Introduction

Colorectal cancer is an important component of the burden of disease in developed countries because it is widespread and has a high morbidity and mortality rate [1]. In Europe, the annual age-standardized incidence (world population) of colorectal cancer is between 20 and 45 per 100 000 among males and between 15 and 30 per 100 000 among females [2]. Incidence rates increase in a regular fashion with age [2][3][4]. “Screening, in the context of colorectal cancer, identifies individuals who are more likely to have colorectal cancer or adenomatous polyps from among those without signs or symptoms of the disease” [1]. Screening may detect cancers at an early stage. If detected at an early stage, treatment may be curative and improve prognosis. Screening may be acceptable to many patients and generally feasible in practice.

In November 1998, a multidisciplinary European expert panel convened in Lausanne, Switzerland, to discuss and develop criteria for the appropriate use of gastrointestinal endoscopy, a widely-used procedure, regarded as highly accurate and safe. The RAND appropriateness method was chosen for this purpose, because it allows the development of appropriateness criteria based on published evidence and supplemented by explicit expert opinion. A detailed description of the RANTD appropriateness method, including the literature search process [5], and of the whole process, as well as the global results of the panel [6], are published as separate articles in the issue of the Journal. The literature review was based on a systematic search of Medline, Embase and the Cochrane Library conducted up to the end of 1997 and completed with some key articles published in 1998. Updating and revision of the literature review is currently ongoing.

This article contains three parts: 1. the review of the literature that was used by the panelists to support their ratings of appropriateness of use of colonoscopic screening for colorectal cancer in asymptomatic patients without personal history of colorectal cancer or polyps; 2. an overview of the main panel results; 3. a summary of the published evidence and of the panel based appropriateness criteria.

1. Literature Review

A general description of the epidemiology of colorectal cancer can be found in the article on surveillance after curative intent resection of a colorectal cancer published in a joint article in this issue of the Journal [7].

Screening Strategies

Although the question being examined is the appropriateness of use of colonoscopy in screening for colorectal cancer, the effectiveness of other tests for early diagnosis of colorectal cancer has also been examined (e.g. Fecal Occult Blood Test). The following literature review section consists of two parts: screening in people at average or increased risk for colorectal cancer.

Persons with Average-Risk

Fecal Occult Blood Test (FOBT)

The rationale for this examination is based on the observation that the cancerous mucosa bleeds more than normal mucosa. As polyps also bleed, particularly if they are large, and are more common than cancers, FOBT will also detect bleeding from polyps. The sensitivity is increased and the specificity decreased with rehydrated tests. The sensitivity also increases with the number of stool samples studied [1].

Accurary of the Screening Test

A multicenter prospective study [8] analysed the accuracy of fecal occult blood screening for colorectal neoplasm and found it to be a poor indicator of the presence of neoplasm. Most cancers, and the vast majority of polyps, remained undetected. The sensitivity of Hemoccult® for colorectal cancer was 26 % and, for polyps greater then 1 cm, 11 - 20 %. The positive predictive value for cancer was about 5 - 8 %, although the study population comprised persons with higher-than-average risk for colorectal cancer.

Another study [9] in 8104 patients from a colorectal cancer screening programme analysed three different tests: Hemoccult II®, Hemoccult IISensa® and HemeSelect® with diet. Each patient received the three tests. No rehydration was done. The main results are presented in Table [1].

A guideline of the American College of Physicians [10][11] proposed techniques for FOBT and its interpretation. In spite of the large randomized controlled trials already carried out, no consensus exists concerning such important technical issues such as diet restriction, rehydration of slides, frequency of screening, strategy for evaluating those who tested positive. Stronger evidence is available for the following technical aspects of colonoscopy: a complete colonic evaluation should be done after a positive test result because the possibility of finding an important neoplasm is high enough to warrant a work-up. Colonoscopy or flexible sigmoidoscopy plus high-quality air-contrast barium enema are both suitable methods for complete colonic evaluation. Only moderate evidence exists as to whether a follow-up colonoscopy should be done in patients without high-risk lesions. An interval of five years before a control colonoscopy has been prosposed. Moderate evidence is available for surveillance with periodic colonoscopy if a high-risk lesion is found at baseline colonoscopy.

Efficacy of FOBT

The best evidence currently available concerning FOBT comes from large RCTs [12][13][14]. Weaker observational evidence supports the findings of these studies: three case-control studies have suggested a reduction in mortality from colorectal cancer through screening with FOBT (odds ratios of 0.4 - 0.9 indicated a protective effect, but bias cannot be excluded) [15][16][17]. In cohort studies, a population-based screening study in people aged 45 - 74 in France [18] reached a participation rate of 43.4 % for the FOBT. The FOBT positive rate was 2.8 % (2,020/71,307; non-rehydrated Hemoccult II) and 79.4 % of these 2,020 patients subsequently underwent a colonoscopy.

Another ongoing study [19] screened 36 034 asymptomatic individuals aged 50 - 80 with FOBT (Hemoccult II, with diet), followed by colonoscopy if positive, over a period of 14 years. The overall participation rate was 47.8 %. During follow-up, 115 colorectal cancers were diagnosed in positive patients with 82 % being early-stage and 18 % being late-stage (Dukes C, D) tumours. During the same period, 312 colorectal cancers were diagnosed outside of the screening study in symptomatic patients: 57 % had early-stage and 43 % late-stage colorectal cancer.

Randomized Controlled Trials

The Minnesota study [13] showed a 33 % reduction in mortality from colorectal cancer in the annually screened group as compared to the control group after 13 years of follow-up (but no reduction in the biennially screened group at this stage), whereas the Funen study [12] indicated that biennal screening reduced colorectal cancer mortality by 18 %, independently of age and gender, although without a significant reduction in overall mortality. In the Nottingham study [14], although overall mortality was not significantly different among the tested and control groups, patients who refused the first screening test showed a higher rate of mortality. No difference in mortality could be found in the Göteborg study [20]. In all but the Göteborg study, a significant reduction in the incidence of advanced-stage cancers (C and D) was observed among the screened groups. Further details are given in Tables [2] and [3]. Table [3] includes results from the Swedish study as well as results from the French cohort study.

Table [3] shows that rehydration increases and that a special diet decreases the number of cases detected (with a reverse effect on the positive predictive value [PPV] of the test [rehydration decreases the PPV and diet increases it]).

Participation in screening programmes was generally situated between 50 and 70 % [21][22][23]. Compliance was higher when the test was proposed by a GP (85 - 94.0 %) than when it was sent by post (26.0 - 34 %). Adherence to screening sigmoidoscopy after a positive FOBT ranged from 11 - 53 % [23]. A 38 % adherence rate by physicians to complete diagnostic evaluation in positive FOBT patients was reported [24].

Recommendations of Others

The American Gastroenterological Association (AGA) has proposed, together with other societies, guidelines for colorectal cancer screening: for people at average risk (asymptomatic, age ≥ 50 years, no other risk factors for colorectal cancer) FOBT should be proposed each year [1]. In patients tested positive, an accurate examination of the entire colon and rectum by colonoscopy is recommended. An alternative is double-contrast barium enema associated with flexible sigmoidoscopy. As the AGA recommendations did not include studies published after September 1996, the final results of the Funen and of the Nottingham study were not included. In 1996, the US Preventive Services Task Force proposed annual FOBT for all persons aged 50 and older, as one option for colorectal cancer screening [25]. In 1994, the Canadian Task Force on the periodic health examination concluded that there was not enough evidence to recommend FOBT as a screening test for colorectal cancer [26].

Follow-up Screening After a Negative Result at Complete Colorectal Evaluation

People with positive FOBT and subsequent negative colon examination, are at low short-term risk for colorectal cancer [27]. The period during which a patient will remain at low risk is uncertain; Ransohoff and Lang proposed to postpone follow-up screening for five years [11].

In a group of patients with positive FOBT who were referred for further evaluation, upper gastrointestinal lesions were identified more frequently than colonic lesions. The value of upper GI endoscopy in asymptomatic patients with positive FOBT in these circumstances is controversial [28]. Occult gastrointestinal bleeding can be detected in about half of patients with celiac sprue. Therefore, it might be worthwhile to suggest duodenal biopsies in patients with positive FOBT [29].

Negative Consequences of FOBT

The relatively low sensitivity of FOBT implies that a high proportion of colorectal cancers will be false negatives or not be detected at screening, thus giving a false sense of security to doctors and patients. In a 14-year cohort study of 27,466 patients screened, 13/21 (62 %) of patients with advanced cancers (Dukes C, D) tested negative initially [30].

The Serendipity Effect of Colonoscopy

Adenomas found through screening programmes with FOBT are likely to be detected because of the high prevalence of adenomas and the frequency of colonoscopy rather than because of the ability of FOBT to detect adenomas [26]. When discussing the Minnesota Colon Cancer Control Study, Lang and Ransohoff observed that “some of the benefit of FOBT screening may come from ‘chance’ selection of persons for colonoscopic examination because of the high positivity rate of FOBT that may occur for reasons other than a bleeding cancer or polyp” and concluded that up to 33 - 55 % of the mortality reduction could have resulted from random colonoscopy in false positive Hemoccult patients [31]. Bond responded that only 6 - 11 % of the observed reduction in colorectal cancer mortality could be explained by this mechanism [32].

To summarize, three randomized trials from three countries demonstrated that people screened with FOBT annually or biennially, and with subsequent colonoscopy in the event of positive testing, have colorectal cancer detected at an earlier stage and have a better survival rate from colorectal cancer. No study has, however, demonstrated to date any significant reduction in the overall mortality or in the incidence of colorectal cancer in the subjects screened.

Flexible Sigmoidoscopy

Flexible sigmoidoscopy allows a large proportion of polyps to be detected [33][34]. In several studies in asymptomatic individuals, the yield by flexible sigmoidoscopy for adenomas ranged from 6 - 26 % and for colorectal cancer from 0.2 - 2 % [35][36][37][38][39][40]. In two studies that examined both tests, 44 - 55 % of adenomas detected at colonoscopy were beyond the reach of sigmoidoscopy [36][41]. In an RCT, the sensitivity of flexible sigmoidoscopy to detect an adenoma or a colorectal cancer in the rectosigmoid region has been estimated to be 87 % and 85 %, respectively [42]. A prospective controlled study [39] analysed the efficacy of polypectomy by flexible sigmoidoscopy screening in 400 screened and 399 control patients from the general population (no exclusion of symptomatic patients or patients with a colorectal cancer) aged 50 - 59 during ten years of follow-up. In a pragmatic analysis, the population submitted to endoscopy showed significantly less cases of colorectal cancer (0/324) than the controls and non-attending individuals in the group screened (5/475).

Impact of Sigmoidoscopy Screening on Patient Outcome

Three case-control studies [43][44][45] suggest that sigmoidoscopy screening can reduce mortality from cancer of the distal colon and rectum. Selby [46] analysed evidence from a randomized trial of multiphase screening (Kaiser Multiphasic Evaluation Study) and stated that no conclusion could be drawn from this study as to whether screening sigmoidoscopy reduces mortality from colorectal cancer. Atkin conducted a long-term study [47] in patients who had adenomas removed by sigmoidoscopy up to 30 years previously and estimated that 80 % of rectal cancers had been prevented by adenoma removal. A prospective controlled study [39] with ten years of follow-up suggested a 75 % reduction in colorectal cancer incidence; the study sample was, however, too small to have sufficient power to yield a significant result.

It has been estimated that screening sigmoidoscopy should result in the detection of one or two carcinomas per 1000 patients with carcinomas being detected at an earlier stage [26].

Screening Intervals

A prospective study [48] of 217 patients with negative FOBT found a 1-year surveillance sigmoidoscopy to yield only 1 % more neoplasia as compared to the baseline examination. Selby [43] showed a long duration of the protective effect of polypectomy, suggesting an interval as great as ten years. This interval is also suggested by a controlled follow-up study in Norway [39].

A recent case-control study [45] found that polypectomy retained its efficacy over five to six years. Rex et al. [49] found an adenoma prevalence of 6 % but no colorectal cancer after two years in asymptomatic average-risk men aged ≥ 50 years with a high socio-economic status. He suggested a screening interval of five years after a negative examination.

Compliance

A survey [50] explored obstacles to compliance with screening sigmoidoscopy and found fear of pain to be the main barrier to compliance, and that clinician advice, a family member with a history of screening sigmoidoscopy and the perceived benefit of the test were associated with a higher degree of utilisation. Although low compliance (12 - 27 %) has been found in community-based interventions [38][51], much higher acceptance rates (87 - 95 %) have been observed in particular settings (City of Boston employees, US State Department personnel) [37][52].

Recommendations of Other Authors

According to the AGA recommendation, a screening sigmoidoscopy every five years should be proposed [1]. Polyps < 1 cm should be biopsied, and if adenomatous polyps or cancer are found, the patients should be offered colonoscopy as well; if large (> 1 cm) polyps are found, colonoscopy should be recommended. The Canadian Task Force on the periodic Health Examination considered, in 1994, that the evidence was insufficient to recommend screening with flexible sigmoidoscopy in people > 40 years (grade C recommendation) [26]. The US Preventive Services Task Force proposed sigmoidoscopy, without any interval being specified, as one option for colorectal cancer screening [25].

An alternative prosposal is to perform a single screening sigmoidoscopy at age 55. This rationale stems from the observation that the prevalence of distal adenomas increases rapidly after the age of 50 and stabilizes at age 60. Moreover, adenoma in the distal colon are considered to be indicators of lesions in the proximal colon [53]. A multicentre randomized trial [54] to evaluate “once-only” flexible sigmoidoscopy is currently in progress in the UK. With attendance rates of up to 74 %, preliminary results show a detection rate of 0.007 for colorectal cancer (higher that expected) and of 0.1 for adenomas.

Combined Fecal Occult Blood Test and Flexible Sigmoidoscopy

Screening including both tests has been recommended by the AGA group [1] and by the WHO Collaborating Center for the Prevention of Colorectal cancer [55]. FOBT once a year and sigmoidoscopy every three years [55] or five years [1] have been prosposed. Using indirect evidence from a mathematical model, Eddy [56] suggested that screening patients aged 50 - 75 years using FOBT plus flexible sigmoidoscopy every three to five years should reduce the risk of dying from colorectal cancer by 10 - 75 % depending on the screening option.

Double-Contrast Barium Enema

Reports of RCT's of screening for colorectal cancer with double-contrast barium enema (DCBE) have not been found. One strategy proposed consists of offering double-contrast barium enema every five to ten years [1]. This method is considered safer than colonoscopy or sigmoidoscopy but small polyps may go undetected. For colorectal cancers and large polyps (≥ 1 cm), the accuracy of double-contrast barium enema has been considered almost as good as that for colonoscopy [1]. However, in a retrospective study assessing the relative sensitivity of colonoscopy and barium enema for detection of colorectal cancer, colonoscopy was found to be more sensitive (97 % vs. 87 %) [57]. Another study [58] found a sensitivity of DCBE of 80 % and 77 % for colorectal cancer and adenomas, respectively.

Double-contrast barium enema is considered a good technique for colorectal cancer surveillance in Hereditary Non-Polyposis Colorectal Cancer, ulcerative colitis and flat adenoma syndrome (i.e. in situations in which small details of the mucosa should be analysed) [59].

Combining double-contrast barium enema and flexible sigmoidoscopy in patients with a positive FOBT provides sensitivities for detecting a colorectal cancer ranging from 90 - 98 % and from 96 - 99 % for adenomas ≥ 1 cm [42][58].

Perforation and mortality rates of 1 : 10 000 and 1 : 50 000, respectively, have been reported for flexible sigmoidoscopy [60].

Colonoscopy

Apparently, no controlled study has evaluated whether colonoscopy alone reduces the incidence or mortality from colorectal cancer in people at average risk of the disease [1]. To perform a screening colonoscopy every 10 years has been proposed [1]. Uncontrolled prospective studies of screening colonoscopy in average-risk asymptomatic persons have been examined by Rex in a non-systematic review [61]. The yield of adenomas diagnosed ranged from 13 - 41 %; the figure was 0 - 2.2 % for colorectal cancer [36][41][62][63][64]. Similar data were observed in another study [40]. However, most data come from diagnostic evaluations and surveillance, and not from screening studies. Small polyps (< 5 mm) could be missed in 25 % of cases and polyps > 1 cm in up to 10 % of cases [1][65]. In two prospective studies of screening colonoscopy, the caecum was reached in over 98 % of cases [63][66].

Impact of Screening Colonoscopy on Patient Outcome

Despite the high yields from screening colonoscopy for adenomas and colorectal cancer, the absence of a control group or of follow-up does not permit analysis of the effectiveness of screening by colonoscopy [26]. Some authors advocate screening colonoscopy in average-risk males ≥ 60 years old, taking into account the higher prevalence of adenomas in this subgroup [61]. Indirect evidence from a case-control study [45] supports the efficacy of colonoscopy in reducing the risk of colorectal cancer. The AGA recommendations indicate that a colonoscopy every ten years is an option in average-risk individuals [1].

The use and benefit of colonoscopic polypectomy is described in detail in the joint article on colorectal polyps [67].

People at Increased Risk of Colorectal Cancer

The use of colonoscopy for surveillance in patients with inflammatory bowel disease is examined in another article published in the same issue of the Journal [68].

People with Close Relatives who Have Had Colorectal Cancer or an Adenomatous Polyp

In seven prospective trials examining the yield from screening colonoscopy in persons with a positive family history of colorectal cancer, adenomas were found in 18 - 36 % and colorectal cancer in 0 - 2 % of the patients [61]. In a prospective controlled trial [69] from the Telemark Polyp Study, 53 % of adenomas were found in relatives vs. 33 % in the general population (historical controls), and 1.7 % of colorectal cancer vs. 0.3 % respectively. Thirty-five percent of polyps were located proximally.

A non-randomized controlled trial [64] compared screening colonoscopy in asymptomatic first-degree relatives of colorectal cancer patients with asymptomatic controls without a personal or a family history of colorectal cancer: 14.4 % of study patients had adenomas vs. 8.4 % of controls, and 48 % and 25 % respectively were beyond the reach of flexible sigmoidoscopy. Age, gender and a history of colorectal cancer in first-degree relatives were independently associated with the presence of adenomas. However, another controlled trial [70] found no relationship between a family history of a single first-degree relative and a higher prevalence of colorectal cancer or adenomas, except if the cancer in the relative was diagnosed before the age of 60. According to Rex, the yield of colonoscopy to detect colorectal neoplasia in patients with a positive family history has been surprisingly “low” so far [61]. Indeed, we could find no study showing decreased overall mortality from colorectal cancer in first-degree relatives undergoing a screening programme.

A non-systematic review [71] found compliance rates for screening colonoscopy between 42 % and 69 %, in first-degree relatives of persons with colorectal cancer. Compliance was lower (30 %) in an Italian pilot cohort study of 802 first-degree relatives of colorectal cancer patients [72], but higher (82 %) in an Norwegian study [69].

Recommendations of Others

According to the AGA, first-degree relatives of individuals with colon cancer should be offered the same options as average-risk people but beginning at age 40 years. If the close relative was diagnosed with colorectal cancer before the age of 55 years or with an adenomatous polyp before age 60, special efforts should be made to ensure that screening takes place [1]. Similar proposals have been made by the WHO Collaborating Center for the Prevention of Colorectal Cancer [55].

People with a Family History of Familial Adenomatous Polyposis (FAP)

These patients should receive genetic counseling and consider undergoing genetic testing to see if they are gene carriers. A negative genetic test result rules out FAP only if an affected family member has an identifiable mutation. Gene carriers or indeterminate cases should be offered flexible sigmoidoscopy every twelve months beginning at puberty to see if they are carrying the gene. If polyposis is present, they should begin to consider at what stage they should have a colectomy. FAP has an autosomal dominant inheritance and people with FAP have a 100 % chance of developing colorectal cancer [1].

People with a Family History of Hereditary Nonpolyposis Colorectal Cancer (HNPCC)

HNPCC is the most common hereditary form of colorectal cancer (nearly 10 % of all colorectal cancer cases). An accelerated adenoma-carcinoma progression has been suggested because of the presence of larger adenomas with more dysplasia than in non-familial cases [73]. Two-thirds of cancers occur in the right colon (nearly 70 % proximal to the splenic flexure) and a majority of synchronous and metachronous cancers have been observed [73]. An examination of the entire colon is thus necessary.

We only found one prospective controlled study [74] of screening for colorectal cancer in HNPCC families. Most of the available evidence is otherwise observational. Järvinen [74] and colleagues found that a 3-year interval colonoscopy (or double-contrast barium enema and flexible sigmoidoscopy) screening in families with HNPCC reduces the colorectal cancer rate of 62 % within a 10-year follow-up period. In the screening group, six colorectal cancers were identified (6/133) and in the control group 14 (14/118). A probable reduction in the colorectal cancer death rate should be achieved, but the reduction in this study did not reach statistical significance.

No consensus exists with regard to the interval betwen consecutive examinations [75]. Lynch et al. recommend initiating colonoscopy at age 20 - 25 years with intervals of one to two years [73].

Recommendations of Other Authors

People with a family history of colorectal cancer in multiple close relatives and across generations, especially if cancers occur at a young age, should receive genetic counselling and consider undergoing genetic testing for hereditary non-polyposis colorectal cancer. They should be offered an examination of the entire colon every one to two years starting between the ages of 20 and 30 years, and every year after age 40 [1]. An US task force (Cancer Genetics Studies Consortium [CGSC]) [76] recommends a full colonoscopy every one to three years beginning at age 20 to 25 years, for individuals known to be carriers of HNPCC-associated mutations (and for those with a substantial likelihood of being mutation carriers).

2. Panel Results

Considering the above review of relevant literature, the panel evaluated 37 specific theoretical patient scenarios related to screening for colorectal cancer in asymptomatic patients without a personal history of colorectal cancer or polyps, using an explicit two round modified Delphi panel expert method (RAND appropriateness method) which is described in a joint publication [5].

Definition of Terms Used

The terms and definitions were reviewed and approved by the panelists, they are listed in Table [4].

Clinical Variables Used

The clinical variables used to create and rate detailed patient scenarios to evaluate the appropriateness of use of colonoscopy for screening for colorectal cancer in asymptomatic individuals and for screening for colorectal cancer in asymptomatic individuals with a positive FOBT screening are shown in Table [5].

General Panel Results for Screening for Colorectal Cancer in Asymptomatic Patients

Colorectal cancer screening in asymptomatic patients was assessed in 34 clinical scenarios for five categories of risk of developing colorectal cancer, according to age and interval since a previous colonoscopy. Three additional scenarios referred to positive FOBT screening in asymptomatic patients. Of the 37 scenarios, the panel rated 22 % (8) as inappropriate, 32 % (12) as uncertain and 46 % (17) as appropriate. The rate of overall agreement was high (72 % of the scenarios).

Specific Clinical Panel Results for Screening for Colorectal Cancer in Asymptomatic Patients

The main results are worded as overall statements representing several clinical scenarios. In some cases, the same scenario may apply to more than one statement. Over 90 % (34/37) indications could be characterized by the six overall statements given in Table [6]. Detailed appropriateness and necessity criteria, including voting distribution for all 37 theoretical scenarios, are available in a computerized form accessible via Internet at the EPAGE web site (http://www.epage.ch).

Description of Necessity

Ten out of 37 scenarios (27 %) were judged necessary. All necessary indications for screening for colorectal cancer in asymptomatic patients were related to patients at moderate or high risk to develop colorectal cancer (Table [7]).

Conclusions

The AGA guideline, published in 1997, considers that there is strong evidence to support screening for colorectal cancer. “Evidence exists that reductions in colorectal cancer mortality can be achieved through detection and treatment of early-stage colorectal cancers and the identification and removal of adenomatous polyps, the precursor of these cancers” [1]. According to these recommendations, several tests offer a sufficiently high level of performance and effectiveness, so that one should take into account the patient's preferences, the patient's age and comorbidity as well as local resources and expertise availability [1].

However, these recommendations have been considered too optimistic, given for instance that the Nottingham [14] and the Funen studies [12], conducted among the general population (as opposed to the all-volunteer Minnesota study [13]) indicated a reduction of 15 - 18 % in mortality due to colorectal cancer, a result that may well be much lower when screening is applied in daily practice [77].

To perform a colonoscopy every ten years has been proposed as an effective option for colorectal screening. This is a very expensive option and there is no solid data to support this proposal to date.

Using colonoscopy as a screening tool for colorectal cancer in asymptomatic individuals at no increased risk was judged inappropriate in patients aged under 50 by the EPAGE panel, whereas the indication was rated uncertain in individuals aged 50 or more. When a screening FOBT was positive, colonoscopy was considered appropriate but not necessary.

Colonoscopy was more often judged appropriate in patients with at least moderate risk for developing colorectal cancer. However, the EPAGE panel was uncertain about the appropriateness of use of colonoscopy to screen for colorectal cancer in patients at increased risk because of HNPCC. Given the relative disagreement of the EPAGE panel with the (limited) published evidence, the panelists will be formally asked to reexamine their position during the Summer of 1999. In patients with FAP, colonoscopy was considered appropriate and necessary in almost every clinical scenario proposed. Full panel results can be consulted at the EPAGE web site (http://www.epage.ch).

Acknowledgement

The authors gratefully acknowledge the selfless commitment and invaluable contribution of the expert panel members, who made this project possible: Marcello Anti (IT), Peter Bytzer (DK), Mark Cottrill (UK), Micheal Fried (CH), Roar Johnsen (NO), Gerd Kanzler (DE), François Lacaine (FR), Cornelis Lamers (NL), Roger J. Leicester (UK), Mattijs E. Numans (NL), Javier P. Piqueras (ES), Jean-François Rey (FR), Giacomo Sturniolo (IT), Robert P. Walt (UK).

This work was supported by the EU BIOMED II Programme (BMH4-CT96-1202), the Swiss National Science Foundation (32.40522.94) and the Swiss Federal Office of Education and Science (95.0306-2).

1 Winawer SJ, Fletcher RH, Miller L, Godlee F, Stolar MH, Mulrow CD, Woolf SH, Glick SN, Ganiats TG, Bond JH et al. Colorectal cancer screening - clinical guidelines and rationale (1997; 112: 594).
Gastroenterology  1997; 112: 1060-1063  
Search for contribution in: PubMed

2 Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J. Cancer incidence in five continents. .
Lyon VII. IARC Scientific Publications  1997; 143: 

3 Cooper GS, Yuan Z, Landefeld CS, Johanson JF, Rimm AA. A national population-based study of incidence of colorectal cancer and age. Implications for screening in older Americans.
Cancer  1995; 75: 775-781  
Search for contribution in: PubMed

4 Capocaccia R, De Angelis R, Frova L, Gatta G, Sant M, Micheli A, Berrino F, Conti E, Gafa L, Roncucci L et al. Estimation and projections of colorectal cancer trends in Italy.
Int J Epidemiol  1997; 26: 924-932  
Search for contribution in: PubMed

5 Vader JP, Burnand B, Froehlich F, Dubois RW, Bochud M, Gonvers JJ. The European Panel on Appropriateness of Gastrointestinal Endoscopy (EPAGE): Project and methods.
Endoscopy  1999; 31: 572-578  
Search for contribution in: PubMed

6 Vader JP, Froehlich F, Dubois RW, Beglinger C, Wietlisbach V, Pittet V, Ebel N, Gonvers JJ, Burnand B. The European Panel on the Appropriateness of Gastrointestinal Endoscopy (EPAGE): Conclusions and WWW site.
Endoscopy  1999; 31: 687-694  
Search for contribution in: PubMed

7 Bochud M, Burnand B, Froehlich F, Dubois RW, Vader JP, Gonvers JJ. Appropriateness of colonoscopy: Surveillance after curative resection of colorectal cancer.
Endoscopy  1999; 31: 664-671  
Search for contribution in: PubMed

8 Ahlquist DA, Wieand HS, Moertel CG, McGill DB, Loprinzi CL, O'Connell MJ, Mailliard JA, Gerstner JB, Pandya K, Ellefson RD. Accuracy of fecal occult blood screening for colorectal neoplasia. A prospective study using Hemoccult and HemoQuant tests.
JAMA  1993; 269: 1262-1267  
Search for contribution in: PubMed

9 Allison JE, Tewaka IS, Ransom LJ, Adrain AL. A comparison of fecal occult-blood tests for colorectal-cancer screening.
N Engl J Med  1996; 334: 155-159  
Search for contribution in: PubMed

10 Ransohoff DF, Lang CA. Suggested technique for fecal occult blood testing and interpretation in colorectal cancer screening.
Ann Intern Med  1997; 126: 808-810  
Search for contribution in: PubMed

11 Ransohoff DF, Lang CA. Screening for colorectal cancer with the fecal occult blood test: a background paper. American College of Physicians.
Ann Intern Med  1997; 126: 811-822  
Search for contribution in: PubMed

12 Kronborg O, Fenger C, Olsen J, Jorgensen OD, Sondergaard O. randomized study of screening for colorectal cancer with faecal-occult-blood test.
Lancet  1996; 348: 1467-1471  
Search for contribution in: PubMed

13 Mandel JS, Bond JH, Church TR, Snover DC, Bradley GM, Schuman LM, Ederer F. Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota colon cancer control study.
N Engl J Med  1993; 328: 1365-1371  
Search for contribution in: PubMed

14 Hardcastle JD, Chamberlain JO, Robinson MH, Moss SM, Amar SS, Balfour TW, James PD, Mangham CM. randomized controlled trial of fecal-occult-blood screening for colorectal cancer.
Lancet  1996; 348: 1472-1477  
Search for contribution in: PubMed

15 Selby JV, Friedman GD, Quesenberry CP, Weiss NS. Effect of fecal occult blood testing on mortality from colorectal cancer. A case-control-study.
Ann Intern Med  1993; 118: 1-6  
Search for contribution in: PubMed

16 Saito H, Soma Y, Koeda J, Wada T, Kawaguchi H, Sobue T, Aisawa T, Yoshida Y. Reduction in risk of mortality from colorectal cancer by fecal occult blood screening with immunochemical hemagglutination test. A case-control study.
Int J Cancer  1995; 61: 465-469  
Search for contribution in: PubMed

17 Wahrendorf J, Robra BP, Wiebelt H, Oberhausen R, Weiland M, Dhom G. Effectiveness of colorectal cancer screening: results from a population-based case-control evaluation in Saarland, Germany.
Eur J Cancer Prev  1993; 2: 221-227  
Search for contribution in: PubMed

18 Launoy G, Herbert C, Vallee JP, Desoubeaux N, Reaud JM, Ollivier V, Bouvier V, Flachs A, Arsene D, Valla A et al. Mass screening for colorectal cancer in France. Experience in 165 000 people in the department of Calvados (French). .
Gastroenterol Clin Biol  1996; 20: 228-236  
Search for contribution in: PubMed

19 Pitsch RJ, Becker JM, Dayton MT. The occurrence of colon cancer in patients with known premalignant colonic mucosal diseases.
Journal of Surgical Research  1994; 57: 293-298  
Search for contribution in: PubMed

20 Kewenter J, Brevinge H, Engaras B, Haglind E, Ahren C. Follow-up after screening for colorectal neoplasms with fecal occult blood testing in a controlled trial.
Dis Colon Rectum  1994; 37: 115-119  
Search for contribution in: PubMed

21 Tazi MA, Faivre J, Dassonville F, Lamour J, Milan C, Durand G. Participation in faecal occult blood screening for colorectal cancer in a well defined French population: results of five screening rounds from 1988 to 1996.
J Med Screen  1997; 4: 147-151  
Search for contribution in: PubMed

22 Sommer T, Kronborg O. Possible causes of a positive haemoccult-ii test in a population screening study for colorectal neoplasia.
Eur J Cancer Prev  1996; 5: 181-187  
Search for contribution in: PubMed

23 Vernon SW. Adherence to colorectal cancer screening. A brief overview (review, 19 refs).
Annals of the New York Academy of Sciences  1995; 768: 292-295  
Search for contribution in: PubMed

24 Myers RE, Balshem AM, Wolf TA, Ross EA, Millner L. Screening for colorectal neoplasia: physicians' adherence to complete diagnostic evaluation.
American Journal of Public Health  1993; 83: 1620-1622  
Search for contribution in: PubMed

25 U.S.P.S.T.F. Screening for Colorectal Cancer in “Guide to Clinical Preventive Services: Report of the U.S. Preventive Services Task Force”.
Baltimore USPSTF  1996; 2: 89  

26 Solomon MJ, McLeod RS. Periodic health examination, 1994 update: 2. Screening strategies for colorectal cancer. Canadian Task Force on the Periodic Health Examination.
Canadian Medical Association Journal  1994; 150: 1961-1970  
Search for contribution in: PubMed

27 Winawer SJ, Zauber AG, Ho MN, O'Brien MJ, Gottlieb LS, Sternberg SS, Waye JD, Schapiro M, Bond JH, Panish JF et al. Prevention of colorectal cancer by colonoscopic polypectomy. The national polyp study workgroup.
N Engl J Med  1993; 329: 1977-1981  
Search for contribution in: PubMed

28 Rockey DCKJCJPSLLMK. Relative frequency of upper gastrointestinal and colonic lesions in patients with positive fecal occult-blood tests.
New England Journal of Medicine  1998; 339: 153-159  
Search for contribution in: PubMed

29 Fine KD. The prevalence of occult gastrointestinal bleeding in celiac sprue (see comments).
New England Journal of Medicine  1996; 334: 1163-1167  
Search for contribution in: PubMed

30 Schnell T, Aranha GV, Sontag SJ, Tode R, Reid S, Chejfec G, Karpf J, Levine G. Fecal occult blood testing: a false sense of security?.
Surgery  1994; 116: 798-803  
Search for contribution in: PubMed

31 Lang CA, Ransohoff DF. Fecal occult blood screening for colorectal cancer. Is mortality reduced by chance selection for screening colonoscopy?.
JAMA  1994; 271: 1011-1013  
Search for contribution in: PubMed

32 Bond JH. Screening for colorectal cancer: confuting the refuters.
Gastrointest Endosc  1997; 45: 105-109  
Search for contribution in: PubMed

33 Dubow RA, Katon RM, Benner KG, van Dijk CM, Koval G, Smith FW. Short (35 cm) vs. long (60 cm) flexible sigmoidoscopy: a comparison of findings and tolerance in asymptomatic patients screened for colorectal neoplasia.
Gastrointest Endosc  1985; 31: 305-308  
Search for contribution in: PubMed

34 Wilking N, Petrelli NJ, Herrera Ornelas L, Walsh D, Mittelman A. A comparison of the 25 cm rigid proctosigmoidoscope with the 65 cm flexible endoscope in the screening of patients for colorectal carcinoma.
Cancer  1986; 57: 669-671  
Search for contribution in: PubMed

35 Neugut AI, Forde KA. Screening colonoscopy: has the time come?.
Am J Gastroenterol  1988; 83: 295-297  
Search for contribution in: PubMed

36 Foutch PG, Mai H, Pardy K, DiSario JA, Manne RK, Kerr D. Flexible sigmoidoscopy may be the ineffective for secondary prevention of colorectal cancer in asymptomatic, average-risk men.
Dig Dis Sci  1991; 36: 924-928  
Search for contribution in: PubMed

37 Wherry DC, Thomas WM. The yield of flexible fiberoptic sigmoidoscopy in the detection of asymptomatic colorectal neoplasia.
Surg Endosc  1984; 8: 393-395  
Search for contribution in: PubMed

38 Olynyk JK, Aquilia S, Fletcher DR, Dickinson JA. Flexible sigmoidoscopy screening for colorectal cancer in average-risk subjects: a community-based pilot project.
Med J Aust  1996; 165: 74-76  
Search for contribution in: PubMed

39 Hoff G, Sauar J, Vatn MH, Larsen S, Langmark F, Moen IE, Foerster A, Thiis-Evensen E. Polypectomy of adenomas in the prevention of colorectal cancer. 10 years' follow-up of the Telemark Polyp Study I. A prospective, controlled population study.
Scan J Gastroenterol  1996; 31: 1006-1010  
Search for contribution in: PubMed

40 Reilly JM, Ballantyne GH, Fleming FX, Zucker KA, Modlin IM. Evaluation of the occult blood test in screening for colorectal neoplasms. A prospective study using flexible endoscopy.
Am Surg  1990; 56: 119-123  
Search for contribution in: PubMed

41 Lieberman DA, Smith FW. Screening for colon malignancy with colonoscopy.
Am J Gastroenterol  1991; 86: 946-951  
Search for contribution in: PubMed

42 Jensen J, Kewenter J, Asztely M, Lycke G, Wojciechowski J. Double contrast barium enema and flexible rectosigmoidoscopy: a reliable diagostic combination for detection of colorectal neoplasm.
Br J Surg  1990; 77: 270-272  
Search for contribution in: PubMed

43 Selby JV, Friedman GD, Quesenberry CP, Weiss NS. A case-control study of screening sigmoidoscopy and mortality from colorectal cancer.
N Engl J Med  1992; 326: 653-657  
Search for contribution in: PubMed

44 Newcomb PA, Norfleet RG, Storer BE, Surawicz TS, Marcus PM. Screening sigmoidoscopy and colorectal cancer mortality (see comments).
Journal of the National Cancer Institute  1992; 84: 1572-1575  
Search for contribution in: PubMed

45 Muller AD, Sonnenberg A. Prevention of colorectal cancer by flexible endoscopy and polypectomy: a case-control study of 32 702 veterans (see comments).
Ann Intern Med  1995; 123: 904-910  
Search for contribution in: PubMed

46 Selby JV, Friedman GD, Collen MF. Sigmoidoscopy and mortality from colorectal cancer: the Kaiser Permanente Multiphasic Evaluation Study.
J Clin Epidemiol  1988; 41: 427-434  
Search for contribution in: PubMed

47 Atkin WS, Morson BC, Cuzick J. Long-term risk of colorectal cancer after excision of rectosigmoid adenomas (see comments).
N Engl J Med  1992; 326: 658-662  
Search for contribution in: PubMed

48 Brint SL, DiPalma JA, Herrera JL. Colorectal cancer screening: is one-year surveillance sigmoidoscopy necessary?.
Am J Gastroenterol  1993; 88: 2019-2021  
Search for contribution in: PubMed

49 Rex DK, Lehmann GA, Ulbright TM, Smith JJ, Hawes RH. The yield of a second screening flexible sigmoidoscopy in average-risk persons after one negative examination.
Gastroenterology  1994; 106: 593-595  
Search for contribution in: PubMed

50 Lewis SF, Jensen NM. Screening sigmoidoscopy. Factors associated with utilization (see comments).
Journal of General Internal Medicine  1996; 11: 542-544  
Search for contribution in: PubMed

51 Senore C, Segnan N, Rossini FP, Ferraris R, Cavallero M, Coppola F, Pennazio M, Atkin WS. Screening for colorectal cancer by once only sigmoidoscopy: a feasibility study in Turin, Italy.
J Med Screen  1996; 3: 72-78  
Search for contribution in: PubMed

52 Schroy PC, Wilson S, Afdhal N. Feasibility of high-volume screening sigmoidoscopy using a flexible fiberoptic endoscope and a disposable sheath system.
Am J Gastroenterol  1996; 91: 1331-1337  
Search for contribution in: PubMed

53 Atkin WS, Cuzick J, Northover JM, Whynes DK. Prevention of colorectal cancer by once-only sigmoidoscopy.
Lancet  1993; 341: 736-740  
Search for contribution in: PubMed

54 Atkin WS, Hart A, Edwards R, Mcintyre P, Aubrey R, Wardle J, Sutton S, Cuzick J, Northover JA. Uptake yield of neoplasia, and adverse effects of flexible sigmoidoscopy screening.
Gut  1998; 42: 560-565  
Search for contribution in: PubMed

55 Winawer SJ, St John DJ, Bond JH, Rozen P, Burt RW, Waye JD, Kronborg O, O'Brien MJ, Bishop DT, Kurtz RC et al. Prevention of colorectal cancer: guidelines based on new data. WHO collaborating center for the prevention of colorectal cancer.
Bulletin of the World Health Organization  1995; 73: 7-10  
Search for contribution in: PubMed

56 Eddy DM. Screening of colorectal cancer.
Ann Intern Med  1990; 113: 373-384  
Search for contribution in: PubMed

57 Rex DK, Rahmani EY, Haseman JH, Lemmel GT, Kaster S, Buckley JS. Relative sensitivity of colonoscopy and barium enema for detection of colorectal cancer in clincial practice.
Gastroenterology  1997; 112: 17-23  
Search for contribution in: PubMed

58 Kewenter J, Brevinge H, Engaras B, Haglind E. The yield of flexible sigmoidoscopy and double-contrast barium enema in the diagnosis of neoplasms in the large bowel in patients with a positive hemoccult test.
Endoscopy  1995; 27: 159-163  
Search for contribution in: PubMed

59 Toribara NW, Sleisenger MH. Screening for colorectal cancer.
N Engl J Med  1995; 332: 861-867  
Search for contribution in: PubMed

60 Gelfand DW, Ott DJ. The economic implications of radiologic screening for colonic cancer.
AJR  1991; 156: 939-943  
Search for contribution in: PubMed

61 Rex DK. Colonoscopy: a review of its yield for cancers and adenomas by indication.
Am J Gastroenterol  1995; 90: 353-365  
Search for contribution in: PubMed

62 Johnson DA, Gurney MS, Volpe RJ, Jones DM, VanNess MM, Chobanian SJ, Avalos JC, Buck JL, Kooyman G, Cattau ELJ. A prospective study of the prevalance of colonic neoplasms in asymptomatic patients with an age-related risk.
Am J Gastroenterol  1990; 85: 969-974  
Search for contribution in: PubMed

63 Rex DK, Lehman GA, Hawes RH, Ulbright TM, Smith JJ. Screening colonoscopy in asymptomatic average-risk persons with negative fecal occult blood tests.
Gastroenterology  1991; 100: 64-67  
Search for contribution in: PubMed

64 Guillem JG, Forde KA, Treat MR, Neugut AI, O'Toole KM, Diamond BE. Colonoscopic screening for neoplasms in asymptomatic first-degree relatives of colon cancer patients. A controlled, prospective study.
Dis Colon Rectum  1992; 35: 523-529  
Search for contribution in: PubMed

65 Hixson LJ, Fennerty MB, Sampliner RE, Garewal HS. Prospective blinded trial of the colonoscopic miss-rate of large colorectal polyps.
Gastrointest Endosc  1991; 37: 125-127  
Search for contribution in: PubMed

66 Rogge JD, Elmore MF, Mahoney SJ, Brown ED, Troiano FP, Wagner DR, Black DJ, Pound DC. Low-cost, office-based, screening colonoscopy (see comments).
Am J Gastroenterol  1994; 89: 1775-1780  
Search for contribution in: PubMed

67 Bochud M, Burnand B, Froehlich F, Dubois RW, Vader JP, Gonvers JJ. Appropriateness of colonoscopy: Surveillance after polypectomy.
Endoscopy  1999; 31: 654-663  
Search for contribution in: PubMed

68 Froehlich F, Larequi-Lauber T, Gonvers JJ, Dubois RW, Burnand B, Vader JP. Appropriateness of colonoscopy: Inflammatory bowel disease.
Endoscopy  1999; 31: 647-653  
Search for contribution in: PubMed

69 Sauar J, Hausken T, Hoff G, Bjorkheim A, Foerster A, Mowinckel P. Colonoscopic screening examiantion of relatives of patients with colorectal cancer. I. A comparison with an endoscopically screened normal population.
Scan J Gastroenterol  1992; 27: 661-666  
Search for contribution in: PubMed

70 Rex DK, Lehman GA, Ulbright TM, Smith JJ, Pound DC, Hawes RH, Helper DJ, Wiersema MJ, Langefeld CD, Li W. Colonic neoplasia in asymptomatic persons with negative fecal occult blood tests: influence of age, gender, and family history.
Am J Gastroenterol  1993; 88: 825-831  
Search for contribution in: PubMed

71 Brewer DA, Fung CL, Chapuis PH, Bokey EL. Should relatives of patients with colorectal cancer be screened? A critical review of the literature.
Dis Colon Rectum  1994; 37: 1328-1338  
Search for contribution in: PubMed

72 Colombo L, Corti G, Magri F, Marocchi A, Brambilla P, Crespi C, Manieri L, Ghezzi S, Giannone D, Merlino L et al. Results of a pilot study of endoscopic screening for first degree relatives of colorectal cancer patients in Italy.
Journal of Epidemiology & Community Health  1997; 51: 453-458  
Search for contribution in: PubMed

73 Lynch HT, Smyrk T, Lynch J, Fitzgibbons RJ, Lanspa S, McGinn T. Update on the differential diagnosis, surveillance and management of hereditary non-polyposis colorectal cancer.
Eur J Cancer  1995; 31A (7 - 8): 1039-1046  
Search for contribution in: PubMed

74 Jarvinen HJ, Mecklin JP, Sistonen P. Screening reduces colorectal cancer rate in families with hereditary nonpolyposis colorectal cancer (see comments).
Gastroenterology  1995; 108: 1405-1411  
Search for contribution in: PubMed

75 Vasen HF, Mecklin JP, Watson P, Utsunomiya J, Bertario L, Lynch P, Svendsen LB, Cristofaro G, Muller H, Khan PM et al. Surveillance in hereditary nonpolyposis colorectal cancer: an international cooperative study of 165 families. The International Collaborative Group on HNPCC.
Dis Colon Rectum  1993; 36: 1-4  
Search for contribution in: PubMed

76 Burke W, Petersen G, Lynch P, Botkin J, Daly M, Garber J, Kahn MJ, McTiernan A, Offit K, Thomson E et al. Recommendations for follow-up care of individuals with an inherited predisposition to cancer. I. Hereditary nonpolyposis colon cancer. Cancer genetics studies consortium.
JAMA  1997; 277: 915-919  
Search for contribution in: PubMed

77 Simon JB. Should all people over the age of 50 have regular fecal occult-blood tests? Postpone population screening until problems are solved.
N Engl J Med  1998; 338: 1151-1152  
Search for contribution in: PubMed

78 Brevinge H, Lindholm E, Buntzen S, Kewenter J. Screening for colorectal neoplasia with faecal occult blood testing compared with flexible sigmoidoscopy directly in a 55 - 56 years old population.
Int J Colorectal Dis  1997; 12: 291-295  
Search for contribution in: PubMed

¹ The European Panel on Appropriateness of Gastrointestinal Endoscopy (EPAGE, Lausanne, Switzerland)

Dr. Bernard Burnand

IUMSP
Bugnon 17
CH-1005 Lausanne
Switzerland
Phone: +41-21-3144954
eMail: Bernard.Burnand@inst.hospvd.ch