Stroke Prevention Research
Dr. J. David Spence
&
Dr. Daniel Hackam

Management of patients:

Beginning in 1992, we have developed a new paradigm for management of patients with atherosclerosis, using plaque measurements in the carotid arteries to determine whether therapy has been successful. To us, trying to manage arteries without knowing how they are doing is like trying to manage hypertension without measuring blood pressure, or hyperlipidemia without measuring serum cholesterol.

In the past, we measured the total plaque area, taking the sum of the cross-sectional area of all plaques seen in both common, internal and external carotid arteries. As discussed below, we are now, in collaboration with Dr. Aaron Fenster of the Robarts Imaging Research Laboratory, measuring plaque volume by 3-D ultrasound. The figure below shows measurement of a long plaque in the right common carotid artery.

(From Spence JD, Eliasziw M, DiCicco M, Hackam DG, Galil R, Lohmann T. Carotid Plaque Area: A Tool for Targeting and Evaluating Vascular Preventive TherapyStroke. 2002;33:2916-2922. )

Plaque increases steeply with age; at any age women have less plaque than men.

( from Iemolo F, Martiniuk A, Steinman DA, Spence JD. Sex differences in carotid plaque and stenosis. Stroke 2004 Feb;35(2):477-81 )

 

We have published evidence that the amount of plaque is a very strong predictor of outcomes: after adjustment for all the traditional risk factors and some new risk factors (age, sex, blood pressure, cholesterol, smoking, diabetes, homocysteine and treatment of blood pressure and cholesterol), patients in the top quartile of plaque area at baseline have 3.5 times the risk of stroke, death or heart attack (MI) compared to patients in the bottom quartile of plaque area ( Spence JD, Eliasziw M, DiCicco M, Hackam DG, Galil R, Lohmann T. Carotid Plaque Area: A Tool for Targeting and Evaluating Vascular Preventive Therapy. Stroke. 2002;33:2916-2922.)

The graph below, copied from that paper, shows the survival free of stroke, death or MI by quartile of baseline plaque area.

We also found that plaque progression was a strong predictor of outcome: patients with plaque progression in the first year had twice the risk of those with stable plaque or regression, as shown in the figure below:

(From Spence JD, Eliasziw M, DiCicco M, Hackam DG, Galil R, Lohmann T. Carotid Plaque Area: A Tool for Targeting and Evaluating Vascular Preventive TherapyStroke. 2002;33:2916-2922. )

This knowledge allows us to use plaque progression or regression in management of patients. If the patient has had plaque regression for successive years, this means that we have not missed anything, the treatment is working, and things are on track. Thus the patient can continue with the family physician, and does not need to return annually to clinic; usually we book a followup in 3-5 years depending on the age, risk level, and other factors.

If the plaque is progressing because known risk factors are not yet well controlled, then efforts are intensified to control known risk factors, and we look for new risk factors. Efforts at smoking cessation, a Mediterranean diet, exercise, target levels of cholesterol, HDL, and control of diabetes are intensified.

New known risk factors that are looked for include homocysteine (including deficiency of vitamin B12), hypothyrodism, elevated, Lipoprotein (a) [Lp(a)].

If the plaque is progressing despite good control of all traditional and new putative risk factors, then we have to consider genetic testing in the family, if the family is large enough (discussed below)

 

GENETICS OF PREMATURE ATHEROSCLEROSIS

We are working on discovering new genetic causes of atherosclerosis and new treatments to prevent atherosclerosis in our Premature Atherosclerosis Clinic. Measurement atherosclerosis progression by 3-D ultrasound, together with regression modeling with the known risk factors, permits the identification of individuals that are highly likely to bear new genetic causes of atherosclerosis. The regression model allows measurement of a quantitative trait, unexplained atherosclerosis progression, in family members, to increase the power of genetic linkage analysis, by permitting quantitative trait locus (QTL) mapping.

Unexplained Atherosclerosis: a quantitative trait

The figure above shows the distribution of measured plaque vs the amount of plaque predicted in multiple regression with age, sex, blood pressure, smoking and cholesterol as the independent variables. The patients represented as solid red triangles have “Unexplained atherosclerosis”, a quantitative trait for genetic analysis. The green squares represent patients who are protected.

We used plasma homocysteine to test the hypothesis that patients in the top 10% of residual scores in the model (i.e. those whose atherosclerosis is least explained by age, sex, blood pressure, cholesterol and smoking, represented by the red triangles squares in the figure) have new causes of atherosclerosis. Compared to 20% of the Canadian urban population and 20.1% of the volunteers in the original stress study, and with 33% of patients in the premature atherosclerosis clinic, 47% of patients with unexplained atherosclerosis in our model had plasma homocysteine >14 mol/L. We showed that plasma homocysteine, but not MTHFR genotype, predicted plaque area independently (i.e. after controlling in multiple regression for age, sex, systolic pressure, lipids and smoking.

Table 2. Multiple regression with plaque area as dependent variable, age, sex, systolic pressure, pack-years of smoking, cholesterol, and treatment for hypertension or hyperlipidemia as predictors.

*lipid levels are not predictive, probably because of treatment; it is for that reason that the variable LIPMED1, representing treatment of lipids, is a significant predictor, as is HTHYP, which represents history of or treatment of hypertension. SYSBP1 = baseline systolic pressure; PKYRS = pack-years of smoking.
** a cube root transform was used to normalize the distribution of plaque area, as multiple regression
requires normally distributed variables. Using this transformation, the normal probability plot is
linear.

The model is extremely robust, as shown by the cumulative probability plot below.

Figure 3. Cumulative probability plot for the regression model.

 

Spence JD, Barnett PA, Bulman DE, Hegele R. An approach to ascertain probands with a non traditional risk factor for carotid atherosclerosis. Atherosclerosis 1999; 144: 429-434.

Spence JD, Barnett PA, Hegele RA, Marian AJ, Freeman D, Malinow MR. Plasma Homocyst(e)ine, but not MTHFR genotype, is associated with variation in carotid plaque area. Stroke 1999; 30: 969-973.

Hegele RA, Ban MR, Anderson CM, Spence JD. Infection-susceptibility alleles of mannose-binding lectin are associated with increased carotid plaque area. J Invest Med 2000; 48: 198-202.

Spence JD, Ban MR, Hegele RA. Lipoprotein lipase (LPL) gene variation and
progression of carotid artery plaque. Stroke 2003;34:1178-1182.

Spence JD, Hegele RA Noninvasive Phenotypes of Atherosclerosis: Similar Windows but Different Views Stroke 2004; 35: 649 - 653.

Identifying new genes that are protective or increase susceptibility to atherosclerosis will lead to new therapeutic targets. An example of this is the new treatment based on the Apo A1 Milano gene.

CLINICAL RESEARCH WITH NEW TREATMENTS FOR ATHEROSCLEROSIS

In collaboration with the Imaging Research Group at the Robarts Research Institute, we have developed 3-D ultrasound methods for measuring the rate of progression of atherosclerosis in the carotid arteries. This method is ideal for testing new treatments, since the rate of growth of plaque in 3 dimensions is more than 2.5 times faster than the growth in thickness, and the measurements are non-invasive. Sample sizes required to demonstrate efficacy of treatments are much smaller (approximately 150 patients per group, followed for 2-3 years) than with alternative methods.

Plaque volume is measured by Disc Segmentation, shown in the figure below:

Measurements are made in the axial plane (c); the perimeter of the plaque in each slice is traced (d and e); slices are made at 1-mm intervals, and the slices are stacked up to form a volume, and the surface contour is mapped for measurement of plaque surface roughness (f). 19-22

In the figure below, the top plaque has been measured by disc segmentation; the lower plaque has had contours applied for measurement of a new therapeutic target, "plaque surface roughness", that we believe is related to plaque instability.

 

This view shows two plaques in the left carotid, in a patient with unstable carotid plaque manifested by microemboli on transcranial Doppler. The plane shown at left (in black) is a slice in the common carotid; the plaques shown extend into the internal carotid. The upper plaque (326.75mm 3 ) has had each slice traced in cross-section (plaque envelopes in yellow); the next step is mapping of the contours, shown in the lower plaque (407.25mm3). Longitudinal and cross-sectional views of this patient's ulcerated left carotid artery are shown below.

(From Spence JD, Blake C, Landry A, Fenster A. Measurement of carotid plaque and effect of vitamin therapy for total homocysteine. Clin Chem Lab Med. 2003; 41: 1498-504. )

Ulceration in a patient with microemboli on TCD (23)

Longitudinal view of the left internal carotid in a man with 70% stenosis and microemboli on TCD. Ulcers and fissures (below) likely explain the occurrence of microemboli

 

(From Spence JD, Tamayo A, DiCicco M. Unstable carotid plaque. CMAJ. 2002;166(9):1189.)

 

The figure below shows that it is now possible to measure effects of therapy on progression of plaque volume in 3 months, in 20 patients per group. The figure shows a significant difference in rate of progression on placebo vs regression on atorvastatin 80mg daily. This approach will make it possible to test new therapies much more efficiently than in the past.