CRP: another piece of the heart-disease puzzle? (Cover Story).

"Study: Inflammation twice as bad as cholesterol for the heart," announced USA Today last November. Suddenly, Americans had a new set of initials to learn: CRP.

C-reactive protein is a measure of low-level inflammation in the body. In November, Paul Ridker and his colleagues at Harvard University reported that a high level of either CRP or LDL ("bad") cholesterol doubled the risk of heart disease. (It's not clear how USA Today concluded that high CRP was twice as bad.)

Does that mean you can forget about cholesterol? "We bent over backwards to say that we're not talking about replacing cholesterol tests with CRP tests," says Ridker. "We're saying that both LDL and CRP predict quite nicely who will get a heart attack."

Here's what you need to know about CRP, LDL, and your risk of heart disease.

"CRP has arrived," Ishwarlal Jialal, professor of internal medicine and pathology at the University of California at Davis, told American Medical News last December. "It is time for CRP testing to be used in primary prevention."

C-reactive protein is a measure of inflammation going on somewhere in the body (see September 2000, cover story). "It's such a low level of inflammation that the person has no symptoms and isn't sick," explains Paul Ridker, director of the Center for Cardiovascular Disease Prevention at Harvard Medical School.

It's not clear how inflammation may be linked to heart disease. The leading theory is that it makes cholesterol-filled plaques in arteries more likely to trigger a heart attack.

The initial "injury" may occur when cholesterol in the blood enters the artery wall and the body's immune system sends cells to repair the damage. "Over time, immune cells, cholesterol, and muscle cells form a plaque that is covered with a fibrous cap," says Ridker. "Inflammation can degrade the cap, making it more likely to rupture."

When the plaque ruptures, a blood clot forms. If the clot cuts off the flow of blood to the heart muscle, the person has a heart attack. If it cuts off blood to the brain, the person has a stroke.

"People with a high CRP are more likely to have unstable plaques, which are more prone to rupture," says Ridker. Researchers hope that by lowering CRP they can prevent heart attacks.

At least that's the theory.

"It's a very viable hypothesis," says Lori Mosca, professor of medicine at Columbia University and director of Preventive Cardiology at New York-Presbyterian Hospital. Mosca wrote an editorial that the New England Journal of Medicine published along with Ridker's new study. "I'm optimistic that it will pan out. But I was also optimistic that antioxidants and estrogen could prevent heart disease."

Millions of people took antioxidants or were prescribed estrogen based on that optimism. But both bombed when they were finally tested in clinical trials.

Beyond Framingham

It's not often that researchers find a new risk factor for heart disease.

"The first data from the Framingham Heart Study were presented in 1956," explains Jeremiah Stamler, a pioneering heart disease researcher who is now professor emeritus at Northwestern University Medical School in Chicago. "Before then, we knew nothing."

It took only four years for researchers to detect a clear pattern among the residents of the Massachusetts town: People with high cholesterol or high blood pressure had a higher risk of heart disease. "Smoking hadn't yet emerged as a risk factor," says Stamler.

Today, smoking is factored into the Framingham risk calculation, as are HDL ("good") cholesterol, age, and gender (see p. 7). And the National Cholesterol Education Program also tells doctors to consider high triglycerides, family or past history of heart disease, large waist size, and diabetes.

The question is: Does CRP tell doctors more about a person's risk than those classic risk factors?

"We found that among women with a given Framingham risk, those with a high CRP have a higher risk of cardiovascular disease than those with a lower CRP," says Ridker. (1)

But CRP wasn't as useful in predicting heart disease in studies like the Atherosclerosis Risk In Communities (ARIC). (2) "In our research, CRP has some predictive power, but it's not going to separate people a heckuva lot more than other risk factors," says lead investigator Aaron Folsom of the University of Minnesota School of Public Health.

"Once we took HDL into account, CRP wasn't a very strong predictor" for heart disease risk.

One reason why Ridker's results may differ from Folsom's: Ridker studied fairly healthy women--nurses, doctors, and other health professionals--while ARIC sampled the general population.

"Some people speculate that CRP may be more predictive in populations that don't smoke and are generally healthy, so their CRPs are lower," says Folsom. "In populations with higher CRP values, CRP may be less predictive."

Nevertheless, it seems clear that some people have a high CRP and few other risk factors. Wouldn't you want to know if you were one of them?

"I don't see how we can deny patients information that might tell them they're at risk," says Ridker. "It's like denying a woman a mammogram because it might lead to a biopsy she doesn't need." (Ridker holds patents on the use of CRP to diagnose heart disease.)

Who Needs a CRP Test

Should everyone get a CRP test?

"If someone has either a high or very low risk anyway, a CRP test wouldn't change what a physician recommends," says Folsom.

Physicians can tell who has a high or low risk using Framingham scores, which estimate your odds of having a heart attack over the next ten years (see p. 7). "If your Framingham score is less than five percent and a high CRP doubles it, the risk still won't be high, so your CRP doesn't matter," explains Ridker.

"And if your risk is already over 20 percent, a high CRP won't matter because your physician should already be treating you aggressively."

But measuring CRP might help physicians figure out how to treat people with an intermediate risk.

In January, the American Heart Association and the Centers for Disease Control and Prevention (CDC) issued new guidelines on CRP. (3) "For someone with a 10 percent to 20 percent risk for heart attack in the next 10 years, a CRP test might tip the scale to help a physician decide on moderate or more intensive treatment," says Thomas Pearson of the University of Rochester School of Medicine and Dentistry, who co-chaired the AHA-CDC panel. CRP would also help doctors decide how to treat people with the metabolic syndrome (see p. 5). But there is "no need for CRP screening of the entire adult population," Pearson adds. So CRP is not in the same category as cholesterol and high blood pressure.

In practice, though, most physicians may simply order a CRP test at the same time they order an LDL and HDL cholesterol test. Says Ridker: "Once you've stuck a needle in the patient's arm to test for HDL and LDL, it may not make sense to have them come back for a second visit to get another blood test. And that visit might cost them $200 for a $12 test."

What CRP Means

"CRP screening is only bad if it leads to inappropriate testing and treatment," says Columbia University's Lori Mosca. Here's what CRP results do and don't mean.

* What's high. A CRP below 1 (milligram per liter of blood) is "low," a CRP above 3 is "high," and anything in the middle (that is, 1 to 3) is "intermediate."

* One high CRP isn't reliable. If your CRP is high, don't panic. If it's over 10, the CDC advises doctors to throw out the results and look for a cause of inflammation or infection, which can temporarily raise your CRP.

"I recommend that patients shouldn't get their CRP tested unless they've been free from infection for at least two weeks," says the University of California's Ishwarlal Jialal. "If you've had a cold or had any trauma injury like a sprained ankle, it's a waste of money to get your CRP tested."

But even if your CRP is under 10 and you feel fine, once is not enough. "The CDC has concluded that doctors should measure CRP twice, about two or three weeks apart, and use the average," says Ridker. "That's nothing surprising. They're also supposed to measure cholesterol twice."

* Don't forget other risk factors. Don't assume that a low CRP is a carte blanche to ignore your cholesterol, blood pressure, weight, and exercise habits (see "The Big Picture," p. 5). "A low CRP could be falsely reassuring," says Mosca, if it leads you to assume that your overall risk is low.

The buzz about CRP may also distract people from getting those risk factors measured. "From a public health perspective, we can't afford to screen for everything," Mosca argues. "We're not even screening for the established risk factors adequately.

"A third of the people who are admitted to hospitals with a heart attack have never had their cholesterol measured. That should be the focus, not CRP."

* Remember what's not proven. People with a high CRP have a higher risk of heart disease than similar people with a low CRP, but that's only part of the equation. "We still don't know whether lowering CRP will prevent heart attacks," says the University of Minnesota's Aaron Folsom.

The Big Picture

If your CRP is consistently high, you've got four surefire ways to lower it: lose excess weight, stop smoking, exercise, or take statin drugs (see "Cut the CRP," p. 4). Some experts worry that most people will opt for Door Number 4.

Statins lower CRP (as well as LDL), they appear to be safe for most people, and they're easy to take. But they're expensive and they don't address the root causes of heart disease: obesity, high blood pressure, smoking, inactivity, and, perhaps most of all, a diet that's loaded with too much meat, cheese, french fries, and other foods that are high in saturated or trans fat.

"Physicians are more comfortable with numbers like CRP than with talking about nutrition and exercise," says Mosca. "We would have lower rates of heart disease if we could get people to follow our advice about diet and exercise."

The answer, says Ridker, is for doctors to think twice before they write out a prescription for medication. "We don't want to put 25 million Americans on statins without evidence that lowering CRP works." That's the estimated number of people with high CRP and low LDL.

Instead, he says, people should use CRP as one more reason to make lifestyle changes that are indisputably worthwhile, CRP or no CRP. "Our hope is that if a person has a CRP over three, it will motivate him or her to lose weight, exercise, and quit smoking," says Ridker.

(1) New Eng. J. Med. 347: 1557, 2002.

(2) American Heart Journal 144: 233, 2002.

(3) Circulation 107: 499, 2003.

The Risk of High CRP

[GRAPHIC OMITTED]

The risk of a heart attack rises in people as their LDL cholesterol (purple arrow) or their CRP (green arrow) climbs. But a person's overall risk is also influenced by other factors--like blood pressure, HDL, triglycerides, blood sugar, and weight.

Source: Circulation 107: 363, 2003.

Cut the CRP

It's still not clear if lowering your CRP will protect your heart. But if you do things that lower your CRP and keep you healthy for other reasons, you've got nothing to lose. Here's what lowers--or may lower--your CRP.

What Lowers CRP

* Not smoking. Smokers have higher CRP levels. It's not clear how quickly CRP drops when people quit. But quitting does cut the risk of heart disease, stroke, emphysema, bronchitis, and cancers of the bladder, esophagus, lung, mouth, and voicebox. Isn't that enough?

* Losing excess weight. Overweight people have higher CRP. And losing even 15 to 20 pounds can bring down not just CRP, but blood pressure, LDL ("bad") cholesterol, and the risk of diabetes. (1) "We have an enormous epidemic of obesity that we can't ignore," says Columbia University's Lori Mosca.

* Taking statins. Statin drugs like Lipitor, Pravachol, and Zocor lower CRP. (2) They also lower LDL and the risk of heart disease, stroke, and possibly Alzheimer's disease. But it's too early to assume that everyone with a moderate or high CRP should take statins.

"In one of our earliest studies, we found that statins lower CRP," says Harvard Medical School's Paul Ridker. "But we can't put 25 million people on statins based on that study because it wasn't designed to test statins on CRP.

"We need a randomized clinical trial before we can be sure that statins will help people with a low LDL and a high CRP." He has one in the works.

* Staying active. Studies have found that people who are more physically fit have lower CRP. (3) In One small study, CRP fell in people who were put on a training program to prepare for a marathon. (4) Of course, the list of reasons to exercise is so long that people hardly need another.

What May Lower CRP

* Lowering your glycemic load. People who eat a diet with a high glycemic load have higher CRP levels. (5) High-carbohydrate foods that raise blood sugar rapidly--like french fries, chips, colas, and white bread--have a high glycemic load.

But so far, no one has put people on a diet with a low glycemic load and measured their CRP levels. "We're in the process of planning that kind of study now," says Simin Liu of Harvard Medical School.

* Taking fish oil. Some studies have found lower CRP levels in people with higher levels of omega-3 fats in their blood cells. But when researchers gave either a placebo or a high dose of fish oil pills (4 grams a day) to 48 overweight people, it made no difference to their CRP levels. (6)

(1) Arterioscler. Thromb. Vasc. Biol. 21: 968, 2001.

(2) Circulation 106: 1447, 2002.

(3) Arterioscler. Thromb. Vasc. Biol. 22: 1869, 2002.

(4) International Journal of Sports Medicine 21: 21, 2000.

(5) Amer. J. Clin. Nutr. 75: 492, 2002.

(6) Clinical Chemistry 48(6 Pt 1): 877, 2002.

The Big Picture

Data from the Framingham Heart Study can help you gauge your risk of getting heart disease within the next ten years (see p. 7). So can your CRP level. But neither takes other risk factors into account (see "Read My Lipids," October 2001, cover story). Among them:

The metabolic syndrome

It's also called "syndrome X" and insulin resistance, and it raises your risk of both heart disease and diabetes. You've got it if you have at least three of these five risk factors:

1. Abdominal obesity
     Men                                 more than 40-inch waist *
     Women                               more than 35-inch waist
2. Triglycerides (mg/dL)                 150 or higher
3. HDL cholesterol (mg/dL)
     Men                                 under 40
     Women                               under 50
4. Blood pressure (mm Hg)                130 (systolic) or higher or
                                         85 (diastolic) or higher
5. Fasting blood sugar (mg/dL)           110-125

* In some men, a 37- to 39-inch waist can be a risk factor.

What to do: If you have the metabolic syndrome, lose weight (if your waist is too large) and exercise (even if you don't shed a pound).

History of heart disease or diabetes

If you have any of the following, your risk of having a heart attack or stroke is high:

* Clinical heart disease. Signs of heart disease include having had a previous heart attack, angina pectoris (chest pain when you walk or run), or procedures to prevent a heart attack (like coronary bypass surgery or angioplasty).

* Symptomatic carotid artery disease. The "symptoms" are a stroke or a TIA (transient ischemic attack, also called a temporary or mini-stroke) that the doctor can trace to clogged carotid (neck) arteries.

* Peripheral arterial disease. If the arteries leading to your legs or feet are clogged, it can cause cramping or fatigue when you're active. People with this disease have a higher risk of heart attacks and strokes.

* Abdominal aortic aneurysm. An aneurysm is a ballooning out of a section of a blood vessel, much like a bulge in an over-inflated inner tube. If the bulge occurs in a section of the aorta (the largest artery in your body) that passes through your abdomen, it's called an abdominal aortic aneurysm.

* Diabetes. People with diabetes are as likely to have a heart attack as people who have already had one. Diabetics also need aggressive treatment because they are more likely to die during or soon after a heart attack than non-diabetics.

What to do: Use diet or, if necessary, drugs to lower your LDL to 100 mg/dL or less.

A family history of premature heart disease

If your father or brother had heart disease before age 55 or your mother or sister had it before age 65, you're at increased risk.

What to do: If you know how to change this risk factor, let us know.

High LDL

Framingham uses total cholesterol instead of LDL ("bad") cholesterol to assess risk, because the study has more data for total and because total cholesterol mirrors LDL. However, it's good to know how your LDL results stack up.

LDL ("bad") cholesterol (mg/dL)

under 100     optimal
100-129       near optimal
130-159       borderline high
160-189       high
190 or more   very high

What to do: To lower your LDL, minimize saturated and trans fats. That means less red meat, cheese, full-fat ice cream, french fries, fried chicken and fish, pies, pastries, doughnuts, whole or 2% milk, cream, eggs, butter, shortening, stick margarine, and chocolate.

The Bottom Line

* If your Framingham score shows that you have a 10 to 20 percent risk of having a heart attack over the next 10 years (see p. 7) or if you're getting a blood test for other reasons, it's worth getting your CRP tested.

* If your CRP is low (below 1), don't worry. If your CRP is high (over 3), have it retested a month or two later.

* If your CRP is consistently high (over 3) or even intermediate (1 to 3), use it as motivation to lose excess weight, exercise, and stop smoking.

* Don't forget about other ways to cut your risk of heart disease. Cut back on saturated and trans fat and salt and shoot for eight to ten servings of fruits and vegetables a day.

What's Your Ten-Year Heart Attack Risk?

Women

Age      Points     Age     Points

20-34      -7      55-59       8
35-39      -3      60-64      10
40-44       0      65-69      12
45-49       3      70-74      14
50-54       6      75-79      16

   Total                         Points
Cholesterol     Age      Age      Age       Age      Age
  (mg/dL)      20-39    40-49    50-59     60-69    70-79

below 160        0        0        0         0        0
160-199          4        3        2         1        1
200-239          8        6        4         2        1
240-279         11        8        5         3        2
280 or more     13       10        7         4        2

                                Points
               Age      Age      Age       Age      Age
 Smoking      20-39    40-49    50-59     60-69    70-79

Non-smoker      0        0        0         0        0
Smoker          9        7        4         2        1

HDL (mg/dL)    Points    HDL (mg/dL)    Points

60 or more       -1         40-49         1
50-59             0       below 40        2

Systolic Blood
Pressure (mm Hg)    If Untreated    If Treated

below 120                0              0
120-129                  1              3
130-139                  2              4
140-159                  3              5
160 or more              4              6

        Risk Score

Point Total    10-Yr Risk (%)

less than 9     less than 1
     9                1
    10                1
    11                1
    12                1
    13                2
    14                2
    15                3
    16                4
    17                5
    18                6
    19                8
    20               11
    21               14
    22               17
    23               22
    24               27
25 or more       30 or more

Men

Age      Points     Age     Points

20-34      -9      55-59       8
35-39      -4      60-64      10
40-44       0      65-69      11
45-49       3      70-74      12
50-54       6      75-79      13

   Total                         Points
Cholesterol     Age      Age      Age       Age      Age
  (mg/dL)      20-39    40-49    50-59     60-69    70-79

below 160        0        0        0         0        0
160-199          4        3        2         1        0
200-239          7        5        3         1        0
240-279          9        6        4         2        1
280 or more     11        8        5         3        1

                                Points
               Age      Age      Age       Age      Age
 Smoking      20-39    40-49    50-59     60-69    70-79

Non-smoker      0        0        0         0        0
Smoker          8        5        3         1        1

HDL (mg/dL)    Points    HDL (mg/dL)    Points

60 or more       -1         40-49         1
50-59             0       below 40        2

Systolic Blood
Pressure (mm Hg)   If Untreated    If Treated

below 120               0              0
120-129                 0              1
130-139                 1              2
140-159                 1              2
160 or more             2              3

        Risk Score

Point Total    10-Yr Risk (%)

less than 0     less than 1
     0               1
     1               1
     2               1
     3               1
     4               1
     5               2
     6               2
     7               3
     8               4
     9               5
    10               6
    11               8
    12              10
    13              12
    14              16
    15              20
    16              25
17 or more       30 or more

To estimate your risk of having a heart attack over the next ten years, add up your points for age, total cholesterol, smoking status, HDL ("good") cholesterol, and systolic blood pressure (the higher of your two blood pressure numbers). These Framingham scores use total cholesterol, rather than LDL ("bad") cholesterol, because the study has better data for total cholesterol.

Framingham scores don't address every risk factor. When physicians assess your overall risk, they should also consider your (or your family's) history of heart disease and whether you have diabetes, as well as your LDL, triglycerides, waist size, and fasting blood sugar level (see "The Big Picture," p. 5).

(For a more precise estimate of your risk, go to www.nhlbi.nih.gov/guidelines/cholesterol/index.htm and click on 10-Year Risk Calculator--online version.)