Epidurals and risk: it all depends
Risk: it all depends, doesn't it? It's one of those irregular verb things, is risk; I am a perfectly safe driver, you are a bit dodgy, he is an accident waiting to happen. But if I drive a modern car with seat belts and airbags, at moderate speed, on quiet roads, during the day, while you drive an old banger on the motorway, and he drives an even worse banger without seatbelts home from the pub when dead drunk, the point is made. It all depends.
When it comes to risks in medical procedures, it all depends, but we don't often have examples of just how much the setting can change the risks. For at least one example we now have an inkling of the effect of setting.
Bandolier 152 discussed the risks following use of an epidural during childbirth. Briefly, there was a risk of deep epidural infection in 1 in 145,000, epidural haematoma in 1 in 170,000, and persistent neurological injury lasting more than one year in 1 in 240,000 women. The rates were abstracted from larger observational studies, and the risk came both from spontaneous occurrence and from the epidural cather and anything injected through it, but combining all of them the risks were low.
Two further systematic reviews [1,2] have examined similar outcomes in two very different circumstances, postoperatively after cardiovascular surgery, and in chronic pain.
The methods of the two new systematic reviews [1,2] were similar to those described for obstetric patients in Bandolier 152. The same outcomes were sought, namely epidural haematoma, infection, and neurological injury in the shorter (less than one year) and longer term. Studies with fewer than 200 patients were not sought, as they were unlikely to report on rare events, and case reports would be unlikely to have a denominator to calculate rate.
Table 1 reprises the main findings for the four outcomes in all three settings, as well as providing brief information about the patients, duration of epidural catheter use, and number of patients on whom information was available. The 'rule of 3' (Bandolier 23) was used to calculate maximum event rates when there were no events; this says that the upper 95% confidence interval that the event will not occur is 1 in the total divided by three (n/3).
Table 1: Risks associated with epidural catheter in different circumstances
|Duration of catheter
often single shot
|Number of patients|
|Epidural haematoma||Number of events|
|Deep infection||Number of events|
|Neurological injury >1 year||Number of events|
|Neurological injury <1 year||Number of events|
In 12 studies were 14,105 patients, of whom 5,026 (36%) had vascular surgery, 4,971 (35%) cardiac surgery, and 4,108 (29%) thoracic surgery. There were no cases of epidural haematoma (Table 1), giving maximum (worst case) risks following epidural anaesthesia in cardiac, thoracic, and vascular surgery of 1 in 1,700, 1 in 1,400 and 1 in 1,700 respectively.
In all these surgery types combined the maximum expected rate would be 1 in 4,700. In all these patients combined there were eight cases of transient neurological injury, a rate of 1 in 1,700 (95% confidence interval 1 in 3,300 to 1 in 850). There were no cases of persistent neurological injury (maximum expected rate 1 in 4,700).
In this chronic pain review epidural catheters had to be in place for seven days or more. Twelve studies provided information on 4,628 patients. There was no information on epidural haematoma or neurological injury.
There were 57 deep infections (1.2%). Ten of the 12 studies had deep infection rates of 2% or less (Figure 1). The incidence of deep infection was 1 per 2391 days of treatment, or 0.4 per 1000 catheter treatment days. In nine studies (1503 patients), predominantly in cancer, and with average catheter duration of 74 days, the deep infection rate was 2.8%. The proportion of patients with infection of any type was higher in cancer patients with longer catheter duration.
Figure 1: Deep infection rates in individual studies of patients with chronic pain
It is obvious that the three situations are not the same. Obstetric epidural use involves young, mainly healthy, women in whom epidural catheters are used mainly for a short period. Cardiovascular patients will tend to be older, and sicker, and while epidural catheters will generally be used only for a few days, anticoagulation will make their use more risky. In chronic pain, very long term use is often in very ill patients.
In these three circumstances our evaluation of the risk understandably differs. The most glaring difference is the amount of information we have, with 100 times more for obstetrics than cardiovascular surgery. This makes any estimate of risk more credible.
Indeed, the number of events was so few for cardiovascular surgery and chronic pain settings that we have to invoke the rule of three to estimate a maximum risk. Some events were just not reported in chronic pain patients, probably because they were not very important in the circumstances.
As best we can judge, the risks themselves differ between circumstances. Figure 2 shows the estimated risk of neurological injury lasting one year or more in obstetrics and cardiovascular surgery using a Paling Perspective Scale. The carry home message, though, is not how good we are at estimating risk, but about how little we seem to know.
Figure 2: Paling perspective scale showing risk of long term neurological problems with epidural catheter with obstetric use and in thoracic surgery
- W Ruppen et al. Incidence of epidural haematoma and neurological injury in cardiovascular patients with epidural analgesia/anaesthesia: systematic review and meta-analysis. BMC Anesthesiololgy 2006 6:10.
- W Ruppen et al. Infection rates associated with epidural indwelling catheters for seven days or longer: systematic review and meta-analysis. BMC Palliative Care 2007 6:3.