Int Urol Nephrol (2010) 42:1069–1075 DOI 10.1007/s11255-009-9616-z
NEPHROLOGY - ORIGINAL PAPER
Modulation of the immune response to HBV vaccination by hemodialysis membranes Fatih Dede Æ Ayla Yıldız Æ Deniz Aylı Æ Nujen C ¸ olak Æ Ali Rıza Odabas¸ Æ Hadim Akog˘lu Æ Erdal Eskiog˘lu Æ Adrian Covic
Received: 1 April 2009 / Accepted: 30 June 2009 / Published online: 19 July 2009 Ó Springer Science+Business Media, B.V. 2009
Abstract Introduction Seroconversion response to Hepatitis B virus (HBV) vaccination is limited in uremic patients because of impaired humoral and cellular immune activity. Recent studies show that high-flux (HF) hemodialysis (HD) membranes can improve T cell functions and decrease the proinflammatory cytokine activation more effectively than low-flux (LF) membranes. In regard to HF membranes may have immune modulator effects; we compared the antibody responses to hepatitis B vaccination between HF HD and LF HD membranes. Methods One thousand four hundred and eight patients on HD programmed for 4 h three times a week at three different centers with HF or LF membranes were scanned. Anti-HBs levels of these patients who were vaccinated with recombinant DNA HBV vaccine on the 0, 1st, 2nd and 6th month from F. Dede (&) D. Aylı A. R. Odabas¸ H. Akog˘lu Department of Nephrology, Ankara Numune Training and Research Hospital, Samanpazari/Ankara 06530, Turkey e-mail:
[email protected] A. Yıldız N. C¸olak E. Eskiog˘lu Department of Internal Medicine, Ankara Numune Training and Research Hospital, Ankara, Turkey
the beginning were recorded on the 0, 3rd, 6th, 9th and 12th month of vaccination schedule. Results Seroconversion rate was 84.2% in HF group and 52.7% in LF group on the 6th month (P \ 0.01). Ratio of the patients who had [100 IU/l antibody titers in HF group was 22.8%, while it was 10.9% in LF group (P \ 0.01). Also on the 9th, 12th and 24th month; seroconversion rates in HF group were higher than LF group: 91.1–70.9% (P \ 0.05), 95.0–81.8% (P \ 0.05), 92.1–83.7% (P [ 0.05), respectively. Ratio of the patients who did not show any seroconversion was higher in LF group than HF group; on the 6th month as 15.8–47.3% (P \ 0.001), on the 9th month 8.9–29.1% (P \ 0.05), on the 12th month as 5.0–18.2% (P \ 0.05), and on the 24th month as 7.9– 16.4% (P [ 0.05), respectively. Conclusion We showed that ratio of anti-HBs seroconversion response to hepatitis B vaccination in patients receiving HD with HF membranes was higher than LF membranes. This finding suggests that HF membranes may improve immune modulator effects. Keywords Hemodialysis Hepatitis B virus infection High-flux dialyzer Immune modulator effect Vaccine
Introduction A. Covic Department of Nephrology Clinic and Dialysis and Transplantation Center, ‘‘C. I. PARHON’’ University Hospital, Lasi, Romania
Hepatitis B virus (HBV) infection is still a major clinical problem in patients with ESRD requiring
123
1070
maintenance hemodialysis (HD), and can lead to many serious consequences such as severe hepatic complications and decreased transplantation rates. Increasing compliance to vaccination schedules, strict adherence to universal precaution measures, and routine utilization of erythropoietin are likely the major causes for the decreasing prevalence of HBV infection in HD patients [1]. Although vaccination is the most effective prevention method with a rate of protective antibodies in healthy subjects of 95%, its effect is significantly limited by the impact of ESRD an important decrease to a 50–80% rate of protective antibodies response [2–4]. Impaired humoral and cellular immune responses in HD patients are the most important reasons for the insufficient seroconversion response to HBV vaccination [1, 5]. Recent studies support that high flux (HF) HD membranes can lower all cause mortality rates [6], improve T cell functions [7], reduce oxidative stress, and suppress chronic inflammation by inhibiting the activation of pro-inflammatory cytokines [8], compared to low flux (LF) membranes. We hypothesized that there might be a significant difference between the impact of long-term dialysis with HF membranes vs LF on the seroconversion rates after HBV vaccination. More specifically, we analyzed a large database to test if HF improves the effectiveness of HBV vaccination.
Subjects and methods All hemodialysis patients (n = 1408) from three different centers from 1995 to 2005, were analyzed retrospectively for hepatitis status. Among these patients, 71 female and 85 male subjects whose serologic markers of hepatitis B and C (HBsAg, antiHBc, HBeAg, anti-HBe, anti-HCV) were negative and anti-HBs titers were \10 IU/l, were selected for the study (n = 156). For the selection of membrane; body surface area, the adequacy of dialysis and body mass index criteria were taken into account. One hundred and one study subjects were dialyzed using HF polysulphon 1 m2/m2 body surface area membranes (either Fresenius F60 and F80)—the HF group, while 55 patients were dialyzed using LF 1 m2/m2 body surface area membranes (Fresenius F5, F6, F8)—the LF group, all for 4 h three times a week. All of the study subjects had been on HD with the same type of
123
Int Urol Nephrol (2010) 42:1069–1075
membrane (HF or LF dialyzers) at least for 6 months before the beginning of the vaccination schedule. Their prescribed diet consisted of 1.2 g/kg/day protein, indigent phosphorus, and vitamin B and C complex. All patients received additive medication, such as antihypertensive agents, active vitamin D, essential aminoacid supplements, oral phosphorus binding agents, iron or erythropoietin was given according to their laboratory and clinical follow up. Patients with serum albumin levels \3.5 mg/dl and hemoglobin\9 g/dl; and who had malignancy except skin cancer, who required blood transfusion during vaccination schedule, needed additional HD session because of volume overload, and whose Kt/V \ 1.2 [9, 10] and URR (urea reduction rate) \65 were not included to the study. One hundred and fifty-six patients with negative Hepatitis B and C serologic markers; not having any inadequacy of dialysis and malnutrition existence, and those which were found to be appropriate considering the excluding criteria, were included in the research. Patients who received immunosuppressive treatment were excluded from the study. All the patients completed the vaccination schedule. All 156 subjects of the study were vaccinated with 40 mcg recombinant DNA hepatitis B vaccine (Hepavax-GeneÒinj, GreenCross Vaccine Corp, Berna). The schedule for vaccine administration was: 0, 1st, 2nd, and 6th month. Some patients who had not shown anti-HBs seroconversion were given a single rappel dose of 40 mcg on the 12th and 18th month according to their anti-HBs titers. Serum anti-HBs levels of all subjects that were measured using the microparticle enzyme immunoassay technique (Abbot-Aoxyme system) from samples drawn at 0, 3rd, 6th, 9th, 12th month and on the 24th month. Antibody titers[10 IU/l were considered as achieved seroconversion, 10–100 IU/l as low protection level, and [100 IU/l as high protection level.
Statistical analyses All analyses were done using the SPSS 15.0 package (SPSS Inc., Chicago, IL) for windows. Disparities between categorical variables were analyzed using the Pearson Chi-square test; congruity of the variables was assessed with the Kolmogorov–Smirnov test; the relationship between parameters was examined using the Spearman correlation and linear regression. For all
Int Urol Nephrol (2010) 42:1069–1075
1071
statistical tests, a P value under 0.05 was considered as significant.
Results Patients’ ages varied between 25 and 89 (58.2 ± 14.9 years, median age = 60 years). Mean age of the HF and LF groups was 57.6 ± 4.9 and 59.5 ± 5.0, respectively (P = NS). Gender distribution was similar in the two subgroups: 55.5% males in the HF group, 52.7% males in the LF group. Patients’ characteristics are shown in Table 1. There were not
Table 1 Baseline characteristics High flux
Low flux
n
%
n
%
56
55.5
29
52.7
45
44.5
26
47.3
\40
13
12.9
5
9
40–49
17
16.8
7
12.7
50–59
20
19.8
15
27.3
C60
51
50.5
28
51
Mean ± SD
57.6 ± 14.9
59.5 ± 15.0
DM
24
23.8
18
HT
11
10.9
3
5.5
CGN
8
7.9
2
3.6
ID
51
50.5
25
45.6
NTH
2
1.9
5
9
CTIN
5
5
2
3.6
Sex Male Female Age
Etiology
Vascular access AVF
32.7
81
80.2
45
81.8
AVG
5
4.9
3
5.5
TCC
15
14.9
7
12.7
Dialysis period (year) \5
42
41.6
35
63.6
C5
59
58.4
20
36.4
Mean ± SD
6.6 ± 4.6
4.0 ± 3.0
AVF arterio venoz fistula, AVG arterio venoz graft, TCC tunneled cuffed catheter, CGN chronic glomerulonephritis, CTIN chronic tubulo-interstitial nephritis, DM diabetes mellitus, HT hypertension, ID idiopathic, NTH nephrolithiasis, SD standard deviation
any significant differences between the two groups in biochemical parameters or dialysis efficiency (Table 2). Anti-HBs seroconversion occurred in none of the patients at the 3rd month following vaccination. Seroconversion rates at the 6th, 9th and 12th month were 73.1, 84.0 and 90.4%, respectively. Higher titers of antibody development were seen on the 9th month for all patients with a ratio of 40.4% and followed by the 12th and 24th month as 45.5 and 41.6%, respectively. The ratio of the patients in whom seroconversion had not occurred was the lowest (9.6%) on the 12th month. These findings were concluded in Table 3. Anti-HBs seroconversion rate in the HF group was 84.2% at the 6th month while it was only 52.7% in the LF group (P \ 0.01). The rate of antibody response over 100 IU/l was 22.8% in the HF group, and 10.9% in LF group (P \ 0.01) (Table 4; Fig. 1). Statistically significant differences among the groups were maintained at the 9th and 12th month. Anti-HBs titers over 100 IU/l were seen in 41.6% of the HF patients and 38.2% of the LF patients (P \ 0.05) on the 9th month, and in 49.5 and 38.2% patients of the HF and LF groups, respectively on the 12th month (P \ 0.05) (Table 4; Fig. 1). Anti-HBs seroconversion was not seen in 29.1% of LF patients and 8.9% of HF patients (P \ 0.05) on the 9th month; also in 18.2% of LF patients and 5% of HF patients on the 12th month (P \ 0.05). According to our records, none of the patients had adverse reactions or complications due o vaccination.
Discussion Our results showed that usage of HF HD membranes improves the effects of HBV vaccination with earlier and higher titer antibody response. Our results thus support previous findings related to a positive immunomodulating effect of these membranes [7, 8]. Compared to other studies, anti-HBs seroconversion rates in our study were better for all patients on the 12th and 24th month from the beginning of the vaccination schedule at 90.4 and 83.2%, respectively [2–4]. Similarly to recent studies in which the frequency of patients who could not achieve seroconversion on the 12th and 24th month was 5–26% [2], we found a proportion of 16.8%.
123
1072
Int Urol Nephrol (2010) 42:1069–1075
Table 2 Distribution of some parameters of the patients in the high flux hemodialysis and low flux hemodialysis groups High flux Mean ± SD
Low flux Mean ± SD
3rd month
10.7 ± 1.4
10.9 ± 1.4
6th month
11.4 ± 1.4
10.9 ± 1.6
9th month
10.7 ± 1.5
10.7 ± 1.6
12th month
10.8 ± 0.5
11.4 ± 1.1
24th month
11.2 ± 1.4
11.4 ± 1.3
3rd month
32.4 ± 4.2
32.9 ± 4.5
6th month
33.7 ± 4.4
32.7 ± 4.6
9th month
31.6 ± 3.8
31.5 ± 1.3
12th month
32.1 ± 5.2
33.5 ± 3.7
24th month
31.5 ± 4.2
33.0 ± 3.8
3rd month
42.9 ± 12.7
41.6 ± 11.3
6th month
41.6 ± 12.9
41.1 ± 13.7
9th month
45.5 ± 9.1
44.0 ± 13.6
12th month
51.3 ± 6.6
43.9 ± 13.0
24th month
49.9 ± 6.5
49.9 ± 11.5
3rd month
154.7 ± 44.9
133.5 ± 30.3
6th month
170.0 ± 28.3
192.1 ± 36.2
9th month
110.0 ± 85.4
179.2 ± 39.5
12th month
215.7 ± 36.2
179.2 ± 26.8
24th month
200.4 ± 45.9
198.8 ± 37.4
Hemoglobin (g/dl)
Hematocrit (%)
Ca 9 P
Pth (pmol/l)
Albumin (g/l) 3rd month
39 ± 3.2
38 ± 2.8
6th month
40 ± 4.4
39 ± 3.7
9th month
39 ± 3.8
39 ± 3.8
12th month
40 ± 3.9
38 ± 2.9
24th month
41 ± 4.2
40 ± 4.1
3rd month
1.3 ± 0.1
1.3 ± 0.1
6th month
1.4 ± 0.2
1.4 ± 0.2
9th month
1.5 ± 0.2
1.4 ± 0.2
12th month
1.4 ± 0.2
1.3 ± 0.1
24th month
1.4 ± 0.2
1.4 ± 0.2
3rd month
78.0 ± 6.3
73.1 ± 3.4
6th month
72.9 ± 6.4
73.2 ± 3.3
9th month
76.2 ± 6.0
76.9 ± 5.2
12th month
75.8 ± 5.8
75.0 ± 4.6
24th month
74.4 ± 5.8
73.4 ± 7.1
Kt/V#
URR
Ca Calcium, P Phosphorus, Pth Parathormone, URR Urea reduction rate #
[9, 10]
123
Duration of HD may cause accumulation of middle and high molecular weight toxins and loss of residual renal functions necessary for their clearance [6]. In our study, although HD duration in HF patients was longer than LF patients’ (6.6 ± 4.6 and 4.0 ± 3.0 months, respectively), anti-HBs seroconversion rates on the 6th, 9th, and 12th month were higher. Additionally on the same months, the ratio of the high protective antibody response was higher in the HF group. Antibody response in the 24th month was observed to be higher in HF group. However, when the seroconversion ratios of the two groups were compared for the 24th month, the rate of patients who received dialysis with HF membrane and observed to develop antibody response to the vaccine was higher than the group with LF membrane. But this difference was not statistically significant. The disappearance of the difference in favor of HF membrane in earlier months can be related to the weakening (attenuation) in the humoral and cellular immune responses which develop with the increase in the duration of dialysis; or it can be linked to the decrease in patients’ number because of the discontinuations of 55 patients to the study at the end of 24th month. Despite this, the rate of patients who received dialysis with high-flux membrane and observed to develop antibody response to the vaccine was higher than the group with low-flux membrane. The immune response to HBV infection depends on T-lymphocytes and a low/lack of response to HBV vaccination is a sign of immune deficiency. The cellular immune response to HBV vaccine is defective in HD patients possibly due to a reduction in CD4(?) T-cell and T-cell receptor numbers; decreased production of T-cell derived interleukins (IL-2 and IL-4) and interferon gamma; impaired functions of monocytes presenting antigen to T cells [11–13]. Uremia and HD are also associated with proinflammatory cytokines activation, which play an important role in the immune response to vaccination. Proinflammatory cytokines, such as interleukin-6 and tumor necrosis factor alpha, are some of the causes of an impaired cellular immune response [14, 15]. Recent studies associated the inadequate immune response to vaccination seen in HD patients with male gender, advanced age, malnourishment and low body weight, blood transfusions, existence of antiHCV antibodies and some haplotypes of the major histocompatibility complex inhibiting the effective
Int Urol Nephrol (2010) 42:1069–1075
1073
Table 3 Anti-HBs titers according to the months after the beginning of vaccination programmed 3rd month (n = 156)
6th month (n = 156)
9th month (n = 156)
12th month (n = 156)
24th month (n = 101)a
Mean ± SD
0.4 ± 1.4
107.8 ± 237.2
248.4 ± 348.7
257.3 ± 333.2
210.9 ± 289.4
Median
0
23.3
76.1
96
81
Min–max
0–9
0–1000
0–1000
0–1000
0–1000
Seroconversion (%) \10 IU/l
100.0
26.9
16.0
9.6
16.8
10–100 IU/l
0.0
54.5
43.6
44.9
41.6
[100 IU/l
0.0
18.6
40.4
45.5
41.6
Total seroconversion
0.0
73.1
84
90.4
83.2
SD standard deviation, Min minimum, Max maximum Anti-HBs titers \10 IU/l: no seroconversion; anti-HBs titers 10–100 IU/l: low protection; anti-HBs titers [100 IU/l: high protection a
Anti-HBs titers on the 24th month of 55 patients were unavailable
Table 4 Seroconversion rates according to high-flux and lowflux hemodialysis membranes
60
Seroconversion (%) \10
P 10–100 [100
n
%
3rd
156
100
High-flux
101
64.7
100.0
–
–
Low-flux
55
35.3
100.0
–
–
6th month
156
100
High-flux
101
64.7
15.8
61.4
22.8
Low-flux
55
35.3
47.3
41.8
10.9
9th month
156
100
High-flux
101
64.7
8.9
49.5
41.6
Low-flux
55
35.3
29.1
32.7
38.2
\0.05
5.0 18.2
45.5 43.6
49.5 38.2
\0.05
12th month
156
100
High-flux Low-flux
101 55
64.7 35.3
24th month
a
a
Anti HBs titers
50
low flux
101
100
77
76.2
7.9
27.7
64.4
Low-flux
24
23.8
16.4
25.5
58.2
38,2
40 30 20
#
22,8
10,9
10 0
High-flux
49,5
high flux
6. month
12. month
membrane
\0.01
[0.05
a
Anti-HBs titers on the 24th month of 55 patients were unavailable
presentation of HBV antigen to T-lymphocytes [1, 5, 16]. The effects of HD membranes on the clearance of uremic toxins and inhibition of inflammatory activation may influence the antibody response to HBV vaccination in HD patients. In one study, it was observed that the immune functions improved in a group of HD patients when dialysis efficiency and the
Fig. 1 Comparisons of the anti-HBs titers of the hemodialysis patients in the high-flux group and low-flux group on the 6th and 12th month after beginning of vaccination programmed (#: P \ 0.01, .: P \ 0.05)
Kt/V value increased to 1.16 from 0.96 (which in reality means insufficient hemodialysis) [12]. HF HD membranes remove uremic toxins more effectively [17]. In the HEMO study, HF membranes were established to reduce both cardiac and all causes of mortality rates compared to LF membranes [6]. Activity of interferon gamma, which is an essential molecule for macrophage activation and cellular immune response, were shown to recover by using HF membranes in HD [7]. Another study comparing HF to LF membranes revealed that neutrophil functions improved more in the HF group [18]. HF HD membranes were found to have positive influences on apoptosis induced by proinflammatory cytokines resulting from increased oxidative stress in uremic patients [19]. Similarly, use of high biocompatible
123
1074
membrane equipments and sterile dialysate lowers inflammatory activation [20, 21]. Patients receiving HD with ultrapure dialysates achieved higher antibody titers and seroconversion rates to following HBV vaccination, compared to patients receiving HD with conventional dialysate [22]. These findings support that HF HD membranes may have positive immune modulator effects on the immune system deficiency by removing intermediate molecular weight ureic toxins, improving T-cell functions, reducing oxidative stress, inhibiting pro-inflammatory cytokine activation, and chronic inflammation. Some limitations of our study should be discussed. Because the study is retrospective and nonrandomized, this situation may prevent any conclusion about causality. Moreover, as it was a retrospective study, cytokines reflecting the inflammatory process and oxidative stress parameters, both of which is led by IL showing T lymphocyte functions could not be included in the study. Evaluation of these markers in dialysate and in blood can contribute to further explain these findings. We propose further prospective randomized studies with larger sample size and long term follow up contribute immune modulator effects of HF HD membranes. In conclusion, high flux hemodialysis (HF HD) membranes improve antibody response to HBV vaccination in HD patients, which might indicate that HF HD membranes have immune modulator effects. Our study is the first to show that HF HD membranes may improve antibody response and reduce lack of/low response to hepatitis B virus vaccination in HD patients.
References 1. Fabrizi F, Martin P, Dixit V, Bunnapradist S, Dulai G (2004) Meta-analysis: the effect of age on immunological response to hepatitis B vaccine in end-stage renal disease. Aliment Pharmacol Ther 20:1053–1062 2. Tsouchnikas I, Dounousi E, Xanthopolou K, Papakostantinou S, Thomoglou V, Tsakiris D (2007) Loss of hepatitis B immunity in hemodialysis patients acquired either naturally or after vaccination. Clin nephrol 68(4):228–234 3. Stevens CE, Alter HJ, Taylor PE et al (1984) Hepatitis B vaccine in patients receiving hemodialysis. Immunogenicity and efficacy. N Engl J Med 311:496 4. Buti M, Viladomiu L, Jardi R et al (1992) Long-term immunogenicity and efficacy of hepatitis B vaccine in hemodialysis patients. Am J Nephrol 12:144
123
Int Urol Nephrol (2010) 42:1069–1075 5. Fabrizi F, Martin P (2001) Hepatitis B vaccine and dialysis: current issues. Int J Artif Organs 24:683–694 6. Cheung AK, Levin NW, Greene T, Agodoa L, Bailey J, Beck G, Clark W, Levey AS, Leypoldt JK, Ornt DB, Rocco MV, Schulman G, Schwab S, Teehan B, Eknoyan G, for the HEMO Study Group (2003) Effect of high-flux hemodialysis on clinical outcomes: results of the HEMO Study. J Am Soc Nephrol 14:3251–3263 7. Lonneman G, Novick D, Rubinstein M, Passlick-Deetjen J, Lang D, Dinarello CA (2003) A switch to high-flux helixone membranes reverses suppresed interferon gamma production in patients on low-flux dialysis. Blood Purif 21:225–231 8. Bordoni V, Piroddi M, Galli F et al (2006) Oxidant and carbonyl stres-related apoptosis in end-stage kidney disease: impact of membrane flux. Blood Purif 24:149–156 9. Daugirdas JT (1995) Simplified equations for monitoring Kt/V, PCRn, eKt/V, and ePCRn. Adv Ren Replace Ther 2(4):295–304 10. Grzegorzewska AE, Banachowicz W (2008) Evaluation of hemodialysis adequacy using online Kt/V and single-pool variable-volume urea Kt/V. Int Urol Nephrol 40(3):771– 778 11. Verkade MA, van Druningen CJ, Op de Hoek CT, Weimar W, Betjes MG (2007) Decreased antigen-spesific T-cell proliferation by moDC among hepatitis B vaccine non-responders on hemodialysis. Clin Exp Med 7:65–71 12. Girndt M, Ko¨hler H (2002) Hepatitis B virus infection in hemodialysis patients. Semin Nephrol 22(4):340–350 13. Stachowski J, Pollok M, Barth C, Maciejewski J, Baldamus CA (1994) Non-responsiveness to hepatitis B vaccination in haemodialysis patients: association with impaired TCR/ CD3 antigen receptor expression regulating co-stimulatory processes in antigen presentation and recognition. Nephrol Dial Transplant 9:144–152 14. Girndt M, Kohler H, Schiedelm-Weick E, Schlaek J, Meyer Zum Buschenfelde KH, Fleicher B (1995) Production of interleukin-6, tumor necrosis factor and interleukin-10 in vitro correlates with the clinical immune defect in chronic hemodialysis patients. Kidney Int 47:559–565 15. Descamps-Latscha B, Chatenoud L (1996) T cells and B cells in chronic renal failure. Semin Nephrol 16:183–191 16. Caillat ZS, Gimenez JJ, Wambergue F (1998) Distinct HLA class II alleles determine to antibody response to vaccination with hepatitis B surface antigen. Kidney Int 53:1626–1630 17. Cheung AK, Rocco MV, Yan G et al (2006) Serum beta-2 microglobulin levels predict mortality in dialysis patients: results of the HEMO study. J Am Soc Nephrol 17(2):546– 555 18. Vanholder R, Ringoir S, Dhondt A, Hakim R (1991) Phagocytosis in uremic and hemodialysis patients: a prospective and cross sectional study. Kidney Int 39:320–327 19. Ward RA, Ouseph R, McLeish KR (2003) Effect of highflux hemodialysis on oxidant stress. Kidney Int 63:353– 369 20. Girndt M, Heisel O, Ko¨hler H (1999) Influence of dialysis with polyamide versus hemophane hemodialysis on monokines and complement activation during a four month long term study. Nephrol Dial Transplant 14:676–682
Int Urol Nephrol (2010) 42:1069–1075 21. Schindler R, Lonnemann G, Schaffer J (1994) The effect of ultrafiltered dialysate on the cellular content of interleukin1 receptor antagonist in patients on chronic hemodialysis. Nephron 68:229–233
1075 22. Schiffl H, Wendinger H, Lang SM (2002) Ultrapure dialysis fluid and response to hepatitis B vaccine. Nephron 91:530–531
123
