To the Editor:—

It was indeed reassuring to read the objective commentary of the application of epidural steroids by Abram, 1which provided a brief but explicit review of the subject. I would like to point out that symptoms of radiculopathy may be caused by tumors, infections, vascular malformations, neuropathy spinal stenosis, facet joint arthropathy, a bulging disc in a patient with short pedicles, and by other diagnoses in a nonoperated spine. Postlaminectomy radiculopathy may occur, for example, from periroot scarring, a loose fragment of the herniated disc, arachnoiditis, or osteophytes. Even if we know that there is a herniated nucleus pulopsus, unless we define that the herniation of the disc is central, paracentral, or lateral, or that it is 2, 3, or 4 mm and that it is effacing the dural sac or compressing the root, we do not know whether epidural steroids are indicated. We do not claim originality regarding this concern; in 1944, Rovenstine and Hershey 2stated that “Much discredit has come to nerve blocking from the too frequent practice of ‘trying a block’ without accurate diagnosis.”

The diagram shown in figure 1 of Abram’s article 1illustrates a lateral herniated nucleus pulposus compressing the root against the posterior wall of the laminae; this is one of the situations in which epidural steroids have less chance to succeed, and it is precisely a situation that will most likely necessitate surgery. Usually when there is a space-occupying lesion, the epidural space is reduced or absent, and therefore attempts to enter the space at the same level have greater chances for dural puncture.

Two preparations mentioned by Abram, 1Aristocort Forte (triamcinolone; ESI Lederle Generics, Philadelphia, PA) and Depo-Medrol (methylpredisolone; Pharmacia & Upjohn, Kalamazoo, MI), in multiple-dose vials (5 ml), contain benzylic alcohol, in addition to polyethyleme glycol, both of which have been shown to be neurotoxic and should therefore be avoided.

Abram 3noted no untoward effects from subarachnoid injections of methylprednisolone in 37 patients with sciatica. Kikuchi et al.  4successfully treated patients with persistent pain from postherpetic neuralgia with use of subarachnoid injections of methylprednisolone. Our experience with intrathecal methylprednisolone administered after removal of the supernatant fluid and dilution of the remnants with iced saline has continued to provide satisfactory analgesia, without evidence of neurotoxicity (as indicated by neurologic examination and annual magnetic resonance imaging studies). 5,6Abram et al.  7administered four intrathecal steroid injections into rats at 3-week intervals without finding histologic abnormalities postmortem. We believe that limited observations should be continued to explore further the indications, safety, and effectiveness of intrathecal steroids, because extradural disease cannot be treated by intrathecal steroids, 3and intrathecal disease cannot be treated by peridural steroids.

Abram SE: Treatment of lumbosacral radiculopathy with epidural steroids. A nesthesiology 1999; 91:1937–41
Rovenstine EA, Hershey SG: Therapeutic and diagnostic nerve blocking: A plan for organization. A nesthesiology 1944; 5:574–6
Abram SE: Subarachnoid corticosteroid injection following inadequate response to epidural steroids for sciatica. Anesth Analg 1978; 57:313–6
Kikuchi A, Kotani N, Sato T, Takamura K, Sakai I, Matsuki A: Long-term analgesia by intrathecal prednisolone acetate in intractable postherpetic neuralgia. Reg Anesth Pain Med 1999; 24:287–93
Aldrete JA, Sued JA, Aldrete VT: Applications of iced saline in pain management. Pain Digest 1998; 8:173–6
Aldrete JA: Intrathecal injections of methylprednisolone “sans preservative.” Reg Anesth Pain Med 1999; 24:67
Abram SE, Marsala M, Yaksh TL: Analgesic and neurotoxic effects of intrathecal corticosteroids in rats. A nesthesiology 1994; 81:1198–205