The chest was one of the first targets of routine radiographic studies. The plain chest film continues to be common, although it is being used more selectively as a result of recent trials.^1-3 As more digital equipment is installed, techniques such as computer-aided detection and temporal subtraction studies are improving its utility. But chest radiography is being supplanted by other modalities. This article looks at some of the newer developments, with an emphasis on areas of controversy.

Spiral CT is now the method of first choice for purposes such as detection and characterization of pulmonary nodules, staging of lung cancer, examination after chest trauma, assessment of congenital anomalies, and evaluation for fistulas and dehiscenses. High-resolution CT is more sensitive than chest films for detecting and characterizing chronic infiltrative lung disease.⁴ Multidetector scanners permit virtual bronchoscopy, which compares well with fiberoptic bronchoscopy in the detection and measurement of all but flat mucosal lesions.^5,6

Two possible indications for spiral CT are the subject of intense debate.

PULMONARY EMBOLISM: YES OR NO?

"Pulmonary embolism is more often diagnosed post mortem by pathologists than in vivo by clinicians" is the way two observers described the problem presented by this potentially fatal disorder.⁷ The classic imaging protocol has been chest radiography together with a ventilationperfusion (V/Q) scan. According to the first Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED I),⁸ 88% of patients with a high-probability V/Q scan actually have pulmonary emboli, as judged by pulmonary angiography. Unfortunately, only a minority of patients with pulmonary emboli have high-probability scans, so V/Q "by itself is limited in its ability to lead to a conclusive diagnosis of pulmonary embolism."⁹

At many medical centers, spiral CT pulmonary angiography (CTPA) is now an important part of the work-up for pulmonary embolism. This modality has several benefits: it directly depicts many emboli, it provides a view of the pulmonary parenchyma and other chest wall anatomy that may suggest an alternative explanation for the patient's symptoms, and it allows venography as part of the same study, permitting evaluation of the pelvic and leg veins for the thrombi that give rise to pulmonary emboli.¹⁰ The likelihood of a nondiagnostic study is relatively low. A meta-analysis of 12 trials enrolling 1,250 patients indicated a sensitivity of 74.1% and a specificity of 89.5% for CTPA in the diagnosis of pulmonary emboli.¹¹

The availability of CTPA has had dramatic effect on the diagnostic approach to pulmonary embolism in the last decade,¹² although its exact role has not been defined. At The Canberra Hospital in Australia, a V/Q scan is the initial study because of its low radiation dose, with CTPA being performed if the scan shows an intermediate probability of embolism or if the scan indicates a low probability but there is strong clinical evidence of pulmonary embolism.¹³ At the University of Vienna, CTPA is combined with ultrasonography of the legs to search for evidence of deep venous thrombosis.¹⁴ The V/Q scan is used primarily to exclude rather than confirm pulmonary embolism.¹⁵ On the basis of their meta-analysis, Safriel and Zinn of the State University of New York suggested that CTPA would be a good first study for suspected pulmonary embolism.¹¹

Because of their faster scan times (of particular benefit in these patients, who usually have difficulty with breathholding) and their ability to depict more-peripheral arteries (and thus smaller emboli), multidetector CT scanners hold promise of even greater utility in the diagnosis of pulmonary embolism. The Medical Research Group Equipe d'Accueil recently compared pulmonary angiograms obtained by spiral and multidetector scanners.¹⁶ Motion artifacts were less common with the latter, and more examinations could be interpreted to the level of the subsegmental arteries. Use of 1-mm reconstructions permits detection of more subsegmental emboli and reduces disagreement among the radiologists.¹⁷

Arnaud Perrier, MD, and colleagues at the Geneva University Hospital in Switzerland recently examined the question of how to use the available modalities to achieve a diagnosis at the lowest cost.¹⁸ In the experience at that institution, when the clinical picture indicates a low probability of embolism, the most cost-effective strategy is a V/Q scan, ultrasonography of the leg veins, and laboratory D-dimer assay. In patients with an intermediate or high clinical probability of pulmonary embolism, either spiral CT or angiography would be performed for patients with a nondiagnostic V/Q scan. However, if the sensitivity of CT exceeded 85%, as might be expected with a multidetector scanner, the most cost-effective strategy for all patients would be leg ultrasonography, D-dimer assay, and CT. As a single modality, spiral CT was not cost-effective.

Enthusiasm for spiral CT for the diagnosis of pulmonary embolism is not universal. In a recent review, Matthew S. Johnson, MD, assistant professor, radiology, at Indiana University School of Medicine, Indianapolis, noted that many of the reported trials have design flaws and that both false-positive and false-negative studies are possible even with multidetector scanners.⁹ In its most recent clinical guidelines,¹⁹ the American College of Chest Physicians described spiral CT as "still under investigation" and said that "no firm general conclusions [about its value] can be made without more extensive experience." Both of these articles call attention to the magnitude of the contrast dose that will be required if the CT scan is nondiagnostic and pulmonary angiography must be performed. Johnson also commented that the view of traditional pulmonary angiography as highly risky and therefore best avoided is not justified by recent series, in which the mortality rate is less than 1% and the major complication rate 1%.⁹

To establish the role of spiral CT, the National Heart, Lung, and Blood Institute began PIOPED II in the fall of 2000. Unlike PIOPED I, in which pulmonary angiography was the gold standard, PIOPED II is using composite endpoints consisting of various combinations of V/Q scans, compression ultrasonography of the legs, digital subtraction pulmonary angiography, and contrast venography.²⁰ The results are expected to be available next year.

LOW-DOSE CT SCREENING

The first attempts to screen high-risk patients for lung cancer were reported more than 20 years ago. The hope was that more cancers could be discovered while they were still small and curable. One study, the NCI Cooperative Early Lung Cancer Group trial, which used a combination of chest radiography and sputum examination, did indeed find more cancers in low stage: 40% vs 15%. The 5-year survival rate of the patients was improved from 15% to 35%. However, there was no change in the overall survival rate for the series. Enthusiasm for screening faded.

Interest has returned with the availability of low-dose and high-resolution CT, but the practice is controversial. Early data demonstrate that CT can indeed find small lung cancers. For example, at the Mayo Clinic, in the first year of a study that enrolled 1,520 asymptomatic subjects, 25 non-small-cell lung cancers were identified, 23 of them by CT. The average size of the lesions was 17 mm, and 56% were stage I.²¹ An update presented at the 2002 Scientific Meeting of the Radiological Society of North America described the finding of 40 cancers during the first 2 years, with 63% being in Stage I. In Nagano, Japan, 87 cancers were found during the first 3 years of a study of 7,847 persons. The mean size of these tumors was 13 mm.²² In a third series, reported from Germany, 12 cancers were found at baseline in 817 subjects, with slightly more than half in Stage I.²³

Many more years of follow-up will be needed to determine the impact of such early discovery on lung cancer survival, but these results might seem to make screening advisable. However, no major organization has endorsed the practice to date. One reason is the high prevalence of findings that need to be followed up with additional studies. For example, the Mayo study identified a total of 2,244 noncalcified (potentially malignant) lesions during the first year, with 66% of the subjects having at least one lesion. In the Japanese series, 745 "suspicious" lesions were identified in the first 3 years. In the German series, 43% of the subjects had noncalcified nodules, a total of 858 lesions, on their first scan. Although computer analysis is helping to clarify the nature of many such nodules,^22,24 further imaging, and sometimes invasive procedures, may be required to determine whether a given lesion is malignant.

There are significant costs associated with such follow-up. Obviously, there is a considerable monetary cost. Also, there is a biologic cost. According to a recent estimate,²⁵ annual screening scans for lung cancer with follow-up studies to determine the nature of all of the lesions discovered could impose an annual radiation dose as high as 40 mSv, which "could create a significant risk of developing fatal and non-fatal cancers."²⁵

To determine the value of regular screening as a means of reducing mortality from lung cancer, the US National Cancer Institute has organized an 8-year study, the National Lung Cancer Screening Trial, comparing chest radiography and low-dose spiral CT. The two arms of the trial are the Lung Screening Study, directed by the NCI, and the Contemporary Screening for the Detection of Lung Cancer, directed by the American College of Radiology Imaging Network (ACRIN). Enrollment of 50,000 current or former smokers began in September 2002 and is expected to be complete within 2 years.

Assuming CT screening can indeed identify lung cancers in a curable stage, is the cost justified? Investigators in Canada and the United States have suggested that it could be cost-effective. Chirikos and colleagues of the H. Lee Moffitt Cancer Center and Research Institute at the University of South Florida in Tampa, whose model was based on a worst-case scenario (maximum cost, lowest yield), determined that if CT allowed half of all lung cancers to be detected when they are localized, screening would cost approximately $48,000 per life-year gained and would be cost-effective if more than half of the new cancers could be detected in a low stage.^26,27 In their view, "cost factors should not be used to deter definitive trials of clinical effectiveness of lung cancer screening with CT."²⁷ Marshall and associates at the Center for the Evaluation of Medicines at McMaster University in Hamilton, Ontario, found that annual screening of high-risk elderly patients (aged 60 to 74 years) might be cost-effective "under optimal conditions."²⁸

However, a just-published study by investigators at MEDTAP International²⁹ is more discouraging. According to their computer model, which explored the costs of 20 years of annual screening in 60-year-old current, quitting, and former heavy smokers, the incremental cost-effectiveness for current smokers was $116,300 per quality-adjusted life year gained. The corresponding costs for quitting and former smokers were $558,600 and $2,322,700, respectively. Summarizing their findings, the investigators wrote: "Given the current uncertainty of benefits, the harms from invasive testing, and the high costs associated with screening, direct to consumer marketing of helical CT is not advisable."

USES FOR MRI AND SONOGRAPHY

Magnetic resonance imaging is finding its own place in chest imaging, as it promises to permit comprehensive anatomic and functional assessment of the lungs. Inhaled contrast agents such as perfluorocarbons and hyperpolarized noble gases provide high spatial and temporal resolution images for purposes such as evaluation of the regional severity of chronic obstructive pulmonary disease.^30,31 Early work suggests that MRI ventilation studies with MR angiography will be useful in the diagnosis of pulmonary embolism.^32-34

Because of the anatomy of the chest, ultrasonography continues to play a relatively minor role in chest imaging. However, its freedom from ionizing radiation and ease of bedside use have guaranteed the modality a role in the detection and assessment of air and fluid collections and in guidance of aspiration.^35,36 In critically ill patients, sonography may be helpful in the diagnosis of pulmonary embolism.³⁷

CONCLUSION

Significant improvements in CT and MR and the introduction of new techniques such as digital radiography and PET have changed the diagnostic approach to many chest diseases, even though the best combinations of studies remain to be defined for some conditions. The improved imaging has also increased our understanding of the pathophysiology of lung disease. Once again, as in the early days of roentgen rays, chest imaging is at the forefront of radiology.

Judith Gunn Bronson, MS, is a contributing writer for Decisions in Imaging Economics.

References:

1. Rothrock SG, Green SM, Costanzo KA, Fanelli JM, Cruzen ES, Pagane JR. High yield criteria for obtaining non-trauma chest radiography in the adult emergency department population. J Emerg Med. 2002;23:117–124.
2. Pacharn P, Heller DN, Kammen BF, et al. Are chest radiographs routinely necessary following thoracostomy tube removal? Pediatr Radiol. 2002;32:138-142.
3. Houghton D, Cohn S, Schell V, Cohn K, Varon A. Routine daily chest radiography in patients with pulmonary artery catheters. Am J Crit Care. 2002;11: 261"265.
4. Swensen SJ, Aughenbaugh GL, Douglas WW, et al. High-resolution CT of the lungs: findings in various pulmonary diseases. AJR Am J Roentgenol. 1992; 158:971-979.
5. Finkelstein SE, Summers RM, Nguyen DM, Stewart JH IV, Tretler JA, Schrump DS. Virtual bronchoscopy for evaluation of malignant tumors of the thorax. J Thorac Cardiovasc Surg. 2002;123: 967-972.
6. Hoppe H, Walder B, Sonnenschein M, Vock P, Dinkel HP. Multidetector CT virtual bronchoscopy to grade tracheobronchial stenosis. AJR Am J Roentgenol. 2002; 178:1195-1200.
7. Pistolesi M, Miniati M. Imaging techniques in treatment algorithms of pulmonary embolism. Eur Respir J Suppl. 2002; 35:28s-39s.
8. PIOPED Investigators. Value of ventilation/perfusion scan in acute pulmonary embolism: results of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). JAMA. 1990; 263:2753-2759.
9. Johnson MS. Current strategies for the diagnosis of pulmonary embolus [review]. J Vasc Intervent Radiol. 2002;13:13-23.
10. Bloomgarden DC, Rosen MP. Newer diagnostic modalities for pulmonary embolism: pulmonary angiography using CT and MR imaging compared with conventional angiography. Emerg Med Clin North Am. 2001;19:975-994.
11. Safriel Y, Zinn H. CT pulmonary angiography in the detection of pulmonary emboli: a meta-analysis of sensitivities and specificities. Clin Imaging. 2002;26:101-105.
12. Washington L, Goodman LR, Gonyo MB. CT for thromboembolic disease. Radiol Clin North Am. 2002;40:751-771.
13. Chen SW, Mouratidis B. Comparison of lung scintigraphy and CT angiography in the diagnosis of pulmonary embolism. Australas Radiol. 2002;46:47-51.
14. Herold CJ. Spiral computed tomography of pulmonary embolism. Eur Respir J Suppl. 2002;35:13s-21s.
15. Zettinig G, Baudrexel S, Leitha T. Introduction of helical computed tomography affects patient selection for V/Q lung scan. Nuklearmedizin. 2002;41:91"94.
16. Remy-Jardin M, Tillie-Leblond I, Szapiro D, et al. CT angiography of pulmonary embolism in patients with underlying respiratory disease: impact of multislice CT on image quality and negative predictive value. Eur Radiol. 2002;12: 1971-1978.
17. Schoepf UJ, Holzknecht N, Helmberger TK, et al. Subsegmental pulmonary emboli: improved detection with thin-collimation multi-detector row spiral CT. Radiology. 2002;222:483-490.
18. Perrier A, Nendaz MR, Sarasin FP, Howarth N, Bounameaux H. Cost-effectiveness analysis of diagnostic strategies for suspected pulmonary embolism including helical computed tomography. Am J Respir Crit Care Med. 2003;
19. ACCP Consensus Committee. Opinions regarding the diagnosis and management of venous thromboembolic disease. Chest. 1998;113:499-504.
20. Gottschalk A, Stein PD, Goodman LR, Sostman HD. Overview of Prospective Investigation of Pulmonary Embolism Diagnosis II. Semin Nucl Med. 2002; 32:173-182.
21. Swensen SJ, Jett JR, Sloan JA, et al. Screening for lung cancer with low-dose spiral computed tomography. Am J Respir Crit Care Med. 2002;165:508-513.
22. Li F, Sone S, Abe H, MacMahon HM, Li Q, Doi K. Low-dose CT screening for lung cancer: HRCT findings, clinical outcome and CAD schemes in benign versus malignant nodules [abstract 160CH-p]. Radiology. 2002;225P:148.
23. Diederich S, Wormanns D, Semik M, et al. Screening for early lung cancer with low-dose spiral CT: prevalence in 817 asymptomatic smokers. Radiology. 2002; 222:773-781
24. Armato SG III, Li F, Giger ML, MacMahon H, Sone S, Doi K. Lung cancer: performance of automated lung nodule detection applied to cancers missed in a CT screening program. Radiology. 2002; 225:685-692.
25. Goodman PC. Radiation risk in CT screening for lung cancer [abstract 2]. Radiology. 2002;225P:233.
26. Chirikos TN, Hazelton T, Tockman M, Clark R. Screening for lung cancer with CT: a preliminary cost-effectiveness analysis [see comment]. Chest. 2002; 121:1507-1514.
27. Chirikos T, Hazelton TR, Tockman MS, Clark RA. Screening for lung cancer with CT: a preliminary cost-effectiveness analysis [abstract 163CH-p]. Radiology. 2002;225P:148-149.
28. Marshall D, Simpson KN, Earle CC, Chu CW. Economic decision analysis model of screening for lung cancer. Eur J Cancer. 2001;37:1759-1767.
29. Mahadevia PJ, Fleisher LA, Frick KD, Eng J, Goodman SN, Powe NR. Lung cancer screening with helical computed tomography in older adult smokers: a decision and cost-effectiveness analysis. JAMA. 2003;289:313-322.
30. Salerno M, de Lange EE, Altes TA, Truwit JD, Brookeman JR, Mugler JP III. Emphysema: hyperpolarized helium 3 diffusion MR imaging of the lungs compared with spirometric indexes: initial experience [see comment]. Radiology. 2002; 222:252-260.
31. Kauczor HU, Hanke A, van Beek EJ. Assessment of lung ventilation by MR imaging: current status and future perspectives. Eur Radiol. 2002;12:1962-1970.
32. Hatabu H, Uematsu H, Nguyen B, Miller WT Jr, Hasegawa I, Gefter WB. CT and MR in pulmonary embolism: a changing role for nuclear medicine in diagnostic strategy. Semin Nucl Med. 2002; 32:183-192.
33. Oudkerk M, van Beek EJ, Wielopolski P, et al. Comparison of contrast-enhanced magnetic resonance angiography and conventional pulmonary angiography for the diagnosis of pulmonary embolism: a prospective study. Lancet. 2002;359: 1643-1647.
34. Vrachliotis TG, Bis KG, Shetty AN, Ravikrishan KP. Contrast-enhanced three-dimensional MR angiography of the pulmonary vascular tree. Int J Cardiovasc Imaging. 2002;18:283-293.
35. Beckh S, Bolcskei PL, Lessnau KD. Real-time chest ultrasonography: a comprehensive review for the pulmonologist. Chest. 2002;122:1759-1773.
36. Krejci CS, Trent EJ, Dubinsky T. Thoracic sonography. Respir Care. 2001; 46:932-939.
37. Lechleitner P, Riedl B, Raneburger W, Gamper G, Theurl A, Lederer A. Chest sonography in the diagnosis of pulmonary embolism: a comparison with MRI angiography and ventilation perfusion scintigraphy. Ultraschall Med. 2002;23:373-378.