Should preoperative chest computed tomography be performed in all patients with colorectal cancer?

This study evaluated the accuracy of chest computed tomography (CCT) in the preoperative staging of patients with potentially resectable colorectal cancer (CRC). An attempt was made to determine whether CCT has a significant influence on management and clinical outcome.

A review was conducted of all preoperative CCTs and X-rays performed in patients submitted to elective resection of CRC at our institution between 2005 and 2012. All scans were revised by an independent radiologist who was unaware of the circumstances of each case. The findings were classified as benign, malignant or indeterminate. Patients were followed for at least 12 months after surgery to assess the clinical evolution of any lesion found on CCT and the oncological outcome.

Two hundred and twenty-three patients were included. The CCT showed normal or benign findings in 157 (70.4%) patients, a malignant lesion in 17 (7.6%), and an indeterminate lung lesion (ILL) in 49 (22%). Of the 30 patients with proven lung metastases, a plain X-ray detected lesions in only 11 (36.7%) patients. During the postoperative follow-up, 14 (28.6%) of the 49 patients with ILL demonstrated malignant progression of the lung lesions. Among all 223 patients, only six (2.7%) underwent lung resection.

CCT is superior to plain X-ray for the detection of lung metastases. Despite the medical and financial implications of preoperative CCT, only a small number of patients will eventually undergo lung resection. We call into question the role of routine CCT in the staging of patients with CRC. A more selective approach, reserving CT for patients at high risk of lung metastases, is suggested.