1

Towards Improving HD
Efficiency
HD Membranes update
HESHAM ELSAYED
PROFESSOR OF NEPHROLOGY
AIN SHAMS UNIVERSITY - ESNT 2018
HD improvements the 3
“3” Unknowns UFR

Integrated Approach
Solid Kt/V
Knowns Time

Mortality
+ RCT
QOL

Fluxes

Unknown Known
Toxins
Unknowns Unknowns
HD improvements the 4
“3” Unknowns Solid
Knowns
Mortality
+
QOL

Unknown
More Proteomics to comeKnown
to
Unknowns
Known UnknownsUnknowns
Men and Women are equal on St Kt/V Women need More HD Dose with BSA correction

KT/V urea is still the official marker

5
Nature Review of Nephrology 2018

6
 7

Pushing
Towards
A Better
Dialysis
To Know the
Unknowns
Measures of HD Adequacy Thirty Years: Lessons Learned in Dialysis
8
Daugirdas :Semin on Dialysis Jan 2017

Additional Measure QOL against Renal Tx

phosphorus and beta-2-

microglobulin (β2-M) removal

Secondary Measure volume status and

ultrafiltration rate

Embraced by the KDOQI adequacy

guidelines
Primary Measure
Kt/V Embraced by the
Caregivers
PAYERS
Dying to Feel Better: The Central Role of Dialysis–
9

Induced Tissue Hypoxia (CJASN April 07, 2016 )

IMPROVING HD

Control of Smooth UFR Better Improve

Uremic Avoid Organs Recovery Depression / Higher QOL
Symptoms Stunning time Post HD dependence

Different
HD Techniques
Outcome
Missing points in clinical practice (Am J Kidney Dis. 2014 Jul;64) 10

Post HD Recovery Time ( 6000 Patients )

How long does it take you to recover from a dialysis session?
% of Patients Recovery Time
45%

Longer recovery times after dialysis is associated with poorer

> 50 % Have Delayed Recovery
40%
41%
35% HRQOL and 20% have greater 50% increase in mortality,
30% 32%
25%
20%
15% 17%
10%
5%
10%
0%
< 2 Hours 2- 6 Hours 7 - 12 Hours > 12 Hours
 Identify patients Have
Recovery Time
Poor QOL
High Risk of Dying

11
 "Optimal" Dialysis Dose prescription Should 12

Comfortable HD

High
More
Permeable
Frequent?
Better Control of BP , membranes
Anemia , CKD-MBD Better
Goal Survival
And HRQOL

Prolonged Time ?
Better control of Uremic Toxins
and Volume state
Therapeutic Approach Kidney International (2014)
13
Haemodialysis Therapy: A Never-Ending Journey 14

LF Dialysis = Small solute KT/V Sunset of KT/V

• High Flux Dialysis and HDF ?? HD dose prescription

• Medium Cut off Dialysis ?? HD dose prescription

• High Cut off Dialysis ?? HD dose prescription

 ULTRAPURE DIALYSATE 15

Microbes and ET are

not allowed

Ultrapure dialysis water obtained

with additional ultrafilter may
reduce inflammation in patients on
hemodialysis.
Journal of Nephrology December 2017, Volume 30
Achieving High Quality water system for Ultrapure Dialysate 16
( UPD)

Standard Ultrapure
Standard Water Ultrapure Water Sterile Dialysate
Dialysate Dialysate
Microbiological Standards for Water and Dialysis Fluid Purity

Bacterial limits a,
< 100-200 < 100-200 < 0.1 < 0.1 < 10-6
CFU/mL

Endotoxin limits b,
< 0.25-2 < 0.25 < 0.03 < 0.03 < 0.03
EU/mL

UPW+UPD
 17
HD Membranes Physiochemical Structures
Before Riding the Road

Blood contact 3 km every

15 sec
nce
i sta
d
k m
3
r is
ze
al y
Di

18
Membrane for Hemodialysis Less Clotting
HD Membranes Physiochemical Structures 20

The Structure

Back Bone and a Hydrophilic Components

The Function

Fluxes and Ultrafiltration

Biocompatibility

As Smooth as an Endothelium
HD Membranes Physiochemical Structures SEM of Ps 21

Asymmetrical Structure

Back Bone of PS
Hydrophobic

Internal skin Layer

Sieving Layer
HD Membranes Physiochemical Structures 22

Internal skin Layer

Sieving Layer 1 um

Define the Molecule that

can Pass
PS Polymer in the
39 um Thickness Should be smooth as an
Endothelium

Only Area of Blood contact

HD Membranes Physiochemical Structures

23
24
PVP Based PS Hydrogel
Formation
Smooth as an Endothelium

25
Computer Simulation of Flow Design

Flow Design

Blood / Dialysate
Match

Blood Entry
Clearance and Ko

26
 Clearance and Ko
Diffusive Permeability

27
Thrombogenicity
Front. Immunol., 25 January 2018

Blood Contact with Membrane

28
Thrombogenicity

Blood cells
Activations

Front. Immunol., 25 January 2018

29
 Nature Reviews Nephrology
Thrombogenicity volume 13, 285–296 (2017)

recruits platelets and

Contact with surface Protein changes Activation
catalyses fibrin formation

Ab, antibody; Bb, activated factor B; C5aR, C5a anaphylatoxin chemotactic receptor 1; FcγR, Fcγ receptor; HK, high molecular weight kininogen; PAR1/4,
proteinase-activated receptor 1/4; P2Y1/12, P2Y purinoceptor 1/12.

30
 Augmentation
Model of innate
Contact Activation
Thrombosis immunity activation
as a cause of
cardiovascular
disease during
EC Atheroma Calcifications dialysis

transfer of
inflammation to EC
31
How to Choose a Dialyzers ?
Fibers • Material and Flux

Size • Surface area in m2

Dialyzer • Biocompatibility

Steriliz • Steam – Gamma –

Ebeam EO
clotting • Better Rhology

Performance • Clearance values

Micro-undulation ?

32
HD membranes in clinical uses
Clinical uses Determine the outcome

33
 Man Power / Equipment / patients

High Quality HD machines

High Quality Disposables

Adequate HD ??
HD safety

Dialysis unit in a convenient mode

Blood speed in the
middle and the
periphery
Improving clearance by better Blood &
dialysate flow geometry
Blood in
Blood to Dialysate matching
to increase the clearance
Header

Hollow fiber

1.9 cm/sec
1.3 cm/sec
Casing or jacket

Dialysate

Blood out
Blood Flow speed is higher in the Dialysate Flow is higher in the
center periphery
 Towards Better Dialysis current achievements
38

Higher Convection Extended and Expanded HD

Membranes Adsorption
TTT Frequent Sessions HDx

Improve compartmental
Higher SC Amplified Clearance Removal of bigger Remove Bigger molecules
Removal
molecule
Than HFD
Removing MM HE-HDF More Physiological Protein Bound Toxins

More Porous Towards Larger Towards Higher Fluxes

Membranes Molecules Removal

Line of improvements
FLUX AND SOLUTE PERMEABILITY 39

cut off = SC 0.1

Super Plasma
LF MF HF MCO HCO
Flux Filter

Cut off
MW
D

5000 8000 20000 3000 45000 65000 2M

Higher MW Uremic Toxins

Beyond B2m
PB Toxins

MM

Small

40
 Cytokine Reduction with HF Dialysis
“STUDY OF THE EFFECTS OF HEMODIAFILTRATION VERSUS Elsayed H , etal 2017
HEMODIALYSIS ON DNA METHYLATION AND INDOXYL SULFATE
REMOVAL Elsayed H , etal 2018
Dialyzer 1 Dialyzer 2 30.00% Dialyzer 1 Dialyzer 2
0.5

0.45 27.10%
0.45 25.00%
0.4

0.35 20.00%
0.3
0.28 15.00%
0.25

0.2
10.00% 10%
0.15

0.1
5.00%
0.05

0 0.00%
Myoglobin SC IL-1 B RR
Indirect correlation (r= -0.922, P < 0.001)
 42

Medium Dialyzer
Cut Off
membrane “Pore
diameter”
Targeting Beyond B2m in MW

43
Medium cut-off membranes - closer to the natural
44
kidney removal function ( MCO )
Medium cut off Membrane MCO 45

M
M BM
Albumin
Bigger pore
Radius
 HDF
MCO Membrane Permeability
Higher Membrane
46
+
Permeability TMP dragging

R
Pe emov Pe Rem
rm rm
eab al by eab oval
ilit hig ilit by
ya h y+
lo n TM
e P

Quantification is Quantification is
needed calculated

More RCT is needed RCT on longterm

benefits
HD Backfiltration internal HDF 47

Pressure Drop

Back filtration volume is risky for

Pyrogen transfer `… volume ??
HDF and dose optimization 48
HDF Dose 49
Depends on : Filtration Fraction in
the Safe Limits
 1-Blood volume between 25 – 30 %
processed. FF = UFR / QB
 2- UF and substitution Volume of
Volume of
convection
volume volume
processed Blood
 HDF
Requirements for Sterile
Dialysate infusion

2
Barriers
• Using 2 ERF
3 • Using 3 ERF
Barriers

ELITE NEPHROLOGISTS SCHOOL (ENES) 2018 50

 HDF benefit over High Flux HD 51
Elsayed H in press 2018

RR %
90

80
82 82.2
70 76
60
67.4
50

40

30

20

10

0
URR % B2M RR%
HDF HFHD
 EFFECT OF OL-HDF on DNA methylation 52
Elsayed H, Elsharkawey M etal 2018 in press
DNA Methylation

Correlation coefficient p value

HFHD -0.156 0.510

HDF

Indoxyl
sulfate
0.004 0.987
DNA methylation
r P
Substitution volume -0.922 <0.001**

**; High Statistical Significant difference

ct correlation (r= -0.922, P < 0.001)

 Hemodiafiltration Reduces All-Cause and 53
Cardiovascular Mortality in Incident Hemodialysis
Patients: A Propensity-Matched Cohort Study
October 2017, Vol.46, No. 4
study cohort comprised 3,075 incident dialysis patients treated

Compared with patients on high-flux HD, those on online hemodiafiltration

received a median replacement volume of 23.45 L/session
Median follow-up period was 2.54 (1.09–4.46) years
Manifested 24 and 33% reductions in all-
cause and cardiovascular mortality

54
Crude model: The univariate analysis, only October 2017, Vol.46, No. 4
including the convective volume;

The crude model adding the Td and the QB

Adjusted analysis all the variables recorded for the study

HDF Dose
dependent
HR

55
Improving Erythropoiesis Stimulating Agent 56
Hyporesponsiveness in Hemodialysis Patients: The
Role of Hepcidin and Hemodiafiltration Online

2018, Vol.45, No. 1-3

HDF latest 11 October 2017 57
Mortality risk in patients on hemodiafiltration versus hemodialysis:
a ‘real-world’ comparison from the DOPPS
% of Patients on HDF
60%
% of patients Subst volumes
60%
50%
52%
50%
40%
50%
40%
30%

30%
20% 23% 27%
20%
10% 13% 16%
10%
0%
0%
7%
SWEEDEN GERMANY OVERALL
4-15 L 15-20 L > 20 L Missing
From: Mortality risk in patients on hemodiafiltration versus hemodialysis: a ‘real-world’ comparison from
the DOPPS NDT 2017
58
 Nephrologists’ perception on clinical indications for HDF 59
use

% of Patients

85%
84%
75% 77%
75%
65%
63%
55% 58% 58%
51%
45%

35%

25% 28%

15%

5%
Role of dialysis in amyloidogenesis (DRA) 60

Dialysate
Impurities
Retention

Type of
Dialyzers
LF – HF
Longer duration
of HD
Elevated
levels of
cytokines

Overproduction
B2M
inflammation Deposition
2293 patients with a minimum of 2 years of follow-up were analysed
61
Parameter Value
62
Treatment time per session (min; 241.5±14.0
mean±s.d.)
Patients on thrice weekly treatments (%) 96

Effective blood flow (ml/min; mean±s.d.) 404.9±69.4

Dialyzer surface (m2; mean±s.d.) 1.54±0.20

Post-dilution HDF (%) 100%
Effective dialysate flow (ml/min; 506.1±53.6
mean±s.d.)

Substitution flow (ml/min; mean±s.d.) 91.4±13.6

Substitution flow normalized to BSA 52.2±11.4

(ml/min/m2, mean±s.d.)

Kidney Int. 2015 Nov;

MCO MEMBRANE 63
Expanded HD ( Expanded than HF removal )
ERA-EDTA 2017 MADRID
they allow for removal of an
expanded range of uremic
toxins compared to
conventional high-flux
membranes.
 Potential Benefits of MCO membranes 64

Expanded
HD
Albumin Loss in
conventional HF
Higher Albumin loss 2.9
– 7 gm
is below 0.4 gm

downregulate the expression of Transcription of pro-

Greater clearance of limited effect on removal of soluble inflammatory cytokines in
both IL-6 and tumor necrosis
λFLC (45 kDa in size protein-bound toxins mediators is enhanced peripheral leukocytes is
factor-α mRNA
markedly reduced
 65
Performance of hemodialysis
depends of membrane and
technique

Nephrol Dial Transplant (2017) 32

(1): 165-172
Performance of 66
hemodialysis depends of
membrane and technique

Nephrol Dial Transplant

(2017) 32 (1): 165-172
HDX versus HD RR % ERA-EDTA 2017 MADRID 67

Adequacy update :Targeting Beyond B2m in MW

 Dose of removal depends on HD Membrane 68
international journal of artificial organVol. 40 Issue 7 | Jul 2017 | pp. 313 - 366

Dialyzer type Water Sieving coefficient

permeability
(mL/(m2*h*mmH
g)) Higher SC = Higher Albumin Loss

ß2- Albumin Loss

Microglobulin Grams
Low-flux 10-20 - zero
High-flux 200-400 0.7-0.8 0.2 0.4 HD
1-4 HDF
Medium cut-off 600-850 1.0 3-7
High cut-off 1100 1.0 > 7 HD
Molecule Rh [nm] Comments
69
β2 microglobulin calculated from the diffusion
1.7 coefficient in free solution

depending on its aggregation

Tumor necrosis factor (TNFα) 1.9 − 2.3 state, influenced by
concentration and pH

Free light chains (FLC) Stokes’ radius determined by

monomeric state (mostly κ- 2.3 chromatography
FLC)

Free light chains (FLC) Stokes’ radius determined by

dimeric form (mostly λ-FLC) 2.8 chromatography

Albumin 3.51 Maximum Pore Radius

 Performance of hemodialysis with MCO dialyzers
70
NDT Volume 32, Issue 1, 1 January 2017

MCO AA HD MCO BB HD High-flux HD HDF

Complement 63.0 (1.73)* 66.7 (1.73)* 32.9 (1.73) 46.3 (1.73)

factor D
Myoglobin 67.9 (2.34)* 71.6 (2.34)* 37.2 (2.34) 59.3 (2.37)
β2- 78.5 (1.32)**,**** 78.9 (1.32)** 73.5 (1.32) 80.6 (1.33)
microglobulin
Creatinine 73.5 (1.45) 73.2 (1.45) 71.7 (1.45) 73.7 (1.45)
Phosphate 52.8 (2.13) 48.8 (2.13) 48.4 (2.13) 51.0 (2.13)
Urea 80.7 (1.33)***** 80.3 (1.33)**** 79.4 (1.33) 81.6 (1.33
71
Medium Cut-Off (MCO) Membranes 72

Reduce Inflammation in Chronic Dialysis

Patients—A Randomized Controlled
Clinical Trial
 The randomized crossover trial in 48 patients compared
MCO-Ci dialysis to High-flux dialysis of 4 weeks duration
each plus 8 weeks extension phase

PLoS One. 2017; 12(1):

 73
p MCO vs HF
High-flux MCO
T=0 T = 4 weeks T=0 T = 4 weeks

Primary
endpoint

TNF-α mRNA 1.19 ± 0.57 1.02 ± 0.49* 0.92 ± 0.34 0.75 ± 0.31** < 0.001

IL-6 mRNA 0.86 ± 0.68 0.83 ± 0.67 0.78 ± 0.80 0.60 ± 0.43** 0.001

PLoS One. 2017; 12(1):

 74

High-flux MCO p MCO vs HF

T=0 T = 4 weeks T=0 T = 4 weeks

Albumin g/l 36.6 ± 3.2 37.5 ± 2.7 37.0 ± 3.6 35.3 ± 3.7** < 0.001

CRP mg/l 13.4 ± 25.5 9.6 ± 15.7 15.3 ± 30.0 9.3 ± 14.5 n.s

IL-6 pg/ml 9.8 ± 20.5 5.5 ± 4.5* 9.0 ± 13.2 6.0 ± 5.9** n.s.

PLoS One. 2017; 12(1):

Comparison of hemodialysis with medium 75

cut-off dialyzer and on-line

hemodiafiltration on the removal of small
and middle-sized molecules
 10 patients treated first with ol-HDF who were
thereafter switched to MCO-HD over a 1-year
period

Clin Nephrol. 2017 Aug 30

 Comparison of hemodialysis with medium cut-off 76
dialyzer and on-line hemodiafiltration on the removal
of small and middle-sized molecules
90
MCO- HD HDF
80
81.5 81.6
70

60
60 61
50

40

30

20

10

0
B2m RR % Myoglobin RR %

Clin Nephrol. 2017 Aug 30

Combined Therapy with Expanded HD 77

Expanded HD and HDF ?

Expanded HD and Extended HD ?
Expanded HD and Frequent sessions ?
Expanded HD and Incremental HD ?
Expanded HD membrane in HDF technique ? 78

The MCO Membrane should NOT be used in HDF

Loss of Free Hb
Increase will induce
High TMP
Albumin loss Blood leak
alarm

Innovative Clinical Approach in Dialysis. Contrib Nephrol. Basel, Karger, 2017, vol 191,
 Albumin Leakage during HD : 79

is it worth or Beneficial ?
27 April 2017

Implications of Albumin Leakage for Survival in

Maintenance Hemodialysis Patients: A 7-year
Observational Study 740 patients follow up
Data is toward accepted 3 gm loss / session
More RCTS is still needed
 80
Haemodialysis Prescription for Incident Patients:
Twice Seems Nice, But Is It Incremental?
American Journal of Kidney Diseases 2016

WE ALL DO INCRIMENTAL HD BEFORE THE TERMINOLOGY IS BORN

 Nephrol Dial Transplant (2015) 30 (10):
81
3 Phases Approach for
Augmented HD HD Dose presription
Frequency
Incre
Duration ase d
o
nece se when
Flux ssary
Evaluate
HDF Monthly
Incremental 82
hemodialysis
prescription with
adjustment of
hemodialysis dose

Expanded
or HDF
based on Residual
kidney function
Nephrol Dial Transplant (2015) 30 (10):
The progressive approach to HD 83

Progressive HD should be
considered a bridge between
conservative therapy and full
renal replacement therapy
(RRT).
A user-friendly tool for incremental
haemodialysis prescription ( 05
January 2018 )

‘SPEEDY’, by using the acronym of KRUn Native kidney normalized

urea clearance

its whole definition: Spreadsheet

eKt/V
for the Prescription of incrEmental
haEmoDalYsis. Session Length
SPEEDY Td
QB and Dialyzer
KoA

nPCR
84
 a l
e n
r
r y
fo a p
ls er
de h
o tt
r m e n
te em
u
p lac
o m p
c re

85
 86
HD Frequency
HD Time

The finding of two frequency peaks of

sudden death, immediately before and after
the first weekly haemodialysis session,
suggests that daily haemodialysis, or at
least the abolition of the long interdialytic
interval during the weekend, could be
helpful in reducing this kind of mortality.
Adverse Effects of Conventional Thrice-Weekly Hemodialysis: Is It Time
to Avoid 3-Day Interdialytic Intervals?

Am J Nephrol 2015;41:400-408

87
88
Hemodialysis Treatment Time: As Important as it Seems?
89
16 January 2017

largely avoid
ultrashort
dialysis
HD > 4 hours Parking Time
Dialysis Frequency versus Time : That is the Question 90
Extended-hours hemodialysis is associated with lower mortality risk in 91
patients with end-stage renal disease Kidney international December 2016Volume 90,
compare mortality risk among 1206
extended-hours hemodialysis had a individuals undergoing thrice weekly
33% lower adjusted risk of death extended-hours hemodialysis or 111,707
patients receiving conventional
Better compartmental hemodialysis treatments
dialysis

The crude mortality rate with extended-hours

hemodialysis was 6.4 deaths per 100 patient-
years compared with 14.7 deaths per 100 patient-
years for conventional hemodialysis

the follow-up period for each patient was

divided into successive 91-day periods from the
date of first dialysis; follow-up was available for
up to 20 periods
LIMITATIONS OF Frequent HD 92
More frequent hemodialysis does not effectively clear protein-bound azotemic solutes
derived from gut microbiome metabolism Kidney international May 2017Volume 91
Relative protein binding
Relative protein binding
Azotemic compound
(%)
p-Cresol glucuronide < 40
Hippuric acid 40–50
Phenylacetylglutamine 40–50
Indoxyl glucuronide 50–60
Phenylacetic acid 60–70
Indole acetic acid 90–95
p-Cresol sulfate 90–95
Indoxyl sulfate > 95
3-carboxy-4-methyl-5-
propyl-2-furanpropanoic > 95
acid
May 2017 Volume 91, Issue 5,

93
94
In conclusion 95
Secondary Measures of Dialysis Adequacy
Additional Measure Additional Measure
QOL
QOL
Secondary
Secondary Measure
Measure
B2m – UFR – etc
B2m – UFR – etc

Primary Measure Kt/V

Kt/V
 96

Pushing Towards
A Better Dialysis

To Know the Unknowns

Thank you