Introduction

Blood pressure (BP) remains one of the most important risk factors for the development or progression of chronic kidney disease (CKD) in patients not requiring renal replacement therapy [ 1,2 ]. The link between achieving better BP control and improving renal outcomes (and thereby directly and indirectly improving overall outcomes) is clear for all to see; it is one of the areas where although the evidential base is not over-whelming in volume, or quality, our clinical practice guidelines are strongest [ 3 ] and dissent least.

This near unanimity of understanding and cohesion of practice, which we see for CKD stages 3 and 4, is something that, however, does not extend to patients who are already receiving chronic renal replacement therapy (CKD stage 5D). For this type of patient, there is much less certainty, far less is known, and as a result, we see greater variation in clinical practice [ 4 ]. We know that there are precious few interventions, which have been shown (by means of a randomized controlled trial—RCT), to influence beneficially the dreadful cardiovascular mortality in this cohort [ 5 ]. Furthermore, we also know that there is ‘reverse causality’ when we try to associate risk factors with outcomes (such that obesity is ‘favoured’ and cholesterol elevation linked to better outcomes than are reduced cholesterol levels [ 6 ]). In fact increasingly strong evidence linking a lack of effect between cholesterol-lowering strategies and outcome stands as a worrisome perspective for BP; one has to be reminded that in non-CKD cohorts and from meta-analyses of trials involving some mild-to-moderate CKD cohorts, a CV benefit for using statins was suggested, but in CKD stage 5D we now have two large RCTs [ 7,8 ] which indicate no benefit whatsoever from deploying statins in dialysis subjects.

And the same is in essence true for BP, as despite the importance of BP elevation in driving the development and progression of pre-dialytic CKD, there are many epidemiological studies which strongly suggest that the link between low BP and poor outcome in CKD stage 5D is stronger than that between high BP and poor outcome, strikingly so in fact. There is a concept of the ‘J-curve’ where at some point the benefit from a therapy, such as BP reduction, is balanced by the harm ensuing from BP reduction (e.g. under-perfusion of vital organs). This is controversial and not accepted as valid by all commentators, but its absence needs to be demonstrated by trials and by careful interventional assessments in a patient population so potentially clinically fragile as the typical dialysis cohorts are. We must also recognize the difference between low BP achieved by ‘intervention’, which may be of benefit, and low BP achieved because of for example a serious underlying cardiomyopathy, or as a result of some haemodynamic insult which arises on dialysis, where the low BP is a marker of, or may result in, poor critical organ function. This mirrors the debate about haemoglobin (Hb) levels—spontaneous low ESA requirements or high Hb may have different clinical implications from therapeutically driven high Hb levels.

Translating epidemiological pointers into solid interventional protocols

Epidemiological clues, however, need to be confirmed or refuted by RCTs—unfortunately, in so far as data suggest a possible link between reduction in BP and increased incidence of strokes in patients with CKD. Toyoda and colleagues showed, in a careful 22-year single-centre follow-up study, a worrying association between dialysis sessions and cerebrovascular accidents [ 9 ]. Very recently, Sozio et al. have reported the CHOICE study [ 10 ] where careful observation of stroke incidence was made in dialysis patients from many centres; the rate of stroke was very significant, there were more ischaemic than haemorrhagic strokes, but importantly, comparing the BP and cholesterol values of patients with and without stroke, there was no difference. Actually, the pulse pressure (by virtue of LOW diastolic BP) was greater in patients who developed stroke than in those who did not. Only if we carefully follow selected patient cohorts (such as Tassin or Manchester—[ 11 ]), do we then find that over extended periods of time there is a tendency for higher BP to be associated with worse outcomes. The big confounding factor here, however, is that these are ‘survivor’ cohorts, including patients with fewer comorbidities who are often dialyzed and managed in a distinct way—differing greatly from current practice in mainstream RRT programmes. These apparently conflicting reports serve as a warning that automatic acceptance of the possibility that BP treatment must automatically be followed by improved outcomes, in all patients over their whole dialytic vintage, may be a worrisome simplication [ 12 ].

Salt and water overload is a key promoter of raised BP in dialysis patients [ 13 ], while at the same time, increased sympathetic nervous system activity [ 14 ], increased activity of the renin–angiotensin–aldosterone system [ 15 ] and reduced biological activity of the NO systems [ 16 ], all conspire to cause increased vasoconstriction and impaired vasorelaxation. As a result of these complementary mechanisms conspiring to raises BP, we theoretically have a spectrum of ‘therapeutic options’, which reflect these aetiological factors by either tackling salt and water by dialysis and diet, or BP levels directly by using cardiovascular-acting drugs. Unlike the pre-dialysis CKD population, there is another concern for CKD stage 5D patients: whatever measures we employ to control inTER-dialytic BP must not at the same time increase the tendency to inTRA-dialytic haemodynamic instability [ 17 ].

The nature of the problem

There are three huge obstacles for an accepted cohesion of practice in CKD 5D. First, as pointed out, we still have to prove that reducing BP levels is beneficial and not harmful. Secondly, we have to answer to the simple question of ‘which targets to aim for?’ This is really harder than it sounds, as we do not really know when is it best to take the BP in the context of haemodialysis sessions (before, after, in between, etc). Thirdly, but not least, we should decide on the ‘perfect recipe’: how much less salt (by diet and removal), how many, when and which antihypertensive medications to use? This is again much tougher than it appears; we do not yet have a consensus on the best way to assess ‘optimal’ or ‘dry’ weight—and indeed, what exactly these phrases mean. This is so different from what appears to be the case for subjects without kidney disease, with or without significant cardiovascular disease burdens, as a very recent mega-meta-analysis of nearly 1 000 000 subjects has reported. Law et al. showed [ 18 ] that any form of BP reducing therapy profoundly reduced cardiovascular outcome burdens (acute coronary syndromes, heart failure and stroke). Given the appalling burden that these malign outcomes impose on patients with kidney disease, it is disappointing to conclude that we do not know the right therapeutic interventions to bring to bear and that our evidence base is miniscule by comparison (these two facts of course are connected).

This complexity of outcome is mirrored by a difficulty in choosing a representative BP for risk assessment in patients receiving three haemodynamically challenging dialysis treatment sessions each week. It is clear from many recent studies that pre-dialysis BP, though convenient to measure, is a poor guide to true BP values; this has recently been definitively shown by Agarwal et al. , from whose work we can see that ambulatory, and home, BP values are more predictive of outcomes than those office BP measurements associated with dialysis sessions [ 19 ]. A very recent post-hoc report from Dialysis Morbidity and Mortality Wave 2 Study [ 20 ] suggests that BP behaviour during dialysis (in this case, a clinically significant rise) is associated with adverse outcome in incident dialysis patients; but this was found only in those patients whose pre-dialysis BP was NOT elevated—clearly making the point that in assessing BP completely in dialysis cohorts we need to know the pre, peri and post dialysis BP trends. One can add to this the realization that increased pulse pressure typically characterizes patients with diabetes, or CKD, and especially patients with CKD stage 5D, a facet of a marked increase in vascular stiffening [ 21 ]. Use of aortic rather than brachial artery BP values (as in the CAFÉ substudy of ASCOT [ 22 ]) would be more likely to take a full account of the abnormal ventriculo-arterial coupling characteristics which we know now strongly to be associated with adverse outcomes in dialysis patients [ 23 ].

So, given all of these considerations, and problems, it would take several well-conducted placebo-controlled double-blinded trials in CKD stage 5D patients, establishing the utility and feasibility of BP-lowering therapy as judged by reduction in hard/relevant endpoints before we could universally embrace the concept of safe BP reduction for all patients regardless of their age, clinical complexity or dialysis vintage. Let alone have begun to understand whom to treat, when, at what BP level to intervene and what BP level to aim for. There is at least the theoretical possibility that for some patients, for some of the time, a less stringent approach to BP reduction might be preferable. This is not an issue for the non-CKD population where such distinctions do not appear important [ 18 ]. The clear dilemma this poses Guidelines groups is shown by the fact that in the UK, the 3rd Clinical Standards document from the UK Renal Association [ 24 ], which recommended 140/90 mmHg pre and 130/80 mm Hg post dialysis, was superseded by the 4th Clinical Practice Guidelines, where no BP target value was offered, on account of the lack of available evidence of clinical benefit [ 25 ].

Is meta-analysis better analysis?

Entering into this confused picture, we have what at first sight might appear to be a solution in 2009—not one, but two, separate meta-analyses [ 26,27 ] of the few existing randomized trials of BP alteration in CKD stage 5D patients! The two meta-analytic approaches are rather different, and surprisingly, the conclusions of these new statistical offerings have some important differences too, which we believe both highlight the benefits and the dangers of using this statistical approach to compensate for an intrinsically weak and flawed evidence base.

The first meta-analysis is by Heerspink et al. [ 26 ]. Their main conclusion was that agents that lower BP should routinely be considered for individuals undergoing dialysis (as they detected a positive effect on overall mortality)—just as Law's paper concluded albeit with around 500 times the subjects included and in a different population [ 18 ]. This seems to us to be a sweeping statement to be applied to over 1,500,000 patients on dialysis worldwide, based on eight small studies of 1679 subjects. One potentially unwanted consequence of the Heerspink meta-analysis conclusions is the concept that using cardiovascularly active drugs, classified as antihypertensives (i.e. with a tendency to lead to BP reduction), is important irrespective of BP levels for all patients, including those with severely compromised LV function or in the increasingly very elderly patients at a risk of cerebral/cardiac hypoperfusion. Before advocating a blanket use of such drugs in these diverse clinical settings, some careful safety tolerability and feasibility analyses need to be done—it is one thing to employ these drugs in a careful trial setting, with titration and frequent clinical examination, but potentially quite another to use these interventions as normal clinical care, without similar care and time expended in monitoring.

The second meta-analysis was by Agarwal and Singh [ 27 ], who in their effort based on fewer trials and subjects showed a CV benefit, which was confined to hypertensive subjects and tempered their recommendations accordingly. An additionally confusing point is the huge clinical heterogeneity between the studies included in both meta-analyses—some were primary prevention, some were secondary prevention, some included subjects regardless of BP levels, others only hypertensive subjects, some required LVH, or LVF as inclusion criteria, while others did not discriminate on the grounds of LV morphology or function (see 25, 26 and Table 1 ). There was no such recorded heterogeneity in Law et al. 's meta-analysis [ 18 ]. Furthermore, some trials examined BB, some CCB, some ACEIs and some ARBs. There was no ‘target’ BP in many studies; as indeed, the entry or achieved BP levels were not regarded as relevant to the trial outcomes. The end-points of the individual studies also differed (different CV composite bundles of outcomes). There is just one consistent feature: in none of the eight individual studies were there any reports of harm derived from use of cardiovascularly active interventions. Finally, it is important to underline that trial inclusion criteria can produce cohorts whose clinical status may differ from the majority of dialysis patients. In Table  1 , we summarize the main points from both meta-analyses, comparing and contrasting key factors in their design and approach. These two meta-analyses have been commented on in their respective journals [ 28,29 ].

Table 1

The main points from meta-analyses of Heerspink and Agarwal and Sinha

  Heerspink et al. [ 26 ]   Agarwal and Sinha [ 27 ]  
Publication date February 2009 March 2009 
Primary source of data Pubmed, Embase, Cochrane, www.ClinicalTrials. gov Pubmed, Cochrane, Bibliographic review 
Date covered 1950–11/2008 1/1996–10/2008 
Search criteria Renal dialysis, kidney failure, cardiovascular disease, anti-hypertensive drugs Hypertension, dialysis, published study 
Identified screened abstracts 1179 195 
Rejected abstracts 1152 166 
Detailed assessment 28 29 
Accepted abstracts 
 Cice 2003 
 Li 2003 
 Cice 2006 
 Takahashi 2006 
 Zannad 2006 
 Nakao 2007 
 Suzuki 2008 
 Tepel 2008 
Clinical heterogeneity 
 Raised BP in all cases? 
 LVH pre-requisite? 
 LVF pre-requisite? 
 ACE/ARB 
 BB 
 CCB 
Patients included 1679 1202 
Cardiovascular events 495  
Conclusions/Confidence Intervals values 
 CV events 0.71 (0.55–0.92) 0.69 (0.54–0.84) 
 Overall mortality 0.80 (0.66–0.96)  
 CV mortality 0.71 (0.50–0.99)  
Outcome analysed by No Yes and dissimilar 
 Hypertension   
  Normotension 0.86 (0.67–1.12) 
  Hypertension 0.49 (0.35–0.67) 
Heterogeneity analysis Negative  Positive ( P < 0.006)  
Publication bias No Yes 
  Heerspink et al. [ 26 ]   Agarwal and Sinha [ 27 ]  
Publication date February 2009 March 2009 
Primary source of data Pubmed, Embase, Cochrane, www.ClinicalTrials. gov Pubmed, Cochrane, Bibliographic review 
Date covered 1950–11/2008 1/1996–10/2008 
Search criteria Renal dialysis, kidney failure, cardiovascular disease, anti-hypertensive drugs Hypertension, dialysis, published study 
Identified screened abstracts 1179 195 
Rejected abstracts 1152 166 
Detailed assessment 28 29 
Accepted abstracts 
 Cice 2003 
 Li 2003 
 Cice 2006 
 Takahashi 2006 
 Zannad 2006 
 Nakao 2007 
 Suzuki 2008 
 Tepel 2008 
Clinical heterogeneity 
 Raised BP in all cases? 
 LVH pre-requisite? 
 LVF pre-requisite? 
 ACE/ARB 
 BB 
 CCB 
Patients included 1679 1202 
Cardiovascular events 495  
Conclusions/Confidence Intervals values 
 CV events 0.71 (0.55–0.92) 0.69 (0.54–0.84) 
 Overall mortality 0.80 (0.66–0.96)  
 CV mortality 0.71 (0.50–0.99)  
Outcome analysed by No Yes and dissimilar 
 Hypertension   
  Normotension 0.86 (0.67–1.12) 
  Hypertension 0.49 (0.35–0.67) 
Heterogeneity analysis Negative  Positive ( P < 0.006)  
Publication bias No Yes 

One additional important factor lies in the nature of hypertension itself in a typical dialysis population. We speculate that there may be (at least) two different ways to acquire elevated BP—one might even describe these as ‘BP phenotypes’: a ‘salt-dominant’ one, where we contend that targeted use of newly developed and refined bioimpedance techniques [ 30,31 ] could facilitate a comprehensive and reliable clinical evaluation, a real physiologically meaningful target, and a quantifiable means to achieve it. The second is a ‘resistance dominant’ phenotype, which, if properly identified, dissected and tracked, might actually contain a more predictable BP–mortality relationship. It is potentially the aetiological admixture of the two different phenotypes, and our inability reliably to differentiate them, which we believe might be a key confounder in interpreting BP-related outcomes and for planning effective interventions. This might spur further characterization of elevated BP and better clinical context for its evaluation—for example, as now, elevated BP with or without LVH, and we believe, we should add for the future, elevated BP with or without volume expansion, vasoconstriction, increased vessel stiffness. In both putative ‘phenotypes’, rigorous salt restriction would most likely continue to be highly sensible, if very hard to achieve in practice. However, knowing more about the aetiology of BP elevation in individual patients might also present the opportunity for more individualized interventions.

Final thoughts

We tend to err on the ‘side’ of Agarwal and Singh's view expressed in their manuscript, from which one can tentatively conclude or surmise that selected hypertensive subjects may actually benefit from BP reduction (howsoever achieved using drugs). It is now a matter of major importance that the very many uncertainties surrounding this important area of clinical practice—how and when to measure BP, how and when to try to bring BP down, and by how much—should individually be subjected to RCTs, which then in time, may lead to more insightful and applicable conclusions from subsequent meta-analyses. Given the ‘life-long’ nature of CKD, patients can and often do start with CKD, some progress to needing dialytic therapies and a lucky subset then are transplanted. Raised BP is common to all three clinical situations (itself perhaps not a surprise as the aetiopathogenesis of raised BP in these situations is crucially different). There is no guarantee that the management of his cardiorenal risk factor would be the same at these different stages of the patient pathway—a very recent meta-analysis from the Cochrane group in Australia of the (again) very small evidence base for the treatment of raised BP in renal transplant recipients comes to a conclusion which differs from that recommended in CKD or in CKD-5D—namely that dihydropyridine calcium-channel blockers are preferred, at least when mainly short-term kidney endpoints are considered [ 32 ].

Recently in early 2009, over 50 nephrologists and BP experts convened in New York as part of a K-DIGO ‘Controversies Conference’ focusing on treatment of BP in CKD Stage 5D patients [http://www.kdigo.org/meetings_ events/BP_Controversies_Conference.php]. This conference will in time generate a ‘Position Statement’ to which we can all eagerly look forward. For now, though, it will be remain the decision of the nephrologist and his or her clinical team to try to establish what the best way forward for individual patients with BP elevation on dialysis might be.

Conflict of interest statement. DG has received speaking honoraria and consultancy fees from Merck, Sanofi-Synthalabo, Pfizer and Takeda. All of those companies make BP lowering drugs, and DG has spoken on the value of BP lowering for these companies.

References

1
Jafar
TH
Stark
PC
Schmid
CH
, et al.  . 
AIPRD Study Group
Progression of chronic kidney disease: the role of blood pressure control, proteinuria, and angiotensin-converting enzyme inhibition: a patient-level meta-analysis
Ann Intern Med
 , 
2003
, vol. 
139
 (pg. 
244
-
252
)
2
Mentari
E
Rahman
M
Blood pressure and progression of chronic kidney disease: importance of systolic, diastolic, or diurnal variation
Curr Hypertens Rep
 , 
2004
, vol. 
6
 (pg. 
400
-
404
)
3
National Collaborating Centre for Chronic Conditions
Chronic kidney disease: national clinical guideline for early identification and management in adults in primary and secondary care
  
London: Royal College of Physicians, September 2008.
4
Pickering
TG
Target blood pressure in patients with end-stage renal disease: evidence-based medicine or the emperor's new clothes?
J Clin Hypertens
 , 
2007
, vol. 
8
 (pg. 
369
-
375
)
5
Covic
A
Gusbeth-Tatomir
P
Goldsmith
D
Negative outcome studies in end-stage renal disease: how dark are the storm clouds?
Nephrol Dial Transplant
 , 
2008
, vol. 
23
 (pg. 
56
-
61
)
6
Flanders
WD
Augestad
LB
Adjusting for reverse causality in the relationship between obesity and mortality
Int J Obes
 , 
2008
, vol. 
32
 
Suppl 3
(pg. 
S42
-
S46
)
7
Wanner
C
Krane
V
März
W
, et al.  . 
German Diabetes and Dialysis Study Investigators. Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis
N Engl J Med
 , 
2005
 
Jul 21; 353(3): 238–48. Erratum in: N Engl J Med 2005 Oct 13; 353(15): 1640
8
Fellström
BC
Jardine
AG
Schmieder
RE
, et al.  . 
AURORA Study Group
Rosuvastatin and cardiovascular events in patients undergoing hemodialysis
N Engl J Med
 , 
2009
, vol. 
360
 (pg. 
1395
-
1407
)
9
Toyoda
K
Fujii
K
Fujimi
S
, et al.  . 
Stroke in patients on maintenance hemodialysis: a 22-year single-center study
Am J Kidney Dis
 , 
2005
, vol. 
45
 (pg. 
1058
-
1066
)
10
Sozio
SM
Armstrong
PA
Coresh
J
, et al.  . 
Cerebrovascular disease incidence, characteristics, and outcomes in patients initiating dialysis: the choises for healthy outcomes in caring for ESRD (CHOICE) study
Am J Kidney Dis
 , 
2009
, vol. 
54
 (pg. 
468
-
477
)
11
Charra
B
Calemard
E
Guy
L
Importance of treatment time and blood pressure control in achieving long-term survival on dialysis
Am J Nephrol
 , 
1996
, vol. 
16
 (pg. 
35
-
44
)
12
Weiner
DE
Tighiouart
H
Levey
AS
, et al.  . 
Lowest systolic blood pressure is associated with stroke in stages 3 to 4 chronic kidney disease
J Am Soc Nephrol
 , 
2007
, vol. 
18
 (pg. 
960
-
966
)
13
Charra
B
Fluid balance, dry wieght, and blood pressure in dialysis
Hemodial Int
 , 
2007
, vol. 
11
 (pg. 
21
-
31
)
14
Klein
IHHT
Ligtenberg
G
Neumann
J
Sympathetic nerve activity is inappropriately increased in chronic renal disease
J Am Soc Nephrol
 , 
2003
, vol. 
14
 (pg. 
3239
-
3244
)
15
Gross
E
Rothstein
M
Dombek
S
, et al.  . 
Effect of spironolactone on blood pressure and the renin-angiotensin-aldosterone system in oligo-anuric hemodialysis patients
Am J Kidney Dis
 , 
2005
, vol. 
46
 (pg. 
94
-
101
)
16
Kusano
E
Akimoto
T
Hemodialysis and nitric oxide
J Artif Organs
 , 
2001
, vol. 
4
 (pg. 
423
-
429
)
17
Santos
S
Mendes
R
Profile of Interdialytic blood pressure in hemodialysis patients
Am J Nephrol
 , 
2003
, vol. 
23
 (pg. 
96
-
105
)
18
Law
MR
Morris
JK
Wald
NJ
Use of blood pressure lowering drugs in the prevention of cardiovascular disease: meta-analysis of 147 randomised trials in the context of expectations from prospective epidemiological studies
BMJ
 , 
2009
, vol. 
338
 pg. 
b1665 [epub ahead of print]
 
19
Agarwal
R
Kariyanna
SS
Light
RP
Prognostic Value of Circadian Blood Pressure Variation in Chronic Kidney Disease
Am J Nephrol
 , 
2009
, vol. 
30
 
6
(pg. 
547
-
553
)
20
Inrig
JK
Patel
UD
Toto
RD
, et al.  . 
Association of Blood Pressure Increases During Hemodialysis With 2-Year Mortality in Incident Hemodialysis Patients: A Secondary Analysis of the Dialysis Morbidity and Mortality Wave 2 Study
Am J Kidney Dis
 , vol. 
54
 (pg. 
881
-
890
)
21
London
GM
Marchais
SJ
Guerin
AP
, et al.  . 
Arterial structure and function in end-stage renal disease
Nephrol Dial Transplant
 , 
2002
, vol. 
17
 (pg. 
1713
-
1724
)
22
Williams
B
Lacy
PS
Thom
SM
, et al.  . 
CAFE Investigators; Anglo-Scandinavian Cardiac Outcomes Trial Investigators; CAFE Steering Committee and Writing Committee
Differential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes: principal results of the Conduit Artery Function Evaluation (CAFE) study
Circulation
 , 
2006
, vol. 
113
 (pg. 
1213
-
1225
)
23
Antonini-Canterin
F
Carerj
S
Di Bello
V
, et al.  . 
Arterial stiffness and ventricular stiffness: a couple of diseases or a coupling disease? A review from the cardiologist's point of view
Eur J Echocardiogr
 , 
2009
, vol. 
10
 (pg. 
36
-
43
)
24
Standards and Audit Measures—Third Edition Renal Association
Treatment of Adults and Children with Renal Failure: Standards and Audit Measures
 , 
2002
3rd edn
London
Royal College of Physicians of London and the Renal Association
 
25
 
The fourth (current) edition of the Renal Association Clinical Practice Guidelines. http://www.renal.org/pages/pages/guidelines/current/haemodialysis.php
26
Heerspink
HJ
Ninomiya
T
Zoungas
S
, et al.  . 
Effect of lowering blood pressure on cardiovascular events and mortality in patients on dialysis: a systematic review and meta-analysis of randomised controlled trials
Lancet
 , 
2009
, vol. 
373
 (pg. 
1009
-
1015
)
27
Agarwal
R
Sinha
AD
Cardiovascular protection with antihypertensive drugs in dialysis patients. systematic review and meta-analysis
Hypertension
 , 
2009
, vol. 
5
 (pg. 
860
-
866
)
28
Goldsmith
DJ
Covic
A
Blood pressure in dialysis patients-look before we leap
Lancet
 , 
2009
, vol. 
373
 pg. 
1945
 
29
Goldsmith
DJ
Covic
AC
Meta-analysis of the effects of treating blood pressure on cardiovascular outcomes of dialysis patients
Hypertension
 , 
2009
, vol. 
54
 pg. 
e6
 
30
Wabel
P
Moissl
U
Chamney
P
, et al.  . 
Towards improved cardiovascular management: the necessity of combining blood pressure and fluid overload
Nephrol Dial Transplant
 , 
2008
, vol. 
23
 (pg. 
2965
-
2971
)
31
Wizemann
V
Wabel
P
Chamney
P
, et al.  . 
The mortality risk of overhydration in haemodialysis patients
Nephrol Dial Transplant
 , 
2009
, vol. 
24
 (pg. 
1574
-
1579
)
32
Cross
NB
Webster
AC
Masson
P
, et al.  . 
Antihypertensives for kidney transplant recipients: systematic review and meta-analysis of randomised controlled trials
Transplantation
 , 
2009
, vol. 
88
 (pg. 
7
-
18
)

Comments

0 Comments