’Original article The diagnostic yield of preoperative staging computed tomography of the thorax in colorectal cancer patients without hepatic metastases Gabriella Yonguea, Alexander Hotourasa, Jamie Murphyb, Hasan Mukhtara, Chetan Bhana and Christopher L. Chanc Background National UK guidelines recommend preoperative computed tomography (CT) examination of the chest for all patients with colorectal cancer (CRC) as identification of thoracic metastatic disease may influence treatment plans. The diagnostic yield of this approach for patients without radiological evidence of hepatic metastases remains unclear. The aim of this study was to establish the impact upon treatment of chest CT examination for CRC patients without metastatic liver disease. Patients and methods All patients diagnosed with CRC without liver metastases disease between December 2006 and December 2013 in a single institution were identified from a prospectively recorded departmental database. Data collected included patient demographics, disease stage and chest CT findings. Results There were 358 CRC patients (183 male, 175 female, median age 70 years, range 24–96 years) without liver metastases. Colon cancer accounted for 68% of cases with 32% of patients diagnosed with rectal cancer. The majority of patients (n = 292, 81.6%) had no evidence of thoracic pathology during the study period. Lung metastases were identified in five (1.4%) patients, one of them subsequently diagnosed with benign pulmonary nodule. In 61 (17.0%) patients the lung nodules were classified as intermediate with only three subsequently reclassified as pulmonary metastases (n = 1) or primary lung cancer (n = 2). Conclusion In the absence of liver metastases, CRC rarely spreads to the lung. Consequently, preoperative chest CT examination for patients without evidence of liver metastases may not be necessary in the majority of patients. Future studies are required to identify patients at high risk for pulmonary metastases who may benefit from this preoperative investigation. Eur J Gastroenterol Hepatol 27:467–470 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Introduction
Over 40 000 new cases of colorectal cancer (CRC) are diagnosed annually in the UK [1], of which 30% are classified as metastatic at presentation [2]. The National Institute of Clinical Excellence (NICE) in the UK recommends routine staging computed tomography (CT) of the chest as well as the abdomen and pelvis to establish the disease burden and guide treatment plans within a multidisciplinary setting [3]. This diagnostic approach, however, has been questioned because of the fact that pulmonary metastases are uncommon in patients without intra-abdominal metastases [4]. The colon is drained by the portal venous system and, thus, the liver is generally European Journal of Gastroenterology & Hepatology 2015, 27:467–470 Keywords: colorectal cancer, computed tomography thorax, pulmonary metastases a Department of Colorectal Surgery, Whittington Health NHS Trust, University College London, London, bPhysiology Unit, St Marks Hospital, Harrow and c Academic Surgical Unit, The Royal London Hospital, Bart’s Health NHS Trust, London, UK
Correspondence to Alexander Hotouras, BSc, MSc, MRCS (Eng), MD(Res), Academic Surgical Unit, Barts Health NHS Trust, National Centre for Bowel Research and Surgical Innovation, Barts and the London School of Medicine and Dentistry, 1st Floor, Abernethy Building, 2 Newark Street, London E1 2AT, UK Tel: + 44 207 882 8751; e-mail: [email protected] Received 2 December 2014 Accepted 20 January 2015
the first site of CRC metastasis, except in distal rectal disease, when metastatic spread can bypass the portal system and directly enter the systemic circulation through the inferior and middle rectal veins [5]. It is reported that 10–15% of CRC metastases are to the lung and, of these, only 10% occur in the absence of hepatic spread [6], with the incidence being double in rectal cancer compared with colon cancer [7]. Recent studies have advocated that routine chest CT should not be a part of the routine preoperative staging investigations for CRC [6,8]. The rationale for this approach includes the low incidence of pulmonary metastasis, the minimal influence on treatment plan and the risk associated with increased radiation exposure [6]. The main objective of this study was to evaluate the clinical utility of preoperative chest CT examination for patients diagnosed with CRC in the absence of liver metastases and the impact of the clinical findings on subsequent treatment plans. Patients and methods
All patients with a histological diagnosis of CRC in our institution between December 2006 and December 2013 were identified from a prospectively recorded departmental database. This included information such as patient demographics, date of diagnosis, stage of disease at
0954-691X Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
DOI: 10.1097/MEG.0000000000000315
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
467
468
April 2015 • Volume 27 • Number 4
European Journal of Gastroenterology & Hepatology
Table 1. Patient demographics, colorectal tumour location and staging Age Median Range Sex Female Male Tumour sites [n (%)] Rectum Rectosigmoid Sigmoid Descending colon Transverse colon Ascending colon Caecum Disease stage Stage 1 Stage 2 Stage 3 Stage 4
73 24–96 175 183 113 45 69 17 16 39 59
(31.6) (12.6) (19.3) (4.7) (4.5) (10.9) (16.5)
61 129 156 12
presentation and treatment plans following discussion in a multidisciplinary team (MDT) meeting. All patients diagnosed with a colorectal malignancy underwent preoperative contrast or noncontrast (if contrast was contraindicated) CT of the chest, abdomen and pelvis. Imaging was obtained using a Siemens Somatom Helical CT scanner (Siemens, Germany). Volumetric imaging of the thorax, abdomen and pelvis was taken following intravenous contrast (unless contraindicated) of 90 ml Omnipaque 200 at 2 ml/s and oral contrast of 300 ml of 1.5% Gastrografin. Slice reconstructions of 3–5 mm thickness were created. Lung lesions were radiologically classified as benign, intermediate or malignant. All images were reviewed and reported by two radiologists in the colorectal MDT meeting with the results prospectively recorded on our departmental database. Pulmonary lesions that could not be defined as either benign or malignant were termed ‘intermediate nodules’. Patients with intermediate lung nodules underwent follow-up CT scan examination of the chest within 6 months. They were considered benign if subsequent follow-up scans, for at least 1 year, showed no signs of malignant changes or growth. Conversely, lesions that increased in size or number were considered malignant and confirmation was sought from histological or cytological examination. Patients with non-CRCs such as appendiceal cancer, neuroendocrine tumours, anal squamous cell carcinomas and gastrointestinal stromal tumours were excluded from this study. Results
A total of 574 patients were discussed in the CRC MDT during the study period. Two hundred and sixteen patients were excluded from this study as they did not fulfil the selection criteria (n = 131 had liver metastases, n = 85 had non-CRC malignancies). The remaining 358 patients (183 male, 175 female, median age 70 years, range 24–96 years) were included in the analysis. Table 1 summarizes patient demographics and tumour characteristics. The cohort included patients with adrenal (n = 2), kidney (n = 1), peritoneal (n = 2), mesenteric (n = 1) and bone (n = 1) metastases.
There were 292 (81.6%) patients without any radiological evidence of pulmonary nodules (Fig. 1). Sixty-one (17.0%) patients had pulmonary nodules classified as indeterminate on initial CT assessment. On serial imaging, one of these patients was deemed to have pulmonary metastatic disease, whereas a further two patients were diagnosed with incidental primary lung cancers. The remaining 58 patients were deemed to have benign lesions. Finally, five patients were initially diagnosed with pulmonary metastases on initial review of chest CT imaging. In one case these lesions were subsequently reclassified as benign. In total, 7/358 (2.0%) patients had pulmonary metastases or a primary lung malignancy in the absence of liver metastases. The median age of patients (three women, two men) with pulmonary metastases was 81 years (range 61–87 years). The TNM staging of the CRC in each one of these cases is shown in Table 2. All tumours were located either in the rectum or the rectosigmoid junction. Three patients had metastases to other nonhepatic sites that included nonregional lymph nodes (n = 3) and the adrenal glands (n = 1). The median number of lung nodules in each patient was nine (range 2–10) and the median maximum size of the nodules was 13 mm (range 5–26 mm). None of these patients underwent pulmonary metastasectomy. Four patients underwent bowel resection and received chemoradiotherapy. Comparatively, the median age of patients (31 women, 27men) with intermediate nodules was 74 years (range 40–89 years). Thirty (52%) patients had rectal or rectosigmoid disease, of these, 16 (53%) had positive local lymph nodes. The median number of lung nodules in each patient was 1 (range 1–5) and the median maximum size of the nodules was 3 mm (range 1–12 mm). Discussion
This study has demonstrated that the incidence of pulmonary metastases for patients with CRC in the absence of hepatic metastases is very low. All five patients in this study with pulmonary metastases had either rectal or rectosigmoid primary tumours. Thus, the prevalence of pulmonary metastases for patients with rectal cancer was 2.7% (3/113), whereas in the rectosigmoid cancer cohort it was 4.4% (2/45). All these patients had a T3 or T4 cancer with regional or nonregional lymph node involvement. This finding is in keeping with the work of Kim et al. [4], who retrospectively assessed 319 CRC patients without liver metastases and a negative chest radiograph (CXR). Kim et al. [4] reported that lymph node involvement is a risk factor for pulmonary metastatic disease and thus concluded that chest CT should not be routinely offered to patients without evidence of liver or lymph node spread. Brent et al. [9] and Christoffersen et al. [10] have also published similar findings. During our study period, 61 patients were initially reported to have intermediate pulmonary nodules as demonstrated by CT examination of the chest; however, one (1.6%) of these patients was later diagnosed with metastatic disease on surveillance CT imaging. Conversely, 1/5 (20%) patients who were initially diagnosed with pulmonary metastases was later downgraded to a benign lesion on review of interval chest CT examination.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
The diagnostic yield of preoperative staging CT Yongue et al.
www.eurojgh.com
469
Colorectal cancer patients with no hepatic metastasis (n = 358, 100%)
Clear lungs on initial staging CT chest (n = 292, 81.6%)
Intermediate lung nodules on initial CT chest (n = 61, 17.0%)
Likely lung metastasis on initial CT chest (n = 5, 1.4%)
Follow-up CT chest : Benign nodules (n = 58, 95.1%) Pulmonary metastases (n = 1, 1.6%) Primary lung cancer (n = 2, 3.3%)
Follow-up CT chest: Benign nodules (n = 1, 20.0%) Pulmonary metastases (n = 4, 80.0%)
Fig. 1. Study flow chart demonstrating radiological diagnosis of pulmonary nodules on computed tomography (CT).
Table 2. TNM staging and pulmonary nodule characteristics in patients with pulmonary metastases (n = 5) Locations
Disease stage
Rectal Rectosigmoid Rectal Rectosigmoid Rectal
T4N2M1b T4N1M1a T3N1M1a T4N0M1b T3N0M1b
Number of lung nodules
Maximum size of nodules (mm)
10 2 9 10 3
5 9 13 26 19
It should be noted that although all patients in our study had treatment for their primary malignancy, none had surgical management of their pulmonary metastatic disease. This is because none of these individuals met the criteria for metastasectomy, which include absence of extrathoracic disease (or extrathoracic disease that is resectable), pulmonary disease that is completely resectable, or adequate cardiopulmonary reserve [11–13]. A further two (0.5%) patients with intermediate lesions were diagnosed with synchronous primary malignancies on rereview of preoperative chest CT imaging. It has been reported that the incidence of synchronous lung and CRC primary malignancies is less than 0.6% [14], which is in keeping with the results of our study. However, identification of this small cohort is significant as identification of a pulmonary lesion as a primary lung cancer rather than as a metastatic CRC significantly changes patient management both in terms of chemotherapy and in terms of surgical options to resect pulmonary disease. Despite the limitations in diagnostic accuracy outlined above, CT has gradually become the preferred modality of imaging the chest because of its high sensitivity. CXR has a sensitivity of only 33% in comparison with 75% for CT [10,15,16], with chest CT examinations capable of reliably identifying nodules of ∼ 2–3 mm, whereas CXR can only demonstrate lesions that are 5–10 mm in size [17]. Although the use of chest CT imaging allows the detection of more subtle metastatic disease, there is also the
downside of increased incidental findings such as intermediate lung nodules as discussed above. These lesions are often too small to characterize with other forms of investigations such as fluorodeoxyglucose-PET scanning (53% of nodules < 1 cm are incorrectly identified) [18] or biopsy. Fine-needle biopsy/aspiration also carries its own morbidity including 24% risk of haemorrhage [19], 12% chance of pneumothorax [19] and a 5% chance of a 3-day hospital admission as well as costing £468 to perform [20]. Consequently, the CRC MDT process is forced to recommend follow-up imaging with serial high-resolution CT scans at 3–12-month intervals [21], exposing patients to significant additional radiation [21]. Furthermore, this approach also increases costs for the hospital or healthcare system and creates anxiety for the patient and their family. Yet, the majority of these lesions are benign as demonstrated by the Mayo Lung Trial, in which the incidence of cancer in intermediate lung nodules less than 9 mm was 0.69% [22]. Additionally, in a study of 100 CRC patients with liver metastases and negative CXR by Povoski et al. [23], the positive predictive value of chest CT was only 36% with a positive yield of only 4%. Consequently, it can be argued that CT imaging of the chest should not be a part of routine preoperative staging for CRC patients and instead should only be considered for patients at high risk for pulmonary metastases. These thoughts have been echoed in studies by Quyn et al. [24] and Hogan et al. [25], who demonstrated that intermediate nodules in patients with small localized primary CRC (i.e. stage T1/T2 or negative lymph nodes) did not progress into metastases. Furthermore, MDT meetings should consider the value of follow-up CT imaging or other more invasive investigations and whether management options would change in the rare scenario that a pulmonary nodule is proven to be malignant [26]. Perhaps as Kim et al. [27] suggest, a model to predict the progression of intermediate pulmonary nodules would be of benefit, first in deciding which patients are at risk for
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
470
European Journal of Gastroenterology & Hepatology
pulmonary metastases and thus requiring a chest CT and second to decide whether further investigations are necessitated should a nodule be found. Conclusion
The incidence of pulmonary metastases for patients with CRC in the absence of hepatic metastases is extremely low. As a result, the clinical utility of preoperative chest CT examination for all patients undergoing surgery for CRC is questionable. Prospective clinical trials are necessary to determine whether this preoperative investigation should be restricted to a subgroup of patients with an increased risk of pulmonary metastases. Acknowledgements Conflicts of interest
There are no conflicts of interest. References 1 Cancer Research UK. Bowel cancer incidence statistics; 2014. Available at: http://www.cancerresearchuk.org/cancer-info/cancerstats/types/bowel/inci dence/uk-bowel-cancer-incidence-statistics [Accessed 1 November 2014]. 2 Midgley R, Kerr D. Colorectal cancer. Lancet 1999; 353:391–399. 3 National Institute for Health and Clinical Excellence. NICE clinical guideline 131: The diagnosis and management of colorectal cancer. London: National Institute for Health and Clinical Excellence; 2011. Available at: http://www.nice.org.uk/nicemedia/live/13597/56998/56998.pdf. 4 Kim HY, Lee SJ, Lee G, Song L, Kim SA, Kim JY, et al. Should preoperative chest CT be recommended to all colon cancer patients? Ann Surg 2014; 259:323–328. 5 McCormack PM, Attiyeh FF. Resected pulmonary metastases from colorectal cancer. Dis Colon Rectum 1979; 22:553–556. 6 Pihl E, Hughes ES, McDermott FT, Johnson WR, Katrivessis H. Lung recurrence after curative surgery for colorectal cancer. Dis Colon Rectum 1987; 30:417–419. 7 Tan KK, Lopes Gde L Jr, Sim R. How uncommon are isolated lung metastases in colorectal cancer? A review from database of 754 patients over 4 years. J Gastrointest Surg 2009; 13:642–648. 8 Grossmann I, Avenarius JK, Mastboom WJ, Klaase JM. Preoperative staging with chest CT in patients with colorectal carcinoma: not as a routine procedure. Ann Surg Oncol 2010; 17:2045–2050. 9 Brent A, Talbot R, Coyne J, Nash G. Should indeterminate lung lesions reported on staging CT scans influence the management of patients with colorectal cancer? Colorectal Dis 2007; 9:816–818. 10 Christoffersen MW, Bulut O, Jess P. The diagnostic value of indeterminate lung lesions on staging chest computed tomographies in patients with colorectal cancer. Dan Med Bull 2010; 57:A4093.
April 2015 • Volume 27 • Number 4
11 Greelish JP, Friedberg JS. Secondary pulmonary malignancy. Surg Clin North Am 2000; 80:633–657. 12 Pfannschmidt J, Klode J, Muley T, Dienemann H, Hoffmann H. Nodal involvement at the time of pulmonary metastasectomy: experiences in 245 patients. Ann Thorac Surg 2006; 81:448–454. 13 Inoue M, Ohta M, Iuchi K, Matsumura A, Ideguchi K, Yasumitsu T, et al. Thoracic Surgery Study Group of Osaka University. Benefits of surgery for patients with pulmonary metastases from colorectal carcinoma. Ann Thorac Surg 2004; 78:238–244. 14 Evans HS, Møller H, Robinson D, Lewis CM, Bell CM, Hodgson SV. The risk of subsequent primary cancers after colorectal cancer in southeast England. Gut 2002; 50:647–652. 15 Hwang M, Jung S, Kim J, Shim M. Diagnostic usefulness of routinely performed preoperative chest computed tomography for colorectal cancer. J Korea Soc Coloproctol 2010; 26:211–216. 16 McIntosh J, Sylvester PA, Virjee J, Callaway M, Thomas MG. Pulmonary staging in colorectal cancer – is computerised tomography the answer? Ann R Coll Surg Engl 2005; 87:331–333. 17 Schaner EG, Chang AE, Doppman JL, Conkle DM, Flye MW, Rosenberg SA. Comparison of computed and conventional whole lung tomography in detecting pulmonary nodules: a prospective radiologic–pathologic study. Am J Roentgenol 1978; 131:51–54. 18 Bryant AS, Cerfolio RJ. The maximum standardized uptake values on integrated FDG-PET/CT is useful in differentiating benign from malignant pulmonary nodules. Ann Thorac Surg 2006; 82:1016–1020. 19 Schneider T, Puderbach M, Kunz J, Bischof A, Giesel FL, Dienemann H, et al. Simultaneous computed tomography-guided biopsy and radiofrequency ablation of solitary pulmonary malignancy in high-risk patients. Respiration 2012; 84:501–508. 20 Baldwin DR, Eaton T, Kolbe J, Christmas T, Milne D, Mercer J, et al. Management of solitary pulmonary nodules: how do thoracic computed tomography and guided fine needle biopsy influence clinical decisions? Thorax 2002; 57:817–822. 21 Kernstine KH, Grannis FW Jr, Rotter AJ. Is there a role for PET in the evaluation of subcentimeter pulmonary nodules? Semin Thorac Cardiovasc Surg 2005; 17:110–114. 22 Swensen SJ, Jett JR, Hartman TE, Midthun DE, Mandrekar SJ, Hillman SL, et al. CT screening for lung cancer: five-year prospective experience. Radiology 2005; 235:259–265. 23 Povoski SP, Fong Y, Sgouros SC, Kemeny NE, Downey RJ, Blumgart LH. Role of chest CT in patients with negative chest x-rays referred for hepatic colorectal metastases. Ann Surg Oncol 1998; 5:9–15. 24 Quyn AJ, Matthews A, Daniel T, Amin AI, Yalamarthi S. The clinical significance of radiologically detected indeterminate pulmonary nodules in colorectal cancer. Colorectal Dis 2012; 14:828–831. 25 Hogan J, O’Rourke C, Duff G, Burton M, Kelly N, Burke J, Coffey JC. Preoperative staging CT thorax in patients with colorectal cancer: its clinical importance. Dis Colon Rectum 2014; 57:1260–1266. 26 Parnaby CN, Bailey W, Balasingam A, Beckert L, Eglinton T, Fife J, et al. Pulmonary staging in colorectal cancer: a review. Colorectal Dis 2012; 14:660–670. 27 Kim CH, Huh JW, Kim HR, Kim YJ. Indeterminate pulmonary nodules in colorectal cancer: follow-up guidelines based on a risk predictive model. Ann Surg 2014. [Epub ahead of print].
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
Introduction
Over 40 000 new cases of colorectal cancer (CRC) are diagnosed annually in the UK [1], of which 30% are classified as metastatic at presentation [2]. The National Institute of Clinical Excellence (NICE) in the UK recommends routine staging computed tomography (CT) of the chest as well as the abdomen and pelvis to establish the disease burden and guide treatment plans within a multidisciplinary setting [3]. This diagnostic approach, however, has been questioned because of the fact that pulmonary metastases are uncommon in patients without intra-abdominal metastases [4]. The colon is drained by the portal venous system and, thus, the liver is generally European Journal of Gastroenterology & Hepatology 2015, 27:467–470 Keywords: colorectal cancer, computed tomography thorax, pulmonary metastases a Department of Colorectal Surgery, Whittington Health NHS Trust, University College London, London, bPhysiology Unit, St Marks Hospital, Harrow and c Academic Surgical Unit, The Royal London Hospital, Bart’s Health NHS Trust, London, UK
Correspondence to Alexander Hotouras, BSc, MSc, MRCS (Eng), MD(Res), Academic Surgical Unit, Barts Health NHS Trust, National Centre for Bowel Research and Surgical Innovation, Barts and the London School of Medicine and Dentistry, 1st Floor, Abernethy Building, 2 Newark Street, London E1 2AT, UK Tel: + 44 207 882 8751; e-mail: [email protected] Received 2 December 2014 Accepted 20 January 2015
the first site of CRC metastasis, except in distal rectal disease, when metastatic spread can bypass the portal system and directly enter the systemic circulation through the inferior and middle rectal veins [5]. It is reported that 10–15% of CRC metastases are to the lung and, of these, only 10% occur in the absence of hepatic spread [6], with the incidence being double in rectal cancer compared with colon cancer [7]. Recent studies have advocated that routine chest CT should not be a part of the routine preoperative staging investigations for CRC [6,8]. The rationale for this approach includes the low incidence of pulmonary metastasis, the minimal influence on treatment plan and the risk associated with increased radiation exposure [6]. The main objective of this study was to evaluate the clinical utility of preoperative chest CT examination for patients diagnosed with CRC in the absence of liver metastases and the impact of the clinical findings on subsequent treatment plans. Patients and methods
All patients with a histological diagnosis of CRC in our institution between December 2006 and December 2013 were identified from a prospectively recorded departmental database. This included information such as patient demographics, date of diagnosis, stage of disease at
0954-691X Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
DOI: 10.1097/MEG.0000000000000315
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
467
468
April 2015 • Volume 27 • Number 4
European Journal of Gastroenterology & Hepatology
Table 1. Patient demographics, colorectal tumour location and staging Age Median Range Sex Female Male Tumour sites [n (%)] Rectum Rectosigmoid Sigmoid Descending colon Transverse colon Ascending colon Caecum Disease stage Stage 1 Stage 2 Stage 3 Stage 4
73 24–96 175 183 113 45 69 17 16 39 59
(31.6) (12.6) (19.3) (4.7) (4.5) (10.9) (16.5)
61 129 156 12
presentation and treatment plans following discussion in a multidisciplinary team (MDT) meeting. All patients diagnosed with a colorectal malignancy underwent preoperative contrast or noncontrast (if contrast was contraindicated) CT of the chest, abdomen and pelvis. Imaging was obtained using a Siemens Somatom Helical CT scanner (Siemens, Germany). Volumetric imaging of the thorax, abdomen and pelvis was taken following intravenous contrast (unless contraindicated) of 90 ml Omnipaque 200 at 2 ml/s and oral contrast of 300 ml of 1.5% Gastrografin. Slice reconstructions of 3–5 mm thickness were created. Lung lesions were radiologically classified as benign, intermediate or malignant. All images were reviewed and reported by two radiologists in the colorectal MDT meeting with the results prospectively recorded on our departmental database. Pulmonary lesions that could not be defined as either benign or malignant were termed ‘intermediate nodules’. Patients with intermediate lung nodules underwent follow-up CT scan examination of the chest within 6 months. They were considered benign if subsequent follow-up scans, for at least 1 year, showed no signs of malignant changes or growth. Conversely, lesions that increased in size or number were considered malignant and confirmation was sought from histological or cytological examination. Patients with non-CRCs such as appendiceal cancer, neuroendocrine tumours, anal squamous cell carcinomas and gastrointestinal stromal tumours were excluded from this study. Results
A total of 574 patients were discussed in the CRC MDT during the study period. Two hundred and sixteen patients were excluded from this study as they did not fulfil the selection criteria (n = 131 had liver metastases, n = 85 had non-CRC malignancies). The remaining 358 patients (183 male, 175 female, median age 70 years, range 24–96 years) were included in the analysis. Table 1 summarizes patient demographics and tumour characteristics. The cohort included patients with adrenal (n = 2), kidney (n = 1), peritoneal (n = 2), mesenteric (n = 1) and bone (n = 1) metastases.
There were 292 (81.6%) patients without any radiological evidence of pulmonary nodules (Fig. 1). Sixty-one (17.0%) patients had pulmonary nodules classified as indeterminate on initial CT assessment. On serial imaging, one of these patients was deemed to have pulmonary metastatic disease, whereas a further two patients were diagnosed with incidental primary lung cancers. The remaining 58 patients were deemed to have benign lesions. Finally, five patients were initially diagnosed with pulmonary metastases on initial review of chest CT imaging. In one case these lesions were subsequently reclassified as benign. In total, 7/358 (2.0%) patients had pulmonary metastases or a primary lung malignancy in the absence of liver metastases. The median age of patients (three women, two men) with pulmonary metastases was 81 years (range 61–87 years). The TNM staging of the CRC in each one of these cases is shown in Table 2. All tumours were located either in the rectum or the rectosigmoid junction. Three patients had metastases to other nonhepatic sites that included nonregional lymph nodes (n = 3) and the adrenal glands (n = 1). The median number of lung nodules in each patient was nine (range 2–10) and the median maximum size of the nodules was 13 mm (range 5–26 mm). None of these patients underwent pulmonary metastasectomy. Four patients underwent bowel resection and received chemoradiotherapy. Comparatively, the median age of patients (31 women, 27men) with intermediate nodules was 74 years (range 40–89 years). Thirty (52%) patients had rectal or rectosigmoid disease, of these, 16 (53%) had positive local lymph nodes. The median number of lung nodules in each patient was 1 (range 1–5) and the median maximum size of the nodules was 3 mm (range 1–12 mm). Discussion
This study has demonstrated that the incidence of pulmonary metastases for patients with CRC in the absence of hepatic metastases is very low. All five patients in this study with pulmonary metastases had either rectal or rectosigmoid primary tumours. Thus, the prevalence of pulmonary metastases for patients with rectal cancer was 2.7% (3/113), whereas in the rectosigmoid cancer cohort it was 4.4% (2/45). All these patients had a T3 or T4 cancer with regional or nonregional lymph node involvement. This finding is in keeping with the work of Kim et al. [4], who retrospectively assessed 319 CRC patients without liver metastases and a negative chest radiograph (CXR). Kim et al. [4] reported that lymph node involvement is a risk factor for pulmonary metastatic disease and thus concluded that chest CT should not be routinely offered to patients without evidence of liver or lymph node spread. Brent et al. [9] and Christoffersen et al. [10] have also published similar findings. During our study period, 61 patients were initially reported to have intermediate pulmonary nodules as demonstrated by CT examination of the chest; however, one (1.6%) of these patients was later diagnosed with metastatic disease on surveillance CT imaging. Conversely, 1/5 (20%) patients who were initially diagnosed with pulmonary metastases was later downgraded to a benign lesion on review of interval chest CT examination.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
The diagnostic yield of preoperative staging CT Yongue et al.
www.eurojgh.com
469
Colorectal cancer patients with no hepatic metastasis (n = 358, 100%)
Clear lungs on initial staging CT chest (n = 292, 81.6%)
Intermediate lung nodules on initial CT chest (n = 61, 17.0%)
Likely lung metastasis on initial CT chest (n = 5, 1.4%)
Follow-up CT chest : Benign nodules (n = 58, 95.1%) Pulmonary metastases (n = 1, 1.6%) Primary lung cancer (n = 2, 3.3%)
Follow-up CT chest: Benign nodules (n = 1, 20.0%) Pulmonary metastases (n = 4, 80.0%)
Fig. 1. Study flow chart demonstrating radiological diagnosis of pulmonary nodules on computed tomography (CT).
Table 2. TNM staging and pulmonary nodule characteristics in patients with pulmonary metastases (n = 5) Locations
Disease stage
Rectal Rectosigmoid Rectal Rectosigmoid Rectal
T4N2M1b T4N1M1a T3N1M1a T4N0M1b T3N0M1b
Number of lung nodules
Maximum size of nodules (mm)
10 2 9 10 3
5 9 13 26 19
It should be noted that although all patients in our study had treatment for their primary malignancy, none had surgical management of their pulmonary metastatic disease. This is because none of these individuals met the criteria for metastasectomy, which include absence of extrathoracic disease (or extrathoracic disease that is resectable), pulmonary disease that is completely resectable, or adequate cardiopulmonary reserve [11–13]. A further two (0.5%) patients with intermediate lesions were diagnosed with synchronous primary malignancies on rereview of preoperative chest CT imaging. It has been reported that the incidence of synchronous lung and CRC primary malignancies is less than 0.6% [14], which is in keeping with the results of our study. However, identification of this small cohort is significant as identification of a pulmonary lesion as a primary lung cancer rather than as a metastatic CRC significantly changes patient management both in terms of chemotherapy and in terms of surgical options to resect pulmonary disease. Despite the limitations in diagnostic accuracy outlined above, CT has gradually become the preferred modality of imaging the chest because of its high sensitivity. CXR has a sensitivity of only 33% in comparison with 75% for CT [10,15,16], with chest CT examinations capable of reliably identifying nodules of ∼ 2–3 mm, whereas CXR can only demonstrate lesions that are 5–10 mm in size [17]. Although the use of chest CT imaging allows the detection of more subtle metastatic disease, there is also the
downside of increased incidental findings such as intermediate lung nodules as discussed above. These lesions are often too small to characterize with other forms of investigations such as fluorodeoxyglucose-PET scanning (53% of nodules < 1 cm are incorrectly identified) [18] or biopsy. Fine-needle biopsy/aspiration also carries its own morbidity including 24% risk of haemorrhage [19], 12% chance of pneumothorax [19] and a 5% chance of a 3-day hospital admission as well as costing £468 to perform [20]. Consequently, the CRC MDT process is forced to recommend follow-up imaging with serial high-resolution CT scans at 3–12-month intervals [21], exposing patients to significant additional radiation [21]. Furthermore, this approach also increases costs for the hospital or healthcare system and creates anxiety for the patient and their family. Yet, the majority of these lesions are benign as demonstrated by the Mayo Lung Trial, in which the incidence of cancer in intermediate lung nodules less than 9 mm was 0.69% [22]. Additionally, in a study of 100 CRC patients with liver metastases and negative CXR by Povoski et al. [23], the positive predictive value of chest CT was only 36% with a positive yield of only 4%. Consequently, it can be argued that CT imaging of the chest should not be a part of routine preoperative staging for CRC patients and instead should only be considered for patients at high risk for pulmonary metastases. These thoughts have been echoed in studies by Quyn et al. [24] and Hogan et al. [25], who demonstrated that intermediate nodules in patients with small localized primary CRC (i.e. stage T1/T2 or negative lymph nodes) did not progress into metastases. Furthermore, MDT meetings should consider the value of follow-up CT imaging or other more invasive investigations and whether management options would change in the rare scenario that a pulmonary nodule is proven to be malignant [26]. Perhaps as Kim et al. [27] suggest, a model to predict the progression of intermediate pulmonary nodules would be of benefit, first in deciding which patients are at risk for
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
470
European Journal of Gastroenterology & Hepatology
pulmonary metastases and thus requiring a chest CT and second to decide whether further investigations are necessitated should a nodule be found. Conclusion
The incidence of pulmonary metastases for patients with CRC in the absence of hepatic metastases is extremely low. As a result, the clinical utility of preoperative chest CT examination for all patients undergoing surgery for CRC is questionable. Prospective clinical trials are necessary to determine whether this preoperative investigation should be restricted to a subgroup of patients with an increased risk of pulmonary metastases. Acknowledgements Conflicts of interest
There are no conflicts of interest. References 1 Cancer Research UK. Bowel cancer incidence statistics; 2014. Available at: http://www.cancerresearchuk.org/cancer-info/cancerstats/types/bowel/inci dence/uk-bowel-cancer-incidence-statistics [Accessed 1 November 2014]. 2 Midgley R, Kerr D. Colorectal cancer. Lancet 1999; 353:391–399. 3 National Institute for Health and Clinical Excellence. NICE clinical guideline 131: The diagnosis and management of colorectal cancer. London: National Institute for Health and Clinical Excellence; 2011. Available at: http://www.nice.org.uk/nicemedia/live/13597/56998/56998.pdf. 4 Kim HY, Lee SJ, Lee G, Song L, Kim SA, Kim JY, et al. Should preoperative chest CT be recommended to all colon cancer patients? Ann Surg 2014; 259:323–328. 5 McCormack PM, Attiyeh FF. Resected pulmonary metastases from colorectal cancer. Dis Colon Rectum 1979; 22:553–556. 6 Pihl E, Hughes ES, McDermott FT, Johnson WR, Katrivessis H. Lung recurrence after curative surgery for colorectal cancer. Dis Colon Rectum 1987; 30:417–419. 7 Tan KK, Lopes Gde L Jr, Sim R. How uncommon are isolated lung metastases in colorectal cancer? A review from database of 754 patients over 4 years. J Gastrointest Surg 2009; 13:642–648. 8 Grossmann I, Avenarius JK, Mastboom WJ, Klaase JM. Preoperative staging with chest CT in patients with colorectal carcinoma: not as a routine procedure. Ann Surg Oncol 2010; 17:2045–2050. 9 Brent A, Talbot R, Coyne J, Nash G. Should indeterminate lung lesions reported on staging CT scans influence the management of patients with colorectal cancer? Colorectal Dis 2007; 9:816–818. 10 Christoffersen MW, Bulut O, Jess P. The diagnostic value of indeterminate lung lesions on staging chest computed tomographies in patients with colorectal cancer. Dan Med Bull 2010; 57:A4093.
April 2015 • Volume 27 • Number 4
11 Greelish JP, Friedberg JS. Secondary pulmonary malignancy. Surg Clin North Am 2000; 80:633–657. 12 Pfannschmidt J, Klode J, Muley T, Dienemann H, Hoffmann H. Nodal involvement at the time of pulmonary metastasectomy: experiences in 245 patients. Ann Thorac Surg 2006; 81:448–454. 13 Inoue M, Ohta M, Iuchi K, Matsumura A, Ideguchi K, Yasumitsu T, et al. Thoracic Surgery Study Group of Osaka University. Benefits of surgery for patients with pulmonary metastases from colorectal carcinoma. Ann Thorac Surg 2004; 78:238–244. 14 Evans HS, Møller H, Robinson D, Lewis CM, Bell CM, Hodgson SV. The risk of subsequent primary cancers after colorectal cancer in southeast England. Gut 2002; 50:647–652. 15 Hwang M, Jung S, Kim J, Shim M. Diagnostic usefulness of routinely performed preoperative chest computed tomography for colorectal cancer. J Korea Soc Coloproctol 2010; 26:211–216. 16 McIntosh J, Sylvester PA, Virjee J, Callaway M, Thomas MG. Pulmonary staging in colorectal cancer – is computerised tomography the answer? Ann R Coll Surg Engl 2005; 87:331–333. 17 Schaner EG, Chang AE, Doppman JL, Conkle DM, Flye MW, Rosenberg SA. Comparison of computed and conventional whole lung tomography in detecting pulmonary nodules: a prospective radiologic–pathologic study. Am J Roentgenol 1978; 131:51–54. 18 Bryant AS, Cerfolio RJ. The maximum standardized uptake values on integrated FDG-PET/CT is useful in differentiating benign from malignant pulmonary nodules. Ann Thorac Surg 2006; 82:1016–1020. 19 Schneider T, Puderbach M, Kunz J, Bischof A, Giesel FL, Dienemann H, et al. Simultaneous computed tomography-guided biopsy and radiofrequency ablation of solitary pulmonary malignancy in high-risk patients. Respiration 2012; 84:501–508. 20 Baldwin DR, Eaton T, Kolbe J, Christmas T, Milne D, Mercer J, et al. Management of solitary pulmonary nodules: how do thoracic computed tomography and guided fine needle biopsy influence clinical decisions? Thorax 2002; 57:817–822. 21 Kernstine KH, Grannis FW Jr, Rotter AJ. Is there a role for PET in the evaluation of subcentimeter pulmonary nodules? Semin Thorac Cardiovasc Surg 2005; 17:110–114. 22 Swensen SJ, Jett JR, Hartman TE, Midthun DE, Mandrekar SJ, Hillman SL, et al. CT screening for lung cancer: five-year prospective experience. Radiology 2005; 235:259–265. 23 Povoski SP, Fong Y, Sgouros SC, Kemeny NE, Downey RJ, Blumgart LH. Role of chest CT in patients with negative chest x-rays referred for hepatic colorectal metastases. Ann Surg Oncol 1998; 5:9–15. 24 Quyn AJ, Matthews A, Daniel T, Amin AI, Yalamarthi S. The clinical significance of radiologically detected indeterminate pulmonary nodules in colorectal cancer. Colorectal Dis 2012; 14:828–831. 25 Hogan J, O’Rourke C, Duff G, Burton M, Kelly N, Burke J, Coffey JC. Preoperative staging CT thorax in patients with colorectal cancer: its clinical importance. Dis Colon Rectum 2014; 57:1260–1266. 26 Parnaby CN, Bailey W, Balasingam A, Beckert L, Eglinton T, Fife J, et al. Pulmonary staging in colorectal cancer: a review. Colorectal Dis 2012; 14:660–670. 27 Kim CH, Huh JW, Kim HR, Kim YJ. Indeterminate pulmonary nodules in colorectal cancer: follow-up guidelines based on a risk predictive model. Ann Surg 2014. [Epub ahead of print].
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
