We appreciate the insightful comments from Herron and colleagues and the opportunity to expand on a few points.
Our retrospective study1 demonstrates that the prevalence of malignancy among patients with diverticulitis diagnosed on high-resolution (64-slice) CT scan was 1.4%, similar to that of average-risk population.2
The data period was chosen based on the time when high-resolution CT came into widespread use. Since then, 64-slice CT has remained the practice standard at our centre as well as at many others, thus maintaining the applicability of our results.
Ideally, contrast-enhanced CT should be used when acute diverticulitis is suspected, but this is often limited by contrast allergies and impaired renal function in practice. Although the quality of the studies are affected by lack of contrast, radiologists can still make the appropriate interpretations based on the clinical context provided by requesting clinician. We fear that excluding patients who were unable to receive contrast would have introduced substantial selection bias. To ensure that the included patients had findings consistent with acute diverticulitis, all of the CT scans in this study were retrospectively reviewed by a single radiologist specialized in abdominal imaging.
The premalignant findings in 23 patients were not specifically compared with the CT scans in this study. Unlike CT colonography, which invovles bowel preparation to rid of fecal matter and enteral contrast to distend the colon as well as to enhance the appearance of polyps, plain CT scan is not designed to assess intraluminal pathology.
None of the 23 patients with premalignant findings had undergone colorectal cancer (CRC) screening in the form of endoscopy (it was unknown if they had previous screening in the form of annual fecal occult blood tests) despite a mean age of 61.5 years owing to opportunistic screening being the primary strategy at that time. This finding, together with the fact that all 4 patients with malignancy also did not have previous CRC screening, underscores the importance of age-appropriate screening. We therefore recommend endoscopy-naive patients undergo follow-up endoscopic evaluation. On the other hand, if a patient already had high-quality colonoscopy with no evidence of polyp within a reasonable time before diverticulitis was diagnosed, it is conceivable that a repeat colonoscopy would be redundant.
One of the strengths of this study is the use of a provincial cancer registry to capture any CRC that may have risen since the diagnosis of acute diverticulitis. Absence of additional cases of CRC in the registry among those who did not have follow-up endoscopy lends support to the idea that not all patients with acute diverticulitis require follow-up endoscopy to rule out underlying malignancy.
Based on the results of our study, we recommend selective endoscopic evaluation in the following patient populations after a diagnosis of acute diverticulitis on high-resolution CT scan: patients ≥ 50 years of age who are due for CRC screening/polyp surveillance in the form of colonoscopy based on recommended intervals,3 and those with suspicious CT findings, such as a mass lesion with obstruction.