Association between herpes zoster and end
stage renal disease
entrance in chronic kidney disease patients: a
population-based
cohort study
S.-Y. Lin & J.-H. Liu & H.-C. Yeh & C.-L. Lin & I.-J. Tsai & P.-C. Chen & F.-C. Sung &
Y.-F. Yang & C.-C. Huang & D. E. Morisky & Y.-J. Chang & C.-H. Kao
Introduction
Chronic kidney disease (CKD) is becoming a major concern to global public health, and may result in adverse cardiovascular events, premature death, and end stage renal disease (ESRD) [1]. Early detection and strategic recommendations have been proposed to alleviate thisworldwide burden [2]. The progressive nature of CKD involves susceptibility, initiation and subsequent progression factors; sharing many similarities with atherosclerosis [3]. Beyond the traditional risks of cardiovascular disease,
infectious agents such as herpes virus and chylamydia might also cause atherosclerosis, in addition to endothelial damage [4]. However, there have been no studies investigating possible links between infection and the progression of CKD.
Herpes zoster (HZ), clinically manifested as painful shingles, is caused by reactivation of latent varicella zoster virus (VZV) within the sensory ganglia and dorsal roots [5]. It often occurs in elderly and immunocompromised patients [6]. Several studies have shown ESRD patients being vulnerable to
HZ attack [7, 8]. The sequela of HZ includes post-herpetic neuralgia, encephalitis, and VZV vasculopathy of the central nervous system (CNS) [9]. HZ is also associated with stroke, speculated due to accelerating atherosclerosis development [10]. Given the tight links between atherosclerosis and CKD [11–13], HZ might also potentially accelerate CKD progression
and ESRD development. Although VZV-related vessel diseases have been reported, the data concerning risks of developing ESRD in CKD patients with HZ are lacking.
Taiwan has an extremely high incidence and prevalence of CKD compared with other countries and thus provides excellent
data on CKD investigation [14]. We aim to use a nationwide population database to assess the risks of developing ESRD in CKD patients after a herpes zoster attack. Materials and methods
Data source
The National Health Insurance (NHICPN) program in Taiwan is a universal health insurance system covering 99 % of all 23 million citizens and contracted with 97 % of hospitals and clinics in Taiwan [15]. The NHI Research Database contains comprehensive health care data including files of inpatient care, outpatient care, prescription drugs, medical services, registration of catastrophic illness, and basic demographic information from1996 to 2008. In this study, we used a subset up to one million beneficiaries randomly selected from the entire population of 23 million in the NHI Research Database, which provides longitudinal medical history of each individual. The diagnoses were coded using the International Classification of Diseases 9th Revision of the Clinical Modification
(ICD-9-CM). Ethics statement
Because identification numbers of patients had been encrypted, patient consent was not required for this study. This study was approved by the Research Ethic Committee at ChinaMedical University (CMU-REC-101-012). The committee waived the requirement for consent.
Study population
Using the NHI Research database over the period 1996–2008, we enrolled 22,928 patients who had been diagnosed with CKD (ICD-9 code 585) for the first time and were older than
18 years. Exclusion was those who ever received renal transplantation (n=197, ICD-9-CM code V42.0). The remaining
CKD patients with the first HZ (ICD-9 code 053.0–053.9) were defined as the HZ cohort (n = 1,144). The comparison
group was matched on age, sex, and index year of HZ group, based on a 1:4 ratio. The index date was the first date of herpes zoster coding. The end point was the date of ESRD diagnosis (based on the Catastrophic Illness Patient Database).
All subjects were followed up until the occurrence of ESRD, withdrawal from NHI, death, or December 31, 2008, whichever came first. The residential areas were grouped into four levels, based on population density (persons/km2), population ratio of education level, elderly and agriculture workers
and the density of physicians (per 100,000 people). Level 1 is the most urbanized, whereas level 4 is the least. The ESRD related co-morbidities included heart disease (ICD-9-CM: 140–429), hypertension (ICD-9-CM: 401–405), diabetes mellitus (ICD-9-CM: 250), hyperlipidemia (ICD-9-CM: 272), and malignancy (ICD-9 codes 140–208). Medications which could affect the progression of CKD were included as analysis variables, such as NSAIDs, angiotensin II receptor blockers (ARB) and angiotensin–converting enzyme inhibitors (ACEI). We defined prescriptions of NSAID and
ACEI/ARB as those being prescribed after the index date of
herpes zoster diagnosis till the end point. Use of immunosuppressive medications was defined as 3 months before the
diagnosis of herpes zoster in order to clarify the effects of
immunosuppresive medications on HZ attack. Receiving antiviral therapy was defined as prescription after the diagnosis
of herpes zoster and classified into the types of topical, injection, and oral medications. Statistical analysis
The differences in demographic characteristics, comorbidities and prescribed medications between study and comparison cohorts were examined by using chi–square test
for categorical variables and Wilcoxon signed rank test two samples for continuous variables. Person–years of follow up were calculated for each study subject from the index date to the date of ESRD occurrence, censored, or the end of study. Incidence rates were calculated by dividing the number of patients with ESRD by total
person years of follow up. Cox proportion hazard model was used to calculate hazard ratios (HRs) and their 95 %
confidence intervals (95 % CIs). Crude HRs and multivariate adjust HRs were calculated. We plotted the
Kaplan–Meier curve to estimate the probability of subsequent ESRD between HZ and comparison cohort. Log–
rank test was used to examine the significance of difference between the 2 cohorts. We performed all statistical
analysis using SAS 9.1 statistical software (SAS Institute, Inc., Cary, NC, USA). A p value of <0.05 was considered significant. Results
We enrolled 1,144 CKD patients who had HZ over the period 1996–2008 as HZ cohort and 3,855 age- and sex-matched CKD patients without HZ as comparison cohort. A total of 396 patients developed ESRD till the end of follow up. The mean follow up years were 3.0±2.3 years in the HZ cohort and 3.3±2.4 years in the comparison cohort (data not shown). Table 1 displays the difference in socio-demographic characteristics, comorbidities, and medications among the two
cohorts. HZ cohort had a significantly higher rate than comparison cohort of malignancy (9 % vs. 6.1 %, p = 0.0008),
hyperlipidemia (56.5 % vs. 53.5 %, p=0.048), and NSAID consumption (96.9 % vs. 95.4 %, p=0.031). There were no significant differences in the distribution of income, urbanization level, diabetes mellitus, hypertension, heart disease, and
systemic lupus erythematosus between the two cohorts. Table 2 exhibits the details of crude and adjusted HRs for ESRD between the two cohorts. After adjusting for age, sex, and comorbidities, the HR for developing ESRD during the follow up years was 1.36 (95 % CI, 1.09–1.70) for the HZ cohort as compared with controls. The incidence of ESRD in the HZ cohort was highest in the groups of age >64 years (95% CI, 1.15–2.05), women (95 % CI, 1.16–2.08), higher income (95 % CI, 1.06–3.45), and NSAID use (95 % CI, 1.16–1.83). The log rank test showed that HZ cohort had significantly higher cumulative incidence rates of ESRD than the comparison group (p = 0.0071). Figure 1 displays the results of Kaplan– Meier survival analysis between patients with or without HZ. We further analyzed the interaction between HZ and comorbidities in the subsequent risks of ESRD. Herpes zoster
interacted with hypertension and diabetes in developing ESRD (the P value for interaction test: 0.0498). The HR of developing ESRD increased to 8.71 (95 % CI, 5.23–14.5) for CKD patients with HZ, diabetes, and hypertension concomitantly (Table 3). Table 4 shows the incidence and hazard ratio
of ESRD in HZ patients receiving different types of treatment. Comparing with HZ patients under topical medication, patients who received systemic injection of antiviralmedications
had a higher risk of ESRD (HR 2.56, 95 % CI 1.99–5.99). Discussion
To our knowledge, this is the first cohort study which
demonstrated a relationship between HZ attacks and subsequent ESRD risks of CKD patients using a populationbased
database. We have several unique findings: CKD
patients with HZ had a HR of 1.36 in developing subsequent ESRD; CKD patients with HZ which required systemic
injection of antiviral therapy had a HR of 2.56 for
developing ESRD compared with those who used topical ointment only; and the risk of subsequent ESRD increased to 8.71 in CKD patients, especially those with HZ and concomitant hypertension and diabetes.
Interpretation
HZ attacks are more common in CKD patients than in the general population [7]. Worsened renal function status may lead to immune dysfunction, decrease lymphocyte counts and activation, alter cytokine production, weaken VZV–related immunity, and result in increased risk of herpes zoster (HZ) [16, 17]. Sato et al. have reported that the incidence of HZ increased with the progression of CKD stages [7]. Therefore, CKD patients who had HZ attacks might have more advanced underlying renal failure and progress to ESRD much faster than patients with incipient renal dysfunction. Moreover, it might be difficult to conclude that topical skin lesions might have systemic influences and possible detrimental effects on organs far away. To eliminate this possible bias, we further analyzed the HZ cohorts concerning the type of antiviral therapy. Our data showed those CKD patients with HZ who required systemic injection of antiviral drugs had a HR of 2.56
for developing ESRD compared with those who received topical ointment. Besides, the association between HZ and increased risk of developing future ESRD was found among subjects aged more than 64 years. The systemic antiviral therapy is recommended for herpes patients with non-truncal
involvement or increase in severity of diseases [18]. The consideration of systemic antiviral therapy is seldom directly
related with the status of patient’s renal function. Thus, we suspect that the increasing risk of ESRD might result from the clinical advancement of HZ which causes a further systemic burden for originally damaged or aging kidneys. CKD as well as aging have generally manifested systemic vascular pathology due to various accumulating insults [19, 20]. Progressive microvascular atherosclerosis of kidney is common, under-recognized, and tightly associated with renal function decline [21–24]. Further, it has
been noted that VZV vasculopathy is not always confined within the CNS, and localized herpes zoster attacks would impact vessels distant from the lesions [9].Multiple peripheral thrombotic events, renal vasculitis, renal and/or coronary arterial stenosis and/or multiple aneurysms have been reported to be associated with herpes virus infection [25–27]. Various reports indicated that ‘herpes burden’
accelerates atherosclerosis development and causes cardiovascular events as well as stroke [4, 10, 28, 29]. The result
that the risk of subsequent ESRD increased to 8.71 in CKD patients with HZ and concomitant hypertension and diabetes strengthens our speculation considering that hypertension and diabetes are two important atherogenetic factors.
Further studies are recommended to investigate the underlying interactions among HZ, atherosclerosis, and renal
function deterioration.
Our HZ patients with HTN have the increasing risk of developing future ESRD. This finding might partly result from sympathetic excitation. Immuno-compromised groups, like CKD patients, might have greater chances of postherpetic complications, thus indicating more lasting pain and
excitation may lead to higher blood pressure, and if not
interrupted, may establish a vicious cycle among CKD patients with sympathetic over-excitation, sustained hypertension, and further advancing renal function decline [31].
In addition to the effects of comorbidities, the therapeutic agents of herpes and related pain might also have nephrotoxic potential [32]. Our data showed that patients who have used NSAID had more chances of subsequent ESRD. Analgesics, particularly NSAIDs, can cause renal injury, including acute
interstitial nephritis, medulla ischemia, and/or nephrotic syndrome [33]. Since CKD status is vulnerable to potential
nephrotoxins, these acute renal insults would tend to be irreversible, permanent, and be associated with faster progression
of renal deterioration [34]. Strengths and limitations
Our study has several strengths. It is a nationwide populationbased study. The use of comprehensive electronic medical
information, provided by the representative nation-based data, gives details about the temporal relationship between HZ attacks and ESRD development among CKD patients. Based on NHIRD in Taiwan, several studies have been conducted well in investigating other associated sequels of HZ or CKD [35–39]. Thus, the diagnosis of CKD as well as HZ infection (by ICD-9 codes) has been well validated in this claim database. However, the study has several limitations: patients
might have used over-the-counter medications; and the potential confounding variables including baseline renal function,
smoking, family history of renal disease, lack of CKD severity, and obesity are associated with ESRD development but not available in this database. Since these variables are not identified as risk factors of HZ, it is reasonable to assume that they are distributed equally in both cohorts so that the possible bias would be lessened.
Conclusion
This is the first population-based cohort study investigating the association between HZ attacks in CKD patients
and subsequent ESRD. CKD patients with HZ had a 1.36-fold higher risk of developing ESRD than the
non-HZ CKD patients and the risk of developing ESRD increased to 8.71 in CKD patients, especially those with HZ and concomitant hypertension and diabetes. Additionally, HZ patients who were more than 64 years old
and who required systemic injection of antiviral therapy are at increased risks of developing ESRD. Further studies are needed to investigate the underlying mechanism of HZ on renal function deterioration.
