Elsevier

Vaccine

Volume 28, Issue 22, 14 May 2010, Pages 3773-3777
Vaccine

Review
The effect of diabetes mellitus on immunological response to hepatitis B virus vaccine in individuals with chronic kidney disease: A meta-analysis of current literature

https://doi.org/10.1016/j.vaccine.2010.03.038Get rights and content

Abstract

Background

Patients with chronic kidney disease (CKD) often fail to produce protective antibodies to hepatitis B virus (HBV) surface antigen after vaccination. Diabetes mellitus (DM) is the most common cause of CKD; however it is not clear whether it affects immunological response to HBV vaccine in these patients.

Aims

We aimed to evaluate the immunological response to HBV vaccine in diabetic patients with CKD by conducting a meta-analysis of the current literature.

Methods

Only studies that evaluated the seroprotection rate for diabetic against non-diabetic CKD patients or the immunological response of these groups to HBV vaccine were included. We applied the random effects model of DerSimonian and Laird, with heterogeneity (Q statistic), publication bias (Egger and Begg test) and sensitivity analyses. The rate of patients showing seroprotective anti-HBsAg titers (>10 IU/mL) at completion of HBV vaccination schedule in the diabetic versus the non-diabetic CKD patients was set as our end-point of interest.

Results

We identified seven studies that fulfilled our inclusion criteria involving 15,073 unique patients with CKD. Aggregation of study results showed a significant decrease in response rates among the diabetic versus the non-diabetic patients [pooled odds ratio = 0.58 (95% CI 0.37–0.89), Q(6) = 11.3, I2 = 50%]. The P-value was 0.07 for our test of heterogeneity.

Conclusions

Our meta-analysis determined that HBV vaccination's seroprotection rate in diabetic CKD patients is significantly lower than that in non-diabetic CKD patients. Therefore, using vaccine adjuvants such as oral levamisole, granulocyte macrophage-colony stimulating factor or intradermal injection might be advisable in these patients.

Introduction

Chronic kidney disease (CKD) patients, including end stage renal disease patients and those who are on maintenance dialysis, are highly vulnerable to be infected with blood born viruses such as hepatitis B virus (HBV). To avert HBV infection in these patients, vaccination is the most effective primary preventive strategy [1], [2], [3]. Nonetheless, we well know that the compromised immune system in these patients results in poor and non-persistent immunological responses to HBV vaccination in comparison with individuals with intact renal function [4], [5], [6]. Today, diabetic nephropathy is the most common cause of CKD in both developing and industrialized world [7]. It is well known that diabetic patients have a compromised immune system, and their immunological response to HBV vaccine is less optimal than non-diabetic individuals; nevertheless, the influence of this metabolic disease on seroconversion after HBV vaccination is not well investigated in CKD patients and only a few studies on HBV vaccination of the renal disease patients with diabetes mellitus (DM) has been reported [8]. Therefore, in this review, we aimed to investigate whether the suboptimal response to HBV vaccination in CKD patients was further exacerbated by DM.

Section snippets

Search strategy and data extraction

We conducted searches of MEDLINE, Scopus, ISI and Cochran Central Register of Clinical Trials from 1995 to 2009 and confined these to English language studies to identify the relevant literature. The keywords we used were different combinations of “Hepatitis B vaccine” or “HBV vaccine” with the following terms: “ESRD”; “renal failure”; “dialysis” or “hemodialysis”; “chronic kidney disease” or “CKD”. Data were extracted by a single investigator; rechecked twice; and entered in to Excel

Criteria for inclusion

We included both prospective and retrospective peer reviewed published studies comparing the response rate in diabetic CKD patients (study group) vs. non-diabetic CKD patients (control group). We also included studies that had reported response rates in both diabetic and non-diabetic subjects separately. Studies that recruited dialysis patients were considered still eligible. Trials of the plasma-derived and recombinant DNA HBV vaccine were included. All dose schedules and routes of vaccine

Ineligible studies

Studies that reported inadequate data on measures of response, or included the individuals with positive serology for hepatitis B virus surface antigen (HBsAg), antibodies to HBsAg (HBsAb) and antibodies to hepatitis B virus core antigen (HBcAb) or HIV and patients with concurrent administration of immunosuppressive medicines were excluded.

End-points of interest

We evaluated the serological response to HBV vaccine in study and control groups and compared the rate of patients with protective anti-HBs titers after completion of HBV vaccination schedule in diabetic vs. non-diabetic patients. The level of antibody production that defines seroprotection was 10 IU/mL across the studies.

Statistical methods

We omitted the data from patients who did not complete the vaccination program thus; analysis was made per protocol, not by intention to treat. The pooled odds ratios (OR) and their 95% confidence interval (CI) for seroprotection rate after the completion of vaccine schedule in diabetic vs. non-diabetic patients were computed by using the random effects model according to DerSimonian and Laird method. The Q statistic was used for quantifying the heterogeneity; I2 was used to provide a measure

Results

We reviewed 139 citations and based on their titles and abstracts, we obtained full texts of 59 potentially relevant study reports. Only 8 studies had reported the seroprotection rate after HBV vaccination in diabetic and non-diabetic patients with CKD separately. One study was excluded because it only provided data for subgroup of DMs that caused nephropathy [9]. Only one study had compared diabetic and non-diabetic patients in two different arms. A total of 15,073 individual patients were

Patients’ characteristics

Some demographic and clinical characteristics of the subjects enrolled are presented in Table 1, Table 2, Table 3. All trials were published in English language from 1995 to 2008. Three studies were from the USA and two from Canada [10], [11], [13], [14], [15], one each from Turkey and Taiwan [8], [12]. All patients had CKD. 92% (13,939/15,073) of subjects underwent standard hemodialysis and 8% (1134/15,073) of subjects were receiving medical treatment. Mean age of subject cohort ranged from 50

Summary estimates of outcome

Table 4 represents seroprotection rates in both study and control groups. Fig. 2 lists individual ORs with 95% confidence intervals for seroprotection rates after HBV vaccination (diabetic vs. non-diabetic subjects). The pooled OR was 0.58 (95% CI 0.37–0.89) according to a random effects model. The test for heterogeneity was significant [Q = 11.3 (df = 6), p = 0.07, I2 = 50%]; the publication bias assessment according to Begg's and Egger's formulas was not significant (PBegg = 0.2, PEgger = 0.16). The

Discussion

It is previously determined that in individuals with normal renal function, diabetic patients show a lower seroprotection rate than non-diabetic patients after HBV vaccination [16]. The presence of DR3 and DR7 human leukocyte antigen (HLA) alleles in diabetic patients, and increased tolerance to stimulation and reduced cytokine secretions of peripheral blood mononuclear cells (PBMCs) are proposed as mechanisms that underlie this impaired immunological response [17], [18]. Diabetic nephropathy

Conclusion

Diabetes mellitus significantly decreases the seroprotection rate of HBV vaccine in chronic kidney disease patients. Using vaccine adjuvants such as oral levamisole, granulocyte macrophage-colony stimulating factor or intradermal injection might be advisable in these patients.

Conflict of interest statement: The authors declare that they have no conflicts of interest relevant to the manuscript.

References (22)

  • R. Vanholder et al.

    Influence of uraemia and haemodialysis on host defence and infection

    Nephrol Dial Transplant

    (1996)
  • Cited by (0)

    View full text