1. Cardiovascular disease and vascular calcification in CKD
Professor Philip A Kalra
Consultant and Honorary Professor of Nephrology
Salford Royal Hospital and University of Manchester, UK

2. Key topics Epidemiology of CVS risk Non-traditional CVS risk factors
In dialysis patients
Non-dialysis CKD
SCD
Non-traditional CVS risk factors
Cardiac structural changes
CKD-MBD : importance of phosphate

3. Rates of death and cardiovascular events rise as renal function declines
Death from any cause
Cardiovascular events
Age-standardised rate per 100 person years
Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N.Engl.J Med. 2004;351:
Estimated GFR (ml/min/1.73 m2)
Go et al et al. NEJM : 351(13): 3 3

4. Chronic Renal Insufficiency Standards Implementation Study (CRISIS)
Mean age 65 yrs
eGFR 31 ml/min
Diabetes 32%
CVS disease (baseline) 47%
1325 patients with mean FU of 34 months

5. CRISIS : survival

6. Cause of death (ONS) %

7. Cardiovascular Mortality Rates are Higher among Dialysis Patients
100 10 1
Dialysis: male
0.1
Dialysis: female
General
population: male
0.01
General
population: female
0.001
Adapted from Levey AS et al. Am J Kidney Dis 1998; 32: 7

8. 4D Study: Primary composite endpoint
Relative Risk Reduction 8 % (95 % CI: -23%, +10%, P=0.37) 60
N=1255 HD pts with type 2 diabetes
Cardiac death, non-fatal MI or stroke
Mean follow-up 4 years 50 40
Cumulative incidence (%) 30
Placebo 20
Atorvastatin 20 mg 10 1 2 3 4 5
5.5 years
Years from Randomization
Placebo Atorvastatin
Wanner et al NEJM 2005;353:

9. Cardiovascular Disease in CKD : Multifactorial Pathogenesis
Elevated PTH/ 2°HPT
Duration of dialysis
Dyslipidemia
Oxidative stress
Hypertension
Exogenous vitamin D/deficit
Cardiovascular Disease
Diabetes Mellitus
Chronic inflammation
Genetics
Hyperphos-phatemia
This diagram represents the many factors that may contribute to cardiovascular calcification in CKD. A good number of these factors are “traditional”—that is, they are at work in the general population as well as the CKD population.
1Qunibi QY, Nolan CA, Ayus JC.
2Hujairi NM, Afzali B, Goldsmith DJ.. Am J Kidney Dis. 2004; 43:
3Goodman WG, et al. N Engl J Med. 2000; 342: 1478 –1483.
4Locatelli F. et al. Nephrol Dial Transplant. 2001; 16:
5Chertow GM, et al.. Kidney Int. 2002;62:
6Guerin AP, et al.. Nephrol Dial Transplant :
Smoking
Exogenous Ca intake
Increased homocysteine levels
Elevated Ca × P product
Traditional risk factors Non Traditional risk factors 9 9

10. Definition of Sudden Cardiac Death (SCD)
Sudden cardiac death is the unexpected natural death from a cardiac cause within one hour of the onset of symptoms in a person not known to have a condition that is potentially fatal

11. Epidemiology of SCD : general population
1 in every 1000 deaths thought to be due to SCD
SCD is usually the 1st cardiac event that a patient will suffer
80% have abnormal coronary arteries
Risk is > in immediate post-MI period
Poor LV function (particularly due to ischemic cardiomyopathy) and a documented history of significant ventricular arrhythmia, are the strongest predictors of SCD

12. Mechanism of SCD : general population
Myocardial infarction and poor left ventricular function both lead to risk of re-entrant ventricular tachycardia (VT) :
MI : by post-infarction scarring
LV failure : by abnormal fibrotic myocardial remodelling
These areas of abnormal tissue may still contain functioning myocytes, but the surrounding scar tissue is thought to cause bundle branch block, and predispose to subsequent re-entrant tachycardia

13. Epidemiology of SCD : CKD populations
CKD stages 3-5 (not dialysis) SCD risk ↑ by HR of 1.1 for every 10ml/min decline in eGFR
Event rate 0.8% per yr in non-dialysis CKD
In non-diabetic dialysis patients, rate is 7% in 1st yr of RRT
SCD risk is > for HD than PD patients during 1st 6 months of dialysis, but equalises thereafter
CKD
Dialysis
General

14. Monday or Tuesday (greatest risk last 12 hrs before dialysis)
Karnik JA et al (Kidney International 2001:60: ) : Characteristics associated with arrest on haemodialysis
Monday or Tuesday (greatest risk last 12 hrs before dialysis)
Low potassium dialysate
Older age
Diabetic
Catheter for access

15. CVS risk factors in CKD Cardiac structural changes – LVH and CCF CAD
Vascular calcification/arterial stiffness
Phosphate
Vitamin D deficiency
Anaemia
Metabolic changes
Inflammation

16. Prevalence of Left Ventricular Hypertrophy in Relation to Creatinine Clearance
p <0.003 (trend analysis)
In a prospective multicentre study, Canadian investigators evaluated the prevalence of LVH in relation to creatinine clearance in this group of 246 patients with mild to moderate renal insufficiency.
The investigators divided the patients into discrete groups by predefined categories of mild (Ccl = mL/min), moderate (Ccl = mL/min), or severe (Ccl <25 mL/min) renal insufficiency and dialysis.
As this slide shows, the prevalence of LVH is related to the degree of renal insufficiency (p <0.003 [trend analysis])
Levin A, Thompson CR, Ethier J, et al. Left ventricular mass index increase in early renal disease: impact of decline in hemoglobin. Am J Kidney Dis 1999; 34:
Patients with diabetes = 24%
Adapted from Levin A et al. Am J Kidney Dis 1999; 34: 16

19. Intra-dialytic myocardial ischaemia
C McIntyre and colleagues (Derby, UK) :
Haemodialysis induces reversible intra-dialytic myocardial stunning
↑stunning associated with greater propensity to arrhythmia
↑stunning associated with worse mortality
Some relationship between endotoxaemia and myocardial ischaemia

20. Calcification of the coronary arteries
Pre-contrast
Post-contrast
Khogali and Townend NEJM 2002;347:1584

21. Arterial Medial Calcification in ESRD
London GM, et al. Nephrol Dial Transplant. 2003;18:

22. Prevalence of Vascular Calcification in CKD
Patients New to Dialysis and Established Patients
***
Stage 3-4 CKD ** *
*Russo et al AJKD 2004 (CrCl =33 ml/min)
**Spiegel D et al. Hemod Internat 2004: 8:265
***Chertow et al KI 2002

23. Probability of All-Cause Survival According to Calcification Status
0.00
0.25
0.50
0.75
1.00
Calcification Score: 0
Calcification Score: 1
Calcification Score: 2
Probability of Survival
Calcification Score: 3
Prospective studied 110 ESRD HD patients
Ultrasound looking for arterial calcification at the carotid artery, abdomainal aorta, iliofemoral axis and legs.
Follow up 53 months
All cause mortality increased 90 % per 1 site increase in calcifacation score, CV mortality increased 260%
The presence and extent of vascular calcifications were strong predictors of cardiovascular and all cause mortality. Carotid incremental elastic modulus gave ADDItiONAL predictive value.
Calcification Score: 4 20 40 60 80
Duration of Follow-Up (Months)
*Comparison Between Curves Was Highly Significant (x2=42.66, P<0.0001) Source: Blacher A, et al. Hypertension: , October 2001 23 23

24. Augmentation index : Applanation Tonometry
CRISIS – chronic renal insufficiency standards implementation study – running since 2002 up to now - clinical and biochemical parameters of 792 pre-dialysis patients in the Hope Hospital CKD programme, with information on cardiovascular risk factors and current medications
Cross sectional analysis was carried out on 605 patients Technique involved applanation tonometry – probe held against radial artery over maximal area of pulsation (as shown in the picture) – waveform recorded on computer interfaced with SphygmoCor machine – 2 consecutive readings taken – quality controlled
Ethical approval from COREC & informed consent
Aortic Augmentation Index (%) = ∆P x 100
(AIx) PP

26. Importance of phosphate

27. Serum Phosphorus and Mortality in Hemodialysis Patients
1.50
1.00
1.08
1.25
1.42
1.68
2.03
0.5 1
1.5 2
2.5
<3
3-4
4-5
5-6
6-7
7-8
8-9
>9
Serum Phosphorous Concentration (mg/dL)
Relative Risk of Death*
n = 40,538
P <
*Multivariable Adjusted
Block G, J Am Soc Neph 15: , 2004

28. CRISIS study : analysis of serum phosphate (Eddington H et al, CJASN 2010)
1213 patients
Baseline demographics – Phosphate divided into quartiles
Cox regression
Baseline phosphate and survival
Time-averaged phosphate and survival

29. Baseline demographics
All
PO4 <1.01
N=318
PO – 1.16
N=300 PO
N=293
PO4 ≥1.34
N=302
P value
Age
64.2 (13.9)
64 (14)
65 (14)
62 (14)
0.037
Female
sex
429 (35.4%)
76 (24%)
109 (36%)
124 (42%)
120 (40%)
<0.0001
Calcium
2.29 (0.14)
2.29 (0.13)
2.29 (0.19)
2.30 (0.13)
2.28 (0.19) ns
PTH
89 (86)
58 (42)
77 62)
86 (77)
135 (124) Hb
124 (18)
135 (18)
125 (16)
123 (14)
114 (17)
eGFR
31.6 (15)
40 (14)
34 (13)
31 (14)
20 (11)
Proteinuria
1.1 (1.8)
0.5 (0.7)
0.8 (1.2)
0.9 (1.2)
2.1 (2.7)
CVD
380 (31%)
99 (31%)
99 (34%)
73 (24%)
0.009 DM
385(32%)
84 (27%)
89 (29%)
90 (29%)
122 (40%)
0.002

30. Baseline phosphate and survival
Adjusted for eGFR, Age, Gender, Hb, Diabetes, CVD, proteinuria, PTH
Mean follow-up 4.3 years
Phosphate <1.01
Phosphate
Phosphate
Phosphate >1.33
Hazard ratio 1.8
P = 0.04
n= n= n= n= n=136

31. 12mth time-average PO4 survival
Adjusted for eGFR, Age, Gender, Hb, Diabetes, CVD, proteinuria, PTH
Mean follow-up 3.6 years
Phosphate <1.01
Phosphate
Phosphate
Phosphate >1.34
Phosphate <1.01
Phosphate
Phosphate
Phosphate >1.34
Hazard ratio 2.59
P = 0.006
Hazard ratio 2.12
P = 0.01
n= n= n= n=136

32. Survival according to previous KDOQI phosphate guidelines
Below Target
In Target
Above Target
Hazard ratio:
In target 1.9 ( ) P = 0.08
Above target 2.6 ( ) P = 0.03

33. Phosphate : general population

34. CARDIA (Coronary artery risk development in young adults)
Prospective multi-centre observational study of CVS disease development in fit young adults (age yrs)
in 4 US regions (Birmingham, Alabama; Chicago, Illinois; Minneapolis,Minnesota; Oakland, California)
5113 participants

35. CARDIA (Coronary artery risk development in young adults)
Various baseline variables assessed
LVMI assessed by echocardiography 5 years after entry
Coronary artery calcification assessed by CT scan 15 years later

36. Left ventricular hypertrophy (LVH) : Foley RN et al Kid Blood Press Res 2009; 32(1):37-44
4005 of 5113 participants underwent echocardiography
Baseline data
Mean age 25 years
Mean phosphate 3.7 mg/dl
eGFR ml/min/1.73m2
Results
Each SD of baseline phosphate above the mean was associated with ↑presence of LVH 5 years later (AOR 1.301, p=0.0018)

37. Coronary artery calcification (CAC) : Foley RN et al J Am Soc Neph 2009; 20(2): 397-404
3015 of 5113 participants underwent CT at 15 years
Baseline data
Mean age 25.2 years
mean phosphate 3.6 mg/dl, calcium 9.5 mg/dl
Mean eGFR ml/min/1.73m2
0.2% with eGFR < 60 ml/min/1.73m2

38. Coronary artery calcification (CAC) : Foley RN et al J Am Soc Neph 2009; 20(2): 397-404
Year 15 CAC scores
Minimal %
Mild %
Moderate %
Severe > %

39. P-Spline plot relating adjusted odds ratio of CAC ≥ 100 and serum phosphorus
AOR, with 95% confidence intervals. Adjusted for all variables
except calcium-phosphorus product and diastolic blood pressure

40. Studies of phosphate in general population : conclusions
Phosphate levels even at the upper end of normal range appear to be a risk factor for :
Coronary artery calcification (surrogate of coronary atherosclerosis)
Left ventricular hypertrophy
? Pathogenetic effect or association

41. FGF-23/Klotho: New players in CKD-MBDPi
1,25D
Small bowel
Reduces Ca and Pi absorption in small bowel
 1,25D
1α(OH)D3
Inhibits 1α-hydroxylase
FGF-23
Parathyroid
Possibly stimulates
Phosphaturia
+ Klotho
Stimulates
Skeleton
Possibly inhibits mineralization
FGF23 inhibits PTH mRNA transcription and protein secretion
Kidney
Kuro-o. Keynote lecture from ERA-EDTA 2008, ASN 2008
Adapted from: Emmett M, et al. Kidney International 2008;73:3–5

42. Haemodialysis patients within the highest range of FGF-23 levels had nearly 6x greater risk of death
Gutierrez OM et al; N Engl J Med 2008 :359;

43. Temporal aspects of mineral disorders in progressive CKD and post transplantation
Wolf JASN 2010

44. How might FGF-23 be associated with CVS risk?
↑ FGF-23 associated with ↓vitamin D, CKD progression and mortality in CKD
↑ FGF-23 associated with ↑ LVH (Gutierrez OM et al, Circulation 2009; 119 : )
↑ FGF-23 associated with ↑ ADMA (asymmetric di-methyl arginine; an inhibitor of NO synthase)
↑ FGF-23 associated with ↓ flow-mediated dilatation (FMD) in CKD patients (Yilmaz MI et al, Kidney Int, 2010; 78 : )

45. Vitamin D levels very low in dialysis patients
52 Vitamin D naïve haemodialysis patients (>90% ‘deficient’)
Mean PTH 345pg/ml ± 37 (245)
Mean 25(OH)D 14.2 ± 1 (13.5)
London GM et al JASN 2007: 18; (latitude 48o)

46. Vitamin D levels assoc with arterial function
London GM et al JASN 2007: 18; (latitude 48o)

47. Studies of intervention for vascular calcification

48. Treat-to-Goal Study : Prevalent haemodialysis patients
Sevelamer
Calcium * *
Median percentage change
*Within treatment P<0.0001; between treatment groups P=0.02
Chertow et al. Kidney Int. 2002

49. ADVANCE :Study Endpoints
Percentage change from baseline in CAC score at week 52
Primary Endpoint
Secondary Endpoints
Percentage change from baseline in CAC score at week 52
Secondary Endpoints
Absolute change in CAC score at week 52
Absolute and percentage change from baseline in
Aortic calcification at week 52
Aortic valve calcification at week 52
Laboratory parameters at end of study (weeks 44 through 52)
Proportion of patients achieving > 15% progression of CAC at week 52
Safety
Data on file, Amgen [ADVANCE Protocol ; 6/22/2007].
Laboratory parameters = PTH, calcium, phosphorus and Ca x P
Safety = nature, frequency, severity and relationship to treatment of adverse events
Inclusion Criteria
Informed Consent
Adults ≥ 18 years of age
On hemodialysis for ≥ 3 months prior to study day 1
iPTH > 300 pg/mL (31.8 pmol/L) OR
iPTH  150 pg/mL and ≤ 300 pg/mL (15.9 – 31.8 pmol/L)
and receiving treatment with vitamin D analogs at time of PTH assessment
and corrected serum Ca x P > 50 mg2/dL2 (3.9 mmol2/L2)
Corrected serum Ca ≥ 8.4 mg/dL (2.1 mmol/L)
Screening CAC score ≥ 30
Subjects were subsequently stratified: ≥ 30–399, ≥ 400–999, and ≥ 1000
Concomitenet Therapy
Changes in active vitamin D brand were to be avoided
Calcium-containing phosphate binders
Were to remain as stable as possible
Best practice guidelines were to be followed
Dialysate calcium concentration and cholesterol-lowering medications were to remain stable throughout study
Not permitted
Non–calcium containing phosphate binders [sevelamer HCl, lanthanum carbonate, aluminum-based phosphate binders]*
Non-study cinacalcet
Other investigational agents
Raggi P et al. Poster presented at the 2010 Clinical Meeting of the National Kidney Foundation, Orlando, FL, April 13-17, 2010.
Floege J et al. Poster presented at the 2010 ISN Nexus Meeting, Kyoto, Japan, April 15-18, 2010. 49

50. Patient characteristics
737 patients were screened and 360 were randomized, 180 to each group.
Mean (SD) age was 61.5 (12.7) years, 58% were male and 24% were black
Median (P10, P90) time on hemodialysis was 36.7 (9.5, 107.0) months.
The efficacy analysis included 235 subjects:
115 assigned to cinacalcet plus low dose vitamin D
120 assigned to flexible doses of vitamin D sterols
Raggi P et al. Poster presented at the 2010 Clinical Meeting of the National Kidney Foundation, Orlando, FL, April 13-17, 2010.
Floege J et al. Poster presented at the 2010 ISN Nexus Meeting, Kyoto, Japan, April 15-18, 2010.

51. Median % Change (P10, P90) in CAC
Percent Change in Total Coronary Artery Calcification Score (CAC) – Agatston
Primary Analysis
Median % Change (P10, P90) in CAC
Cinacalcet
(n=115)
Control group
(n=119)
p-value
24 (-22, 119)
31 (-9, 179)
0.073
Primary analysis based on a generalised Cochran-Mantel-Haenszel test on ranks
Agatston Table:
Volume Table
Raggi P et al. Poster presented at the 2010 Clinical Meeting of the National Kidney Foundation, Orlando, FL, April 13-17, 2010.
Floege J et al. Poster presented at the 2010 ISN Nexus Meeting, Kyoto, Japan, April 15-18, 2010. 51

52. Percent Change in Total Coronary Artery Calcification Score (CAC) - Agatston
Analysis adjusted for baseline phosphorus
Geometric Mean % Change (95% CI) in CAC
Cinacalcet
(n=115)
Control group
(n=119)
p-value
26 (16, 36)
42 (31, 54)
0.031
Supportive analysis (as planned in the protocol) using a generalised linear model to adjust for the baseline imbalance in phosphorous levels between treatment groups.
Agatston Table:
Volume Table
Raggi P et al. Poster presented at the 2010 Clinical Meeting of the National Kidney Foundation, Orlando, FL, April 13-17, 2010.
Floege J et al. Poster presented at the 2010 ISN Nexus Meeting, Kyoto, Japan, April 15-18, 2010. 52

53. Does reducing vascular calcification translate into survival benefit?

54. DCOR study: Primary Endpoint
0.6
RR 0.91 ( ), p = 0.30
0.5
n=2103
0.4
Cumulative Incidence of
All-Cause Mortality
0.3
0.2
Results of the DCOR trial were inconclusive for the primary end-point of all-cause mortality across the entire patient cohort (RR 0.91; p = 0.3)
Overall survival curve indicated need for treatment time analysis
Cox proportional hazards models assumes that the treatment effect is constant over time. Survival curve indicates otherwise—change in curve beginning at 2 years
Calcium
0.1
Sevelamer
0.0 1 2 3 4
Time in Study (Years)
Suki et al, Kidney Int 2007;72: 54

55. DCOR : All-Cause Mortality in Patients ≥ 65 years
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Sevelamer therapy resulted in a statistically significant reduction in the relative risk for all-cause mortality in pre-specified subset
[RR 0.78 ( )]
Cumulative Incidence of All-Cause Mortality
↓ 22%
p = 0.03
Sevelamer
Calcium 1 2 3 4
Time on Study (Years)
No. at Risk
Calcium
Sevelamer

56. Final KDIGO Grading of Recommendations
During the development of the CKD-MBD guideline, KDIGO continued to develop a system for rating the strength of recommendations and the overall quality of evidence supporting those recommendations. A task force had been formed that ultimately made recommendations to the KDIGO Board. After extensive discussion and debate, the KDIGO Board of Directors in 2008 unanimously approved a modification of the Grades of Recommendation Assessment, Development, and Evaluation (GRADE) system. The system that was adopted allows provision of guidance even if the evidence base is weak, but makes the quality of the available evidence transparent and explicit. It is described in detail in the present CKD-MBD guideline (Chapter 2).
Each recommendation is rated 1 or 2, with 1 being a “We recommend …” statement implying that most patients should receive the course of action, and 2 being a “We suggest …” statement implying that different choices will be appropriate for different patients. In addition, each statement is assigned an overall grade for the quality of evidence, A (high), B (moderate), C (low), or D (very low). The grade of each recommendation depends on the strength of the evidence, but also on additional considerations.
A key issue is whether to include guideline statements on topics which can not be subjected to systematic evidence review. KDIGO has decided to meet this need by including some statements that are not graded. Typically, ungraded statements provide guidance that is based on common sense, e.g., reminders of the obvious, and/or recommendations that are not sufficiently specific enough to allow application of evidence. Examples include the frequency of laboratory testing and the provision of routine medical care.
The CKD-MBD guideline encompasses many aspects of care for which there is little or no evidence to inform recommendations. Indeed, there are only three recommendations in the CKD-MBD guideline for which the overall quality of evidence was graded “A”, while 12 were graded “B”, 23 were graded “C”, and 11 were graded “D”. Although there are reasons other than quality of evidence to make a grade 1 or 2 recommendation, in general, there is a correlation between the quality of overall evidence and the strength of the recommendation. Thus, there are 10 recommendations graded “1” and 39 graded “2”. There were two recommendations graded “1A”, five were “1B”, three were “1C”, and none were “1D”. There was one graded “2A”, seven were “2B”, 20 were “2C”, and 11 were “2D”. There were 11 statements that were not graded.
The grades should be taken seriously. The lack of recommendations that are graded “1A” suggests that there are few opportunities for developing clinical performance measures from this guideline. The preponderance of “2” recommendations suggests that patient preferences and other circumstances should be strongly considered for most recommendations. The lack of “A” and “B” grades of overall quality of evidence is a result of the lack of patient-centered outcomes as end-points in the majority of trials in this field, and thus suggests strongly that additional research is needed in CKD-MBD. Indeed, the extensive review that led to this guideline often exposed significant gaps in our knowledge. The Work Group made a number of specific recommendations for future research needs. This will hopefully be of interest to future investigators and funding agencies.
Grading Options: 1A, 1B, 1C, 1D, 2A, 2B, 2D, 2D, & “not graded” 56

57. KDOQI Mineral and PTH targets
Stage 3
Stage 4
Stage 5
Calcium
Normal range
mmol/l
Phosphate
Ca x P
<3.6 mmol/l
<3.6 mmo/l/
< 4.3 mmol/l
PTH
pmol/l
National Kidney Foundation. Am J Kidney Dis 2003;42:S1-S202 57

58. Diagnosis of CKD-MBD: Vascular Calcification
In patients with CKD Stages 3-5D, we suggest a lateral abdominal radiograph can be used to detect the presence or absence of vascular calcification, and an echocardiogram can be used to detect the presence or absence of valvular calcification, as reasonable alternatives to computed tomography (CT)-based imaging (2C).
We suggest that patients with CKD Stages 3-5D with known vascular/valvular calcification be considered at highest cardiovascular risk (2A). It is reasonable to use this information to guide management of CKD-MBD (not graded).

59. Summary
Patients with CKD are at high CVS risk and CKD-MBD is a major contributor
Observational data show the importance of several factors (low vitamin D, ↑ Phosphate, ? ↑ Calcium dose, ↑ PTH)
Early phosphate rise seems to be important in earlier CKD and even in the general population (relevance of FGF-23?)
Interventional studies suggest that calcification can be slowed
Further interventional studies (eg EVOLVE) are necessary to guide optimal treatment in CKD-MBD