This post is the second in a series by Dr. Michael Klein. You can find the first installment of this series here.

Research into the consequences of epidural use
Whether the benefits of epidural analgesia outweigh the potential risks has been the subject of many controversies over recent decades. In my department we have studied these controversies extensively. On a regular basis we looked at our own performance in caring for laboring women. The Department of Family Practice at British Columbia Women's Hospital in Canada (familiarly known as 'BC Women's) is made up of over 100 family doctors, who all attend births. BC Women's is the largest maternity hospital in Canada, with more than 7,000 births per year, and family doctors are responsible for almost half of these births, despite the hospital also being the tertiary care referral centre for the province. This makes us the largest group of family doctors attending births in Canada.

We knew from the literature that epidural analgesia use early in labour, before the fetus was well down in the pelvis, could cause malpositioning (occiput posterior or transverse) (7) due to extension of the fetal head. If the fetal head is extended, it cannot rotate or descend. We found that physicians who ordered epidurals frequently and early in labour had more patients with malpositioned fetuses.(8) They also had more patients who received greater amounts of synthetic oxytocin augmentation of labour. They had fewer spontaneous births and more cesarean sections than those in the department who used epidurals less often. Also, surprisingly, high epidural rates were associated with increased numbers of newborns with low 5-minute Apgar scores and more babies admitted to the newborn intensive care unit.

We found that physicians with mean epidural rates under 40% for women having their first baby, had cesarean section rates of about 10%. In contrast, those family doctors with mean epidural rates of 71-100% had cesarean section rates of 23.4%, the others having rates between the two extremes. The women cared for by the three groups were similar. Thus it appeared that only physician practice difference could have accounted for such large differences in outcome. Interestingly, the caesarean rates of women who were having their second or more births were unaffected by the way in which their doctors prescribed epidurals.

Our departmental experience was similar to results from an observational study in which we compared outcomes at a nearby community hospital with our tertiary care centre. (9) In the community care setting, mean epidural analgesia rates were 15.4% compared with 67.2% in the tertiary care center, for comparable women. The odds of having a cesarean section were 3.4 times greater at the tertiary care centre than in the community hospital. The increased and earlier use of epidural analgesia in the tertiary care setting almost completely explained this difference. The community hospital setting encouraged the use of other non-epidural pain coping techniques, resulting in later epidural placements compared to those in the tertiary hospital setting. We were also interested to note that those physicians who ordered epidurals less often actually spent more time with their patients, even though on average their patients spent lesstime in hospital. The time they spent with their patients involved more intimate, hands-on, supportive care.

It is because of these studies that we had trouble accepting the results of the 2004 Cochrane meta-analysis that concluded that epidural analgesia did not raise the cesarean section rate.(10) This conclusion was the same in the most recent Cochrane meta-analysis,(11) this new one deeply flawed by the inclusion of many studies of women who suffered from complex medical conditions, and many studies that randomized women late, particularly since conventional practice is to use epidurals earlier. Clearly, any meta-analysis is only as good as the individual studies included - illustrating the well-known principle: garbage in, garbage out.

Epidural Analgesia has transformed birth:
In fact, it appeared to us that the increasing use of epidural analgesia was transforming birth. This observation was confirmed by a report from the Canadian Institute for Hospital Information, which indicated that 4 in 5 Canadian women received one or more major obstetrical interventions, with epidurals high on the list at rates of 40-50% of births in various Canadian settings. (12)

We decided to look more closely at earlier Cochrane (10) individual studies that made up the meta-analysis addressing the effect of epidural analgesia on c-section rates. These studies revealed that, epidural analgesia increased the length of the first stage of labour by 4.3 hours. Similarly, the second stage of labour was increase by 1.4 hours. Malpositions were found in 15% of cases where epidurals were used but in only 7% of cases where narcotics were used. Synthetic oxytocin augmentation of labour was found in 52% of women with epidurals and in 7% of women who had narcotic analgesia. Instrumentation (forceps and vacuum) was found in 27% of epidural cases compared with a rate of 16% among women not getting an epidural. Maternal fever was dramatically higher in the epidural versus narcotic analgesia groups - 24% and 6%, respectively.

Maternal fever is a common side effect of epidural analgesia because it interferes with the sympathetic and parasympathetic nervous system balance, altering the body's normal methods to eliminate heat. And since it is hard to know if the fever is due to infection or an epidural effect, a septic work-up is usually carried out following birth on both the mother and baby, including blood and possibly cerebral spinal fluid tests, along with frequent placement on prophylactic antibiotics. Additionally, increased maternal temperature can cause a rapid fetal heart rate and, thus, prompt a caesarean section due to concerns for fetal well-being.

Additionally, several studies have shown that perineal trauma increased two-fold in women who had had an epidural, due in part to an increased use of forceps and vacuum, which in and of themselves are associated with more perineal trauma (with or without epidurals). (13, 14)

Given all the other increases in intervention rates, we found it hard to understand why cesarean section rates were not also higher in theCochrane meta-analysis. In fact, when we separated out the studies that made up the 2004 Cochrane meta-analysis, we found that, in those studies that showed no difference in cesarean section rates, epidurals had been administered after labour was well established (in the active phase at 4-5cm or more of cervical dilation). In the studies where epidurals were given early on in labour, before the active phase (before 4-5cm of cervical dilation) - the cesarean section rate increased more than 2.5 times. (15, 16)

Inadvertently, the Cochrane meta-analysis of epidural analgesia has caused more frequent use of epidurals, resulting in more continuous electronic fetal monitoring, immobility of the labouring woman, increased instrumentation and perineal trauma, and an increase in the cesarean section rate. Because more women will have received a cesarean section, another consequence will be an increase in problems in subsequent pregnancies relating to placentation issues (previa, accrete, percreta, abruption), infertility, and ectopic pregnancy.(17-20) In most maternity care settings, these down-stream consequences ('collateral damage') from epidural use are not discussed.

Published: February 24, 2011

Tags