Calcific Aortic Valve disease and Lp(a) Kate Kearney 4/17/2012.

Calcific Aortic Valve disease and Lp(a)

Kate Kearney 4/17/2012

Outline Calcific Aortic Valve Disease (CAVD)—an

active process

Lp(a) is a lipid that appears to be important in CAVD

AGES-Reykjavik is an elderly Icelandic cohort

CAVD scored on CT scan correlates with Lp(a)

Calcific Aortic Valve Disease Aortic Sclerosis

Thickened, calcified valve

Lack of outflow obstruction Velocity <2.5m/s

across AV 25% in those >65yo1

Associated w/ 50% increased risk CV event2

Subset progress to Aortic Stenosis 1-2% per year

Aortic Stenosis AV velocitity >2.5m/s 2.5% in >65yo age

group Associated 80% 5-yr

risk progression to Heart failure Need for AV

Replacement Death

1. Stewart et al. Cardiovascular Health Study. 2. Otto et al N Engl J Med 1999

Calcific Aortic Valve Disease Age is a clear factor, but not the only one Progressive disease with no medical treatment

Valve replacement only definitive treatment, timing of surgery can be difficult

Increasing evidence for role of endothelial injury (inflammatory, high tensile stress) and lipid deposition in the early phases

Significant overlap with risk factors in CAD

CAVD risk factors overlap with CAD

Cardiovascular Health Study (Stewart et al JACC 1997)

Framingham Offspring Study (Thanassoulis et al JACC 2010)

Variable OR (95% CI)

Age (75th vs 25th)

2.18 (2.15-2.2)

Male Gender 2.03 (1.7-2.5)

Lp (a) (10 unit incr)

1.23 (1.14-1.32)

h/o HTN 1.23 (1.1-1.4)

Present Smoking

1.35 (1.1-1.7)

LDL (10 unit incr)

1.12 (1.03-1.23)

Variable OR (95% CI)

Age, per SD 3.25 (2.76-3.82)

Female gender 0.56 (0.41-0.76)

Total CH 1.74 (1.5-2.01)

Mean HDL 0.77 (0.66-0.91)

BMI 1.21 (1.05-1.4)

Cigarettes daily 1.23 (1.08-1.41)-Framingham CHD risk score also significantly correlated

with prevalence and severity of Aortic Valve Calcium

Aortic Stenosis pathophysiology Inflammatory Cells and

Lipoproteins, including Lp(a) co-localize in AV lesions, sim to Coronary atherosclerosis

Macrophages and T lymphocytes found in valves

LDL and Lp(a) deposits Oxidized lipids found in

human lesions, particularly developing calcifications

Calcification is active process

O'Brien K D et al. Arterioscler Thromb Vasc Biol 1996;16:523-532

apo B

apo (a)

macrophages

Ca++

Lipids and Calcific Aortic Valve Disease Strong association with CAD, CAD risk factors

Perhaps similar treatment will be effective?

Multiple trials looked at Statins and Aortic Stenosis

Three randomized prospective trials were negative for effect Ranging in size and f/u time 2-5yr, Atorvastatin, Simva

+Ezetimibe and Rosuvastatin failed to effect Jet velocity or peak gradient

Asymptomatic mild-moderate disease in SEAS and ASTRONOMER, SALTIRE included those with severe AS

Too little? Too Late?

Lipoprotein(a) consists of an LDL-like particle to which apolipoprotein(a) is covalently linked.

Nordestgaard B G et al. Eur Heart J 2010;eurheartj.ehq386

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2010. For permissions please email: [email protected]

Lp(a) CV risk factor with unclear role

Recommended by some authors to screen in high-risk patients

Appears important in setting elevated LDL Appears to be particularly important in those with

elevated LDL1

Niacin is currently available treatment Dose dependent reduction in levels 30-40%

Determinants: 90% genetics, mostly stable over lifetime Genetics LPA locus variations (encoding Lp(a)) strongly

associated with CAD Strong relationship with LPA locus recently demonstrated

1. Luc Lipoprotein (a) as a predictor of coronary heart disease: the Prime Study. Atherosclerosis 2002; 163(2):377-84.

GWAS demonstrates SNP at LPA locus associated with CAVD LPA locus encoding Lp(a) lipoprotein

Lp(a) levels known to be ~90% genetically determined

This SNP previously shown to associate with CAD risk

This study confirmed association with Lp(a) levels

Strongly associated with Aortic Valve Calcium Scores

Holds across multiple Euro-descent cohorts (including AGES), verified in other cultural groups (MESA database) Each risk allele confers 68% increased risk Correlated with clinical outcomes, need AVR

Thanassoulis G, Campbell CY, Owens DS et al . A genome wide association study identifies novel loci for aortic valvular and mitral annular calcium and implicates LPA in the development of Aortic Stenosis. (Unpublished)

AGES-R Study Age, Gene/Environment Susceptibility-Reykjavik

Study The Reykjavik study is an established population-

based cohort, established 1967 Men and Women born 1907-1935, followed by Icelandic

Heart Association for prospective study of CVD AGES initiated in 2002 to examine risk factors and

interaction with complex traits, using improved phenotypes based off subclinical disease

Reykjavik provides mid-life data, collected prospectively and followed therefore less biased than retrospective recall

AGES examining same subjects in old age, examining for clinical events and subclinical disease (labs, imaging)

Population

Our Study… Hypothesis: The lipoprotein Lp(a) is associated

with prevalence and severity of calcific aortic valve disease (CAVD) and interacts with individual LDL cholesterol levels.

Population: AGES-R elderly cohort with mid-life LDL levels and followup LDL and Lp(a) levels, CT scans scored for AVC. Exclusions: prior AVR, known CAD, statin use Cross sectional evaluation of interaction of LDL

and Lp(a) with Aortic Valve Calcification

CAVD scored by CT-Agatston method Areas of calcium defined by brightness

(Hounsfield un) in contiguous pixels Includes valve leaflets only, not aorta or coronary

arteries Scores can be used as continuous or

dichotomous variable Does not provide functional data, but scores

above a threshold predict stenosis or severe disease Suggestion of prognostic information Validated with and Complementary to TTE

Detects subclinical disease Important in our model of CAVDMessika-Zeitoun et al. Eval and clinical implications of aortic valve calcification measured by electron-beam CT. Circ 2004

Our Study… Evaluate relationship between Lp(a) and AVC

scores within these LDL groups Test variable interactions using multivariable

regression with the presence and severity of AVC, as measured by the Agatston method from AGES-Reykjavik baseline CT scans.

Control for confounders based off prior epidemiological data: Age, gender, smoking history, Statin and Niacin use (if not excluded) use anti-HTN, diabetic medications

LDL Distribution in AGES-R Separate subjects

into 3 groups based off clinical LDL cutpoints: <130, 130-160, >160 Approximate tertiles

Note persistence of elevated LDL in Elderly (lack expected

survival bias) After excluding statin-

users

01

002

003

004

00

70 80 90 100 70 80 90 100

Male Female

LDL cholesterol (mg/dL Fitted values

Subject age (years)

Graphs by Sex (0=Male 1=Female)

Results…Pending

But, we can see looking at preliminary data that Lp(a) correlates with AV calcium scores in the AGES-R cohort:

To follow with analysis of cross-sectional Lp(a) effect followed by multivariate regression

Follow-up CT scan w/ Agaston scores available

Association w/ AVC

OR (p<.0001)

Plasma Lp(a) 1.34( 1.25-1.43)

LPA SNP 1.75 (1.59-1.91)

Summary Calcific aortic valve disease and athersclerosis

have similar risk factors and overlap We now know lipids are involved in the

development of CAVD Lp(a) is similar to LDL, largely genetically

determined, and shown to correlate and demonstrate causal relationship with CAVD

LDL and Lp(a) appear to interact, but this has not been studied in a clinical cohort

Lp(a) is associated with CAVD in the AGES-R cohort