Pain is usually the predominant symptom in all phases of HZ disease, being reported by up to 90% of patients 28 . The intensity of pain in patients with HZ and PHN is assessed using subjective self-reporting. Patients are asked to rate their level of HZ/PHN pain (for on average or at its worst) on a numerical rating scale from 0 (no pain) to 10 (pain as bad as you can imagine).

Several typical symptoms of HZ can be debilitating for patients, even among those who do not experience clinically meaningful pain 43 . The main characteristic of the acute phase of HZ is the rash which, together with subsequent scarring, may be unsightly 28 .

HZ can also give rise to non-pain complications, many of which can increase the risk of permanent or long-standing physical impairment 20 . Complications can be ophthalmic (for example, herpes zoster ophthalmicus), neurological (for example, cranial and peripheral nerve palsies), dermatological (for example, bacterial super-infection) or visceral (for example, pneumonia; Table 2) 20 . A degree of motor deficit is common in patients with HZ; severe and long-lasting paresis may rarely occur if HZ affects the cervical and lumbosacral dermatomes 44 . HZ patients may also be at greater risk of stroke. Two recent retrospective population-based studies in Taiwan found that HZ and herpes zoster ophthalmicus patients had a 1.31-fold and 4.52-fold higher risk of stroke, respectively, in the following year 45 46 . Almost one in 10 immunocompetent patients with HZ is affected by at least one non-pain complication 47 . The incidence of complications becomes more common with increasing age: patients aged >65 years have four-times the number of complications than do individuals younger than 35 years 48 .

Complications associated with HZ and PHN place a heavy burden on individuals, who may suffer a permanent reduction in QoL even if the original HZ symptoms fully resolve 24 .

A survey by Weinke et al. was designed to capture the full course of HZ episodes from rash onset to pain resolution. The findings demonstrate the major and wide-ranging impact that HZ, and in particular PHN, have on patients' QoL 49 . Eleven thousand interviews were conducted in Germany to evaluate patient-rated pain in historic HZ episodes and interference with ADL. Screening questions identified 280 patients ≥50 years old with painful HZ diagnosed during the previous 5 years, of whom 32 patients developed PHN. Patients with PHN scored significantly worse across all outcomes related to pain and QoL than those with HZ 49 . The mean interference of pain on different aspects of QoL was highest for sleep, mood and work 49 . Three-quarters of patients experienced problems in carrying out daily activities, including work, studies, housework, family and leisure activities 49 . Approximately 60% of employed interviewees had to stop work at some point during the active disease period 49 .

When patients' partners were asked how HZ or PHN had affected patients' QoL, most replied that the symptoms interfered considerably with patients' ADL. This was particularly evident for people who developed PHN 50 . Not only patients are affected by this disease: caring for patients with chronic diseases places a burden on their partners' personal lives and finances 51 . Patients' partners and family members may need to fill additional roles. Caring for a person with HZ or PHN may cause family members to take time off work 52 . Patients and their partners are less likely to attend or participate in social occasions 24 51 .

HZ-related hospitalizations may contribute to the burden of HZ and PHN on patients and their families 53 . It is estimated that, in Europe, approximately 12,000 hospitalizations each year are due to HZ and its complications 38 48 54 55 56 . HZ complications which are likely to lead to hospitalization are PHN, encephalitis or meningitis and ocular complications 57 . Of 88,650 estimated annual cases of HZ in individuals ≥60 years old in England and Wales, an estimated 1750 (2%) hospitalizations are due primarily to HZ 53 . In Connecticut, USA, the mean annual hospitalization rate due to principal or secondary diagnosis of HZ over a 10-year period is 16.1/100,000 (range, 14.5-18.2) 58 . Almost 67% of these patients were aged >64 years.

The impact of acute HZ disease on patients' QoL is considerable 36 , equalling that seen with common debilitating chronic conditions such as congestive heart failure, diabetes mellitus, myocardial infarction and clinical depression 59 (Table 3). HZ patients have an extremely low score for role limitations due to physical problems (scoring 19 out of 100, Table 3). They cannot accomplish daily duties or fulfil their roles as well as they would normally expect because of the acute effects of HZ. In patients with PHN, pain ranks very high compared with pain from other chronic conditions such as atypical facial pain, osteoarthritis and rheumatoid arthritis 30 .

QoL analysis has been undertaken using data from a general-population group (including healthy people and those with a wide range of chronic conditions such as arthritis, chronic lung disease, congestive heart failure, diabetes, myocardial infarction or angina and hypertension; Leplège, unpublished data). Compared with the mean QoL level for each domain in the general population, patients suffering from HZ have poorer QoL scores (Figure 3) 38 . During the acute phase of HZ, the impact is clinically meaningful for some domains. For patients suffering from complications or PHN, the impact is higher and all the domains are affected (Figure 3) 38 .

Treatment of patients with HZ aims to accelerate rash healing, relieve pain and reduce complications 5 . Current treatments are effective in some patients, but therapies may show limited efficacy and poor tolerability in others (particularly in the elderly; Table 4).

Despite the development of treatment guidelines by the International Herpes Management Forum 60 , HZ management is complex 61 . The guidelines recommend oral antiviral agents for 7 days in patients with HZ who are at risk of developing PHN (patients aged >50 years with severe acute pain, severe rash, or significant prodromal symptoms) 60 . Studies have shown that aciclovir, valaciclovir and famciclovir decrease viral shedding, relieve acute phase pain, reduce the formation of new lesions and accelerate healing 5 62 63 . In general, these treatments are well tolerated, even by older patients 37 . By reducing viral replication, they may decrease the duration of acute-phase pain and PHN 4 5 .

To be effective, antiviral agents must be administered within 72 h of acute-symptom onset 60 , which may be difficult because of delays in patients seeking medical advice and delays in reaching diagnosis (for example, due to presentation of unusual symptoms) 8 . Often, viral activity and neuronal damage have been ongoing for several days before appropriate treatment is given 4 . Even with prompt treatment, antiviral treatment does not prevent all cases of PHN 37 64 . A recent Cochrane Review demonstrated that oral acyclovir does not significantly reduce PHN incidence 64 . The Cochrane review highlighted that there is insufficient evidence from randomized controlled trials to determine if other antiviral agents prevent PHN and if these treatments have a beneficial impact on QoL 64 . Twenty studies of antiviral treatment for acute HZ were identified, but 14 were excluded from the Cochrane review, mainly because of limitations with the study design.

A recent Canadian study of patients with HZ rash or pain showed that pain resolution is associated with recovery of QoL to a pre-morbid state (personal communication). Data showed that the median duration of pain was equal to the median duration of QoL impairment. Once established, the pain associated with HZ or PHN is difficult to treat 65 . PHN remains largely refractory to pharmacological treatments and prevention strategies 66 . Data show that >50% of patients require more than one prescription drug for PHN 13 . Despite treatment, symptomatic relief is obtained in only 50% of patients 67 . Patients with PHN experience significantly less pain relief than those with HZ, despite receiving more pain medications (mean, 2.2 versus 1.6) 49 . Studies have assessed the 'number needed to treat' for existing PHN therapies (how many patients need to be treated n order to achieve at least 50% pain relief in one patient compared with the controls). Data highlight the limited pain relief obtained in PHN patients. For some standard treatments, approximately four patients with PHN need to be treated in order to achieve pain relief in one patient (Figure 4) 68 .

As patients with PHN may find it difficult to tolerate pain therapies, drug doses are often titrated. This may result in several weeks of suboptimal treatment, during which it is unclear if (or when) the drugs will become effective 67 . Therapeutic compliance is also difficult to achieve. Almost 50% of patients with PHN do not discuss their symptoms regularly with physicians 13 . One in 10 is troubled substantially by the side-effects of PHN treatments 13 24 .

Treatment dissatisfaction is high in patients with PHN. In one study, only 14% of patients aged >65 years were satisfied with pain medication for PHN 13 . In the survey by Weinke et al. 49 , patient-reported satisfaction with HZ treatments was significantly lower in patients with PHN versus those with HZ (6.8 versus 8.3; scale from 0, 'not satisfied' to 10, 'very satisfied').

Current treatment strategies for HZ and PHN are only partially effective, so reducing the burden of HZ disease remains an area of considerable need 11 . A solution is for proactive strategies to focus on HZ prevention 67 . At least one study has shown that, compared with people who have not experienced HZ, those who have had HZ or PHN place higher value on preventing this disease. In this study, both groups questioned (community members and patients with HZ or PHN) said that they would be willing to pay substantial amounts of money to prevent HZ 69 .

The clinical rationale for the development of an HZ vaccine has centred on reducing the severity of HZ and PHN. Studies have shown that the live attenuated Oka/Merck VZV vaccine ('zoster vaccine') also reduced the incidence of HZ and PHN in addition to burden of illness (a severity-by-duration measure of the total pain and discomfort associated with a disease). Evidence that the zoster vaccine reduces HZ burden in an ageing population is derived from the Shingles Prevention Study 70 . The Shingles Prevention Study had a randomized, double-blind, placebo-controlled design and assessed HZ burden of illness and PHN incidence (pain rated as ≥3 on a 0-10 scale ≥90 days after rash onset) in >38,000 people aged ≥60 years who received zoster vaccine or placebo. Compared with placebo, vaccination significantly reduced HZ burden of illness by 61.1%. Vaccine recipients had a significantly reduced incidence of HZ (-51%) and PHN (-67%) compared with placebo 70 71 . HZ symptoms in the vaccine group were, in general, milder and of shorter duration than those in the placebo group 70 71 . Vaccination reduced PHN incidence defined using alternative cut-off times for the duration (persistence) of pain; the reduction in incidence was 72.9% at 180 days (Figure 5) 70 71 . Although the reduction in HZ incidence with vaccine compared with placebo was less apparent in patients who were aged ≥70 years versus those aged 60-69 years, the reductions in HZ burden of illness and PHN were preserved in the older population 70 .

In an adverse event substudy of the Shingles Prevention Study, rates of serious adverse events were higher in the vaccine group (1.9%) than in the placebo group (1.3%) 70 . Injection site reactions were more frequent among vaccine recipients than those taking placebo (48.3% versus 16.6%, respectively), but were generally mild.

The impact of vaccination on QoL was assessed in the entire study population of the Shingles Prevention Study using ZBPI and ZIQ questionnaires. Mean ZBPI and ZIQ scores, which rated the 'interference' of HZ and PHN with ADL, were used to calculate a 'severity of interference score'. For the entire study population, zoster vaccination reduced the ZBPI severity of interference by 66% (95% confidence interval; CI: 55, 74) and the ZIQ severity of interference by 68% (95% CI: 57, 77). In individuals who developed HZ, vaccination reduced the ZBPI severity of interference by 31% (95% CI: 12, 51) and the ZIQ severity of interference by 35% (95% CI: 13, 57). Zoster vaccination therefore reduced the burden of HZ-related interference with ADL by about two-thirds in a population of older adults and by about one-third in vaccine recipients who developed disease 72 . Much of the effectiveness of the zoster vaccination was due to the prevention of HZ episodes 70 71 72 73 .