Australian gastroenterologists have for the first time reported ‘real world’ data to benchmark the accuracy and quality of colonoscopy for excluding colorectal cancer (CRC).

Their prospective cohort study investigated the number of  post colonoscopy colorectal cancer (PCCRC) cases occurring after an index colonoscopy at the Canberra Hospital.

PCCRCs are potentially the most important markers of colonoscopy quality, and are defined as a colorectal cancer arising within 6-60 months of a negative colonoscopy recommended for any indication. This is different to interval colorectal cancer, a term reserved for cancers arising within a defined period after a negative examination performed primarily for screening.

In the study, gastroenterologists prospectively followed a cohort of 7818 people who underwent colonoscopy at the Canberra Hospital between 2001 and 2008. Colonoscopy results were then correlated with cancer histories from regional cancer registries and hospital records in the ACT as well as those in NSW to capture those patients who had moved between states during the minimum five-year follow up period.

Among 7818 patients who underwent colonoscopies, 537 were diagnosed with CRC, and some 15 individuals were found to have developed PCCRC.

The PCCRC incidence rate was 0.4/1000 person-years while the 5-year PCCRC risk was 0.192%.

The data should be shared with patients in the informed consent process before screening and diagnostic colonoscopy, said lead study investigator Dr Doug Taupin, senior staff specialist in Gastroenterology and Hepatology at The Canberra Hospital.

“This is the first time this has been done in Australia and with a methodology where we have tried to avoid linkage failure,” Dr Taupin told the limbic.

Linkage failure was a factor that contributed to the wide variance of reported PCCRC rates, he explained.

The study also revealed a number of factors associated with these ‘missed’ cancers. Patients with PCCRC tended to have a presence of diverticulitis disease at the index colonoscopy, poor bowel preparation, or a previous history of polyps.

But while these factors were statistically significant, Dr Taupin added that the relative risk for diverticulitis and PCCRC was in the order of 3-4.

“This means the number needed to treat to prevent one post colonoscopy colorectal cancer with the presence of diverticular disease at the index colonoscopy would be 333. In order to achieve a halving of the PCCRC ratio, through three-yearly surveillance colonoscopy for example, you would have to treat 666 patients with diverticular disease to the new schedule to prevent one PCCRC.”

Describing the findings as “meaningful” Dr Taupin said he’s not sure that yet translates to practice.

“It might be that if we found diverticular disease and lets say  bowel preparation was poor, we might say come back for a repeat colonoscopy”.

There was also a higher frequency of right-sided PCCRC, which may arise from CRC being more biologically aggressive or from right-sided precursor lesions being more likely to be undetected or incompletely removed at prior examinations Dr Taupin suggested.

And while it didn’t reach statistical significance, there was a trend for PCCRC to be more frequent where a trainee had performed the index colonoscopy.

“An accurate PCCRC rate means we’re now aware of quality indicators, said Dr Taupin.

“As colonoscopies get better, as we do better with bowel preparation, as we become more skilful with the endoscope and get better with polypectomies,  I would suggest that when we do the same review in five years’ time with the current database we will see a reduction in that rate.”

The findings are published in BMJ Open.