Association Between Cytomegalovirus Disease and Dementia: A Population-Based Cohort Study

Background The association between human cytomegalovirus (CMV) and dementia remains controversial. Previous studies have suggested that CMV serostatus assessed by serum immunoglobulin G plays a role in neurodegeneration with cognitive impairment. We aimed to evaluate the association between CMV tissue-invasive diseases and vascular dementia (VD) and Alzheimer’s disease (AD). Methods The International Classification of Diseases (ICD) 10th revision codes from the National Health Insurance Database covering entire people in Republic of Korea were used to classify patients: case group (n = 687), age ≥40 years with CMV tissue-invasive end-organ disease and control group (n = 3,435), without CMV disease, matched by age and sex (ratio, 1:5). The case and control subjects selected during 2010–2014 with a washout period of the previous 4 years were followed-up until December 2016. of Chlamydia increased high-sensitive C-13


Abstract Background
The association between human cytomegalovirus (CMV) and dementia remains controversial.
Previous studies have suggested that CMV serostatus assessed by serum immunoglobulin G plays a role in neurodegeneration with cognitive impairment. We aimed to evaluate the association between CMV tissue-invasive diseases and vascular dementia (VD) and Alzheimer's disease (AD).

Methods
The International Classification of Diseases (ICD) 10th revision codes from the National Health Insurance Database covering entire people in Republic of Korea were used to classify patients: case group (n = 687), age ≥40 years with CMV tissue-invasive end-organ disease and control group (n = 3,435), without CMV disease, matched by age and sex (ratio, 1:5). The case and control subjects selected during 2010-2014 with a washout period of the previous 4 years were followed-up until December 2016.
Conclusions CMV disease may be associated with the risk of VD.

Background
Dementia is a neurologic disease defined as a gradual decline in memory and cognitive function [1].
The worldwide dementia burden, estimated to be 47 million cases in 2015, is expected to double every 20 years [1,2]. Although the aetiology and pathogenesis of dementia are not yet clear, it could be caused by brain damage due to vascular ischemia and hereditary, sporadic, or age-related factors.
Moreover, the hypothesis that infection may be a risk factor for Alzheimer's disease (AD) has been supported by neuroinflammation and disease pathology [3][4][5]. In vascular dementia (VD), the second most common cause of dementia, neurogenic inflammation due to vascular damage could develop into cognitive dysfunction [6][7][8].
Past ubiquitous human cytomegalovirus (CMV) infection can present in several forms, from life-long asymptomatic latency to tissue-invasive end-organ disease [9]. Because CMV is associated with several chronic inflammatory diseases, particularly vascular disorders and immunodysfunction, an association between CMV and dementia has been suggested [8,[10][11][12]. CMV is associated with AD development and declining global cognition [10,11]. CMV deoxyribonucleic acid was more dominant in the brain tissue of case group subjects (with VD) than in the control group subjects [12]. However, several studies on CMV and dementia were based on seroprevalence or the serostatus of latent CMV, assessed using serum CMV immunoglobulin G (IgG). Serum CMV IgG was suggested to play a role in neurodegeneration with cognitive impairment [8,10,11], but a significant association between CMV and dementia was not observed in a recent meta-analysis [13,14]. The Republic of Korea has high rates of CMV seropositivity [15][16][17]. The Republic of Korea strictly applies the registration of CMV tissue-invasive disease diagnosis using the national health insurance system. Therefore, a confirmation of the association between dementia and CMV tissue-invasive end-organ diseases would indicate active replication of the whole virus in patients, not latent past infection status; this can provide clearer evidence than that from previous studies based on the known seroprevalence. This study aimed to evaluate the association between CMV tissue-invasive end-organ diseases and type of dementia.

Extraction of nationwide population data
In the Republic of Korea, the national health insurance service involves mandatory registration for the entire population. These compulsory subscribers pay for health insurance based on their income level. All medical institution present claims through the diseases diagnosed using the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) codes [18].
Rare incurable disorders (RIDs), such as human immunodeficiency virus infection, other specific viral diseases including CMV disease, and cancer, are considered special disorders by the Korean National Health Insurance Service [19]. Patients with RID receive significant health coverage by paying 5% of their healthcare costs. Therefore, the process of diagnosing RID and receiving medical claims should be based on strict and accurate data. The Korean Health Insurance Review and Assessment (HIRA) plays an important inspector role in this process. We used a medical dataset extracted from the National Health Insurance Database (NHID), which was submitted to the HIRA. The NHID includes big data consisting of 1.3 trillion records with information regarding medical diagnosis, treatment results, long-term care insurance for elderly patients, registration information on cancer or RIDs, and status of medical institutions [20,21]. Health information-related data of the total population of the Republic of Korea were collected. This study was approved by the Institutional Review Board of the Gangnam Severance Hospital, Yonsei University College of Medicine, with the waiver of the informed consent as well as the relevant permission forms have been obtained from the NHI Sharing Service.

Study population and design
From January 2010 to December 2014, data of 1,557 patients diagnosed with CMV tissue-invasive end-organ diseases and enrolled with RID were extracted from the NHID. A 4-year washout period was considered (between January 2006 and December 2009) to evaluate the effect of CMV on new-onset dementia. Forty-five patients diagnosed with dementia during the washout period were excluded.
Among the remaining 1,512, 687 with CMV tissue-invasive diseases and age ≥40 years were selected for the case group with CMV disease. For the case-control study, 3,435 age-and sex-matched controls with no CMV tissue-invasive diseases were selected at a 1:5 ratio. Participants from both groups were followed-up until December 2016. The incidence rate (IR) per 1,000 individuals was calculated by dividing the number of events (new-onset dementia) with the total follow-up duration.

Definition
The criterion for diagnosing CMV disease as RID is 'tissue invasion'. The diagnosis of CMV tissueinvasive end-organ diseases can be established if the following are observed: (1) histopathologic features during immunohistochemical staining for CMV and the presence of inclusion bodies, (2) pp65 antigen measurement, and (3) detection and quantification of CMV or CMV-related genes in body fluids or tissue using the antigenemia test, culture test, or nucleic acid amplification test.
CMV tissue-invasive diseases are given a unique V104 code for RID registration. The V104 code is consistent with the specific B25 codes in the ICD-10 provided by the World Health Organization, including all types of CMV tissue-invasive end-organ diseases, such as cytomegaloviral pneumonitis

Statistical analysis
Categorical and continuous data were presented as numbers (percent) and mean ± standard deviation, respectively. Matched groups were compared using the McNemar and paired t-tests.
Kaplan-Meier curves adjusted for age and sex, were established to analyse the incidence and probability of dementia according to the presence of CMV diseases. Multivariate logistic regression analyses using model 1 (M1, non-adjusted), model 2 (M2, adjusted for age and sex), and model 3 (M3, adjusted for age, sex, low-income status, NIDDM, hypertension, and dyslipidaemia) were performed to evaluate the impact of CMV diseases on dementia development. Statistical analyses were performed using the Statistical Analysis System (SAS) program (version 9.2; SAS Institute, Cary, NC). Two-tailed P values<0.05 were considered significant.

Effect of CMV disease on dementia development
The case group (14.8/1,000 individuals) had twofold higher IRs for all types of dementia than the control group (7.5/1,000) per M1 with an odds ratio (OR) of 2.0 (95% confidence intervals [CI], 1.4-2.9), per M2 with an OR of 2.1 (95% CI, 1.5-3.1), and per M3 with an OR of 2.0 (95% CI, 1.4-3.0). In the subgroup analysis based on dementia type, the IR of AD in the case group (9.0/1,000) was 1.7 times higher than that in the control group (5.5/1,000) with an OR of 1.7 (95% CI, 1.1-2.8) in M2. The IR of VD in the case group (2.7/1,000) was up to threefold higher than that in the control group (0.9/1,000) with an OR of 3.3 (95% CI, 1.3-8.4) in M2. The IR of other dementia except AD and VD in the case group (3.1/1,000) was 2.8 times higher than that in the control group (1.1/1,000) ( Table 2).

Incidence probability of new-onset dementia
During the 7-year follow-up, the incidence probability of all types of dementia (P<0.001) was significantly higher in the case group ( Figure 1A). The incidence probabilities of AD (P=0.034) and VD (P=0.011) were also significantly higher in the case group ( Figure 1B, 1C). The slope of the Kaplan-Meier curves indicates that the incidence probability of AD was fairly different between the groups 5 years after follow-up initiation. Contrastingly, VD mostly occurred within 3 years of CMV tissueinvasive end-organ disease diagnosis ( Figure 1B, 1C).

Effect of CMV diseases on dementia according to sex and age
The IRs of all dementia types in the case and control groups were not significantly different between men and women (OR in M2, 2.

Discussion
This study showed that the new-onset incidence of all types of dementia and each type of AD or VD were significantly higher in subjects with CMV tissue-invasive end-organ disease. After adjusting for age, sex, low-income status, NIDDM, hypertension, and dyslipidaemia which can be the risk factors of dementia [5,6], patients diagnosed with CMV disease had a higher risk for VD than for AD (OR, 3.3 vs. 1.7). Previous studies suggested that Helicobacter pylori, Chlamydia pneumoniae, and human Tlymphotropic virus-1 infection could be associated with VD [22][23][24]. The chronic infection of murine CMV (MCMV) in blood vessel could progress atherosclerotic plaque, and then VD could be induced by atherosclerosis pathogenesis in mouse model [25]. However, it is difficult to simulate the pathogenesis of HCMV disease in a MCMV infection mouse model because human CMV and MCMV are genetically different, and mice could not be infected with HCMV [26]. Even though the relevant infectious agents of VD have not been studied more than those of AD, the stronger association of CMV disease with VD than AD in this study will be interesting finding. This result suggests that CMV may play a possible role in complex VD pathogenesis because active recurrent intermittent CMV replication and/or latent CMV infection status is associated with chronic inflammatory atherosclerosis, vascular damage, and ultimately, cerebrovascular/cardiovascular diseases (CVDs) [27,28].
Apolipoprotein E (APOE), produced in the kidney, liver, and brain, is involved in lipid transport and uptake [3,29,30]. The ε4 allele of APOE, a genetic risk factor for AD and CVD, induces oxidative damage in the central nervous system in herpes simplex virus infection. Neurotoxicity and neuroinflammation by APOE ultimately results in neurodegeneration and dementia [3,30,31]. When APOE ε4 is expressed, β-amyloid is aggregated in the brain, and amyloid plaques can cause cerebral amyloid angiopathy [32]. CMV and β-herpesviridae might also be associated with APOE. Further studies are required to evaluate the expression of the APOE genotype in dementia patients with CMV tissue-invasive disease. Generally, persistent systemic infection and inflammation can cause chronic neurodegeneration. Particularly, decreased CD8+ T lymphocyte counts and cognitive dysfunction due to CMV infection are characteristic features of AD [33][34][35].
Because AD is also associated with aging, age matching was essential for determining the association between the case and control groups. The IR of AD in the case group was higher in those aged 40-59 years than in those aged ≥60 years. Early-onset dementia in individuals aged <65 years was associated with head trauma, alcohol abuse, and human immunodeficiency virus-1 infection [36]. The association between CMV serostatus or disease and new-onset dementia has not been studied yet.
However, one study revealed that cognitive impairment was six-times higher in patients with three risk factors of low educational level, APOE ε4 allele, and Herpesviridae including CMV infection [37].
Moreover, the APOE ε4 allele was associated with new-onset AD or a decline in cognition in the earlier clinical stages [38][39][40][41].

Consent for publication: Not applicable.
Availability of data and materials: The authors are not allowed to share the analysis datasets of the current study due to data regulations.

Competing interests:
The authors declare that they have no competing interests.