Abstract
Background
This study aimed to investigate the impact of new-onset herpes zoster (HZ) on the risk of cardiorenal outcomes in patients with type 2 diabetes mellitus (T2DM) and to evaluate whether subsequent initiation of sodium-glucose cotransporter-2 (SGLT2) inhibitors modified this risk.
Methods
Patients with T2DM were identified from a territory-wide cohort from 2016 to 2021. The primary outcomes were primary renal composite outcomes (defined as a composite of end-stage renal failure, a 50% decline in estimated glomerular filtration rate [eGFR], or all-cause mortality) and primary cardiovascular composite outcomes (defined as a composite of heart failure hospitalization, myocardial infarction, stroke or all-cause mortality).
Results
New-onset HZ occurred in 3,306/171,474 patients (incidence rate 5.70/1,000 person-years) and was independently associated with higher risks of the primary renal composite outcomes (adjusted HR 1.42, 95% CI 1.30–1.55) and primary cardiovascular composite outcomes (adjusted HR 1.59, 95% CI 1.42–1.77) in patients with T2DM. This association was more pronounced within the first three months following infection and attenuated thereafter. Among patients who developed HZ, the initiation of SGLT2 inhibitors was associated with a lower risk of adverse renal (HR 0.57, 95% CI 0.42–0.78) and cardiovascular (HR 0.54, 95% CI 0.32–0.91) outcomes.
Conclusions
New-onset HZ was associated with increased cardiorenal risk among patients with T2DM, and subsequent SGLT2 inhibitors initiation was associated with a reduced risk of these adverse events. These findings underscore the significance of HZ prevention and the potential benefits of early initiation of SGLT2 inhibitors post-HZ to mitigate cardiorenal adverse events in patients with T2DM.
Clinical trial number
Not applicable.
Supplementary information
The online version contains supplementary material available at 10.1186/s12902-026-02263-5.
Keywords: Herpes zoster, Type 2 diabetes mellitus, Primary renal composite outcomes, Primary cardiovascular composite outcomes, Prognosis, SGLT2 inhibitors
Background
Type 2 diabetes mellitus (T2DM) has evolved into a global pandemic, projected to affect approximately 693 million individuals by 2045 [1]. Among patients with T2DM, renal and cardiovascular complications are highly prevalent and represent the leading causes of morbidity and mortality. While there is a growing body of evidence on the mechanistic insights and implications of cardiovascular-kidney-metabolic (CKM) syndrome [2], the interplay between infection and CKM syndrome remains underexplored.
Herpes zoster (HZ) is a common neurocutaneous disorder with an estimated lifetime incidence of nearly one in three individuals. Patients with T2DM, especially older individuals, are known to experience a higher incidence and greater severity of HZ [3, 4]. A recent large-scale United States observational study of 91 million patients revealed that patients with T2DM were associated with an 84% higher risk of HZ, alongside a significant increase in healthcare costs of $5216 compared to those with T2DM alone [5]. Beyond the well-established neurologic, cutaneous, ophthalmic and visceral complications of HZ, there is a scarcity of evidence on the potential systemic sequalae of HZ, in particular its impact on the progression of CKM health, in high-risk populations such as patients with T2DM.
Among various therapeutic agents to delay kidney disease progression and to lower cardiovascular disease risk, sodium-glucose cotransporter-2 (SGLT2) inhibitors are widely prescribed for patients with T2DM due to their favorable safety profile, glycemic efficacy, and established cardiorenal benefits [6]. No study has yet estimated the association between the initiation of SGLT2 inhibitors following an incident HZ and the subsequent risk of adverse cardiovascular and renal events in patients with T2DM. Therefore, using a well-validated, territory-wide database, this study aimed to assess the impact of incident HZ on primary renal and cardiovascular composite outcomes in patients with T2DM. Second, we sought to conduct an exploratory analysis to evaluate the initiation of SGLT2 inhibitors post-HZ on cardiorenal risk in this cohort.
Methods
Data source
This retrospective study was conducted using data obtained from the Clinical Data Analysis and Reporting System (CDARS), a territory-wide database established by the Hong Kong Hospital Authority (HA) in 1993. The Hong Kong HA is the sole statutory body responsible for managing all public hospitals and healthcare institutions, serving a population of 7.4 million residents in Hong Kong [7]. CDARS prospectively captures comprehensive and detailed patients’ information, including diagnoses, laboratory test results, drug prescriptions, hospitalization records, and mortality data. The diagnoses in this database were coded by the International Classification of Diseases, Ninth Version (ICD-9), which has been demonstrated to have high coding accuracy [8, 9]. Numerous high-quality studies were conducted with this database [10–12]. To safeguard patient privacy, all patient names and identification numbers were deidentified, and a unique reference key was assigned to each individual.
Study design
A total of 266,892 patients with T2DM were identified from the CDARS database for the period between January 2016 to December 2021. Patients aged < 18 years (N = 272), with a history of HZ (N = 12,186), and individuals with human immunodeficiency virus (N = 39) were excluded. Additionally, patients without baseline hemoglobin A1c (HbA1c) (N = 25,743), estimated glomerular filtration rate (eGFR) (N = 12,045) and body mass index (BMI) measurement (N = 45,133) were also excluded. Finally, a total of 171,474 patients were included for further data analysis (Supplementary Figure S1).
For the analysis of incident HZ, the index date was defined as the first HZ diagnosis date for exposed patients and a random clinical visit date for patients without HZ. For the exploratory analysis of SGLT2 inhibitors initiation following HZ, the index date for SGLT2 inhibitors users was defined as the first prescription date of SGLT2 inhibitors. For non-SGLT2 inhibitors users within the HZ cohort, a randomly selected clinical visit date after their HZ diagnosis was assigned. To address potential immortal time bias and balance follow-up duration, we employed a prescription time-distribution matching (PTDM) approach, a method that has been demonstrated to effectively mitigate such bias and is widely used in population-based observational studies [13–15]. In this approach, the time intervals from the cohort entry date or HZ diagnosis date to the index date were adjusted as a covariate.
Study covariates
Medical records of each patient were traced for extracting information such as age, sex, interval between cohort entry date and index date, duration of diabetes, and various comorbidities, including atrial fibrillation, cerebrovascular disease, heart failure, peripheral artery disease, hypertension, hyperlipidemia, malignancy, diabetic microvascular complications (diabetic retinopathy, diabetic neuropathy, and diabetic nephropathy), ischemic heart disease, rheumatic disease, acute kidney injury, smoking and alcoholism. All comorbidities were identified using ICD-9 codes (Supplementary Table S1). Additionally, we gathered baseline data on medication history in the year prior to index date, which encompassed the use of insulin, metformin, sulfonylureas, thiazolidinediones, dipeptidyl peptidase 4 (DPP-4) inhibitors, SGLT2 inhibitors, angiotensin converting enzyme inhibitors (ACEIs), angiotensin II receptor blockers (ARBs), beta blockers, calcium channel blockers, diuretics, statins, aspirin, and oral corticosteroids. BMI and laboratory test results for HbA1c and eGFR levels, were ascertained in the year prior to the index date, with the measurement closest to the index date used for data analysis. Obesity was defined using ICD-9 codes or a BMI ≥ 25 kg/m2, based on the Asian-pacific criteria [16]. For assessment of renal outcomes, eGFR measurements during the follow-up period were collected. Detailed definitions of study covariates are provided in Supplementary Table S1.
Follow-up endpoints
The primary renal composite outcomes was defined as a composite of end-stage renal failure, a 50% decline in eGFR, or all-cause mortality. The primary cardiovascular composite outcomes was defined as a composite of heart failure hospitalization, myocardial infarction, stroke, or all-cause mortality. All-cause mortality was incorporated in the primary composite outcomes in order to reduce adjudication bias and risk of informative censoring, while providing a comprehensive assessment of the net effect of HZ in patients with T2DM. The secondary outcomes were renal-specific composite events (defined as a composite of end-stage renal failure, a 50% decline in eGFR, or kidney mortality), major adverse cardiovascular events (MACE, defined as a composite of heart failure hospitalization, myocardial infarction, stroke, or cardiovascular mortality), and all-cause mortality. Cardiovascular mortality was defined based on death resulting from ischemic heart disease, cerebrovascular disease, hypertensive disorders without heart disease, atherosclerosis, aortic aneurysm and dissection, and other diseases of the arteries, arterioles, and capillaries (ICD-10 codes: I00-I13, I20-I51, and I60-I78); and kidney mortality was defined based on death due to chronic kidney disease, acute kidney failure, or unspecified kidney failure (ICD-10 codes: N17-N19). Patients were followed up until the occurrence of the primary outcomes, death, or end of study date (December 31, 2022), whichever came first.
Statistical analyses
Inverse probability of treatment weighting (IPTW) was employed to create a pseudo-population to balance baseline covariates between patients with and without incident HZ, as well as between SGLT2 inhibitors users and non-users following HZ diagnosis. The propensity score for the development of incident HZ or the initiation of SGLT2 inhibitors were derived via the covariate balancing propensity score (CBPS) methodology, incorporating the aforementioned covariates into multivariable logistic regression models [17]. The covariates were deemed as balanced when the standard mean difference (SMD) was ≤ 0.10 after IPTW. A multivariable Cox proportional hazards model was applied to the weighted cohort to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between HZ or initiation of SGLT2 inhibitors following HZ and the risks of primary or secondary outcomes. We further assessed the temporal association between HZ and the subsequent risks of adverse outcomes by subdividing the two-year post-infection period into two distinct follow-up intervals:1–90 days and 91–730 days.
The incidence rates (IRs) of all adverse outcomes were expressed as the number of events per 1000 person-years. The comparative risk of adverse events between patients with and without incident HZ was visualized using weighted cumulative incidence curves.
Subgroup and sensitivity analyses
We performed subgroup analyses based on the following factors: age (≤ 50 or > 50 years), sex (female or male), history of CVD, duration of diabetes (≤ 5.0 or > 5.0 years), use of oral corticosteroids, and HbA1c (≤ 7.0% or > 7.0%). To strengthen the robustness of findings from primary analysis, we performed a sensitivity analysis with modified endpoint definitions. In this sensitivity analysis, the primary renal composite outcomes was redefined as a composite of end-stage renal failure, a sustained 50% decline in eGFR, or all-cause mortality, and the renal-specific composite outcomes was redefined as a composite of end-stage renal failure, a sustained 50% decline in eGFR, or kidney mortality. A laboratory abnormality was considered sustained only if a repeated test confirmed it at least four weeks after the initial measurement. If a confirmatory test was not available, such events were only counted if the patient died subsequently [18]. Moreover, a second sensitivity analysis was conducted with initiation of SGLT2 inhibitors modeled as a time-dependent covariate to investigate the impact of SGLT2 inhibitors on the risks of primary and secondary outcomes. All statistical analyses were performed using R (version 4.3.1).
Results
Incident HZ versus no-incident HZ
Baseline clinical characteristics
This study included 171,474 patients with T2DM (mean age 66 years, 48.6% male). During follow-up, 3,306 individuals (1.93%) developed HZ, corresponding to an incidence rate of 5.70 per 1000 person-years. Patients with incident HZ were found to be older and had a higher prevalence of atrial fibrillation, cerebrovascular disease, heart failure, ischemic heart disease, malignancy, rheumatic disease, and acute kidney injury. In addition, patients with incident HZ were more likely to use insulin, diuretics, aspirin, and oral corticosteroids, and exhibited lower eGFR and higher HbA1c levels, compared to those without HZ. After implementing IPTW, all baseline covariates were balanced across the comparison groups (Table 1).
Table 1.
Baseline characteristics of patients before and after IPTW
| No-incident HZ (N = 168168) | Incident HZ (N = 3306) | SMD before IPTW | SMD after IPTW | |
|---|---|---|---|---|
| Age, years | 68 ± 11 | 68 ± 11 | 0.17 | < 0.01 |
| Male, n (%) | 81,739 (48.6) | 1659 (50.2) | 0.03 | < 0.001 |
| Diabetes duration, years | 0.06 | < 0.01 | ||
| < 1 | 12,297 (7.3) | 217 (6.6) | ||
| 1–5 | 58,023 (34.5) | 1087 (32.9) | ||
| 5–10 | 46,370 (27.6) | 906 (27.4) | ||
| > 10 | 51,478 (30.6) | 1096 (33.2) | ||
| Hypertension, n (%) | 123,273 (73.3) | 2476 (74.9) | 0.04 | < 0.01 |
| Hyperlipidemia, n (%) | 81,362 (48.4) | 1626 (49.2) | 0.02 | 0.01 |
| Atrial fibrillation, n (%) | 4473 (2.7) | 178 (5.4) | 0.14 | < 0.01 |
| Cerebrovascular disease, n (%) | 12,174 (7.2) | 395 (11.9) | 0.16 | < 0.01 |
| Heart failure, n (%) | 3821 (2.3) | 178 (5.4) | 0.16 | < 0.01 |
| Ischemic heart disease, n (%) | 13,763 (8.2) | 389 (11.8) | 0.12 | < 0.01 |
| Peripheral artery disease, n (%) | 1078 (0.6) | 38 (1.1) | 0.05 | < 0.01 |
| Malignancy, n (%) | 14,619 (8.7) | 620 (18.8) | 0.30 | < 0.01 |
| Microvascular complications, n (%) | 6593 (3.9) | 194 (5.9) | 0.09 | 0.02 |
| Rheumatic disease, n (%) | 11,136 (6.6) | 318 (9.6) | 0.11 | < 0.01 |
| Acute kidney injury, n (%) | 1802 (1.1) | 102 (3.1) | 0.14 | < 0.01 |
| Insulin, n (%) | 17,263 (10.3) | 536 (16.2) | 0.18 | < 0.01 |
| Metformin, n (%) | 134,191 (79.8) | 2582 (78.1) | 0.04 | < 0.01 |
| Sulfonylureas, n (%) | 73,543 (43.7) | 1528 (46.2) | 0.05 | < 0.01 |
| Thiazolidinediones, n (%) | 8166 (4.9) | 150 (4.5) | 0.02 | < 0.001 |
| DPP-4 inhibitors, n (%) | 15,794 (9.4) | 337 (10.2) | 0.03 | < 0.01 |
| SGLT2 inhibitors, n (%) | 2914 (1.7) | 46 (1.4) | 0.03 | < 0.01 |
| ACEIs, n (%) | 54,418 (32.4) | 1056 (31.9) | < 0.01 | < 0.01 |
| ARBs, n (%) | 40,302 (24.0) | 850 (25.7) | 0.04 | < 0.01 |
| Beta blockers, n (%) | 47,986 (28.5) | 1011 (30.6) | < 0.05 | < 0.001 |
| Calcium channel blockers, n (%) | 103,196 (61.4) | 2036 (61.6) | < 0.01 | < 0.001 |
| Diuretics, n (%) | 6103 (3.6) | 275 (8.3) | 0.20 | 0.01 |
| Statins, n (%) | 126,120 (75.0) | 2432 (73.6) | 0.03 | < 0.01 |
| Aspirin, n (%) | 27,903 (16.6) | 705 (21.3) | 0.12 | < 0.01 |
| Oral corticosteroids, n (%) | 5506 (3.3) | 249 (7.5) | 0.19 | 0.01 |
| Smoking, n (%) | 46,743 (27.8) | 966 (29.2) | 0.03 | < 0.01 |
| Alcoholism, n (%) | 54,563 (32.4) | 1165 (35.2) | 0.06 | < 0.001 |
| eGFR, ml/min/1.73m2 | 88.4 (72.2–98.6) | 83.4 (62.5–96.1) | 0.31 | < 0.01 |
| HbA1c, % | 6.8 (6.4–7.4) | 6.9 (6.5–7.6) | 0.10 | 0.01 |
| Obesity, n (%) | 93,092 (55.4) | 1763 (53.3) | 0.04 | < 0.01 |
Abbreviations: ACEIs, angiotensin-converting-enzyme inhibitors; ARBs, angiotensin II receptor blockers; DPP-4 inhibitors, dipeptidyl peptidase-4 inhibitors; eGFR, estimated glomerular filtration rate; HbA1c, glycated haemoglobin; HZ, herpes zoster; IPTW, inverse probability of treatment weighting; SGLT2 inhibitors, sodium–glucose cotransporter-2 inhibitors; SMD, standardized mean difference.
Clinical outcomes
During a median follow-up of 3.19 (2.58–3.36) years, 17,197 patients developed the primary renal composite outcomes, 13,205 developed the renal-specific composite outcomes, 12,352 developed the primary cardiovascular composite outcomes, 8,856 had MACE, and 4,291 had all-cause mortality. Among patients without incident HZ, the incidence rates for primary renal composite outcomes, renal-specific composite outcomes, primary cardiovascular composite outcomes, MACE, and all-cause mortality were 85.08, 67.29, 44.08, 27.48, and 19.28 per 1000 person-years, respectively. While among patients with incident HZ, the incidence rates of all outcomes were higher, with 140.16, 104.47, 78.68, 46.53 and 38.31 per 1000 person-years, respectively. New-onset HZ was associated with a higher risk of the following outcomes compared to no-incident HZ: primary renal composite outcomes (adjusted HR 1.42, 95% CI 1.30–1.55) (Fig. 1A; Table 2), renal-specific composite outcomes (adjusted HR 1.33, 95% CI 1.21–1.48), primary cardiovascular composite outcomes (adjusted HR 1.59, 95% CI 1.42–1.77) (Fig. 1B; Table 2), MACE (adjusted HR 1.53, 95% CI 1.33–1.76), and all-cause mortality (adjusted HR 1.73, 95% CI 1.47–2.05) (Table 2).
Fig. 1.
Weighted cumulative incidence of primary renal composite outcomes and primary cardiovascular composite outcomes associated with new-onset HZ. Shown are weighted cumulative incidence curves for primary renal composite outcomes (A) and primary cardiovascular composite outcomes (B). Abbreviations: HZ, herpes zoster
Table 2.
Association of incident herpes zoster with subsequent risk of adverse events
| No-incident HZ | Incident HZ | P value | |
|---|---|---|---|
| Primary renal composite outcomes | |||
| Events, n/N (%) | 16,447/168,168 (9.8) | 750/3306 (22.7) | |
| IR (per 1000 person-years) | 85.08 (84.13–86.04) | 140.16 (138.77-141.56) | |
| Multivariate HR (95% CI)* | Ref | 1.42 (1.30–1.55) | < 0.001 |
| Renal-specific composite outcomes | |||
| Events, n/N (%) | 12,646/168,168 (7.5) | 559/3306 (16.9) | |
| IR (per 1000 person-years) | 67.29 (66.44–68.14) | 104.47 (103.27-105.68) | |
| Multivariate HR (95% CI)* | Ref | 1.33 (1.21–1.48) | < 0.001 |
| Primary cardiovascular composite outcomes | |||
| Events, n/N (%) | 11,931/168,168 (7.1) | 421/3306 (12.7) | |
| IR (per 1000 person-years) | 44.08 (43.40-44.77) | 78.68 (77.64–79.73) | |
| Multivariate HR (95% CI)* | Ref | 1.59 (1.42–1.77) | < 0.001 |
| MACE | |||
| Events, n/N (%) | 8607/168,168 (5.1) | 249/3306 (7.5) | |
| IR (per 1000 person-years) | 27.48 (26.94–28.02) | 46.53 (45.73–47.34) | |
| Multivariate HR (95% CI) * | Ref | 1.53 (1.33–1.76) | < 0.001 |
| All-cause mortality | |||
| Events, n/N (%) | 4086/168,168 (2.4) | 205/3306 (6.2) | |
| IR (per 1000 person-years) | 19.28 (18.83–19.74) | 38.31 (37.59–39.05) | |
| Multivariate HR (95% CI) * | Ref | 1.73 (1.47–2.05) | < 0.001 |
Abbreviations: ACEIs, angiotensin-converting-enzyme inhibitors; ARBs, angiotensin II receptor blockers; CI, confidence interval; DPP-4 inhibitors, dipeptidyl peptidase-4 inhibitors; eGFR, estimated glomerular filtration rate; HbA1c, glycated haemoglobin; HR, hazard ratio; HZ, herpes zoster; IR, incidence rate; MACE, major adverse cardiovascular events; SGLT2 inhibitors, sodium–glucose cotransporter-2 inhibitors.
Primary renal composite outcomes: a composite of end-stage renal failure, a 50% decline in eGFR, or all-cause mortality
Renal-specific composite outcomes: a composite of end-stage renal failure, a 50% decline in eGFR, or kidney mortality
Primary cardiovascular composite outcomes: a composite of heart failure hospitalization, myocardial infarction, stroke or all-cause mortality
MACE: a composite of heart failure hospitalization, myocardial infarction, stroke, or cardiovascular mortality
*Adjusted for age, sex, diabetes duration, diabetic microvascular complications, malignancy, cerebrovascular disease, ischemic heart disease, chronic heart failure, atrial fibrillation, hypertension, hyperlipidemia, peripheral artery disease, acute kidney injury, rheumatic disease, insulin, metformin, sulfonylureas, DPP-4 inhibitors, thiazolidinediones, SGLT2 inhibitors, ACEIs, ARBs, beta blockers, calcium channel blockers, diuretics, statins, aspirin, oral corticosteroids, smoking, alcoholism, HbA1c, eGFR, and obesity at baseline
The risk between HZ and the primary renal composite outcomes appeared to be more pronounced in the early phase post-infection (1–90 days: adjusted HR 1.50, 95% CI 1.33–1.70). After 3 months post-infection, patients with HZ still showed a significantly increased risk of primary renal composite outcomes compared to those without HZ, but the risk was of a lower magnitude than in the early phase (91–730 days: adjusted HR 1.32, 95% CI 1.14–1.53). A similar pattern was observed for the primary cardiovascular composite outcomes, the association was more pronounced during the first 90 days post-infection (1–90 days: adjusted HR 1.72, 95% CI 1.43–2.08) and the risk attenuated in the later phase (91–730 days: adjusted HR 1.52, 95% CI 1.29–1.78). The association between HZ and all the secondary outcomes followed a similar temporal relationship (Supplementary Table S2).
Subgroup and sensitivity analyses
The results of subgroup analyses are summarized in Supplementary Figure S2 and Supplementary Figure S3. Incident HZ was consistently associated with higher risk of the primary renal composite outcomes and primary cardiovascular composite outcomes across all subgroups, including age (≤ 50 or > 50 years), sex, presence of CVD, duration of diabetes (≤ 5 or > 5 years), corticosteroids usage, and glycemic control (HbA1c ≤ 7.0% or > 7.0%). The risk of primary renal composite outcomes was more pronounced in patients without pre-existing cardiovascular diseases, with a shorter duration of diabetes, and with an HbA1c level ≤ 7.0% (all Pinteraction <0.05). Conversely, the risk of primary cardiovascular composite outcomes was greater among those with established cardiovascular diseases (Pinteraction <0.05). Findings from the sensitivity analysis, which redefined adverse renal outcomes by sustained decline in eGFR, were consistent with the findings from primary analysis (Supplementary Table S3).
SGLT2 inhibitors users versus non-SGLT2 inhibitors users
Baseline clinical characteristics
Of the 3,306 patients with incident HZ, 46 were excluded due to SGLT2 inhibitors use in the year prior to incident HZ. The final exploratory analysis cohort comprised 3,260 patients, among whom 321 (9.8%) initiated SGLT2 inhibitors following their HZ diagnosis. Compared to non-SGLT2 inhibitors users, patients initiating SGLT2 inhibitors were younger, more likely to be male, obese and smokers, and had a longer duration of diabetes. They also had a higher prevalence of heart failure, ischemic heart disease, and microvascular complications, and were more frequently prescribed metformin, sulfonylureas, thiazolidinediones, DPP-4 inhibitors, ACEIs, ARBs, beta blockers, aspirin, corticosteroids, and use of insulin. Additionally, SGLT2 inhibitors users also presented with significantly higher baseline HbA1c level. After implementing IPTW, all baseline covariates were balanced across the comparison groups (Table 3).
Table 3.
Baseline characteristics of patients with herpes zoster stratified by initiation of SGLT2 inhibitors before and after IPTW
| Non-SGLT2 inhibitors users (N = 2939) | SGLT2 inhibitors users (N = 321) | SMD before IPTW | SMD after IPTW | |
|---|---|---|---|---|
| Age, years | 68 ± 11 | 63 ± 10 | 0.47 | 0.03 |
| Male, n (%) | 1455 (49.5) | 179 (55.8) | 0.13 | < 0.001 |
| Diabetes duration, years | 0.34 | < 0.001 | ||
| < 1 | 198 (6.7) | 16 (5.0) | ||
| 1–5 | 1010 (34.4) | 70 (21.8) | ||
| 5–10 | 803 (27.3) | 89 (27.7) | ||
| > 10 | 928 (31.6) | 146 (45.5) | ||
| Hypertension, n (%) | 2275 (77.4) | 219 (68.2) | 0.21 | < 0.001 |
| Hyperlipidemia, n (%) | 1545 (52.6) | 128 (39.9) | 0.26 | < 0.001 |
| Atrial fibrillation, n (%) | 166 (5.6) | 22 (6.9) | 0.05 | < 0.001 |
| Cerebrovascular disease, n (%) | 392 (13.3) | 32 (10.0) | 0.11 | < 0.01 |
| Heart failure, n (%) | 144 (4.9) | 38 (11.8) | 0.25 | < 0.001 |
| Ischemic heart disease, n (%) | 334 (11.4) | 61 (19.0) | 0.21 | < 0.001 |
| Peripheral artery disease, n (%) | 38 (1.3) | 4 (1.2) | < 0.01 | < 0.001 |
| Malignancy, n (%) | 599 (20.4) | 36 (11.2) | 0.25 | < 0.001 |
| Microvascular complications, n (%) | 169 (5.8) | 32 (10.0) | 0.16 | < 0.001 |
| Rheumatic disease, n (%) | 294 (10.0) | 37 (11.5) | 0.05 | < 0.001 |
| Acute kidney injury, n (%) | 114 (3.9) | 11 (3.4) | 0.02 | < 0.001 |
| Insulin, n (%) | 423 (14.4) | 94 (29.3) | 0.37 | < 0.001 |
| Metformin, n (%) | 2256 (76.8) | 284 (88.5) | 0.31 | < 0.01 |
| Sulfonylureas, n (%) | 1270 (43.2) | 227 (70.7) | 0.58 | < 0.01 |
| Thiazolidinediones, n (%) | 107 (3.6) | 27 (8.4) | 0.20 | < 0.001 |
| DPP-4 inhibitors, n (%) | 237 (8.1) | 79 (24.6) | 0.46 | < 0.01 |
| ACEIs, n (%) | 903 (30.7) | 131 (40.8) | 0.21 | < 0.001 |
| ARBs, n (%) | 737 (25.1) | 96 (29.9) | 0.11 | < 0.001 |
| Beta blockers, n (%) | 884 (30.1) | 113 (35.2) | 0.11 | < 0.001 |
| Calcium channel blockers, n (%) | 1826 (62.1) | 184 (57.3) | 0.10 | < 0.001 |
| Diuretics, n (%) | 236 (8.0) | 33 (10.3) | 0.08 | < 0.001 |
| Statins, n (%) | 2157 (73.4) | 239 (74.5) | 0.03 | 0.08 |
| Aspirin, n (%) | 607 (20.7) | 84 (26.2) | 0.13 | < 0.001 |
| Oral corticosteroids, n (%) | 209 (7.1) | 33 (10.3) | 0.11 | < 0.001 |
| Smoking, n (%) | 837 (28.5) | 114 (35.5) | 0.15 | < 0.001 |
| Alcoholism, n (%) | 1016 (34.6) | 126 (39.3) | 0.10 | < 0.001 |
| eGFR, ml/min/1.73m2 | 83.3 (62.5–95.6) | 83.3 (62.2–97.9) | 0.06 | < 0.001 |
| HbA1c, % | 6.8 (6.4–7.4) | 7.7 (6.9–9.1) | 0.67 | < 0.01 |
| Obesity, n (%) | 1546 (52.6) | 192 (59.8) | 0.15 | < 0.001 |
Abbreviations: ACEIs, angiotensin-converting-enzyme inhibitors; ARBs, angiotensin II receptor blockers; DPP-4 inhibitors, dipeptidyl peptidase-4 inhibitors; eGFR, estimated glomerular filtration rate; HbA1c, glycated haemoglobin; IPTW, inverse probability of treatment weighting; SGLT2 inhibitors, sodium–glucose cotransporter-2 inhibitors; SMD, standardized mean difference
Clinical outcomes
The incidence of adverse outcomes was lower in patients who initiated SGLT2 inhibitors after HZ diagnosis compared to non-users. For the primary renal composite outcomes, renal-specific composite outcomes, primary cardiovascular composite outcomes, MACE, and all-cause mortality, the incidence rates were 85.59, 74.29, 41.99, 22.61, and 12.91 per 1000 person-years, respectively, in the SGLT2 inhibitors users, versus 149.46, 122.46, 65.19, 40.56 and 28.66 per 1000 person-years in non-SGLT2 inhibitors users. SGLT2 inhibitors users were associated with a lower risk of primary renal composite outcomes (adjusted HR 0.57, 95% CI 0.42–0.78), renal-specific composite outcomes (adjusted HR 0.61, 95% CI 0.43–0.85), primary cardiovascular composite outcomes (adjusted HR 0.54, 95% CI 0.32–0.91), MACE (adjusted HR 0.54, 95% CI 0.30–0.98), and all-cause mortality (adjusted HR 0.45, 95% CI 0.22–0.94) (Table 4). The second sensitivity analysis, in which initiation of SGLT2 inhibitors was modeled as a time-dependent covariate, also demonstrated a lower risk among SGLT2 inhibitors users for primary renal composite outcomes (adjusted HR 0.64, 95% CI 0.45–0.91) and primary cardiovascular composite outcomes (adjusted HR 0.52, 95% CI 0.30–0.89) (Supplementary table S4).
Table 4.
Association of SGLT2 inhibitors and risk of subsequent adverse events in patients with T2DM and herpes zoster
| Non-SGLT2 inhibitors users | SGLT2 inhibitors users | P value | |
|---|---|---|---|
| Primary renal composite outcomes | |||
| Events, n/N (%) | 685/2939 (23.3) | 53/321 (16.5) | |
| IR (per 1000 person-years) | 149.46 (139.57-159.87) | 85.59 (78.51–93.14) | |
| Multivariate HR (95% CI)# | Ref | 0.57 (0.42–0.78) | < 0.001 |
| Renal-specific composite outcomes | |||
| Events, n/N (%) | 505/2939 (17.2) | 46/321 (14.3) | |
| IR (per 1000 person-years) | 122.46 (113.52-131.91) | 74.29 (67.70-81.34) | |
| Multivariate HR (95% CI)# | Ref | 0.61 (0.43–0.85) | < 0.01 |
| Primary cardiovascular composite outcomes | |||
| Events, n/N (%) | 313/2939 (10.7) | 17/321 (5.3) | |
| IR (per 1000 person-years) | 65.19 (58.71–72.18) | 41.99 (37.08–47.37) | |
| Multivariate HR (95% CI)# | Ref | 0.54 (0.32–0.91) | 0.02 |
| MACE | |||
| Events, n/N (%) | 229/2939 (7.8) | 14/321 (4.4) | |
| IR (per 1000 person-years) | 40.56 (35.49–46.15) | 22.61 (19.05–26.64) | |
| Multivariate HR (95% CI) # | Ref | 0.54 (0.30–0.98) | 0.04 |
| All-cause mortality | |||
| Events, n/N (%) | 193/2939 (6.6) | 8/321 (2.5) | |
| IR (per 1000 person-years) | 28.66 (24.42–33.42) | 12.91 (10.26–16.05) | |
| Multivariate HR (95% CI) # | Ref | 0.45 (0.22–0.94) | 0.03 |
Abbreviations: ACEIs, angiotensin-converting-enzyme inhibitors; ARBs, angiotensin II receptor blockers; CI, confidence interval; DPP-4 inhibitors, dipeptidyl peptidase-4 inhibitors; eGFR, estimated glomerular filtration rate; HbA1c, glycated haemoglobin; HR, hazard ratio; IR, incidence rates; SGLT2 inhibitors, sodium–glucose cotransporter-2 inhibitors; MACE, major adverse cardiovascular events; T2DM, type 2 diabetes mellitus
Primary renal composite outcomes: a composite of end-stage renal failure, a 50% decline in eGFR, or all-cause mortality
Renal-specific composite outcome: a composite of end-stage renal failure, a 50% decline in eGFR, or kidney mortality
Primary cardiovascular composite outcomes: a composite of heart failure hospitalization, myocardial infarction, stroke, or all-cause mortality
MACE: a composite of heart failure hospitalization, myocardial infarction, stroke, or cardiovascular mortality
#Adjusted for age, sex, diabetes duration, diabetic microvascular complications, malignancy, cerebrovascular disease, ischemic heart disease, chronic heart failure, atrial fibrillation, hypertension, hyperlipidemia, peripheral artery disease, acute kidney injury, rheumatic disease, insulin, metformin, sulfonylureas, DPP-4 inhibitors, thiazolidinediones, ACEIs, ARBs, beta blockers, calcium channel blockers, diuretics, statins, aspirin, oral corticosteroids, smoking, alcoholism, HbA1c, eGFR, and obesity at baseline
Discussion
Using a large territory-wide population-based cohort of 171,474 patients with T2DM, our study demonstrated that new-onset HZ was independently associated with increased risks of adverse outcomes compared to patients without HZ. Specifically, the risks were elevated by 42% for primary renal composite outcomes, 59% for primary cardiovascular composite outcomes, and 73% for all-cause mortality, compared to those without HZ, irrespective of age, sex, presence of CVD, duration of diabetes, corticosteroids usage, and glycemic control. Results were consistent across secondary outcomes, including renal-specific composite outcomes, MACE, and all-cause mortality, and were further corroborated by robust sensitivity analyses. Notably, initiation of SGLT2 inhibitors following HZ was associated with a reduced risk of these adverse renal and cardiovascular events.
Hyperglycemia in T2DM has long been recognized as a driving force in various metabolic, hemodynamic, inflammatory, and fibrotic processes [19] that contribute to the progression of CKM syndrome. Despite recent pharmacological advancements in diabetes therapy, study has shown that intensive management of hyperglycemia only modestly lowers the risk of microvascular complications or their progression in individuals with long-term diabetes [20]. Recent studies have suggested that immunity and inflammation play a paramount role in the pathogenesis of diabetic kidney disease and thus may be potential targets in slowing the progression of CKM syndrome [21, 22]. HZ emerges as one of the most prevalent infections in this group of patients, especially among those with obesity and poor diabetes control [4, 5]. Moreover, recent studies have revealed an increasing global incidence of HZ, driven by aging global population [23, 24]. A retrospective cohort study that included 5.9 million patients with T2DM and 85.4 million patients without T2DM, revealed a higher incidence of HZ among those with T2DM (9.8 per 1000 person-years versus 2.6 per 1000 person-years) [5]. This association was supported by a meta-analysis, which reported a higher HZ risk among patients with diabetes, with an incidence rate of 7.2 per 1000 person-years in patients with diabetes [4]. In line with prior evidence, our study showed a similar HZ incidence rate of 5.9 per 1000 person-years among patients with T2DM. Furthermore, we identified that patients experiencing incident HZ were older, with multiple comorbidities, polypharmacy, lower eGFR and higher HbA1c levels at baseline. These findings highlight a potentially frail and high-risk subgroup within T2DM population characterized by poor renal function, inadequate glycemic control, and multimorbidity, which stands to gain significant benefits from targeted interventions aimed at preventing HZ and its sequalae.
Associations between HZ and adverse systemic complications have been reported across various heterogeneous populations. A handful of studies have shown that HZ was associated with an increased risk of stroke, transient ischemic attack (TIA), coronary artery disease (CAD), and arrhythmia [25–27]. Similarly, a retrospective observational study of 4999 patients with CKD revealed that HZ was associated with an increased risk of end-stage renal disease (HR 1.36, 95% CI 1.09–1.70). However, this study was limited by a relatively small sample size and inadequate control for key confounders, including baseline renal function, BMI, and smoking status [28]. Leveraging a well-validated, territory-wide database, we longitudinally tracked the cardiovascular and renal outcomes, including eGFR, in a large cohort of 171,474 real-world patients with T2DM. This analysis demonstrated for the first time that incident HZ might potentiate the metabolic-heart-kidney interactions in this group of patients, with significantly elevated associated risk of 42% and 59% in adverse primary renal and cardiovascular outcomes respectively. In particular, these associations were more pronounced during the first 90 days following infection and attenuated over time, a pattern consistent with previous studies [25, 29].
The mechanism linking HZ to an increased risk of renal and cardiovascular events are multifactorial which may involve pathways such as systemic inflammation, vasculopathy and hemodynamic alterations. Inflammatory response to VZV infection may potentially promote a hypercoagulable state that elevates thrombosis risk, which is an established pathway for CVD [30, 31]. This inflammation state may also interact with conventional cardiovascular risk factors, thereby amplifying the risk of adverse cardiovascular outcomes. Furthermore, VZV can directly infect blood vessels, causing vasculopathy that may manifest as renal artery stenosis, vasculitis, or vascular wall damage, aneurysm or rupture, which could lead to renal function deterioration or CVD [32, 33]. Hemodynamic alterations secondary to zoster-related pain and stress may also further contribute to the progression of renal impairment and development of CVD.
We further investigated the role of SGLT2 inhibitors following HZ on the risk of adverse renal and cardiovascular events. SGLT2 inhibitors are widely used in patients with T2DM, and their prescription rate is increasing [34]. A substantial body of evidence from previous studies has demonstrated that SGLT2 inhibitors confer cardiovascular and renal protective effects [35, 36]. A large-scale observational study (N = 605,345) showed that SGLT2 inhibitors was associated with a lower risk of a composite renal outcome compared to treatment with sulfonylureas or DPP-4 inhibitors [36]. Randomized clinical trials have confirmed SGLT2 inhibitors reduced the risk of heart failure hospitalization, cardiovascular death and all-cause mortality [37, 38]. Our exploratory analysis demonstrated that initiation of SGLT2 inhibitors following HZ was associated with significantly lower risks of adverse cardiovascular and renal events in patients with T2DM. We hypothesize that this protective effect may be attributable to the pleiotropic properties of SGLT2 inhibitors, encompassing anti-inflammatory, antioxidative, and vasodilatation actions [39, 40]. However, given the observational nature of this study, these mechanistic interpretations remain speculative and require validation through future experimental or interventional investigations.
Clinical implications
Given the significant cardiorenal complications of HZ in patients with T2DM, prevention of HZ in this subset of patients should be greatly emphasized. Zoster vaccines, including zoster vaccine live (ZVL) and recombinant zoster vaccine (RZV), have been demonstrated to effectively reduce the risk of HZ [41, 42]. In particular, previous studies have found that HZ vaccination was also associated with a lower risks of MACE in patients with diabetes [43] and in the general population [44]. Despite its potential benefits, global zoster vaccine coverage remains alarmingly low [45]. While the Centers for Disease Control and Prevention (CDC) currently recommends 2 doses of RZV in individuals ≥ 50 years and immunocompromised individuals aged ≥ 19 years [46], we found from this study that incident HZ was associated with a higher risk of cardiovascular and renal outcomes in patients with T2DM both above and under 50 years of age. These findings suggest that HZ vaccination should be considered for all patients with T2DM, especially in those with suboptimal diabetes control, poor renal function, or concomitant CVD, irrespective of age. Furthermore, our exploratory analysis revealed that initiation of SGLT2 inhibitors following HZ was associated with a lower risk of adverse cardiovascular and renal events. This supports the consideration of prioritizing early initiation of SGLT2 inhibitors in this group of patients, but these findings merit further evaluation in future randomized trials.
Strengths and limitations
A key strength of this study is its use of a well-validated, territory-wide database (CDARS), which prospectively collects comprehensive patient data, including detailed diagnoses, laboratory test results, drug prescriptions, hospitalization records, and death. This enabled the longitudinal assessment of disease trajectories. Moreover, the meticulous nature of this data collection helps mitigate selection biases and residual confounding, which are common limitations in observational studies.
We acknowledge some potential limitations of our data and its interpretation. First, our study lacks information on certain risk factors, such as vaccination, frailty, and socioeconomic factors. Given that vaccination is associated not only with a lower incidence of HZ but also with a decreased risk of subsequent cardiovascular events [41, 43], its absence in our dataset introduces residual confounding. Additionally, the risk associated with frailty, which includes a higher incidence of HZ and adverse cardiovascular events, may be compounded by socioeconomic factors, as lower income was demonstrated to be associated with a greater burden of frailty [47]. Consequently, not adjusting for these factors may have led to an overestimation of cardiorenal risk. Second, it is noteworthy that our exploratory analysis on initiation of SGLT2 inhibitors after HZ may be subject to confounding and indication bias, particularly given the small number of SGLT2 inhibitor users and baseline differences between groups. Although we rigorously minimized immortal time bias through the application of PTDM and mitigated confounding by using IPTW to balance baseline characteristics, some residual bias may persist. Third, the generalizability of our findings may also be limited, as the study population was drawn from a single territory-wide healthcare system in Hong Kong. Accordingly, further studies are warranted to validate the findings from our study in other ethnic populations.
Conclusion
This study demonstrated that new-onset HZ was associated with a greater risk of primary renal and cardiovascular composite outcomes in patients with T2DM, whereas initiation of SGLT2 inhibitors following HZ was associated with a lower risk of adverse renal and cardiovascular events. These findings underscore the significance of HZ prevention and the potential benefits of early initiation of SGLT2 inhibitors post-HZ to mitigate cardiorenal adverse events in patients with T2DM.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We thank the Hong Kong Hospital Authority for granting access to the data.
Author contributions
All authors critically reviewed and contributed to the intellectual content of the manuscript. KHY and MZW were involved with the conception of the study. Initial data preparation was done by MZW and CTWT. Statistical analyses and interpretation of data were undertaken by MZW and CTWT. CP was responsible for verifying the data reported in the manuscript. CP, JYH, RG, YKT, YHC, CHL, JNZ, WLG, XLL, QWR undertook revision of the manuscript. KHY provided clinical expertise. All authors have read and approved the final version of the manuscript.
Funding
This work was supported by the National Natural Science Foundation of China (No. 82270400); the Natural Science Foundation of Guangdong Province (No. 2023A1515010731); Sanming Project of Medicine in Shenzhen (No. SZSM202411021); the National Natural Science Foundation of China (No. 82570457); and the General Guidance Project of Guangzhou Municipal Health Commission (No. 20251A011017).
Data availability
Data are available upon reasonable request by contacting Prof. Yiu Kai-Hang.
Declarations
Ethics approval and consent to participate
This study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of the University of Hong Kong and the West Cluster of the Hong Kong Hospital Authority (IRB Ref. No.: UW 24–679). Given the retrospective nature of the study, the requirement for informed consent was waived by the ethics committee in accordance with relevant laws and regulations.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Mei-zhen Wu and Christopher Tze-Wei Tsang contributed equally to the work and should be considered as co-first authors.
References
- 1.Cho NH, Shaw JE, Karuranga S, Huang Y, da Rocha Fernandes JD, Ohlrogge AW, Malanda B. IDF Diabetes Atlas: Global estimates of diabetes prevalence for 2017 and projections for 2045. Diabetes Res Clin Pract. 2018;138:271–81. [DOI] [PubMed] [Google Scholar]
- 2.Ndumele CE, Neeland IJ, Tuttle KR, Chow SL, Mathew RO, Khan SS, Coresh J, Baker-Smith CM, Carnethon MR, Després JP, et al. A Synopsis of the Evidence for the Science and Clinical Management of Cardiovascular-Kidney-Metabolic (CKM) Syndrome: A Scientific Statement From the American Heart Association. Circulation. 2023;148(20):1636–64. [DOI] [PubMed] [Google Scholar]
- 3.Mareque M, Oyagüez I, Morano R, Casado MA. Systematic review of the evidence on the epidemiology of herpes zoster: incidence in the general population and specific subpopulations in Spain. Public Health. 2019;167:136–46. [DOI] [PubMed] [Google Scholar]
- 4.Lai SW, Liu CS, Kuo YH, Lin CL, Hwang BF, Liao KF. The incidence of herpes zoster in patients with diabetes mellitus: A meta-analysis of cohort studies. Med (Baltim). 2021;100(16):e25292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Poirrier JE, Meyers JL, Nagar SP, Patterson BJ, Glasser LI, Jabbour SA. Herpes Zoster Incidence and Burden in Adults With Type 2 Diabetes in the U.S.: A Retrospective Database Analysis. Diabetes Care. 2022;45(11):2585–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lam CSP, Chandramouli C, Ahooja V, Verma S. SGLT-2 Inhibitors in Heart Failure: Current Management, Unmet Needs, and Therapeutic Prospects. J Am Heart Assoc. 2019;8(20):e013389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.The Hospital Authority. Hospital authority statistical report. 2021–2022.
- 8.Cheung KS, Chen L, Chan EW, Seto WK, Wong ICK, Leung WK. Statins reduce the progression of non-advanced adenomas to colorectal cancer: a postcolonoscopy study in 187 897 patients. Gut. 2019;68(11):1979–85. [DOI] [PubMed] [Google Scholar]
- 9.Chan EW, Lau WC, Leung WK, Mok MT, He Y, Tong TS, Wong IC. Prevention of dabigatran-related gastrointestinal bleeding with gastroprotective agents: a population-based study. Gastroenterology. 2015;149(3):586–e595583. [DOI] [PubMed] [Google Scholar]
- 10.Huang JY, Tse YK, Li HL, Chen C, Zhao CT, Liu MY, Wu MZ, Ren QW, Yu SY, Hung D, et al. Prediabetes is associated with increased risk of heart failure among patients with atrial fibrillation. Diabetes Care. 2023;46(1):190–6. [DOI] [PubMed] [Google Scholar]
- 11.Ren Q-W, Yu S-Y, Teng T-HK, Li X, Cheung K-S, Wu M-Z, Li H-L, Wong P-F, Tse H-F, Lam CSP, et al. Statin associated lower cancer risk and related mortality in patients with heart failure. Eur Heart J. 2021;42(32):3049-59. [DOI] [PMC free article] [PubMed]
- 12.Wu MZ, Guo R, Chandramouli C, Liu L, Tung AM, Tsang CT, Tse YK, Chan YH, Lee CH, Huang JY, et al. Urinary tract infection and continuation of sodium-glucose cotransporter-2 inhibitors in diabetic patients. Eur Heart J. 2025. [DOI] [PubMed]
- 13.Ramos R, Comas-Cufí M, Martí-Lluch R, Balló E, Ponjoan A, Alves-Cabratosa L, Blanch J, Marrugat J, Elosua R, Grau M. Statins for primary prevention of cardiovascular events and mortality in old and very old adults with and without type 2 diabetes: retrospective cohort study. bmj. 2018;362:k3359. [DOI] [PMC free article] [PubMed]
- 14.Zhou Z, Rahme E, Abrahamowicz M, Pilote L. Survival Bias Associated with Time-to-Treatment Initiation in Drug Effectiveness Evaluation: A Comparison of Methods. Am J Epidemiol. 2005;162(10):1016–23. [DOI] [PubMed] [Google Scholar]
- 15.Mohanty A, Tate JP, Garcia-Tsao G. Statins Are Associated With a Decreased Risk of Decompensation and Death in Veterans With Hepatitis C-Related Compensated Cirrhosis. Gastroenterology. 2016;150(2):430–40.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Organization WH. The Asia-Pacific perspective: redefining obesity and its treatment. 2000;17–20.
- 17.Imai K, Ratkovic M. Covariate balancing propensity score. J Royal Stat Society: Ser B: Stat Methodol 2014:243–63.
- 18.Causland FRM, Vaduganathan M, Claggett BL, Kulac IJ, Desai AS, Jhund PS, Henderson AD, Brinker M, Perkins R, Scheerer MF, et al. Finerenone and Kidney Outcomes in Patients With Heart Failure. JACC. 2025;85(2):159–68. [DOI] [PubMed] [Google Scholar]
- 19.Tuttle KR, Agarwal R, Alpers CE, Bakris GL, Brosius FC, Kolkhof P, Uribarri J. Molecular mechanisms and therapeutic targets for diabetic kidney disease. Kidney Int. 2022;102(2):248–60. [DOI] [PubMed] [Google Scholar]
- 20.Wong MG, Perkovic V, Chalmers J, Woodward M, Li Q, Cooper ME, Hamet P, Harrap S, Heller S, MacMahon S, et al. Long-term Benefits of Intensive Glucose Control for Preventing End-Stage Kidney Disease: ADVANCE-ON. Diabetes Care. 2016;39(5):694–700. [DOI] [PubMed] [Google Scholar]
- 21.Pichler R, Afkarian M, Dieter BP, Tuttle KR. Immunity and inflammation in diabetic kidney disease: translating mechanisms to biomarkers and treatment targets. Am J Physiol Ren Physiol. 2017;312(4):F716–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Koshino A, Schechter M, Sen T, Vart P, Neuen BL, Neal B, Arnott C, Perkovic V, Ridker PM, Tuttle KR, et al. Interleukin-6 and Cardiovascular and Kidney Outcomes in Patients With Type 2 Diabetes: New Insights From CANVAS. Diabetes Care. 2022;45(11):2644–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Chen J, Abrahamson PE, Ke Y, Ong CR, Parikh R, Shantakumar S. A systematic literature review of the epidemiology and burden of herpes zoster in selected locales in Asia Pacific. Hum Vaccin Immunother. 2024;20(1):2344983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Thompson RR, Kong CL, Porco TC, Kim E, Ebert CD, Acharya NR. Herpes Zoster and Postherpetic Neuralgia: Changing Incidence Rates From 1994 to 2018 in the United States. Clin Infect Dis. 2021;73(9):e3210–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wu PY, Lin CL, Sung FC, Chou TC, Lee YT. Increased risk of cardiovascular events in patients with herpes zoster: a population-based study. J Med Virol. 2014;86(5):772–7. [DOI] [PubMed] [Google Scholar]
- 26.Breuer J, Pacou M, Gauthier A, Brown MM. Herpes zoster as a risk factor for stroke and TIA: a retrospective cohort study in the UK. Neurology. 2014;82(3):206–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Curhan SG, Kawai K, Yawn B, Rexrode KM, Rimm EB, Curhan GC. Herpes Zoster and Long-Term Risk of Cardiovascular Disease. J Am Heart Assoc. 2022;11(23):e027451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lin SY, Liu JH, Yeh HC, Lin CL, Tsai IJ, Chen PC, Sung FC, Yang YF, Huang CC, Morisky DE, et al. Association between herpes zoster and end stage renal disease entrance in chronic kidney disease patients: a population-based cohort study. Eur J Clin Microbiol Infect Dis. 2014;33(10):1809–15. [DOI] [PubMed] [Google Scholar]
- 29.Minassian C, Thomas SL, Smeeth L, Douglas I, Brauer R, Langan SM. Acute Cardiovascular Events after Herpes Zoster: A Self-Controlled Case Series Analysis in Vaccinated and Unvaccinated Older Residents of the United States. PLoS Med. 2015;12(12):e1001919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Grau AJ, Urbanek C, Palm F. Common infections and the risk of stroke. Nat Reviews Neurol. 2010;6(12):681–94. [DOI] [PubMed] [Google Scholar]
- 31.Nagareddy P, Smyth SS. Inflammation and thrombosis in cardiovascular disease. Curr Opin Hematol. 2013;20(5):457–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Gilden D, Cohrs RJ, Mahalingam R, Nagel MA. Varicella zoster virus vasculopathies: diverse clinical manifestations, laboratory features, pathogenesis, and treatment. Lancet Neurol. 2009;8(8):731–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Nagel MA, Jones D, Wyborny A. Varicella zoster virus vasculopathy: The expanding clinical spectrum and pathogenesis. J Neuroimmunol. 2017;308:112–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Adhikari R, Jha K, Dardari Z, Heyward J, Blumenthal RS, Eckel RH, Alexander GC, Blaha MJ. National Trends in Use of Sodium-Glucose Cotransporter-2 Inhibitors and Glucagon-like Peptide-1 Receptor Agonists by Cardiologists and Other Specialties, 2015 to 2020. J Am Heart Assoc. 2022;11(9):e023811. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Svanström H, Mkoma GF, Hviid A, Pasternak B. SGLT-2 inhibitors and mortality among patients with heart failure with reduced ejection fraction: linked database study. BMJ. 2024;387:e080925. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Xie Y, Bowe B, Gibson AK, McGill JB, Maddukuri G, Yan Y, Al-Aly Z. Comparative Effectiveness of SGLT2 Inhibitors, GLP-1 Receptor Agonists, DPP-4 Inhibitors, and Sulfonylureas on Risk of Kidney Outcomes: Emulation of a Target Trial Using Health Care Databases. Diabetes Care. 2020;43(11):2859–69. [DOI] [PubMed] [Google Scholar]
- 37.Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, Mattheus M, Devins T, Johansen OE, Woerle HJ, et al. Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes. N Engl J Med. 2015;373(22):2117–28. [DOI] [PubMed] [Google Scholar]
- 38.Wiviott SD, Raz I, Bonaca MP, Mosenzon O, Kato ET, Cahn A, Silverman MG, Zelniker TA, Kuder JF, Murphy SA, et al. Dapagliflozin and Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med. 2019;380(4):347–57. [DOI] [PubMed] [Google Scholar]
- 39.Preda A, Montecucco F, Carbone F, Camici GG, Lüscher TF, Kraler S, Liberale L. SGLT2 inhibitors: from glucose-lowering to cardiovascular benefits. Cardiovasc Res. 2024;120(5):443–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Girardi ACC, Polidoro JZ, Castro PC, Pio-Abreu A, Noronha IL, Drager LF. Mechanisms of heart failure and chronic kidney disease protection by SGLT2 inhibitors in nondiabetic conditions. Am J Physiology-Cell Physiol. 2024;327(3):C525–44. [DOI] [PubMed] [Google Scholar]
- 41.Lal H, Cunningham AL, Godeaux O, Chlibek R, Diez-Domingo J, Hwang SJ, Levin MJ, McElhaney JE, Poder A, Puig-Barberà J, et al. Efficacy of an adjuvanted herpes zoster subunit vaccine in older adults. N Engl J Med. 2015;372(22):2087–96. [DOI] [PubMed] [Google Scholar]
- 42.Silverii GA, Clerico A, Fornengo R, Gabutti G, Sordi V, Peruzzi O, Tafuri S, Mannucci E, Dicembrini I. Efficacy and effectiveness of Herpes zoster vaccination in adults with diabetes mellitus: a systematic review and meta-analysis of clinical trials and observational studies. Acta Diabetol. 2023;60(10):1343–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Kornelius E, Lo SC, Huang CN, Wang CC, Wang YH, Yang YS. Association of herpes zoster vaccination and cardiovascular risk in patients with diabetes: long-term insights from a retrospective cohort study. BMJ Open. 2025;15(2):e090428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Xu X, Ray I, Tang E, Arnold BF, Acharya NR. Protective Effects of Recombinant Zoster Vaccine and Antiviral Therapy Against Cardiovascular Disease Following Herpes Zoster Infection. J Infect Dis. 2025;232(2):465–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Wang X, Shang S, Zhang E, Dai Z, Xing Y, Hu J, Gao Y, Fang Q. Unraveling Herpes Zoster Vaccine Hesitancy, Acceptance, and Its Predictors: Insights From a Scoping Review. Public Health Rev. 2024;45:1606679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.CDC. What Everyone Should Know about the Shingles Vaccine (Shingrix). 2023.
- 47.Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, Seeman T, Tracy R, Kop WJ, Burke G, et al. Frailty in older adults: evidence for a phenotype. J Gerontol Biol Sci Med Sci. 2001;56(3):M146–156. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon reasonable request by contacting Prof. Yiu Kai-Hang.