The case reported by Dietrich and Smith1again demonstrates that performing steroid epidural injections under fluoroscopic guidance does not absolutely prevent perforation of the dura by the needle tip, because the needle is usually advanced before the next bolus of dye is injected. Measurements of skin to epidural space in magnetic resonance imaging films2showed that the posterior epidural space at C6–C7 averages 3 mm in adults; in the case in question, it was not visible in the magnetic resonance imaging in figure 1 or in the computerized tomography scan in figure 3. There are two possible explanations. One is shown in figures 1 and 2 demonstrating that the patient had Chiari I syndrome usually accompanied by a narrow posterior cervical epidural and intrathecal compartments. The other is the C6–C7 space, where a herniated nucleus pulposus is still present, displacing the dural sac posteriorly. There is no posterior epidural space in either figure 1 or figure 3, as noted before.3 

As far as how the mass got there, if the steroid was injected epidurally, the substance loculated anteriorly where there was more room. Because the epidural space stops at the foramen magnum, it is likely that some of it went intrathecally through the previously made orifice, distributing through the subdural space above the clivus and other areas (figs. 1 and 2). However, 4 weeks is too soon to develop a granuloma, which was not seen at the time of surgery. Most likely, what the authors called collections  is more likely the “depo” vehicle of triamcinolone preparation. Interestingly enough, when this type of steroid is deposited epidurally, the steroid fraction is absorbed within 2 days into the circulation; it does not cross the dura, as long as it remains intact. The depo vehicle may stay in the epidural compartment for 2–6 weeks. Three doses of 60 mg triamcinolone given within 1 month may be responsible for the accumulation of this substance in the anterior cervical epidural space and the smaller fractions shown intracranially (even after the drainage of the anterior epidural mass). The so-called intracranial hypotension was leakage of cerebrospinal fluid through the hole made at the time of the last epidural steroid injection.

The hanging drop method is not an appropriate technique in the absence of cervical epidural space, although it can be distended if a solution is injected from below. There is no solid evidence that depositing the steroid medication precisely in the intervertebral space where pathologic findings have been reported produces better results than if injected one or two spaces away or, for that reason, if steroids are deposited paravertebrally. Cervical epidural steroid injection can be performed safely and effectively at C7–T1, where there is consistently a wider epidural space that can be reached in more than 85% of the patients with a 11/2-in-long needle2without danger of perforating the dura.

Without doubt, a “lightening bolt” sensation with radicular distribution, while the physician is looking for the epidural space, means paraesthesia4on one of the intrathecal nerve roots, because there are not nerve roots in the posterior epidural space. If there is a “wet tap,” the injection of steroids should be deleted because every steroid preparation available in the United States has preservatives and triamcinolone has polyethylene glycol and benzylic alcohol that may enter the subarachnoid space, initiating an inflammatory reaction in the arachnoid.4,5These are not urgent procedures, and the usual option of trying one space above is not applicable because the medication may pass through the previously made hole, as in this case. One hopes that the autologous blood and the fibrin, both well-known central nervous system irritants,6injected in the anterior epidural space will not produce arachnoiditis at the operative level. After all, it was neither a granuloma nor a case of primary intracranial hypotension.

Aldrete Pain Care Center, Inc., Birmingham, Alabama.

Dietrich CL, Smith CE: Epidural granuloma and intracranial hypotension resulting from cervical epidural steroid injection. Anesthesiology 2004; 100:445–7
Aldrete JA, Zapata JC, Ghaly RF: Skin to cervical epidural space distances as determined by MRI: Consideration of the “Hump Pad.” J Clin Anesth 1998; 10:309–13
Aldrete JA, Ghaly RF: Need for precise diagnosis prior to epidural steroids. Anesthesiology 2000; 93:565–6
Aldrete JA: Neurologic deficits and arachnoiditis following neuroaxial anesthesia. Acta Anaesthesiol Scand 2003; 47:3–12
Aldrete JA: Corticosteroids, Arachnoiditis: The Silent Epidemic. Edited by Aldrete JA. Denver, Futuremed, 2000, pp 125–34Corticosteroids,Aldrete JA
Aldrete JA: Corticosteroids, Blood in the CSF, Arachnoiditis: The Silent Epidemic. Edited by Aldrete JA. Denver, Futuremed, 2000, pp 125–34Blood in the CSF,Aldrete JA