Abstract
Background
Our study aimed to determine the risk of herpes zoster reactivation and coronavirus disease 2019 (COVID-19) vaccination (mRNA vaccine [BNT162b2] and adenovirus-vectored vaccine [ChAdOx1 nCoV-19]).
Methods
This retrospective study analyzed herpes zoster cases diagnosed between 26 February 2021 and 30 June 2021 and registered in the National Health Insurance Service database. A matched case-control study with a 1:3 matching ratio and a propensity score matching (PSM) study with a 1:1 ratio of vaccinated and unvaccinated individuals were performed.
Results
In the matched case control analysis, BNT162b2 was associated with an increased risk of herpes zoster reactivation (first dose adjusted odds ratio [aOR], 1.11; 95% confidence interval [CI], 1.06–1.15; second dose aOR, 1.17; 95% CI, 1.12–1.23). PSM analysis revealed a statistically significant increase in risk within 18 days following any vaccination (adjusted hazard ratio [aHR], 1.09; 95% CI, 1.02–1.16). BNT162b2 was associated with an increased risk at 18 days postvaccination (aHR, 1.65; 95% CI, 1.35–2.02) and second dose (aHR, 1.10; 95% CI, 1.02–1.19). However, the risk did not increase in both analyses of ChAdOx1 vaccination.
Conclusions
mRNA COVID-19 vaccination possibly increases the risk of herpes zoster reactivation, and thus close follow-up for herpes zoster reactivation is required.
Keywords: COVID-19 vaccination, herpes zoster, mRNA vaccines
The risk of herpes zoster reactivation was increased after COVID-19 mRNA vaccination with statistical significance but not increased after adenovirus-vectored vaccination. Individuals administered COVID-19 mRNA vaccination should be closely monitored for herpes zoster reactivation.
Various vaccines have been developed since the coronavirus disease 2019 (COVID-19) pandemic started. Vaccine-related adverse events have also been reported globally. Herpes zoster reactivation reportedly related to COVID-19 infection or COVID-19 vaccination is a notable concern [1–4]. The national surveillance systems of United Kingdom and the United States have reported the relationship between herpes zoster and COVID-19 vaccine [5, 6]. There have also been many case reports of herpes zoster after vaccination, particularly with mRNA vaccines [7]. Large-scale analytical studies were also conducted to investigate the relationship between COVID-19 vaccination and herpes zoster, and they revealed statistically significant associations [5, 8–11]. However, a limited number of studies have been conducted in Asia. Moreover, considering the potential influence of multiple confounders on the occurrence of adverse reactions, further research is warranted.
Herpes zoster itself is often mild, but can cause complications in severe cases and has significant morbidity and cost [12]. Identifying the exact relationship between COVID-19 vaccine and adverse events could help in selecting the appropriate vaccine and in eliminating distrust about vaccine adverse events. Thus, this study aimed to determine the risk of herpes zoster reactivation in 2 types of COVID-19 vaccines (BNT162b2 and ChAdOx1) approved in South Korea.
METHODS
Study Design and Data Source
This retrospective study analyzed data from the National Health Insurance Service (NHIS). The NHIS database includes both NHI and Medical Aid data and contains personal information of each beneficiary [13]. However, the NHIS database does not contain information about vaccination. Therefore, information on COVID-19 vaccination was collected from the Korea Disease Control and Prevention Agency (KDCA) database and matched by personal registration number.
Case Definition
Herpes zoster patients were identified using the Korean Standard Classification of Diseases (KCD) codes B02 (herpes zoster) and G53.0 (postzoster neuralgia). The KCD codes are updated regularly based on the International Classification of Diseases-10 codes. Zoster cases recorded between 27 February 2021 and 30 June 2021 were included because COVID-19 vaccination with ChAdOx1 nCoV-19 and BNT162b2 commenced in South Korea on 26 February 2021 [14]. Herpes zoster cases diagnosed at the date of vaccination were censored. Only adult patients aged ≥19 years and prescribed antiviral drugs at least once within 6 days of initial diagnosis were enrolled in the zoster group. The antiviral drugs were classified as acyclovir, valacyclovir, and famciclovir. The prescription of antiviral drugs within the study period was identified using drug codes. Diagnosis of herpes zoster in the 3 years before the index day of the study period was considered as the washout period, and patients diagnosed during this period were excluded.
Age groups were stratified into 19–49, 50–59, 60–69, 70–79, and >80 years based on the increased risk of herpes zoster after age 50 years [15]. Underlying diseases and immunocompromised state were designated using KCD codes (Supplementary Table 1). Prescription of immunosuppressive agents was also determined (Supplementary Table 2). Individual history of COVID-19 vaccination was matched from KDCA data. Data on type of vaccine (ChAdOx1 or BNT162b2), dose of vaccination (first or second), and date of vaccination were collected. Individuals who received other vaccine types (Ad26.COV2.S or others) or were cross-vaccinated were excluded.
Statistical Analysis
Data were analyzed using matched case-control and propensity score matching (PSM). In the matched case-control study, confounding factors were recorded, and these were matched between the control (no herpes zoster) and the case (herpes zoster) according to the weight of confounding factors (matching variables) until the sample size of the control was as intended. The case group was categorized by the number and type of vaccination with newly diagnosed herpes zoster. The control group was initially defined as the whole population in the study period, and the same washout periods and exclusion criteria as for the case group were applied. The control group was matched by sex, age groups, underlying diseases, insurance type, and immunocompromised state until the sample size ratio reached 1:3. Conditional logistic regression analysis was performed to calculate the odds ratio of herpes zoster reactivation by variables, including vaccine type and number.
The PSM cohort study was designed to derive differences in the incidence of herpes zoster by comparing between the vaccine group and the matched nonvaccine group. Herpes zoster that occurred within 30 days of COVID-19 vaccination was regarded as a vaccine adverse reaction. Data for vaccinations administered between 26 February 2021 and 31 May 2021 were collected. After excluding people vaccinated in June 2021, the population was divided into the vaccine and nonvaccine groups. The same washout periods and exclusion criteria were applied in both groups. The remaining individuals were subjected to PSM by sex and age (perfectly matched), underlying diseases, insurance type, and immunocompromised state until the sample size ratio became 1:1. We used standardized difference to confirm the appropriateness of comparability between 2 groups after matching. The difference due to the covariates between the 2 groups were considered as negligible when the value was less than 0.1 [16]. Follow-up after the first dose vaccination was censored at the 30th day or at the date of second dose if the event occurred within 30 days. Multivariate Cox regression analysis was performed to adjust matching variables and estimate the adjusted hazard ratio (HR) of herpes zoster by vaccine type and doses.
Subgroup analysis by sex and age group was performed using both conditional logistic regression and Cox regression. When the analysis conducted for any specific vaccine type or number, people with another vaccination were excluded. The Z-test was used to confirm that several variables did not meet the assumption of proportional hazard in the PSM cohort study, and, thus, the extended Cox regression model was additionally conducted in such variables by dividing the period after vaccination into 18 days within and after. Dichotomous variables were compared and calculated using the χ2 test. Continuous variables were presented as the mean and standard deviation and compared using the Mann-Whitney U test. All statistical analyses were performed using SAS/STAT software, version 9.4 (SAS Institute Inc). Forest plots were generated by GraphPad Prism, version 9.5 (GraphPad Software). A P value of <.05 was considered statistically significant.
Ethics
This study was approved by the institutional review board of Korea University Guro Hospital (2021GR0304) and was conducted according to the tenets of the Declaration of Helsinki. The need for informed consent was waived owing to the retrospective nature of the study.
RESULTS
Participant Characteristics
In total, NHIS data for 45 038 541 adults from 27 February 2021 to 30 June 2021 were analyzed in the study. The pattern of vaccination showed that 9 453 405 (21.0%) and 935 901 (2.1%) adults received 1 and 2 doses of ChAdOx1, respectively (Supplementary Table 3). BNT162b2 vaccines were administered in 1 and 2 doses to 765 005 (1.7%) and 3 004 275 (6.7%) adults, respectively. The other participants received more than 1 dose of other COVID-19 vaccines (n = 1 333 237, 3.0%), were cross-vaccinated between ChAdOx1 and BNT162b2 (n = 409, <0.1%), or nonvaccinated (n = 29 546 309, 65.8%). The second dose of ChAdOx1 was administered 8–12 weeks after the first dose; the second dose of BNT162b2 was administered 3 weeks after the first dose.
Matched Case-Control Analysis
Overall, 400 523 participants were diagnosed with herpes zoster during the study period. Supplementary Figure 1 shows the participant selection flowchart. A total of 199 066 cases and 597 198 controls were included. There were more women than men (60.5% vs 39.5%). The analyses for specific vaccine types and doses were performed in the same manner (Supplementary Figures 2–5). Table 1 and Supplementary Tables 4 and 5 show the baseline characteristics of the case and control groups after 1:3 matching. The mean age in both groups was 55.0 years. The mean time lag from vaccination to herpes zoster (or matched date in control group) was not different between the case and control groups (28.3 days vs 28.1 days, P = .378), suggesting that they were well matched (Supplementary Table 6).
Table 1.
Case Control Analysis: Baseline Patient Characteristics of Case (Zoster) and Control (Nonzoster) Groups After 1:3 Matching
Characteristic | Case Group (n = 199 066) | Control Group (n = 597 198) | P Value |
---|---|---|---|
COVID-19 vaccination event | 24 708 (12.4) | 73 676 (12.3) | .378 |
Age group, y | |||
Mean ± SD | 55.0 ± 16.0 | 55.0 ± 16.0 | 1.000 |
19–29 | 15 024 (7.5) | 45 072 (7.5) | 1.000 |
30–39 | 21 626 (10.9) | 64 878 (10.9) | |
40–49 | 33 173 (16.7) | 99 519 (16.7) | |
50–59 | 45 854 (23.0) | 137 562 (23.0) | |
60–69 | 47 057 (23.6) | 141 171 (23.6) | |
70–79 | 24 404 (12.3) | 73 212 (12.3) | |
≥80 | 11 928 (6.0) | 35 784 (6.0) | |
Sex | |||
Male | 78 673 (39.5) | 236 019 (39.5) | 1.000 |
Female | 120 393 (60.5) | 361 179 (60.5) | |
Insurance type | |||
Medical Aid | 5949 (3.0) | 17 847 (3.0) | 1.000 |
National Health Insurance | 193 117 (97.0) | 579 351 (97.0) | |
Comorbid conditions | |||
Any comorbidity | 130 901 (65.8) | 392 703 (65.8) | 1.000 |
≥2 chronic conditions | 83 643 (42.0) | 247 357 (41.4) | <.001 |
Diabetes | 53 261 (26.8) | 159 783 (26.8) | 1.000 |
Hypertension | 67 332 (33.8) | 201 996 (33.8) | 1.000 |
Chronic heart disease | 31 600 (15.9) | 89 874 (15.0) | <.001 |
Chronic lung disease | 51 082 (25.7) | 148 568 (24.9) | <.001 |
Chronic liver/gastrointestinal disease | 18 787 (9.4) | 57 598 (9.6) | .007 |
Chronic kidney disease | 10 184 (5.1) | 29 607 (5.0) | .005 |
Chronic neurologic disease | 38 190 (19.2) | 113 038 (18.9) | .012 |
Rheumatologic disease | 15 977 (8.0) | 40 918 (6.9) | <.001 |
Metabolic disease | 3024 (1.5) | 8210 (1.4) | <.001 |
Thyroid disease | 26 938 (13.5) | 76 603 (12.8) | <.001 |
Hematologic disease | 581 (0.3) | 1391 (0.2) | <.001 |
Immunocompromised state | |||
Solid organ transplant | 395 (0.2) | 975 (0.2) | .001 |
Hematologic malignancy | 746 (0.4) | 1426 (0.2) | <.001 |
Solid cancer | 14 818 (7.4) | 45 098 (7.6) | .114 |
Immune deficiency/HIV | 85 (0.0) | 225 (0.0) | .325 |
Immunosuppressive drug | 4477 (2.2) | 6642 (1.1) | <.001 |
Data are presented as the No. (%) unless otherwise indicated.
Abbreviations: COVID-19, coronavirus disease 201;9 HIV, human immunodeficiency virus.
Conditional multiple logistic regression analysis to determine the adjusted odds ratio (aOR) of herpes zoster by vaccination dose, number, and other variables is shown in Table 2, Figure 1, and Supplementary Figure 11. The crude ORs of subgroup analyses are shown separately in Supplementary Table 7. Any vaccine (aOR, 1.02; 95% confidence interval [CI], 1.00–1.04) and ChAdOx1 vaccine (first dose aOR, 0.98; 95% CI, .96–1.01; second dose aOR, 0.93; 95% CI, .86–1.00) were not associated with herpes zoster infection. However, the first dose (aOR, 1.11; 95% CI, 1.06–1.15) and second dose (aOR, 1.17; 95% CI, 1.12–1.23) of BNT162b2 were associated with an increased risk of infection (Table 2).
Table 2.
Case Control Analysis: Adjusted Odds Ratio of Herpes Zoster Reactivation by Sex and Age Subgroups According to Vaccination Types and Doses Calculated by Multivariate Conditional Logistic Regression
Subgroup | Any Vaccinationa | P Value | ChAdOx1 First dosea |
P Value | ChAdOx1 Second dosea |
P Value | BNT162b2 First dosea |
P Value | BNT162b2 Second dosea |
P Value |
---|---|---|---|---|---|---|---|---|---|---|
Vaccination | 1.02 (1.00–1.04) | .128 | 0.98 (.96–1.01) | .161 | 0.93 (.86–1.00) | .049 | 1.11 (1.06–1.15) | <.001 | 1.17 (1.12–1.23) | <.001 |
Subgroup analysis by sex | ||||||||||
Male | 1.05 (1.01–1.08) | .016 | 1.00 (.96–1.05) | .872 | 0.94 (.81–1.10) | .442 | 1.13 (1.05–1.21) | .001 | 1.22 (1.13–1.33) | <.001 |
Female | 1.00 (.98–1.03) | .865 | 0.97 (.95–1.00) | .071 | 0.93 (.85–1.01) | .072 | 1.09 (1.04–1.15) | .001 | 1.15 (1.08–1.22) | <.001 |
Subgroup analysis by age group, y | ||||||||||
19–49 | 1.07 (1.03–1.12) | .001 | 1.09 (1.04–1.14) | <.001 | 1.10 (.98–1.23) | .116 | 1.05 (.92–1.19) | .488 | 1.00 (.84–1.19) | .995 |
50–59 | 1.00 (.96–1.05) | .858 | 0.99 (.95–1.04) | .802 | 1.06 (.94–1.21) | .335 | 1.06 (.88–1.28) | .539 | 0.86 (.69–1.08) | .196 |
60–69 | 0.94 (.90–.98) | .004 | 0.95 (.91–.99) | .015 | 0.58 (.48–.69) | <.001 | 0.67 (.53–.85) | .001 | 0.60 (.45–.79) | <.001 |
70–79 | 0.98 (.93–1.03) | .351 | 0.92 (.85–.99) | .025 | 0.76 (.47–1.22) | .250 | 1.01 (.94–1.08) | .818 | 1.16 (1.07–1.25) | <.001 |
≥80 | 1.13 (1.07–1.20) | <.001 | 0.69 (.61–.79) | <.001 | 0.64 (.43–.96) | .029 | 1.25 (1.17–1.33) | <.001 | 1.32 (1.23–1.42) | <.001 |
Data are presented as the odds ratio (95% confidence interval).
aAdjusted for variables including vaccination, comorbid conditions (chronic heart disease, chronic lung disease, chronic liver/gastrointestinal disease, chronic kidney disease, chronic neurologic disease, rheumatologic disease, metabolic disease, thyroid disease, and hematologic disease), immunocompromised state (hematologic malignancy, solid cancer, immune deficiency/HIV), and use of immunosuppressive drugs.
Figure 1.
Forest plot for the odds ratios and hazard ratios of herpes zoster in the 2 analytic methods. A, Adjusted odds ratios of vaccination and subgroup variables (age and sex) calculated by multivariate logistic regression in the matched case-control analysis. B, Adjusted hazard ratios of vaccination and subgroup variables (age and sex) calculated by multivariate Cox regression and extended Cox regression analysis in the propensity score matching analysis. Each variable is divided by the period after vaccination within or after 18 days when the assumption of proportional hazard is violated.
In subgroup analysis, male sex was more strongly associated with a high risk of herpes zoster (aOR, 1.05; 95% CI, 1.01–1.08), but both sexes were associated with a high risk of herpes zoster infection when the BNT162b2 vaccine was used. The 19–49 years (aOR, 1.07; 95% CI, 1.03–1.12) and >80 years (aOR, 1.13; 95% CI, 1.07–1.20) age groups showed a higher risk of zoster after any vaccination. Interestingly, the 19–49 years age group showed a higher OR after the first dose of ChAdOx1 (aOR, 1.09; 95% CI, 1.04–1.14), whereas the >80 years group showed a higher OR in the first (aOR, 1.25; 95% CI, 1.17–1.33) and second (aOR, 1.32; 95% CI, 1.23–1.42) doses of BNT162b2 (Table 2). Supplementary Table 8 shows the risks of herpes zoster by underlying diseases.
Propensity Score Matching Cohort Analysis
Among the 45 038 541 participants, we excluded those with vaccinations in June 2021 and divided the remaining participants into the COVID-19 vaccine group (n = 5 117 355) and the nonvaccine group (n = 29 559 420) (Supplementary Figure 6). After excluding those with other vaccinations, cross-vaccinations, and those in the washout period, the propensity score was derived from each participant, and score matching was performed until the ratio of vaccine and nonvaccine group reached 1:1. Finally, 2 993 584 participants remained in the vaccine and nonvaccine groups. Similar methods were performed for the analysis of specific vaccine types and doses (Supplementary Figures 7–10).
The baseline characteristics of the vaccine and nonvaccine groups before and after matching are shown in Table 3 and Supplementary Tables 9–12. The matched groups were perfectly matched for age and sex. Other variables showed similar proportions between the matched groups, despite differences in insurance type, comorbid conditions, and immunocompromised states. All standardized differences were under 0.1, suggesting that the differences by covariates were negligible. The mean age of the entire cohort was 46.1 years. There was a difference in mean time lag from vaccination (or matched date in nonvaccine group) to herpes zoster reactivation between the vaccine and nonvaccine groups (15.3 days vs 14.4 days, P = .049; Supplementary Table 13).
Table 3.
Propensity Score Matching Analysis: Baseline Characteristics of the Cohorts Before and After 1:1 Matching
Characteristic | Entire Cohort (n = 34 676 775) | Total Cohort Before Matching | Propensity Score Matched | Standardized Difference After Matching | ||||
---|---|---|---|---|---|---|---|---|
Vaccinated (n = 5 117 355) | Unvaccinated (n = 29 559 420) | P Value | Vaccinated (n = 2 993 584) | Unvaccinated (n = 2 993 584) | P Value | |||
Age group, y | ||||||||
Mean ± SD | 46.1 ± 16.8 | 65.8 ± 17.1 | 42.7 ± 14.2 | 58.5 ± 17.7 | 58.5 ± 17.7 | .000 | ||
19–29 | 6 787 764 (19.6) | 158 373 (3.1) | 6 629 391 (22.4) | <.001 | 153 919 (5.1) | 153 919 (5.1) | 1.000 | |
30–39 | 6 038 081 (17.4) | 366 473 (7.2) | 5 671 608 (19.2) | 350 623 (11.7) | 350 623 (11.7) | |||
40–49 | 7 692 621 (22.2) | 503 104 (9.8) | 7 189 517 (24.3) | 475 888 (15.9) | 475 888 (15.9) | |||
50–59 | 8 353 134 (24.1) | 603 473 (11.8) | 7 749 661 (26.2) | 554 673 (18.5) | 554 673 (18.5) | |||
60–69 | 2225.607 (6.4) | 872 187 (17.0) | 1 353 420 (4.6) | 623 752 (20.8) | 623 752 (20.8) | |||
70–79 | 1 911 408 (5.5) | 1 419 438 (27.7) | 491 970 (1.7) | 429 598 (14.4) | 429 598 (14.4) | |||
≥80 | 1 668 160 (4.8) | 1 194 307 (23.3) | 473 853 (1.6) | 405 131 (13.5) | 405 131 (13.5) | |||
Sex | .000 | |||||||
Male | 16 895 086 (48.7) | 2 176 220 (42.5) | 14 718 866 (49.8) | <.001 | 1 260 636 (42.1) | 1 260 636 (42.1) | 1.000 | |
Female | 17 781 689 (51.3) | 2 941 135 (57.5) | 14 840 554 (50.2) | 1 732 948 (57.9) | 1 732 948 (57.9) | |||
Insurance type | .027 | |||||||
Medical Aid | 997 838 (2.9) | 294 851 (5.8) | 702 987 (2.4) | <.001 | 182 814 (6.1) | 202 459 (6.8) | <.001 | |
National Health Insurance | 33 678 937 (97.1) | 4 822 504 (94.2) | 28 856 433 (97.6) | 2 810 770 (93.9) | 2 791 125 (93.2) | |||
Comorbid conditions | ||||||||
Any comorbidity | 14 140 875 (40.8) | 3 312 064 (64.7) | 10 828 811 (36.6) | <.001 | 1 761 057 (58.8) | 1 638 009 (54.7) | <.001 | .083 |
≥2 chronic conditions | 9 088 295 (26.2) | 2 920 731 (57.1) | 6 167 564 (20.9) | <.001 | 1 422 480 (47.5) | 1 299 289 (43.4) | <.001 | .083 |
Diabetes | 5 713 567 (16.5) | 1 843 207 (36.0) | 3 870 360 (13.1) | <.001 | 898 228 (30.0) | 815 403 (27.2) | <.001 | .061 |
Hypertension | 7 441 119 (21.5) | 2 686 680 (52.5) | 4 754 439 (16.1) | <.001 | 1 185 217 (39.6) | 1 143 938 (38.2) | <.001 | .028 |
Chronic heart disease | 3 230 750 (9.3) | 1 221 044 (23.9) | 2 009 706 (6.8) | <.001 | 563 048 (18.8) | 556 796 (18.6) | <.001 | .005 |
Chronic lung disease | 6 673 347 (19.2) | 1 419 581 (27.7) | 5 253 766 (17.8) | <.001 | 795 650 (26.6) | 692 405 (23.1) | <.001 | .080 |
Chronic liver/gastrointestinal disease | 2 353 774 (6.8) | 524 784 (10.3) | 1 828 990 (6.2) | <.001 | 344 747 (11.5) | 276.185 (9.2) | <.001 | .075 |
Chronic kidney disease | 1 128 036 (3.3) | 394 006 (7.7) | 734 030 (2.5) | <.001 | 226 255 (7.6) | 192 620 (6.4) | <.001 | .044 |
Chronic neurologic disease | 4 145 517 (12.0) | 1 706 174 (33.3) | 2 439 343 (8.3) | <.001 | 761 474 (25.4) | 749 622 (25.0) | <.001 | .009 |
Rheumatologic disease | 1 616 714 (4.7) | 388 107 (7.6) | 1 228 607 (4.2) | <.001 | 245 684 (8.2) | 178 249 (6.0) | <.001 | .088 |
Metabolic disease | 386 005 (1.1) | 73 862 (1.4) | 312 143 (1.1) | <.001 | 51 280 (1.7) | 43 141 (1.4) | <.001 | .022 |
Thyroid disease | 3 053 502 (8.8) | 643 935 (12.6) | 2 409 567 (8.2) | <.001 | 413 201 (13.8) | 336 182 (11.2) | <.001 | .078 |
Hematologic disease | 62 356 (0.2) | 14 271 (0.3) | 48 085 (0.2) | <.001 | 9911 (0.3) | 8880 (0.3) | <.001 | .006 |
Immunocompromised state | ||||||||
Solid organ transplant | 41 550 (0.1) | 11 792 (0.2) | 29 758 (0.1) | <.001 | 9314 (0.3) | 8420 (0.3) | <.001 | .006 |
Hematologic malignancy | 61 437 (0.2) | 11 135 (0.2) | 50 302 (0.2) | <.001 | 7427 (0.2) | 7583 (0.3) | .203 | .001 |
Solid cancer | 1 474 694 (4.3) | 420 156 (8.2) | 1 054 538 (3.6) | <.001 | 230 327 (7.7) | 223 260 (7.5) | <.001 | .009 |
Immune deficiency/HIV | 13 861 (0.0) | 1583 (0.0) | 12 278 (0.0) | <.001 | 1204 (0.0) | 874 (0.0) | <.001 | .006 |
Immunosuppressive drug | 306 512 (0.9) | 77 214 (1.5) | 229 295 (0.8) | <.001 | 52 586 (1.8) | 41 301 (1.4) | <.001 | .030 |
Data are presented as the No. (%).
Abbreviation: HIV, human immunodeficiency virus.
Table 4, Table 5, Figure 1, and Supplementary Tables 14 and 15 show the absolute risk difference, and crude and adjusted HRs (aHR) for herpes zoster after COVID-19 vaccination calculated by multivariate Cox regression analysis. The absolute risk difference increased at the first (30.9 cases/100 000) and second (27.3 cases/100 000) dose mRNA vaccination, and the vaccine doses required to generate 1 additional zoster case were 3236 doses and 3662 doses, respectively (Table 4). Many underlying diseases were associated with a high risk of herpes zoster development after any vaccination (Supplementary Table 15). Those conditions were matched for the multivariate Cox regression analysis. When the proportional hazard assumption was violated in such variables, extended Cox regression was performed by dividing the period after vaccination into within 18 days and 18 days or longer. The aHR was high with any vaccines after 18 days of vaccination (aHR, 1.09; 95% CI, 1.02–1.16), and 18 days after the first (aHR, 1.65; 95% CI, 1.35–2.02) and second (aHR, 1.10; 95% CI, 1.02–1.19) dose of BNT162b2. The HR was low within 18 days of any vaccination (aHR, 0.86; 95% CI, .82–.90) and first dose of ChAdOx1 (aHR, 0.80; 95% CI, .76–.85) (Table 5). In subgroup analysis, men had a high HR after 18 days of any vaccination (aHR, 1.21; 95% CI, 1.08–1.35), and both sexes showed a high HR after 18 days of the first BNT162b2 dose. The 19–49 years age group presented a high HR in all types and numbers of vaccines. The 50–59 years and >80 years groups also showed high HRs with BNT162b2 vaccination (Table 5).
Table 4.
Propensity Score Matching Analysis: Incidence, Relative Risk, and Absolute Risk Difference of Herpes Zoster Reactivation
Characteristic | Total No. | No. of Herpes Zoster Reactivations | P Value | Relative Risk | Absolute Risk Difference (Cases Per 100 000 People) | |
---|---|---|---|---|---|---|
Vaccinated | Unvaccinated | |||||
All | 2 993 584 | 5072 (0.17) | 5272 (0.18) | .049 | 0.96 | −6.7* |
ChAdOx1 first dose | 2 479 236 | 4149 (0.17) | 4709 (0.19) | <.001 | 0.88 | −2.3* |
ChAdOx1 second dose | 457 702 | 690 (0.15) | 736 (0.16) | .223 | 0.94 | −10.1 |
BNT162b2 first dose | 653 585 | 1073 (0.16) | 871 (0.13) | <.001 | 1.23 | 30.9* |
BNT162b2 second dose | 629 781 | 1353 (0.21) | 1181 (0.19) | .001 | 1.15 | 27.3* |
Age group, y | ||||||
19–29 | 153 919 | 109 (0.09) | 94 (0.16) | .292 | 1.16 | 0.5 |
30–39 | 350 623 | 427 (0.12) | 339 (0.10) | .002 | 1.26 | 2.9* |
40–49 | 475 888 | 628 (0.13) | 524 (0.11) | .002 | 1.20 | 3.5* |
50–59 | 554 673 | 921 (0.17) | 831 (0.15) | .031 | 1.11 | 3.0* |
60–69 | 623 752 | 1365 (0.22) | 1851 (0.30) | <.001 | 0.74 | −16.2* |
70–79 | 429 598 | 1024 (0.24) | 1118 (0.26) | .042 | 0.92 | −3.1* |
≥80 | 405 131 | 598 (0.15) | 515 (0.13) | .013 | 1.16 | 2.8* |
Sex | ||||||
Male | 1 260 636 | 1851 (0.15) | 1769 (0.14) | .173 | 1.05 | 2.7 |
Female | 1 732 948 | 3221 (0.19) | 3503 (0.20) | .001 | 0.92 | −9.4* |
Insurance type | ||||||
Medical Aid | 182 814 | 260 (0.14) | 286 (0.14) | .937 | 0.91 | −0.9 |
National Health Insurance | 2 810 770 | 4812 (0.17) | 4986 (0.18) | .035 | 0.97 | −5.8* |
Comorbid conditions | ||||||
Any comorbidity | 1 761 057 | 3391 (0.19) | 3489 (0.21) | <.001 | 0.97 | −3.3* |
≥2 chronic conditions | 1 422 480 | 2847 (0.20) | 2948 (0.23) | <.001 | 0.97 | −3.4* |
Diabetes | 898 228 | 1816 (0.20) | 1810 (0.22) | .005 | 1.00 | 0.2* |
Hypertension | 1 185 217 | 2243 (0.19) | 2545 (0.22) | <.001 | 0.88 | −10.1* |
Chronic heart disease | 563 048 | 1127 (0.20) | 1299 (0.23) | <.001 | 0.87 | −5.7* |
Chronic lung disease | 795 650 | 1683 (0.21) | 1508 (0.22) | .410 | 1.12 | 5.8 |
Chronic liver/gastrointestinal disease | 344 747 | 666 (0.19) | 626 (0.23) | .004 | 1.06 | 1.3* |
Chronic kidney disease | 226 255 | 459 (0.20) | 409 (0.21) | .502 | 1.12 | 1.7 |
Chronic neurologic disease | 761 474 | 1452 (0.19) | 1530 (0.20) | .063 | 0.95 | −2.6 |
Rheumatologic disease | 245 684 | 576 (0.23) | 526 (0.30) | <.001 | 1.10 | 1.7* |
Metabolic disease | 51 280 | 111 (0.22) | 91 (0.21) | .855 | 1.22 | 0.7 |
Thyroid disease | 413 201 | 956 (0.23) | 884 (0.26) | .006 | 1.08 | 2.4* |
Hematologic disease | 9911 | 13 (0.13) | 21 (0.24) | .090 | 0.62 | −0.3 |
Immunocompromised state | ||||||
Solid organ transplant | 9314 | 21 (0.23) | 20 (0.24) | .867 | 1.05 | 0.0 |
Hematologic malignancy | 7427 | 17 (0.23) | 32 (0.42) | .038 | 0.53 | −0.5* |
Solid cancer | 230 327 | 500 (0.22) | 548 (0.25) | .047 | 0.91 | −1.6* |
Immune deficiency/HIV | 1204 | 4 (0.33) | 1 (0.11) | .406 | 4.00 | 0.1 |
Immunosuppressive drug | 52 586 | 142 (0.27) | 135 (0.33) | .111 | 1.05 | 0.2 |
Data are presented as the No. (% for incidence risk among each group). Asterisks (*) indicate statistically significant values.
Abbreviation: HIV, human immunodeficiency virus.
Table 5.
Propensity Score Matching Analysis: Adjusted Hazard Ratio of Herpes Zoster Reactivation in Subgroup of Sex and Age by Vaccination Types and Doses Calculated by Multivariate Cox Regression Analysis
Any Vaccinationa | P Value | ChAdOx1 First Dosea |
P Value | ChAdOx1 Second Dosea |
P Value | BNT162b2 First Dosea |
P Value | BNT162b2 Second Dosea |
P Value | |
---|---|---|---|---|---|---|---|---|---|---|
Vaccination | 0.93 (.90–.97) | <.001 | 0.87 (.83–.90) | <.001 | 0.94 (.85–1.04) | .250 | 1.16 (1.06–1.27) | .002 | 1.10 (1.09–1.19) | .017 |
≤18 d of vaccinationb | 0.86 (.82–.90) | <.001 | 0.80 (.76–.85) | <.001 | … | … | 1.06 (.96–1.17) | .283 | … | … |
>18 d of vaccinationb | 1.09 (1.02–1.16) | .013 | 0.98 (.92–1.05) | .527 | … | … | 1.65 (1.35–2.02) | <.001 | … | … |
Subgroup analysis by sex | ||||||||||
Male | 1.02 (.95–1.09) | .586 | 0.95 (.89–1.02) | .173 | 0.99 (.78–1.25) | .928 | 1.18 (1.01–1.38) | .043 | 1.12 (.98–1.29) | .101 |
≤18 d of vaccinationb | 0.93 (.86–1.01) | .068 | 0.88 (.81–.97) | .007 | … | … | 1.02 (.86–1.22) | .811 | … | … |
>18 d of vaccinationb | 1.21 (1.08–1.35) | .001 | 1.07 (.96–1.20) | .220 | … | … | 2.13 (1.47–3.10) | <.001 | … | … |
Female | 0.89 (.85–.93) | <.001 | 0.82 (.78–.87) | <.001 | 0.92 (.82–1.04) | .174 | 1.14 (1.02–1.28) | .019 | 1.09 (.99–1.20) | .090 |
≤18 d of vaccinationb | 0.82 (.77–.87) | <.001 | 0.76 (.71–.81) | <.001 | … | … | 1.07 (.95–1.21) | .278 | … | … |
>18 d of vaccinationb | 1.03 (.95–1.11) | .521 | 0.93 (.86–1.01) | .098 | … | … | 1.46 (1.15–1.86) | .002 | … | … |
Subgroup analysis by age group, y | ||||||||||
19–49 | 1.22 (1.12–1.33) | <.001 | 1.17 (1.07–1.27) | .001 | 1.28 (1.08–1.51) | .004 | 1.34 (.92–1.95) | .124 | 1.47 (1.01–2.13) | .044 |
≤18 d of vaccinationb | … | … | … | … | … | … | 1.05 (.70–1.56) | .830 | … | … |
>18 d of vaccinationb | … | … | … | … | … | … | 15.03 (1.98–113.78) | .009 | … | … |
50–59 | 1.11 (1.01–1.22) | .033 | 1.02 (.93–1.12) | .639 | 1.10 (.92–1.32) | .277 | 1.69 (1.05–2.71) | .030 | 1.80 (1.13–2.86) | .014 |
60–69 | 0.73 (.68–.78) | <.001 | 0.72 (.68–.78) | <.001 | 0.47 (.38–.60) | <.001 | 1.29 (.69–2.39) | .427 | 0.74 (.46–1.20) | .218 |
≤18 d of vaccinationb | 0.62 (.56–.67) | <.001 | 0.67 (.61–.73) | <.001 | … | … | … | … | … | … |
>18 d of vaccinationb | 0.95 (.85–1.06) | .358 | 0.82 (.74–.92) | .001 | … | … | … | … | … | … |
70–79 | 0.85 (.77–.92) | <.001 | 0.76 (.69–.84) | <.001 | 0.31 (.07–1.49) | .144 | 0.94 (.80–1.11) | .457 | 0.78 (.67–.90) | .001 |
≤18 d of vaccinationb | 0.73 (.66–.81) | <.001 | … | … | … | … | 1.14 (1.00–1.31) | .056 | … | … |
>18 d of vaccinationb | 1.17 (1.01–1.37) | .043 | … | … | … | … | 1.74 (1.32–2.28) | <.001 | … | … |
≥80 | 1.09 (.97–1.23) | .140 | 0.61 (.49–.77) | <.001 | 0.21 (.04–1.06) | .059 | 1.24 (1.10–1.40) | .001 | 1.30 (1.17–1.44) | <.001 |
≤18 d of vaccinationb | … | … | … | … | … | … | 1.14 (1.00–1.31) | .056 | … | … |
>18 d of vaccinationb | … | … | … | … | … | … | 1.74 (1.32–2.28) | <.001 | … | … |
Data are presented as the hazard ratio (95% confidence interval).
aAdjusted for variables including vaccination, insurance type, comorbid conditions (chronic heart disease, chronic lung disease, chronic liver/gastrointestinal disease, chronic kidney disease, chronic neurologic disease, rheumatologic disease, metabolic disease, thyroid disease, and hematologic disease), immunocompromised state (hematologic malignancy, solid cancer, immune deficiency/HIV), and immunosuppressive drug use.
bExtended Cox regression analysis was done by dividing the period after vaccination into 18 days within and after when the proportional hazard assumption is violated.
DISCUSSION
Evidence on the risk of herpes zoster infection after COVID-19 vaccination has been conflicting. This study found that the mRNA vaccine was associated with a high risk of herpes zoster occurrence after vaccination. In the matched case-control analysis, the first and second doses of BNT162b2 increased the risk of herpes zoster after vaccination. Similarly, the PSM cohort analysis showed a high risk after the first and second dose of BNT162b2. However, the risk of herpes zoster did not increase in the adenovirus-vectored vaccine ChAdOx1. To our best knowledge, this study is the first to analyze large-scale nationwide data of both adenovirus-vectored and mRNA vaccination with millions of doses and to provide evidence of their related risk of herpes zoster reactivation.
Although there were differences in several variables between PSM cohort analysis and matched case-control analysis, the increased risk of herpes zoster in BNT162b2 was established in both study methods and this increase in risk was consistent with previous findings [5, 8, 11, 17, 18]. It is unclear why adenovirus-vectored vaccine had a lower risk of herpes zoster reactivation than mRNA vaccine. The results might be related to the differences in cytokine activation and T-cell immune responses among vaccines [19, 20].
In subgroup analysis by age, the risks of herpes zoster after COVID-19 vaccination varied. In particular, PSM analyses showed that the risks after 18 days were higher in all relevant variables. The cutoff of 18 days might reflect the time of vaccine response and delayed varicella-zoster virus (VZV)-associated T-cell dysfunction after vaccination. Interestingly, the risk of herpes zoster infection after the first and second dose of ChAdOx1 was increased only in the 19–49 years group. This could be partially explained by the fact that majority of young people were vaccinated with ChAdOx1 instead of BNT162b2 (first dose, 130 697 of ChAdOx1 vs 27 676 of BNT162b2 in the 19–49 years group), and the bias of health-seeking behavior might have influenced the results. This group had high interest in the vaccine side effects and might be more conscious of the occurrence of herpes zoster after vaccination and seek hospital treatment. However, there is also a possibility that this result is the effect of the ChAdOx1 vaccine itself. Thus, additional research to investigate the effect of adenovirus-vectored vaccines in young and middle-aged individuals is necessary.
The risk of zoster reactivation was lower in the 60–69 years age group. The risk reduction was pronounced in the first dose of the ChAdOx1 vaccine among those aged >60 years. This trend might be associated with the protective effect of zoster vaccine in the elderly group. In Korea, although not included in the national immunization program, zoster vaccine is recommended for those aged >60 years. Nevertheless, the high risk of zoster infection in mRNA vaccination is considered to be due to the overcoming of these effects.
The mechanism by which herpes zoster reactivation occurs with COVID-19 vaccination is yet to be clarified. However, given that T lymphocytes control VZV in the latent phase, T-cell dysfunction caused by vaccination might be involved in this occurrence. It is known that COVID-19 itself induces lymphopenia and lymphocyte exhaustion that decreases the function of CD4+, CD8+ and T reg cells [3]. COVID-19 vaccination can also induce robust T-cell response and enhance proinflammatory action profile on potent antigen-presenting cells via the spike protein, which might cause VZV reactivation [21]. Moreover, a large shift of naive CD8+ cells to create specific CD8+ cells might temporarily disable the control of VZV by VZV-specific CD8+ cells [22]. Additionally, changes in Toll-like receptor signaling due to vaccination or the effects of cytokines such as type I interferon might be involved [23, 24]. Further research is needed to establish the exact effect of COVID-19 vaccines with various platforms and doses.
Our study has some limitations. First, the analysis simply depended on the disease code, and there is a concern of overestimation of herpes zoster. However, we introduced a 3-year washout period and corrected the diagnosis by using a history of antiviral agents to reduce inaccuracies. Selection bias is also minimized as all Korean citizens are obligated to register in the NHI. Second, there might be a bias in health-seeking behavior among the participants. Specifically, vaccinated people might have taken better care of their health and well-being. This could act as a bias for both case-control and PSM cohort analyses. Third, several variables, including history of herpes zoster vaccination in the elderly and zoster reactivation before the 3-year washout period, are not reflected in the analysis. Medical procedures and drugs that were not covered by insurance were also not included in the analysis. Fourth, the severity of herpes zoster infection, hospitalization, neurologic sequalae other than postherpetic neuralgia and mortality could not be identified in the study. A proportion of cases of zoster reactivation might have occurred in the form of Bell's palsy, acute neuritis, or zoster sine herpete, which were not included in the zoster diagnosis code. Fifth, the result of P value and statistical significances might be misinterpreted in the case of large-data analysis. The significance of 95% confidence intervals and P values was due to the large sample size; thus, it is difficult to assert a true causal relationship between vaccines and adverse events. Actually, the absolute risk increase is very small and the occurrence of 1 additional zoster case required more than 3000 doses of mRNA vaccination. Future research into pathophysiologic mechanism and prospective clinical trials would help eliminate the limitations.
In conclusion, COVID-19 mRNA vaccination might be associated with a risk of herpes zoster reactivation. Thus, individuals, especially the elderly, should be carefully followed up for herpes zoster reactivation after mRNA vaccination.
Supplementary Data
Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Supplementary Material
Contributor Information
Jin Gu Yoon, Division of Infectious Diseases, Department of Internal Medicine, Guro Hospital, Korea University College of Medicine, Seoul, South Korea.
Young-Eun Kim, Big Data Department, National Health Insurance Service, Wonju, South Korea.
Min Joo Choi, Department of Internal Medicine, International St Mary's Hospital, Catholic Kwandong University College of Medicine, Incheon, South Korea.
Won Suk Choi, Division of Infectious Diseases, Department of Internal Medicine, Ansan Hospital, Korea University College of Medicine, Ansan, South Korea.
Yu Bin Seo, Division of Infectious Diseases, Department of Internal Medicine, Kangnam Sacred Heart Hospital, Hallym University College of Medicine, Seoul, South Korea.
Jaehun Jung, Artificial Intelligence and Big-Data Convergence Center, Gachon University College of Medicine, Incheon, South Korea.
Hak-Jun Hyun, Division of Infectious Diseases, Department of Internal Medicine, Guro Hospital, Korea University College of Medicine, Seoul, South Korea.
Hye Seong, Division of Infectious Diseases, Department of Internal Medicine, Guro Hospital, Korea University College of Medicine, Seoul, South Korea.
Eliel Nham, Division of Infectious Diseases, Department of Internal Medicine, Guro Hospital, Korea University College of Medicine, Seoul, South Korea.
Ji Yun Noh, Division of Infectious Diseases, Department of Internal Medicine, Guro Hospital, Korea University College of Medicine, Seoul, South Korea.
Joon Young Song, Division of Infectious Diseases, Department of Internal Medicine, Guro Hospital, Korea University College of Medicine, Seoul, South Korea.
Woo Joo Kim, Division of Infectious Diseases, Department of Internal Medicine, Guro Hospital, Korea University College of Medicine, Seoul, South Korea.
Dong Wook Kim, Department of Information and Statistics, Department of Bio and Medical Big Data, Research Institute of Natural Science, Gyeongsang National University, Jinju, South Korea.
Hee Jin Cheong, Division of Infectious Diseases, Department of Internal Medicine, Guro Hospital, Korea University College of Medicine, Seoul, South Korea.
Notes
Author contributions. H. J. C., D. W. K., J. Y. S., and J. J. conceptualized the study. Y. E. K., D. W. K., and J. J. acquired, linked, and managed the data and analyzed the results. All authors contributed to the analysis and interpretation of the results and accessed to the study process. J. G. Y. and Y. E. K. wrote the first draft and revised the manuscript with all other authors until the final version.
Acknowledgment . This study used the database of the Korea Disease Control and Prevention Agency and the National Health Insurance Service of Korea for policy and academic research. The research number of this study is KDCA-NHIS-2022-1-448.
Data sharing. The data are not publicly available in this study.
Financial support. This work was supported by the Korea Disease Control and Prevention Agency (grant number 20220313E05-0).
References
- 1. Cho SI, Lee DH, Park YM. Identification of herpes zoster high-risk group using Charlson comorbidity index: a nationwide retrospective cohort study. J Dermatol 2020; 47:47–53. [DOI] [PubMed] [Google Scholar]
- 2. Marra F, Parhar K, Huang B, Vadlamudi N. Risk factors for herpes zoster infection: a meta-analysis. Open Forum Infect Dis 2020; 7:ofaa005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Diez-Domingo J, Parikh R, Bhavsar AB, Cisneros E, McCormick N, Lecrenier N. Can COVID-19 increase the risk of herpes zoster? A narrative review. Dermatol Ther (Heidelb) 2021; 11:1119–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Katsikas Triantafyllidis K, Giannos P, Mian IT, Kyrtsonis G, Kechagias KS. Varicella zoster virus reactivation following COVID-19 vaccination: a systematic review of case reports. Vaccines (Basel) 2021; 9:1013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Gringeri M, Battini V, Cammarata G, et al. Herpes zoster and simplex reactivation following COVID-19 vaccination: new insights from a vaccine adverse event reporting system (VAERS) database analysis. Expert Rev Vaccines 2022; 21:675–84. [DOI] [PubMed] [Google Scholar]
- 6. Medicines and Healthcare Products Regulatory Agency . Coronavirus vaccine - summary of Yellow Card reporting. Available at: https://www.gov.uk/government/publications/coronavirus-covid-19-vaccine-adverse-reactions/coronavirus-vaccine-summary-of-yellow-card-reporting. Accessed 4 July 2023.
- 7. Iwanaga J, Fukuoka H, Fukuoka N, Yutori H, Ibaragi S, Tubbs RS. A narrative review and clinical anatomy of herpes zoster infection following COVID-19 vaccination. Clin Anat 2022; 35:45–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Barda N, Dagan N, Ben-Shlomo Y, et al. Safety of the BNT162b2 mRNA COVID-19 vaccine in a nationwide setting. N Engl J Med 2021; 385:1078–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Birabaharan M, Kaelber DC, Karris MY. Risk of herpes zoster reactivation after messenger RNA COVID-19 vaccination: a cohort study. J Am Acad Dermatol 2021; 87:649–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Shasha D, Bareket R, Sikron FH, et al. Real-world safety data for the Pfizer BNT162b2 SARS-CoV-2 vaccine: historical cohort study. Clin Microbiol Infect 2022; 28:130–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Wan EYF, Chui CSL, Wang Y, et al. Herpes zoster related hospitalization after inactivated (CoronaVac) and mRNA (BNT162b2) SARS-CoV-2 vaccination: a self-controlled case series and nested case-control study. Lancet Reg Health West Pac 2022; 21:100393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Pan CX, Lee MS, Nambudiri VE. Global herpes zoster incidence, burden of disease, and vaccine availability: a narrative review. Ther Adv Vaccines Immunother 2022; 10:25151355221084535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Bahk J, Kang HY, Khang YH. Age- and cause-specific contributions to the life expectancy gap between Medical Aid recipients and National Health Insurance beneficiaries in Korea, 2008–2017. PLoS One 2020; 15:e0241755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Oh HK, Kim EK, Hwang I, et al. COVID-19 vaccine safety monitoring in the Republic of Korea: February 26, 2021 to April 30, 2021. Osong Public Health Res Perspect 2021; 12:264–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Patil A, Goldust M, Wollina U. Herpes zoster: a review of clinical manifestations and management. Viruses 2022; 14:192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009; 28:3083–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Hertel M, Heiland M, Nahles S, et al. Real-world evidence from over one million COVID-19 vaccinations is consistent with reactivation of the varicella-zoster virus. J Eur Acad Dermatol Venereol 2022; 36:1342–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Chu CW, Jiesisibieke ZL, Yang YP, Wu PC, Lin HL, Tung TH. Association of COVID-19 vaccination with herpes zoster: a systematic review and meta-analysis. Expert Rev Vaccines 2022; 21:601–8. [DOI] [PubMed] [Google Scholar]
- 19. Sadarangani M, Marchant A, Kollmann TR. Immunological mechanisms of vaccine-induced protection against COVID-19 in humans. Nat Rev Immunol 2021; 21:475–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Wei J, Pouwels KB, Stoesser N, et al. Antibody responses and correlates of protection in the general population after two doses of the ChAdOx1 or BNT162b2 vaccines. Nat Med 2022; 28:1072–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Trougakos IP, Terpos E, Alexopoulos H, et al. Adverse effects of COVID-19 mRNA vaccines: the spike hypothesis. Trends Mol Med 2022; 28:542–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Desai HD, Sharma K, Shah A, et al. Can SARS-CoV-2 vaccine increase the risk of reactivation of varicella zoster? A systematic review. J Cosmet Dermatol 2021; 20:3350–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. West JA, Gregory SM, Damania B. Toll-like receptor sensing of human herpesvirus infection. Front Cell Infect Microbiol 2012; 2:122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Psichogiou M, Samarkos M, Mikos N, Hatzakis A. Reactivation of varicella zoster virus after vaccination for SARS-CoV-2. Vaccines (Basel) 2021; 9:572. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.