Postherpetic neuralgia is the most frequent complication of herpes zoster. Treatment of this neuropathic pain syndrome is difficult and often disappointing. We assessed the effectiveness of a single epidural injection of steroids and local anaesthetics for prevention of postherpetic neuralgia in older patients with herpes zoster. We randomly assigned 598 patients older than 50 years, with acute herpes zoster (rash <7 days) below dermatome C6, to receive either standard therapy (oral antivirals and analgesics) or standard therapy with one additional epidural injection of 80 mg methylprednisolone acetate and 10 mg bupivacaine. The primary endpoint was the proportion of patients with zoster-associated pain 1 month after inclusion. Analyses were by intention-to-treat. This study is registered as an International Standard Randomised Controlled Trial, number ISRCTN32866390. At 1 month, 137 (48%) patients in the epidural group reported pain compared with 164 (58%) in the control group (relative risk [RR] 0.83, 95% CI 0.71-0.97, p=0.02). After 3 months these values were 58 (21%) and 63 (24%) respectively (0.89, 0.65-1.21, p=0.47) and, at 6 months, 39 (15%) and 44 (17%; 0.85, 0.57-1.13, p=0.43). We detected no subgroups in which the relative risk for pain 1 month after inclusion substantially differed from the overall estimate. No patient had major adverse events related to epidural injection. A single epidural injection of steroids and local anaesthetics in the acute phase of herpes zoster has a modest effect in reducing zoster-associated pain for 1 month. This treatment is not effective for prevention of long-term postherpetic neuralgia.