Cutaneous fluid leakage after epidural catheter removal.

Sir,

Epidural analgesia is commonly used to provide effective perioperative analgesia. Cerebrospinal fluid (CSF)-cutaneous fistula is a very rare complication of epidural analgesia.[1] We present a case of persistent cutaneous fluid leak from the epidural puncture site after catheter removal on the 5th post-operative day.

A 55-year-old female patient was scheduled for an exploratory laparotomy for cecal carcinoma. Patient's past medical history was not significant. General anesthesia with thoracic epidural was planned for her. In the operating room, an epidural catheter was placed at the first attempt in the T9-T10 inter-space by the midline approach in the sitting position using loss of resistance to saline with a 16G Tuohy needle. A transparent dressing was applied and general anesthesia was induced. An epidural loading dose of 10 ml of 0.25% bupivacaine was administered after a test dose of 3 ml of 2% lignocaine-adrenaline solution. This was maintained with a continuous epidural infusion of 0.125% bupivacaine at 4 ml/h and the surgery was completed in 6 h without any complications. The epidural was used satisfactorily post-operatively with boluses of bupivacaine 0.125% for pain relief. On the 4th post-operative day, the epidural catheter was removed intact.

On the 5th post-operative day, the anesthetist was called to evaluate a “clear fluid leak” from the epidural puncture site. On examination, clear fluid was leaking slowly from the epidural puncture site with soakage of patient's bed sheet. Although patient had no complaints, generalized edema over the entire body was noted. There were no neurological or meningeal signs. Blood investigations were within the normal limits except for a low albumin level (serum albumin - 1.6). A sterile colostomy bag was put at the puncture site. Within an hour, 40 ml of a clear, pale yellow fluid collected in the bag [Figure 1]. A sample of this fluid was sent for biochemical analysis along with a venous blood sample taken at the same time.

Figure 1

Oedema fluid collected in colostomy bag

The fluid glucose was 4.8 mmol/l, chloride was 110 mmol/l and protein was 60 g/l. The venous sample also had similar values (glucose 4.9 mmol/l, chloride 110 mmol/l, protein 62 g/l). A 16 gauge hypodermic needle was inserted 7 cm lateral to the puncture site and 1 cm deep. Clear fluid started dripping from this needle too which was collected and sent for biochemistry [Figure 2]. The appearance, glucose, chloride and protein levels of this fluid were similar to the leak fluid and a diagnosis of interstitial fluid was made and a compression dressing was applied over the epidural puncture site. Gradually, the leak decreased and stopped completely on the 10th post-operative day.

Figure 2

Leak fluid collection with hypodermic needle

CSF-cutaneous fistulae have been associated with a variety of clinical situations such as post-neurosurgery, lumbar spinal drains, trauma, tumor and infection.[2] Although rare, CSF fistulae have been reported in anesthesia practice too, secondary to subarachnoid or epidural punctures.[12]

We present a patient who had a fluid leak from the epidural puncture site and was assumed to be a CSF-cutaneous fistula. The correct diagnosis of edema fluid was later made after conducting a few simple tests.

The leak fluid was sent for biochemical analysis. The values obtained were then compared with patient's blood values and normal values of CSF, plasma and interstitial fluid. The biochemical analysis of the leak fluid was comparable with the normal values of interstitial fluid and also patient's blood values. Another test that aided our diagnosis was inserting a needle away from the epidural site and analyzing the fluid collected from there, as performed by Ennis and Brock-Utne[2] This fluid had the same biochemical values as the leak fluid, hence ruling out CSF.

The low pressure and distensibility of epidural space makes it a probable site for the transudate accumulation and in our case, hypoalbuminemia probably aggravated the edema formation. CSF-cutaneous fistulae usually occur within 48 h, while this leak was first seen on the 5th post-operative day.[2] A CSF leak of this magnitude would also cause severe headache, but our patient had no headache even in the sitting position.

A simple bedside test to differentiate edema fluid from CSF is to determine the pH of the fluid. A more alkaline pH is in favor of edema fluid[23] [Table 1]. The diagnosis of CSF can be made by testing the fluid for CSF specific acetyl cholinesterase by protein electrophoresis.[4] β2 transferrin immunofixation assay confirms CSF and can be performed on fluid samples as low as 0.1 ml.[5] A neurosurgery opinion with a computed tomography scan can be done to investigate the possible cause of an increase in CSF pressure.[2] If sufficient fluid cannot be obtained, then a radioisotope myelography can diagnose a CSF fistula.[67] These tests are expensive and may not be easily available.

Table 1

Biochemical values of CSF, interstitial fluid, plasma and leak fluid

However, if even 0.5 ml of the leak fluid can be collected, then a simple chemical analysis for chloride, glucose and proteins can determine the fluid origin and avoid expensive tests.[78]

In conclusion, our case shows that every epidural site leak need not be CSF and simple inexpensive tests are adequate for differentiating between CSF and interstitial fluid.