Skip to main content
Have a personal or library account? Click to login
Multimodality CT imaging contributes to improving the diagnostic accuracy of solitary pulmonary nodules: a multi-institutional and prospective study Cover

Multimodality CT imaging contributes to improving the diagnostic accuracy of solitary pulmonary nodules: a multi-institutional and prospective study

Open Access
|Feb 2023

Figures & Tables

Figure 1

Flow chart of patient selection.

SPNs = solitary pulmonary nodules

Figure 2

Technology roadmap of multimodality computed tomography (CT) imaging for evaluating solitary pulmonary nodules (SPNs).

BF = blood flow; BV = blood volume; CECT = contrast enhanced CT; CTPI = CT perfusion imaging; DECT = dual-energy CT; IC = iodine concentration; MPR = multiplanar reconstruction; NECT = non-contrast enhanced CT; MTT = mean transit time; NIC = normalized iodine concentration; OL = iodine overlay; PS = permeability surface

Table 1

Patient characteristics in 285 patients with solitary pulmonary nodules

Benign SPNs (n = 118)Malignant SPNs (n = 167)
Gender*
Male56 (47.46%)96 (57.49%)
Female62 (52.54%)71 (42.51%)0.1170
Age (years)#50.84 ± 19.6052.93 ± 20.300.3952
Smoking status*
Yes52 (44.07%)81 (48.50%)
No66 (55.93%)86 (51.50%)0.4721
Tumor history*
Yes17 (14.41%)28 (16.77%)
No101 (85.59%)139 (83.23%)0.6245
Tumor biomarkers*
Normal103 (87.29%)139 (83.23%)0.4026
Abnormal15 (12.71%)28 (16.77%)

[i] SPNs = solitary pulmonary nodules; *compared with the Fisher’s exact test; #compared with the two independent samples t-test

Table 2

Pathological results of the 285 solitary pulmonary nodules (SPNs) included in this study

SPNsPathologyDatum (%)
Benign SPNs (n = 118)
Tuberculosis46 (29.0%)
Acute and chronic inflammation32 (27.1%)
Inflammatory pseudotumor14 (11.9%)
Hamartoma9 (7.6%)
Pulmonary sclerosing hemangioma6 (5.1%)
Sequestration4 (3.4%)
Bronchogenic cyst3 (2.5%)
Rheumatoid arthritis2 (1.7%)
Granulomatosis with polyangiitis2 (1.7%)
Malignant SPNs (n = 167)
Primary pulmonary carcinoma116 (69.5%)
Solitary metastasis23 (13.8%)
Primary lung neuroendocrine tumor21 (12.6%)
Primary pulmonary lymphoma7 (4.2%)
Figure 3

A solitary pulmonary nodule (SPN) with the size of 17.0 × 19.0 mm located in the middle lobe of right lung of a 59 years old male. Non-contrast enhanced CT (NECT), both upper (A) and (B) images, showed that there was sign of smooth margin, but without signs of lobulation, spiculation, vacuole, cavitation, air bronchogram, calcification, fat, pleural indentation, vessel convergence, or adjacent bronchial changes. Contrast enhanced CT (CECT) showed that there were mild and obvious enhancements in the arterial (C) and venous (D) phases, respectively. The patient was scanned with a CT-guided percutaneous lung biopsy procedure, and the pathological result showed chronic inflammatory disease. This SPN disappeared after a week of antibiotic therapy.

Figure 4

A solitary pulmonary nodule (SPN) with the size of 25.0 × 27.0 mm located in the superior lobe of left lung of a 61 years old female evaluated by CT perfusion imaging (CTPI). Blood volume (BV) (A), blood flow (BF) (B), mean transit time (MTT) (C), and permeability surface (PS) (D) for the SPN were 6.16 ml/100 g, 34.16 ml/100 g/min, 10.94 s, and 25.97 ml/100 g/min, respectively. Pathology of the SPN after the surgery confirmed the diagnosis of an adenocarcinoma.

Figure 5

Receiver operating characteristic curve for distinguishing benign from malignant nodules using CT perfusion imaging parameter of permeability surface (PS).

Figure 6

A solitary pulmonary nodule (SPN) with the size of 24.0 × 26.0 mm located in the superior lobe of left lung of a 57 years old female evaluated by arterial (A) and venous phases (B) of dual-energy CT (DECT). Iodine concentration at the arterial phase (aIC), Iodine concentration at the venous phase (vIC), normalized iodine concentration at the arterial phase (aNIC), and normalized iodine concentration at the venous phase (vNIC) were 2.409 mg/mL, 10.23 mg/mL, 0.17 (2.409/14.18), 0.53 (10.23/19.18) respectively. Pathology of the SPN after the surgery confirmed the diagnosis of an adenocarcinoma.

Figure 7

Receiver operating characteristic curves for distinguishing benign from malignant nodules using dual-energy CT parameters.

aOL = Iodine overlay at the arterial phase (AUC = 0.636); vOL = Iodine overlay at the venous phase (AUC = 0.638); aIC = Iodine concentration at the arterial phase (AUC = 0.657); vIC = Iodine concentration at the venous phase (area under the curve [AUC] = 0.703); aNIC = normalized iodine concentration at the arterial phase (AUC = 0.728); vNIC = normalized iodine concentration at the venous phase (AUC = 0.790); all p values < 0.05

Table 3

Solitary pulmonary nodules evaluated with non-contrast enhanced CT

CT findings*Benign SPNs (n = 118)Malignant SPNs (n = 167)P Values
Smooth margin
Yes84 (71.19%)49 (29.34%)< 0.0001
No34 (28.81%)118 (70.66%)
Lobulated sign
Yes29 (24.58%)115 (68.86%)
No89 (75.42%)52 (31.14%)< 0.0001
Spiculated sign
Yes25 (21.19%)108 (78.81%)
No93 (64.67%)59 (35.33%)< 0.0001
Vacuole sign
Yes13 (11.02%)38 (22.75%)0.0120
No105 (88.98%)129 (77.25%)
Cavity sign
Yes9 (7.63%)16 (9.58%)
No109 (92.37%)151 (90.42%)0.6727
Air bronchogram
Yes33 (27.97%)56 (33.53%)0.3643
No85 (72.03%)111 (66.47%)
Calcification
Yes10 (8.47%)6 (3.59%)
No108 (91.53%)161 (96.41%)0.1149
Fat
Yes6 (5.08%)4 (2.40%)0.3277
No112 (94.92%)163 (97.60%)
Pleural indentation
Yes25 (21.19%)89 (53.29%)< 0.0001
No93 (78.81%)78 (46.71%)
Vessel convergence
Yes33 (27.97%)84 (50.30%)0.0002
No85 (72.03%)83 (49.70%)
Adjacent bronchial changes
Yes26 (22.03%)43 (25.75%)0.4867
No92 (77.97%)124 (74.25%)
Location
Upper and middle lobes53 (44.92%)79 (47.31%)0.7186
Inferior lobe65 (55.08%)88 (52.69%)
Size (mm)
15–2029 (24.58%)46 (27.54%)0.5883
20–3089 (75.42%)121 (72.46%)

[i] SPNs = solitary pulmonary nodules; *compared with the Fisher’s exact test

Table 4

Non-enhanced computed tomography (NECT) in evaluating solitary pulmonary nodules (SPNs) with various categories in 285 patients

ItemsCategory ICategory IICategory IIICategory IVCategory V
Benign SPNs (n = 118)39948220
Malignant SPNs (n = 167)05541989

[i] Category I = most likely benign; Category II = possibly benign; Category III = uncertain benign or malignant; Category IV = possibly malignant; Category IV = most likely malignant

Table 5

Solitary pulmonary nodules evaluated with CT perfusion imaging

Parameters*Benign SPNs (n = 118)Malignant SPNs (n = 167)P values
BF (ml/100 g/min)49.34 (27.78, 72.81)58.44 (24.91, 80.47)0.1022
BV (ml/100 g)4.79 (2.87, 7.66)4.84 (2.90, 7.74)0.1829
MTT (s)6.71 (3.05, 9.58)7.66 (3.83, 10.54)0.2034
PS (ml/100 g/min)8.89 (4.94, 12.45)14.37 (11.50, 16.29)< 0.0001

[i] BF = blood flow; BV = blood volume; CT = computed tomography; MTT = mean transit time; PS = permeability surface; SPNs = solitary pulmonary nodules; *compared with the Mann Whitney rank sum test

Table 6

Solitary pulmonary nodules (SPNs) evaluated with dual-energy CT

Parameters*Benign SPNs (n = 118)Malignant SPNs (n = 167)P Values
aOL (HU)13.24 (10.97, 21.58)19.58 (13.29, 26.07)< 0.0001
vOL (HU)11.09 (10.09, 14.86)14.99 (10.59, 23.98)< 0.0001
aIC (mg/ml)0.69 (0.47, 0.985)1.13 (0.70, 1.56)< 0.0001
vIC (mg/ml)0.55 (0.44, 1.00)0.97 (0.50, 1.46)< 0.0001
aNIC0.10 (0.06, 0.13)0.18 (0.11, 0.25)< 0.0001
vNIC0.23 (0.13, 0.32)0.54 (0.43, 0.65)< 0.0001

[i] aIC = Iodine concentration at the arterial phase; aNIC = normalized iodine concentration at the arterial phase; aOL = Iodine overlay at the arterial phase; CT = computed tomography; HU = Hounsfield unit; SPNs = solitary pulmonary nodules; vIC = Iodine concentration at the venous phase; vNIC = normalized iodine concentration at the venous phase; vOL = Iodine overlay at the venous phase; *compared with the Mann Whitney rank sum test

Table 7

Solitary pulmonary nodules evaluated with dual-energy CT

ParametersAUCThresholdSensitivitySpecificity95% CIP values
aOL (HU)0.63613.8970.6661.020.577–0.692< 0.001
vOL (HU)0.63812.7959.8872.030.580–0.694< 0.001
aIC (mg/ml)0.6570.6567.6669.490.599–0.712< 0.001
vIC (mg/ml)0.7030.8568.8671.190.646–0.755< 0.001
aNIC0.7280.1267.6674.580.672–0.778< 0.001
vNIC0.7900.3575.4580.510.738–0.8360.0001

[i] aIC = Iodine concentration at the arterial phase; aNIC = normalized iodine concentration at the arterial phase; aOL = Iodine overlay at the arterial phase; AUC = area under the curve; CT = computed tomography; HU = Hounsfield unit; vIC = Iodine concentration at the venous phase; vOL = Iodine overlay at the venous phase; vNIC = normalized iodine concentration at the venous phase

95 % CI = 95 % confidence interval

Table 8

Pairwise comparison of AUC of dual energy CT parameters in 285 patients with solitary pulmonary nodules

Parameters*Z statisticP value
aOL vs vOL0.09950.9207
aOL vs aIC0.8130.4162
aOL vs vIC2.4850.0129
aOL vs aNIC3.1710.0015
aOL vs vNIC5.170< 0.0001
vOL vs aIC0.7020.4829
vOL vs vIC2.5670.0103
vOL vs aNIC3.2800.0010
vOL vs vNIC5.345< 0.0001
aIC vs vIC2.0340.0420
aIC vs aNIC2.7550.0059
aIC vs vNIC4.728< 0.0001
vIC vs aNIC1.0360.3001
vIC vs vNIC3.2270.0013
aNIC vs vNIC2.7080.0068

[i] aIC = Iodine concentration at the arterial phase; aOL = Iodine overlay at the arterial phase; aNIC = normalized iodine concentration at the arterial phase; AUC = area under the curve; CT = computed tomography; vIC = Iodine concentration at the venous phase; vNIC = normalized iodine concentration at the venous phase; vOL = Iodine overlay at the venous phase; *compared with the DeLong test

Table 9

Solitary pulmonary nodules evaluated with multimodality CT imaging

Methods*Sensitivity (%)Specificity (%)Accuracy (%)PPV (%)NPV (%)
Method A83.2363.5675.0976.3772.82 %
Method B85.6367.8078.2579.0176.92 %
Method C84.4366.1076.8477.9075.00 %
Method A+B94.6174.5886.3284.0490.72 %
Method A+C92.8177.9786.6785.6488.46 %
Method B+C95.8181.3689.8287.9193.20 %
Method A+B+C97.6088.1493.6892.0996.30 %

[i] CT = computed tomography; Method A = non-contrast enhanced CT combined with contrast enhanced CT; Method B = non-contrast enhanced CT combined with dual energy CT; Method C = non-contrast enhanced CT combined with CT perfusion imaging; Method A + B = non-contrast enhanced CT combined with contrast enhanced CT and dual energy CT; Method A + C = non-contrast enhanced CT combined with contrast enhanced CT and CT perfusion imaging; Method B + C = non-contrast enhanced CT combined with dual energy CT and CT perfusion imaging; Method A + B + C = non-contrast enhanced CT combined with contrast enhanced CT, dual-energy CT, and CT perfusion imaging; NPV = Negative predictive value; PPV = Positive predictive value; *compared with the McNemar’s test

Table 10

Step-wise approach of multimodality CT imaging for evaluating solitary pulmonary nodules1, 46, 47

1. Solid
Density2. Subsolid
3. Round or oval
Shape4. Triangular or polygonal
5. Smooth
Margins6. Lobulated
7. Spiculated
Non-contrast enhanced CT8. Fat
Internal characteristics9. Calcification
10. Cavitation
11. Pleural retraction
12. Air bronchogram
Some complex findings13. Bubble like lucencies (pseudocavitation)
14. Cystic airspace
15. Vascular convergence
Contrast enhanced CTParameter (s)16. Degree of enhancement
CT perfusion imagingParameter (s)17. Permeability surface
Dual-energy CTParameter (s)18. Normalized iodine concentration at the venous phase
DOI: https://doi.org/10.2478/raon-2023-0008 | Journal eISSN: 1581-3207 | Journal ISSN: 1318-2099
Language: English
Page range: 20 - 34
Submitted on: Aug 15, 2022
Accepted on: Dec 5, 2022
Published on: Feb 17, 2023
Published by: Association of Radiology and Oncology
In partnership with: Paradigm Publishing Services
Publication frequency: 4 issues per year

© 2023 Gaowu Yan, Hongwei Li, Xiaoping Fan, Jiantao Deng, Jing Yan, Fei Qiao, Gaowen Yan, Tao Liu, Jiankang Chen, Lei Wang, Yang Yang, Yong Li, Linwei Zhao, Anup Bhetuwal, Morgan A. McClure, Na Li, Chen Peng, published by Association of Radiology and Oncology
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.