Original Paper

Blood Purif 2012;34:3–9 Received: December 23, 2011

Accepted: April 17, 2012
DOI: 10.1159/000338923
Published online: June 13, 2012

Advanced Glycation End Products and

␤2-Microglobulin as Predictors of Carpal Tunnel
Syndrome in Hemodialysis Patients
Martin Busch Andreas Schwenzky Sybille Franke Günter Stein Gunter Wolf
Department of Internal Medicine III, Jena University Hospital – Friedrich Schiller University, Jena, Germany

Key Words Introduction

Advanced glycation end products ⴢ ␤2-Microglobulin ⴢ
Carpal tunnel syndrome ⴢ Dialysis-related amyloidosis ⴢ Besides enhanced cardiovascular risk and renal oste-
Hemodialysis opathy, dialysis-related amyloidosis (DRA) is a common
complication in patients with chronic kidney disease
(CKD) stage 5. ␤2-Microglobulin (␤2M) is the major con-
Abstract stitutional protein of this type of amyloid [1]. A frequent
Background/Aims: Carpal tunnel syndrome (CTS) is a com- clinical sign of DRA is the carpal tunnel syndrome (CTS).
mon clinical presentation of dialysis-related amyloidosis. It CTS is highly prevalent in patients who are on hemodi-
was determined whether ␤2-microglobulin (␤2M) and ad- alysis (HD) for more than 10 years [2]. Exact diagnosis of
vanced glycation end products in serum are predictors of DRA is made by the histological confirmation of amy-
CTS in dialysis patients. Methods: A total of 385 hemodialysis loid. Depositions of ␤2M amyloid are a common finding
patients were screened for CTS. ␤2M in serum was deter- [3]. In CKD, there is a progressive increase in serum ␤2M
mined by a competitive enzyme-linked immunoassay, CML with a significantly higher concentration of serum ␤2M
by a competitive enzyme-linked immunosorbent assay and in HD patients compared to healthy subjects [4].
total pentosidine by reverse-phase high-performance liquid Chemical modifications of ␤2M may cause the transi-
chromatography. Results: 127 patients (33%) were treated tion from an early asymptomatic to the symptomatic
with biocompatible membranes, 174 (45%) with high-flux di- stage of DRA. Advanced glycation end products (AGEs)
alysis. 122 patients (31.7%) had clinical signs of CTS. Signifi- contribute to such modifications [5]. AGEs are chemical
cant predictors of CTS were: age, female gender, serum ␤2M, modifications of proteins, lipids, peptides, amino acids
total protein, dialysis with non-biocompatible high-flux di- and nucleic acids by carbohydrates/reducing sugars in-
alysis compared to non-biocompatible low-flux dialysis, cluding reactive carbonyl compounds as metabolic inter-
Kt/V and serum concentration of CML (OR 2.47 for the 3rd vs. mediates [6]. Modifications of ␤2M have been described
1st quartile, 95% CI 1.229–4.961, p = 0.011). Conclusion: The for pentosidine and CML [5].
prevalence of CTS as a possible manifestation of dialysis-re- In this study, we tested whether serum concentrations
lated amyloidosis is still high. Serum concentration of CML of CML and pentosidine were related to the presence of
may be a predictor of CTS besides ␤2M and malnutrition. DRA as diagnosed by clinically confirmed CTS in HD
Copyright © 2012 S. Karger AG, Basel patients.
169.230.243.252 - 3/23/2015 12:54:53 PM

© 2012 S. Karger AG, Basel Martin Busch, MD

UCSF Library & CKM

0253–5068/12/0341–0003$38.00/0 Department of Internal Medicine III

Fax +41 61 306 12 34 Jena University Hospital, Friedrich Schiller University
Downloaded by:

E-Mail [email protected] Accessible online at: Erlanger Allee 101, DE–07747 Jena (Germany)
www.karger.com www.karger.com/bpu Tel. +49 3641 932 4621, E-Mail martin.busch @ med.uni-jena.de
 Methods interassay coefficient of variation of !8 and !9%. Parameters
such as total protein or creatinine were determined using auto-
Study Population mated standardized laboratory techniques.
This study was conducted in 5 outpatient dialysis units in Ger-
many. Between 2003 and 2004, 385 patients (204 female, 181 male Statistical Methods
patients, mean age 62.5 8 14.1) on maintenance HD treatment Results are given as means with standard deviations and me-
were enrolled for participation. Each participant provided written dians. Kruskal-Wallis H test was used for the comparison be-
informed consent, and the study was approved by the local ethics tween independent groups, Mann-Whitney U test for the com-
committee. parison between two independent groups, ␹2 statistics were used.
Spearman rank correlation test was used for estimating relation-
Diagnosis of CTS ships between variables. Stepwise logistic regression models were
During a physician-guided interview and by the help of a constructed. Covariates with p ! 0.10 were included in the final
structured questionnaire, patients were asked for any history of multivariate model. A p value ^0.05 was considered to indicate
CTS with four different questions: clinical signs of CTS such as statistical significance. Statistics were done using the Statistical
typical pain or paresthesia, any known pathological results of a Package of Social Science software (SPSS 15.0, 2006; SPSS Inc.,
nerve conduction testing of the median nerve, any surgical cor- Chicago, Ill., USA).
rection of CTS, or known muscular atrophy. Clinical examination
to confirm positive information completed each interview. Only
complaints starting after the initiation of chronic dialysis treat-
ment were considered. Results

Dialyzer/Dialysate Clinical and Laboratory Data

Dialyzers were classified into low flux (LF) or high flux (HF)
according to their ultrafiltration coefficient (UFC) [7]. Based on
Clinical data for the total group are shown in table 1.
current definitions [8], a UFC of !10 ml/hⴢmm Hg was classified There were no gender-related differences. Significant re-
as LF, a UFC of 110 ml/hⴢmm Hg as HF allowing dichotomization lationships of ␤2M, CML and pentosidine with selected
of data. However, 41 dialyzers from type F6–F8 HPS (Fresenius parameters in the total group are shown in table 2. Patient
Polysulfone쏐, Fresenius Medical Care, St. Wendel, Germany) properties in the groups of different dialyzers are given
with a UFC of 13–18 ml/hⴢmm Hg were also classified as low flux
in table 3.
since the manufacturer sized them explicitly as LF [7]. The most-
ly used dialyzers were as follows (only dialyzers used in 610 pa-
tients are given, data taken from Uhlenbusch-Körwer et al. [7]): Dialyzer, ␤2M and AGEs
Nipro Sureflux 110L (Cellulose Triacetate, UFC 9.4 ml/hⴢmm Hg, The mean UFC of all membranes was 18.3 8 16.0 (me-
n = 43), Hospal Diacepal 20 (Cellulose Diacetate, 17.7, 39), Cobe dian 12, 5.2–65) ml/hⴢmm Hg, 9.1 8 3.4 ml/hⴢmm Hg in
HG500 (Cellulose/Hemophan쏐, 6.4, 33), Cobe HG600 (Cellulose/ LF (n = 209), and 29.1 8 18.1 ml/hⴢmm Hg in HF (n =
Hemophan, 7.4, 29), Fresenius F6HPS (Polysulfone, 13, 22), Gam-
bro GFSPlus12 (Cellulose/Hemophan, 6.8, 21), GFSPlus20 (Cel-
176).
lulose/Hemophan, 11.4, 21), Fresenius F60(S) (Polysulfone, 40, In 67% (n = 247), non-biocompatible membranes were
17), Gambro Polyflux 14S (Polyamide STM, 62, 16), Nipro Sure- used, 99 of them were HF, 148 LF. Remaining patients
flux130E (Cellulose Triacetate, 17.8, 16), Gambro GFSPlus16 (Cel- (n = 125) were treated with biocompatible membranes, 75
lulose/Hemophan, 9.4, 13), Fresenius F7HPS (Polysulfone, 16, 13), with HF, 50 with LF. Patients treated with biocompatible
Nipro FB-150H/A (Cellulose Diacetate, 12.4, 10), remaining (n =
membranes showed significantly lower ␤2M concentra-
92). Biocompatibility was classified as described [7]. Efficacy of
dialysis treatment and dosing was calculated as single-pool Kt/V tion compared to non-biocompatible membranes (medi-
using the equation of Daugirdas [9]. an 38.2 vs. 43.2 mg/l, p = 0.026). Patients on LF dialyzers
At the time of data collection, all patients were treated by a had lower ␤2M and pentosidine levels than those treated
sterile and nonpyrogenic dialysis fluid containing less than 100 with HF (median 38.2 and 1,148 in LF vs. 43.0 mg/l and
colony forming units per liter (CFU/l), and less than 0.025 endo- 1,409 pmol/ml in HF, p ^ 0.008). In non-biocompatible
toxin units per ml (EU/ml) [10].
HF, ␤2M levels were significantly increased compared to
Laboratory Analysis any other group (p ! 0.001, respectively; table 3).
Total pentosidine (free and protein bound) was measured us-
ing high-performance liquid chromatography, levels of serum Prevalence of CTS
CML by enzyme-linked immunosorbent assay [11]. The intra- In 31.7% (n = 122), CTS was diagnosed. By the ques-
and interassay coefficients of variation are !5 and !7%, respec-
tionnaire, typical pain was specified in 14.5% (n = 56), a
tively. Reference values for serum AGE concentrations were ob-
tained from a group of 51 healthy volunteers [11]. Serum concen- pathological nerve conduction test in 9.6% (n = 37), and
tration of ␤2M was measured with enzyme-linked immunoassay muscular atrophy in 21.3% (n = 82). Thirty-four patients
(R&D Systems, Minneapolis, Minn., USA) having an intra- and (8.8%) underwent surgical release intervention. The pro-
169.230.243.252 - 3/23/2015 12:54:53 PM

4 Blood Purif 2012;34:3–9 Busch/Schwenzky/Franke/Stein/Wolf

UCSF Library & CKM
Downloaded by:
Table 1. Clinical data from 385 patients on maintenance HD Table 2. Correlations (Spearman rho test) of ␤2M, CML and pen-
tosidine with selected clinical parameters
Age, years 63.1814.1
65.1 (20.4–96) ␤2M CML Pentosidine
Gender, female/male 204 (53%)/181 (47%)
Body mass index –0.171 –0.268 –0.26
Diabetes mellitus 161 (42%) 0.001 <0.001 <0.001
Body mass index 25.484.9 Dialysis duration 0.357 0.164 0.357
24.9 (15.5–39.5) <0.001 0.001 <0.001
Dialysis duration, months 42.59838.32 Residual diuresis –0.514 –0.255 –0.329
31.44 (0.46–239) <0.001 <0.001 <0.001
Dialysis time/week, h 12.882.1 Dialysis time/week 0.264 0.142 0.193
12.9 (4–20) <0.001 0.006 <0.001
Residual diuresis, ml/day 502.98691.4 Kt/V 0.258 0.203 0.216
200 (0–3,600) <0.001 <0.001 <0.001
Dialysis membrane (HF/LF) 176 (45.7%)/209 (54.3%) Creatinine 0.423 0.172 0.267
<0.001 0.001 <0.001
Kt/V 1.380.34
1.29 (0.54–3.11) Total protein NS 0.227 NS
<0.001
Creatinine, ␮mol/l 8888279
887 (161–1,621) ␤2M 0.346
<0.001
Hematocrit 0.3480.04
0.34 (0.22–0.46) Total pentosidine 0.437 0.620
<0.001 <0.001
Intact PTH, ng/l 2798331
167 (2–2,202)
Total protein, g/l 70.787.2
70.1 (48.5–100.3)
␤2M, mg/l1 43.1832.1 CTS and Dialyzers/Kt/V
40.5 (5.7–572) Thirty-four percent (n = 85) of patients treated with
Total pentosidine, pmol/ml2 1,4598949 non-biocompatible dialyzers (n = 254), showed CTS com-
1,274 (259–10,066) pared to only 28% (n = 35) of those treated with biocom-
CML, pmol/ml3 4,06882,532 patible membranes (n = 127), which was not significantly
3,616 (169–15,902) different (p = 0.292). However, 71% (n = 85) of the patients
suffering from CTS, were treated by non-biocompatible
Data are mean 8 standard deviation and median (range). dialyzers. Patients with HF showed CTS in 35.6% com-
1
Reference range: 0.9–2 mg/l.
2 Reference range: 71–200 pmol/ml. pared with 29.1% on LF dialysis (not significant).
3 Reference range: 1,500–2,800 pmol/ml. For logistic regression analysis, membranes were di-
vided into 4 groups: non-biocompatible LF, biocompati-
ble LF, non-biocompatible HF and biocompatible HF (ta-
ble 3). Figure 2 shows their different rate of use in patients
with CTS (n = 122). Figure 3 illustrates the differential use
of dialyzers in relation to dialysis duration. Thus, the rate
portion of CTS was nearly twice as high in women com- of HF and the use of biocompatible dialyzers is increasing
pared to men (66.3 vs. 33.7%, p ! 0.001). Prevalence of with longer time on dialysis treatment.
CTS increased with longer time on dialysis (fig. 1). Pa- There were significant differences in Kt/V between pa-
tients with CTS were older than patients without CTS tients with and without CTS (1.42 in CTS vs. 1.25 without
(66.4 vs. 60.7 years, p ! 0.001), also at the start of dialysis CTS, p ! 0.001). Patients with CTS had a significantly low-
treatment (62.3 vs. 57.4 years, p = 0.002). CTS patients had er residual diuresis compared to those without (378 8 547
a higher ␤2M concentration compared to non-CTS (me- vs. 561 8 743 ml; p = 0.003). Women had a significantly
dian 43.0 vs. 38.9 mg/l, p = 0.034). higher Kt/V (1.42 8 0.34 vs. 1.17 8 0.29; p ! 0.001).
169.230.243.252 - 3/23/2015 12:54:53 PM

Advanced Glycation End Products and Blood Purif 2012;34:3–9 5

UCSF Library & CKM

Carpal Tunnel Syndrome

Downloaded by:
 39.6* 38.7
40 34.7 40
Prevalence of CTS (%)

32.8
28.7
30 30
22.9*

Rate (%)
19.3
20 20
9.2
10 10

0 0
0.5–14.3 14.3–31.4 31.4–60.3 60.3–239 LF non- LF HF non- HF
(n = 96) (n = 98) (n = 95) (n = 96) biocompatible biocompatible biocompatible biocompatible
Quartiles of dialysis duration (months) Membrane type

Fig. 1. Prevalence of CTS in relation to quartiles of dialysis dura- Fig. 2. Use of different types of dialyzers in CTS patients (n = 122).
tion. * p = 0.015.

70 LF non-biocompatible
61.7 LF biocompatible
60 HF non-biocompatible
HF biocompatible
50 46.7

40 38.7
Rate (%)

34.1

28.3 28.3 28.7

30 26.4

20.2 21.3
20 16
14.1
10.9 11 11.7
10
2.1
0
0.5–14.3 14.3–31.4 31.4–60.3 60.3–239
(n = 96) (n = 98) (n = 95) (n = 96)
Fig. 3. Use of different types of dialyzers in Quartiles of dialysis duration (months)
385 patients on hemodialysis treatment in
relation to quartiles of dialysis duration.

Predictors of CTS Discussion

In univariate analysis, older age, female gender and
longer dialysis duration were related to the presence of The present study confirms that the prevalence of CTS
CTS (p ! 0.015, respectively). Furthermore, correlations is high in HD patients. Depending on the criteria used for
were shown for residual diuresis !1,000 ml/day, lower diagnosis, the prevalence of CTS in our HD cohort was
creatinine (p ^ 0.045, respectively), higher Kt/V and twice as high when compared with the general popula-
lower total protein (p ! 0.009, respectively) as well as tion in which prevalence varies from 5 to 16% [12]. CTS
higher ␤2M, and CML levels (p ^ 0.044, respectively). is known to be the earliest marker for DRA [1]. Most like-
Predictors for CTS as assessed with multivariate analysis ly, the higher prevalence of CTS in our cohort can be re-
are shown in table 4. lated to the presence of DRA. Hence, the rate of DRA-
associated symptoms in HD patients is still high. The oc-
curence of CTS related to DRA increases 3–5 years after
169.230.243.252 - 3/23/2015 12:54:53 PM

6 Blood Purif 2012;34:3–9 Busch/Schwenzky/Franke/Stein/Wolf

UCSF Library & CKM
Downloaded by:
Table 3. Selected parameters (means 8 SD) in different types of dialysis membranes

LF LF HF HF p value1
non-biocompatible biocompatible non-biocompatible biocompatible
(n = 148) (n = 50) (n = 99) (n = 75)

Age, years 66.7813 61.8814,6 60.2814.3 57.9813.7 <0.001

Body mass index 24.684.8 25.785.6 25.384.3 2784.8 0.007
Dialysis duration, months 30.7831.8 41.9851.9 44.1830 65.2841.1 <0.001
Residual diuresis, ml/day 6698765 6148725 3998593 2128364 <0.001
Kt/V 1.2980.41 1.2680.34 1.3380.27 1.3480.27 0.150
Creatinine, ␮mol/l 7698270 8148268 9848256 1,0558204 <0.001
Total protein, g/l 70.387.75 68.286.95 71.986.26 71.587.66 0.053
Hematocrit 0.3380.38 0.3480.41 0.3480.4 0.3480.33 0.338
Intact PTH, ng/l 2348273 2478257 2728309 3838441 0.126
␤2M, mg/l 39816.9 38.5819 48.8816.3 46.8864.9 <0.001
CML, pmol/ml 4,05782,512 4,34083,187 4,18182,411 4,01882,335 0.982
Total pentosidine, pmol/ml 1,46781,190 1,2748705 1,4868673 1,5768896 0.059
1 According to Kruskal-Wallis test.

Table 4. Multivariate logistic regression for the presence of CTS (n = 122) in 385 patients on maintenance HD

Indicators Regression Standard Odds ratio 95% confidence p

coefficient error interval

Age1 (per increase of 1 year) 0.035 0.010 1.035 1.016–1.055 <0.001

Kt/V1 (per increase of 0.1) 0.136 0.039 1.146 1.062–1.236 <0.001
Dialyzer1 (non-biocompatible
HF dialyzer vs. non-biocompatible LF dialyzer) 0.687 0.309 1.987 1.085–3.639 0.026
CML1 (3rd vs. lowest quartile) 0.904 0.356 2.470 1.229–4.961 0.011
Total protein1 –0.052 0.019 0.949 0.915–0.984 0.005
Gender2 (female vs. male) 0.626 0.237 1.870 1.176–2.972 0.008
Dialysis duration2 (per increase of 1 month) 0.007 0.003 1.007 1.002–1.013 0.012
␤2M3 (3rd vs. lowest quartile) 0.991 0.368 2.695 1.311–5.540 0.007
␤2M3 (highest vs. lowest quartile) 0.808 0.377 2.243 1.072–4.692 0.032

CML quartiles: lowest 169–2,236; 2nd 2,236–3,621; 3rd 3,621–5,268; highest 5,268–15,902 pmol/ml; ␤2M quartiles: lowest 5.7–29.1;
2nd 29.1–40.5; 3rd 40.5–53.1; highest 53.1–572 ␮g/ml; Dialyzers: non-biocompatible LF, biocompatible LF, non-biocompatible HF,
biocompatible HF.
1 Adjusted for gender, dialysis duration, residual diuresis and other variables denoted with 1.
2 Adjusted for residual diuresis and other variables denoted with 2 .
3
Adjusted for residual diuresis, CML, biocompatibility LF/HF, ␤2M and other variables denoted with 3.

starting HD [2]. In the present study, patients with CTS tionnaire and clinical examination. Thus, any prevalence
were dialyzed for a mean of 4.1 8 3.5 years. For the diag- of non-dialysis-related CTS cannot be ruled out com-
nosis of DRA, the histological proof of ␤2M amyloid de- pletely, also because the clinical signs of CTS from DRA
posits with a positive Congo red staining and typical dou- and non-DRA-related CTS are not different. The higher
ble refraction under polarized light is the gold standard prevalence of CTS in women might be an indicator for
[13] which cannot be used routinely in a clinical setting. that since, in the general population, women are affected
The diagnosis of CTS in our study was made by a ques- twice as often as men [12].
169.230.243.252 - 3/23/2015 12:54:53 PM

Advanced Glycation End Products and Blood Purif 2012;34:3–9 7

UCSF Library & CKM

Carpal Tunnel Syndrome

Downloaded by:
 The question whether DRA is really decreasing re- sate retards and decreases long-term complications of
mains unanswered since reliable clinical data are lacking HD like DRA [18].
[3]. HF was shown to decelerate the occurrence of DRA Extending dialysis time resulted in a better removal of
[1]. In our study, the prevalence of CTS was not different ␤2M, not related to Kt/V [20]. In our study, higher Kt/V
in HF vs. LF. Moreover, the use of non-biocompatible HF was predictive of DRA, which can be regarded as a para-
dialyzers increased the probability for CTS as compared dox result. But there was also positive correlation of ␤2M
to non-biocompatible LF. However, HF was used signifi- and AGEs with Kt/V. These findings might have various
cantly more in patients who have been longer on dialysis. explanations since the formula of Kt/V (urea clearance K
In a Japanese cohort, patients with longer dialysis time ! dialysis time t/urea distribution volume V) results in
showed higher rates of DRA and CTS independent of HF a lower distribution volume leading to higher Kt/V val-
or LF [14]. Patients treated with HF had significantly low- ues. Probably, this was also the case for women in our
er concentrations of ␤2M than patients with LF [15]. In study as two thirds of the patients with CTS were female.
the present study, however, ␤2M was lower in LF, which Moreover, lower distribution volume in ESRD often re-
was possibly related to the significantly lower residual re- sults from malnutrition and inflammation as it is highly
nal function and a longer dialysis time in the HF group. evident in patients on long-term dialysis treatment [21].
In addition, long-term dialysis patients on HF might have Supporting this, a higher concentration of total protein
used LF for several years and then were switched to HF was protective against CTS in the current study.
when CTS was already established. AGEs are independently related to a decrease in the
The difference between the rate of DRA in non-bio- glomerular filtration rate, and are a class of uremic tox-
compatible vs. biocompatible membranes was not statis- ins. In our study, serum concentrations of CML and pen-
tically significant. But most of the patients with CTS tosidine were 1.5 and 7 times higher compared to healthy
were treated with non-biocompatible membranes. Thus, subjects confirming previous findings [11]. Modifica-
the biocompatibility of dialysis membranes might play a tions of ␤2M by AGEs are a major component of HD-
role in the formation of DRA, which is in accordance associated amyloidosis, and are said to induce the clinical
with previous findings [2, 15]. ␤2M levels were lower manifestation of DRA [5]. The positive correlation of
when biocompatible membranes were used [16]. As in ␤2M and AGEs in our patients might suggest this rela-
the present study, the majority of LF membranes are not tionship. By means of univariate and multivariate logistic
biocompatible, contributing to a higher rate of DRA in regression, increased serum levels of CML were shown to
LF [2]. increase the probability of CTS in our group of HD pa-
␤2M is thought to play a major role in the pathogen- tients. Thus, it could be shown that increased serum or
esis of DRA [17]. In our study, the concentration of ␤2M plasma concentrations of AGEs such as CML might sug-
was up to 30 times higher than in healthy subjects. Pa- gest the presence of CTS complaints as a possible mani-
tients with CTS revealed a significantly higher concen- festation of DRA confirming experimental findings [5].
tration of ␤2M, which increased the probability for CTS In contrast, cardiovascular end points in CKD were not
confirming previous findings [3, 15, 17]. The lower re- related to higher serum concentration of AGEs [6]. How-
sidual renal function in CTS might have contributed to ever, DRA formation and modification of ␤2M by AGEs
this. In gaining lower ␤2M levels, improvements in di- is a long-term process reflecting tissue accumulation of
alysis membrane together with the use of ultrapure di- AGEs as well. Pentosidine showed no association with
alysate are recommended [3, 15]. Furuya et al. [18] found DRA as also shown by others [22]. Older age and longer
a significantly lower ␤2M concentration by the use of dialysis time could be confirmed as known predictors of
ultrapure dialysate. The use of ultrapure dialysate was CTS/DRA [2].
accompanied by a lower prevalence of DRA, indepen- In conclusion, our results demonstrate that CTS/DRA
dent of HF or LF [19]. Although the use of ultrapure di- in dialysis patients is still important. Apart from the con-
alysate was already a common standard at data collec- firmation of known predictors of CTS/DRA, serum con-
tion (see ‘Methods’), the study is limited by its data centration of the AGE CML seems to play a role besides
collection time between the years 2003 and 2004. increased ␤2M. Moreover, malnutrition might be a piv-
Considering the inconstant use of ultrapure dialysate in otal factor in DRA formation.
the past, influence of dialysate purity on the present re-
sults is likely since DRA formation is a long-lasting pro-
cess. Nowadays, the consequent use of ultrapure dialy-
169.230.243.252 - 3/23/2015 12:54:53 PM

8 Blood Purif 2012;34:3–9 Busch/Schwenzky/Franke/Stein/Wolf

UCSF Library & CKM
Downloaded by:
 Acknowledgements Disclosure Statement

The authors wish to thank the following nephrologists and The authors state that they have no conflict of interest.
dialysis outpatient centers for their outstanding collaboration: Dr.
Christoph C. Haufe and Dr. Ralf Czerwinski, KfH Dialysis Cen-
ter, Erfurt; Dr. André Schip, Dr. Andreas Biermann and Dr. Hans-
Peter Holzapfel, DM Sylvia Pirstat, Dialysis Center Thomaseck,
Erfurt; Dr. Norbert Jung, Dr. Michael Scholl and DM Michael
Hildebrandt, Dialysis Centers in Mühlhausen and Bad Langen-
salza; Klaus W. Ochlich†, formerly Dialysis Center Eichsfeld, Hei-
ligenstadt; Dr. Ingo Brauns and Dr. Dieter Voigt, Dialysis Center,
Gotha.

References
1 Saito A, Gejyo F: Current clinical aspects of 9 Daugirdas JT: Second generation logarith- 16 Hakim RM, Wingard RL, Husni L, Parker
dialysis-related amyloidosis in chronic dial- mic estimates of single-pool variable volume RA, Parker TF 3rd: The effect of membrane
ysis patients. Ther Apher Dial 2006; 10: 316– Kt/V: an analysis of error. J Am Soc Nephrol biocompatibility on plasma beta 2-micro-
320. 1993;4:1205–1213. globulin levels in chronic hemodialysis pa-
2 Jadoul M: Dialysis-related amyloidosis: im- 10 Canaud B, Bosc JY, Leray H, Morena M, Stec tients. J Am Soc Nephrol 1996; 7:472–478.
portance of biocompatibility and age. F: Microbiologic purity of dialysate: ratio- 17 Drueke TB, Massy ZA: Beta2-microglobu-
Nephrol Dial Transplant 1998; 13(suppl 7): nale and technical aspects. Blood Purif 2000; lin. Semin Dial 2009;22:378–380.
61–64. 18:200–213. 18 Furuya R, Kumagai H, Takahashi M, Sano K,
3 Yamamoto S, Kazama JJ, Narita I, Naiki H, 11 Busch M, Franke S, Muller A, Wolf M, Gerth Hishida A: Ultrapure dialysate reduces plas-
Gejyo F: Recent progress in understanding J, Ott U, Niwa T, Stein G: Potential cardio- ma levels of beta2-microglobulin and pento-
dialysis-related amyloidosis. Bone 2009; vascular risk factors in chronic kidney dis- sidine in hemodialysis patients. Blood Purif
45(suppl 1):S39–S42. ease: ages, total homocysteine and metabo- 2005;23:311–316.
4 Stein G, Schneider A, Thoss K, Ritz E, Schae- lites, and the C-reactive protein. Kidney Int 19 Lonnemann G, Koch KM: Beta(2)-micro-
fer K, Huller M, Sperschneider H, Marzoll I: 2004;66:338–347. globulin amyloidosis: effects of ultrapure di-
Beta 2-microglobulin serum concentration 12 Aroori S, Spence RA: Carpal tunnel syn- alysate and type of dialyzer membrane. J Am
and associated amyloidosis in dialysis pa- drome. Ulster Med J 2008;77:6–17. Soc Nephrol 2002; 13(suppl 1):S72–S77.
tients. Nephrol Dial Transplant 1991;6(suppl 13 Yamamoto S, Gejyo F: Historical back- 20 Eloot S, Van Biesen W, Dhondt A, Van de
3):57–61. ground and clinical treatment of dialysis-re- Wynkele H, Glorieux G, Verdonck P, Van-
5 Miyata T, Taneda S, Kawai R, Ueda Y, Horiu- lated amyloidosis. Biochim Biophys Acta holder R: Impact of hemodialysis duration
chi S, Hara M, Maeda K, Monnier VM: Iden- 2005;1753:4–10. on the removal of uremic retention solutes.
tification of pentosidine as a native structure 14 Otsubo S, Kimata N, Okutsu I, Oshikawa K, Kidney Int 2008;73:765–770.
for advanced glycation end products in beta- Ueda S, Sugimoto H, Mitobe M, Uchida K, 21 Himmelfarb J, Stenvinkel P, Ikizler TA, Ha-
2-microglobulin-containing amyloid fibrils Otsubo K, Nitta K, Akiba T: Characteristics kim RM: The elephant in uremia: oxidant
in patients with dialysis-related amyloidosis. of dialysis-related amyloidosis in patients on stress as a unifying concept of cardiovascu-
Proc Natl Acad Sci USA 1996; 93:2353–2358. haemodialysis therapy for more than 30 lar disease in uremia. Kidney Int 2002; 62:
6 Busch M, Franke S, Ruster C, Wolf G: Ad- years. Nephrol Dial Transplant 2009; 24: 1524–1538.
vanced glycation end-products and the kid- 1593–1598. 22 Takahashi M, Hoshino H, Kushida K, Mu-
ney. Eur J Clin Invest 2010;40:742–755. 15 Schiffl H, Fischer R, Lang SM, Mangel E: rata H, Baba S, Inoue T: The advanced glyca-
7 Uhlenbusch-Körwer I, Bonnie-Schorn E, Clinical manifestations of ab-amyloidosis: tion endproduct, pentosidine, in the carpal
Grassmann A, Vienken J (eds): Understand- effects of biocompatibility and flux. Nephrol ligament in patients with carpal tunnel syn-
ing Membranes and Dialysers. Lengerich, Dial Transplant 2000;15:840–845. drome undergoing hemodialysis: compari-
Pabst, 2004, vol 5. son with idiopathic carpal tunnel syndrome.
8 Hoenich NA: Membranes and filters for hae- Nephron 1998;80:444–449.
modiafiltration. Contrib Nephrol 2007; 158:
57–67.
169.230.243.252 - 3/23/2015 12:54:53 PM

Advanced Glycation End Products and Blood Purif 2012;34:3–9 9

UCSF Library & CKM

Carpal Tunnel Syndrome

Downloaded by: