In this study, we showed that the seroconversion rate of our dialysis patients was 70.5% after four double doses (40
μg per dose) of the hepatitis B vaccine administered over 6
months. We also found that responders were significantly younger than non-responders (P
0.003). Seroconversion to HBV vaccination and one-month post-vaccination anti-HBs titers did not predict all-cause mortality in our maintenance dialysis patients. After adjusting for age, DM, albumin levels, gender, and dialysis modality, non-seroconversion following HBV vaccination could independently predict mortality due to infection in ESRD dialysis patients. This novel and interesting finding demonstrates the possible relationship between the vaccination response and the specific cause of mortality in ESRD patients.
Our study dialysis subjects displayed a similar suboptimal immune response to patients in previous studies. In ESRD patients, the response rate to HBV vaccination is approximately 50%–80%, which is lower than the 90% response rate in the general population, with advanced age being a significant negative factor associated with this poor immune response [11
]. Numerous factors associated with the impaired immune response following HBV vaccination in ESRD, such as age, albumin levels, HCV infection, and gender, have been well studied [8
]. Current studies have focused on the insufficient acquired immunity due to uremia to account for the suboptimal response rates following HBV vaccination [14
]. Recent studies have proposed that the immunodeficiency in ESRD results from the impaired function of antigen-presenting cells, alteration of toll-like receptors, weakened activation of T lymphocytes, and the imbalance of T-helper lymphocytes (Th 1/Th 2 ratio) [14
In the present study, we demonstrated that DM, age, low albumin levels, and non-response following HBV vaccination could predict mortality due to infection.
Several negative determinants of mortality in ESRD, such as old age, DM, and malnutrition, have been documented [7
]. Immune dysfunction in uremia is also recognized as a determinant of mortality. However, our findings are important because they demonstrate a link between infection-cause mortality and insufficient immune responses to HBV vaccination in ESRD, since these two clinical presentations involve homogenous but varied interactions between the immune, innate immune and the adaptive immune systems. A recent review indicated that in dialysis patients innate immunity plays an important role in fighting infections, while acquired immunity (via antibody production) also makes a significant contribution [14
]. We therefore hypothesize that T cell dysfunction may be the main link between the high mortality due to infection and the insufficient antibody response to vaccination in ESRD patients. Impaired cytotoxic CD8 T cell response limits the fight of the infected host defense system against the pathogens, while the inadequate function of helper CD4 T cells weakens the humoral immune system. The results of studies by Alvarez-lara et al. and Meier P et al. support this hypothesis, finding increased apoptosis of T cells in uremia [17
]. An alternative interpretation is that the key linking factor between high mortality due to infection and the impaired vaccination response lies upstream of T-cell activation, i.e., the pattern of recognition and the antigen presenting cell or secondary dysregulation or activation of cytokines [6
It has been proposed that both the innate and acquired immune systems may play a role in the progression of atherosclerosis, the major pathogenic factor responsible for most deaths in ESRD patients [21
]. Interestingly, we did not find an association between seroconversion and all-cause mortality or between seroconversion and CV-cause mortality in dialysis patients. The correlations among persistent inflammation, CVD, immune dysfunctions, and infections in ESRD have become apparent in recent years [5
Chronic inflammation as a result of hypercytokinemia and an imbalance in the pro-inflammatory to anti-inflammatory ratio has been well recognized in ESRD [25
]. The dysregulation of various cytokines due to the uremia milieu may link inflammation-atherosclerosis to impaired immunity inflammation in dialysis patients [14
]. Several studies have attempted to link the levels and types of cytokines with clinical outcomes. Girndt et al. determined that higher levels of IL-6 and TNF-α correlated with an immunocompromised status, that is, non-seroconversion to HBV vaccinations in HD patients [26
]. Kimmel et al. assessed immune parameters such as cytokines in 230 HD patients, with a three-year follow up. Their results showed that increased levels of IL-6, TNF-α, IL-1, and IL-13 correlated with a higher relative mortality risk. High levels of IL-2, IL-4, IL-5, and IL-12, and T cell functions were associated with survival in HD patients. The authors did not analyze the causes of mortality in their patients or further investigate the associations between the causes of mortality and cytokines. Interestingly, they noted that the patterns of cytokines involved may be more important than the individual cytokine levels in interpreting survival outcomes [27
Among the other studies on cytokines’ ability to predict mortality in ESRD, the study by Badiou et al. measured levels of both pro-inflammatory cytokines and anti-inflammatory cytokines in 134 HD patients and found that IL-6 levels could strongly predict CV mortality. In addition, (IL-4+ IL-6+ IL-10)/(IL-2
IFN-γ) was associated with non-CV mortality. The authors speculated that examination of the Th1/Th2 cytokine relation might be more relevant for predicting non-CV mortality [28
]. These findings indicate that the laboratory immune parameter, especially the Th1/Th2 cytokine ratio that implies defective immunity, may be able to predict clinical outcomes. The clinical immune parameters and the immune response following HBV vaccination may be able to correlate with mortality in ESRD patients.
The study by Fernandez et al. of 64 HD patients noted that low levels of albumin negatively influenced the response to HBV vaccination. In the study’s survival analysis of 31 patients, non-responders had higher mortality and morbidity [29
]. The authors did not adjust for albumin levels or age in the survival analysis of these HD subjects. However, our Cox regression survival analysis of 156 dialysis patients reveals that only old age and low levels of albumin have significantly higher mortality but that non-response following HBV vaccination did not. Since low albumin levels are recognized as an independent factor of mortality in ESRD [30
], we speculate that low albumin levels account for the concomitant impaired immune response following HBV vaccination and increased mortality in the Fernandez et al. study.
The Kimmel et al. study suggests that better T cell function and humoral immunity are associated with a survival advantage in HD patients [27
]. Unlike the predictive value of the cytokine ratio on mortality mentioned above, our study demonstrated no significant relationship between the clinical immune response following HBV vaccination and all-cause mortality. The levels and types of cytokines in dialysis patients were the result of the interacting influences of residual renal function, dialysis adequacy, comorbidity, and current treatment [31
]. Our dialysis patients received HBV vaccinations at the initiation of renal replacement therapy. Some of our study subjects, especially the PD patients, retained some residual renal function, which would affect the immune response to HBV vaccinations. In the Badiou et al. and Kimmel et al. studies, the levels of circulating cytokines (either Th1 or Th2-related) were significantly higher in the HD patients than in the control patients [27
]. The effects of these unusual levels of circulating cytokines may not correlate well with their baseline local immunological functions [31
]. These circulating cytokines may operate in a more sophisticated pathophysiologic network, causing erythropoietin resistance, hypoalbuminemia, frailty, atherosclerosis, and dyslipidemia [31
]. Therefore, the clinical outcomes, survival and immune response following HBV vaccination may be the diverse results of the differing interactions of various cytokines at specific cellular, organ, and system targets. We therefore could not determine whether a direct association exists between seroconversion and all-cause mortality in dialysis patients despite the well-recognized immunodeficiency, hypercytokinemia, and high mortality in ESRD populations [1
We did not check the antibody of the core HBV antigen (anti-HBc). The finding of a positive anti-HBc in the absence of HBs Ag or anti-HBs is relatively uncommon [35
]. Chen et al. found that the responsiveness rates of the hepatitis B vaccination were the same between isolated anti-HBc positive and normal subjects [36
We recognize that our study has certain limitations, such as the relatively small number of dialysis patients included to determine the infection-based and CV mortality. Moreover, infection was the leading cause of mortality among our study subjects, despite being the second leading cause of deaths in the ESRD population as a whole. The immune status of dialysis patients may change after losing residual renal function. The results of our study need to be replicated in prospective studies of using larger populations of chronic, stable dialysis patients.