Background: Postherpetic neuralgia (PHN) is a debilitating complication of herpes zoster infection.
Although various risk factors have been identifi ed, the role of iron status in PHN development remains
unexplored.
Methods: This retrospective cohort study used data from the TriNetX database to investigate the
association between serum iron levels and PHN risk. Patients diagnosed with herpes zoster between
2010–2024, who had serum iron measurements within one month post-infection, were included. Patients
were divided into an iron defi ciency (ID) group (serum iron < 50 μg/dL) and a control group (≥ 50 μg/
dL). After 1:1 propensity score matching, 4,610 patients were analyzed per group. The primary outcome
was PHN development at 1–3 months post-infection, with a secondary outcome at 3–6 months.
Results: Patients with ID had a significantly higher risk of developing PHN at 1–3 months compared
to controls (26.4% vs. 23.0%; hazard ratio = 1.24, 95% confi dence interval: 1.14–1.34, P < 0.001). No
signifi cant difference was observed at 3–6 months. In patients receiving antiviral therapy and those aged
≥ 65 years, sensitivity analyses showed a signifi cantly increased PHN risk in the ID group during both the
early and late follow-up periods. Notably, antiviral therapy did not signifi cantly reduce PHN risk in irondefi cient patients. Pre-infection iron status was not associated with subsequent PHN development.
Conclusion: ID during acute herpes zoster infection is associated with an increased risk of PHN. These
fi ndings suggest that iron status may infl uence post-infection neural infl ammation and potentially modify
the effectiveness of antiviral treatment. Future prospective studies are needed to establish causality and
determine the optimal intervention strategies for PHN.
Keywords: herpes zoster, iron defi ciency, postherpetic neuralgia, retrospective cohort study, risk factors

Herpes zoster, commonly known as shingles, results from the reactivation of the varicella-zoster virus (VZV), which leads to herpetic pain by inflammation and direct injury to the peripheral nervous system. When herpetic pain persists for more than three months, it is known as postherpetic neuralgia (PHN), which can last for months or even years. PHN is not only the most common complication, but also a debilitating complication of herpes zoster infection. Currently, pharmacological therapies often do not provide satisfactory pain relief for some patients with PHN [1,2]. Identifying and preventing PHN is crucial because it is a chronic pain condition that can significantly impact quality of life.

Iron is an essential micronutrient that plays an important role in various biological processes, including the antiviral functions of natural killer (NK) cells [3,4], synthesis of monoamine neurotransmitters [5], and myelin formation [6]. Iron deficiency (ID) may result in immunological dysfunctions [3,4], neurochemical dysregulation, demyelination, and axonal degeneration [7-9]. A case-control study has demonstrated that low dietary iron intake increases the risk of herpes zoster in individuals over 60 years old [10]. Peptic ulcer disease (PUD) is associated with an increased risk for both herpes zoster and PHN [11,12]. Patients with PUD often experience ID due to chronic gastrointestinal blood loss [13,14]. Importantly, ID has been found to increase acute and chronic pain responses as well as peripheral neuropathy in mice [15-18].

The pathophysiology of PHN involves neuronal excitability and persistent VZV-induced ganglionitis [19,20]. In a rat model of PHN, VZV infection upregulated inducible nitric oxide synthase (iNOS), which produces nitric oxide (NO) and leads to spinal astrocytic activation. Activated astrocytes significantly increase the expression of interleukin-1 beta (IL-1β), which subsequently activates N-methyl-d-aspartate (NMDA) receptors in spinal dorsal horn neurons [21]. Activation of NMDA receptors in dorsal horn neurons enhances neuronal excitability, resulting in mechanical allodynia, a prominent and characteristic symptom of neuropathic pain [22]. Notably, ID through increasing the expression and activity of iNOS [23] can promote “NO—Astrocyte—Cytokine—NMDAR—Neuron” pathway [21], contributing to the development and maintenance of PHN. Furthermore, ID decreases NK cell immunity, leading to increased viral loads and profound inflammation because NK cells require iron to effectively eliminate virus-infected cells [3,4]. Subsequently, chronic low-grade productive virus infection in the ganglia leads to chronic VZV-induced ganglionitis and persistent herpetic pain in the hosts [24]. Overall, ID may play a critical role in the pathological mechanisms of PHN.

To our knowledge, no study has investigated whether patients with shingles and coexisting ID have an increased risk of PHN. Therefore, we aimed to assess the potential role of iron status in the development of PHN using the TriNetX database.

This retrospective cohort study utilized data from the TriNetX database, a global federated research network that provides access to electronic health records from healthcare organizations. The TriNetX platform allows real-world data analysis while maintaining patient privacy through de-identification protocols in compliance with HIPAA regulations. The database incorporates comprehensive patient information, including demographic details, diagnoses, procedures, medications, and laboratory results from millions of patients across multiple healthcare systems, enabling robust population-level analyses. The study protocol was reviewed and approved by the Institutional Review Board of Chi Mei Medical Center (IRB number: 11310-E04), which waived the requirement for informed consent. This waiver was granted owing to the use of de-identified data, which minimizes the risk to patient privacy. Individual patient identifiers were not accessible to the researchers at any point during the study period.

We conducted a retrospective cohort study to identify patients diagnosed with herpes zoster infection between January 1, 2010, and December 31, 2024. From this initial population, we included only patients who had at least one serum iron level measurement within one month following their herpes zoster diagnosis, as this laboratory value served as our primary exposure variable of interest.

Patients were stratified into two groups based on their serum iron levels within one month of herpes zoster infection. The ID group comprised patients with serum iron levels < 50 μg/dL, while the control group included patients with serum iron levels ≥ 50 μg/dL. This cutoff was selected based on a previous study that considered values below 50 μg/dL as indicative of ID in adults [25].

To ensure clear group separation and to minimize classification errors, additional exclusion criteria were applied to each group. Patients in the ID group were excluded if any serum iron measurement exceeded 50 μg/dL across multiple measurements within the specified timeframe. Conversely, in the control group, patients were excluded if any serum iron level fell below 50 μg/dL across multiple measurements within the same period.

To properly identify new cases of PHN, we excluded patients with documented use of medications commonly prescribed for neuropathic pain [20] (i.e., pregabalin, gabapentin, opioid analgesics, tricyclic antidepressants, capsaicin cream, or lidocaine patch) within one month prior to their herpes zoster diagnosis. This exclusion criterion helped ensure that the medication use observed after herpes zoster diagnosis was likely related to the management of PHN rather than pre-existing neuropathic pain.

We extracted patient demographic characteristics from the TriNetX database, including age at the index date (defined as the date of herpes zoster diagnosis), sex, race, and body mass index (BMI). Clinical data included comorbid conditions identified using ICD10 codes: essential (primary) hypertension (I10), chronic kidney disease (N18), neoplasms (C00–D49), diabetes mellitus (E08–E13), ischemic heart diseases (I20–I25), overweight and obesity (E66), cerebrovascular diseases (I60–I69), liver diseases (K70– K77), chronic obstructive pulmonary disease (J44), malnutrition (E40–E46), nicotine dependence (F17), systemic connective tissue disorders (M30–M36), and alcohol-related disorders (F10). Additionally, relevant laboratory parameters were collected, including serum albumin and hemoglobin A1c levels. Vaccination history was also obtained, including administration of COVID-19 vaccines, VZV glycoprotein E vaccine, and live attenuated VZV vaccine (Oka-Merck strain).

To minimize potential confounding effects and selection bias, propensity score matching was used.Propensity scores were calculated using logistic regression, incorporating all the variables listed in Section 2.3. We used a 1:1 greedy nearest-neighbor matching algorithm without replacement and with a caliper width of 0.1 of the standard deviation of the logit of the propensity score. The quality of matching was assessed using standardized mean differences (SMD), with values < 0.1, indicating an adequate balance between the groups.

Our primary outcome measure was the development of PHN at 1–3 months following herpes zoster infection. The secondary outcome measure was the development of PHN at 3–6 months following herpes zoster infection. PHN was identified based on the prescription of medications commonly used for its management [20] (i.e., pregabalin, gabapentin, opioid analgesics, tricyclic antidepressants, capsaicin cream, or lidocaine patch) following the herpes zoster diagnosis in the absence of such prescriptions in the month preceding the diagnosis.

We conducted two sensitivity analyses to assess the robustness of our findings. Model I was restricted to patients who received antiviral therapy (i.e., acyclovir, valacyclovir, and famciclovir) within one month of herpes zoster diagnosis. Antiviral therapy is strongly advised as first-line treatment in the first 72 hours of rash onset for all immunocompetent patients fulfilling certain criteria [26], although there is limited evidence regarding antiviral therapy and the risk of developing PHN [27]. This analysis helps to evaluate the association in a subgroup of patients who received antiviral treatment for herpes zoster, reducing the potential risk of PHN variability as a confounder. Model II was restricted to patients aged ≥ 65 years. This analysis addressed the higher risk and prevalence of PHN in older adults. Age is one of the strongest risk factors for PHN development, and restricting the analysis to older adults helps to ensure that our findings are applicable to higher-risk populations.

Iron plays a vital role in immune function and viral replication, and ID may alter the effectiveness of antiviral medications through changes in viral kinetics, host immune response, or drug metabolism [3,4].To evaluate the effectiveness of antiviral therapy in patients with ID, we performed a subgroup analysis to examine the impact of antiviral therapy on the risk of PHN in patients with ID within one month of herpes zoster infection.

Table 4. Subgroup Analysis of the Association Between Anti-Viral Therapy and Risk of Postherpetic Neuralgia in
Patients With ID

To investigate whether ID preceding herpes zoster infection might also influence PHN risk, we conducted an additional analysis examining the relationship between iron biomarkers measured the month before herpes zoster diagnosis and subsequent PHN development. For this analysis, we defined iron status using two different biomarkers. Serum iron levels were used, with ID defined as serum iron levels < 50 μg/dL, whereas the control group had serum iron levels ≥ 50 μg/dL. Serum ferritin levels were also analyzed, with low ferritin defined as serum ferritin levels < 30 ng/mL, while the control group had serum ferritin levels ≥ 30 ng/mL.

Continuous variables are reported as mean ± standard deviation, and categorical variables are presented as counts and percentages. We used Cox proportional hazards regression models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between iron status and the risk of developing PHN. For the 1–3 months outcome assessment, we calculated the time-to-event from the date of herpes zoster diagnosis to the date of PHN diagnosis occurring from 1 to 3 months post-infection or to the end of the 3-month follow-up period, whichever occurred first. Similarly, for the 3–6 months assessment, time-to-event was calculated from the date of herpes zoster diagnosis to the date of PHN diagnosis occurring from 3 to 6 months post-infection or to the end of the 6-month follow-up period. All statistical analyses were performed using the TriNetX platform. A two-sided P-value < 0.05 was considered statistically significant.

The initial database search identified patients diagnosed with herpes zoster infection between January 1, 2010, and December 31, 2024. After applying our inclusion and exclusion criteria, we identified 5,148 patients in the ID group and 6,742 patients in the control group before propensity score matching (Figure). Following 1:1 propensity score matching, we retained 4,610 patients in each group for the primary analysis.

Table 1 presents the baseline characteristics of the study population before and after propensity score matching. Before matching, patients in the ID group were older (66.5 ± 17.0 vs. 62.8 ± 16.9 years, SMD = 0.221) and had a higher prevalence of several comorbidities compared to the control group, such as chronic kidney disease (35.1% vs. 28.6%), diabetes mellitus (33.5% vs. 26.4%), and ischemic heart diseases (30.9% vs. 21.8%). The ID group also had a lower proportion of patients with albumin ≥ 3.5 g/ dL (63.5% vs. 68.5%). After propensity score matching, all baseline characteristics were well balanced between the two groups, with all SMD being less than 0.10. The mean age of the matched cohort was 65.7 ± 17.3 years in the ID group and 65.6 ± 16.2 years in the control group (SMD = 0.003).

Figure. Flowchart for Patient Selection
Abbreviation: ID, iron deficiency.

Table 1. Baseline Characteristics of Patients With Herpes Zoster Infection Before and After Propensity Score
Matching

In the matched cohort, patients in the ID group had a significantly higher risk of developing PHN within 1–3 months following herpes zoster infection compared to the control group (26.4% vs. 23.0%, HR = 1.24, 95% CI: 1.14–1.34, P < 0.001) (Table 2). However, there was no statistically significant difference in the risk of PHN within 3–6 months (21.9% vs. 22.4%, HR = 1.08, 95% CI: 0.99–1.18, P = 0.074).

Table 2. Association Between Serum Iron Levels Within One Month of Herpes Zoster Infection and the Risk of
PHN

To assess the robustness of our findings, we conducted two sensitivity analyses (Table 3). In Model I, which was restricted to patients who received antiviral therapy within one month of herpes zoster diagnosis (n = 2,619 per group), the ID group had a significantly higher risk of PHN during both the 1–3 month period (HR = 1.20, 95% CI: 1.08–1.34, P < 0.001) and the 3–6 month period (HR = 1.15, 95% CI: 1.03–1.29, P = 0.012) than the control group. In Model II, which was restricted to patients aged ≥ 65 years (n = 3,122 per group), the results were consistent with those of Model I. The ID group had a significantly higher risk of PHN within 1–3 months (HR = 1.20, 95% CI: 1.09–1.33, P < 0.001) and a marginally higher risk within 3–6 months (HR = 1.12, 95% CI: 1.01–1.25, P = 0.037) than the control group. In patients receiving antiviral therapy and those aged ≥ 65 years, sensitivity analyses showed a significantly increased PHN risk in the ID group during both the early and late follow-up periods. The consistency across models in the sensitivity analyses confirmed the initial findings, suggesting that the increased risk is robust and not merely a product of the characteristics of the initial cohort.

Table 3. Sensitivity Analysis of the Association Between Serum Iron Levels Within One Month of Herpes Zoster
Infection and the Risk of Postherpetic Neuralgia

Subgroup analyses were performed to evaluate the effectiveness of antiviral therapy in patients with ID (Table 4). Among patients in the ID group, antiviral therapy did not show a significant protective effect against PHN within either 1–3 months (HR = 1.06, 95% CI: 0.93–1.20, P = 0.375) or 3–6 months (HR = 1.04, 95% CI: 0.90–1.20, P = 0.604). These findings suggest that ID may attenuate the efficacy of antiviral therapy in preventing PHN.

To explore whether ID preceding herpes zoster infection influences the risk of PHN, we analyzed the relationship between iron biomarkers measured the month before herpes zoster diagnosis and subsequent PHN development (Table 5). For serum iron levels (n = 1,686 per group), there was no significant difference in the risk of PHN between the ID group and the control group within either 1–3 months (HR = 0.99, 95% CI: 0.86–1.14, P = 0.872) or 3–6 months (HR = 1.02, 95% CI: 0.89–1.18, P = 0.741). Similarly, for serum ferritin levels (n = 1,098 per group), there was no significant difference in the risk of PHN between patients with low ferritin levels (< 30 ng/mL) and those with normal ferritin levels (≥ 30 ng/mL) within either 1–3 months (HR = 1.01, 95% CI: 0.54–1.22, P =0.922) or 3–6 months (HR = 0.96, 95% CI: 0.79–1.15, P = 0.647).

Table 5. Relationship Between Iron Biomarkers One Month Before Herpes Zoster and the Risk of Postherpetic Neuralgia

Our study is the first to demonstrate a significant association between post-infection ID and an increased risk of PHN. After propensity score matching, patients with low serum iron within one month of herpes zoster diagnosis showed higher PHN risk during the early follow-up period. This association persisted in the sensitivity analyses of patients receiving antiviral therapy and older adults. Notably, antiviral therapy showed reduced efficacy in preventing PHN in iron-deficient patients, suggesting that iron status may influence antiviral treatment outcomes. Pre-infection ID showed no association with PHN, indicating the importance of acute-phase iron metabolism rather than chronic ID. These findings suggest a critical role of iron in immune function, viral clearance, and neural repair, which contribute to PHN pathophysiology following herpes zoster infection.

The current study showed a significantly greater risk of PHN in patients with low serum iron during the 1–3 months following herpes zoster infection. Age is a critical factor for herpes zoster infection and the development of PHN. The effectiveness of antiviral therapy in reducing the risk of PHN remains controversial [27]. To validate the robustness of the association between ID and a higher risk of PHN, two sensitivity analyses of patients receiving antiviral therapy and older adults were conducted. Consistency across models in the sensitivity analyses confirmed the initial findings, suggesting that the increased risk of PHN is robust and not merely a product of the characteristics of the initial cohort. We provide explanations linking the proposed pathophysiology of PHN [19,20]. First, low iron availability heightens the risk of pain by promoting pro-inflammatory cytokines, such as iNOS, and chronic inflammation in response to infection [28-31]. ID enhances the expression and activity of iNOS [23], promoting the “NO—Astrocyte—Cytokine—NMDAR—Neuron” pathway [21] and contributing to the development and maintenance of PHN. Second, ID diminishes NK cell immunity, resulting in an increased viral load and inflammation [3,4], which can lead to chronic VZV-induced ganglionitis and persistent herpetic pain. Third, the biosynthesis of monoamines (dopamine and noradrenalin) relies on the rate-limiting enzyme tyrosine hydroxylase, which requires ferrous iron for its activity [32]. Consequently, ID reduces spinal dopamine levels, leading to mechanical allodynia, one of the hallmarks of neuropathic pain [33]. Fourth, iron is a crucial cofactor for enzymes involved in myelin synthesis. ID activates heme-regulated eIF2α kinase, triggering a maladaptive integrated stress response from disrupted mitochondrial metabolism in Schwann cells [7,8]. This dysfunction causes aberrant lipid metabolism, resulting in demyelination and axonal degeneration, ultimately leading to peripheral neuropathy [9]. All the molecular mechanisms underlie the potential links between ID and the development of neuropathic pain, supporting the finding that shingled patients with coexisting ID had a higher risk of PHN during the early follow-up period.

Acute infections can transiently alter iron metabolism as part of the inflammatory response [29]. Our retrospective study could not establish whether the observed ID was a cause of acute herpes zoster infection or a result of shingles. However, pre-infection iron biomarkers, including serum iron and ferritin levels, showed no association with subsequent PHN development during either the 1–3 month period or the 3–6 month period. The results indicate that acute alteration in iron metabolism, not chronic iron status, influences the outcomes of herpes zoster. A recent study discovered that persistent inflammation and low serum iron were seen as early as two weeks post-infection in individuals with long COVID, irrespective of COVID-19 severity [31]. The authors demonstrated that long COVID patients with early iron dysregulation displayed a compromise in T and B cell immunity, NK cell activation, and neutrophil antimicrobial activity, which resulted in a longer time to recovery [31]. Furthermore, post-COVID pain has also been found to be one of the prevalent long COVID symptoms [30]. Low serum iron levels in both herpes zoster and COVID-19 may potentially affect patient outcomes associated with the interplay between iron metabolism, inflammation, and immune response. More studies are required to investigate the issues regarding the risk of PHN and shingles patients experiencing early iron dysregulation.

These findings from our study and other studies [30,31] raise the question of whether iron supplementation during acute herpes zoster infection could be a preventive strategy for PHN. A case age/sex-matched control study in a tertiary referral medical center revealed that maintenance hemodialysis patients receiving iron therapy had a significantly lower risk of developing herpes zoster reactivation [34]. However, iron is known to participate in virus replication in living host cells. A large trial [35] found that routine iron supplementation in children aged 1–35 months with high malaria rates increased illness and death. Some studies [36,37] discovered that ID was associated with lower risks of malaria, and mortality. Conversely, a Cochrane review including 35 trials in children [38] concludes that iron therapy does not increase the risk of clinical malaria under regular malaria prevention or efficient management. Another example is that iron promotes the translation of the hepatitis C virus (HCV) [39]. This contradicts evidence that iron can inhibit the activity of HCV RNA polymerase NS5B, leading to reduced viral replication. This inhibition occurs because iron is a cofactor for NS5B, and excessive iron can disrupt the structure and function of NS5B [40]. Overall, it is still uncertain whether iron therapy inhibits or enhances HCV replication [41]. Similarly, it remains unclear whether supplementing iron for shingles patients with ID during the herpes zoster infection period would modify PHN risk or worsen their condition. The efficacy and safety of iron therapy in patients with shingles with ID need to be carefully explored.

Antiviral drugs are effective in treating acute herpes zoster; however, their effectiveness in preventing PHN is uncertain and subject to ongoing research. A meta-analysis suggested that future studies on the effectiveness of antiviral drugs in preventing PHN should be conducted among various subgroups of individuals [27]. In the present study, sensitivity analyses restricted to patients who received antiviral therapy within one month of herpes zoster infection showed that the ID group had a significantly higher risk of PHN during both the 1–3 month period and the 3–6 month period. Intriguingly, antiviral therapy in patients experiencing ID within one month of herpes zoster infection did not show a significant protective effect against PHN during the early or late follow-up period. These findings suggest that ID may potentially attenuate the efficacy of antiviral therapy in preventing PHN, a confounding factor of the efficacy of antiviral therapy. Thus, the serum iron status in patients with shingles should be taken into account regarding the effectiveness of antiviral therapy in preventing PHN in future trials.

Our study has several limitations that should be considered when interpreting the results. First, as this was a retrospective study using electronic health records, we could not control for all potential confounding factors. While propensity score matching helped balance the observed covariates between the groups, unmeasured confounders might still exist. Second, we identified PHN based on medication prescriptions [20], which might not capture all PHN cases, particularly milder cases managed without prescription medications. Additionally, these medications could have been prescribed for other neuropathic pain conditions that developed after herpes zoster infection, potentially leading to misclassification. Third, the timing of serum iron measurement within one month of herpes zoster diagnosis varied among patients, and a single measurement might not reflect the overall iron status throughout the course of infection and recovery. Multiple measurements over time would provide a more accurate picture of how iron dynamics influence PHN development. Fourth, our study could not determine whether the observed ID was a cause or consequence of the acute herpes zoster infection itself. Finally, although our database included patients from multiple healthcare systems, selection bias may still exist because patients who had their serum iron levels measured might differ systematically from those who did not, potentially limiting the generalizability of our findings.

This study demonstrated that ID during acute herpes zoster infection is associated with an increased risk of PHN development. The relationship appears strongest for early PHN (1–3 months post-infection) but persists for later PHN (3–6 months) in specific populations, including older adults and those receiving antiviral therapy. Additionally, our findings suggest that ID may reduce the effectiveness of antiviral treatment in preventing PHN. These results highlight a potentially modifiable risk factor for PHN and suggest new avenues for preventive strategies. However, given the complex role of iron in infection and inflammation, future prospective studies are needed to establish causality and determine optimal intervention strategies.

None.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

This work was funded by Chi Mei Medical Center, Tainan, Taiwan (CMHCR112043).