| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 British Columbia Cancer Agency, Vancouver, British Columbia, Canada; 2 Department of Medicine, Respiratory Division, University of British Columbia, Vancouver, British Columbia, Canada; and 3 Department of Radiology, Vancouver General Hospital, Vancouver, British Columbia, Canada
Correspondence and requests for reprints should be addressed to Annette McWilliams, M.D., Clinical Assistant Professor, University of British Columbia, British Columbia Cancer Agency, 675 West 10th Avenue, Vancouver, BC, V5Z 1l3 Canada. E-mail: amcwilli{at}bccancer.bc.ca
ABSTRACT
Our purpose was to review the reported behavior and malignant risk of small noncalcified pulmonary nodules detected by computed tomography (CT). A review of published clinical guidelines and studies using CT scan for lung cancer screening was performed. Small pulmonary nodules are found in 5 to 60% of patients in published CT screening studies. The detection rate is influenced by the CT scan technique used, definition of a significant nodule, and the population of subjects screened. There is limited published systematic longitudinal observation of all nodules of any size. The malignancy rate of small nodules detected in smokers is likely less than 1 to 2%, and predictors of malignancy include semisolid appearance, diameter greater than or equal to 10 mm or persistent growth on greater than or equal to two CT scans. There is a wide variation in the performance of positron emission tomography (PET) scan in screening detected lung cancers. In summary, multidetector row CT detects greater than or equal to 1 nodule in most high-risk patients. The risk of malignancy for a single nodule appears to be low, but is increased by serial growth, diameter greater than or equal to 10 mm, and semisolid appearance. The role of PET in evaluating these nodules needs further exploration. Serial follow-up for 24 months in a high-risk cohort appears reasonable based on present data, but further longitudinal information is required.
Key Words: lung cancer • screening • computed tomography • pulmonary nodules
The rapid technological advances in computed tomography (CT) scanners, especially the development of multi-track detectors and sub second gantry rotation time, has greatly increased the speed of image acquisition. This has allowed a substantial decrease in the thickness or width of CT image slices. For example, since a single-breath-hold chest scan must be acquired in less than 15 seconds, a single detector row CT scanner is limited to 7- to 10-mm-thick slices, while a 16-track scanner can deliver 1-mm slices. Thinner slices reduce the effect of volume averaging, improving spatial resolution and allowing detection of smaller lung nodules. In addition, radiation dose modulation systems have been developed to reduce radiation exposure while maintaining image quality.
However, this improved technology has resulted in the detection of large numbers of small noncalcified pulmonary nodules (< 10mm) on thoracic CT scans. This effect has been previously noted in the literature, especially in studies using CT for lung cancer screening. In reported studies, the frequency of detecting one or more small pulmonary nodules on a screening CT scan ranges from 5 to 60%, with higher rates noted using thinner slices (1–18). On a single examination, the clinical significance of small lung nodules is unknown. The best way to establish the benign etiology of a nodule is to document no growth on follow-up CT scans. However, the optimum follow-up interval and duration for serial CT scans is controversial (19). The medical community has generally recommended 24-month follow-up with no growth as proof of a benign nodule. However, this recommendation is based on plain film studies and is not based on published longitudinal CT data. Guidelines for the follow-up of lung nodules detected using CT have been published by the Fleischner Society and are based on the consensus opinion of a panel of experts (19) (Table 1). These guidelines have been widely adopted in clinical practice.
|
LUNG CANCER SCREENING STUDIES
Over the last decade there has been renewed interest in lung cancer screening due to the widespread introduction of CT scanners capable of imaging the entire chest within a single breath hold. It has been shown that because of increased contrast sensitivity and the cross-sectional imaging perspective, noncontrast chest CT can reliably detect pulmonary nodules between 5 and 8 mm in diameter, approximately one half the size that are reliably detected using the chest radiograph. Multiple nonrandomized, single-arm, observational CT lung cancer screening studies have been reported (Table 2). The populations screened have varied, and although most studies have concentrated on high-risk current or former smokers, some have included significant proportions of lifelong nonsmokers (4–7, 13).
|
The majority of nodules detected were small, being less than 5 mm in maximum long axis diameter (8–12, 14–16, 18). Some CT lung cancer screening studies did not record or systematically follow small nodules or used CT scans providing 10-mm slice width, making small nodules undetectable due to volume-averaging effects (1–3, 17, 18). As a result, the majority of these studies have not reported a lung cancer arising in nodules less than 5 mm in diameter. The BCCA study, using thinner slice width (1.25 and 1 mm) and systematic observation of all detectable nodules, reported that nearly 20% of malignancies were less than 4mm diameter when first seen on CT. This result indicates that size is not a guarantee of benign nature (16).
In addition, in some studies using 7- to 10-mm slice width, the presence of more than six noncalcified nodules in a subject was presumed to indicate benign pulmonary disease, and individual nodules were not cataloged or serially followed (1–3, 8–11, 20). However, the increased rate of nodule detection using thinner slices challenges this assumption. Further data obtained in longitudinal lung cancer screening studies will be necessary to determine if the presence of multiple nodules confers a lower risk of malignancy.
Therefore, the measured malignancy rate per nodule depends on the definition of a significant nodule, slice thickness provided by the CT scan protocol and the successful follow-up of nodules to detect growth. If all detectable nodules are counted, including those less than 5 mm in diameter, the overall malignancy rate appears to be less than 1 to 2%, with higher rates associated with larger diameter lesions (14–16, 8–13) (Tables 2–4
).
|
|
). A substantial number of nodules are also seen in contact with the major and minor pleural fissures. These nodules, referred to a perifissural opacities, are seen as solid, polygonal nodules in a subpleural location (Figure 4). The overwhelming majority of these nodules are benign (22). These lesions likely represent subpleural lymph nodes.
|
|
|
|
According to the current published evidence, most small pulmonary nodules, even those found in current or previous heavy smokers, are benign, with an overall malignancy rate of less than 1 to 2%. In one previous report of systematic longitudinal observation of 2,408 screening-detected pulmonary nodules, 95% of lesions remained stable, resolved, or were seen to be benign, and 4% showed some suspicious behavior that required additional observation or investigation over 24 months of follow-up (16).
Lesion characteristics on serial CT scans that are worrisome for malignancy include persistent growth on two sequential CT scans, increase in the density or solid component of a nonsolid or semisolid lesion, lesions greater than or equal to 10 mm diameter, and semisolid nodules (11, 14–16, 20, 21, 23) (Figures 5A, 5B, 6A, and 6B).
|
|
Positron emission tomography (PET) has an established role in the clinical diagnosis and staging of lung cancer. However, its performance is limited by the size, appearance and histopathology of the suspicious nodule (24–28). There is little data published regarding its role in the evaluation of early lung cancers detected by CT screening for lung cancer or those incidentally found in research studies. A few published reports regarding the diagnostic performance in screening detected lung cancers varies from a sensitivity of 43 to 90% and specificity of 57 to 91% (17, 29–31). These studies suggest that a negative PET examination in an enlarging pulmonary nodule on serial CT is not reassuring for benign disease (31).
CURRENT MANAGEMENT PROTOCOLS
In lung cancer screening subjects, current follow-up protocols recommend 24-month observation of detected nodules. In some protocols, nodules less than 5 mm are not thought to be significant, but in others all nodules are observed and documented (1–3, 14–16, 20, 32). However, even small nodules less than 5 mm in size undergo an annual repeat screen according to one study's published protocol (3, 20). The interval of the repeat CT scan is generally dictated by the size and appearance of the largest pulmonary nodule detected. For lesions less than 5 mm in diameter, a follow-up CT at 12 and 24 months appears reasonable (3, 16, 20, 32). For larger lesions (> 5 mm), a repeat CT at 3 to 6 months and then at 12 and 24 months from initial detection is generally used in screening studies. Suspicious lesions are investigated more immediately using either biopsy, PET/CT, and/or resection. Any new lesion that appears during follow-up or a lesion that shows growth requires a repeat CT scan in 1 to 3 months depending on its size and appearance.
The Fleischner Society guidelines were developed to guide follow-up CT scans in non–lung cancer screening patients with incidentally detected nodules. They are not based on published longitudinal studies of CT-detected nodules, but rather are based on previous chest X-ray follow-up of pulmonary lesions. These guidelines recommend no follow-up for nodules less than or equal to 4mm in low-risk patients and only 12 months follow-up for lesions less than or equal to 4mm in high-risk patients (19) (Table 1). High-risk patients are defined as those with a history of smoking or other risk factors. The recommendation of 12 months follow up of less than or equal to 4 mm nodules in high-risk patients may be too short, and longitudinal CT data is required to further assess this recommendation.
CONCLUSIONS
The detection of small noncalcified pulmonary nodules is now a frequent coincidental finding on thoracic CT scans performed in a clinical or research setting. The significance and best follow-up protocol of these lesions remains uncertain, as there is limited CT longitudinal data published in the literature. In the absence of evidence, we believe that follow-up duration of 24 months in high-risk patients is reasonable with current data.
The overall risk of malignancy in a detected nodule is small, but only proof of stability over time is completely reassuring. Negative findings at PET scan cannot be used to confirm the benign nature of small growing nodules. Worrisome nodule characteristics include growth on two serial follow-up examinations, no change in size but an increase in density or development of a solid component in a nonsolid or semisolid nodule, semisolid appearance, and diameter larger than 10 mm. Further research is required to fully characterize the natural history of these small lung nodules.
FOOTNOTES
Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
(Received in original form September 25, 2008; accepted in final form October 11, 2008)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
