Extensive Spinal Epidural Abscess Associated With an Unremarkable Recovery

SUMMARY

Spinal epidural abscess is a rare complication of epidural catheter insertion, with an incidence reported to be as low as 0.02%, but with a high morbidity and mortality. This reflects the difficulty in diagnosis of the condition, as early symptoms and signs are usually non-specific with late neurological manifestations. We report a case of spinal epidural abscess in a diabetic patient who underwent splenectomy and distal pancreatectomy. Early investigation with magnetic resonance imaging was effective in demonstrating an extensive epidural abscess involving the thoracic spine, with extension to the base of the skull and associated cord compression. Extensive multilevel laminectomies with thorough irrigation and washout of the epidural space were successful in treating this patient and preventing the development of permanent neurological sequelae.

Key Words: ANAESTHETIC TECHNIQUES, EPIDURAL: infection, abscess, diabetes, splenectomy

Spinal epidural abscess is a rare complication of epidural catheter insertion. The incidence is estimated to be between 0.02 to 0.12%1-3. The condition is associated with high morbidity and mortality; with only half of treated patients leaving hospital with minimal or no residual symptoms2. These figures indicate that the greatest challenge is early diagnosis and hence treatment of this condition. Patients usually present with back pain and associated systemic signs such as pyrexia, with investigations demonstrating elevated white cell count, C-reactive protein (CRP) or erythrocyte sedimentation rate. Unfortunately, all these markers are non- specific and common after many types of surgery. Neurological signs such as impaired sensation, leg weakness, or bladder or bowel disturbance usually occur at a late stage when irreversible damage may have already occurred.

We report a rare case of an extensive spinal epidural abscess involving the spinal canal from T5 to T12, with extension to the cranio-vertebral junction associated with cord compression at C2, in a patient following distal pancreatectomy and splenectomy. Early diagnosis and surgical intervention was successful at treating the condition and the patient was discharged free of any residual symptoms.

CASE REPORT

A 50-year-old female was admitted for resection of a microcystic adenoma of the pancreas. Past medical history included diet- controlled diabetes, hypothyroidism and a previously treated benign breast abscess; routine medication consisted of thyroxine 50 g daily. Before induction of anaesthesia a thoracic epidural catheter was inserted under sterile conditions at around T9/10. Lignocaine 5 ml 1% was used for skin anaesthesia; a 16 gauge Tuohy needle with a loss of resistance to saline technique was used to identify the epidural space at 4 cm, and an initial bolus of 15 ml of 0.5% bupivacaine was given in 5 ml aliquots to establish an epidural block. Subsequently, an infusion of 0.1% bupivacaine with 0.1 mg/ml diamorphine was commenced at a rate of 5 ml/h. The anaesthetist wore a cap and sterile gloves and gown. The skin was prepared using alcoholic chlorhexidine. Intravenous prophylactic antibiotics were given perioperatively and included 1.5g cefuroxime and 500 mg metronidazole. The operative procedure involved distal pancreatectomy and splenectomy, following which the patient was returned to the ward where the epidural infusion was continued. The patient initially made an uneventful recovery, but was noted to experience intermittent confusion with associated agitation and respiratory compromise on day 3 postoperatively. Although the confusion persisted intermittently, the other symptoms resolved and the epidural analgesia was stopped at day 5. Due to the timing of thromboprophylaxis administration at 1800h daily (dalteparin 2500IU- Fragmin, Pharmacia and Upjohn, Kalamazoo, MI, U.S.A.), the epidural catheter was removed on the following day (day 6 postoperatively).

The patient continued to make a good recovery but eight days postoperatively developed symptoms of headache and neck stiffness with associated pyrexia of 39C. Complete neurological examination at this time did not detect any focal abnormalities, however signs of meningism were noted, along with a small tense abscess at the insertion site of the previous epidural cannula. Incision and drainage of this showed only a small superficial collection. A possible diagnosis of epidural abscess was made with associated meningitis. Initial investigations included haematology and biochemistry which showed a white cell count of 54.4 10^sup 9^/l, CRP 293.4 mg/l and culture of pus from the abscess site as well as blood, both subsequently resulted in a heavy growth of Staphylococcus aureus sensitive to flucloxacillin. Lumbar puncture was considered, but it was felt an MRI scan of the thoracolumbar spine should be carried out initially, and this showed a posterolateral epidural collection extending from T5 to T12 (Figure 1). Further images of the cervical spine were then viewed and extension of the epidural abscess to the base of the skull with associated cord compression at the level of C2 were identified (Figure 2).

FIGURE 1: (1a) Sagittal magnetic resonance (MR) image of the thoracolumbar spine revealing an epidural abscess (indicated by the longer arrow) extending from T5/6 to T11/12. Spinal cord distortion can be noted from T6 to T12 due to posterior compression by the abscess. Furthermore, the actual site of insertion of the previous epidural cannula can be seen, with extruding pus, indicated by the shorter arrow. (1b) Axial MR image identifying a posterior epidural abscess, indicated by the arrow, impinging on the spinal cord at the T8/9 intervertebral disc level.

Following diagnosis of the epidural abscess, surgical treatment was undertaken on day 11 post catheter insertion. This involved multiple level laminectomies at C1, C3/4, C7/T1, T4/5 and T7/8. At each level a soft catheter was advanced cranially as well as caudally to washout the entire epidural space. The patient was started on chloramphenicol according to microbiological advice and transferred to intensive care. The patient continued to have small amounts of discharge from the wound while in intensive care but this responded to regular dressings. Antibiotics were changed on the basis of continuing bacteriological culture and sensitivity to pipericillin and tazobactam, and the patient was discharged to the general ward 12 days following spinal surgery. The patient continued to make a steady recovery on the ward and was finally discharged home 30 days after spinal surgery (41 days following her initial operation). At time of discharge she was completely symptom-free and mobilizing safely. The patient was also apyrexial with the wound no longer discharging.

FIGURE 2: (2a) Sagittal MR image of the cranio-vertebral junction revealing extension of the epidural abscess to the level of the base of the skull. The arrows identify the epidural collection lying anteriorly, posterior to the clivus, as well as posteriorly. (2b) Axial MR image at the level of C1/2 revealing the presence of an epidural abscess, indicated by the arrow, surrounding the dens.

DISCUSSION

In addition to its widespread use during labour4, epidural analgesia is also frequently used as a method of pain control following major surgery. This can be attributed to its success at relieving labour pain5 as well as pain in the postoperative period. This fact is supported by the findings of Dolin et al6 who undertook a meta-analysis of 410 papers and confirmed that less pain is experienced in the postoperative period by patients receiving epidural analgesia compared with patient controlled analgesia (PCA) or intramuscular analgesia. Furthermore, the use of postoperative epidural analgesia, but not systemic opioids, may attenuate some of the adverse pathophysiological effects associated with surgery, and result in decreased morbidity and mortality7, particularly if following abdominal aortic8 or orthopaedic procedures9. Additionally, epidural analgesia is associated with fewer side- effects compared with morphine PCA, which commonly causes nausea and vomiting, along with sedation and possible respiratory depression.

There has been an increase in the number of cases of epidural abscess being reported and this may be attributed either to the increasing use of this analgesia technique, or a previously underestimated incidence. Also there is agreement in previous studies that this condition is very difficult to diagnose due to the early signs and investigations being nonspecific. These signs include pyrexia and back pain, with an elevated white cell count or inflammatory markers1-3,10,11. This delay in diagnosis is associated with a greater likelihood of permanent neurological sequelae as well as increased mortality2. Furthermore, although epidural abscesses usually present five days post catheter insertion, delayed presentation up to 32 days later has been reported10,12-15. This raises the question whether it is direct spread of bacteria from the skin at time of insertion, or late haematogenous spread from a transient bacteraemia that actually results in formation of the abscess. Nonetheless, increased suspicion is necessary for a significant length of time, even long after catheter removal. Although local signs of infection may be notedat the site of previous insertion, these may not necessarily precede epidural abscess development.

Increased suspicion is also necessary in particular groups of patients, such as those with diabetes mellitus, concomitant infection, intravenous drug use, immune-compromise or chronic medical illness, all being associated with an increased risk of developing epidural abscesses10. Diabetes has been identified as the most important risk factor16, with one study demonstrating that 13% of patients who presented with epidural abscesses without prior spinal injections had diabetes mellitus17. Our patient had a history of diabetes mellitus (albeit diet-controlled), which may have increased her susceptibility. In addition, our patient was rendered asplenic postoperatively and this may have been a contributory factor. However, there have been no previous case reports of epidural abscess in patients post-spenectomy. The infective organism in our patient was S.aureus and this has not been noted to have increased pathogenicity in asplenic patients unlike encapsulated bacteria18. Identifying S.aureus is not unusual, with this being the most commonly identified organism in cases of epidural abscess2. Other organisms that have been implicaed are Proteus mirabilis, Streptococcus pneumoniae, Escherichia coli, Staphylcoccus epidermidis and Streptococcus milleri.

Prevention of epidural abscess is the best way to prevent morbidity and mortality. However, it can be noted from a review of the literature that cases may still occur in centres where set protocols are in place to reduce the risk. These include scrubbing before performing catheter insertion, and wearing masks, gown and gloves as well careful skin preparation. Most centres also have a length of time limit for leaving a catheter in situ, even though reports have not proved this to be linked to abscess formation, with majority of cases being seen in less than five days of use10. Therefore, the risk of epidural abscess remains despite precautions, and careful consideration of the risks and benefits should be undertaken in each patient19. Once an abscess is suspected, the key to successful treatment is early diagnosis and this is best done by undertaking magnetic resonance imaging, which is very sensitive at identifying epidural abscesses2. MRI should be mandatory in patients who develop back pain or neck stiffness with an associated pyrexia and raised inflammatory markers in the post epidural cannulation period. Furthermore, there is benefit in undertaking this investigation urgently, with transfer of patients to specialist centres if necessary, even if neurological symptoms or signs are absent. Epidural abscesses most commonly affect the thoracic spine, followed by the lumbar spine2 and this is probably why, depending on area of clinical suspicion, MRI localized to one of these regions is often undertaken. Cervical epidural abscesses are very rare; this case had extensive involvement from lower thoracic spine to cranio- cervical junction and has never been reported before, but nonetheless, it does demonstrate why it is important to image the entire spine. Multilevel involvement has a far worse prognosis and requires wider surgical debridement.

Our experience, in common with others, is that surgical decompression and debridement should be the first line of treatment. This should be carried out as soon as possible after diagnosis and involve undertaking a laminectomy at the appropriate level13,16. Multilevel laminectomies should be undertaken if the epidural abscess is extensive. A sample of pus should be taken and sent for microscopy and culture to guide postoperative antibiotic use. Thorough washout of the epidural space should then be undertaken with irrigation via a catheter. Immediate wound closure can be undertaken but regular review should follow in the postoperative period to confirm resolution of infection. Appropriate antibiotics should be initiated and continued for at least six to eight weeks postoperatively. Apart from clinically monitoring the patient, white cell count and inflammatory markers may also be useful in the evaluation. Repeat MRI scan may be necessary to confirm the absence of further collection developing.

In summary, although epidural abscess is a rare complication of epidural catheter insertion, it is difficult to diagnose and hence is associated with significant morbidity and mortality. Increased vigilance is thus necessary to diagnose this condition early, with MRI of the spine being a very useful diagnostic tool. Early surgical decompression and debridement should be undertaken, with a long course of postoperative antibiotics and careful monitoring.

REFERENCES

1. Royakkers AA, Willigers H, van der Ven AJ, Wilmink J, Durieux M, van Kleef M. Catheter-related epidural abscesses-don’t wait for neurological deficits. Acta Anaesthesiol Scand 2002; 46:611-615.

2. Joshi SM, Hatfield RH, Martin J, Taylor W. Spinal epidural abscess: a diagnostic challenge. Br J Neurosurg 2003; 17:160-163.

3. Phillips JM, Stedeford JC, Hartsilver E, Roberts C. Epidural abscess complicating insertion of epidural catheters. Br J Anaesth 2002; 89:778-782.

4. 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; 169:851-858.

5. Leighton BL, Halpern SH. The effects of epidural analgesia on labor, maternal, and neonatal outcomes: a systematic review. Am J Obstet Gynecol 2002; 186:869-877.

6. Dolin SJ, Cashman JN, Bland JM. Effectiveness of acute postoperative pain management: I. Evidence from published data. Br J Anaesth 2002; 89:409-423.

7. Wu CL, Caldwell MD. Effect of post-operative analgesia on patient morbidity. Best Pract Res Clin Anaesthesiol 2002; 16:549- 563.

8. Park WY, Thompson JS, Lee KK. Effect of epidural anaesthesia and analgesia on perioperative outcome: a randomized, controlled Veterans Affairs cooperative study. Ann Surg 2001; 234:560-569.

9. Rodgers A, Walker N, Schug S et al. Reduction of postoperative mortality and morbidity with epidural or spinal anaesthesia: results from overview of randomised trials. BMJ 2000; 321:1493-1504.

10. Hernandez JM, Coyle FP, Wright CD, Ballantyne JC. Epidural abscess after epidural anesthesia and continuous epidural analgesia in a patient with gastric lymphoma. J Clin Anesth 2003; 15:48-51.

11. Evans PR, Misra U. Poor outcome following epidural abscess complicating epidural analgesia for labour. Eur J Obstet Gynecol Reprod Biol 2003; 109:102-105.

12. Dawson P, Rosenfeld JV, Murphy MA, Hellyar AG. Epidural abscess associated with postoperative epidural analgesia. Anaesth Intensive Care 1991; 19:569-572.

13. Hlavin ML, Kaminski HJ, Ross JS, Ganz E. Spinal epidural abscess: a ten-year perspective. Neurosurgery 1990; 27:177-184.

14. Wang LP, Hauerberg J, Schmidt JF. Incidence of spinal epidural abscess after epidural analgesia: a national 1-year survey. Ancsthcsiology 1999; 91:1928-1936.

15. Kindler CH, Seeberger MD, Staender SE. Epidural abscess complicating epidural anesthesia and analgesia. An analysis of the literature. Acta Anaesthesiol Scand 1998; 42:614-620.

16. Reihsaus E, Waldbaur H, Seeling W. Spinal epidural abscess: a meta-analysis of 915 patients. Neurosurg Rev 2000; 23:175-204.

17. Baker AS, Ojemann RG, Swartz MN, Richardson EP Jr. Spinal epidural abscess. N Engl J Med 1975; 293:463-468.

18. Davidson RN, Wall RA. Prevention and management of infections in patients without a spleen. Clin Microbiol Infect 2001; 7:657- 660.

19. Rigg JRA, Jamrozik K, Myles PS et al. Epidural anaesthesia and analgesia and outcome of major surgery: a randomised trial. Lancet 2002; 359:1276-1282.

A. ANSARI*, D. W. L. DAVIES[dagger], J. W. G. LOHN[double dagger], P. CULPAN, G. ETHERINGTON**

Department of Orthopaedics and Department of Anaesthetics, University College Hospital, Grafton Way, London, United Kingdom

* B.Sc., M.B., B.S., Orthopaedic SHO.

[dagger] D.R.C.O.G., D.C.H., F.R.C.A., Consultant Anaesthetist.

[double dagger] B.Sc., M.B., B.S., Orthopaedic SHO.

F.R.C.S., Orthopaedic Specialist Registrar.

** F.R.A.C.S., Consultant Spinal Surgeon.

Address for reprints: Dr A. Ansari, 21 Mount Pleasant Road, Chigwell, Essex IG7 5 EP, United Kingdom.

Accepted for publication on July 20, 2004.

Copyright Australian Society of Anaesthetists Dec 2004