Since pathologists and anatomists first began examining the heart, they realized that a connection existed between deposits of calcium and disease. When x-rays were discovered, calcium was again recognized as a disease marker. In fact, for most of the 20th century, calcium, because of its density, was the only feature that stood out on radiographs of the heart. In the 1950s, heart disease became more recognized as a significant cause of mortality in the United States. Along with this recognition came numerous publications about the ability to detect calcifications in the coronary arteries with radiography. In some ways, this period can be thought of as the first age of importance for calcium detection in the heart.
This period came to an end with the widespread acceptance of coronary angiography and other less invasive tests, such as stress thallium testing. If an actual stenosis or area of ischemia could be detected, attempts to qualitatively detect calcium with radiography or fluoroscopy seemed primitive. The advent of angioplasty and stent placement in the treatment of arterial stenoses seemed to herald the end of calcium detection.
Why, then, should this or any other article present information about detecting calcification in the coronary arteries? The answer is threefold.
First and foremost, calcium is a marker for a diseased artery.
The second is related to the recent revolution in CT scanning. Electron-beam CT (EBCT) was the first technique to provide a real breakthrough in the quantitation of calcium in the coronary arteries. Although this examination is valuable, the cost of the machines limited its use, and, by association, its impact. Some time afterward came helical, or spiral, CT. This technique was further improved with the addition of twin- and even quad-detector arrays. These machines allowed truly fast, completely noninvasive examination of the average person. During this period the scanners were still not quite sophisticated enough to allow direct visualization of the coronary arteries while filled with contrast material. This continued to focus attention on the capabilities and significance of calcium scoring.
Advances in CT technology have continued with the development of 16- and 64-slice scanners. With these scanners, more attention was directed to coronary artery CT angiography, but the use of calcium scoring in preventive cardiology had solidified. The latest scanners are volume-type 320-detector machines that can scan in a heartbeat two. The questions involving these procedures now have changed from "Can we do this?" to "When should we do this?" 
See the CT scan images of coronary artery calcification below.
Third, according to the American Heart Association, coronary artery disease caused 20% of all deaths in the United States in 2004, with mortality being 451,326. It is estimated that in 2008, 770,000 people in the United States will experience a first heart attack and 430,000 will experience a recurrent attack. In 2004, cardiovascular disease mortality in women was about 460,000, more than the combined deaths from lower respiratory disease, Alzheimer’s disease, accidents, and diabetes mellitus combined. 
Pathophysiology of Calcium in the Coronary Arteries
In an early study of autopsy findings in 2,500 patients, calcium in the coronary arteries and the total plaque burden were shown to be correlated. Patients who died of coronary artery disease were found to have 2-5 times as much calcium as those who died of other causes.
In June 2000, the American College of Cardiology (ACC) and American Heart Association (AHA) Consensus Panel wrote the following in the Journal of the American College of Cardiology: "Coronary calcium is part of the development of atherosclerosis; …it occurs exclusively in atherosclerotic arteries and is absent in the normal vessel wall." Simply put, the presence of calcification in the epicardial coronary arteries indicates that the patient has coronary atherosclerosis. 
This observation is of great significance, because atherosclerotic coronary artery disease is the number 1 cause of death in the Western world. Our ability to screen for coronary artery disease and, hopefully, prevent the sequelae of myocardial infarction and sudden cardiac death has traditionally depended on the assessment of atherosclerotic risk factors and on tests of coronary flow reserve.
Atherosclerotic risk factors have been evaluated in multiple longitudinal epidemiologic studies, such as the Framingham Heart Study. These studies have defined advancing age, male sex (or better stated, the absence of protective female hormones), hypertension, dyslipidemias, diabetes, cigarette smoking, and family history as predictors of subsequent cardiac events and angiographically demonstrated coronary artery disease.
Tremendous overlap exists, and sensitivities and specificities vary, even when multiple risk factors are applied. Novel risk factors have been proposed in an effort to enhance disease detection, particularly in asymptomatic patients. As a result, clinicians now may measure levels of homocysteine, fibrinogen, lipoprotein subunits (eg, lipoprotein A), C-reactive protein, and other biochemical markers of coronary atherosclerosis and subsequent cardiovascular events.
The stress test has been used for many years to noninvasively identify coronary artery disease and to screen patients who are at risk for subsequent cardiac events. Although it is valuable in populations in whom atherosclerotic risk factors may produce obstructive coronary lesions, stress tests—even those performed with associated nuclear and echocardiographic imaging techniques—frequently fail in the identification of patients who are at risk for subsequent cardiac events.
Why does this testing sometimes fail? First, Bayesian analysis reveals that the usefulness of any test depends on the pretest likelihood of the presence of disease. Therefore, if stress testing is used in a population of asymptomatic individuals, it lacks both sensitivity and specificity, because the prevalence of obstructive coronary artery disease is low in this group. More important, the mechanism of cardiac events (ie, myocardial infarction, sudden cardiac death) is not detectible with the stress test or any measure of coronary flow reserve. Multiple angiographic and epidemiologic studies have shown that the mechanism of myocardial infarction and/or sudden cardiac death in asymptomatic patients is plaque rupture with superimposed thrombosis. In most cases, the plaque burden is not flow limiting; therefore, the patient does not have a positive stress-test result or even a significantly abnormal coronary angiogram.
These facts have renewed our interest in imaging techniques that can be used to detect a coronary atherosclerotic plaque at a point in its natural history when flow-limiting obstructive disease does not exist. Coronary calcification can begin in patients as young as 10-20 years. The calcification itself is calcium phosphate (hydroxyapatite), which is similar to that in bone. Such calcium deposition was believed to be the result of a degenerative process, but evidence now suggests an active process, perhaps a response to injury, that is regulated in the fashion similar to bone mineralization. At this point, the mechanism of calcium deposition in areas of atherosclerotic plaque is not completely understood. 
Types of CT Scanners
The initial investigation of coronary artery calcification with CT was made possible with the development of the electron-beam CT (EBCT) scanner in the late 1980s. The speed of this machine was vastly superior to that of existing CT scanners. With this speed, it had the ability to "stop" heart motion enough to allow measurement of the amount of calcium in a coronary artery. Another revolution in CT has was the development of ultrafast spiral CT. [5, 6]
Principles of EBCT
One of the factors that limit the speed of a conventional CT scanner is necessary rotation of the tube around the patient. EBCT completely avoids this problem because the machine does not have any moving parts. A beam of electrons is generated and then focused with a series of electromagnets. The beam is directed onto 1 of 4 tungsten targets under the patient. The resultant fan-shaped X-ray beam passes through the patient and is collected by a 210° arc of detectors above the patient. More than 3,000 detectors are used in this process.
EBCT allows the acquisition of 1.5- to 3-mm sections, with an exposure time of 100 milliseconds. The images are gated to the end of diastole, and the entire examination is performed during 1 breath hold by the patient. Usually, 40-60 sections are obtained with this method.
Principles of multisection helical CT
Helical, or spiral, CT has renewed interest in many applications for CT. The major advantage to this technique is that it is faster than conventional CT. In addition, CT manufacturers have been able to put as many as 320 sets of detectors into the conventional donut configuration.
During a single-section spiral CT examination, the patient moves at a rate of about 5 mm/s while the tube rotates at a rate of up to 3 revolutions per second. Typically, a scanning length of 8-11 cm is used. Most existing spiral scanners are capable of a sub 100-ms acquisition time. Although this still hasn't equaled the original temporal resolution of EBCT, the other aspects of the multislice scanners make them overwhelmingly better. All major CT manufacturers now have 128-slice or greater machines available, with the latest being 320-slice.
Although each manufacturer has different protocols, the basic techniques are similar. No patient preparation is required. Blood samples do not have to be obtained, and no contrast material is used. Some manufacturers recommend the removal of any metal object that may be near the chest region. Examples include metal buttons, bras with underwires, and necklaces. Metal objects are removed because they cause non-linear x-ray scatter that can produce artifacts in the images.
Asking patients to practice holding their breath may be helpful, not because a long breath hold is needed (usual duration, 15-30 s), but because reproducibility of their breath-holding is enhanced. Many centers ask the patient to complete a risk-assessment questionnaire to aid in the overall interpretation of the study. The patient lies supine on the scanner gantry with the arms over the head. If the patient cannot raise the arms, an acceptable scan can be obtained with the patient's arms at his or her sides.
Settings for the scanner depend on the manufacturer's recommendations. A typical protocol for a quad-slice multidetector CT would be 165 ma, 120 kVp, 0.5 pitch, and quad X 2.5 mm.
The use of cardiac gating is an area of current disagreement. Some manufacturers do not use it at all, while others disagree about whether it should be used prospectively or retrospectively. Although the addition of gating is not difficult, it requires more patient preparation than that of the simple CT scanning. Leads must be placed on the patient's chest; at some centers, the patient may need to wear a hospital gown.
Coronary segments with a luminal obstruction of greater than 50% are likely to have some calcification that is detectable with electron-beam CT (EBCT). In one trial, a 0 calcium score had a 100% predictive value in the exclusion of angiographic evidence of obstructive epicardial coronary lesions. The higher the calcium score, the more likely the presence of angiographic obstructive disease. In another study,  a calcium score greater than 371 had a 90% specificity in the detection of a luminal obstruction of greater than 70%. Specificity tends to decrease with advanced patient age, but it increases with the number of calcified vessels as well as the total calcium score. 
In a study in which calcium scores and thallium stress test results were compared, almost one half of the patients with scores greater than 400 had a normal thallium stress result.  Such testing may not be contradictory in terms of the pathophysiology; thallium detects inducible ischemia, not plaque burden.
Coronary calcification is strongly associated with the prognosis. Indeed, the extent of coronary atherosclerosis (total calcium score) is the most powerful predictor of subsequent or recurrent cardiac events. This was true in the early days when calcium was detected with fluoroscopy and conventional CT.
When EBCT calcium scores became available, the prognostic value of coronary calcification was again affirmed. The higher the calcium score, the worse the prognosis. [10, 11, 12] The degree of coronary calcium was a good predictor of the development of symptomatic cardiovascular disease. In a study by Agatston et al, the mean calcium score for patients with a cardiovascular event was 399, compared with a mean score of 76 in those without such an event. One study suggested that the detection of coronary calcification at EBCT was a better predictor of subsequent events than many traditional risk factors, including those evaluated in the Framingham database. 
Cardiac events do occur in patients with low calcium scores, but the incidence is low. Intravascular ultrasonographic studies show that as many 30% of coronary plaques are devoid of calcium. In an autopsy study,  the benefit of combined assessment of coronary artery calcification and risk factors (Framingham Risk Index) in predicting sudden cardiac death was apparent. In the study, 79 consecutive adults with sudden cardiac death were evaluated by using a Framingham Risk Index and histologic findings of coronary calcification. The risk classifications with the 2 techniques agreed in a majority of the patients. Patients with plaque erosion (as opposed to plaque rupture) who were dying of sudden cardiac death had significantly less coronary calcification and lower Framingham Risk Indexes.
Clearly, in establishing the cardiac risk, traditional coronary artery disease risk factors and coronary calcification may be most useful when used in combination. Whether risk stratification is further enhanced with the use of novel risk factors is yet to be determined. [10, 15, 16, 17]
Calcium scoring can be accomplished without cardiac gating, but most of the current work is devoted to either prospective or retrospective gating. At this time, every major manufacturer has or is working on both of these methods. Retrospective gating may be proven to be the most accurate technique, because it allows the operator to choose the optimum time during diastole for image selection. [18, 19, 20]
In terms of nontechnical aspects, the most important work being performed now is the formation of large databases. Only long-term analysis of this data will reveal the ultimate value and role for this procedure.
The most exciting possibility with calcium scoring may be CT angiography in the coronary arteries. As the scanners become faster and as the 3-dimensional computer postprocessing workstations become more powerful, this examination may become a reality. Already, preliminary studies are being performed in Europe to evaluate the feasibility of CT angiography of the coronary arteries.