Does an open access system properly utilize endoscopic resources? Gregory Zuccaro, Jr., MD, Kimberly Provencher, RN Cleveland, Ohio
Background: In an effort to maintain procedure volumes and control consult~tion costs, many gastrointestinal endoscopists and primary care providers have implemented systems of open access endoscopy. In these systems, specialists in digestive diseases perform endoscopy without prior consultation. The purpose of this study is to determine if indications for upper endoscopic procedures requested in an open access system conform to national practice guidelines and to establish the yield of diagnostic information relevant for patient care in this system. Methods: Procedural indications and results for 3715 upper endoscopic procedures performed in an open access system were recorded in a computer database. The practice guideline "Appropriate Use of Gastrointestinal Endoscopy" (AUGE) of the American Society for Gastrointestinal Endoscopy was used to determine appropriateness of procedural indications. Results: Eighty-four percent of procedures were performed for indications listed in the AUGE, and 59% resulted in findings relevant to patient care. Specialists requested endoscopy more frequently for "approved" indications than did nonspecialists 0o = .004) and more frequently had findings relevant to patient care 0o < .001). Findings relevant to patient care are significantly more frequent for some indications listed in the AUGE compared to others (p < .001). Conclusions: Adherence to practice guidelines can and does occur in an open access system. Specialists request endoscopy more frequently for appropriate indications compared to nonspecialists and have a higher yield of information relevant to patient care. Further refinement and better definition of some indications within the AUGE are needed to increase the clinical utility of this document. (Gastrointest Endosc 1997;46:15-20.)
The appropriate use of gastrointestinal endoscopy is an area of increasing concern for physicians, patients, and health care administrators. Endoscopy is an effective diagnostic and therapeutic tool, is well tolerated, and is safe. However, these same advantages m a y lead to increased costs through overutilization. In an effort to control costs, primary care Received July 18, 1996. For revision October 6, 1996. Accepted December 26, 1996. From the Department of Gastroenterology, Cleveland Clinic Foundation, Cleveland, Ohio. This work was supported in part by a research grant from the American Society for Gastrointestinal Endoscopy. Reprint requests: Gregory Zuccaro, Jr., MD, Cleveland Clinic Department of Gastroenterology, 9500 Euclid Ave., Cleveland, OH 44195. Copyright © 1997 by the American Society for Gastrointestinal Endoscopy 0016-5107/97/$5.00 + 0 3'//1/80211 VOLUME 46, NO. 1, 1997
providers may choose alternative imaging studies (e.g., barium radiography), learn and perform endoscopy themselves, request consultation by a specialist in digestive disease without an imaging study or, conversely, request endoscopy from specialists without prior consultation. In "open access" systems, primary care providers in essence request upper endoscopy as they would a radiologic study. Some gastrointestinal endoscopists have accepted these systems as a way of maintaining procedure volume and patient flow from primary care providers. A successful system of open access endoscopy assumes that the physician requesting the procedure understands its indications and contraindications and is capable of interpreting endoscopic findings and their implications for the patient. Practice guidelines and guidelines for referral to specialists are among the ways proposed to control utilization GASTROINTESTINAL ENDOSCOPY 15
G Zuccaro, Jr, K Provencher
Does an open access system properly utilize endoscopic resources?
Other 9% Abnorm=
~pepsia
Surveillance (Barreff~ ~tevious GU, etc 7%
GI bleeding/a~ 22%
16%
Persistent nausea/vomiting 6%
Evaluation of GE reflux symptoms 6%
Figure 1. Indications for UGI endoscopy. a n d d e c r e a s e u n n e c e s s a r y p r o c e d u r e s . T h e s e guidelines m a y be g e n e r a t e d b y p r o f e s s i o n a l societies, consensus between referring physicians and gastroi n t e s t i n a l specialists, m a n a g e d c a r e o r g a n i z a t i o n s , or o t h e r t h i r d p a r t i e s . F e w d a t a a r e a v a i l a b l e to v a l i d a t e t h e s e guidelines. A t t h e C l e v e l a n d Clinic, g a s t r o i n t e s t i n a l endoscop y is p r a c t i c e d in a closed s y s t e m w i t h o p e n r e f e r r a l . All p r o c e d u r e s a r e p e r f o r m e d b y fully t r a i n e d gast r o e n t e r o l o g i s t s or g a s t r o i n t e s t i n a l s u r g e o n s in t h e G I e n d o s c o p y suite. A n y p r o f e s s i o n a l s t a f f m e m b e r m a y r e f e r a p a t i e n t for u p p e r e n d o s c o p y w i t h o u t f o r m a l c o n s u l t a t i o n w i t h a g a s t r o i n t e s t i n a l specialist. T h i s s y s t e m p r o v i d e s t h e o p p o r t u n i t y to a s s e s s v a r i o u s a s p e c t s of t h e o p e n access s y s t e m of g a s t r o i n t e s t i n a l endoscopy, i n c l u d i n g a c o m p a r i s o n of utilization and results between gastroenterologists, other internists, and surgeons.
METHODS Relevant information for each upper endoscopic procedure performed in the Department of Gastroenterology over a 1-year period was recorded into a database built on VAX Datatrieve operating software. This included the indication for the procedure, results, and specialty of the physician requesting the endoscopy (i.e., gastroenterologist, other internist, or surgeon). Endoscopy was performed in an open access system. Consultation with a gastroenterologist was not necessary prior to performance of upper endoscopy. Because the Cleveland Clinic functions with a fully integrated charting system, the endoscopist had access to all of the patients' records and results of laboratory and radiologic studies. One of 12 staff endoscopists performed each procedure. The endoscopist would typically perform the procedure whether or not he or she might agree with the indication as noted by the requesting physician. One exception would be if there was an obvious contraindication (e.g., suspicion of perforation or unacceptable sedation/analgesia risk). The practice guideline of the American Society for Gastrointestinal Endoscopy entitled "Appropriate Use of Gastrointestinal Endoscopy "1 (AUGE) was used to assess the appropriateness of each referral for upper endoscopy 16
GASTROINTESTINAL ENDOSCOPY
Table 1. Appropriate indications for upper endoscopy as listed in the AUGE Upper abdominal pain/distress despite an appropriate therapeutic trial, or if serious organic disease is suggested Dysphagia/odynophagia Gastroesophageal reflux symptoms despite therapy Persistent vomiting of unknown cause Other system disease in which the presence of upper gastrointestinal pathology might influence management plans Surveillance for conditions such as Barrett's esophagus, familial adenomatous polyposis syndrome, or sporadic adenomatous polyps Confirmation of abnormal findings on UGI x-ray (e.g., suspected neoplasm, obstruction, potentially malignant ulcer) Gastrointestinal bleeding (includes range of active bleeding, recurrent bleeding, suspicion of aortoenteric fistula, contemplation of surgery, or chronic bleeding when UGI source suspected) Sampling of duodenal/jejunal contents Assess injury after caustic ingestion Therapy for bleeding, foreign bodies, removal of selected polyps, placement of feeding tubes, dilation of strictures/stenotic lesions, palliative treatment of obstructing neoplasms Document healing of selected ulcers of the UGI tract
(Table 1). Therefore, patients undergoing upper endoscopy could be categorized as follows: 1. A traditional group seen in consultation by a gastroenterologist who, after clinical evaluation, determined that upper endoscopy should be performed. 2. A group evaluated by a nongastroenterologist internist who, after clinical evaluation, determined that upper endoscopy should be performed. This procedure was done by a staff endoscopist without prior consultation by a gastroenterologist. 3. A group evaluated by a gastrointestinal surgeon who, after clinical evaluation, determined that upper endoscopy should be performed. This procedure was done by a staff endoscopist without prior consultation by a gastroenterologist. The endoscopic database also provided an opportunity to examine the results or findings from upper endoscopy. Before analyzing the database the authors established, for each indication for endoscopy, which results (i.e., endoscopic findings) would be considered "positive" (i.e., yielding information potentially significant for future patient care). The following criteria were established: (1) a malignant or premalignant condition (e.g., Barrett's esophagus) was considered a positive endoscopic finding for any indication; and (2) an endoscopic finding that potentially explained the patient's symptoms or indication for endoscopy and was unlikely to improve with short-term, empiric medical therapy was also considered positive. Therefore, a given endoscopic finding could be considered positive when associated with some indications, but not others. For example, duodenal ulcer was considered a "significant" finding when endoscopy was performed for evaluation of abdominal pain refractory to therapy, but not when endoscopy was performed for dyspepsia when a trial of therapy had not preceded endoscopy. A "normal" endosV O L U M E 46, NO. 1, 1997
Does an open access system properly utilize endoscopic resources?
Table 2. Request for upper endoscopy in an open access system Requesting specialty
Number
Total (%)
Gastroenterologists Nongastroenterologist internists Surgeons
1933 1139 643
52 31 17
Total
3715
copy was not considered a positive result, even if this i n f o r m a t i o n provided r e a s s u r a n c e for the p a t i e n t or
guided further medical care. A nonspecific finding, such as gastritis, was :motconsidered positive for any indication if it was the sole endoscopic finding. Data analysis included the overall frequency of "appropriate" referral for upper endoscopy in an open access system, and a comparison of appropriate referral by gastreenterologists, other internists, and surgeons. The yield of positive findings for upper endoscopy for gastroenterologists, other internists, and surgeons was compared. Also examined was the likelihood of a positive result of endoscopy for common indications listed in the AUGE. All statistical comparisons were performed using the chisquared test.
RESULTS Procedural indications A total of 3715 upper endoscopic procedures were entered into the database. A few indications accounted for the majority of upper endoscopic examinations (Fig. 1). Overall, abdominal pain/dyspepsia, dysphagia/odynophagia, and acute or Chronic gastrointestinal bleeding accounted for two thirds of the indications for endoscopy. The utilization of endoscopy by requesting physician is listed in Table 2. For upper endoscopy, 52% of cases were requested by gastroenterologists and 48% by other physicians. Table 3 lists the percentage of approved indications as per the AUGE for upper endoscopy. The overall rate of appropriate requests for upper endoscopy in this open referral system was 84%. The rate of approprial~e requests for endoscopy was 85% for gastroenterologists and 87% for gastrointestinal surgeons. Although the percentage of appropriate requests for upper endoscopy by nongastroenterologist internists was high (81%), it was significantly lower than the gastroenterologists (p = .004).
Endoscopic findings The overall frequency of positive endoscopic findings, i.e., findings potentially relevant to patient care, was 59%. The frequency of positive findings according to the specialty of the requesting physician is listed in Table 4. The frequency of positive V O L U M E 46, ND. 1, 1997
G Zuccaro, Jr, K Provencher
Table 3. "Appropriate" requests for upper endoscopy Total Total 95% endoscopies appropriate Confidence requested indications (%) interval
Requesting specialty Gastroenterolog/sts Nongastroenterologist internists Surgeons Total
1933 1139
1639 (85)* 919 (81)*
83.2%, 86.4% 78.4%, 83.0%
643 3715
558 (87) 3116 (84)
84.2%,89.4%
Difference between gastroenterologists and nongastroenterologist internists is 4.1%. 95% CI [1.3%, 6.9%]. *p = . 0 0 4 .
Table 4. "Positive" endoscopic findings per specialty Total Total procedures positive requested findings (%)
Requesting specialty Gastroenterologists Nengastroenterologist internists Surgeons Total
95% Confidence inter~zal
1933 1139
1193 (62)* 59.6%, 63.9% 589 (52)* 48.8%, 54.6%
643 3715
407 (63) 2189 (59)
59.6%,67.0%
Difference between gastroenterologists and nongastroenterologist internists is 10%. 95% CI [6.4%, 13.6%]. *p < .001. endoscopic findings in procedures requested by the gastroenterologists was 62%. Although the percentage of positive results for upper endoscopy requested by nongastroenterologist internists is reasonably high (52%), here again it is significantly lower than the gastroenterologists' (p = <.001).
Clinical utility of the AUGE Figure 2 depicts the frequency of positive endoscopic results for specific indications in the AUGE. There was variability in the frequency of positive endoscopic results for appropriate indications as listed in this practice guideline. The frequency of a positive result for dysphagia/odynophagia, therapy for acute gastrointestinal bleeding, clarification of an abnormality on UGI x-ray (e.g., mass, ulcer, stricture), or gastroesophageal reflux net responding to therapy was 74% to 84%. The likelihood of a positive result for these four indications was significantly greater than for the other indications listed in this Figure 2 (p < .001). For the other indications, evaluation of chronic gastrointestinal bleeding/ anemia, abdominal pain/distress not responding to therapy, abdominal pain/distress associated with weight loss/anorexia, or nausea/vomiting, the frequency of a positive result was 40% to 53%. Here again, the likelihood of a positive endoscopic result GASTROINTESTINAL ENDOSCOPY
17
G Zuccaro, Jr, K Provencher
Does an open access system properly utilize endoscopic resources?
Therapy for acute UG] bleeding Abnormal UGI Xray GE Reflux
despite therapy Dysphagla/ odynophagla
P <,go1
Chronic GI
bleeding/anemia Abdominal Pain despite therapy Abdominal Pain with anorexialwt loss Nausea/
Vomiting I
0
20
I
I
40 60 Percent Positive Results
I
F
80
100
Figure 2. Frequency of positive results for selected indications found in the AUGE. performed for evaluation of chronic gastrointestinal bleeding or abdominal pain not responding to therapy was significantly greater than for abdominal distress associated with weight loss or nausea/ vomiting (p < .001). Sixteen percent of upper endoscopic procedures were performed for indications not considered appropriate by the ASGE document. These indications included dyspeptic symptoms in which a trial of medical therapy had not been provided prior to endoscopy, pain believed to be most likely functional (where the procedure was performed to reassure the patient and/or physician), evaluation of symptoms of gastroesophageal reflux where medical therapy had resulted in satisfactory resolution of symptoms, and surveillance of conditions such as atrophic gastritis, healed esophagitis, or nonadenomatous gastric polyps. For the most common of these indications, namely, dyspepsia without a prior trial of medical therapy, a positive endoscopic finding occurred in only 35% of cases. DISCUSSION
In our experience with an open access system for upper gastrointestinal endoscopy, 84% of procedures were performed for approved indications listed in the practice guideline "The Appropriate Use of Gastrointestinal Endoscopy. "1 The majority of procedures (59%) yielded information potentially relevant to patient care. For both parameters (indications and results), specialists did significantly better (by comparative statistics) than nonspecialists. However, the difference between appropriate requests for upper endoscopy by gastroenterologists and other internists was only 4% (85% to 81%). Although this difference is statistically significant, it m a y not be clinically significant, i.e., the cost of having all patients requiring endoscopy be seen first by a gastroenterologist m a y exceed the cost savings in 18
GASTROINTESTINAL ENDOSCOPY
decreased "inappropriate" procedures. Our data suggest that an open access endoscopy system, properly structured and supported, m a y be one effective w a y for gastroenterologists to balance the problems of access, cost, and resource utilization in today's systern of health care. We are not surprised that our analysis shows essentially no difference between gastroenterologists and gastrointestinal surgeons in the appropriate utilization of endoscopy, as both have detailed training in digestive diseases. There were significant differences between gastroenterologists and nongastroenterologist internists in the rate of appropriate referral for endoscopy (85% vs 81%, p = .004), and the frequency of positive endoscopic results (62% vs 52%, p < .001). In this era of cost control through limitation of access to specialists, gastroenterologists will no doubt find some vindication in these findings. However, the lower frequency of appropriate endoscopy and positive endoscopic results requested by nongastroenterologist internists m a y be more related to differences in patient population than clinical acumen. Many patients originally seen by internists for symptoms such as dysphagia or UGI bleeding, and who have a high likelihood of a positive endoscopy, will be referred directly to a gastroenterologist or gastrointestinal surgeon for evaluation and management. In our system, the gastroenterologist or surgeon will then receive credit for requesting the endoscopy. Therefore, the gastroenterologist and surgeon have better statistics, but not necessarily because of a better sense of the appropriate use of gastrointestinal endoscopy. Open access endoscopy has been successful in other health care systems. Ready access to willing endoscopists was feared as a predictable method to increase costs and the rate of unnecessary endoscopy. However, the data on so-called open access endoscopy do not universally support this assumption. An early report from Holdstock et al. ~ did conclude that open access endoscopy in Southhampton, U.K., increased procedure numbers without a tangible increase in the quality of care. However, a subsequent report by Kerrigan et al. 3 showed no increase in the number of negative examinations performed in an open access system when compared to a consultative service in Sheffield, U.K. Experiences in both Australia and Italy support an open access system for endoscopy. 4' 5 Specifically, in an open access system in Italy, Minoli et al. ~ report that 32% of esophagogastroduodenoscopies requested by family physicians were performed for inappropriate indications as per the AUGE, compared to 19% of procedures requested by surgeons and 14% by gasV O L U M E 46, NO. 1, 1997
Does an open access system properly utilize endoscopic resources?
G Zuccaro, Jr, K Provencher
troenterologists (p < 0.001). Although we found similar differences in appropriateness of endoscopic utilization between generalists and specialists, our overall rate of inappropriate indications for nonspecialists was lower than this Italian experience. These differences may be due to the multicenter design of the Minoli study or differences in practice patterns of family physicians vs internists in different health care systems. Can findings on open access endoscopy at the Cleveland Clinic be generalized to other patient populations? In our study, 16% of endoscopic procedures were performed for inappropriate indications. In a RAND corporation study, Kahn et al. 7 applied a carefully developed list of indications for upper endoscopy to 1585 Medicare patients undergoing this procedure in 1981. Seventeen percent of procedures in their analysis were done for inappropriate indications (essentially the same as our experience). In other publications, this same group found that appropriateness of indications cannot explain geographic variability in the use of procedures such as upper endoscopy, and that patient, physician, or hospital demographics do not readily predict the likelihood of procedure appropriateness, s' 9 We believe our findings and conclusions on open access endoscopy are relevant to other patient populations and U.S. healthcare delivery systems. Another unique aspect of our study is the analysis of the clinical utility of the AUGE with respect to the diagnostic yield of endoscopy for specific indications listed in the practice guideline. The limitations of the AUGE of the American Society for Gastrointestinal Endoscopy with respect to procedure indications was addressed recently by Kahn et al. 1° They compared the indication list developed by RAND/ UCLA Health Services Utilization Study to those of the AUGE in the assessment of procedure indications for the same 1585 upper endoscopic tests in Medicare patients. They found that the AUGE criteria could only be applied to 70% of the endoscopic procedures, i.e., more specific criteria were needed if the AUGE were universally applied. The exact role of Helicobact, er pylori in the evaluation of patients with ulcerlike symptoms is one specific example lacking in the current AUGE document. Specific criteria as to what constitutes an "appropriate trial of medical therapy" for dyspepsia or reflux symptoms are also needed. In our analysis of procedure results, significantly higher diagnostic yields were associated with the least subjective of those indications listed in the AUGE (e.g., dysphagia or therapy for acute UGI bleed). In other words, there is likely less variability in clinical circumstance and differences in interphysician interpretation of "therapy
for acute UGI bleeding" compared to "nausea/ vomiting" or "abdominal distress unresponsive to therapy." In our analysis, a normal examination was never considered a positive endoscopic finding. Perhaps in a perfect world, a normal endoscopy would never be performed. Realistically, however, normal endoscopic procedures will be performed for the most universally accepted indications, e.g., upper endoscopy for refractory nonulcer dyspepsia or surveillance colonoscopy for adenomatous polyps. It can also be argued that a negative endoscopy m a y be valuable to patients and clinicians alike. In an analysis of 483 endoscopic procedures by Naji et al., 11 69% of procedures were characterized as positive, but over 80% were considered by the gastroenterologist as helpful in patient management. Our analysis therefore could underestimate the overall contribution of endoscopy to patient care. A persistent concern regarding practice guidelines is their potential restriction of physician judgement and instincts. Most physicians believe that there are clinical circumstances where a test or procedure m a y be warranted, even if practice guidelines suggest that the diagnostic yield is low. This concern m a y well be justified, as practice guidelines often reflect far more expert opinion than the result of controlled clinical studies. As discussed above, our percentage of upper endoscopy for inappropriate indications is virtually identical to that of other published series. Again, our study reinforces the need for future guidelines to be based more on outcome data than opinion. The leadership of the major gastroenterology societies have recognized this, as evidenced by the increased emphasis on funding of endoscopic outcomes research. There are limitations with our analysis. The frequency of positive endoscopic results must be interpreted with caution. First, a positive result on endoscopy does not always mean that the procedure was essential to patient care. For example, a patient with iron deficiency anemia might be referred for upper endoscopy and found to have an asymptomatic nonsteroidal-induced antral ulceration. This would be interpreted from the database as a positive result. If this patient was subsequently found to have another, more significant finding (e.g., colon cancer), the database analysis for upper endoscopy would not recognize this. A second limitation relates to the method by which indications were coded into the database. Procedure indications were recorded by the endoscopist at the time of the procedure. Because this was an open referral system involving 12 endoscopists, there was the potential for marked variability among endoscopists for recording indica-
V O L U M E 46, NO. 1, 1997
GASTROINTESTINAL ENDOSCOPY
19
G Zuccaro, Jr, K Provencher
tions into the database. Indeed, as gastrointestinal endoscopists are likely quite familiar with the content of the AUGE, it is n a t u r a l to assume a tendency to record an approved indication over an unapproved one. Therefore, the authors reviewed a random sample of 10 medical records from each endoscopist. In our judgement, the procedure indication and the medical record were concordant 93% of the time. We conclude t h a t adherence to practice guidelines for procedure indications can and does occur in an open access endoscopy system. This is associated with a reasonably high yield of positive procedure results. Gastrointestinal specialists request endoscopy more frequently for appropriate indications compared to other internists and have a higher yield of information relevant to patient care. F u r t h e r refinement and better definition of some indications within the AUGE is needed to further increase the clinical utility of this document. REFERENCES 1. Committee on Standards of Practice. Appropriate use of gastrointestinal endoscopy. Manchester, Massachusetts: American Society for Gastrointest Endoscopy, 1992.
Does an open access system properly utilize endoscopic resources?
2. Holdstock G, Wiseman M, Loehry CA. Open-access endoscopy service for general practitioners. BMJ 1979;1:457-9. 3. Kerrigan DD, Brown SR, Hutchinson GH. Open access gastroscopy: too much to swallow? BMJ 1990;300:374-6. 4. Goy JA, Herold E, Jenkins PJ, Colman JC, Russell DM. "Open-access" endoscopy for general practitioners. Med J Aust 1986;144:71-4. 5. Mansi C, Mela GS, Savarino V, Mele MR, Valle F, Celle G. Open access endoscopy: large-scale analysis of its use in dyspeptic patients. J Clin Gastroenterol 1993;16:149-53. 6. Minoli G, Prada A, Gambetta G, Formenti A, Schalling R, Lai L, et al. The ASGE guidelines for the appropriate use of upper gastrointestinal endoscopy in an open access system. Gastrointest Endosc 1995;42:387-9. 7. Kahn KL, Kosecoff J, Chassin MR, Solomon DH, Brook RH. The use and misuse of upper gastrointestinal endoscopy. Ann Intern Med 1988;109:664-70. 8. Brook RH, Park RE, Chassin MR, Solomon DH, Keesey J, Kosecoff J. Predicting the appropriate use of carotid endarterectomy, upper gastrointestinal endoscopy, and coronary angiography. N Engl J Med 1990;323:1173-7. 9. Chassin MR, Kosecoff J, Park RE, Winslow CM, Kahn KL, Merrick NJ, et al. Does inappropriate use explain geographic variations in the use of health care services? JAMA 1987;258: 2533-7. 10. Kahn KL, Park RE, Vennes J, Brook RH. Assigning appropriateness ratings for diagnostic upper gastrointestinal endoscopy using two different approaches. Med Care 1992;30: 1016-28. 11. Naji SA, Brunt PW, Hagen S, Mowat NA, Russell IT, Sinclair TS, et al. Improving the selection of patients for upper gastrointestinal endoscopy. Gut 1993;34:187-91.
Bound volumes available to subscribers B o u n d volumes of G a s t r o i n t e s t i n a l E n d o s c o p y are available for the 1997 issues from the P u b l i s h e r at a cost of $78.50 ($93.50 i n t e r n a t i o n a l , $100.05 C a n a d a ) for Vol. 45 ( J a n u a r y - J u n e ) a n d Vol. 46 (July-December). Shipping charges are included. Each b o u n d volume contains a subject a n d a u t h o r index a n d all advertising is removed. Copies are shipped w i t h i n 60 days after publication of the last issue of the volume. The b i n d i n g is durable b u c k r a m with the j o u r n a l n a m e , volume n u m b e r , a n d year s t a m p e d i n gold on the spine. P a y m e n t m u s t a c c o m p a n y all orders. Contact Mosby, Subscription Services D e p a r t m e n t , 11830 Westline I n d u s t r i a l Drive, St. Louis, MO 63146-3318, USA; phone (800)453-4351 or (314)453-4351.
20
GASTROINTESTINAL ENDOSCOPY
VOLUME 46, NO. 1, 1997