Tải bản đầy đủ - 0 (trang)
8 Hypertension, Cardiac Morphology and Progression of CKD

8 Hypertension, Cardiac Morphology and Progression of CKD

Tải bản đầy đủ - 0trang


J. Cunningham

VDR activators, though not by manipulation of dietary calcium and phosphorous

such as to normalise the circulating concentrations of these minerals [32]. This

remains an incomplete story, though the available data, still lacking positive outcomes in patients enrolled in to RCT’s, has led some to propose hypertension as

another indication for vitamin D replacement [33].

Cardiac morphology is floridly deranged in many subjects with CKD and this has

prompted investigation of the effects of vitamin D on this potentially critical aspect

of cardiovascular health. In children Patange et al. showed by multiple regression

analyses that 25-hydroxyvitamin D and systolic blood pressure were independent

predictors of increased left ventricular mass index [34]. The same authors had previously shown an association between low 25-hydroxyvitamin D and arterial stiffness,

providing a plausible explanation the cardiac changes described a year later [21]. A

small prospective study of 26 non diabetic subjects with CKD brachial artery flow

mediated dilation and pulse wave velocity and documented favourable effects on

both in response to two large doses of cholecalciferol [18]. The PRIMO study was

an RCT designed to probe the effects of paricalcitol on cardiac structure as assessed

by MRI and found no influence on the primary endpoint, namely left ventricular

mass index [35]. In a post hoc analysis of the same study it was reported that paricalcitol treated patients apparently benefited from reduced left atrial volume and

attenuated brain natriuretic peptide rises [36]. A different type of study in children

looked at associations between nutritional vitamin D status and structural cardiac

parameters, including left ventricular mass index (LVMI) and identified

25-hydroxyvitamin D as an independent predictor of LVMI [34].

de Borst et al. [37] reviewed studies examining an important related matter,

namely renal protection. This is relevant because local RAAS activation is a major

determinant of renal damage. The surrogate examined was proteinuria and randomised controlled trials of active vitamin D compounds in patients with CKD

reporting effects on proteinuria were included, provided the samples size was

greater than 50. Of 904 citations retrieved, only 6 were ultimately included (4 using

paricalcitol and 2 calcitriol) providing data on a total of 688 patients. Most patients

(84 %) received RAAS blockade throughout the study. Active vitamin D compounds

reduced proteinuria by a mean of 16 % (95 % CI of 13–18 %), with no effect seen in

controls. There was no evidence of superiority of either one of the studied active

compounds, calcitriol and paricalcitol, over the other. Nor did studies using higher

doses of paricalcitol show any additional benefit. There was a fairly good level of

consistency across the studies with no evidence of a greater or lesser effect in diabetic nephropaths. Examination of the relationship between vitamin D parameters

(25-hydroxyvitamin D and calcitriol) showed that low concentrations of either were

associated with increased odds of having albuminuria [16] and in a prospective

observational study investigators found that treatment of diabetics with nephropathy

with cholecalciferol significantly decreased albuminuria and urinary TGF-β1 at 2

and 4 months [38]. A similar phenomenon was also seen in the VITAL study in

which Type 2 diabetics with secondary hyperparathyroidism and albuminuria were

treated with 1–2 mcg of paricalcitol, or placebo, per day for 24 weeks [39]. Subjects

on the higher dose showed the expected reduction of parathyroid hormone (PTH)


Vitamin D and Mortality Risk in Chronic Kidney Disease


and also an 18 % reduction of albuminuria (p < 0.014). A post hoc analysis of in

children (Effect of Strict Blood Pressure Control and ACE Inhibition on Progression

of CKD in Paediatric Patients (ESCAPE) suggested that vitamin D sufficiency is

associated with less proteinuria and slower progression compared with those with

lower 25-hydroxyvitamin D concentrations [40]. These concepts are important

because hypertension, proteinuria and level of GFR all provide potentially important links between vitamin D status and cardiovascular outcomes in normal and

CKD populations [41, 42].


Cardiovascular Events and Mortality

Most deaths in CKD patients have a cardiovascular aetiology and in many of the

studies suggesting beneficial non classical effects of vitamin D treatment in CKC,

parallel improvements in total mortality, cardiovascular mortality, and non fatal cardiovascular events are seen. This adds weight to the general notion that the effects

of vitamin D on survival are likely to be mediated by its actions on the cardiovascular system. A role for reduction of PTH is plausible, but not well supported by the

data in which the apparent benefit of VDRA treatment was seen across the full

range of PTH, calcium and phosphate, thereby including subgroups of patients in

whom a relative contra-indication to VDRA treatment exists [4, 9]. Furthermore,

effective lowering of PTH in response to calcimimetic treatment has not been convincingly associated with mortality reductions, at least in the ESRD population

[43]. These observations support the view that any survival gain attributable to vitamin D is mediated by other means, possibly by direct action on cardiovascular and

other tissues expressing the VDR [13].

After the initial publications showing apparently beneficial effects of active

VDRA compounds [3, 4], there was a profusion of reports showing similar findings.

Some of these were in incident dialysis populations [5] and others in pre-dialysis

CKD subjects [8]. For example, patients with CKD stage III and IV treated with

calcitriol and mild hyperparathyroidism exhibited a 26 % mortality reduction compared to those who were not treated. Again, the baseline PTH, did not appear to

influence this effect, even where PTH was low and there was a greater risk of hypercalcaemia [8, 9]. These and similar studies implied, somewhat counterintuitively,

that perceived contraindications to vitamin D therapy, namely over suppressed PTH,

high calcium and high phosphate, were either wrong, or overwhelmed by the other

positive consequences of VDR activation. In one study of orally administered vitamin D metabolites, there was an inverse dose response relationship between vitamin

D administered and beneficial outcome [6]. The result of these publications was to

galvanise the nephrology community into considering vitamin D treatments as

potentially going far beyond the traditional focus on calcium, phosphorous and calcium regulating hormones [13, 27, 44–49]. The apparent mortality reductions were

dramatic – in the order of 20–30 %, and even higher in some studies [4, 6, 8, 9].

Thus the stage was set for a rapidly expanding research effort to shed light on the


J. Cunningham

credibility of these early observations, their potential applicability in the clinical

arena, and also to potential underlying mechanisms.

Several authors sounded a note of caution however. Despite the profusion of

studies purporting to show beneficial effects of active vitamin D compounds on

survival in CKD patients, this finding is by no means universal and the inherent

weaknesses in much of the published work, most of which falls short of any convincing demonstration of genuine cause and effect, is a matter for concern. These

study designs are vulnerable to various confounders and not all have yielded positive results [7, 50, 51].

A large meta-analysis of randomised controlled studies (RCT’s) by Mann et al.

[52] examined the effects of oral vitamin D compounds (native and active) on cardiovascular outcomes in patients across the range of CKD stages. The intention of

the authors was to include in a meta-analysis only placebo RCT’s reporting original

data of oral vitamin D compounds given as supplementation to adult patients at all

stages of CKD. Transplanted subjects were excluded. The results are telling 4246

abstracts were identified leading to 107 manuscripts initially judged suitable, ultimately leading to only 13 studies (0.3 % of the potentially relevant abstracts) in

which the RCT design was deemed eligible for inclusion. In the subsets of all cause

mortality, cardiovascular mortality and serious cardiovascular events, the relative

risks were respectively 0.84, 0.79 and 1.20, none of them coming close to statistical

significance. Stratification by CKD stage, choice of vitamin D compound, and proportion of diabetics, had no bearing on all cause mortality. A similar analysis

restricted to pre dialysis patients identified five RCT’s comparing active vitamin D

treatment with a placebo or no treatment in which the study outcome included cardiovascular events, blood pressure, cardiac structure and function and proteinuria

[53]. Excluded were non RCT’s, studies using native vitamin D and studies involving dialysis patients. From a total of 780 records screened, only seven ultimately

met the criteria set by the authors. Five compared paricalcitol with control and two

compared calcitriol with control. The results of these analysis showed a striking

reduction of cardiovascular events in vitamin D treated subjects (RR 0.27; 95 % CI

0.13–0.59), and increased the likelihood of reduction of proteinuria (RR 1.9; 95 %

CI 1.34–2.71). There were no effects on other cardiac parameters including LV mass

index and systolic function. Blood pressure was not significantly altered. When considering the different outcome of these two analyses, a crucial question is the choice

of subjects. In the study of Mann et al. nearly a third of the patients were receiving

dialysis. Furthermore, in many of the studies reviewed, the design was principally

aimed at measurement of biochemical and other surrogates and patient level cardiovascular and mortality outcomes, though available, had not been predefined.

Another meta-analysis examining the effect on mortality of vitamin D treatment in

CKD patients was conducted by Duranton et al. [54]. These were observational studies, either prospective or retrospective with no blinding or randomisation. For inclusion the studies required a minimum follow up of 6 months, and sufficient data to

calculate the RR’s and CI’s of all cause and cardiovascular mortality between treated

and untreated subjects. Ultimately 14 articles met the criteria relating to all cause

mortality (extracted from 13 studies including nearly 200,000 patients) and


Vitamin D and Mortality Risk in Chronic Kidney Disease


cardiovascular mortality (from 6 studies and approximately 78,000 patients). The

vitamin D compounds given were calcitriol, paricalcitol, alfacalcidol and doxercalciferol. These agents were given orally or parentally and exposure was defined as receiving any dose of one of these compounds during follow up. The results favoured

patients who had received active vitamin D, whether established on haemodialysis or

predialysis significantly reduced the risk of all-cause mortality (relative risk 0.73,

95 % CI 0.65–0.82). The risk reduction was greatest in patients with the highest PTH

levels (p = 0.01).

A further review of prospective observational studies [55] examined 10 studies

comprising a total of 6853 patients with CKD. There was a higher relative risk of

mortality in subjects with low levels of 25-hydroxyvitamin D. The “dose response”

reduction of mortality was 0.86 (95 % CI 0.82–0.91) with each increase of 25 nmol/L

in the concentration of 25-hydroxyvitamin D. This study examined observational

data only – one nested case control study and nine prospective cohort studies. A

large majority of the patients included were at CKD stage V, although a considerable heterogeneity was evident. Furthermore, in many of the studies reviewed, the

design was aimed principally at measurement of biochemical and other surrogates.

Patient level cardiovascular and mortality outcomes, though available, had not been




A large body of data, almost all of association type, points to deficiency of vitamin

D as a risk for infection. Empirically the use of tuberculosis (TB) sanatoria was

consistent with this view, which is supported further by the demonstration that

phagocytic microbial killing depended on vitamin D dependent regulation of cathelicidin [56]. The association with infection is particularly clear in regard to tuberculosis where certain polymorphisms of the vitamin D binding protein (DBP) gene or

the VDR gene have greater susceptibility to mycobacterial infection. Increasing

evidence indicates that vitamin D has important roles in the regulation of both the

innate and adaptive immune systems. This has implications for infection and autoimmunity [57]. In several small studies in patients with CKD the administration of

cholecalciferol or calcitriol has improved parameters of immune responsiveness

and/or resolution of infection [58, 59]. These studies are mixed, not all showing

significant responses [60, 61].



The evidence that malignancy is influenced by vitamin D status is largely indirect,

but quite persuasive [10]. Vitamin D sufficiency is associated with reduced instance

of colon cancers, prostate cancer and breast cancer [62], though not all studies


J. Cunningham

support this view [63]. Some haematological malignancies may also be influenced

by vitamin D status. Epidemiological studies (prospective and retrospective) indicate that levels of 25-hydroxyvitamin D below 50 nmol/l are associated with a

30–50 % increased risk of incident colon [64] and other cancers. Mortality from

these cancers is higher when vitamin D stores are low. A small study by Lappe et al.

showed lower breast cancer recurrence rates in vitamin D + calcium supplemented

patients than in those given calcium alone [65].

Support for this link also comes from in vitro studies which have shown potentially anti-oncogenic effects resulting from the binding of calcitriol to the

VDR. Several mechanisms appear to be involved and are discussed in more detail

in Chap. 10 and reviewed by Dusso et al. [66] and by Fleet et al. [67]. In the

nephrology world, cancer has been considered normal or even low in CKD populations, and high following transplantation [68]. A recent larger study of 16,400 incident haemodialysis patients confirmed a very high prevalence of vitamin D

deficiency/insufficiency with cancer prevalence also high at 22.1 % [69]. These

frequencies were similar in patients with high and low 25-hydroxyvitamin D levels

and the findings are therefore at variance with other studies in non-renal



Which Vitamin D Should We Use? Native Vitamin D

or Active VDRA Compound? D2 or D3?

Our understanding of vitamin D biology raises the important possibility that the

requirement of cells for local production of calcitriol capable of activating the VDR

in an autocrine or paracrine fashion might not be satisfied by normal concentrations

of hormonal calcitriol, or for that matter, by the sort of calcitriol concentrations

achieved during therapy with active vitamin D compounds [10, 70]. If true, this

would mandate attention to, and treatment of, deficiency or insufficiency of native

vitamin D in CKD patients, including those receiving active vitamin D compounds.

Studies comparing prevailing 25-hydroxyvitamin D concentration with calcitriol

concentration as predictors of mortality have suggested that vitamin D status as

assessed by 25-hydroxyvitamin D is a better predictor of early mortality in haemodialysis patients [5] and for both early mortality and likelihood of progression to

end stage renal disease in predialysis patients [25]. Certainly the body of data concerning 25-hydroxyvitamin D is greater than that of 1,25-dihydroxyvitamin D,

although this may reflect no more than the relative abundance of 25-hydroxyvitamin

D measurements in the data sets examined [55].

Little attention has been given to the choice of D3 over D2, or vice versa, and it

is apparent that, certainly in the case of native vitamin D, the choice has been determined largely by local availability and licencing considerations. There is some evidence that D3 may increase the 25-hydroxylated product more than D2, but this

seems unlikely to have practical implications beyond choice of dose [71].


Vitamin D and Mortality Risk in Chronic Kidney Disease



Unmet Need

Studies capable of demonstrating a positive effect of treatment with vitamin D compounds at the patient level are difficult to undertake if sufficiently robust. Thus virtually all of the available data to hand at the moment fall some way short of being

able to establish clear cut cause and effect in regard to mortality, and also in regard

to specific patient level outcomes that are themselves likely to impact on mortality.

Thus far interventional studies have focused largely on biochemical end points,

although some have examined other surrogates such as mortality and morbid events

that may be closer to the holy grail of patient level outcomes that really matter.

Nevertheless we still lack convincing data from randomized intervention controlled

trials demonstrating that any formulation of vitamin D results in improved patient

level outcomes, although some are planned or are in progress. To be set against this

negative view are the background plausibility as outlined above, and the experimental data and the extensive observational clinical data currently available.

The same limitations apply when it comes to making recommendations in regard

to the indications for vitamin D treatment beyond the classical ones related to bone

and mineral metabolism. This somewhat negative and cautious conclusion that

applies to patients with CKD is close to that drawn by the Institute of Medicine

when making recommendations regarding vitamin D supplementation [72]. That

organisation considered how to ensure that the vitamin D requirements of the general population should be met and concluded that there was sufficiently robust evidence of beneficial effects in relation to classical bone mineral and skeletal

indications to justify current strategies, albeit at slightly higher dose than historically recommended. The Institute of Medicine stopped short of recommending

higher doses such as have been considered in the setting of the pleiotropic actions

of cholecalciferol and also stopped short of routinely recommending treatment with

vitamin D for indications related to infection, autoimmunity or cancer.

For the nephrologist one thing is clear. We need patient level outcome data from

properly designed intervention studies capable of establishing, or refuting, cause and

effect. Less clear is what we should do in the meantime [73]. A pragmatic view, based

on the principle of “do good if you can, provided you do no harm”, is to use active

vitamin D compounds in appropriate pharmacological doses, often supraphysiological, for established indications based on the classical actions of vitamin D on the parathyroids, bone and mineral metabolism. This then leaves two important unanswered

questions. First, should all CKD patients, whether or not manifesting an established

indication, be offered physiological doses of active vitamin D in the hope of reaping

the benefit implied in the results of the many large observational studies already published? Second, should generous supplementation with native vitamin D be offered to

all CKD patients with the aim of supporting widespread extra renal generation of calcitriol and facilitating the putative pleotropic effects of vitamin D that could mitigate

some of the cardiovascular and other attrition faced by these patients (Fig. 24.1)? Such

an approach appears extraordinarily unlikely to do harm and so, for the time being at

least, a cautious and provisional “yes” to both questions appears justified.

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

8 Hypertension, Cardiac Morphology and Progression of CKD

Tải bản đầy đủ ngay(0 tr)