FormalPara Key Summary Points

Why carry out this study?

The burden of herpes zoster complications is poorly studied, with limited epidemiological information, except for postherpetic neuralgia. This impairs the quantitative analysis of complications incidence to be used in health economic and outcomes studies.

A systematic literature review was conducted to summarize the incidence of herpes zoster complications in different populations between 2002 and 2022.

What was learned from the study?

Among 124 retrieved studies, most frequently reported complications were of neurological (n = 110), ocular (n = 48), and cutaneous (n = 38) origin.

Postherpetic neuralgia, ocular complications associated with herpes zoster ophthalmicus, and disseminated zoster were the most reported complications.

High incidence of herpes zoster complications was frequently reported in populations with immunocompromising conditions; however, intra-study comparisons between the general or immunocompetent population and immunocompromised subgroups were mostly not statistically tested.

Inter-study variability of outcomes was high as a result of varying study design, setting, and sample sizes.

Introduction

Herpes zoster (HZ), commonly referred to as shingles, is caused by reactivation of latent varicella-zoster virus (VZV). HZ frequently presents as a painful debilitating rash, including skin inflammation and blisters, and sometimes causes scarring and permanent pigment changes. Treatment of HZ with oral antiviral therapy is recommended within 72 h of rash onset, and usually given for 7 days in the absence of HZ complications. However, substantial pain and discomfort may be prolonged for weeks, months, or even years in severe cases, diminishing patient quality of life (QoL) [1].

The incidence of HZ ranges between 5.2 and 10.9 cases per 1000 person-years [2]. Anyone previously infected with VZV is at risk of develo** HZ. The frequency and severity of HZ episodes increase with age as a result of age-related decline in immunity [3, 4]. Among adults who reach 85 years of age, approximately half will likely have suffered at least one episode of HZ [5]. Besides age, people with immunosuppression from any cause, including hematologic malignancies, human immunodeficiency virus (HIV), and immunosuppressive medications, have a higher risk of HZ [6]. Other commonly identified risk factors include female sex, white race, and presence of comorbid conditions [7,8,9].

HZ can give rise to various complications, some of which are serious or even life-threatening. The most common HZ complication is postherpetic neuralgia (PHN), which is frequently defined as chronic pain persisting or appearing 3 months after initial rash detection or HZ diagnosis. It is associated with an average worst pain score ≥ 3 on the Likert scale, as measured by the Zoster Brief Pain Inventory (ZBPI) [10]. HZ and PHN have been shown to adversely affect healthy aging and QoL [11]. Other common HZ complications include ocular, neurological, and cutaneous conditions [12]. Some of these consequently cause severe health impact, such as Ramsay Hunt syndrome, which can lead to facial paralysis, or delayed contralateral hemiparesis [13]. Various other chronic infections are important in clinical practice [14,15,16,17,18,19,20].

In general, the incidence and burden of HZ complications other than PHN are poorly studied and reliable epidemiological information is limited. Consequently, the incidence and burden of these conditions in many regions are unknown. This limits reliable modelling of complications in health economic evaluations examining the public health impact of HZ on diverse subject groups.

The objective of this systematic literature review (SLR) was to identify and summarize the incidence of HZ complications in different populations.

Methods

Objectives

The main objective of the present SLR was to collect and consolidate information on the incidence of HZ complications in adults ≥ 18 years of age by complication type (and subtype) and for a wide range of populations generally classified as (a) general, (b) healthy (immunocompetent), and (c) with selected comorbidities/underlying conditions (including immunocompromising and/or immunosuppressive ones). The secondary objective was to detect differences in the incidence of complications by (a) age and (b) sex, if supported by available data.

Specific to the main objective, a list of primary HZ complications was developed (see Supplementary material, Appendix SI, Table SI-1); additional HZ complications were extracted as reported in the included body of literature.

Design

SLR design followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and the Cochrane Collaboration guidelines [21, 22]. All articles retrieved from the electronic databases were screened against predefined eligibility criteria (detailed in the Supplementary material, Appendix SI, Table SI-2) using a modified PICO (population, intervention, comparator, outcomes) framework. This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors. Due to this reason an ethics committee approval was not required.

Electronic Databases

The search was conducted on records published between January1, 2002 and October 20, 022 in Medline (accessed via PubMed), Embase, and Virtual Health Library, combining keywords (free-text terms) and Medical Subject Headings (MeSH terms) specific to each database. The start date was chosen in alignment with a recent SLR of HZ incidence worldwide [2] resulting into an extensive search process of manageable size. The exact search strings are documented in the Supplementary material, Appendix SII, Tables SII-1 to SII-3.

Study Selection, Data Extraction, and Quality Control

Letters to the editor, editorials or comments, case reports, and SLRs or meta-analyses were excluded from this review. Additional exclusion criteria encompassed small sample size (fewer than 100 HZ cases in the case of relative incidence, a population sample size with underlying disease lower than 100 in the case of cumulative incidence or incidence rate), animal studies, purely pediatric studies (< 18 years of age), studies conducted in non-representative populations (e.g., military personnel, pregnant women, etc.), phase I/II/III clinical trials, modelling studies, studies on the occurrence of HZ after coronavirus disease 2019 (COVID-19) vaccination or on the relation of HZ with COVID-19 infection, and studies reporting on the risk of another disease after HZ (such as stroke, dementia, coronary artery syndrome, or other malignancy). Pediatric zoster [23,24,25,26,27,28], with an incidence reported as declining from 74 to 38 per 100,000 person-years between 2003 and 2014 [29] consistent with increased participation in pediatric varicella vaccination programs, was explicitly not in scope of this review.

A three-step study selection procedure was implemented as detailed in the Supplementary material, Appendix SIII, along with further data extraction and quality control methodology.

Data extraction was guided by the common definitions outlined in the Supplementary material, Appendix SIV. Critical study quality aspects were documented in a data-extraction sheet, including representativeness of the population, definition of HZ cases and its complications, loss to follow-up, misclassification, calculation of incidence, etc. The details are reported in the Supplementary material, Appendix SV.

Results

Out of 3855 unique search hits, 124 articles were included in this review [30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,162]; the risk of develo** PHN was reported to be between 6% and 45% across the examined immunocompromising conditions (6–41% in HSCT, 7–45% in SOT, 6–40% in HM), and the risk of disseminated zoster in HSCT ranged from 0 to 32% with a median of 3%.

Limitations

Most studies (n = 83) used a retrospective design. Most retrospective database analyses (n = 53) utilized diagnostic codes to identify HZ and its complications, with the possibility of physician miscoding or undercoding. A few retrospective studies [42, 50, 92, 103] utilized diagnostic codes coupled with antiviral prescriptions, while some [31, 55, 57, 100, 118, 127, 141, 146] were based on clinical diagnoses not accompanied by coding. Two retrospective database analyses [84, 112] were based on self-reported outcomes. Several retrospective studies [61, 62, 64,65,66,67, 70, 72, 106, 107, 119, 121, 137, 147, 152] were performed on general practitioner (GP) or GP/outpatient/primary care databases, with the possibility of selection bias in the form of exclusion of older institutionalized subjects. Selection bias in the form of exclusion of unemployed/uninsured persons from a number of (n = 25) health insurance database analyses [30, 33, 35, 36, 41, 42, 44, 47, 52, 56, 58, 59, 73, 75, 77,78,79,80,81,82, 92, 101, 116, 117, 138] is also possible.

In addition, HZ case definition was frequently not available in the study report (n = 17), nor were specific inclusion (n = 15) or exclusion (n = 64) criteria. Several studies did not provide critical baseline characteristics such as participant age range (n = 31) and gender (n = 35). Most studies (n = 97) did not report on antiviral use for HZ episode treatment.

In populations with comorbidities, the specific conditions were not reported in several studies [44, 64, 95, 106, 111, 114, 116].

The majority of studies were conducted in Europe (n = 44), Asia (n = 40), and North America (n = 36). There were only two studies including data from South America [142, 143]. Multicountry studies [142, 143] did not stratify incidence of complications by country or geographical region.

All outcomes summarized in the present SLR should be placed in perspective under the various methodological and operational study limitations outlined above.

Conclusions

This systematic review synthesized the literature on incidence of HZ complications in adults  ≥18 years of age from 124 studies. Complication types were grouped as ocular, neurological, cutaneous, visceral, and miscellaneous other. Neurological complications were the most studied complication type (n = 110), while visceral complications were minimally reported (n = 6). PHN remains the most studied HZ complication.

The incidence of HZ complications was higher in older age groups (although frequently not statistically tested), and it appeared higher in immunocompromised populations compared to the immunocompetent (cell-mediated immunity and VZV reactivation are discussed elsewhere [163]). The relationship between incidence levels in the general population and the immunocompromised was less clear, which may be attributed to varying proportions of immunocompromised individuals in the respective general population samples. Furthermore, HZ complications in immunocompromised populations were reported in only a few studies and the immunocompromising disease or condition was not always clearly defined. In studies examining the burden of HZ by underlying condition/diagnosis, and across many complication types, high incidence of complications was observed in patients with HM, SOT, and HIV. In studies that stratified incidence of complication by gender (n = 41), the association between gender and incidence was unclear.

In general, the incidence of HZ complications could not be sufficiently summarized by age, or gender, as a result of insufficient numbers of studies that provided complete data on these variables. In terms of geographical representation, the present data collection would not be globally representative as most studies were conducted in Europe, Asia, and North America, while there were no studies from Africa.

In short, further research would be required to quantitatively analyze HZ complications incidence by age, gender, and/or location, for effective utilization of aggregated values in health economic and outcomes studies. The collected body of evidence from this work can serve as the starting point towards that end.