Treatment of herpes zoster

Antiviral medication

Most of the placebo-controlled RCTs of antiviral therapy were summarized in a meta-analysis.10 A subanalysis (4 studies11–14 comprising 692 patients) showed that acyclovir (800 mg 5 times daily for 7–10 days) had no statistically significant effect on acute pain after 1 month (pooled odds ratio 0.83, 95% confidence interval [CI] 0.58 to 1.21). Because the various studies used measurement methods that could not be compared, no overall effect on the duration of acute pain could be established. From the separate studies, however, it appeared that the effect of acyclovir on the duration of acute pain did reach statistical significance in a few instances. Pain relief was seen at most a few days earlier. Another placebo-controlled RCT with acyclovir, published after this meta-analysis, also did not demonstrate a statistically significant effect on pain reduction after 1 month.15

A double-blind placebo-controlled RCT with famciclovir (500 or 750 mg 3 times a day for 1 week) reported that the median length of time to the disappearance of acute pain did not differ significantly among the 3 groups.16 It only reported a statistically significant effect on pain in the subgroup of patients with more than 50 skin lesions given famciclovir in the lowest dose. The median length of time for pain to disappear was 20 days in this subgroup compared with 30 days in the placebo group (hazard ratio 1.9, 95% CI 1.3 to 3.0). Data from a double-blind acyclovir-controlled RCT showed that famciclovir administered once (750 mg) or twice (500 mg) daily was as effective as famciclovir (250 mg) 3 times daily, and as effective as acyclovir (800 mg) 5 times daily, in cutaneous healing and reduction of acute pain.17 A small double-blind acyclovir-controlled RCT showed that famciclovir (250 mg 3 times a day) was as effective as acyclovir (800 mg 5 times a day) for healing skin lesions and decreasing acute pain.18 Another study compared valacyclovir with acyclovir.19 The design of that study does not allow for conclusions on the reduction of acute pain.

In most individual studies the time necessary for the vesicles, ulcers, and crusts to disappear was shorter when antiviral medication was used than when placebo was used. The time gain was only 1 or 2 days. Antiviral medication might relieve acute pain and speed healing of skin lesions. These effects are only marginal, though, and thus have little clinical relevance.

All trials reported the same frequencies of side effects (eg, headache in 11% to 23% of patients and nausea in 12% to 16% of patients15,16,19) for both antiviral medication and placebo.

Corticosteroids

There is no thorough meta-analysis of studies on the effects of corticosteroids. One large RCT compared the effects of acyclovir with those of the combination acyclovir and prednisolone.20 The addition of prednisolone (40 mg/d for 3 weeks, tapering) to acyclovir resulted in a statistically significant reduction of pain during the first 2 weeks (average pain score, on a scale of 0 to 5, had diminished by 1.4 in the prednisolone group and by 1.0 in the non-prednisolone group on day 7; the reductions were 1.8 and 1.5, respectively, on day 14). No further statistically significant differences were found after 2 weeks. Moreover, a significantly higher percentage of the skin lesions had been cured in the prednisolone group (P = .02) on day 7 and day 14, although the duration of complete recovery did not differ significantly between the 2 groups. Another RCT compared acyclovir-prednisone, acyclovir-placebo, prednisone-placebo, and placebo-placebo combinations.15 Patients who received prednisone (60 mg/d for 3 weeks, tapering), whether or not in combination with acyclovir, had a 2.3 times (95% CI 1.4 to 3.5) greater chanceof being free of pain after 1 monththan patients who did not receive prednisone. Quicker healing of skin lesions was not observed. Corticosteroids, therefore, only somewhat relieve acute pain.

All studies reported side effects of corticosteroids. The most common were dyspepsia (5% in the prednisolone group, < 1% in the acyclovir group), nausea (2% and 1%, respectively), headache (1% and 0%, respectively), and edema (3% and 0%, respectively).20 The theoretically feared dissemination of the localized HZ was not observed.

Other therapeutic options

An open RCT assessed the effectiveness of a single epidural injection of corticosteroids (80 mg methylprednisolone) and local anesthetics (10 mg bupivacaine) in the acute phase of HZ for HZ-associated pain.21 One month after randomization, 48% of the patients in the intervention group reported HZ-associated pain versus 58% in the control group (relative risk 0.83, 95% CI 0.71 to 0.97). Moreover, intervention reduced the intensity of pain (median score on a 100-point visual analog scale 2 [25th–75th percentile 0 to 23] in the intervention group vs 6 [0 to 32] in the control group [P = .02]). The effect of the epidural injection, however, did not last beyond 1 month. Therefore, epidural injection should only be considered for patients with severe acute pain from HZ who are not responding to standard analgesic therapy.

No patient had serious adverse events related to the epidural injection.

Antiviral medication

There is no convincing evidence that acyclovir influences the incidence or duration of PHN. Three placebo-controlled RCTs on acyclovir (800 mg 5 times daily for 7 days13 and 10 days11,14) reported less pain in the treatment groups in the short-term (1–3 months), but no difference in the long-term (more than 3 months). Another study compared valacyclovir with acyclovir.19 Three groups were used: valacyclovir 1000 mg 3 times a day for 7 days, valacyclovir 1000 mg 3 times a day for 14 days, and acyclovir 800 mg 5 times daily for 7 days. The median length of time for pain to disappear was 38 and 44 days in the groups treated with valacyclovir for 7 or 14 days, respectively, and 51 days in the group treated with acyclovir. There was no statistically significant difference among the various groups with respect to PHN occurrence, although the postherpetic pain lasted longer among those in the acyclovir group than among those in the group treated with valacyclovir for 7 days (hazard ratio 1.3, 95% CI 1.0 to 1.6). Famciclovir also had no effect on PHN incidence. However, a well-designed placebo-controlled RCT did show that the duration of pain in those patients who developed PHN was notably shorter (63 days in the famciclovir group vs 119 days in the placebo group; 63 vs 163 days among patients > 50 years).16 After 6 months, pain was reported by 15% of the patients in the group treated with famciclovir compared with 23% of those in the placebo group. According to these data, 12 patients must be treated with famciclovir in order to gain 1 additional patient free of pain after 6 months.

Other therapeutic options

Corticosteroids have not been shown to influence the occurrence or duration of PHN.15,20,22 A small placebo-controlled RCT showed that treatment with amitriptyline (25 mg before bedtime for 90 days) during the acute stage of HZ reduced the risk of PHN by half.23 Further evidence, however, is lacking. The efficacy of cutaneous (eg, topical local anesthetics) and percutaneous (eg, sympathetic and epidural blocks) interventions on the prevention of PHN has not been established.21,24

Oral antiviral medication

Ophthalmic HZ is a potentially serious disease that can result in severe and lasting pain, particularly among elderly patients. Moreover, without antiviral treatment, about half of all patients will develop various eye disorders. Conjunctivitis, for example, is seen in nearly all HZ patients with ocular involvement. More severe disorders include keratitis, uveitis, and optic neuritis of the affected eye. If these latter disorders are not diagnosed and treated adequately, the patient’s sight might be permanently affected. Early (within 72 hours after rash onset) treatment with acyclovir (800 mg 5 times a day for 7 days) reduces the percentage of eye disorders in ophthalmic HZ patients from 50% to between 20% and 30% (eg, 25% of the patients in the acyclovir group developed stromal keratitis vs 56% in the control group; P = .008).25 Valacyclovir (1000 mg 3 times daily) and famciclovir (500 mg 3 times daily) seem to be as effective as acyclovir in reducing HZ-associated pain,26,27 but their efficacy in reducing eye disorders associated with HZ has not been studied. In clinical practice, however, these second-generation antiviral agents might be more effective than acyclovir because patients are more likely to comply with the treatment regimen (3 rather than 5 daily doses). The efficacy of antiviral medication initiated more than 72 hours after the onset of skin rash has never been confirmed. Nevertheless, prescription of these drugs after this time span should be considered among elderly patients with ophthalmic HZ, because their immune responses are usually delayed and, consequently, viral shedding can last longer (level III evidence).28 As the course of disease is unpredictable, prescription of oral antiviral drugs at the first signs of infection is recommended for all patients with ophthalmic HZ, irrespective of their age or the severity of symptoms.29

Antiviral eye ointment

Although the additional effectiveness of acyclovir eye ointment has never been established, topical acyclovir can be considered in cases of severe eye infection, as much higher concentrations of the drug in the anterior eye segment will be achieved by this route. Monotherapy with antiviral ointment is, however, insufficient and should be adjuvant to oral treatment.30

Other therapeutic interventions

Anesthesia-based interventions, such as stellate ganglion block,31 might produce short-term pain relief but are not done on a regular basis. In addition, no evidence exists for their effectiveness in reducing the risk of protracted pain.