We read with interest C. M. Cameron et al. 's review of epidural morbitity,1having recently published a similar survey of our own.2As in their review, we examined more than 8,000 cases of patients receiving epidural analgesia after surgery over a 6-yr period in our hospital. This compares with their 16 yr, perhaps reflecting not only the different sizes of our two centers, but also differences in practice. Despite this, the results are remarkably similar. We demonstrated an epidural abscess rate of 1:1,350 (vs.  1:1,368) and an epidural hematoma rate of 1:2,700 (vs.  1:4,105). However we also identified 3 cases of meningitis (1:2,700). Although this is a recognized complication of postoperative epidural infusions, Cameron et al.  do not discuss meningitis and seem not to have searched their database for this significant complication. We assume that this was an oversight.

The presenting symptoms did vary between the patients in the two surveys. All of our patients with epidural abscesses displayed back pain, and most had features of meningitic irritation. In contrast, both of these symptoms were rare in their patients. A key point to note from both studies is the delay before clinical presentation of an epidural abscess (6–31 and 5–11 days, respectively). Therefore, almost all patients develop symptoms after the epidural catheter has been removed, and many after discharge home. One of our patients with an epidural abscess did not make a full neurologic recovery. This patient developed symptoms of spinal cord compression at home but did not return to the hospital until paraplegic. We now provide written advice to patients receiving epidural analgesia, detailing specific signs and symptoms that require medical investigation should they occur after discharge.

Chlorhexidine in alcohol has been our standard solution for skin preparation for many years. We note that in 2004, Cameron et al.  reverted from chlorhexidine back to an iodine solution for skin preparation. Since then, their incidence of insertion site infections seems to have increased. We wonder whether they think this increase is significant enough to justify returning to an alcoholic chlorhexidine solution.

Common with most of the published data on epidural infections, the implicated organism in their patients was Staphylococcus aureus . What was striking in our data was the high incidence of methicillin-resistant S. aureus , which was cultured from five of the nine patients with infective complications. Unfortunately, routine methicillin-resistant S. aureus  screening was not performed during our survey. We were thus unable to investigate whether methicillin-resistant S. aureus  colonization of the patient, the staff, or the ward predisposed to these complications. We would now, however, be reluctant to insert an epidural catheter in any patient colonized with methicillin-resistant S. aureus .

Cameron et al.  discovered that epidural abscesses were associated with epidural site infection that in turn was related to the duration of epidural catheterization and abdominal or thoracic surgery (odds ratio, 3.3). Interestingly, in our series, insertion of an epidural at a thoracic level did not increase the risk of developing either an epidural abscess or meningitis (odds ratio, 1.18; 95% confidence interval, 0.3–4.7). Perhaps, as they suggest, it is the duration of the epidural infusion that is the more critical factor. We recommend that epidurals are removed within 3 days of insertion.

Although Cameron et al.  identified two spinal hematomas, only one was epidural. We would be interested to know where the other hematoma was located in relation to the dura and how it presented (i.e. , whether it affected patient 1 or patient 3).

When we presented the results of our survey, both locally and nationally, the frequency with which these complications occur came as a surprise to many. It is reassuring that on the other side of the globe, the incidence seems to be similar. Only through continued audit of our practice can we determine the true risk of the interventions we undertake to benefit rather than harm our patients. The Royal College of Anaesthetists has recently started a national audit of these complications in the United Kingdom, and we would commend to others to consider a similar register.

*Christchurch Hospital, Christchurch, New Zealand. shanemccabe@totalise.co.uk

Cameron CM, Scott DA, McDonald WM, Davies MJ: A review of neuraxial epidural morbidity: Experience of more than 8,000 cases at a single teaching hospital. Anesthesiology 2007; 106:997–1002
Christie IW, McCabe S: Major complications of epidural analgesia after surgery: Results of a six-year survey. Anaesthesia 2007; 62:335–41