Cytomegalovirus Seroprevalence and Titres in Solid Organ Transplant Recipients and Non-transplant Individuals in Seoul, South Korea

Background: Cytomegalovirus (CMV) can cause poor in recipients and The anti-CMV immunoglobulin G (IgG) seroepidemiology is useful for identifying the risk of post-SOT CMV infection or disease as well as immunosenescence or CVD. However, recent CMV seroprevalence and titre have not been fully evaluated with respect to age distribution and comparison between SOT recipients and non-transplant individuals. Methods: We retrospectively retrieved all anti-CMV IgG results from 12658 individuals aged of > one years screened between July 2006 and Nov. 2017 at the Severance Hospital at Seoul. The cohort excluding hematopoietic stem cell transplant recipients and subjects with equivocal values included 2184 SOT recipients and 9738 non-transplant individuals. All IgG results in SOT recipients were measured at pre-transplant period, and only first values among subjects with repeated tests were included in the analyses. Results: The overall IgG seroprevalence and titres were significantly higher among SOT recipients relative to non-transplant individuals (98.7% vs. 94.8%, p < 0.001 and 64.7 ± 44.3 vs. 54.7 ± 38.9 arbitrary units aU/mL, p < 0.001, respectively). The lowest seropositive rates were observed among both SOT recipients and non-transplant individuals aged 11– 15 years (70.6% and 70.2%, respectively). The frequency of seropositivity among adults aged ≥ 41 years increased to ≥ 99% in both groups. Age and SOT group were independently associated factors with higher CMV seroprevalence (OR, 1.12, 95% CI, 1.11-1.12, p < 0.001 and OR 2.43, 95% CI, 1.65-3.59, p < 0.001, respectively). Although liver (99.7%, 80.1 ± 54.5 aU/mL) and lung (100%, 84.3 ± 68.3 aU/mL) transplants had significantly higher seropositive rates and titres compared to kidney (98.3%, 59.4 ± 36.1 aU/mL) and heart (97.0%, 58.1 ± 46.9 aU/mL) transplant (overall p = 0.024 and < 0.001, respectively), the different transplant organs did not affect the seroprevalence in multivariate regression analysis. Conclusions: CMV seropositivity was lowest among school-aged children. IgG testing revealed that most adult SOT recipients in South Korea faced an intermediate serostatus risk of post-transplant CMV infection and disease.


Introduction
The genomic DNA of cytomegalovirus (CMV) is incorporated into the host chromosome after primary infection (usually during early life) and is retained throughout life in a latent status [1][2][3][4]. Various mechanisms can induce temporary or sustained CMV replication, leading to acute cytolytic inflammation [1,5,6]. In both immunocompetent and immunocompromised patients, reactivation of latent CMV can result in direct tissue damage that causes end-organ diseases [1,[7][8][9][10]. The indirect immunomodulatory effects by CMV may cause serious problems such as increased mortality, graft dysfunction or failure, and rejection in recipients of highly immunosuppressive solid organ transplantation (SOT) and hematopoietic stem cell transplantation (HSCT) [8,11,12]. The CMVspecific immune activation in the non-immunocompromised general population could predispose individuals to chronic inflammatory diseases such as cardiovascular diseases and diabetes [13][14][15].
Therefore, surveillance for CMV IgG seropositivity may be epidemiologically important as a means of predicting CMV reactivation and various related chronic diseases in this era of population aging.
The CMV serostatus, which is assessed using an anti-CMV immunoglobulin G (IgG) test, is an important pre-transplant analysis used to predict the risk of post-transplant CMV infection and disease [12,16]. The anti-CMV IgG positivity rate has varied widely by era, geographic distribution, age group, and population characteristics [18][19][20][21]. Therefore, in CMV vaccine-absent era, the regular analyses of CMV seroprevalence by age group in the general population may be needed to obtain information about the epidemiology of CMV, which has various long-term effects on public health. However, no recent report has evaluated CMV seroepidemiology across all ages in South Korea, which has experienced rapid socioeconomic development and advances in public health. In addition, the patterns of pre-transplant CMV IgG seroprevalence and titres in SOT recipients have not been compared to those of non-transplant individuals to identify differences, particularly according to age. Therefore, this study aimed to evaluate the recent CMV serostatus distributions among SOT recipients and non-transplant individuals.

Study design and population
underwent this screening irrespective of the testing purpose between July 2006 and November 2017 from the electric medical records of Severance Hospital, a university-affiliated tertiary care centre in Seoul, South Korea. We excluded the 488 infants under one year of age, because infants could have maternal anti-CMV IgG for a year [22]. The cohort with total 11922 individuals after further exclusion of 34 subjects with equivocal result and 702 HSCT recipients was stratified into two groups of 2184 SOT recipients and 9738 non-transplant individuals (5982 out-patients and 3756 in-patients at CMV IgG measurements) who did not receive any transplantation (SOT or HSCT). Our cohort did not have the SOT recipients who received HSCT before and after SOT or human immunodeficiency virusinfected individuals. If the subject had the repeated anti-CMV IgG tests, we only included the first value in the analyses. For all of SOT recipients, anti-CMV IgG values were evaluated at the pretransplantation within three months of transplantation. The ages at the time of first testing and anti-CMV IgG titres were also recorded. Ages were grouped in five-year intervals up to 40 years and then in 20-year intervals. For these patients, the age, sex and result/titre at the time of the first test were included in the analysis. The study was approved by our institution review board, which waived the requirement for informed consent.

Measurement of anti-human cytomegalovirus IgG antibody
Serum samples were automatically subjected to quantitative anti-CMV IgG testing using an enzymelinked fluorescent immunoassay (VIDAS®, bioMérieux Cop., Marcy-l'Etoile, Lyon, France). Antigens from the CMV AD169 laboratory strain were used as the positive control. The antibody titre was expressed in arbitrary units (aU)/mL. Qualitatively, the results were reported as positive, equivocal and negative if the titres were ≥ 6, 4 to < 6 and < 4 aU/mL, respectively. The results of anti-CMV IgG tests in the analyses were just categorised as positive and negative.

Statistical analyses
We used the independent two-sample T-test and the chi-square test or Fisher's exact test, respectively, to examine differences in continuous and nominal variables between SOT recipients and non-transplant individuals. The post-hoc analyses in nominal variables were performed by adjusted standardized residuals to control for type I error inflation (adjusted p value). An analysis of variance (ANOVA) with the Bonferroni post-hoc test was used for multiple comparisons of IgG titres between transplant organs. The multivariate logistic regression analyses were performed to identify the effect of age, sex and SOT on CMV seroprevalence. Data are expressed as numbers (percentages) or means ± standard deviations (SD) or odds ratio (OR) (95% confidence interval [CI]). SAS version 9.3 (SAS Institute Inc., Cary, NC, USA) was used for the statistical analyses. A two-tailed p value of ≤ 0.05 was considered statistically significant.

CMV IgG results from SOT recipients and non-SOT individuals
The overall anti-CMV IgG positivity rate was 95.5% in total subjects. The CMV seropositivity of male  Table 1). The adult SOT recipients aged > 18 years had significantly lower seronegative rate than non-transplant individuals aged > 18 years (0.8% vs. 3.7%, p < 0.001), but the seroprevalence was similar between two groups aged ≤ 18 years (Table 1).

Regression analyses for the effect of age and SOT on CMV seropositivity
The older age (OR: 1.12, 95% CI: 1.11-1.12, p < 0.001) and SOT (OR: 2.43, 95% CI: 1.65-3.59, p < 0.001) were independent factors associated with higher CMV IgG positive rate. However, sex did not have the independent effect on CMV seroprevalence. Another regression model showed that the transplant organs were not related to CMV seropositive rate (Table 4).

Discussion
Our data demonstrate very high recent CMV seroprevalence (up to 95%) in South Korea, a developed country with a high socioeconomic status and well-organised public health system. As a result, our data may suggest high proportions of both seropositive donors and recipients (D+/R+), which is considered a usual intermediate serostatus risk for CMV infection and disease after SOT through reactivation of latent virus [16,23]. Our IgG seropositive rate is higher than 83% (or 88% of upper level in 95% uncertainty interval) of general CMV seroprevalence worldwide in recent metaregression-based estimation [24]. Seronegative people were rare among those aged ≥ 41 years in non-transplant individuals and extremely rare among those aged ≥ 31 years in SOT recipients. In general, CMV seropositive rates were proportional to age except at teenagers. The lowest seroprevalence observed in subjects aged 11-15 years could be attributed to primary acquisition of CMV at adolescents owing to improved hygiene.
In addition, this analyses reveal high CMV IgG titres even in elderly population. A high CMV IgG titre and persistent immune reactivation caused by an inflation in the population of long-lived, nonclassical CMV-specific effector memory CD8+ T lymphocytes have been associated with chronic inflammatory diseases, including atherosclerosis, stroke and coronary artery disease [3,4,14,15,[25][26][27][28]. Therefore, this report showing fairly high seropositivity and IgG titres at old age might suggest further evaluation for prevention of CMV reactivation in specific population regardless of their immunocompromised status, as this approach could reduce the morbidity and mortality associated with inflammatory vascular diseases in rapidly aging societies.
Despite the international distribution of CMV, seropositivity rates around the world vary widely from 18% to 100%, according to geographical location, ethnicity and specific subpopulation features [24,[29][30][31][32]. In recent study of Li et al, stratification of serological profiles by age group revealed a very high IgG positive rate (97%) even among young individuals (0-14 years), in contrast to our data, that may be associated with high rate of breastfeeding or regional hygienic status [17]. The Netherlands study in 2006-2007 reported that non-Western individuals (76.7%) had a considerably higher seroprevalence than did native Dutch and Western individuals (41.5%) [31]. In general, the very high CMV seroprevalence of South Korea was similar to that of World Health Organization Eastern Mediterranean region, rather than European region or region of the Americas [24,29,31,32]. The different breastfeeding rates and CMV IgM or IgG seropositive rates at women of reproductive age could attribute to various CMV seroprevalence between countries or regions, because mother-toinfant CMV transmission may have a major impact on global epidemiology of CMV [24,33,34].
In a SOT setting reported in Hungary, living organ donors were found to have a CMV seroprevalence of 85% [35]. However, a detailed analysis of the pre-transplant CMV IgG seropositivity rates and titres among SOT recipients according to age groups or transplant organs as well as contrast to nontransplant individuals had not been fully reported. In our study, adult SOT recipients aged ≥ 18 years had a higher IgG seropositive rate and titres relative to non-SOT individuals. The SOT was the independent factor associated with high CMV seroprevalence in multivariate regression analysis. It would be informative to reveal that patients received SOT in end-stage organ disease state have the higher CMV IgG seropositive rates and titres compared to non-transplant individuals. This finding might have any association with asymptomatically intermittent CMV replication and immune boosting against CMV by chronic end-stage medical diseases as well as polyclonal immune activation through pro-inflammatory cytokine release by fulminant acute organ failure leading to SOT [36][37][38][39].
Characteristically, the sub-analyses by transplant organs showed LT and LTx recipients have higher CMV IgG seroprevalence and titres compared to KT and HT recipients. But, multivariate regression model did not show the independent effect of transplant organs on higher seroprevalence. The older mean age in LT and LTx recipients seems to have a relation to their higher CMV IgG seropositive rates and titres.
This study was mainly limited by the retrospective large-scale data extraction. Accordingly, we could not perform further subgroup analyses according to healthy individuals, various underlying morbidities, and other immunosuppressive conditions except SOT and HSCT in non-transplant group.
In addition, the purpose of CMV IgG measurement would be heterogeneous and was hard to obtain this information. But, 38.6% of in-patient test among non-transplant individuals may suggest that the substantial portion of non-transplant group did not receive CMV IgG test for routine health check-up without any illness. Therefore, our non-transplant group may be relevant to compare with IgG values of SOT recipients. In addition, this design did not allow the serial testing of anti-CMV IgG titres in all individuals with initial negative results. However, this study is the first to report a difference in the qualitative and quantitative anti-CMV IgG findings between SOT recipients and non-transplant individuals according to serial age groups as well as transplant organs in SOT recipients.

Conclusion
The overall CMV seropositive rate in South Korea was as high as 95% among both SOT recipients and non-transplant individuals, and the frequency of seronegativity decreased to less than one percentage among adults aged ≥ 41 years. Furthermore, compared to non-transplant individuals, SOT  Data are expressed as mean ± standard deviation. * † ‡ §p-value < 0.05 between two groups in posthoc tests by Bonferroni correction after ANOVA. aadjusted p < 0.05 between two groups using posthoc analyses based on adjusted standardized residuals. Aberrations: aU, arbitrary unit; CMV, cytomegalovirus; IgG, immunoglobulin G