Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Diederich, S.
Right arrow Articles by Peters, P. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Diederich, S.
Right arrow Articles by Peters, P. E.
(Radiology. 1999;213:289-298.)
© RSNA, 1999

Thoracic Imaging

Pulmonary Nodules: Experimental and Clinical Studies at Low-Dose CT1

Stefan Diederich, MD, Horst Lenzen, BSc, Ralf Windmann, MD, Zoltan Puskas, MD, Talat M. Yelbuz, MD, Stefan Henneken, MD, Thomas Klaiber, MD, Mitra Eameri, Nikolaus Roos, MD and Peter E. Peters, MD 2

1 From the Institute of Clinical Radiology (S.D., H.L., R.W., Z.P., S.H., T.K., M.E., N.R., P.E.P.) and the Gerhard Do- magk Institute of Pathology (T.M.Y.), University of Münster, Albert-Schweitzer-Str 33, D-48129 Münster, Germany. Received September 30, 1998; revision requested November 17; revision received January 14, 1999; accepted April 15. Address reprint requests to S.D. (e-mail: diestef@uni-muenster.de).


   Abstract
TOP
 Abstract
Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
PURPOSE: To compare the number of pulmonary nodules detected at helical low- and standard-dose computed tomography (CT) and to investigate the diagnostic value of low-dose CT with a radiation exposure equivalent to that used at chest radiography.

MATERIALS AND METHODS: Two radiologists recorded pulmonary nodules at standard-dose (250 or 100 mA, pitch of 1; 200 mA, pitch of 2) or low-dose CT (50 or 25 mA, pitch of 1 or 2) in five postmortem specimens and 75 patients. Nodules were assessed by size (5 mm or smaller, 6–10 mm, or larger than 10 mm) and by diagnostic confidence ("definite nodule," "definite lesion, not classic nodule," or "questionable lesion, possibly representing a vessel") with the Wilcoxon signed rank test. Artifacts depicted at low-dose CT were recorded.

RESULTS: There were no statistically significant differences in the number of nodules detected at standard- or low-dose CT except in nodules 5 mm or smaller that were assessed as definite nodules at standard- or low-dose CT (25 mA, pitch of 2) (472 vs 397, P < .05). Artifacts that possibly interfered with nodule detection were observed exclusively at CT with 25 mA and a pitch of 2.

CONCLUSION: Pulmonary nodules were detected reliably at CT with 50 mA and pitch of 2 or with 25 mA and a pitch of 1. However, further reduction of the dose to that used at chest radiography was associated with a significant decrease in the number of nodules 5 mm or smaller that were detected, possibly due to artifacts.

Index terms: Computed tomography (CT), comparative studies, 60.12115 • Computed tomography (CT), experimental studies, 60.12115 • Computed tomography (CT), radiation exposure, 60.12115, 60.47 • Lung, nodule, 60.281 • Lung neoplasms, CT, 60.12115, 60.281 • Lung neoplasms, diagnosis, 60.281


   Introduction
TOP
 Abstract
 Introduction
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Findings from comparative studies of chest radiography and computed tomography (CT) have shown CT to be the best imaging modality for the detection of pulmonary nodules in vivo (1). Its role has continuously expanded, particularly since the introduction of spiral technology (2–5).

Compared with other radiologic modalities, CT is associated with a relatively high level of radiation. Therefore, CT is regarded as the main source of medical radiation exposure in industrialized countries (6–10). In the literature, the effective dose equivalent from a single CT examination of the chest is reported to range from 2 to 25 mSv, depending on the CT scanner and the examination protocol used (6,7,9–14). These doses markedly exceed effective dose equivalents received at chest radiography (<=0.3 mSv) by factors of 10–100 (11,13,14). In particular, radiation exposure causes concern in patients, especially children, with benign disease (15,16).

Findings in recent publications suggest that substantial dose reduction is possible at chest CT because of its high inherent contrast and because of the low radiation absorption of the lung (16–21).

The aim of this study was to assess the accuracy of low-dose spiral CT for the detection of pulmonary nodules compared with that of CT protocols with established examination parameters. We were particularly interested in the diagnostic value of low-dose CT with a radiation exposure equivalent to that used at chest radiography.


   MATERIALS AND METHODS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Postmortem Lung Specimens
During a study period of 6 months (April 1995 to October 1995), the lungs were removed at autopsy from five patients who had died of malignant tumors (colonic, breast, thyroid, or testicular carcinoma or non-Hodgkin lymphoma; one each) and who had known pulmonary nodules. The lungs were fixed at end-inspiratory volume by using the method described by Markarian and Dailey (22), which was modified and used for correlation of findings at CT and histopathologic examination (23). The specimens were sealed in waterproof polyethylene bags (Quickpack, Renningen, Germany) and were placed in a Plexiglas phantom filled with dilute contrast medium (15 mg of iodine per milliliter) to simulate the radiation absorption of an adult chest (95% at 120 kV) (24).

Patients
The study population consisted of 75 consecutive patients (36 women, 39 men; age range, 19–78 years; mean age, 56 years) with proved hematogenous metastases from malignant tumors (carcinoma, n = 64; sarcoma, n = 10; melanoma, n = 1) and with known pulmonary nodules. Patients were referred for clinically indicated chest CT during a study period of 15 months (July 1995 to September 1996). Written informed consent was obtained to perform an additional low-dose CT examination of the chest. Patients not well enough to tolerate the additional examination time were excluded from the study. The study design was approved by the local ethics committee and the governmental radiation protection board.

Protocols
All postmortem and patient examinations were performed with the same helical CT scanner (Tomoscan SR 7000; Philips, Eindhoven, the Netherlands). Lung specimens and patients were examined at CT with established protocols (standard-dose CT) and with reduced radiation exposure (low-dose CT) in one session (Figs 1, 2).



View larger version (110K):
[in this window]
[in a new window]
  		Figure 1a. (a-c) Axial images obtained in the identical section position depict a lung with a 4-mm peripheral nodule and perifocal infiltration (arrow) at (a) standard-dose CT (250 mA, 120 kV, 10-mm collimation, pitch of 1), (b) low-dose CT (50 mA, 120 kV, 5-mm collimation, pitch of 2), and (c) pathologic analysis. (d) Photomicrograph of a histologic specimen from the peripheral zone of the lesion shows a granuloma (G) with a fibrous capsule and adjacent infiltration from malignant non-Hodgkin lymphoma (L). (Hematoxylin-eosin stain; original magnification, x2.5.)

 


View larger version (126K):
[in this window]
[in a new window]
  		Figure 1b. (a-c) Axial images obtained in the identical section position depict a lung with a 4-mm peripheral nodule and perifocal infiltration (arrow) at (a) standard-dose CT (250 mA, 120 kV, 10-mm collimation, pitch of 1), (b) low-dose CT (50 mA, 120 kV, 5-mm collimation, pitch of 2), and (c) pathologic analysis. (d) Photomicrograph of a histologic specimen from the peripheral zone of the lesion shows a granuloma (G) with a fibrous capsule and adjacent infiltration from malignant non-Hodgkin lymphoma (L). (Hematoxylin-eosin stain; original magnification, x2.5.)

 


View larger version (124K):
[in this window]
[in a new window]
  		Figure 1c. (a-c) Axial images obtained in the identical section position depict a lung with a 4-mm peripheral nodule and perifocal infiltration (arrow) at (a) standard-dose CT (250 mA, 120 kV, 10-mm collimation, pitch of 1), (b) low-dose CT (50 mA, 120 kV, 5-mm collimation, pitch of 2), and (c) pathologic analysis. (d) Photomicrograph of a histologic specimen from the peripheral zone of the lesion shows a granuloma (G) with a fibrous capsule and adjacent infiltration from malignant non-Hodgkin lymphoma (L). (Hematoxylin-eosin stain; original magnification, x2.5.)

 


View larger version (150K):
[in this window]
[in a new window]
  		Figure 1d. (a-c) Axial images obtained in the identical section position depict a lung with a 4-mm peripheral nodule and perifocal infiltration (arrow) at (a) standard-dose CT (250 mA, 120 kV, 10-mm collimation, pitch of 1), (b) low-dose CT (50 mA, 120 kV, 5-mm collimation, pitch of 2), and (c) pathologic analysis. (d) Photomicrograph of a histologic specimen from the peripheral zone of the lesion shows a granuloma (G) with a fibrous capsule and adjacent infiltration from malignant non-Hodgkin lymphoma (L). (Hematoxylin-eosin stain; original magnification, x2.5.)

 


View larger version (126K):
[in this window]
[in a new window]
  		Figure 2a. Axial images show multiple pulmonary metastases (arrows) from renal cell carcinoma at (a) standard-dose CT (200 mA, 120 kV, 5-mm collimation, pitch of 2) and (b) low-dose CT (25 mA, 120 kV, 5-mm collimation, pitch of 2). Despite the markedly reduced signal-to-noise ratio at low-dose CT, all pulmonary nodules are demonstrated with both techniques.

 


View larger version (129K):
[in this window]
[in a new window]
  		Figure 2b. Axial images show multiple pulmonary metastases (arrows) from renal cell carcinoma at (a) standard-dose CT (200 mA, 120 kV, 5-mm collimation, pitch of 2) and (b) low-dose CT (25 mA, 120 kV, 5-mm collimation, pitch of 2). Despite the markedly reduced signal-to-noise ratio at low-dose CT, all pulmonary nodules are demonstrated with both techniques.

 
Initially, the examination parameters suggested by the manufacturer for chest CT (250 mA; 120 kV; 10-mm collimation; table feed, 10 mm per rotation; 10-mm reconstruction interval) were used for standard-dose CT, and exposure was reduced by decreasing the tube current to 50 mA while keeping the other parameters constant. At that time, 50 mA was the lowest tube current possible. To achieve further dose reduction, the ratio of table feed per rotation to collimation (pitch) was increased. To reduce partial volume effects and to cover the entire chest during one breath hold, we used a collimation of 5 mm and a table feed of 10 mm per rotation (pitch of 2). Thus, all five lung specimens and 39 patients were examined by using a tube current of 50 mA.

For another part of the study, the tube current at standard-dose CT was reduced to 100 mA after an analysis of the findings from the previous studies and from the literature (9,11,12). At this time, the manufacturer provided an experimental 25-mA setting. As with the higher settings for tube current, we initially compared findings at standard-dose CT with findings at low-dose CT with a pitch of 1 (10-mm collimation; table feed, 10 mm per rotation; 10-mm reconstruction interval). For further dose reduction, the pitch was increased again by reducing collimation (5-mm collimation; table feed, 10 mm per rotation; 5-mm reconstruction interval). To keep all of the other parameters between standard- and low-dose CT constant, we also increased the pitch at standard-dose CT, while maintaining a constant signal-to-noise ratio, by increasing the tube current to 200 mA (24). A total of 36 patients were examined by using a tube current of 25 mA.

Thus, three different steps of dose reduction were achieved at low-dose CT, which represented approximately 20%, 10%, and 5% of the dose associated with the original standard-dose CT protocol. Table 1 summarizes the corresponding examination protocols.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Parameters for CT Protocols
 
No magnification or targeted reconstruction was used. Images were photographed at identical window settings (window width, 1,500 HU; window level, -600 HU) on laser film.

Analysis was performed independently and in random order by two board-certified radiologists (S.D., R.W.) who had similar experience with spiral CT in pulmonary nodules. The readers were not aware of the specimen number, the patient's identity, or the precise examination protocol. They could, however, identify standard- and low-dose settings because of differences in image noise. There were 4-week intervals between analysis of corresponding studies from one specimen or patient to prevent recognition of the findings, and corresponding low- and standard-dose CT images were analyzed in alternating fashion to exclude bias from learning effects.

Each pulmonary nodule was recorded in a standardized form: The lesion size was measured in millimeters; particular attention was paid to the correct classification of the nodule into three size categories (<=5 mm, 6–10 mm, >10 mm) by using the caliper on the CT scanner, which was transferred to a strip of cardboard that could be held against the nodule. No electronic measurement was obtained on the monitor. The location of the lesion was noted by recording the section position at which the lesion was best displayed and by marking the location of the lesion at the respective levels within the section in a schematic drawing of the lung.

The degree of diagnostic confidence was also recorded in three categories: An area of attenuation observed by the readers was recorded as a "definite nodule" if it was round or slightly oblong (long axis/short axis < 2) and well defined. It was recorded as a "definite lesion, not classic nodule" if it was more oblong (long axis/short axis > 2) or ill defined but was believed to represent a true lesion. If the reader was uncertain whether an area of attenuation represented a true lesion or a pulmonary vessel that was imaged in cross-section, it was recorded as a "questionable lesion, possibly representing a vessel."

Analysis was performed separately for nodules detected by individual readers and for the total of nodules detected by both readers. No consensus opinion was obtained for the detection of nodules.

Artifacts
During the individual reading session, both readers recorded if the CT image was degraded by artifacts that were believed to potentially interfere with the detection of pulmonary nodules. All examinations in which artifacts were recorded were assessed more precisely during a second reading session by both readers in consensus: The presence of artifacts was recorded for every CT image. To correspond to the classification of nodule size, artifacts were classified by the size of the nodules that they possibly obscured, as follows: 5 mm or smaller, 6–10 mm, or larger than 10 mm. The subject's height and body weight were recorded for correlation with the presence and classification of artifacts.

Histopathologic Analysis
After the CT examination, each lung specimen was cut into slices with a thickness of approximately 10 mm in an orientation identical to that on the CT images (Fig 1c). For correlation of CT and histopathologic findings, a total of 27 nodules (range, three to nine per specimen; mean, 5.4) 10-mm or smaller (range, 2–8 mm; mean, 4.9 mm) that were all recorded as a definite nodule or as a definite lesion, not classic nodule by at least one observer at one dose setting were selected. These nodules were embedded in paraffin, cut on a microtome, and stained with hematoxylin-eosin.

Statistical Analysis
Pulmonary nodules demonstrated in the same lung with different examination protocols represented paired data, which were not normally distributed. Therefore, the Wilcoxon signed rank test was used to assess differences in the detection of nodules at the corresponding CT examinations. P values less than or equal to .05 indicated a statistically significant difference (25).

From the true-positive, false-positive, and false-negative findings, the sensitivity and positive predictive value of low-dose CT were calculated by using findings from the corresponding standard-dose CT as the standard. As small pulmonary nodules are not demonstrated reliably in fixed lung specimens, histopathologic findings were not used as the standard. True-negative findings could not be recorded for individual nodules in lungs with multiple lesions in lung specimens or in patients. Therefore, specificity and negative predictive value could not be calculated (26).

Calculation of Effective Dose Equivalents
Effective dose equivalents for the different standard- and low-dose CT protocols were calculated according to the investigations by Lenzen et al (11), members from our group, who used the same CT scanner that was used in this study. In that study, effective dose equivalents were determined by using a male RANDO phantom (Phantom Laboratory, Salem, NY), thermoluminescence dosimetry, and calculations with conversion factors (27,28). With this CT scanner, a collimation of 5 mm, pitch of 2, and a tube current of 50 mA, effective dose equivalents of 0.6 mSv (men) and 1.1 mSv (women) were found. The differences were due to the fact that the breast contributes substantially to the effective dose equivalent in women (11).


   RESULTS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
The number of nodules detected at standard- and low-dose CT are presented in Tables 2–6.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Nodules Detected in Five Lung Specimens
 

View this table:
[in this window]
[in a new window]
  		TABLE 3. Patient Study 1: Nodules Detected in Seven Patients
 

View this table:
[in this window]
[in a new window]
  		TABLE 4. Patient Study 2: Nodules Detected in 32 Patients
 

View this table:
[in this window]
[in a new window]
  		TABLE 5. Patient Study 3: Nodules Detected in 14 Patients
 

View this table:
[in this window]
[in a new window]
  		TABLE 6. Patient Study 4: Nodules Detected in 22 Patients
 
Examination of Postmortem Specimens
A total of 140 lesions were detected by at least one observer at one dose setting. The number of nodules in one specimen ranged from five to 140 (mean, 57) (Table 2), and lesion size ranged from 2 to 30 mm (mean, 12 mm). The total number of nodules, obtained by adding findings from both observers, was 160 nodules at standard-dose CT and 139 nodules at low-dose CT. The number of nodules detected by reader 1 at standard- and low-dose CT were almost identical (95 and 98, respectively), whereas reader 2 diagnosed more nodules at standard-dose CT than at low-dose CT (65 and 41, respectively) (Table 7).


View this table:
[in this window]
[in a new window]
  		TABLE 7. Nodules Diagnosed by Both Observers
 
The difference was not statistically significant for the total observations of both readers, for individual readers, or for all nodules or individual size categories (P > .05). The marked variation between the findings at low- and standard-dose CT in individual lung specimens was the reason why differences were not statistically significant, although observer 2 diagnosed a much higher number of nodules at standard-dose CT than at low-dose CT.

There was also no significant difference in the proportion of the three categories of diagnostic confidence. For both readers, the proportion of definite nodules increased with lesion size and the number of questionable lesions that possibly represented vessels decreased with lesion size (Table 8). The sensitivity and positive predictive value of low-dose CT, with standard-dose CT as the standard, are presented in Table 9.


View this table:
[in this window]
[in a new window]
  		TABLE 8. Lesions in Different Categories of Diagnostic Confidence
 

View this table:
[in this window]
[in a new window]
  		TABLE 9. Sensitivity and Positive Predictive Value of Low-Dose CT Protocols
 
Histopathologic Analysis
At histologic examination, 26 of the 27 nodules selected for pathologic analysis represented true lesions, as follows: 12 metastases (eight thyroid, three testicular, one colonic carcinoma [Fig 3b]), two nodules from non-Hodgkin lymphoma, 12 focal lesions due to bronchopneumonia, and one tuberculous granuloma (Fig 1d). One lesion recorded at CT corresponded to a pulmonary arteriole that was imaged in cross-section (positive predictive value, 96%).



View larger version (134K):
[in this window]
[in a new window]
  		Figure 3a. Images show correlation of axial CT and histologic findings in a lung with pulmonary metastasis from colonic carcinoma. (a) Low-dose CT image (50 mA, 120 kV, 5-mm collimation, pitch of 2) in a lung shows diffuse infiltration (histologic finding was bronchopneumonia) and a 3-mm, well-defined nodule (arrow). (b) Photomicrograph shows the nodule, which represents a metastasis from colonic carcinoma. (Hematoxylin-eosin stain; original magnification, x2.5.)

 


View larger version (140K):
[in this window]
[in a new window]
  		Figure 3b. Images show correlation of axial CT and histologic findings in a lung with pulmonary metastasis from colonic carcinoma. (a) Low-dose CT image (50 mA, 120 kV, 5-mm collimation, pitch of 2) in a lung shows diffuse infiltration (histologic finding was bronchopneumonia) and a 3-mm, well-defined nodule (arrow). (b) Photomicrograph shows the nodule, which represents a metastasis from colonic carcinoma. (Hematoxylin-eosin stain; original magnification, x2.5.)

 
Patient Studies
In the 75 patients, a total of 1,446 nodules were detected by at least one observer at one dose setting. The number of nodules detected per patient ranged from one to 141 (mean, 19). Nodule size ranged from 2 to 61 mm (mean, 6.5 mm).

By using low-dose CT protocols 1, 2, and 3 (with doses that were 20% or 10% of the initial standard dose), a total of 865 nodules were detected by the two readers, whereas 793 nodules were detected by the readers on images obtained at the corresponding standard-dose CT protocols (Tables 3–5). The number of nodules detected at all three of the low-dose CT protocols was larger than the number detected at the corresponding standard-dose CT protocols, almost exclusively because of the findings of observer 1. Observer 2, on the contrary, diagnosed more nodules at standard-dose CT than at low-dose CT with protocols 1–3. The difference in the number of nodules detected at standard-dose CT and at low-dose CT was 10% or less and was not statistically significant (P > .05) (Table 7). Also, there was no significant difference in the diagnostic confidence between findings at low-dose CT and at standard-dose CT. As with the postmortem studies, the percentage of definite nodules increased with lesion size (Table 8).

The sensitivity of low-dose CT for nodules 5 mm or smaller, with standard-dose CT as the standard, ranged from 36% to 67% for both readers, with marked variation for individual readers (18%–88%). For 6–10-mm nodules, sensitivity ranged from 72% to 89% for both readers, with less variation between individual readers (60%–90%). For lesions larger than 10 mm, sensitivity of low-dose CT ranged from 91% to 100%, with little variation between individual readers (86%–100%). Findings were similar for positive predictive values (Table 9).

However, with the lowest-dose CT protocol (low-dose CT protocol 4, 5% of the dose of the standard protocol), a smaller total number of nodules were diagnosed (983 vs 1,077 at standard-dose CT, P > .05), and there were significantly fewer lesions diagnosed as definite nodules (779 vs 885, P < .05) (Table 6). This was exclusively due to the smaller number of definite nodules that were 5 mm or smaller (397 vs 472, P < .05) (Table 7). When standard-dose CT with 200 mA was used as the standard, the sensitivity of low-dose CT with 25 mA and a pitch of 2 was 78% for all nodules, 67% for nodules 5 mm or smaller, 89% for 6–10-mm nodules, and 100% for nodules larger than 10 mm. Positive predictive values were 86%, 78%, 93%, and 97%, respectively.

Artifacts
No artifacts that interfered with the detection of pulmonary nodules were recorded at low-dose CT in the postmortem studies or patient studies with dose reductions between 20% and 10% (low-dose CT protocols 1–3). The images were, however, more grainy than those obtained at standard-dose CT. Artifacts were recorded only at low-dose CT when the dose was reduced to 5% (low-dose CT protocol 4). With this protocol, no artifacts were recorded in three of the 22 patients examined (body height <= 179 cm; body weight <= 80 kg) (Fig 4a).



View larger version (175K):
[in this window]
[in a new window]
  		Figure 4a. Low-dose axial CT images (25 mA, 120 kV, 5-mm collimation, pitch of 2) obtained 1, 2, 3, or 4 cm below the lung apex show (a) no notable artifacts in a patient with a body weight of 75 kg and (b) increased noise and streak artifacts (arrows), which were recorded as "potentially obscuring nodules 6-10 mm," in a patient with a body weight of 107 kg.

 


View larger version (184K):
[in this window]
[in a new window]
  		Figure 4b. Low-dose axial CT images (25 mA, 120 kV, 5-mm collimation, pitch of 2) obtained 1, 2, 3, or 4 cm below the lung apex show (a) no notable artifacts in a patient with a body weight of 75 kg and (b) increased noise and streak artifacts (arrows), which were recorded as "potentially obscuring nodules 6-10 mm," in a patient with a body weight of 107 kg.

 
In the other 19 patients (height range, 156–187 cm; mean height, 175 cm; body weight range, 55–107 kg; mean body weight, 80 kg), linear streak artifacts were recorded; these artifacts arose from the bony structures of the shoulder girdle and, to a lesser extent, the spine. In no case were artifacts diagnosed that were believed to potentially obscure nodules larger than 10 mm. In three of the 19 patients, only artifacts that potentially obscured nodules 5 mm or smaller were diagnosed.

In 16 patients, artifacts that possibly interfered with the detection of 6–10-mm nodules were recorded; these artifacts involved the uppermost one to 11 (mean, four) images in the lung apex. In only two of the 16 patients did these artifacts involve more than six images; these patients' body weights were 95 kg (11 images) and 107 kg (eight images) (Fig 4b). In three patients, artifacts were present below the level of the aortic arch; these were recorded as artifacts that only potentially obscured nodules that were 5 mm or smaller. In no patient were artifacts seen below the inferior margin of the scapula.

We looked for statistically significant differences in the detection of nodules on standard- and low-dose CT images that were degraded by artifacts by comparing the number of lesions detected above the level of the aortic arch. No significant difference was found between standard- and low-dose CT in the total number of nodules observed (162 vs 159), nodules 5 mm or smaller (110 vs 103), and definite nodules 5 mm or smaller (72 vs 60).

Effective Dose Equivalents at Standard- and Low-Dose CT
By applying the data reported by Lenzen et al (11) to the different CT protocols used in this study, the effective dose equivalents were found to range from 8.0 (men) and 13.5 (women) (250 mA; 120 kV; 10-mm collimation; table feed, 10-mm per rotation, 10-mm reconstruction interval) to 0.3 (men) and 0.55 (women) (25 mA, 120 kV; 5-mm collimation; table feed, 10-mm per rotation, 5-mm reconstruction interval) (Table 10).


View this table:
[in this window]
[in a new window]
  		TABLE 10. Effective Dose Equivalents
 
In their investigation, Lenzen et al (11) found effective dose equivalents of 0.23 and 0.25 for posteroanterior and lateral chest radiographs, respectively, by using 125 kV, a film focus distance of 200 cm, a 250-speed screen-film system (Thoramat; Siemens, Erlangen, Germany), and the same RANDO phantom.


   DISCUSSION
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
Findings in recent publications suggest that substantial dose reduction is possible at chest CT because of the high inherent contrast and the low radiation absorption of the lung (16–21). In an article by Lee et al (17) in patients with chronic infiltrative lung disease, the diagnostic accuracy of a limited number of low-dose thin-section CT examinations was superior to that of chest radiography with similar radiation exposure. Findings of a preliminary study (29) of low-dose spiral CT for the detection of focal lung disease suggest that the diagnostic accuracy is sufficient in different clinical settings. In these studies, different settings for tube current were used, and effective dose equivalents that would allow comparison of radiation doses were not reported.

In our study, dose reduction to 10%–25% of the corresponding dose at standard-dose CT was shown to cause no significant difference in the detection of pulmonary nodules, when compared with detection at standard-dose settings. In fact, when we compared findings at low-dose CT with 5-mm collimation with findings at standard-dose CT with 10-mm collimation (Table 3), we found that more lesions were detected at low-dose CT, possibly because of decreased partial volume effects (30). Also, the readers' diagnostic confidence did not change significantly with these examination protocols. The effective dose equivalent of these settings was in the range of 0.6–2.7 mSv, which corresponded to the effective dose equivalent for the acquisition of two to 10 chest radiographs (two views). Our findings and those of other investigators (29) suggest that these settings are sufficient for the detection of pulmonary nodules in most circumstances.

Although there was no superior standard for assessing the sensitivity and positive predictive value with standard-dose CT and although histopathologic examination of the lung specimens did not allow us to assess the false-negative CT findings, histologic findings in nodules detected at CT suggested that there are few false-positive results at both low- and standard-dose CT. The large number of focal lesions due to bronchopneumonia may explain the high proportion of lesions that were classified as definite lesions not classic nodules in the postmortem studies. The marked variation between the two readers and the two dose protocols is most likely due to the large number of lesions in individual lungs.

Further dose reduction was achieved in our study with a tube current of 25 mA and a pitch of 2, which in our CT scanner resulted in an effective dose equivalent of 0.3 mSv in the men and 0.55 mSv in the women; this was comparable to the dose associated with the acquisition of one to two chest radiographs in two views. When we compared this low-dose CT protocol with the standard-dose protocol, no significant differences were found in the detection of nodules larger than 5 mm. There were, however, significantly fewer nodules detected that measured 5 mm or less. Also, artifacts were observed with this protocol; they degraded some images in the lung apex, particularly in obese patients. The use of this protocol can, therefore, be recommended only if it is acceptable that nodules 5 mm or smaller may be missed, particularly those in the extreme lung apex.

In our study, the sensitivity and positive predictive value of low-dose CT for nodules 10 mm or smaller was better with a 5-mm collimation and pitch of 2 (effective section thickness, 6.5 mm) than with a 10-mm collimation and pitch of 1 (effective section thickness, 10 mm). This was most likely due to a reduction of partial volume effects and slightly overlapping image reconstruction; this was particularly true for the low-dose CT protocol 2, which was compared with a standard-dose CT protocol with 10-mm collimation and no overlapping reconstruction. Certainly, the increased number of images resulting from a decreased collimation and reconstruction interval did not lead to a decrease in the sensitivity or positive predictive value.

A limitation of our study is the fact that the results cannot be automatically transferred to scanners from different manufacturers or to different scanners of the same model, as identical settings for tube current, tube voltage, collimation, and table feed do not necessarily result in identical image quality or effective dose equivalents. For example, with modern scanners that have solid-state detectors instead of gas detectors, further dose reduction can be expected, with no change in the signal-to-noise ratio (31). Therefore, image quality and effective dose equivalents should be determined individually for specific scanners. Ideally, the image quality, as determined by the signal-to-noise ratio, that is required to solve a specific clinical problem should be defined, and the examination protocol should be selected accordingly (32).

Obviously, in our study, low-dose CT of the chest was not studied in children. However, as radiation absorption decreases with the third order of object diameter, the exposure can be reduced to approximately 50% with every 4-cm decrease of the patient's mediolateral diameter, while maintaining an identical signal-to-noise ratio (31). For this reason, we apply low-dose settings of 25 mA with a pitch of 2 in all infants and children who undergo chest CT for pulmonary nodules.

In our study, there was no significant difference in the number of pulmonary nodules detected at chest CT with standard examination parameters and the number detected at low-dose CT with 10%–25% of radiation exposure of the corresponding standard protocol. Only further reduction of doses to the exposure levels used at chest radiography was associated with significantly decreased sensitivity for nodules 5 mm or smaller; there were no significant differences for nodules larger than 5 mm.

The low-dose CT settings used in our study certainly provided images of sufficient quality to exclude or confirm pulmonary nodules in the presence of an equivocal chest radiograph or at screening for benign (eg, pulmonary arteriovenous malformations [33]) or malignant (eg, bronchogenic carcinoma in smokers [34,35]) disease.


   Footnotes
 
{dagger} Deceased Back

Author contributions: Guarantor of integrity of entire study, S.D.; study concepts, S.D., P.E.P.; study design, S.D., H.L.; definition of intellectual content, S.D., H.L., N.R.; literature research, S.D., S.H., T.K.; clinical studies, S.D., R.W., S.H.; experimental studies, S.D., H.L., M.E., Z.P., T.M.Y.; data acquisition and analysis, S.D., T.K., S.H.; statistical analysis, S.D., T.K., S.H.; manuscript preparation and editing, S.D.; manuscript review, S.D., H.L., N.R.


   References
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Davis SD. CT: evaluation for pulmonary metastases in patients with extrathoracic malignancy. Radiology 1991; 180:1-12.[Abstract/Free Full Text]
  2. Collie DA, Wright AR, Williams JR, Hashemi-Malayeri B, Stevenson AJ, Turnbull CM. Comparison of spiral-acquisition computed tomography and conventional computed tomography in the assessment of pulmonary metastatic disease. Br J Radiol 1994; 67:436-444.[Abstract]
  3. Costello P, Anderson W, Blume D. Pulmonary nodule: evaluation with spiral volumetric CT. Radiology 1991; 179:875-876.[Abstract/Free Full Text]
  4. Friese SA, Rieber A, Fleiter T, Brambs HJ, Claussen CD. Pulmonary nodules in spiral volumetric and single slice computed tomography. Eur J Radiol 1994; 18:48-51.[Medline]
  5. Remy-Jardin M, Remy J, Giraud F, Marquette CH. Pulmonary nodules: detection with thick-section spiral CT versus conventional CT. Radiology 1993; 187:513-520.[Abstract/Free Full Text]
  6. Geleijns J, Van Unnik JG, Zoetelief J, Zweers D, Broerse JJ. Comparison of two methods for assessing patient dose from computed tomography. Br J Radiol 1994; 67:360-365.[Abstract]
  7. Poletti JL. Patient doses from CT in New Zealand and a simple method for estimating effective dose. Br J Radiol 1996; 69:432-436.[Abstract]
  8. Shrimpton PC, Wall BF. The increasing importance of x-ray computed tomography as a source of medical exposure. Radiat Prot Dosim 1995; 57:413-415.[Abstract]
  9. Van Unnik JG, Broerse JJ, Geleijns J, Jansen JT, Zoetelief J, Zweers D. Survey of CT techniques and absorbed dose in various Dutch hospitals. Br J Radiol 1997; 70:376-371.
  10. Wade JP, Weyman JC, Goldstone KE. CT standard protocols are of limited value in assessing actual patient dose. Br J Radiol 1997; 70:1146-1151.[Abstract]
  11. Lenzen H, Roos N, Diederich S, Meier N. Dosimetry for low dose computed tomography of the thorax. Radiologe 1996; 36:483-488[German].[Medline]
  12. Nishizawa K, Maruyama T, Takayama M, Okada M, Hachiya J, Furuya Y. Determinations of organ doses and effective dose equivalents from computed tomographic examination. Br J Radiol 1991; 64:20-28.[Abstract]
  13. Kaul A, Bauer B, Bernhardt J, Nosske D, Veit R. Effective doses to members of the public from the diagnostic application of ionizing radiation in Germany. Eur Radiol 1997; 7:1127-1132.[Medline]
  14. Wall BF, Hart D. Revised radiation doses for typical x-ray examinations: report on a recent review of doses to patients from medical x-ray examinations in the UK by NRPB—National Radiological Protection Board. Br J Radiol 1997; 70:437-439.[Medline]
  15. . 1990 Recommendations of the International Commission on Radiological Protection Oxford, England: Pergamon, 1992.
  16. Ambrosino MM, Genieser NB, Roche KJ, Kaul A, Lawrence RM. Feasibility of high-resolution, low-dose chest CT in evaluating the pediatric chest. Pediatr Radiol 1994; 24:6-10.[Medline]
  17. Lee KS, Primack SL, Staples CA, Mayo JR, Aldrich JE, Müller NL. Chronic infiltrative lung disease: comparison of diagnostic accuracies of radiography and low- and conventional-dose thin-section CT. Radiology 1994; 191:669-673.[Abstract/Free Full Text]
  18. Mayo JR, Hartman TE, Lee KS, Primack SL, Vedal S, Müller NL. CT of the chest: minimal tube current required for good image quality with the least radiation dose. AJR 1995; 164:603-607.[Abstract/Free Full Text]
  19. Mayo JR, Whittall KP, Leung AN, et al. Simulated dose reduction in conventional chest CT: validation study. Radiology 1997; 202:453-457.[Abstract/Free Full Text]
  20. Naidich DP, Marshall CH, Gribbin C, Arams RS, McCauley DI. Low-dose CT of the lungs: preliminary observations. Radiology 1990; 175:729-731.[Abstract/Free Full Text]
  21. Zwirewich CV, Mayo JR, Müller NL. Low-dose high-resolution CT of lung parenchyma. Radiology 1991; 180:413-417.[Abstract/Free Full Text]
  22. Markarian B, Dailey ET. Preparation of inflated lung specimens. In: Groskin SA, eds. Heitzman's the lung: radiologic-pathologic correlations. 3rd ed. St Louis, Mo: Mosby, 1993; 4-12.
  23. Roos N, Diederich S, Fahrenkamp A, et al. Postmortem high-resolution computed tomography of the lung: radiologic-morphologic correlation. Radiologe 1993; 33:113-123[German].[Medline]
  24. Drexler G, Panzer W, Widenmann L, Williams G, Zankl M. The calculation of dose from external photon exposure using reference human phantoms and Monte Carlo methods III. Bericht. Munich, Germany: Gesellschaft für Strahlen- und Umweltforschung, 1990.
  25. Goldin J, Sayre JW. Review: A guide to clinical epidemiology for radiologists. II. Statistical analysis. Clin Radiol 1996; 51:317-324.[Medline]
  26. Dwyer AJ. Matchmaking and McNemar in the comparison of diagnostic modalities. Radiology 1991; 178:328-330.[Abstract/Free Full Text]
  27. Zankl M, Panzer V, Drexler G. The calculation of dose from external photon exposure using reference human phantoms and Monte Carlo methods I. The male (ADAM) and female (EVA) adult mathematical phantoms. Munich, Germany: Gesellschaft für Strahlen- und Umweltforschung, 1990.
  28. Zankl M, Panzer V, Drexler G. The calculation of dose from external photon exposure using reference human phantoms and Monte Carlo methods VI. Organ doses from computed tomographic examinations. Munich, Germany: Gesellschaft für Strahlen- und Umweltforschung, 1990.
  29. Gartenschläger M, Schweden F, Gast K, et al. Pulmonary nodules: detection with low-dose vs conventional-dose spiral CT. Eur Radiol 1996; 8:609-614.
  30. Paranjpe DV, Bergin CJ. Spiral CT of the lungs: optimal technique and resolution compared with conventional CT. AJR 1994; 162:561-567.[Abstract/Free Full Text]
  31. Rothenberg LN, Pentlow KS. AAPM tutorial: radiation dose in CT. RadioGraphics 1992; 12:1225-1243.[Abstract]
  32. Jurik AG, Jessen KA, Hansen J. Image quality and dose in computed tomography. Eur Radiol 1997; 7:77-81.[Medline]
  33. Coley SC, Jackson JE. Pulmonary arteriovenous malformations. Clin Radiol 1998; 53:396-404.[Medline]
  34. Kaneko M, Eguchi K, Ohmatsu H, et al. Peripheral lung cancer: screening and detection with low-dose spiral-CT versus radiography. Radiology 1996; 201:798-802.[Abstract/Free Full Text]
  35. Sone S, Takashima S, Li F, et al. Mass screening for lung cancer with mobile spiral computed tomography scanner. Lancet 1998; 351:1242-1245.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Diederich, S.
Right arrow Articles by Peters, P. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Diederich, S.
Right arrow Articles by Peters, P. E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE