Colonoscopy in Nonagenarians Is Safe and May Be Associated With Clinical Benefit

Asher Shafrir, MD; Benjamin Koslowsky, MD; Dov Wengrower, MD; Eran Goldin, MD; Dan M. Livovsky, MD


J Am Geriatr Soc. 2019;67(6):1158-1163. 

In This Article


As individuals live longer, invasive procedures such as colonoscopy are being performed with increased frequency and on older patients. Nonetheless, the data regarding the utility of this test in this population are very limited. To the best of our knowledge, this retrospective case-control study of colonoscopy in nonagenarians compared with octogenarians represents the largest and eldest cohort published to date. We aimed to describe the performance, safety, outcomes, and impact of significant findings (ie, CRC) on the subsequent clinical management and prognosis of these patients. Only one study by Cha et al[9] comparing 76 nonagenarians with 140 75- to 79-year-old patients reported that colonoscopy in this age group is safe and has a high yield; yet they did not report the clinical impact of findings during the procedure in management or prognosis.

Similar to Cha et al,[9] we found that increasing age alters the indications for colonoscopy because nonagenarians were more likely to undergo colonoscopy due to rectal bleeding, anemia, and following an episode of volvulus, and less likely for polyp surveillance. As opposed to other studies,[12] we found that most of the nonagenarians were successfully and safely sedated and had a low procedure-related complication rate. Nonagenarians were more likely to be sedated by fentanyl and propofol, to receive general anesthesia, and altogether needed lower medication doses compared with octogenarians. These findings are in line with previous studies.[6,9,13] Although propofol sedation was shown to be safe in octogenarians,[14] we are presenting the first study showing that propofol sedation is also safe in nonagenarians and that low doses are generally sufficient.

Polyethylene glycol preparation is considered safe in the general geriatric population.[15] Indeed we showed it to be safe also in nonagenarians; nevertheless, bowel cleansing quality was significantly lower than in the octogenarian population. Overall, 17.2% of nonagenarians had a very poor bowel preparation, leading to a low (78.9%) colonoscopy completion rate. A meta-analysis together with previous studies also showed poorer quality of preparation and lower completion rates in extremely old patients.[6,7,9] These findings raise the question of the need for a specialized and tailored preparation for nonagenarians.

Although adverse effects appear to be more common in older patients,[7] colonoscopy is still considered to be safe.[6,8] In our study, two severe adverse events occurred during colonoscopy in each group, representing 0.9% and 1.6% of all procedures. These relatively low rates allow us to consider colonoscopy to be a safe procedure even at these extreme ages. Five nonagenarians (3.9%) and one octogenarian died in the 30 days following the procedure (P = .027). None of these patients died due to a direct complication of the colonoscopy. Four of them died after undergoing urgent surgery for an obstruction or subobstructing mass. Altogether, these findings are in line with previous reports of procedure-related complications.[6–9]

It is well established that increasing age is associated with more numerous and more advanced adenomas.[6,15,16] Despite that, in our study nonagenarians were more likely to have advanced adenomas and a histologic diagnosis of adenocarcinoma; a surprising finding was that octogenarians had more polyps of any kind than nonagenarians. This may be attributed to nonagenarians' poorer preparation, but it may also be the result of different indications for the procedure. Whereas octogenarians undergo colonoscopy for many different indications, nonagenarians mostly undergo colonoscopy due to gastrointestinal bleeding or for decompression of a sigmoid volvulus, procedures where polyps may be overlooked. We found that age, in-hospital procedure, and anemia were the strongest predictors of CRC. Similar to previous reports,[17] changes of bowel movements, abdominal pain, diarrhea, or constipation were not associated with an increased risk of cancer. A total of 10 of 19 nonagenarian patients found to have cancer underwent surgery. We documented that nonagenarians who underwent surgical resection of a CRC had a survival benefit when compared with the patients who received palliative treatment alone. This can be also attributed to the fact that healthier nonagenarians are more likely to be suitable to undergo surgery. Nevertheless, the finding that surgery is feasible and has potential prognostic advantage is important. Overall and in line with previous reports,[18] patients requiring urgent surgery had bad outcomes, whereas age alone was not associated with increased postoperative complications.[19]

We decided to match nonagenarians to octogenarians because we aimed to compare more similar populations who underwent the colonoscopy due to potentially more comparable indications. Notwithstanding we show that although the age difference was relatively small, many significant differences were documented between the groups, and much can be learned by comparing these two special groups.

This study had a few limitations. Because this study was retrospective, the patients studied represented a selected group of patients fit to undergo colonoscopy. This population clearly does not represent nonagenarians as a whole. Also it is important to note that the sample is from a single center and country, further limiting the generalizability of our results. Additionally, due to our retrospective data, we were unable to extract information about comorbidities and chronic medications, and therefore we could not control for these important variables in our statistical analysis.

In conclusion, this is the largest and oldest cohort published to date showing the high yield, safety, and potential benefit of colonoscopy in the extremely older population.