Diagnostic Imaging
November 2002

Consumer Imaging

CT screening: Is technology running ahead of the science?

Although not supported by data, self-referred scans are catching on with patients and are probably here to stay

Recent technical advances in CT scanning have led to burgeoning interest in CT-based screening examinations. The main thrust of CT screening has been in the early detection of coronary artery disease, lung cancer, and colon cancer. Whole-body CT screening is gaining popularity.

An ideal screening test should detect the targeted disease process with a high degree of accuracy at a stage when treatment is relatively effective. It should also be safe, cost-effective, and widely available. Although CT screening has some strengths when judged by these criteria, additional data are needed before reasonable recommendations can be made regarding its long-term safety and efficacy. At the same time, there is a growing demand for these services and self-referred CT screening appears to be here to stay.

CORONARY ARTERY CALCIUM SCORING

Subclinical atherosclerosis can be measured by quantifying calcification in the coronary arteries. Electron-beam CT (EBCT) with its short acquisition times and prospective electrocardiogram-gated image acquisition is well suited for this purpose. Multidetector CT is also being evaluated for use in coronary artery screening (Figure 1). With the recent introduction of 16-detector scanners, the high gantry speed and superior spatial resolution of spiral CT should prove effective in identifying small calcified plaques.

Commercially available software is used to analyze calcium deposits in the coronary arteries. The area of the calcified plaque is multiplied by a coefficient based on the peak pixel attenuation of the lesion to yield a plaque-specific score. These scores are then summed to provide a total coronary artery calcium score (CCS).¹

This noninvasive technique provides an objective score that reflects the degree of coronary atherosclerosis. There is a direct correlation between the amount of calcium in the coronary arteries and the degree of coronary atherosclerosis. Asymptomatic patients with calcium scores below average for their age and sex are less likely to have obstructive atherosclerosis than patients with higher than average scores. A calcium score in the top decile for a given age group is associated with a high likelihood of obstructive coronary atherosclerosis. High calcium scores have also been associated with an increased likelihood of perfusion defects on radionuclide stress tests.

CT-based calcium scoring does have problems, however. Reported interscan variability in calcium scores is relatively high, ranging from 19% to 49%.^2,3 Causes of interscan variability include cardiac motion, partial volume effects, image noise, and intra- and interobserver variability. Technical differences such as choice of mean versus peak attenuation of the calcified plaque and the type of ECG gating used for data acquisition can also result in considerable variation in calcium scores.

This variability decreases the effectiveness of coronary calcium scoring in assessing the progression of atherosclerotic disease and hampers its use in monitoring the response to treatment. Interscan variability may be lower with an alternative measurement called the calcium volume score (CVS),⁴ but the use of CVS complicates comparisons with prior studies, most of which have been based on CCS methodology.

One of the goals of CT screening is to detect early atherosclerotic disease and then implement dietary and lifestyle changes to prevent disease progression. Lipid-lowering drugs are also commonly used to slow the progression of atherosclerotic disease. Interscan variability is more pronounced in subjects with low calcium scores, which could hinder its use in asymptomatic subjects. On the other hand, high calcium scores often result in referrals for coronary angiography and possible angioplasty or coronary stent placement if significant stenoses are identified. Data are currently insufficient to determine whether coronary artery calcium screening leads to a reduction in morbidity and mortality that would justify the increased expense of these procedures.

Another drawback is that significant coronary atherosclerosis and stenosis can be present without detectable calcium on EBCT, particularly in patients younger than 40.⁵ Soft plaques with a lipid core may be unstable and more prone to rupture than the calcified plaques detected with CT screening.

It has been suggested that coronary artery calcium scores may be most useful in patients with intermediate risk of coronary artery disease.⁶ If these patients are found to have high calcium scores, aggressive interventions can be undertaken to reduce the risk of disease progression. Other patients who can potentially benefit from CT screening are young adults and postmenopausal women with single risk factors such as elevated homocysteine, isolated low HDL cholesterol, or elevated LDL cholesterol.

LUNG CANCER

Lung cancer is the leading cause of cancer death among both men and women in the U.S., and approximately 170,000 new cases are diagnosed annually in this country. An estimated 154,900 people in the U.S. will die of lung cancer in 2002, accounting for 28% of all cancer deaths.⁷ The prognosis for patients diagnosed with lung cancer is dismal, with an average five-year survival rate of 15%. The survival rate improves to 48%, however, for cases detected while the disease is still localized.⁷ On this basis alone, the idea of early detection seems very attractive.

Unfortunately, lung cancer screening is complicated by the high prevalence of benign pulmonary nodules (Figure 2), which varies substantially from one geographic region to the next. For example, pulmonary nodules are particularly common in areas where histoplasmosis is endemic, such as the Midwest. Among the 1000 asymptomatic smokers over 60 who were screened in the Early Lung Cancer Action Project, 23% had one or more noncalcified nodules.⁸ Only 12% of those nodules were found to be lung cancers. In a Japanese screening study, suspicious pulmonary lesions were seen in 236 of 5483 subjects, or 4.3% of the study population.⁹ Sixty of those subjects were diagnosed with lung cancer.

Lung cancer screening is based on the premise that detection of small primary tumors will decrease the mortality rate. But there is currently no proof that early detection reduces mortality.^7,10 In a cohort of 510 patients with stage IA lung cancer, Patz et al found no significant correlation between tumor size and survival.¹¹ A possible confounding factor is micrometastases in the bone marrow, which may be present in up to 60% of patients with resectable non-small cell lung cancer.¹² Although detection of early-stage disease would seem to be a worthwhile goal, further research is needed to determine whether CT-based screening will improve survival.

The National Cancer Institute has launched a large multicenter trial to determine whether lung cancer screening with low-dose helical CT reduces cancer-specific mortality relative to screening with chest radiographs in a high-risk cohort. The NCI hopes to enroll 50,000 current or former cigarette smokers between the ages of 55 and 74 in this study.

The American College of Radiology Imaging Network (ACRIN) is sponsoring a similar trial called Contemporary Screening for the Detection of Lung Cancer. As in the NCI study, high-risk subjects will be randomized to receive annual screening with low-dose helical CT or chest radiographs. ACRIN aims to enroll 10,000 smokers between the ages of 55 and 74 in this study. These investigations will provide valuable data about the efficacy of CT-based screening and its potential impact on lung cancer mortality rates.

COLORECTAL CANCER

Colorectal cancer is the third leading cause of new cancer diagnoses and cancer deaths in the U.S. An estimated 148,300 new diagnoses and 56,600 deaths from cancer of the colon or rectum were expected this year.⁷ There is no debate about the value of early detection of colorectal cancer. Screening programs can markedly lower the mortality rate by detecting and removing adenomatous polyps and diagnosing malignant disease at a stage when it can be effectively cured.

Screening for colorectal cancer can be performed with periodic fecal occult blood tests, colonoscopies, and/or barium enema examinations. Despite the proven effectiveness of colorectal cancer screening, these tests are drastically underutilized. In 1999, only 19% of adults 50 and older had had a fecal occult blood test within the preceding year, while only 32% had had sigmoidoscopy or colonoscopy within the preceding five years.⁷

As a result of the poor adherence to colorectal cancer screening guidelines, 63% of cases are diagnosed when the disease has already spread beyond the colon. Several factors contribute to the underutilization of screening tests, including lack of patient education and variable insurance coverage. Another factor is that many patients delay or avoid colonoscopy and barium enema examinations because of concerns about discomfort and potential risks associated with these procedures.

CT colonography mitigates some of these concerns and is currently being explored as a possible alternative screening method (Figure 3). Although safe and well tolerated, CT colonography typically involves a preprocedure cathartic regimen that can cause discomfort. Stool tagging using dilute barium has shown promise in eliminating the need for a cathartic regimen, which should help make the procedure more acceptable to patients.¹³ Another potential source of patient discomfort is colonic air insufflation. About 1500 to 2000 cc of air is typically insufflated to attain adequate colonic distention. Using multislice CT, the entire data set is then obtained during a 20-second breath-hold. A variety of postprocessing methods are used to interrogate the image data, including 2D and 3D multiplanar reconstruction and endoluminal flythrough images (Vitrea 2, Vital Images).

The sensitivity and specificity of CT colonography depend on the chosen threshold for lesion size. Most protocols have targeted polyps 1 cm or greater in diameter.¹⁴ In a cohort of 180 patients, the sensitivity for identifying patients with polyps greater than 1 cm was 85%, while the specificity was 93%.¹⁵ The sensitivity for detecting individual lesions was 75% for polyps greater than 1 cm, but only 47% for polyps between 0.5 and 1 cm. The sensitivity for detecting polyps approaches 100% for lesions greater than 2 cm in diameter.¹⁶

Polyps may be missed on CT colonography because of retained fluid and incomplete colonic distention. These pitfalls can largely be overcome by scanning patients in both the prone and supine positions. Retained stool within the colon can mimic polypoid lesions and result in false-positive diagnoses. Stool tagging techniques may help minimize this problem in the future. The ability to detect extracolonic pathology is an advantage of CT colonography compared with other screening methods. Hara et al found clinically relevant extracolonic abnormalities in 11% of patients who underwent CT colonography.¹⁷

An ACRIN-sponsored multicenter trial is being developed to evaluate the accuracy of CT colonography in detecting colorectal neoplasia in subjects 65 and older. The goals of the study are to validate the efficacy of screening with CT colonography, describe the morphologic features of colonic lesions, and assess patient preferences and cost-effectiveness.

PROS AND CONS

CT-based screening examinations are enormously appealing to our aging and increasingly health-conscious population. They offer great potential for early detection and improved treatment of a variety of devastating diseases. But the general public tends to overlook the possibility that screening asymptomatic subjects can also lead to a plethora of clinically irrelevant findings resulting in unnecessary diagnostic procedures, excess radiation exposure, unwarranted treatments, and great expense.

Radiologists who offer CT screening examinations owe their patients a frank explanation of the current state of knowledge, including the potential for harm due to false-positive diagnoses. It is likely that studies will validate the use of CT-based screening in some settings. The results of these studies should be closely monitored so that screening examinations can be offered to the patients most likely to benefit from them. Third-party payers should then be lobbied to provide reimbursement for these examinations in the appropriate clinical settings.

Dr. Pattekar is an abdominal imaging fellow at the Mallinckrodt Institute of Radiology, and Dr. Brown is director of clinical research and co director of MRI at Washington University in St. Louis.

References

1. Agaston AS, Janowitz WR, Hildner FJ, et al. Quantification of coronary artery calcium using ultra-fast computed tomography. J Am Coll Cardiol 1990;15:827-832.
2. Yoon HC, Goldin JG, Sayre J, et al. Interscan variation in coronary artery calcium quantification in a large asymptomatic patient population. AJR 2000;174:803-809.
3. Wang S, Detrano RC, Secci A, et al. Detection of coronary calcification with EBCT: evaluation of interexamination reproducibility and comparison of three image-acquisition protocols. Am Heart J 1996;132:550-558.
4. Yoon H-C, Emerick AM, Hill JA, et al. Calcium begets calcium: progression of coronary artery calcification in asymptomatic subjects. Radiology 2002;224:236-241.
5. O'Rourke RA, Brundage BH, Froelicher VF, et al. American College of Cardiology/American Heart Association expert consensus document on electron-beam computed tomography for the diagnosis and prognosis of coronary artery disease: committee members. Circulation 2000;102:126-140.
6. Grundy SM. Coronary calcium as a risk factor: role in global risk assessment. J Am Coll Cardiol 2001;37:1512-1515.
7. Cancer facts and figures 2002. Atlanta: American Cancer Society, 2002.
8. Henschke CI, McCauley DI, Yankelevitz DF, et al. Early lung cancer action project: overall design and findings from baseline screening. Lancet 1999;354:99-105.
9. Sone S, Li F, Yang Z-G, et al. Results of three year mass screening programme for lung cancer using mobile low-dose spiral computed tomography scanner. Br J Cancer 2001;84:25-32.
10. Reich JM. Improved survival and higher mortality. The conundrum of lung cancer screening. Chest 2002;122;329-337.
11. Patz EF, Rossi S, Harpole DH, et al. Correlation of tumor size and survival in patients with pathologic stage 1A non-small cell lung cancer. Chest 2000;117:1568-1572.
12. Pantel K, Izbickie J, Passlick B, et al. Frequency and prognostic significance of isolated tumor cells in bone marrow of patients with non-small cell lung cancer without overt metastases. Lancet 1996;347:649-653.
13. Callstrom MR, Johnson CD, Fletcher JG, et al. CT colonography without cathartic preparation: feasibility study. Radiology 2001;219:693-698.
14. Winawer SJ, Fletcher RH, Miller L, et al. Colorectal cancer screening: clinical guidelines and rationale. Gastroenterology 1997;112: 597-642.
15. Fletcher JG, Johnson CD, Welch TJ, et al. Optimization of CT colonography technique: prospective trial in 180 patients. Radiology 2000;216:704-711.
16. Royster AP, Fenlon HM, Clarke PD, et. al. CT colonoscopy of colorectal neoplasms: two-dimensional and three-dimensional virtual-reality techniques with colonoscopic correlation. AJR 1997;169:1237-1242.
17. Hara AK, Johnson CD, MacCarthy RI, Welch TJ. Incidental extracolonic findings at CT colonography. Radiology 2000;215:353-357.