A prediction model for lung metastases in patients with indeterminate pulmonary nodules in newly diagnosed colorectal cancer.

INTRODUCTION: Multidisciplinary teams treating patients with newly diagnosed Colorectal Cancer (CRC) often encounter the appearance of Indeterminate Pulmonary Nodules (IPNs) that warrants follow-up with repetitive medical imaging and anxiety for patients. We determined the incidence of IPNs in patients with newly diagnosed CRC and developed and validated a model for individualized risk prediction of IPNs being lung metastases. MATERIAL AND METHODS: Newly diagnosed CRC who underwent surgery between November 2011 to June 2014 were included to create the risk model, developed using both clinical experience and statistical selection. Discrimination and calibration slopes of the risk score were evaluated in an independent temporal validation sample. A nomogram is presented to assist clinicians in estimating an individual risk score. RESULTS: Out of 2111 CRC patients staged with chest CT, 204 (9.6%) had IPNs and 54/204 (26%) had lung metastases. We identified 4 predictors: "location of primary tumour", "pathological nodal stage", "size of the largest nodule" and "extrapulmonary synchronous metastases at diagnosis". Discrimination of the final model in the validation sample was demonstrated by the difference in mean predicted risk between progressed cases en non-progressed cases (49% versus 21%, p = <0.001). CONCLUSION: A prediction model with 4 clinical risk factors can be used to assist multidisciplinary teams in the prediction of individualized risk of lung metastases and imaging strategy in patients with IPNs and newly diagnosed colorectal cancer. The model performed well in new patients not included in the model development.

[Evaluation of solid pulmonary nodules in patients with bladder cancer with computed tomography based texture analysis method: Is it possible to differentiate metastatic - non-metastatic nodules?] / Bilgisayarli tomografi tabanli doku analizi yöntemi ile mesane kanserli hastalarda solid pulmoner nodüllerin degerlendirilmesi: Metastatik - metastatik olmayan nodül ayrimini yapmak mümkün mü?

INTRODUCTION: Solid Pulmonary Nodule (SPN) is defined as parenchymal radiopacity smaller than 3 cm in diameter. Evaluating the metastatic nature of the SPNs detected in the thorax computed tomography (TCT) examination for staging purposes in cancer patients becomes a fundamental problem for the physician. Invasive procedures, additional imaging or follow-up imaging, are often used to differentiate metastatic and non-metastatic nodules. In this study, we aimed to distinguish SPNs detected in patients diagnosed with bladder cancer (BC) as metastatic and non-metastatic nodules by texture analysis. MATERIALS AND METHODS: TCT images of patients diagnosed with BC in our hospital from January 2007 until December 2017 were retrospectively evaluated. A total of 46 patients with SPN, including metastatic (n= 19) and non-metastatic (n= 27), were included in the study. Short axis diameter, long-axis diameter, nodule volume and volume histogram values of the nodules were obtained. Chisquare test was used to evaluate dependent variables, and the Mann-Whitney U test was used to evaluate independent variables. ROC curves of the obtained data were plotted. Statistically, the significant p-value was determined as less than 0.05. RESULT: A significant difference was found between SPN long axis, short axis and volume values. In the volumetric histogram analysis, the maximum density value and the mean density value were found to be statistically significant. When the average of the highest densities in the volume histogram data was evaluated, the area under the curve value was 0.702 (95% CI, 519-854). The metastatic nodule could be distinguished with a sensitivity of 88% and a specificity of 70% when the volume histogram has the maximum density threshold of 50 HU. CONCLUSIONS: In this study, we concluded that SPN detected on CT images can be distinguished as metastatic and non-metastatic nodules using texture analysis method without invasive procedures.

Artificial intelligence for detecting small FDG-positive lung nodules in digital PET/CT: impact of image reconstructions on diagnostic performance.

OBJECTIVES: To evaluate the diagnostic performance of a deep learning algorithm for automated detection of small 18F-FDG-avid pulmonary nodules in PET scans, and to assess whether novel block sequential regularized expectation maximization (BSREM) reconstruction affects detection accuracy as compared to ordered subset expectation maximization (OSEM) reconstruction. METHODS: Fifty-seven patients with 92 18F-FDG-avid pulmonary nodules (all ≤ 2 cm) undergoing PET/CT for oncological (re-)staging were retrospectively included and a total of 8824 PET images of the lungs were extracted using OSEM and BSREM reconstruction. Per-slice and per-nodule sensitivity of a deep learning algorithm was assessed, with an expert readout by a radiologist/nuclear medicine physician serving as standard of reference. Receiver-operator characteristic (ROC) curve of OSEM and BSREM were assessed and the areas under the ROC curve (AUC) were compared. A maximum standardized uptake value (SUVmax)-based sensitivity analysis and a size-based sensitivity analysis with subgroups defined by nodule size was performed. RESULTS: The AUC of the deep learning algorithm for nodule detection using OSEM reconstruction was 0.796 (CI 95%; 0.772-0.869), and 0.848 (CI 95%; 0.828-0.869) using BSREM reconstruction. The AUC was significantly higher for BSREM compared to OSEM (p = 0.001). On a per-slice analysis, sensitivity and specificity were 66.7% and 79.0% for OSEM, and 69.2% and 84.5% for BSREM. On a per-nodule analysis, the overall sensitivity of OSEM was 81.5% compared to 87.0% for BSREM. CONCLUSIONS: Our results suggest that machine learning algorithms may aid detection of small 18F-FDG-avid pulmonary nodules in clinical PET/CT. AI performed significantly better on images with BSREM than OSEM. KEY POINTS: • The diagnostic value of deep learning for detecting small lung nodules (≤ 2 cm) in PET images using BSREM and OSEM reconstruction was assessed. • BSREM yields higher SUVmaxof small pulmonary nodules as compared to OSEM reconstruction. • The use of BSREM translates into a higher detectability of small pulmonary nodules in PET images as assessed with artificial intelligence.

Near-infrared intraoperative imaging for minimally invasive pulmonary metastasectomy for sarcomas.

BACKGROUND: Complete pulmonary metastasectomy for sarcoma metastases provides patients an opportunity for long-term survival and possible cure. Intraoperative localization of preoperatively identified metastases and identification of occult lesions can be challenging. In this trial, we evaluated the efficacy of near-infrared (NIR) intraoperative imaging using second window indocyanine green during metastasectomy to identify known metastases and to detect occult nodules. METHODS: Thirty patients with pulmonary nodules suspicious for sarcoma metastases were enrolled in an open-label, feasibility study (NCT02280954). All patients received intravenous indocyanine green (5 mg/kg) 24 hours before metastasectomy. Patients 1 through 10 (cohort 1) underwent metastasectomy via thoracotomy to assess fluorescence patterns of nodules detected by traditional methods (preoperative imaging and intraoperative visualization/bimanual palpation). After confirming reliability within cohort 1, patients 11 through 30 (cohort 2) underwent video-assisted thoracic surgery metastasectomy with NIR imaging. RESULTS: In cohort 1, 14 out of 16 preoperatively identified pulmonary metastases (87.5%) displayed tumor fluorescence. Nonfluorescent metastases were deeper than fluorescent metastases (2.1 cm vs 1.3 cm; P = .03). Five out of 5 metastases identified during thoracotomy displayed fluorescence. NIR imaging identified 3 additional occult lesions in this cohort. In cohort 2, 33 out of 37 known pulmonary metastases (89.1%) displayed fluorescence. Nonfluorescent tumors were deeper than 2.0 cm (P = .007). NIR imaging identified 24 additional occult lesions. Of 24 occult lesions, 21 (87.5%) were confirmed metastases and the remaining 3 nodules were lymphoid aggregates. CONCLUSIONS: NIR intraoperative imaging with indocyanine green (5 mg/kg and 24 hours before surgery) localizes known sarcoma pulmonary metastases and identifies otherwise occult lesions. This approach may be a useful intraoperative adjunct to improve metastasectomy.

Robust response to nivolumab in patient with renal cell carcinoma inferior vena cava tumour thrombus.

A 47-year-old previously healthy man presented with acute moderate flank pain. Evaluation revealed left renal cell carcinoma, with inferior vena cava tumour thrombus invasion. Patient had no significant history or risk factors to pre-dispose him to genitourinary cancers. Surgery was deemed to not be appropriate due to distant metastases, but patient received targeted molecular therapy and immunotherapy with striking regression of the thrombus.

Indeterminate Pulmonary Nodules at Diagnosis in Rhabdomyosarcoma: Are They Clinically Significant? A Report From the European Paediatric Soft Tissue Sarcoma Study Group.

PURPOSE: To evaluate the clinical significance of indeterminate pulmonary nodules at diagnosis (defined as ≤ 4 pulmonary nodules < 5 mm or 1 nodule measuring ≥ 5 and < 10 mm) in patients with pediatric rhabdomyosarcoma (RMS). PATIENTS AND METHODS: We selected patients with supposed nonmetastatic RMS treated in large pediatric oncology centers in the United Kingdom, France, Italy, and the Netherlands, who were enrolled in the European Soft Tissue Sarcoma Study Group (E pSSG) RMS 2005 study. Patients included in the current study received a diagnosis between September 2005 and December 2013, and had chest computed tomography scans available for review that were done at time of diagnosis. Local radiologists were asked to review the chest computed tomography scans for the presence of pulmonary nodules and to record their findings on a standardized case report form. In the E pSSG RMS 2005 Study, patients with indeterminate pulmonary nodules were treated identically to patients without pulmonary nodules, enabling us to compare event-free survival and overall survival between groups by log-rank test. RESULTS: In total, 316 patients were included; 67 patients (21.2%) had indeterminate pulmonary nodules on imaging and 249 patients (78.8%) had no pulmonary nodules evident at diagnosis. Median follow-up for survivors (n = 258) was 75.1 months; respective 5-year event-free survival and overall survival rates (95% CI) were 77.0% (64.8% to 85.5%) and 82.0% (69.7% to 89.6%) for patients with indeterminate nodules and 73.2% (67.1% to 78.3%) and 80.8% (75.1% to 85.3%) for patients without nodules at diagnosis ( P = .68 and .76, respectively). CONCLUSION: Our study demonstrated that indeterminate pulmonary nodules at diagnosis do not affect outcome in patients with otherwise localized RMS. There is no need to biopsy or upstage patients with RMS who have indeterminate pulmonary nodules at diagnosis.

Free-breathing ultrashort echo time lung magnetic resonance imaging using stack-of-spirals acquisition: A feasibility study in oncology patients.

OBJECTIVES: To investigate the diagnostic accuracy of lung magnetic resonance imaging (MRI) with a free-breathing three-dimensional ultrashort echo time spoiled gradient echo sequence using a stack-of-spirals acquisition (spiral 3D UTE) for pulmonary nodule detection at 3 T in oncology patients. METHODS: The institutional review board approved this retrospective study. Between June and September of 2017, 32 oncology patients underwent both free-breathing spiral 3D UTE of the lungs and thin-section chest computed tomography (CT) for pulmonary metastasis workups. Semiquantitative analyses of the visible pulmonary vessels, bronchi, mediastinum, and overall image quality on spiral 3D UTE were assessed by two reviewers; CT was used as the reference standard. The probability of nodule presence also was assessed. RESULTS: The mean acquisition duration of the spiral 3D UTE was 327 s (range, 300-465 s). The pulmonary vessels and bronchi were visible nearly consistently up to the sub-sub-segmental branch levels on spiral 3D UTE (96.9% [31/32] and 90.6% [29/32], respectively). >90% of the spiral 3D UTE images had an acceptable or good mediastinal evaluation; >80% had good or excellent overall image quality. Fifty nodules (6.1 ±â€¯5.9 mm) were identified in 13 patients on CT; the overall nodule detection rate of spiral 3D UTE was 86% (43/50). All 20 nodules ≥ 5 mm in diameter were identified on spiral 3D UTE (100%). CONCLUSIONS: Free-breathing spiral 3D UTE had high sensitivity for the detection of pulmonary nodules, a reasonable scan duration, and acceptable image quality, which may make it a potential alternative to CT for oncology patients.

Clinical and Radiological Discrimination of Solitary Pulmonary Lesions in Colorectal Cancer Patients.

OBJECTIVES: The lung is one of the most common organs of metastasis from colorectal cancer (CRC), and we have encountered lung cancer patients with a history of CRC. There have been few studies regarding methods used to discriminate between primary lung cancer (PLC) and pulmonary metastasis from CRC (PM-CRC) based only on preoperative findings. We retrospectively investigated predictive factors discriminating between these lesions in patients with a history of CRC. METHODS: Between 2006 and 2015, 117 patients with a history of CRC (44 patients with 47 PLC and 73 patients with 102 PM-CRC) underwent subsequent or concurrent resection of pulmonary lesions. We compared the clinical and radiological characteristics of 100 patients with solitary lesions (43 PLC and 57 PM-CRC). Using univariate and multivariate analyses, we examined predictive factors for discrimination of these two lesions. RESULTS: All tumors with findings of ground-glass opacity (GGO) were PLC (n = 19). In a multivariate analysis of 81 radiologically solid tumors, two factors were found to be significant independent predictors of PLC: a history of stage I CRC and presence of pleural indentation. All tumors in 26 patients with either GGO or both a stage I CRC history and pleural indentation were PLC, while most tumors in patients without all three factors were PM-CRC (43/44; 97.7%). CONCLUSIONS: The presence or absence of GGO, pathological CRC stage, and pleural indentation could be useful factors to distinguish between PLC and PM-CRC.

Management of pulmonary nodules in head and neck cancer patients - Our experience and interpretation of the British Thoracic Society Guidelines.

BACKGROUND: and purpose of the study: The frequency of lung nodules in the head and neck cancer population is unknown, currently the only guidance available recommends following local policy. The aim of this study was to determine the incidence of pulmonary nodules in our head and neck cancer group and interpret the recently updated British Thoracic Society (BTS) Lung Nodule Guidelines in a head and neck cancer setting. METHODS: 100 patients were diagnosed with head and neck cancer between July 2013-March 2014, clinico-pathological, demographic and radiological data was extracted from the electronic records. Images with lung findings were re-reviewed by a single consultant radiologist for patients with lung pathology on the initial staging CT report. RESULTS: Twenty patients (20%) had discreet pulmonary findings on CT. Eleven (11%) had lung nodules, 6 (6%) had lesions suspicious for metastasis and 3 (3%) had co-incidental bronchogenic primary cancers. These patients were re-imaged between 6 and 18 months and in 1 patient the previously identified 7 mm nodule had progressed to 16 mm at 1 year. There was no set follow up imaging protocol used. CONCLUSION: The MDT in NHS Lothian has reviewed the BTS guidance and now has a local policy for the management of lung nodules in head and neck cancer patients. Lung Nodules in the head and neck cancer population are common >10%. Higher risk patients with larger nodules should be risk assessed with validated assessment tools. PET-CT has a place in the assessment of lung nodules when risk of malignancy is high.

Pulmonary metastasis of a meningioma presenting as a solitary pulmonary nodule: 2 case reports.

Distant metastases of meningioma are rare, especially in grade 1 meningiomas. In a recent literature review, only 115 cases were found. In almost all published cases, the meningioma was treated several years before the metastasis was diagnosed. The lungs are the most frequent site of metastasis. We describe two patients treated for meningioma (one case grade 1, the other grade 3) who were referred to the Respiratory Oncology Unit because of the incidental finding of a pulmonary nodule on routine chest radiography. Both had undergone several neurosurgical procedures but the last operation was more than 7 years before in both cases. Positron emission tomography scan was suggestive of a malignant lung tumour. The lesions were surgically removed. Pathology confirmed meningioma in both cases with the same WHO grade, immunohistochemical and genetic profiles as the original meningioma. Both patients recovered well from thoracic surgery. The patient with grade 3 meningioma died three years later from intracranial recurrence. When a patient previously treated for meningioma develops a nodular lung lesion, metastasis of the meningioma should be in the differential diagnosis list. Because of the occurrence of distant metastasis even in grade I meningiomas, we suggest that the grading system should take into account genetic changes in the meningioma. Chromosome 1p and 14q losses possibly explain the aggressive behaviour of the grade 1 meningioma.

A rare case of pulmonary sclerosing hemagioma with lymph node metastasis and review of the literature.

Pulmonary sclerosing hemagioma (SH) is an uncommon tumor with malignance potential. Clinically this disease is regarded as benign but extremely rare cases can have lymph node metastasis. Up to date, there have been only very few reports concerning SH with lymph node metastasis. In this paper we reported one pulmonary SH case with lymph node metastasis and additionally overviewed the clinical and pathological features of SH. A young-aged female was found incidentally to have a nodule in the right upper lung. This patient presented no cough, no hemoptysis and chest pain. Computed tomography (CT) scan indicated a large mass in the right upper lung and enlarged lymph nodes in the right hilum. The patient underwent lobectomy of the right upper lung. Histologically, the tumor demonstrated typical features of SH and was consisted of angiomatoid areas, sclerosis, papillary structures lined with cuboidal cells and sheets of round to polygonal cells. Polygonal cells in some solid areas presented abnormal enlarged nuclei and increased karyoplasmic ratio; tumor giant cells were noted; whereas mitosis was not observed. One peribronchial lymph node was noted for SH metastasis and the metastatic tissue were consisted of polygonal cells. Immunohistochemistry (IHC) revealed that both surface-lining cuboidal and polygonal cells expressed EMA and thyroid transcription factor 1 (TTF-1), but were negative for CD34, VIII factor, CD68 and Claratinin. The polygonal cells showed relatively higher expression of Ki-67 and p53 than the surface-lining cells. Postoperatively, the patient received no chemotherapy or radiotherapy and no recurrence 2 years after surgery was noted.

Pulmonary nodules in liver transplant candidates with hepatocellular carcinoma: Imaging characteristics and clinical outcomes.

No guidelines exist for the management of pulmonary nodules in patients with hepatocellular carcinoma (HCC) who are being evaluated for liver transplantation. The 172 patients with HCC who were listed for liver transplant at our institution received both pretransplant chest computed tomography (CT) and follow-up CT. Pulmonary nodules on CT were characterized and followed on subsequent scans by a blinded radiologist, with a consensus review with a second radiologist being performed for equivocal cases. Nodule characteristics and outcomes were examined with chi-square tests, and the posttransplant survival of patients with different nodule outcomes was compared. Cumulative probabilities of waiting-list removal for nontransplant patients and cumulative probabilities of undergoing transplantation for all patients were also compared between patients with and without pulmonary nodules. Of all the patients, 76.2% had at least 1 pulmonary nodule on pretransplant CT, with 301 total nodules characterized; 2.7% of nodules represented HCC metastases, 1.0% represented other bronchopulmonary malignancies, and 2.7% represented infections. None of the malignant nodules exhibited a triangular/lentiform shape or calcifications. There were no statistically significant differences in pulmonary nodule outcomes between patients who underwent transplantation and those who did not undergo transplantation. No significant differences in posttransplant survival were found between patients with different nodule outcomes. There was also no significant difference between patients with and without nodules in the cumulative probabilities of waiting-list removal. However, the cumulative probability of undergoing liver transplantation was borderline significantly higher in patients without pulmonary nodules. In conclusion, despite the low prevalence of malignant nodules, all pulmonary nodules besides triangular/lentiform-shaped or calcified nodules should be followed with serial CT while the patient is on the transplant list, with biopsy performed for new and/or enlarged nodules. Both malignancy and active infection must be excluded when one is confronted with enlarged pulmonary nodules. Clinicians should also be aware of the possibility of reactivation of a granulomatous infection after transplantation.

Risk of malignancy in pulmonary nodules: A validation study of four prediction models.

OBJECTIVES: Clinical prediction models assess the likelihood of malignancy in pulmonary nodules detected by computed tomography (CT). This study aimed to validate four such models in a UK population of patients with pulmonary nodules. Three models used clinical and CT characteristics to predict risk (Mayo Clinic, Veterans Association, Brock University) with a fourth model (Herder et al. [4]) additionally incorporating (18)Fluorine-Fluorodeoxyglucose (FDG) avidity on positron emission tomography-computed tomography (PET-CT). MATERIALS AND METHODS: The likelihood of malignancy was calculated for patients with pulmonary nodules (4-30mm diameter) and data used to calculate the area under the receiver operating characteristic curve (AUC) for each model. The models were used in a restricted cohort of patients based on each model's exclusion criteria and in the total cohort of all patients. RESULTS: Two hundred and forty-four patients were studied, of whom 139 underwent FDG PET-CT. Ninety-nine (40.6%) patients were subsequently confirmed to have malignant nodules (33.2% primary lung cancer, 7.4% metastatic disease). The Mayo and Brock models performed similarly (AUC 0.895 and 0.902 respectively) and both were significantly better than the Veterans Association model (AUC 0.735, p<0.001 and p=0.002 respectively). In patients undergoing FDG PET-CT, the Herder model had significantly higher accuracy than the other three models (AUC 0.924). When the models were tested on all patients in the cohort (i.e. including those outside the original model inclusion criteria) AUC values were reduced, yet remained high especially for the Herder model (AUC 0.916). For sub-centimetre nodules, AUC values for the Mayo and Brock models were 0.788 and 0.852 respectively. CONCLUSIONS: The Mayo and Brock models showed good accuracy for determining likelihood of malignancy in nodules detected on CT scan. In patients undergoing FDG PET-CT for nodule evaluation, the highest accuracy was seen for the model described by Herder et al. incorporating FDG avidity.

Noninvasive Differential Diagnosis of Pulmonary Nodules Using the Standardized Uptake Value Index.

OBJECTIVES: We previously showed that the standardized uptake value (SUV) index, which was defined as the ratio of the maximum SUV of the tumor to mean SUV of the liver, was a surrogate marker of lung cancer aggressiveness. In this study of patients with pulmonary nodules (PNs), we explored whether the SUV index could be used to differentiate small malignant from small benign PNs. METHODS: A total of 284 patients with solitary PNs ≤2 cm in size underwent positron emission tomography/computed tomography and surgery. The associations between pathological findings and clinical factors were evaluated. RESULTS: The median SUV indices of lung cancer, metastatic PNs and benign nodules were 1.2, 1.5, and 0.6, respectively (P <0.01). A SUV index cut-off value of 1.2 was used to differentiate benign from malignant nodules. When patients were grouped according to SUV index cut-off values of <1.2 or ≥1.2, the following cases were false-negative: lung adenocarcinoma (P <0.01), kidney as primary site (P <0.01), and metastatic PNs with long disease-free survival (P = 0.02). CONCLUSIONS: As a noninvasive diagnostic marker, the SUV index was found to be useful for differentiating benign from malignant small PNs.

Indeterminate pulmonary nodules in colorectal cancer.

AIM: To investigate the clinicopathologic parameters of pulmonary metastasis in colorectal cancer (CRC) patients after lung operation of indeterminate pulmonary nodules (IPNs). METHODS: From a prospective database of CRC patients, 40 cases that underwent lung operation between November 2008 and December 2012 for suspicious metastatic pulmonary nodules on chest computed tomography (CT) were enrolled. The decision to perform a lung operation was made if the patient met the following criteria: (1) completely resected or resectable primary CRC; (2) completely resectable IPNs; (3) controlled or controllable extrapulmonary metastasis; and (4) adequate general condition and pulmonary function to tolerate pulmonary operation. Lung operation was performed by a thoracic surgeon without CT-guided biopsy for pathologic confirmation. RESULTS: A total of 40 cases of lung resection was performed in 29 patients. Five patients underwent repeated lung resection. The final pathology result showed metastasis from the CRC in 30 cases (75%) and benign pathology in 10 cases (25%). The primary tumor site was the rectum in 26/30 (86.6%) cases with pulmonary metastasis, but only 3/10 (30%) cases in the benign group had a primary rectal cancer (P = 0.001). Positron emission tomography (PET)-CT was performed for 22/30 (73.4%) patients in the lung metastasis group and for 6/10 (60.0%) patients in the benign group. PET-CT revealed hot uptake of (18)fluorine 2-fluoro-2-deoxy-D-glucose with all IPNs in both groups. The group with pulmonary metastasis had a higher incidence of primary rectal cancer (P = 0.001), a more advanced tumor stage (P = 0.011), and more frequent lymphatic invasion of tumor cells (P = 0.005). Six cases with previous liver metastasectomy were present in the lung metastasis group. Serum carcinoembryonic antigen levels before lung operation were not elevated in any of the patients. CONCLUSION: The stage and location of the primary tumor and tumor cell infiltration of lymphatics provide useful indicators for deciding on lung resection of IPNs in CRC.

Sclerosing hemangioma of the lung mimicking pulmonary metastasis.

Sclerosing hemangioma (SH) of the lung is a rare, benign neoplasm, initially thought to be of vascular origin, but immunohistochemical results suggest an origin from primitive respiratory epithelium. We report a case of SH initially misdiagnosed as malignant due to the strong FDG-PET uptake.

Indeterminate pulmonary nodules in colorectal-cancer: do radiologists agree?

BACKGROUND: The clinical significance of indeterminate pulmonary nodules (IPN) at staging computed tomography (CT) for colorectal cancer (CRC), and the optimal diagnostic approach, are debated. This study aimed to analyse variability in radiologists' detection of IPN at staging CT for CRC. METHODS: All patients with CRC referred to our center between 2006 and 2011 were included. Primary staging CT scans were re-evaluated by an experienced thoracic radiologist whose findings were entered into a dedicated database and merged with data from the Danish Colorectal Cancer Group database, the National Patient Registry, the Danish Pathology Registry, and the primary CT evaluation. Inter-reader agreement was calculated by Kappa statistics, and associations between variables and malignancy of pulmonary nodules were analyzed with χ (2) and Mann-Whitney-Wilcoxon tests. Multivariable logistic regression analyses were used to adjust for potential confounding variables. RESULTS: In total, 841 patients were included. The primary CT assessment reported IPN in 9.8 % of patients and pulmonary metastases in 5.1 % of patients compared with 5.6 and 7.0 %, respectively, reported by the experienced thoracic radiologist. Kappa for agreement between the primary assessor and the thoracic radiologist on IPN was 0.31 and 0.65 for pulmonary metastases. Synchronous liver metastases were predictive of malignancy of IPN (adjusted odds ratio 20.1; 95 % confidence interval 2.64-437.66; p = 0.012), whereas no other investigated radiological characteristics or clinicopathological factors were significantly associated with malignancy of IPN. CONCLUSION: The characterization of pulmonary findings on staging CT for CRC varied greatly between the radiologists, and double-reading of scans with IPN is recommended prior to further diagnostic work-up.