Some issues in the article by Iyer et al. [1] are of particular interest to us. The article, which is an overview of imaging in the workup of patients with colorectal cancer, seems to be missing some important facts. On the second page, they say that neoadjuvant chemo- and radiotherapy is increasingly administered in patients with rectal cancer. They also say that local recurrence rates have been reduced by 50% by such neoadjuvant therapy. This information is confusing because the reduction in local recurrence rates is related more to preoperative radiotherapy [2] than to chemotherapy. Furthermore, the 30–50% local recurrence rates they give for stages II and III rectal cancer are not true if surgery is performed according to the principle of total mesorectal excision [3], under which local recurrence rates of 5–15% are more likely. Also, total mesorectal excision is probably one of the most important improvements in rectal cancer treatment during recent years, so it is surprising that this is not mentioned in the article.

Helical CT has developed significantly during the last few years along with multidetector technology; its role in colorectal cancer staging must therefore be regarded as not evaluated using modern scanners. For this reason, it is doubtful that it is relevant to cite a review article [4] reporting 50–70% sensitivity for assessing local invasion with CT as Iyer et al. [1] state on page 6. They also state that adequate luminal distention is essential for rectal imaging with CT and that this may be achieved with water as a contrast agent. In our experience, endoluminal water has little impact on staging rectal cancer with CT. CT colonography for the detection of tumors is performed with endoluminal air, but such bowel distention will have little effect on assessing extramural spread.

The authors [1] also state that staging rectal cancer with MRI using an endorectal coil can be performed with an accuracy of approximately 80%; the articles cited to support this position are also taken from a review and not from original work; the original work is now 10 years old and is based on a small series of patients [5]. In recent years, the same accuracy has been reported with MRI using external array coils alone [6–8]. This fact is not mentioned in the article, and neither are the prognostic implications of extramural extension of tumor growth or extension to the potential surgical resection margin (the mesorectal fascia) and neighboring organs, which are central to assessing the role of MRI and which are critical in selecting patients with rectal cancer for treatment.

Iyer et al. [1] stated that nodal size greater than 1 cm remains the primary criterion for assessing nodal metastases on cross-sectional imaging. For perirectal lymph node metastases, a nodal size of 5 mm has been regarded as a more appropriate criterion [9], although size is not sufficient for diagnosis by itself. In our experience, the architecture of a lymph node on T2-weighted thin-slice MRI is more important, as is the echogenicity of a lymph node when it is being evaluated on sonography.

In conclusion, some recent developments with significant implications for colorectal cancer imaging are missing in the article. The reference list lacks some important developments and should be updated. Furthermore, the figures for recurrence rates of rectal cancer are not those that can be obtained today.