Secondary Logo

Journal Logo


Endoscopy or Radiology in Nonagenarians?

Morini, Sergio MD; Zullo, Angelo MD; Hassan, Cesare MD

Author Information
Journal of Clinical Gastroenterology: April 2007 - Volume 41 - Issue 4 - p 343-344
doi: 10.1097/

Life expectancy has dramatically increased in the last century, shifting a substantial proportion of the general population toward the older ages. It has been estimated that over-60 people will represent 40% and 30% of the European and American population by 2050.1 A quite similar 2-fold rise may be expected for nonagenarian people. Unfortunately, age is a major risk factor for cancer development, and malignancy is becoming a major cause of morbidity and mortality in western societies. Among all the cancer localizations, colon cancer is the most incident in the Europe and the second or third in the different American countries.2 It is well-known that age is the most crucial risk factor for colon cancer development. In particular, an exponential rise in incidence has been consistently reported after age 50, so that a prevalence greater than 2% has been reported in very elderly people.3 Screening in such people is generally not advised, because of the high risk/benefit ratio. For this reason, the main indication for a colonic study in this subgroup of population will be the occurrence of symptoms. Among the latter, the alarm symptoms—rectal bleeding and sideropenic anemia—have been reported to increase the possibility of cancer detection by 10-fold.4 Therefore, up to 20% of symptomatic nonagenarian patients will presumably harbor a cancer in their colon.

When dealing with such a high prevalence of malignant disease, sensitivity of the diagnostic test is essential to reduce the possibility of false negative results. There is no doubt that colonoscopy is by far the most accurate procedure for polyp and cancer detection. However, its clinical efficacy in very elderly people could be hampered by the following reasons: (1) the preventive role of polypectomy is greatly reduced by the limited life expectancy; (2) the overall sensitivity could be reduced by a higher rate of incomplete examinations; (3) the risk of complications for diagnostic and operative colonoscopy could be substantially increased by the frequent presence of comorbidities. Taking all these points together, it is clear that the risk/benefit ratio of colonoscopy in nonagenarian patients should be fully addressed.

On the other hand, radiology would seem as a promising alternative in this subgroup of people. Although sensitivity of either CT-colonography or double-contrast enema for polyps is undoubtedly much poorer than that of colonoscopy, a very high sensitivity for cancer has been reported.5 Moreover, it has been recently shown that the cardiovascular side-effects of radiologic procedures are substantially less than those reported with endoscopy.6 Furthermore, CT-colonography with minimal preparation or fecal tagging are expected to dramatically reduce the false-negative for cancer, a sensitivity of 100% being recently reported.7

Therefore, if we agree that colon cancer, and not the advanced adenoma, is the only valuable target in nonagenarian patients, we are left with an unsolving dilemma: endoscopy or radiology?

The paper by Weiss et al8 published on this issue wholly addressed the endoscopic issues in nonagenarian patients. As expected, colon cancer prevalence in this population is very high, being nearly 20%, having most of the procedures being performed for symptomatic indications. Colonoscopy seemed to be extremely safe in the nonagenarian patients. Although roughly half of them had important comorbidities, no death or hospitalization was reported in the following 48 hours. Therefore, a substantial rate of clinically relevant cardiovascular complications can be reasonably excluded. However, it may not be excluded that the avoidance of major complications is also related to the high rate of failed colonoscopies. We could presume that the endoscopist has been more prudential when facing a nonagenarian patient than a younger control, being aware that a complication requiring surgery—such as perforation—could have had disaster effects on these high surgical-risk patients. Moreover, the relatively small sample size unveils the necessity of further larger studies to confirm this initial data. The second positive point is the possibility of predicting the patients in whom colonoscopy will be technically unsuccessful. Indeed, after a comprehensive statistical analysis including several clinical variables, a poor functional status of the patient seemed to be a strong predictor of a failed procedure. Interestingly, the assessment of such functional status is extremely easy, requiring only information regarding the basic daily activities such as dressing or walking. On the other hand, a disappointing result from this analysis is the high rate of incomplete colonoscopies, which is nearly 40%, whereas the rate reported in younger people is consistent with previous analysis.9 The clinical efficacy of colonoscopy in nonagenarian people may be expected to be heavily hampered by this high failure rate. Indeed, not to assess the right-colon in very elderly patients is much more dangerous than in relatively younger people because of the shifting of cancer localization toward the right colon. Consequently, if we assume that roughly 50% of the malignancies are localized in the right colon, we may assume that 9% of the patients with failed colonoscopies will harbor a right-side cancer, which means a 3.5% of missing rate for malignancy in the nonagenarian patients. If we admit that such rate is clinically relevant, we would need 39 radiologic examinations every 100 patients to identify most of them. On the other hand, if we simulate to perform an initial radiologic examination in the same population, it would come out that only 17 true positive and about 7 to 10 false positives would eventually need an endoscopic examination to confirm the diagnosis of cancer. Therefore, there would be a slight advantage for radiology in the clinical practice. Here is the strength of the present paper! The ability of the authors to identify a predictor for failed colonoscopy with an odds ratio of 5.6 strongly helps in rationalizing the access to diagnostic procedures. It indeed suggests reserving open-access colonoscopy only to those people with a good functional status, although radiology could be a better alternative in those with a poor status. Such an approach could reduce the necessity of a double examination to less than 20% of the population. Moreover, those with a good functional status may also be expected to gain more benefit from the identification of early malignant lesions, which may be treated with minimally invasive surgical approach or with palliative endoscopic procedures.

A further result of the present paper is the clear evidence that a bad preparation is a much more frequent cause of failed colonoscopy in the nonagenarians as compared to the younger controls. This is to be explained not only with the age of the patient, but also with the associated mental conditions and the related drug therapies. Nonagenarians were obviously more frequently hospitalized than younger controls. Colonic cleansing in the hospital may be more troublesome because of the more rigid timetable and the local setting. This should advice the necessity of a different bowel preparation in nonagenarian people, especially in those with poor general conditions.

Interestingly, the authors showed that a failed colonoscopy was associated with a tortuous colon, but not with diverticular disease. This suggests that, even from a clinical point of view, the “aging” of the colon is much more related with a progressive deterioration of the connective structures (ie, collagen deposition, degeneration of the neural components) than with the simple appearance of the submucosal diverticular herniation.

In conclusion, age, as a criterion, should not prevent anyone from potentially life-saving procedures. On the other hand, a more thorough clinical approach allows to identify some predictive factors to rationalize the access to such diagnostic tests. Weiss et al8 should be thanked for having shown that this is a real possibility for open-access colonoscopy.


1. United Nations. World Population Prospects: the 1992 Revision. New York, USA: United Nations; 1993.
2. Boyle P, Autier P, Bartelink H, et al. European code against cancer and scientific justification: third version (2003). Ann Oncol. 2003;14:973–1005.
3. Ries LA, Kosary CL, Hankey BF, et al. SEER Cancer Statistics Review, 1973-1994. Bethesda, MD: National Institutes of Health, National Cancer Institute; 1997. NIH Publication No. 97-2789.
4. Norrelund N, Norrelund H. Colorectal cancer and polyps in patients aged 40 years and over who consult a GP with rectal bleeding. Fam Pract. 1996;13:160–165.
5. Mulhall BP, Veerappan GR, Jackson JL. Meta-analysis: computed tomographic colonography. Ann Intern Med. 2005;142:635–650.
6. Taylor SA, Halligan S, O'Donnell C, et al. Cardiovascular effects at multi-detector row CT colonography compared with those at conventional endoscopy of the colon. Radiology. 2003;229:782–790.
7. Robinson P, Burnett H, Nicholson DA. The use of minimal preparation computed tomography for the primary investigation of colon cancer in frail or elderly patients. Clin Radiol. 2002;57:389–392.
8. Schmilovitz-Weiss H, Weiss A, Boaz M, et al. Predictors of failed colonoscopy in nonagenarians: a single-center experience. J Clin Gastroenterology. 2007;41:388–393.
9. Harewood GC, Sharma VK, de Garmo P. Impact of colonoscopy preparation quality on detection of suspected colonic neoplasia. Gastrointest Endosc. 2003;58:76–79.
© 2007 Lippincott Williams & Wilkins, Inc.