National Institute for Health and Care Excellence
Anaemia management in chronic kidney disease (update)
Stakeholder Comments – Draft Guideline
NOTE:
NICE is unable to accept comments from non-registered organisations or individuals.
If you wish your comments to be considered but are not a registered stakeholder, please register
via the NICE website or contact the registered stakeholder organisation that most closely
represents your interests and pass your comments to them.
Please fill in both the ‘stakeholder organisation’ and ‘name of commentator’ fields below in order
for your comments to be considered.
Stakeholder
Organisation:
Name of
commentator:
Order
number
Document
(For internal
use only)
Indicate if you
are referring to
the
Full version
NICE version
or the
Appendices
The Renal Association and Kidney Research UK
Iain Macdougall, Peter Storey
Page
Number
Line
Number
Comments
Indicate
number or
‘general’ if
your
comment
relates to the
whole
document
Indicate
number or
‘general’ if
your
comment
relates to
the whole
document
Please insert each new comment in a new row.
Please do not paste other tables into this table, as your
comments could get lost – type directly into this table.
Example
Full
16
45
Our comments are as follows ……
Proformas that are not correctly submitted as detailed in the example above may be returned to you.
NICE/Full
General
General
Clearly, an enormous amount of work has gone into the
creation of this updated CKD anaemia guideline, and
thanks and congratulations are due to those who have
spent hours and hours on this huge effort. The
methodological process is particularly impressive.
However, we have a number of concerns, which can be
summarised into misinterpretation of the evidence,
exclusion of relevant evidence, impracticalities of
adopting these guidelines in UK clinical practice, and
poor prioritisation of research recommendations. Further
details and explanations are shown below.
NICE/Full
General
General
We would also bring to your attention the current major
UK trial known as PIVOTAL. This trial aims to produce
evidence around safety in IV iron dosing and will address
all of the research questions arising from the KIDIGO
Controversies Conference on IV iron. The trial ref
numbers are – EUDRACT: 2013-002267-25 and UKCRN
ID: 15250.
Given our concerns about the interpretation of previous
trial findings in this guidance, we are keen to bring
PIVOTAL to your attention as an important research
initiative which should inform subsequent guideline
PLEASE NOTE: The Institute reserves the right to summarise and edit comments received during consultations, or
not to publish them at all, where in the reasonable opinion or the Institute the comments are voluminous,
publication would be unlawful or publication would be otherwise inappropriate.
development. Publication of trial results is anticipated in
2018.
NICE version
10
5
NICE version
10
General
NICE version
10
17
NICE version
NICE version
11
12
17
Paragraph 3
NICE version
NICE version
NICE version
NICE version
13
14
21
21
1:1:3
1:1:4
1:3:18
Last
paragraph
We do not agree with the recommendation that iron
status should be checked monthly in haemodialysis
patients. We are aware that there is no evidence either
way to inform this frequency of testing in this clinical
context, and that this then becomes a matter of opinion.
There are certainly some patients in whom monthly
testing is sensible, but for a large majority, this frequency
of testing is unnecessary. Most haemodialysis units in
the UK currently test iron status every three months, and
there would be logistic and economic implications in
increasing this frequency to monthly, in the absence of
any hard evidence that this is required.
We are aware of the evidence to support measurement
of hypochromic red cells as a marker of iron status,
based largely on Tessitore’s study from 2001. However,
this parameter is very prone to erroneous values if there
is any delay in the blood sample being analysed. In
Tessitore’s study, the blood samples were taken
immediately from the dialysis unit to the lab, which was
sited in close proximity to his dialysis unit. In many
satellite dialysis unit patients, there is a considerable
delay in the samples being analysed. Iain Macdougall
advises that his experience with this parameter is that in
a research study setting it performs excellently, but in
real clinical life it is too unreliable and impractical (N.B.
Iain’s view is that he isinversely conflicted with this
opinion – he introduced measurement of this parameter
to the global stage (Macdougall et al. Detection of
functional iron deficiency during erythropoietin treatment:
a new approach. BMJ 1992; 304: 225-226), but is also
very much aware of its limitations outside the research
setting). Furthermore, the availability of analysers to
perform this measurement is quite limited throughout the
UK. Suggesting CHr as an alternative suffers from the
same limitation, while other reticulocyte indices which are
similar to CHr (and have since become available), such
as Ret-He have not been validated properly in this
setting.
We do not fully agree with recommendation 1.1.4 that
“transferrin saturation or serum ferritin should not be
used alone to assess iron deficiency status in people with
anaemia of CKD”. If a patient has a rock bottom serum
ferritin level (e.g. 14 ug/l), then this patient is absolutely
iron-deficient, and there is no value or need to check the
transferrin saturation.
As above, for comment 2.
“When offering intravenous iron therapy to adults,
consider high-dose low-frequency” — it is not clear
whether this refers to both non-dialysis and
haemodialysis patients. If it includes the latter, then we
suggest there is no evidence to support this statement,
and there is certainly a large observational study to
suggest that this may be harmful (Brookhart et al, JASN
2013)
As above, for comments 2 and 3.
As above, for comment 4.
As above, for comment 3.
Not sure where the “600—1000 mg of iron” comes from.
Although this does not seem an inappropriate amount, it
PLEASE NOTE: The Institute reserves the right to summarise and edit comments received during consultations, or
not to publish them at all, where in the reasonable opinion or the Institute the comments are voluminous,
publication would be unlawful or publication would be otherwise inappropriate.
NICE version
22
1:3:19
NICE version
NICE version
22
23
1:3:20
1:3:23
NICE version
23
1:3:25
NICE version
NICE version
NICE version
24
24
27—30
1:3:27
1:4:2
General –
Research
Recommend
ations
NICE version
Full version
41—46
21
General
1:2:4
Full version
Full version
Full version
Full version
Full version
Full version
Full version
Full version
60
61
62
65
65
66
67
92
5—18
3—4
9—21
25—30
34—36
20—21
10—18
18—20
Full version
93
36—38
may be better to say 500—1000 mg, since the iron
preparations that may be used in this context come in
500 or 1000 mg vials.
“Once percentage hypochromic red blood cells are less
than 6%” should read “Once percentage hypochromic
red blood cells is less than 6%”.
As above, for comments 2 and 3.
We do not agree with this recommendation. Patients who
have end-stage renal failure and are on haemodialysis
are both erythropoietin and iron deficient. We are not
aware of any robust data showing that IV iron alone is
successful in correcting the anaemia in this clinical
scenario, and usually both ESA and iron supplementation
are required. To recommend trying IV iron first is likely to
result in patients suffering from anaemia for longer than
they need to, with very little chance of success.
We are not sure that oral iron should be offered to
haemodialysis patients who choose not to have IV iron.
The chances of this being of benefit are extremely slim
due to no or negligible absorption from the gut, and the
chances of side-effects are high. Clearly, patient choice
is important in this context, but we do not think the
health-care professional should be recommending this
approach.
As above, for comment 6.
As above, for comment 2.
While research into optimising anaemia management in
acute illness and conservative care would undoubtedly
be useful, this pales into insignificance in relation to the
safety of intravenous iron. This is more relevant to a far
larger patient group, and is relevant to all haemodialysis
patients. Ferritin levels in this patient population in the
UK are currently running at levels that have not been
proven to be safe, and there is considerable laboratory
and observational data that suggest that this could be
harmful to a large majority of chronic haemodialysis
patients. For this issue (which has been extensively
commented on in the literature recently) not to have even
been mentioned in the 2015 research recommendations
seems to be a significant omission.
As per comments above.
In line with scientific evidence that has come to light over
the last decade or so, upregulation of hepcidin activity
should be mentioned as contributing to the pathogenesis
of CKD anaemia.
As above, for comments 2 and 3.
As above, for comment 2.
As above, for comments 2 and 3.
As above, for comment 3.
As above, for comment 10.
As above, for comment 14.
As above, for comments 2 and 3.
This previous recommendation is still valid, is helpful,
and should not be deleted. Low serum ferritin levels are
100% diagnostic of absolute iron deficiency, and will
respond to effective iron replacement.
We agree with the GDG that serum hepcidin has not
shown any clinical value in the detection of iron
insufficiency; we are not clear as to why the GDG felt that
this was out of the scope of this guideline. If they are
critically reviewing the evidence for detection of iron
PLEASE NOTE: The Institute reserves the right to summarise and edit comments received during consultations, or
not to publish them at all, where in the reasonable opinion or the Institute the comments are voluminous,
publication would be unlawful or publication would be otherwise inappropriate.
Full version
Full version
Full version
108
109--117
214
20
top
14-15
Full version
Full version
214
228
6:12:6
Table
Full version
234
6.15.3.4
Full version
234
6.15.3.4
Full version
234
Full version
235
6.15.3.4,
5th last line
6.15.3.4
deficiency, then the paper by Tesssitore in 2010
(Nephrology Dialysis Transplantation 2010; 25: 39964002) provides stronger evidence (albeit negative) than
any paper that supports the use of Ret He in the
monitoring of iron status in people with CKD, and yet the
GDG seem happy to make statements recommending
the latter novel reticulocyte parameter.
As above, for comment 3.
As above, for comments 2 and 3.
We do not agree with the statement “that lack of
response to ESA (hypo-responsiveness) may be partly
due to any discontinuation of iron”. It is now fairly clear
that inflammatory cytokines directly suppress erythroid
cell proliferation in the bone marrow, and IL-6-generated
upregulation of hepcidin activity causes iron to be locked
away in the reticuloendothelial system. We do not believe
that discontinuation of iron therapy contributes
meaningfully to ESA hyporesponsiveness.
As above, for comment 17.
“Macdougall” is mis-spelt. Also, the superscript ‘214’ is
awkwardly split.
“Very low quality evidence from one study214 (n=461)
suggested that oral iron performed better compared with
high-dose/low-frequency IV iron when considering
numbers of people receiving blood transfusion(s) and
numbers of people requiring initiation of ESA therapy”.
We do not think there is any evidence from this study to
support this statement (see Appendix for raw data), and
there is also a methodological flaw in making this
assertion. The study was neither designed nor powered
to show this, and more importantly, patients randomised
to the oral iron group were more likely to then be treated
with intravenous iron if they remained anaemic. Many of
these subjects responded quite well to IV iron and did not
need to receive ESA therapy, making it appear (falsely)
as if oral iron was keeping patients off ESA therapy. In
contrast, patients randomised to intravenous iron moved
on to ESA therapy, since they had already exhausted
any benefit from iron management. This therefore
introduced a bias in the oral iron group, rendering any
such analysis meaningless. Thus, interpretation of the
primary endpoint should include the full composite, and
looking at the individual components, as has been
attempted in the draft NICE guidance is scientifically
flawed.
“With regard to other important iron parameters, such as
measures of SF and TSAT, moderate to low quality
evidence suggested little clinically important difference
between the IV iron therapy and oral iron therapy”. We
do not consider the differences in serum ferritin achieved
between the high-dose IV iron group and the oral iron
group in the FIND-CKD study214 to be clinically
insignificant (see Appendix figure).
“the control group were observed” should read “the
control group was observed”…..
“Very low quality evidence from one study214 (n=461)
suggested that low-dose/high-frequency IV iron was
preferable when considering numbers of people requiring
blood transfusion(s), but low quality evidence from the
same study suggested that oral iron was preferable when
considering numbers of people needing to initiate ESA
therapy”. For the same reasons outlined in comment 34
PLEASE NOTE: The Institute reserves the right to summarise and edit comments received during consultations, or
not to publish them at all, where in the reasonable opinion or the Institute the comments are voluminous,
publication would be unlawful or publication would be otherwise inappropriate.
above, this analysis and statement is scientifically flawed.
Full version
236
6.15.4.2
Full version
253
Full version
Full version
254
260
Four lines
above Table
92.
Table
6.15.5.4
Full version
Full version
Full version
Full version
262
270
272
280
6.15.6
6.15.7
Lines 30-31
Line 34
Full version
Full version
284
285
11th last line
Recommend
ation 1
Full version
Full version
286
296
Full version
296
7.5.6
Reference
List
Reference
List
“We searched for randomised trials comparing the
effectiveness of different iron preparations given as either
IV low-dose/high-frequency, IV high-dose/low-frequency
or oral regimens in people receiving ESA”. This reference
appears not to have been used:
Macdougall et al. A randomized comparison of
ferumoxytol and iron sucrose for treating iron deficiency
anaemia in patients with CKD. Clin J Am Soc Nephrol
2014; 9: 705-712.
“hyposensitive” should read hypersensitive”.
“Macdougall” is mis-spelt.
“All-cause mortality rates were lower in people receiving
low-dose/high-frequency IV iron compared with oral iron,
but this a highly uncertain finding based on low quality
evidence214 (n=460). The FIND-CKD study was neither
designed nor powered to examine mortality, thus we do
not think that this statement is valid and may be a
dangerous assumption.
As above, for comment 14.
As above, for comment 17.
As above, for comment 2.
We do not think in 2015 that aluminium toxicity should be
considered as a cause of ESA resistance. For practical
purposes, this complication in dialysis patients has now
become extinct.
“Mircera” is mis-spelt.
We do not think in 2015 that aluminium toxicity should be
considered as a cause of anaemia in ESRD. For practical
purposes, this complication in dialysis patients has now
become extinct.
As above, for comment 17.
For reference 1, the “Canadian Erythropoietin Study
Group” should appear before the title of this paper.
In quite a number of the references, the number of
authors listed appears fairly random. Moreover, the use
of ‘et al’ also seems fairly random, e.g. in reference 210
only the first two authors are included, with no ‘et al’,
while many other references have more authors included
with or without ‘et al’.
Please add extra rows as needed.
Please email this form to: AMCKD@nice.org.uk
Closing date: 5pm 8th January 2015
PLEASE NOTE: The Institute reserves the right to summarise and edit comments received during consultations, or
not to publish them at all, where in the reasonable opinion or the Institute the comments are voluminous,
publication would be unlawful or publication would be otherwise inappropriate.
Appendix
PLEASE NOTE: The Institute reserves the right to summarise and edit comments received during consultations, or
not to publish them at all, where in the reasonable opinion or the Institute the comments are voluminous,
publication would be unlawful or publication would be otherwise inappropriate.
PLEASE NOTE: The Institute reserves the right to summarise and edit comments received during consultations, or
not to publish them at all, where in the reasonable opinion or the Institute the comments are voluminous,
publication would be unlawful or publication would be otherwise inappropriate.