522 Original Article Quality Assurance and Colonoscopy G. Minoli 1, G. Meucci 1, A. Prada 2, V Terruzzi 1, A. Bortoli 2, R. Gullotta 3, F. Rocca 4, E. Lesinigo 4, M. Curzio 3 'Division of Medicine, Valduce Hospital, Como, Italy 2 Division of Medicine, Rho Hospital, Rho, Italy 3 Gastroenterology Unit, Circolo di Varese Hospital, Varese, Italy 4 Medical Unit, Busto Arsizio Hospital, Busto Arsizio, Italy Background and Study Aims: Little is known concerning the usefulness and feasibility of quality assurance programs in gastrointestinal departments. The aim of this study was to identify the indicators of quality in colonoscopy, to check their use in clinical practice, and to identify their threshold values. Materials and Methods: A prospective study was performed in four endoscopic units. In the first phase, a questionnaire was used to identify the indicators that were considered important and easy to record; in the second phase, the selected items were prospectively recorded. Introduction It has been recommended that each endoscopy center should have a proper quality assurance program to allow systematic monitoring of the quality and appropriateness of care, to create opportunities to improve patient care, and to resolve any problems that are identified [1]. As early as 1988, the American Society of Gastrointestinal Endoscopy (ASGE) developed guidelines for quality assurance programs, based on previously released recommendations by the Joint Commission on Accreditation of Healthcare Organizations [2]. In 1993, the Italian Society for Gastroenterology and Gastrointestinal Endoscopy officially initiated the use of this type of quality assurance program in Italy as well [3]. Endoscopy 1999; 31 (7): 522-527 Georg Thieme Verlag Stuttgart New York ISSN 0013-726X Results: Data from 603 colonoscopies were evaluated. The selected indicators were: rate of cecal intubation, rate of examinations with normal findings, rates of complications, appropriateness of indications, use of a washing machine for disinfection, duration of the disinfection procedure, rate of procedures repeated due to poor colon cleansing, rate of operative procedures, length of waiting time, rate of procedures performed for follow-up of known disease, experience of the operator, and rate of procedures performed with the patient under conscious sedation. A striking difference emerged between the technical standards at three centers, which were fairly good, and the standard at the fourth center, which was less satisfactory. The length of the waiting time was high in all centers, as well as the rate of examinations conducted with an inappropriate indication. The rate of procedures performed under conscious sedation varied widely between the centers. Conclusions: The study of the indicators of quality of colonoscopy is feasible and easy to perform in clinical practice, and can be useful for quality assurance programs. With regard to colonoscopy, initiating quality assurance programs appears to be a matter of primary importance. Although the qualifications needed to perform endoscopic procedures have been defined [4,5], it appears that a large number of procedures are being performed by clinicians who do not have an appropriate background or training [4]. Moreover, a substantial proportion of colonoscopies are performed without an appropriate clinical indication [6]. Some preliminary reports have clearly demonstrated the value of quality assurance programs, and a quality assurance review has been shown to result in a reduction in the rate of inappropriate procedures and of deficiencies in procedural reports [6,7]. The major steps in any quality assurance program are to identify indicators of quality and to establish threshold levels for each of them. The aim of the present multicenter study was to assess the feasibility of this process in gastrointestinal units in Italy, and to verify whether this type of procedure can really provide useful information on the technical standards at different centers.
Quality Assurance and Colonoscopy Endoscopy 1999; 31 523 Methods Phase I: Identifying Quality Indicators In this phase, a questionnaire was prepared and sent to endoscopists at seven gastrointestinal departments in Northern Italy, giving a list of 19 parameters that might be regarded as important indicators of quality in colonoscopy. These parameters included institution-related, procedure-related and outcome-related variables (Table 1). Endoscopists were requested to indicate which of these parameters they considered to be important and easy to record, by giving each of them a score ranging from one to ten. The mean scores for each parameter were calculated. According to these, the importance of the parameters was graded as follows: mean score >7: very important; mean score 6-7: fairly important; mean score < 6: not important. Table 1 Potential indicators of the quality of colonoscopy included in the first questionnaire, and mean scores of importance obtained in phase 1 Rate of cecal intubation 8.8* Rate of procedures with normal findings 8.5* Rate of complications 8.2* Appropriateness of indication 8.0* Duration of reprocessing cycle 7.8* Rate of endoscopies repeated due to inadequate 7.5* colon cleansing Rate of operative procedures 7.1* Length of waiting time 6.8* Ratio between the number of colonoscopies per 6.5 formed and the number of endoscopists in the unit Rate of polypectomies attempted without success 6.4 Rate of procedures performed for follow-up of 6.3* known lesions No. of colonoscopies performed by the endoscopist 6.2* during the previous year Rate of endoscopies performed under conscious 6.2* sedation Rate of appointments canceled 5.7 Duration of the procedure 5.7 Rate of polyps removed but not retrieved 5.5 Ratio between no. of colonoscopies performed 5.4 and no. of endoscopes in the unit Frequency of instrument breakage 3.1 Life duration of the endoscopes 1.6 * Parameters selected for prospective evaluation. Phase 2: Prospective Evaluation of Quality Indicators On the basis of the results of the initial questionnaire, 11 parameters that had a mean score higher than 6 were selected and a standardized form for prospective evaluation of these was prepared. This evaluation was then carried out in 783 consecutive colonoscopies performed at four participating institutions. Endoscopists were requested to fill in a copy of the form for each colonoscopy, just after completing the examination. Periodic phone interviews were conducted with the participating centers in order to ensure good compliance. Statistical analysis was carried out using the chi-squared test or Fisher's exact test, as appropriate. Results Phase I The mean scores obtained using the 19 parameters are shown in Table 1. Eight parameters obtained a mean score higher than 7 ("very important"), and all of these were selected for prospective evaluation. Five further parameters had a mean score between 6 and 7 ("fairly important"). Two of these parameters-the ratio between the number of colonoscopies and number of endoscopists in the unit, and the rate of polypectomies attempted without success-were considered unsuitable for systematic review, and were excluded from the prospective evaluation (the first because it would have been the same for all colonoscopies performed at a given center, and the second because it would have been applicable to a very limited number of examinations). The remaining six parameters obtained a mean score lower than 6 ("not important"), and were also excluded. It was decided to include items in the questionnaire relating to whether the instrument was disinfected manually or using a washing machine after the procedure, since it would not make sense to record the duration of the automated reprocessing cycle without recording this item. Phase 2 Overall, 783 consecutive colonoscopies were evaluated at the four participating institutions. The number of colonoscopies performed at each center ranged from 169 (Center 3) to 210 (Center 1). Exploration of only the left colon was required in 180 examinations (23%), and since these procedures could not be regarded as colonoscopies "ab initio," they were recorded but excluded from the analysis. The rate of such examinations varied widely among the centers-23% in Center 1, 8% in Center 2, 49% in Center 3, and 16% in Center 4. The analysis was therefore confined to 603 examinations: 162 at Center 1, 187 at Center 2, 86 at Center 3, and 168 at Center 4 (Table 2). The mean age of the patients undergoing these 603 examinations was 59.6 years (range 17-94); 308 patients (51 %) were male. The colonoscopies were performed as outpatient procedures in 376 cases (62.5%). In 99 cases (16%), an operative procedure was performed (polypectomy in 98, dilation of a stricture in one). Evaluation of the selected parameters gave the following results (Table 2).
524 Endoscopy 1999; 31 G. Minoli et al. Rate of cecal intubation. In 117 procedures (19.4%), no attempt was made to reach the cecum-in 48 cases (8%) because of stenosis, and in 69 (11.4%) because of poor colon cleansing. The cecum was intubated in 443 of the remaining 486 examinations (91.1 %). The percentage of colonoscopies in which no attempt was made to reach the cecum was between 10% and 20% at three centers, but was as high as 54% in the remaining center (Center 3). The rate of cecal intubation was 88-100% at all centers. Normal findings. No organic lesions were found in 207 colonoscopies (34.3%). This percentage was 54% at Center 3 and 22-27% at the other three centers. Complications. There were complications in two of the 603 examinations (0.33%), with no procedure related deaths. Both complications were colon perforations occurring during therapeutic procedures. Indications. The indication for colonoscopy was inappropriate, relative to the ASGE guidelines, in 102 cases (16.9%). In a further 82 cases (13.6%), the indication was not mentioned in the ASGE guidelines. These indications included recent onset constipation, the presence of an abdominal mass, weight loss, tenesmus, and follow-up in patients who had undergone surgery for non-neoplastic colonic disease. The relative percentages for single centers are listed in Table 2: no substantial variations between the centers were found, except for a slightly higher percentage of endoscopies with an appropriate indication at Center 3. Duration of the reprocessing cycle. At Center 4, instruments were always disinfected manually A washing machine was used in 52% of the examinations performed at Center 1, 74% of those performed at Center 2, and 62% of those performed at Center 3. When a washing machine Table 2 Prospective evaluation of selected indicators of quality Overall Center 1 Center 2 Center 3 Center 4 Procedures performed 783 210 204 169 200 Only left colonoscopy required (%) 1 80 (33 %) 48 (22.9%) 1 7 (8.3 %) 83 (49.1 %) 32 (1 6 %) Total colonoscopy required (%) 603 (77 %) 162 (77.1 %) 187 (91.7%) 86 (50.9 %) 1 68 (84 %) Indicators evaluated Rate of cecal intubation Not attempted because of poor 69 (1 1 %) 1 1 (6.8 %) 10(5.3%) 31 (36.1 %) 17 (10.1 %) cleansing (%) Not attempted because of stenosis (%) 48 (8 %) 1 1 (8 %) 1 1 (5.9 %) 16 (1 8.6 %) 10 (6%) Obtained (% of attempts) 443 (91 %) 1 32 (94.3 %) 147 (88.6 %) 39 (100 %) 1 25 (88.7 %) Procedures with normal findings 207 (34 %) 37 (22.8%) 52 (27.8%) 47 (54.7 %) 71 (22.3%) Complications 2 (0.3 %) 1 0 0 1 Appropriateness of indications Appropriate according to the ASGE guidelines (%) 442 (73 %) 1 22 (75.3 %) 1 32 (70.6 %) 72 (83.7 %) 1 16 (69 %) Not appropriate according to the 92 (1 5 %) 24 (14.8%) 26 (1 3.9 %) 1 1 (12.8%) 31 (18.5%) ASGE guidelines (%) Not mentioned in the ASGE guidelines (%) 69 (1 1 %) 16 (9.9 %) 29 (1 5.5 %) 3 (3.5 %) 21 (12.5%) Reprocessing performed by a 360 (60 %) 109 (52 %) 145 (74 %) 106 (62 %) - washing machine Duration of the reprocessing cycle 39±15 60 min 40 min 20 min n.a. Examinations repeated due to poor 69 (1 1 %) 11 (5.3 %) 1 0 (5.4 %) 31 (36.1 %) 17 (10.1 %) colon cleansing (%) Rate of surgical procedures 98 (16 %) 35 (21.6%) 44 (23.5 %) O(-) 1 9 (1 1.3 %) Length of the waiting time: outpatients 28 days 42 22 32 28 (days, mean) Length of the waiting time: inpatients 2.2 days 2.6 2.4 2.9 1.7 (days, mean) Procedures performed for follow-up 1 78 (30 %) 38 (20.4 %) 54 (28.9 %) 39 (45.4 %) 52 (31.1 %) of known lesions (%) Number of colonoscopies performed by the operator during the previous year <100(%) 45 (8 %) 29 (1 8 %) 10 (5.4 %) 6 (7 %) O(-) 100-300(%) 1 62 (27 %) 36 (22.2 %) 25 (13.4%) 67 (78 %) 34 (20 %) >300(%) 396 (65 %) 97 (59.9 %) 152 (81.3%) 13 (15%) 1 34 (80 %) Procedures performed under conscious 289 (18 %) 85 (53.1 %) 1 72 (92 %) 25 (29.1 %) 7 (4.2 %) sedation (%)
Quality Assurance and Colonoscopy Endoscopy 1999; 31 525 was used, the duration of the reprocessing cycle was 60, 40, and 20 minutes, respectively, at these centers. Rate of endoscopies repeated due to inadequate colon cleansing. Overall, 69 examinations had to be repeated (11.4%); the percentage was 36% at Center 3 and 5-11% at the others. Rate of surgical procedures. Operative procedures were carried out during 99 examinations (16.4%); no operative procedures were performed at Center 3, while at the other three centers the rate was 11-23%. Length of waiting time (mean). For outpatients, the mean waiting time was 28.3 days (range 1-90). The means varied between centers from 22 to 42 days. For inpatients, the mean waiting time was 2.2 days (range 0-79). Rate of procedures performed for follow-up of known lesions. Overall, 178 of the 603 procedures were for follow-up (29.6%): 20% at Center 1, 29% at Center 2, 45% at Center 3, and 31 % at Center 4. Number of colonoscopies performed by the operator during the previous year. The figure was more than 300 in 396 cases (65.7%), between 100 and 300 in 162 (26.9%), and less than 100 in 45 (7.4%). The percentage of examinations performed by operators who had carried out more than 300 colonoscopies during the previous year was about 80% at Centers 2 and 4, 65% at Center 1, and 15% at Center 3. No correlation was found between the number of colonoscopies performed by the endoscopist during the previous year and the rate of cecal intubation, which was 307 of 337 cases (91.1%; range 89-100% at single centers) for endoscopists who had carried out more than 300 procedures during the previous year and 37 of 42 (88.9%; 60-100% at single centers) for those who had carried out less than 100 colonoscopies (P-0.343). Rate of endoscopies performed with the patient under conscious sedation. Overall, conscious sedation was used in 291 of the 603 procedures (48.3%), with a very high variability among centers, from 4.2% (Center 4) to 92% (Center 2). The rate was 230 of 443 (52%) when the cecum was reached, 23 of 43 (53%) when cecal intubation was attempted but not achieved and 36 of 117 (31%) when cecal intubation was prevented either by stenosis or poor colon cleansing. All of the remaining procedures were performed without the use of sedation. Discussion Although the importance of quality assurance programs is well recognized little is known regarding the feasibility and usefulness of such programs in endoscopy units. This preliminary study shows that it is easy to assess quality indicators in endoscopic units, and that it provides useful information about the technical standards of different centers. Using such indicators in the context of quality assurance programs thus appears to be both feasible and desirable. Firstly, some comment is needed regarding the rate of procedures in which only an examination of the left colon was required-an indicator not included in the original list, but which was necessarily recorded and revealed important differences between the centers. This rate was less than 25 % in three centers, but was nearly 50% in one of them (Center 3); this may indicate a lower technical standard in the examinations carried out at this center. The rate of procedures during which cecal intubation was not attempted because of "inadequate colon cleansing" was also unacceptably high at Center 3-36%, compared with rates of between 5 % and 11 % at the other three, within the range reported in most controlled trials [8]. In addition, the rate of procedures in which cecal intubation was prevented by the presence of a stenosis was much higher at Center 3 than in the remaining three (18% vs. 6-8%). Although this difference may have been mere chance, it could indicate less experience and selfconfidence among the endoscopists at Center 3, leading to no attempts generally being made to pass a noncritical stricture. The analysis of two other indicators also indicated a lower degree of experience among the operators at Center 3: no therapeutic procedures were conducted at Center 3, whereas the rate of therapeutic procedures was 11-23% at the other centers. In addition, the rate of procedures performed by operators who had carried out more than 300 examinations during the previous year was only 15% at Center 3, compared with between 60% and 80% at the remaining three centers. Although there is no definite consensus regarding the threshold number of procedures required for a trainee to become competent in colonoscopy [9,10], both the ASGE Training Committee [11] and the guidelines of the Union of European Medical Specialists and the European Board of Gastroenterology [12,13] recommend that a minimum of 100 supervised colonoscopies should be performed prior to assessment of competence. Some recent data confirm that this recommendation is a sensible one [14], Since it is reasonable to assume that most clinicians who have performed fewer than 100 colonoscopies during the previous year have nevertheless performed more than 100 colonoscopies during their professional career, it can be concluded that the vast majority of colonoscopies in our units are being performed by operators who have received appropriate preliminary training. As far as ongoing experience is concerned, there is no universal standard indicating how many procedures endoscopists need to perform annually to maintain their skills. However, we can assume that the degree of ongoing experience was satisfactory at three centers in the present study, in which the majority of operators had performed more than 300 procedures during the previous year, and much less satisfactory at Center 3, in which this rate was only 15%. The rate of cecal intubation was 88-100% in all of the centers, a figure that is very close to the level of 90 % gen-
526 Endoscopy 1999; 31 G. Minoli et al. erally considered to be the minimum acceptable standard for formally trained endoscopists [1,14]. Even for procedures performed by less experienced endoscopists (less than 100 colonoscopies during the previous year), this rate was as high as 88%. At Centers 1, 2, and 4, this high rate of cecal intubation is a further confirmation of the satisfactory technical skills of the endoscopists working in these units. On the other hand, little significance can be given to the 100% rate of cecal intubation recorded at Center 3, since it was attempted in less than 20 % of the total number of examinations performed (39 of 169). In summary, the analysis of four indicators (rate of cecal intubation, rate of examinations needing to be repeated due to poor colon cleansing, rate of therapeutic procedures, and number of procedures performed by the operator during the previous year), in conjunction with the rate of procedures in which only exploration of the left colon was required, allows us to recognize a striking difference between the technical standard of three of the participating units (Centers 1, 2, and 4), which appears to be fairly acceptable, and the technical profile of Center 3, which appears to be much less satisfactory. Regarding the other parameters, the following remarks can be made. Firstly, the rate of complications was quite low (0.3%) and within the range reported in the literature [15]. On the other hand, the length of the waiting time for outpatients was unsatisfactory in all centers, approaching three months in some instances. This is a widespread problem in Italian gastrointestinal units, due firstly to the fact that endoscopies are carried out on an open access basis and secondly to the inadequate organization of the Italian health-care system. Only about 70% of colonoscopies were performed on the basis of an appropriate indication according to the ASGE guidelines. In 17% of cases, the indication had to be considered as inappropriate according to these guidelines. We have recently reported a similar rate of inappropriate indications for upper gastrointestinal endoscopies in our area [16]. Again, this may be partly due to the fact that in Italy, gastrointestinal endoscopies are performed on an open access basis, being ordered by family doctors without prior consultation with a specialist. In the remaining 14% of cases, the indication for colonoscopy was not mentioned in the ASGE guidelines, although in most instances it had to be regarded as fully appropriate on the basis of common sense. It has been suggested that the ASGE guidelines should now be regarded as out of date, and that new guidelines need to be developed [17]. European countries are unfortunately still well behind schedule in this task. In any case, continuous review of indications appears to be advisable, since it has been reported that this policy can significantly reduce the rate of inappropriate procedures [6]. It is also noteworthy that that only 291 procedures (48%) were performed with the patients under conscious sedation, a figure that is much lower than those recorded in surveys from the United States, where some form of sedation is used in over 90% of colonoscopies [18]. This is somewhat surprising, since it has been reported that using conscious sedation is considered an important indicator of quality by the majority of Italian endoscopists [3]. However, the rate of procedures performed with the patient under conscious sedation varied widely among centers (from 4% to 92%). This might reflect both different attitudes in giving sedation to patients and different availability of recovery facilities for patients. However, there is still a great deal of controversy on whether conscious sedation should be given to all patients at the beginning of the procedure, or whether on-demand sedation is preferable [19-23]. Further controlled studies on patient satisfaction are needed before a high rate of examinations carried out using conscious sedation can definitely be regarded as an indicator of good quality. The reliability of endoscopic disinfection is a matter of debate [24-26]. Highlevel disinfection after each endoscopic examination is generally recommended [25-27], but the best way to achieve this is not known [26-28], and the time needed for disinfection has been arbitrarily set at 20 minutes in the absence of reliable evidence in favor of this period rather than any other [25]. Surveys performed both in America [27] and Europe [29] indicate that disinfection methods are often less than optimal, so that in a fairly high percentage of cases, instruments are still contaminated at the end of the disinfection procedure [27, 28]. In the present survey, a washing machine was used after 52-74% of colonoscopies at three centers (with a duration of the cycle varying from 20 to 60 minutes) and never in the other center. It is not clear how these indicators should be interpreted, since it is not known whether an automated disinfection procedure can achieve better results than a manual one [28], nor what the optimal duration of the cycle is. At any rate, it is noteworthy that some form of disinfection procedure was carried out after each colonoscopy at all of the centers. By contrast, a survey performed some 10 years ago at 120 units in Western Europe found that no disinfection was carried out after colonoscopy in as many as 13% of institutions [29]. Finally, the significance of the remaining two indicatorsthe rates of examinations with normal findings and of those performed for follow-up of known lesions-is not clear. However, these figures were very similar among Centers 1, 2, and 4 (ranging, for both indicators, between 20% and 30%), and much higher at Center 3-the one that had scored worst for most of the other indicators. It can therefore be suggested that the figures recorded at Centers 1, 2, and 4 might provisionally be used as reference values for subsequent comparisons. In conclusion, the data presented here show that the study of the quality indicators in colonoscopy is feasible and easy to perform in everyday clinical practice, and that it can be useful for quality assurance programs.
Quality Assurance and Colonoscopy Endoscopy 1999; 31 527 Acknowledgements The authors are grateful to Dr. Alessandro Zambelli (Crema), Dr. Pietro Leo (Bologna) and Dr. Angelo Pera (Turin) for participating in the first phase of this study. References 1 Schroeder KW. Quality assurance in gastrointestinal endoscopy. Endosc Clin N Am 1993; 3: 571-580 2 American Society for Gastrointestinal Endoscopy Standards of Practice Committee. Quality assurance of gastrointestinal endoscopy. Manchester, MA: ASGE, 1988 3 Lombardia SIED, Minoli G, Comin U, et al. Identificazione di alcuni indicatori di qualita in endoscopia digestiva. In: Atti XV Congresso Nazionale della Societa di Endoscopia Digestiva. Bologna: Monduzzi, 1993: 33-39 4 Hogan WJ. Hospital credentialing standards for physicians who perform endoscopies. Gastroenterology 1993; 104: 1563-1565 5 American Society for Gastrointestinal Endoscopy Standards of Training and Practice Committee. Methods of granting hospital privileges to perform gastrointestinal endoscopy. Gastrointest Endosc 1992; 38: 765-767 6 Sapienza PE, Levine GM, Pomerantz S, et al. Impact of a quality assurance program on gastrointestinal endoscopy. Gastroenterology 1992; 102: 387-393 7 Mai HD, Sanowsky RA, Waring JP. Improved patient care using the ASGE guidelines on quality assurance: a prospective comparative study. Gastrointest Endosc 1991; 37: 597-599 8 Hsu CW, Imperiale TF. Metaanalysis and cost comparison of polyethylene glycol lavage versus sodium phosphate for colonoscopy preparation. Gastrointest Endosc 1998; 48: 276-282 9 Hogan WJ. What constitutes endoscopic competence? Gastroenterology 1993; 104: 1564-1565 Cass OW, Freeman ML, Peine CJ, et al. Objective evaluation of endoscopy skills during training. Ann Intern Med 1993; 118: 40-44 American Society for Gastrointestinal Endoscopy. Methods of granting hospital privileges to perform gastrointestinal endoscopy. Manchester, MA: ASGE (ASGE publication 1012), 1992 European Board of Gastroenterology. Specialist training in gastroenterology in the European Community: the case for the European Boards. Gut 1994; 35: 135-138 Union Europeenne des Medecins Specialistes (UEMS) Specialist Section, Gastroenterology/European Board of Gastroenterology. Charter on training of medical specialists in the EU: requirements for the specialty gastroenterology. Brussels: UEMS, 1995 Chak A, Cooper GS, Blades EW, Canto M, Sivak MV Prospective assessment of colonoscopic intubation skills in trainees. Gastrointest Endosc 1996; 44: 54-57 Keefe BK, Schrock TR. Complications of gastrointestinal endoscopy. In: Sleisenger MH, Fordtran JS. Gastrointestinal disease. Philadelphia: Saunders, 1993; 5th ed: 301-308 Minoli G, Prada A, Gambetta G, et al. The AGSE guidelines for the appropriate use of upper gastrointestinal endoscopy in an open access system. Gastrointest Endosc 1995; 42: 387-389 Froehlich F, Pache I, Burnand B, et al. Performance of panelbased criteria to evaluate the appropriateness of colonoscopy: a prospective study. Gastrointest Endosc 1998; 48: 128-136 Keeffe EB, O'Connor KW. 1989 ASGE survey of endoscopic sedation and monitoring practices. Gastrointest Endosc 1990; 36: S13-S18 10 11 12 13 14 15 16 17 18 [Anon]. Sedation and monitoring of patients undergoing gastrointestinal endoscopic procedures. Gastrointest Endosc 1995; 42: 626-629 Bianchi-Porro G, Lazzaroni M. Preparation, premedication and surveillance. Endoscopy 1992; 24: 1-8 Cataldo PA. Colonoscopy without sedation. Dis Colon Rectum 1996; 39: 257-261 Phillips MS. Drugs and sedation for colonoscopy. Prim Care 1995; 22: 433-443 Offman MS, Butler TW, Shraver T. Colonoscopy without sedation. J Clin Gastroenterol 1998; 26: 279-282 Knapp DA, Michocki RJ, Mays DA. Endoscopists need to clean up their act-literally. Arch Intern Med 1997; 22: 1919-1920 American Society for Gastrointestinal Endoscopy. Infection control during gastrointestinal endoscopy. Guidelines for clinical application. Gastrointestinal Endoscopy 1999; 49: 836-841 Tremain SC. Cleaning and disinfection of lower gastrointestinal endoscopes. Prim Care 1995; 22: 471-478 Kaczmarec RG, Moore RM, McCrohan J, et al. Multistate investigation of the actual disinfection/sterilization of endoscopes in health care facilities. Am J Med 1992; 92: 257-261 Fraser VJ, Zuckerman G, Clouse RE, et al. A prospective randomized trial comparing manual and automated endoscope disinfection methods. Infect Control Hosp Epidemiol 1993; 14: 383-389 Van Gossum A, Loriers M, Serruys E, Cremer M. Methods of disinfecting endoscopic material: results of an international survey. Endoscopy 1989; 21: 247-250 Corresponding Author G. Minoli, M.D. II Divisione di Medicina Ospedale Generale di Zona Valduce Via Dante 11 22100 Como Italy Fax: +39-031-308047 E-mail: gminoli@valduce.it Submitted: 5 January 1999 Accepted after Revision: 22 April 1999