Epidural abscess complicating insertion of epidural catheters

M. Hearn1 and C. Roberts2

1 Torquay, UK 2 Gloucester, UK

Editor—I was very interested to read the case reports by Dr Phillips and colleagues1 of epidural abscess formation after the use of epidural catheters. They make the point that not only does the incidence appear to be increasing, but also development of signs and symptoms often occurs after the catheter has been removed, making diagnosis difficult. I totally agree with their observations.

In Torbay, we perform about 600 non-obstetric epidurals a year. Over the past 3 years we have diagnosed three abscesses related to epidural catheters, although before this there had been none. Other centres in the south west of England have reported similar events. All three of the cases in Torbay presented after the catheter had been removed. One patient had already been moved to a community hospital for convalescence. The presentation in all three patients included a mild pyrexia, back pain and inflammation at the site where the epidural had been inserted, and an increase in C-reactive protein. All patients had undergone bowel surgery, one for cancer and two for ulcerative colitis. The epidural catheters had been in situ for no more than 5 days. One patient had an inflamed epidural site with pus when the catheter was removed (which grew Staphylococcus aureus). Another patient developed a superficial abscess a week after removal of the catheter, which proved to be because of MRSA. No patient had any neurological signs or symptoms or meningism. The diagnoses were established by means of MRI scanning. Two patients required surgical evacuation of the abscess, and the third patient was managed conservatively. There were no long-term sequelae.

As a result of this increasing problem, we have clarified our guidelines regarding the insertion and management of epidurals, and have developed a new nursing care plan to ensure daily site observation and documentation. We have also produced an algorithm that provides easy instructions for use with suspected cases. Constant education helps to raise awareness of the problem. However, in view of the occasional late presentation, perhaps we should be providing patients and their GPs with written information, to ensure early recognition of all cases in an endeavour to reduce the chance of permanent neurological damage.

It is recognized that epidural infections (deep and superficial) are more likely in patients on ITU, or patients receiving immunosuppression, with diabetes, cancer or an underlying infection.2 Epidurals are now being performed more frequently, particularly on patients with greater infection risks, making the previous figure of 1/19303 inaccurate. I believe that Dr Phillips’ figures are more representative. It is important to realize that in the high-risk groups, the chance of developing an epidural abscess might be as high as 1 in 100–200. This has serious implications regarding performing epidurals in such patients and obtaining their consent.

At the Acute Pain Symposium held in Chester, England in 2002, it was felt that we should develop a national database to collect figures regarding the incidence of major epidural-related complications such as abscesses, haematomas and nerve damage. We are trying to do this in the south west of England, but for many units, collecting the information to provide the denominator is difficult, as it stretches an already over-committed and under-funded service. In view of the seriousness of the possible outcomes, however, I do feel that we should be able to produce accurate national figures to enhance patient safety and improve the information available to patients.

M. Hearn

Torquay, UK

Editor—We thank Dr Hearn for her support of our paper. We wholeheartedly agree that that the occurrence of symptoms of pyrexia, back pain and inflammation at the epidural insertion site should raise suspicion of impending or actual epidural abscess formation. While all patients must receive meticulous care at the time of epidural insertion, it is essential that high standards of infection control and prevention continue during the time the epidural catheter remains in situ. We too ensure that the epidural insertion site is checked daily and that its condition is recorded on the epidural observation chart. Hopefully this brings any potential epidural infection to the attention of the ward nurses and acute pain team. However, as in one of our case reports1 and in all three of Dr Hearn’s reports, epidural site infection can develop after removal of the epidural catheter. It may be that we should continue to look for signs of epidural infection daily until discharge from hospital. Whether this should continue after discharge is difficult to say. Its instigation would certainly prove difficult in practice. We should, however, attempt to raise awareness of the presentation of epidural site infection in the primary care setting.

It may well be that there are groups of patients who are at higher risk of epidural infection. If a national database were set up to prospectively record the incidence of major epidural complications, it might be possible to retrospectively look for such risk factors in these patients. By defining those patients who are at high risk from complications of epidural anaesthesia, we could adjust our clinical practice to minimize the risk of such complications. This would also improve our ability to provide informed consent.

C. Roberts

Gloucester, UK

References

1 Phillips JMG, Stedeford JC, Hartsilver E, Roberts C. Epidural abscess complicating insertion of epidural catheters. Br J Anaesth 2002; 89: 778–82[Abstract/Free Full Text]

2 Kindler CH, Seeberger MD, Staender SE. Epidural abscess complicating epidural anaesthesia and analgesia. Acta Anaesthesiol Scand 1998; 42: 614–20[ISI][Medline]

3 Wang LP, Hauerberg J, Schmidt JF. Incidence of epidural abscess after epidural anaesthesia. A national 1-year survey. Anesthesiology 1999; 91: 928–36[CrossRef]