing to the increased cesarean birth rate in older parturient women. American Journal of Obstetrics & Gynecology 1993 Oct;169(4):936-40
David J. Birnbach, M.D., Steven T. Fogel, M.D., Stephen D. Pratt, M.D.; New Data Debunks Belief that Epidurals Cause Cesarean sections, San Diego, Press Release, American Society of Anesthesiologists, 1998.
Sharma SK, Sidawi JE, Ramin SM, Lucas MJ, Leveno KJ, Cunningham FG. Cesarean Delivery: A randomized trial of epidural versus patient-controlled meperidine analgesia during labor. Anesthesiology 1997; 87:487-94.
Twelve Years and More than 30,000 Nurse Midwife-Attended Births: The Los Angeles County + University of Southern California Women's Hospital Birth Center Experience: Journal of Nurse midwifery Vol 39, No 4, July Aug 94.
Webb and Kantor, 1992
Haire DB. Obstetric drugs: their effects on mother and child. Encyclopedia of Childbearing, New York: Henry Holt and Company, 1993.
[Hetland, et al., 1989]
[Rosay, et al., 1989]
[Parnass and Schmidt, 1990]
[Marx, et al., 1990]
[Philipson, et al., 1989]
[Dunne and Kox, 1991]
[Tanaka, et al., 1993]
[Skidmore, 1991]
[Tanaka, et al., 1993]
[Halpenny, et al., 1992]
[Schmidt and Nolte, 1992]
[Campbell & Varma,1993]
[Webb and Kantor, 1992]
[Tanaka, et al., 1993]
[Berga and Trierweiler, 1989]
[Goucke and Graziotti, 1990]
[Sonden, 1991]
[MacArthur, et al., 1992]
[Rajanna, 1989]
[Anonymous, Lancet, 1990a,b]
[MacArthur, et al., 1990]
McQuay H, Moore A. Epidural anesthesia and low back pain after delivery (letter). BMJ 1996; 312(7030): 581.
[Sepkoski, et al., 1992]
[Sepkoski, et al., 1992]
[Webb and Kantor, 1992]
[Bembridge, et al., 1986]
[Girompaire, et al., 1990]
[Eldor, et al., 1989]
[Veeckman, et al., 1989]
[MacArthur, et al., 1992]
[Anonymous, Lancet, 1989]
[Vanek and Valenta, 1989]
[Fortuna, 1989]
[Lee and Collins, 1990]
[Tembo and Hills, 1990]
[Tabatabia, et al., 1989]
[Clayton, 1989]
[Gazmuri, et al., 1992]
[Branda and Tumbarello, 1990]
[ Basora, et al., 1990]
[Tanaka, et al., 1993]
[MacArthur, et al., 1992]
[MacArthur, et al., 1992]
[Chestnut, et al., 1989]
[Zucker-Pinchoff and Ramanathan, 1989]
[Thomas and Caunt, 1993]
[Parnass and Schmidt, 1990]
[Parnass and Schmidt, 1990]
[MacArthur, et al., 1992]
[MacArthur, et al., 1992]
Greenhalgh CA. Respiratory arrest in a parturient following intrathecal injection of sufentanil and bupivacaine. Anaesthesia 1996; 51(2):173-175.
[Leung, et al., 1993]
Eberle and Norris, 1996, op cit.
DAngelo R, Berkebile BL, Gerancher JC. Prospective examination of epidural catheter insertion. Anesthesiology 1996; 84(1) 88-93.
[Stickes, 1993]
[LeBorgne, et al., 1991]
[Sghirlanzoni, et al., 1989]
[Sklar, et al., 1991]
[Parnass and Schmidt, 1990]
[Sghirlanzoni, et al., 1989 Oct]
[Whiting et al., 1990]
[Ackerman, et al., 1990]
[Gaudin and Lefant, 1990]
[Scott and Hibbard, 1990]
[Hachisuka, et al., 1991]
[Veselis, 1990]
[Van Zundert and Scott, 1989]
[Gild and Crilley, 1990]
[Watanabe, et al., 1990]
[Elstein and Marx, 1990]
[Williams, et al., 1991]
[Xie and Liu, 1991]
[Klaus et al. 1993:47]

Epidural Ups Fever, C-Section Risk

SOURCE: American Journal of Public Health 1999;89:506-510.
Copyright 1999 Reuters Limited
April 6, 1999

NEW YORK (Reuters Health) -- Women who have a fever during labor are much more likely to have a cesarean section, and epidural anesthesia tends to increase the risk of fever, according to a study by Harvard researchers.

The findings suggest that fever may be one reason why women who have an epidural are more likely to have a cesarean rather than vaginal delivery, according to a report published in the American Journal of Public Health, the journal of the American Public Health Association.

A vaginal delivery is generally considered to be preferable to a cesarean section due to a quicker recovery time and a lower risk of complications.

In the study of 1,233 low-risk women undergoing childbirth for the first time at Brigham and Women's Hospital in Boston, Massachusetts, 301 women (24%) developed a fever over 99.5 degrees Fahrenheit.

About 25% of women who developed a fever underwent a cesarean section or an assisted vaginal delivery compared with 7% to 8% of women who did not have a fever.

Nine out of 10 women with elevated temperatures had received epidural anesthesia, but not all women receiving epidural anesthesia developed fever. Of the 736 women in the study who had an epidural, 37% developed an elevated temperature, compared with about 6% of women who did not receive an epidural, report Dr. Ellice Lieberman and colleagues from Brigham and Women's and Harvard Medical School.

While a fever can be associated with an infection, the researchers found that only a few cases of elevated temperature in the study were due to infection.

Previous studies have linked epidural use to a higher rate of cesarean and assisted vaginal deliveries, but the reason for the link was unclear.

"Our data suggest that the temperature elevation that is associated with epidural use may in part explain the higher rates of cesarean and assisted vaginal delivery," the authors conclude. "Given the widespread use of epidural analgesia, we believe that it is essential that further study be conducted to determine whether treatment or prevention of fever during labor represents a practical means of lowering rates of cesarean delivery and operative vaginal delivery."

Epidurals- Real Risks for Mother and Baby
Copyright © Sarah Buckley, MD.
An edited version of this paper was first published in Australia’s Parents magazine, Aug/Sept 1998, as "What you need to know before you have an epidural".

Note: This paper may be copied and circulated, as long as the author is acknowledged. Sarah Buckley, Brisbane, Australia. Nov 1998

Epidural pain relief is an increasingly popular choice for Australian women in the labour ward. Up to one-third of all birthing women have an epidural,(1) and it is especially common amongst women having their first babies.(2) For women giving birth by caesarean section, epidurals are certainly a great alternative to general anaesthetic, allowing women to see their baby being born, and to hold and breastfeed at an early stage: however their use as a part of a normal vaginal birth is more questionable(3)

There are several types of epidural used in Australian hospitals. In a conventional epidural, a dose of local anaesthetic is injected through the lower back into the epidural space, around the spinal cord. This numbs the nerves which bring sensation from the uterus and birth canal. Unfortunately, the local anaesthetic also numbs the nerves which control the pelvic muscles and legs, so with this type of epidural, a woman usually cannot move her legs and, unless the epidural has worn off, cannot push her baby out, in the second stage of labour.

More recent forms of epidurals use a lower dose of local anaesthetic, usually combined with an opiate, such as pethidine, morphine or fentanyl (sublimaze). With this low-dose or combination epidural, most women can move around with support; however the chance of a woman being able to give birth without forceps is still low(4). Another form of epidural, popular in the US, is the CSE, or combined spinal-epidural, where a one-off dose of opiate, with or without local anaesthetic, is injected into the spinal space, very close to the end of the spinal cord. This gives pain relief for around 2 hours, and if further pain relief is needed, it is given as an epidural. These forms of "walking epidural" may seem advantageous, but being attached to a CTG machine to monitor the baby, and hooked up to a drip which is also a requirement when an epidural is in place, can make walking impossible.

Many women have a good experience with epidurals. Sometimes the relief from pain can allow a woman to rest and relax sufficiently to go on and have a good birth experience. However deciding to use an epidural for pain relief can also lead to a "cascade of intervention", where an otherwise normal birth becomes highly medicalised, and a woman feels that she loses her control and autonomy. Often the decision to accept an epidural is made without an awareness of these, and other, significant risks to both mother and baby.

Although the drugs used in epidurals are injected around the spinal cord, substantial amounts enter the mother's blood stream, and pass through the placenta into the baby's circulation. Most of the side effects of epidurals are due to these "systemic", or whole-body effects.

One of the most commonly recognised side effects is a drop in blood pressure. Up to one woman in eight will have this side effect to some degree(5), and for this reason, extra fluids are usually given through a drip to prevent problems. A drop in the mother's blood pressure will affect how much of her blood is pumped to the placenta, and can lead to less oxygen being available to the baby.

An epidural will often slow a woman's labour, and she is three times more likely to be given an oxytocin drip to speed things up(6,7). The second stage of labour is particularly slowed, leading to a three times increased chance of forceps(8). Women having their first baby are particularly affected; choosing an epidural can reduce their chance of a normal delivery to less than 50%(9).

This slowing of labour is at least partly related to the effect of the epidural on a woman's pelvic floor muscles. These muscles guide the baby's head so that it enters the birth canal in the best position. When these muscles are not working, dystocia, or poor progress, may result, leading to the need for high forceps to turn the baby, or a caesarean section. Having an epidural doubles a woman's chance of having a caesarean section for dystocia(10).

When forceps are used, or if there is a concern that the second stage is too long, a woman may be given an episiotomy, where the perineum, or tissues between the vaginal entrance and anus, are cut to enlarge the outlet and hurry the birth. Stitches are needed and it may be painful to sit until the episiotomy has healed, in 2 to 4 weeks.

As well as numbing the uterus, an epidural will numb the bladder, and a woman may not be able to pass urine, in which case she will be catheterised. This involves a tube being passed up from the urethra to drain the bladder, which can feel uncomfortable or embarrassing.

Other side effects of epidurals vary a little depending on the particular drugs used. Pruritis, or generalized itching of the skin, is common when opiate drugs are given. It may be more or less intense and affects at least 1/4 of women(11,12): morphine or diamorphine are most likely to cause this. Morphine also causes oral herpes in 15% of women(13).

All opiate drugs can cause nausea and vomiting, although this is less likely with an epidural (around 30%)(14) than when these drugs are given into the muscle or bloodstream, where larger doses are needed. Up to 1/3 of women with an epidural will experience shivering(15), which is related to effects on the bodies heat-regulating system.

When an epidural has been in place for more than 5 hours, a woman's body temperature may begin to rise(16). This will lead to an increase in both her own and her baby's heart rate, which is detectable on the CTG monitor. Fetal tachycardia, or fast heart rate can be a sign of distress, and the elevated temperature can also be a sign of infection such as chorioamnionitis, which affects the uterus and baby. This can lead to such interventions as caesarean section for possible distress or infection, or, at the least, investigations of the baby after birth such as blood and spinal fluid samples, and several days of separation, observation, and possibly antibiotics, until the results are available(17).

Less common side effects for a woman having an epidural are; accidental puncture of the dura, or spinal cord coverings, which can cause a prolonged and sometimes severe headache (1 in 100)(18) ongoing numb patches, which usually clear after 3 months(1 in 550)(19); and weakness and loss of sensation in the areas affected by the epidural, (4-18 in 10,000) also usually resolving by 3 months(20).

More serious but rare side effects include permanent nerve damage; convulsions and heart and breathing difficulties (1 in 20,000)(21) and death attributable to epidural. (1 in 200,000)(22) When opiates are used, a woman may experience difficulty in breathing which comes on 6 to 12 hours later.(23)

There is a noticeable lack of research and information about the effects of epidurals on babies.(24) Drugs used in epidurals can reach levels at least as high as those in the mother(25), and because of the baby's immature liver, these drugs take a long time- sometimes days- to be cleared from the baby's body.(26) Although findings are not consistent, possible problems, such as rapid breathing in the first few hours(27) and vulnerability to low blood sugar(28) suggest that these drugs have measurable effects on the newborn baby.

As well as these effects, babies can suffer from the interventions associated with epidural use; for example babies born by caesarean section have a higher risk of breathing difficulties.(29) When monitoring of the heart rate by CTG is difficult, babies may have a small electrode screwed into their scalp, which may not only be unpleasant, but occasionally can lead to infection.

There are also suggestions that babies born after epidurals may have difficulties with breastfeeding(30,31) which may be a drug effect, or may relate to more subtle changes. Studies suggest that epidurals interfere with the release of oxytocin(32) which, as well as causing the let-down effect in breastfeeding, encourages bonding between a mother and her young(33).

Epidural research, much of it conducted by the anaesthetists who administer epidurals, has unfortunately focussed more on the pro's and con's of different drug combinations than on possible serious side-effects(34). There have been, for example, no rigorous studies showing whether epidurals affect the successful establishment of breastfeeding(35).

Several studies have found subtle but definite changes in the behaviour of newborn babies after epidural(36,37,38) with one study showing that behavioural abnormalities persisted for at least six weeks(39). Other studies have shown that, after an epidural, mothers spent less time with their newborn babies(40), and described their babies at one month as more difficult to care for.(41)

While an epidural is certainly the most effective form of pain relief available, it is worth considering that ultimate satisfaction with the experience of giving birth may not be related to lack of pain. In fact, a UK survey which asked about satisfaction a year after the birth found that despite having the lowest self-rating for pain in labour (29 points out of 100), women who had given birth with an epidural were the most likely to be dissatisfied with their experience a year later.(42)

Some of this dissatisfaction was linked to long labours and forceps births, both of which may be a consequence of having an epidural. Women who had no pain relief reported the most pain (70 points out of 100) but had high rates of satisfaction.

Pain in childbirth is real, but epidural pain relief may not be the best solution. Talk about other options with your care-givers and friends. With good support, and the use of movement, breathing and sound, most women can give themselves, and their babies, the gift of a birth without drugs.


1) Perinatal Statistics, Queensland 1996. Queensland Health 1998. At the present time, national figures for epidural use are not collected.
2) Dr Steve Chester, Head of Anaesthetics Dept, Royal Women's Hospital, Melbourne. Around 45% of primiparous women at RWH have an epidural. Personal Communication
3) World Health Organisation. Care in Normal birth: A Practical Guide..P 16. WHO 1996
4) Russell R, Reynolds F. Epidural infusion of low-dose bupivicaine and opioid in labour. Does reducing the motor block increase the spontaneous delivery rate? Anaesthesia 1996; 51(5): 266-273
5) Webb RJ, Kantor GS. Obstetrical epidural anaesthesia in a rural Canadian hospital. Can J Anaesth 1991; 39:390-393
6) Ramin SM, Gambling DR, Lucas MJ et al. Randomized trial of epidural versus intravenous analgesia during labor. Obstet Gynecol 1995; 86(5): 783-789
7) Howell CJ. Epidural vs non-epidural analgesia in labour. [Revised 6 May 1994] In: Keirse MJNG, Renfrew MJ, Neilson JP, Crowther C. (eds) Pregnancy and Childbirth Module. In: The Cochrane Pregnancy and Childbirth Database.
(database on disc and CD-ROM ) The Cochrane Collaboration; Issue 2, Oxford: Update Software 1995 (Available from BMJ publishing group, London)
8) Thorp JA, Hu DH, Albin RM, et al. The effect of intrapartum epidural analgesia on nulliparous labor; a randomized, controlled, prospective trial. Am J Obstet Gynecol 1993; 169(4): 851-858
9) Paterson Catherine, Banfield Philip. Epidural analgesia and maternal satisfaction. BMJ 1991 v302: 1079
10) Thorp JA, Meyer BA, Cohen GR et al. Epidural analgesia in labor and cesarean section for dystocia. Obstet Gynecol Surv 1994; 49(5): 362-369
11) Lirzin JD, Jacquintot P, Dailland P, et al. Controlled trial of extradural bupivicaine with fentanyl, morphine or placebo for pain relief in labour. Br J Anaesth 1989; 62: 641-644
12) Caldwell LE, Rosen MA, Shnider SM. Subarachnoid morphine and fentanyl for labor analgesia. Efficacy and adverse effects. Reg Anesth 1994;19:2-8
13) John Paull, Faculty of Anaesthetists, Melbourne. Quoted in: "The perfect epidural for labour is proving elusive" New Zealand Doctor. 21 Oct 1991
14) as above
15) Buggy D, Gardiner J. The space blanket and shivering during extradural analgesia in labour. Acta-Anaesthesiol-Scand 1995; 39(4): 551-553
16) Camann WR, Hortvet LA, Hughes N, et al. Maternal temperature regulation during extradural analgesia for labour. Br J Anaesth 1991;67:565-568.
17) Kennell J, Klaus M, McGrath S, et al. Continuous emotional support during labor in a US hospital. JAMA 1991;265:2197-220
18) Stride PC, Cooper GM. Dural taps revisited: a 20 year survey from Birmingham Maternity Hospital. Anaesthesia 1993; 48(3):247-255
19) Epidurals for pain relief in labour: Informed choice leaflet for women. MIDIRS and the NHS centre for Reviews and dissemination 1997.
20) Epidural pain relief during labour; Informed choice for professionals. MIDIRS and the NHS centre for Reviews and dissemination 1997.
21) see 13
22) see 13
23) Rawal N, Arner S et al Ventilatory effects of extradural diamorphine.Br J Anaesthesia 1982;54:239
24) Howell CJ, Chalmers I. A review of prospectively controlled comparisons of epidural with non-epidural forms of pain relief during labour. Int J Obstet Anaesth 1992;1:93-110
25) Fernando R, Bonello E et al. Placental and maternal plasma concentrations of fentanyl and bupivicaine after ambulatory combined spinal epidural (CSE) analgesia during labour. Int J Obstet Anaesth 1995;4:178-179
26) Caldwell J, Wakile LA, Notarianni LJ et al. Maternal and neonatal disposition of pethidine in child birth- a study using quantitative gas chromatography-mass spectrometry. Lif Sci 1978;22:589-96
27) Bratteby LE, Andersson L, Swanstrom S. Effect of obstetrical regional analgesia on the change in respiratory frequency in the newborn. Br J Anaesth 1979; 51:41S-45S
28) Swanstrom S, Bratteby LE. Metabolic effects of obstetric regional analgesia and of asphyxia in the newborn infant during the first two hours after birth I. Arterial blood glucose concentrations. Acta Paediatr Scand 1981; 70:791
29) Enkin M, Keirse M, Renfrew M, Neilson J. A Guide to Effective Care in Pregnancy and Childbirth. P 287 Oxford University Press 1995
30) Smith A. Pilot study investigating the effect of pethidine epidurals on breastfeeding. Breastfeeding Review, Nursing Mothers Association of Australia. V5 no1 May 1997.
31) Walker M. Do labor medications affect breastfeeding? J Human Lactation 1997;13(2) 131-137
32) Goodfellow CF, Hull MGR, Swaab DF et al. Oxytocin deficiency at delivery with epidural analgesia. Br J Obstet Gynaecol 1983; 90:214-219
33) Insel TR, Shapiro LE. Oxytocin receptors and maternal behavior. In Oxytocin in Maternal Sexual and Social Behaviors. Annals of the New York Academy of Sciences, 1992 Vol 652. Ed CA Pedersen, JD Caldwell, GF Jirikowski
and TR Insel pp 122-141 New York, New York Academy of Science
34) Howell CJ, Chalmers I A review of prospectively controlled comparisons of epidural with non-epidural forms of pain relief during labour. Int J Obstet Anaesth 1992 1: 93-110
35) See 31
36) Scanlon JW, Brown WU, Weiss JB Alper MD. Neurobehavioral responses of newborn infants after maternal epidural anesthesia. Anesthesiology, 1974; 40: 121-128
37) Morikawa S, Ishikawa I, Kamatsuki H, et al. Neurobehavior and mental development of newborn infants delivered under epidural analgesia with bupivicaine. Nippon Sanka 1990; 42: 1495-1502
38) Lester BM, Heidelise A, Brazelton TB. Regional obstetric anesthesia and newborn behavior: a synthesis toward synergistic effects.Child Dev 1982; 53;687-692
39) Rosenblatt DB, Belsey EM, Lieberman BA et al. The influence of maternal analgesia on neonatal behaviour II epidural bupivicaine. Br J Obstet Gynecol 1981 24;649-670
40) Seposki C, Lester B, Ostenheimer GW, Brazelton, TB. The effects of maternal epidural anesthesia on neonatal behavior during the first month. Dev Med Child Neurol 1992:34;1072-1080
41) Murray AD, Dolby RM, Nation RL, Thomas DB.Effects of epidural anesthesia on newborns and their mothers. Child Dev 1981; 82:71-82
42) Morgan BM, Bulpitt CJ, Clifton P, Lewis PJ. Analgesia and satisfaction in childbirth (the Queen Charlotte's 1000 mother survey) Lancet 1982; 2 (Oct 9) 808-810